SEARCH

SEARCH BY CITATION

Keywords:

  • AGAMOUS-LIKE13 ;
  • MADS box gene;
  • fluorescence resonance energy transfer;
  • heterotetrameric protein complex;
  • pollen morphogenesis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Arabidopsis AGL13 is a member of the AGL6 clade of the MADS box gene family. GUS activity was specifically detected from the initiation to maturation of both pollen and ovules in AGL13:GUS Arabidopsis. The sterility of the flower with defective pollen and ovules was found in AGL13 RNAi knockdown and AGL13 + SRDX dominant-negative mutants. These results indicate that AGL13 acts as an activator in regulation of early initiation and further development of pollen and ovules. The production of similar floral organ defects in the severe AGL13 + SRDX and SEP2 + SRDX plants and the similar enhancement of AG nuclear localization efficiency by AGL13 and SEP3 proteins suggest a similar function for AGL13 and E functional SEP proteins. Additional fluorescence resonance energy transfer (FRET) analysis indicated that, similar to SEP proteins, AGL13 is able to interact with AG to form quartet-like complexes (AGL13–AG)2 and interact with AG–AP3–PI to form a higher-order heterotetrameric complex (AGL13–AG–AP3–PI). Through these complexes, AGL13 and AG could regulate the expression of similar downstream genes involved in pollen morphogenesis, anther cell layer formation and the ovule development. AGL13 also regulates AG/AP3/PI expression by positive regulatory feedback loops and suppresses its own expression through negative regulatory feedback loops by activating AGL6, which acts as a repressor of AGL13. Our data suggest that AGL13 is likely a putative ancestor for the E functional genes which specifies male and female gametophyte morphogenesis in plants during evolution.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

vFlower development is thought to be regulated by five classes of MADS box genes designated as ABCD and E in various plant species (Coen and Meyerowitz, 1991; Theissen, 2001; Theissen and Saedler, 2001; Pařenicová et al., 2003). It has been thought that there is a special role for E functional proteins in this model, as they could play a major role as mediators to form a higher-order protein complex with A/B/C/D proteins (Immink et al., 2009; Melzer and Theissen, 2009; Melzer et al., 2009; Smaczniak et al., 2012).

All of the ABCDE MADS box genes are of the MIKC type. The functions of most of the other MIKC-type MADS box genes that have not been assigned to any of the five ABCDE groups remain unknown. For example, AGL6-like genes in the A/E class of the AP1/AGL9 subfamily (Pařenicová et al., 2003) have been cloned from different plant species (Ma et al., 1991; Mena et al., 1995; Tandre et al., 1995; Mouradov et al., 1998; Shindo et al., 1999; Hsu et al., 2003; Chang et al., 2009; Rijpkema et al., 2009; Li et al., 2010). Only a few functional characterizations have been performed for these genes. In the evolutionary tree, AGL6 lineage genes were discovered in gymnosperms (Winter et al., 1999). The AGL6 lineage gene in Norway Spruce, DAL1, has a high expression level in ovuliferous scale primordia of the ovulate cone. In the later developmental stages, DAL1 was expressed in ovuliferous scales (Carlsbecker et al., 2004). In male reproductive organs, DAL1 was detected in whole pollen cone morphogenesis stages (Carlsbecker et al., 2004). The sequences of the AGL6 lineage genes are most similar to the E functional genes of the MADS box family. Interestingly, to date, no E functional genes have been found in gymnosperm plants. In contrast, the AGL6 lineage genes were duplicated and divided into two genes in gymnosperms (Winter et al., 1999; Li et al., 2010; Melzer et al., 2010). This division indicates that the function of the AGL6 lineage genes, such as DAL1, in regulating male and female gametophyte morphogenesis is primitive and conserved in plants. This assumption is supported by the expression patterns in the stamen and pistil of AGL6 lineage genes in monocot grasses (Reinheimer and Kellogg, 2009).

SEP genes may be the redundant counterparts for the AGL6 lineage genes generated by the duplication of the AGL6 lineage genes in angiosperms during evolution. Thus, the AGL6 lineage genes must have a similar function and characteristics as SEP genes. One important remaining question is which gene in angiosperm plants is orthologous to DAL1 and plays the same developmental role as DAL1 in gymnosperm plants. It has been reported that the petunia AGL6 gene (PhAGL6) and rice AGL6 gene (OsMADS6) may have an E functional SEPALLATA-like function in floral patterning (Rijpkema et al., 2009; Li et al., 2011). However, no further molecular or biochemical evidence has been provided. Arabidopsis AGL13, an AGL6-like gene (Pařenicová et al., 2003), has been reported to be expressed in the ovule and anther (Rounsley et al., 1995; Schauer et al., 2009). Therefore, clarifying its function and determining the evolutional and functional relationship between AGL13 and E functional genes during reproductive organ development is important.

In this study, we performed a functional analysis of AGL13 by studying loss-of-function transgenic plants. We found the AGL13 gene may play an important role as an E functional gene and interact with C functional AG proteins to form quartet-like protein complexes to regulate the initiation and formation of ovules, whereas AG–AGL13–AP3/PI heteromeric complexes are formed to control the initiation and formation of pollen throughout flower development. Thus, our characterization of AGL13 provides useful information for understanding the function and the ancestral nature of AGL6-like MADS proteins for E functional genes in plants during evolution.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

GUS staining pattern reveals the specific expression of AGL13 in pollen and ovules

To investigate the expression of AGL13, the construct 13P-1–GUS, containing the entire exon 1, intron 1 and the 5′ part of exon 2 of AGL13 fused to the GUS reporter gene driven by the 1.4 kb promoter sequence of AGL13, was transformed into Arabidopsis. A construct, 35S–GUS, containing a 5′ partial sequence of AGL13 exon 1 fused to the GUS reporter gene driven by the CaMV-35S promoter, was used as a control (Figure S1).

In 35S–GUS plants, GUS was strongly detected throughout the whole plant (Figure S2a). In 13P-1–GUS plants, GUS was only detected in the flowers (Figure S2b). The GUS was specifically detected in the centre of the flower primordia in early stage (2–4) flowers and was observed in the carpel and stamen primordia in stage 4–5 flowers (Figures S2c and S3a–c). Later, GUS was strongly detected in the anthers and developing pollen of the stamen until stage 12 (Figure S2d–g). The expression of GUS in pollen was undetectable after pollen maturation in the flower after stage 12 (Figure S2h,i,k).

When the anthers of different developmental stages were analyzed, a specific pattern for GUS expression was observed. In the anther of stage 8–10 flowers, GUS was specifically detected in the microsporangium, microspore mother cells (stage 8), tapetum, developing pollen (stage 9–10) and the endothecium (Figure S3d–h). In contrast, GUS was absent in the epidermis, vascular bundle and connective tissue (Figure S3d–f). Later, GUS remained only in the tapetum, endothecium and pollen grains of the anther in stage 11 flowers (Figure S3i) and was completely absent in all anther tissues after stage 12 (Figure S3j).

GUS was also detected in the ovules of carpels beginning in stage 8–9 flowers, when the ovule primordia initiate (Figures S2e and S4a). The GUS expression was consistent and strong in the funiculus, nucellus and both the outer and inner integuments of 2- to 8-nucleate embryo sacs during ovule development in stage 12 flowers (Figure S4b). Later, GUS was detected consistently in the mature ovules of stages 12–14 flowers (Figures S2h–j and S4c–e), even after stage 16 when early embryo development had occurred (Figure S2k).

Pollen development is altered in AGL13 RNAi transgenic plants

To explore the role of AGL13, we generated mutants of AGL13 by RNAi. In total, 14 of the 23 independent 35S:AGL13 RNAi plants sharing a similar sterile flower phenotype were obtained (Figure 1a). The siliques failed to elongate during late development (Figure 1b). This phenotype was different from that of the wild-type inflorescence, which contained fully developed siliques (Figure 1a,c). To explore whether the mutant phenotype in 35S:AGL13 RNAi transgenic plants correlates with AGL13 expression, we performed reverse transcription polymerase chain reaction (RT-PCR). As shown in Figure 1(d), AGL13 expression is almost completely suppressed in the 35S:AGL13 RNAi transgenic plants that have severe phenotypes. In contrast, AGL13 expression levels are similar in wild-type plants and 35S:AGL13 RNAi transgenic plants with wild-type phenotypes (Figure 1d). This result indicates that the phenotype of 35S:AGL13 RNAi transgenic Arabidopsis correlates with the suppression of the AGL13 gene. We were also interested in whether the expression of other genes is altered in 35S:AGL13 RNAi transgenic plants. For this purpose, we analyzed the expression of AGL6, the gene with the highest level of identity to AGL13 in the AGL6-like gene family. As shown in Figure 1(d), levels of AGL6 expression do not change in 35S:AGL13 RNAi transgenic plants, indicating that the phenotype of 35S:AGL13 RNAi transgenic Arabidopsis plants is due to the specific suppression of the AGL13 gene alone.

image

Figure 1. Phenotypic analysis of transgenic Arabidopsis plants ectopically expressing AGL13 RNAi.(a) A 40-day-old 35S:AGL13 RNAi plant was sterile and produced short siliques (right), whereas a wild-type plant produced well developed long siliques (left).(b) The inflorescences of a 35S:AGL13 RNAi plant with short siliques (arrowed).(c) The inflorescences of a wild type plant with elongated siliques (arrowed).(d) Detection of AGL13 and AGL6 expression in 35S:AGL13 RNAi Arabidopsis plants. mRNA accumulation for AGL13 and AGL6 was determined by RT-PCR. Plants 4 and 20 (labeled by *) were severe 35S:AGL13 RNAi plants, whereas plants 8 and 23 were wild-type-like transgenic plants. The results indicate that the level of AGL13 expression was significantly reduced and much lower in plants 4 and 20 than in plants 8 and 23. The expression of the AGL6 was similar in wild-type and the four 35S:AGL13 RNAi plants.(e) Flowers in different stages (with number) of wild-type (top row) and 35S:AGL13 RNAi plants (bottom row).(f) The stigma of a stage 14 wild-type flower containing numerous pollen grains (arrowed) released from the dehiscent anther (an) were observed.(g) Pollen grains (arrowed) were not observed in the stigma of a stage 14 35S:AGL13 RNAi flower.(h) The wild-type flower pollinated with either 35S:AGL13 RNAi (middle, wt × 13i) or wild-type (right, wt × wt) pollen developed into elongated siliques. An undeveloped silique (s) from a 35S:AGL13 RNAi flower was shown in the left.(i) The wild-type ovules developed into embryos (arrowed) after manual pollination with 35S:AGL13 RNAi (left, wt × 13i) or wild-type (right, wt × wt) pollen grains.(j) Close-up of the normal (n) embryo from (i).(k) The 35S:AGL13 RNAi flower pollinated with wild-type pollen developed into elongated siliques (arrowed), whereas short siliques (s) were developed in the absence of wild-type pollen pollination.(l) About half of the ovules developed into embryos (arrowed) in the silique (13i × wt) from (k).(m) Close-up of the normal (n) and aborted (arrowed) embryos from (l).

Download figure to PowerPoint

When the flowers were examined, pollination occurred normally in the wild-type flower (Figure 1e,f) but was altered in the 35S:AGL13 RNAi flowers due to defects in the pollen grains attached to the stigmatic papillae (Figure 1e,g). As a result, these 35S:AGL13 RNAi flowers were sterile and no further elongation of the siliques was observed (Figure 1e).

Pollen grains of wild-type Arabidopsis were 30 × 15 micrometres and had three colpi, which gave them a slightly triangular aspect (Figure S5a–c). The surface of the outer exine wall of Arabidopsis pollen has a typical irregular structure and is deposited by the tapetum (Figure S5d). Interestingly, defective and wild-type-like pollen were observed in approximately equal amounts in the dehiscent anther of 35S:AGL13 RNAi flowers (Figure S5e). The defective pollen were smaller in size and had a flat or collapsed shape with aberrations in the exine patterning of the outer wall (Figure S5f–i). Although wild-type-like pollen were produced, breakage and aberrations in the exine patterning of the outer wall were observed (Figure S5g,j,k). To examine the viability of the pollen, the stigmas of the wild-type flowers were pollinated manually with 35S:AGL13 RNAi pollen. We found that silique elongation occurred (Figure 1h) and ovules further developed into embryos (Figure 1i,j). Thus, most of the wild-type-like pollen in 35S:AGL13 RNAi flowers were still functioning. Further analysis indicated that the 35S:AGL13 RNAi pollen were more easily washed out from the stigma than wild-type pollen by a retention assay (Ishiguro et al., 2010) (Figure S6a–d). This result revealed that the aberrations in the exine patterning of the outer wall reduced the ability for these 35S:AGL13 RNAi pollen to stably attach to the stigma, but the ability of these pollen grains to germinate and fertilize with egg cells was not influenced (Figure 1h–j). Furthermore, a similar thickness (approximately 1 μm) of the outer wall between the 35S:AGL13 RNAi wild-type-like (Figure S6g,h) or collapsed pollen (Figure S6i,j) and wild-type pollen (Figure S6e,f) was observed. This result revealed that the mutation in AGL13 causes surface defects with aberrations in the exine patterning but not in the thickness of the outer wall in 35S:AGL13 RNAi pollen.

To further explore the role of AGL13 in pollen formation, the 35S:AGL13 RNAi construct was introduced into the quartet1-2 (qrt1-2) mutants to generate 35S:AGL13 RNAi/qrt1-2 plants. As qrt1-2 mutation produced four attached microspores that derived from a single pollen mother cell during pollen development, which makes it as an excellent approach to perform tetrad analyses (Preuss et al., 1994; McCormick, 2004; Fujiki et al., 2007). The result indicated that normal viability was observed in four pollen grains of each qrt1-2 tetrad (Figure 2a–d). In contrast, only two of the four pollen grains that exhibited a wild-type shape were stained a dark red colour in each 35S:AGL13 RNAi/qrt1-2 tetrad (Figure 2e–h). The production of two defected pollen grains suggests that the development of the pollen that carried the 35S:AGL13 RNAi was arrested after meiosis and the tetrad formation. Further analysis indicated that the two wild-type-like pollen of the 35S:AGL13 RNAi/qrt1-2 tetrad were in tricellular stage in which two sperm cells and one vegetative nucleus were clearly observed (Figure 2i–l). This observation indicates that the pollen mitosis I (PM I) and II (PM II) proceed normally in these two wild-type-like pollen of the 35S:AGL13 RNAi/qrt1-2 tetrad.

image

Figure 2. Alexander's and 4′,6-diamidino-2-phenylindole (DAPI) staining of pollen produced in qrt1-2 and 35S:AGL13 RNAi/qrt1-2 tetrads.(a, b) Tetrads released from the qrt1-2 anther. Four pollen grains (stained in dark red colour) indicated in number were visualized in one tetrad.(c) Pollen grain with the outer wall stained in green colour (arrowed) in a qrt1-2 tetrad.(d) Confocal laser scanning microscopy (CLSM) of four pollen grains (indicated in number) produced in qrt1-2 tetrad. Colpi are arrowed.(e, f) Tetrads released from the 35S:AGL13 RNAi/qrt1-2 anther. Two wild-type-like pollen grains (stained in dark red color) indicated in number were visualized whereas the other two defected pollen grains not stained in color were in collapsed shape (d1, d2). Other tetrads with two visualized dark red color pollen grains were circled.(g) The wild-type-like pollen grain in a 35S:AGL13 RNAi/qrt1-2 tetrad with the outer wall stained in green color (arrowed).(h) CLSM of 35S:AGL13 RNAi/qrt1-2 tetrads that composed of two wild-type-like pollen grains (1 and 2) and two defected pollen grains (d1 and d2). Colpi are arrowed.(i–l) DAPI staining of a 35S:AGL13 RNAi/qrt1-2 tetrad. Two sperm cells (green arrow) and one vegetative nucleus (blue arrow) were observed in two wild-type-like pollen grains (1 and 2). Neither sperm nor vegetative nucleus was observed in defected pollen grains (d1 and d2).(i) Is in bright field. DAPI fluorescence was examined by different focus (j–l) to reveal the nucleus in different pollen grains under the microscope. Bar = 20 μm in (a, e); 10 μm in (b, d, f, h–l) and 5 μm in (c, g).

Download figure to PowerPoint

Ovule development is altered in AGL13 RNAi transgenic plants

To explore whether the formation of the ovule was altered, the carpels were mounted and cleared with clear solution (50 ml water, 5 g phenol, and 5 g chloral hydrate) for 2 days, the ovules were examined with an optical microscope. The ovules were arranged in a satisfactory order in each locule (Figure S6k) in the wild-type ovary. In contrast, 35S:AGL13 RNAi produced only half of the wild-type-like ovules which were clearly improperly arranged in each locule due to the random absence ovules in each row (Figure S6l). These 35S:AGL13 RNAi wild-type-like ovules (Figure S6n) were morphologically indistinguishable from normal ovules in wild-type flowers (Figure S6m). The carpels at stage 12 in both 35S:AGL13 RNAi and wild-type plants were further examined by CLSM. The wild-type ovary is divided into two locules by a central septum (Figure S7a). There are two rows of ovules arranged in a good order in each locule, and a total of 50–60 ovules are produced in four rows (Figure S7a–c). Unlike the wild-type ovary, only half (about 25–30) of the wild-type-like ovules were produced in the ovary of 35S:AGL13 RNAi stage 12 flowers (Figure S7d,e). The arrangement of these ovules into two rows in each locule clearly indicates that about half of the ovules at random positions in the locule aborted early and failed to develop (Figure S7d–f). This arrangement caused the formation of the warped central septum (Figure S7e). The wild-type-like ovules (Figure S7e) were similar in appearance to wild-type ovules (Figure S7b).

When the stigmas of the 35S:AGL13 RNAi flowers were pollinated manually with wild-type pollen, silique elongation occurred (Figure 1k) and approximately half of the ovules developed into embryos (Figure 1l,m). This result supports the observation that about half of the ovules were aborted during early development stage whereas the other half of the wild-type-like ovules are still functioning in 35S:AGL13 RNAi flowers.

Ectopic expression of AGL13 + SRDX causes similar alteration of pollen and ovule development and the sterility of the plants

AGL13 may be functionally redundant with other genes. To test this hypothesis, transgenic dominant loss-of-function mutant plants were generated by fusing a conserved SRDX-suppressing motif that contained a 12 amino acid repressor sequence (LDLDLELRLGFA) to AGL13 (AGL13 + SRDX). This strategy has been thought to provide efficient suppression of the putative target genes in the presence of native and redundant genes, thus resulting in the generation of the mutant phenotypes that could be more severe than RNAi knockdown lines (Koyama et al., 2012). Therefore, this strategy has already been successfully used to generate dominant-negative mutants for studying the function of transcriptional activators with redundant functions (Hiratsu et al., 2003; Eklund et al., 2010; Guo et al., 2010; Tejedor-Cano et al., 2010; Koyama et al., 2012).

Interestingly, a sterile flower phenotype similar to that observed in the 35S:AGL13 RNAi mutants was observed in AGL13 + SRDX plants (Figure 3a,b). Similar to 35S:AGL13 RNAi pollen, normally viable and nonviable pollen with small and collapsed shapes were observed in AGL13 + SRDX flowers (Figure 3c,d). When the stigmas of the AGL13 + SRDX flowers were pollinated manually with wild-type pollen, approximately half of the total number of ovules developed into embryos while the other half aborted during early development (Figure 3e–h). This result supports the observation that AGL13 + SRDX flowers produce only half of the wild-type-like ovules, which are still functioning similar to that observed in 35S:AGL13 RNAi flowers. The similar mutant phenotype in AGL13 + SRDX and 35S:AGL13 RNAi flowers indicates that the AGL13 gene encodes a transcriptional activator.

image

Figure 3. Phenotypic analysis and detection of gene expression in transgenic Arabidopsis plants ectopically expressing AGL13 + SRDX or SEP2 + SRDX.(a, b) The inflorescence (a) and a flower (b) of an AGL13 + SRDX plant. The AGL13 + SRDX flowers were sterile and contained sepals (s), petals (p) and stamens (st) clearly shorter than those in wild-type flowers. c, carpel.(c, d) Pollen with normal viability (w, stained dark red) or non-viability (d) with small and collapsed shapes were observed in AGL13 + SRDX anther.(e, f) The silique elongation occurred in an AGL13 + SRDX flower after pollinating manually with wild-type pollen. Approximately half of the ovules developed into embryos and seeds (s) and the other half aborted (arrowed) during early development.(g, h) An aborted ovule (ov) with funiculus (fu) (g) and a funiculus with breakage (arrowed) where the mature seed separated (h).(i, j) The inflorescence (i) and a flower (j) of an AGL13 + SRDX plant with a more severe phenotype than that in (a). These flowers were sterile and contained much shorter sepals (s), petals (p) and stamens (st) than those in wild-type flowers. c: carpel.(k, l) The inflorescence (k) and a flower (l) of a SEP2 + SRDX plant. These flowers were sterile and produced shorter sepals (s), petals (p) and stamens (st) than those in wild-type flowers. c: carpel.(m, n) Detection of gene expression in AGL13 + SRDX (m) and SEP2 + SRDX (n) plants. mRNA accumulation for AGL13 + SRDX (13-SRDX), SEP2 + SRDX, AG, PI, AP3, AP1 and AP2 was determined by real-time quantitative polymerase chain reaction (PCR). Total RNA was isolated from the flower buds before stage 12 of one wild-type Columbia plant (wt), of one AGL13 + SRDX plant (#214 in m), of two SEP2 + SRDX plants (SEP2# 1 and 3 in n) with mutant phenotypes and of one AGL13 + SRDX plant (#203 in m) with a wild-type-like phenotype. The expression of each gene was relative to that of the wild-type plant, which was set at 1. The error bars represent the standard deviation. Each experiment was repeated twice with similar results.

Download figure to PowerPoint

The AGL13 + SRDX mutant phenotypes resemble the SEP2 + SRDX phenotypes

In addition to the sterility of flowers with defects in pollen and ovule development, some severe lines of the AGL13 + SRDX plants with additional floral organ defects were produced (Figure 3i,j). The lengths of the sepal, petal and anther in these AGL13 + SRDX flowers were considerably shorter (Figure 3i,j) than those in wild-type flowers. Because the most putative redundant genes for AGL13 in regulating flower development could be E functional genes, it is interesting to see if the 35S:SEP+SRDX also showed a similar phenotype. When 35S:SEP2 + SRDX plants were generated, these plants produced phenotypes of flower sterility, with short floral organs (Figure 3k,l) similar to that observed in AGL13 + SRDX plants. The mutant phenotypes were strongly correlated with the down-regulation of the ABC functional genes AP1/AP2/AP3/PI/AG in both AGL13 + SRDX and SEP2 + SRDX Arabidopsis, which showed high expression levels for transgenes (Figure 3m,n). Thus, AGL13 and SEP2 should possess a similar function and target to similar downstream genes once fused with SRDX and ectopically expressed in the flower organs.

AG nuclear localization efficiency is enhanced similarly by the presence of AGL13 and SEP3

E functional proteins have been thought to cooperate with AG to form a higher-order complex regulating stamen/carpel formation (Honma and Goto, 2001; Theissen and Saedler, 2001; Favaro et al., 2003; Kaufmann et al., 2005; Smaczniak et al., 2012). Because the similar phenotype in AGL13 + SRDX and SEP2 + SRDX transgenic plants reveals the possible redundant function for AGL13 and SEPs, the exploration of the relationship between AG and these two proteins is interesting. To test the cellular localization of AG, constructs containing AG–CFP and AG–YFP fusion proteins were generated. The transient expression of AG–YFP in tobacco leaf cells indicated that AG–YFP was detected both in the cytoplasm and nucleus (Figure 4a-1). When AG–YFP was co-expressed with AGL13, the AG–YFP signal was exclusively detected in the nucleus and was undetectable in the cytoplasm (Figure 4a-2). Similarly, in the AG–CFP and SEP3 co-expressing cells, the AG–CFP signal was also exclusively detected in the nucleus (Figure 4a-3). This result indicates that AGL13 and SEP3 could interact with AG and enhance the nuclear localization efficiency of AG in a similar way.

image

Figure 4. The detection of the interaction between AGL13 and AG proteins. (a) Tobacco leaf cells were transfected with AG–YFP (a-1), AG–YFP and AGL13 (a-2) and AG–CFP and SEP3 (a-3). AG–YFP was detected both in the cytoplasm (c) and the nucleus (n) (a-1). The AG–YFP (a-2) and AG–CFP (a-3) signal was exclusively detected in the nucleus (n). Bar = 20 μm (b) Analysis of the interaction between AGL13 and AG proteins through the FRET technique. In the top image, CyPet- and YPet-fused protein pair fluorescence signals are detected in the nucleus (n) and the cytoplasm (c) expressed in tobacco leaves. CFP and YFP channels were exited with a 440 nm laser and these two channels were used to calculate the raw FRET signal. Finally, raw FRET values were divided by CFP signals to calculate the FRET efficiency. The average FRET efficiency values (Bottom) were quantified in multiple samples (n > 4). Empty CyPet and YPet protein pairs were used as a FRET signal control. Image frame = 20 × 20 μm2. (c) Schematic model for the protein interaction combination for AGL13 and AG proteins. AG proteins could form homodimers to enter into the nucleus with low efficiency. AGL13 proteins are not able to form homodimers. The rapid entry of AG proteins into the nucleus is accomplished by the interaction with AGL13 proteins to form a tetrameric complex.

Download figure to PowerPoint

AGL13 can not form homodimers and is able to interact with AG to form quartet-like complexes

To further verify the relationship between AGL13 and AG, the fluorescence resonance energy transfer (FRET) technique was applied to calculate the protein interaction level for these two proteins. Using non-fusion CFP and YFP as negative controls (Figure 4b, column 1), AGL13:CFP and AGL13:YFP showed low FRET efficiency value (<10) in the nucleus (Figure 4b, column 2). This result indicates the low efficiency of AGL13 to form homodimers. The homodimers of AG (AG:CFP and AG:YFP) were detected with the FRET efficiency value of approximately 25–35%, with the signal slightly higher in the nucleus (Figure 4b, column 4) than that in the cytoplasm (Figure 4b, column 3). The nuclear signal of the AG homodimer was enhanced by the addition of the AGL13 protein, when the FRET efficiency value rose to 60% (Figure 4b, column 5). The 50% FRET efficiency value of AG:CFP and AGL13:YFP indicates that the AGL13 protein stably interacts with AG (Figure 4b, column 6). Thus, AG could form homodimers and interact with AGL13 at the same time and the entry ratio of the nucleus rose. The most probable protein complex of AG and AGL13 was a tetramer: two groups of AG–AGL13 heterodimers bound together with the same proteins set diagonally (Figure 4c).

AGL13 interacts with B and C functional MADS box proteins to form heteromeric higher-order MADS-domain protein complexes

Based on the results described above, there is a high possibility that AGL13 has characteristics similar to E functional proteins. The E functional protein SEP3 has been thought to interact with the B functional AP3/PI heterodimer to form quartet-like complexes (Favaro et al., 2003; Kaufmann et al., 2005; Melzer and Theissen, 2009; Melzer et al., 2009). To further explore the relationship between E functional proteins and AGL13, the interaction between the PI/AP3 heterodimer and AGL13 was detected by FRET assay. The co-expression admixture of ‘PI:CFP, AGL13:YFP and AP3’ and ‘AP3:CFP, AGL13:YFP and PI’ revealed similar FRET efficiency values of 50% (Figure 5a, columns 1, 3). These results indicate that, similar to SEP3, the AGL13 protein could interact with the AP3–PI heterodimer (Figure 5b-1,b-3). It has been shown that E functional proteins are also able to coordinate B and C proteins to form heteromeric higher-order complexes (Smaczniak et al., 2012). With the co-expression admixture of ‘PI:CFP, AGL13:YFP and AP3, AG’, the FRET efficiency value decreased from 50 to 35% (Figure 5a, column 2). However, the FRET efficiency value (at 50%) was unchanged with the co-expression admixture of ‘AP3:CFP, AGL13:YFP and PI, AG’ (Figure 5a, column 4). These results indicate that the position of the PI binding site for AGL13 was occupied by AG and AGL13 is located diagonally to the PI protein (Figure 5b-2,b-4). Additionally, the FRET pair AG:CFP and AGL13:YFP exhibited no change in their FRET efficiency value of 50% with the addition of the AP3–PI dimer (Figure 5a, column 5). These data indicate that the AG–AGL13 heterodimer could interact with the AP3–PI heterodimer to form a higher-order heterotetramer (Figure 5b-5).

image

Figure 5. The detection of the interaction between AGL13, AG, AP3 and PI proteins by FRET.(a) In top image, four channels were exhibited in FRET imaging. Different ectopic protein compositions were expressed in tobacco cells and the fluorescence signals in the cell nucleus were detected with a confocal microscope. The average FRET efficiency values (Bottom) were quantified in multiple samples (n > 4). Image frame = 20 × 20 μm2.(b) Schematic model for the protein interaction combination for AGL13, PI, AP3 and AG proteins. The AGL13 protein interacts with the PI–AP3 to form a tetrameric complex (b-1, b-3). The heterotetrameric complex (b-2, b-4, b-5) is formed when AG, AGL13, PI and AP3 are co-expressed in the cells.

Download figure to PowerPoint

AGL13 and AG regulate similar downstream genes controlling anther and pollen development

Because AGL13 can form quartet-like complexes with AG (Figure 4) and form a heterotetramer with AG and AP3/PI (Figure 5), it is reasonable to believe that AGL13 and AG may regulate similar downstream genes during anther (pollen) and ovule development due to their co-existence in the same complexes. It has been reported that AG binds to the upstream element of SPL/NZZ to initiate anther cell layer specification (Ito et al., 2004). To examine whether AGL13 is also involved in regulating the SPL/NZZ pathway similar to AG, the expression levels of the pollen cell layer specification genes (TGA9, TGA10 and SPL/NZZ) and their downstream tapetal development- and pollen wall development-associated genes (DYT1, MS1, ANAC025, MYB99) were analyzed in flowers of AGL13 + SRDX plants (Figure 6a). Notably, the expression levels of all the genes involved in this SPL/NZZ pathway were clearly down-regulated in AGL13 + SRDX flowers (Figure 6b–h). Furthermore, the level of decrease of these genes was correlated with an increase in AGL13 + SRDX expression (Figure 6a). These results clearly support the idea that AGL13 and AG could form a protein complex and regulate the same pathway by targeting similar downstream genes (Figure 7).

image

Figure 6. Analysis of the expression levels of the genes involved in Arabidopsis male gametophyte and tapetum development in AGL13 + SRDX plants.The mRNA accumulation for AGL13 + SRDX (a), anther cell layer-associated genes TGA9 (b), TGA10 (c) and SPL/NZZ (d), tapetal development-associated genes DYT1 (e), MS1 (f), the genes downstream of the MS1 gene ANAC025 (g) and MYB99 (h), AG (i), AGL6 (j) and endogenous AGL13 (k) was determined by real-time quantitative PCR. Total RNA was isolated from the flower buds before stage 12 of one wild-type Columbia plant (wt, line 1), of one AGL13 + SRDX plant (#318) with a severe mutant phenotype (in Figure 3i), of two AGL13 + SRDX plants (#206, 301) with mutant phenotypes (in Figure 3a) and of one AGL13 + SRDX plant (#203) with a wild-type-like phenotype. The transcript levels of these genes were determined using two to three replicates and were normalized using UBQ10. The expression of each gene was relative to that of the wild-type plant, which was set at 1. The error bars represent the standard deviation. Each experiment was repeated twice with similar results.

Download figure to PowerPoint

image

Figure 7. Model for the function of AGL13 in cooperation with AG, AP3 and PI to regulate pollen and ovule development.

In wild-type Arabidopsis anther development, AGL13 interacts with AG and the AP3–PI to form a heterotetrameric complex to regulate the unidentified genes (?) involved in early initiation and the further development of pollen. This AGL13–AG–AP3–PI complex is also involved in the regulation of male gametophyte and tapetum development by activating the anther cell layer-associated genes TGA9, TGA10 and SPL/NZZ, tapetal development-associated genes DYT1 and MS1 and the genes downstream of the MS1 gene, ANAC025 and MYB99. This AGL13–AG–AP3–PI complex could activate AG, AP3 and PI expression through a positive autoregulation mechanism. This AGL13–AG–AP3–PI complex may also need to interact with other unidentified proteins (X) to form a much higher-order complex to perform its function. In wild-type Arabidopsis pistil development, AGL13 interacts with AG to form a tetrameric complex to activate AGL6 expression or regulate unidentified genes (?) to control early initiation and the further development of the ovule. This (AGL13–AG)2 complex could also suppress the expression of AGL13 through negative regulatory feedback loops in which AGL6 acts as a repressor to suppress AGL13 expression. This (AGL13–AG)2 complex could also activate AG expression in pistil through a positive autoregulation mechanism. This (AGL13–AG)2 complex may also need to interact with other unidentified proteins (Y) to form a much higher-order complex to perform its function.

Download figure to PowerPoint

It has been reported that AG has positive autoregulation abilities (Gómez-Mena et al., 2005). Not surprisingly, the AG expression level was clearly down-regulated in AGL13 + SRDX plants (Figure 6i) and the level of decrease was correlated with an increase in AGL13 + SRDX expression (Figure 6a). The possible mechanism for this phenomenon is that AGL13 + SRDX could interact with the AG protein and then, through autoregulation, repress the transcription of AG (Figure 7). It has been reported that AGL6 functions as a repressor to control ovary and ovule development (Schauer et al., 2009; Koo et al., 2010). When the expression patterns for endogenous AGL13 and AGL6 were analyzed in AGL13 + SRDX plants, the AGL6 mRNA accumulation level decreased (Figure 6j) and the endogenous AGL13 expression level was up-regulated (Figure 6k). Our result indicates that AGL6 is possibly activated by AGL13 during ovule development and the AGL6 gene could form the feedback loop to repress the endogenous AGL13 expression level (Figure 7).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

AGL13 is expressed in male and female gametophytes

In this study, we detected the spatial and temporal expression pattern of AGL13 using GUS reporter fusions. GUS was specifically detected in the centre of the flower primordia in early stage 2–3 flowers and in the stamen primordia in stage 4 flowers. In stage 7–10 flowers, GUS was only detected in the microsporangium, archesporial cells, developing pollen and endothecium of the anther. GUS was undetectable in all anther tissues and mature pollen after stage 12. These results indicate that AGL13 is likely to be involved in the regulation of early initiation of the anther and the entire process of microsporogenesis. In addition to the anther, GUS was also specifically detected very early in the ovule primordia at stage 8–9 and until the maturation of the ovule, indicating that AGL13 also functions in the initiation and development of the ovule in Arabidopsis. It is worthwhile to note that the detection of GUS in the developing anther and ovule in 13P-1–GUS Arabidopsis was consistent with the in situ RNA hybridization pattern of AGL13 reported previously (Schauer et al., 2009).

AGL13 regulates pollen and ovule development

Not surprisingly, the production of defective pollen and ovules in 35S:AGL13 RNAi and AGL13 + SRDX dominant-negative mutants correlates well with the specific GUS expression observed during both pollen and ovule development in 13P-1–GUS flowers.

Notably the defect in pollen formation in 35S:AGL13 RNAi plants is incomplete: two types of defective pollen were found in approximately equal numbers. The first type is reduced in size with a flat or collapsed shape, and the second type resembles wild-type but has defects in the outer wall. The fact that 2 of the 4 pollen grains were severely defected in the 35S:AGL13 RNAi/qrt1-2 tetrad indicates that the development of the pollen that carried the 35S:AGL13 RNAi was arrested after meiosis and the tetrad formation. This result indicates that AGL13 plays a specific essential role in regulating pollen development. In addition, clear pollen surface defects, including breakage and aberrations in the exine patterning of the outer wall, were observed in most of the wild-type-like pollen and may have caused them to hardly stick to the stigma of pistil during the pollen adhesion test assay. This surface defect indicates that the process of the deposition from the tapetum to the surface of the pollen requires the expression of AGL13. This assumption was further supported by the down-regulation of the genes controlling tapetal development and pollen wall development (DYT1, MS1, ANAC025, MYB99) in AGL13 + SRDX plants (Figure 6).

Defective ovules were also identified in approximately half of the ovules produced by the 35S:AGL13 RNAi and AGL13 + SRDX ovaries. As only one haploid megaspore from the tetrad eventually develops into a functional megaspore during ovule formation (Webb and Gunning, 1990; Robinson-Beers et al., 1992), we believe that the generation of two types of ovules in the 35S:AGL13 RNAi ovary occurs through the random production of a functional haploid megaspore in the ovule, half of which contain by chance a copy of the 35S:AGL13 RNAi transgene. Severe abortion resulted in the ovules containing a functional megaspore with the AGL13 RNAi transgene. In contrast, normal development occurred in those ovules with a functional megaspore lacking the AGL13 RNAi transgene. The altered ovule formation in 35S:AGL13 RNAi and AGL13 + SRDX flowers indicates that AGL13 also plays a specific essential role in ovule development.

AGL13 is able to interact with AG to form quartet-like complexes and regulate the expression of similar genes involved in pollen morphogenesis

AGL13 expression was initially detected at floral stage 3 and in the regions similar to the expression of the AG gene (Bowman et al., 1991; Busch et al., 1999). The co-existence of their transcripts in the same place indicates the possibility of positive interaction between AGL13 and AG in controlling anther (pollen) and ovule development. This assumption was further supported by four lines of evidence. The first piece of evidence is that the AG nuclear localization efficiency was significantly enhanced by the presence of the AGL13 protein (Figure 4). This result revealed that AGL13 could interact with AG and enters into the nucleus. The second line of evidence comes from the direct interaction between AGL13 and AG to form quartet-like complexes found through FRET analysis (Figure 4). This finding also supports the result reported previously that AGL13 interacts with AG in yeast two-hybrid analysis (de Folter et al., 2005). The third piece of evidence is the similar downstream genes regulated by AGL13 and AG. In pollen morphogenesis (Figure 7), AG binds the upstream element and initiates the expression of the NZZ/SPL gene for sporogenesis (Ito et al., 2004) and DYT1, the gene downstream of NZZ/SPL that controls the initiation of tapetum morphogenesis (Zhang et al., 2006). MS1 is the downstream gene of DYT1, which is expressed in tapetum to regulate the synthesis of exine and to activate additional downstream genes, such as MYB99 and ANAC025 (Figure 7) (Ito et al., 2007; Yang et al., 2007). In addition, two leucine zipper-type transcription factors, TGA9 and TGA10, have been thought to form a heterodimer involved in regulating stamen morphogenesis (Murmu et al., 2010). TGA9 and TGA10 are upstream of DYT1 (Figure 7) with an expression pattern similar to AGL13. Interestingly, the expression levels of NZZ/SPL, DYT1, MS1, MYB99, ANAC025, TGA9 and TGA10 genes were all clearly down-regulated in AGL13 + SRDX dominant-negative mutants in which AGL13 was converted into a repressor (Figure 6). The fourth line of evidence is that AG expression was also down-regulated in AGL13 + SRDX plants. AG has been reported to have positive autoregulation abilities (Gómez-Mena et al., 2005). The binding of AGL13 + SRDX to AG proteins causes the suppression of the autoregulation of AG and the transcription of their downstream genes. This evidence clearly supports the fact that AGL13 and AG could form a protein complex and regulate the same pathway by targeting similar downstream genes.

AGL13 has negative autoregulation abilities by activating AGL6

It is interesting to note that AGL6 has an overlapping expression pattern with AGL13 in the developing ovule (Schauer et al., 2009). How these two AGL6-like genes interact to regulate ovule development is interesting. We found that the expression level of AGL6 clearly decreased in AGL13 + SRDX dominant-negative mutants. This result indicates that AGL13 acts upstream of AGL6 and the endogenous AGL13 could activate the expression of AGL6 in wild-type plants. Further analysis indicates that, with the decrease in AGL6 expression, the expression of endogenous AGL13 was up-regulated. Thus, a possible hypothesis is revealed: AGL13 is the major action officer of AGL6 lineage genes in ovule development and AGL6 plays the role of controller in a feedback loop to restrict the endogenous AGL13 expression (Figure 7).

AGL13 regulates pollen and ovule development through forming higher-order MADS-domain protein complexes with B and C functional MADS box proteins

In this study, a direct interaction between AGL13 and AG to form quartet-like complexes has been demonstrated by FRET analysis (Figure 4). This result may explain the action of AGL13 in regulating anther (pollen) and ovule development through interaction with AG. However, stamen morphogenesis and anther/pollen development are also regulated by the B functional proteins AP3/PI. To explore the possible relationship between AGL13–AP3/PI, three proteins, PI/AP3/AGL13, were transiently co-expressed and the FRET efficiency of both ‘AP3–AGL13’ and ‘PI–AGL13’ was calculated to be 50%. This result indicates that AGL13 has the ability to interact with the PI–AP3 heterodimer. With the co-expression of four proteins, PI/AP3/AG/AGL13, the FRET efficiency of AP3–AGL13 remains at 50%; the FRET efficiency of PI–AGL13, however, decreased from 50 to 30%. This result indicates that the distance between AGL13 and PI was lengthened by adding the AG protein. Thus, AGL13 could form heteromeric higher-order MADS-domain protein complexes with AG and AP3–PI heterodimers (Figure 5). The possible relative position of AGL13 in this complex is next to AG and AP3, but diagonal to PI (Figure 5c). Based on our results, quartet-like complexes consist of two groups of AG–AGL13 heterodimers (AG–AGL13)2 most likely involved in the regulation of ovule development, whereas AG–AGL13–AP3/PI heteromeric complexes may participate in anther and pollen development (Figure 7).

AGL13, a putative E functional gene, specifies male and female gametophyte morphogenesis: from evolution to gene function

Different from that in angiosperm, E functional SEP genes were not discovered in gymnosperms (Winter et al., 1999). It is reasonable to believe that SEP genes may be the redundant counterparts of the AGL6 lineage genes in angiosperms and were generated by duplication from the AGL6 lineage genes during evolution. DAL1, the AGL6 lineage gene in Norway Spruce, is expressed in both male and female reproductive organs (Carlsbecker et al., 2004). This result reveals that the function of AGL6 lineage genes in regulating male and female gametophytes in plants is primitive and conserved.

The identity of the Arabidopsis AGL6 lineage gene AGL13 has remained elusive. In our result, the expression patterns and the function of AGL13 in regulating male and female gametophyte in Arabidopsis were very similar to DAL1 in Norway Spruce. This similarity suggests a possible ancestral nature of AGL13 for the AGL6 lineage and E functional genes. This assumption was further supported by three lines of evidence. The first piece of evidence is the similar mutant flower phenotype generated by AGL13 + SRDX and SEP2 + SRDX dominant negative mutants. This result revealed that AGL13 and SEP2 could target similar downstream genes once expressed in the same places. This targeting is evidenced by the similar down-regulation of the AP1/AP2/AP3/PI/AG genes in both AGL13 + SRDX and SEP2 + SRDX plants (Figure 3). The second line of evidence comes from the finding that the AG nuclear localization efficiency is significantly enhanced similarly by the presence of AGL13 and SEP3 proteins (Figure 4). This result reveals that AGL13 could interact with AG similar to SEP3. The third piece of evidence is the ability for AGL13 to form heteromeric higher-order MADS-domain protein complexes with AG and AP3–PI heterodimers (Figure 5). Recently, it has been reported that SEP3 could also interact with AG and AP3–PI to form similar heteromeric higher-order MADS-domain protein complexes through EMSA analysis (Smaczniak et al., 2012). Interestingly, similar to AGL13, this result also suggests that SEP3 is next to AG and AP3 and diagonal to PI in this complex (Smaczniak et al., 2012). This result strongly supports the idea that AGL13 has a similar E functional biochemical nature as SEP3 to occupy the same position in an AGL13(SEP3)–AG–AP3/PI heteromeric protein complex.

In summary, we characterized a putative AGL6-like MADS box gene AGL13 in Arabidopsis. We found that AGL13 functions as an E functional gene by interacting with AG/AP3/PI in specifically regulating the development of the pollen and ovules. The characterization of AGL13 that was performed in this study provides useful information for understanding the function of AGL6-like genes in controlling reproductive organ development and further clarifies their role as the ancestor for the E functional genes in plants during evolution.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant materials and the construction of various constructs

The plants materials and growth conditions, the plant transformation and transgenic plant analysis, the detailed description of the construction of AGL13:GUS, 35S:AGL13 RNA interference, 35S:AGL13 + SRDX, 35S:SEP2 + SRDX constructs, the real-time PCR analysis, sequences for the primers (Table S1), histochemical GUS assay, light microscopy, semi-thin sections, scanning electron microscopy (SEM) and confocal laser scanning microscopy (CLSM), Alexander's staining, pollen adhesion assay, construction of FRET-associated fusion constructs and the FRET assay were provided in the Methods S1–S9.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was supported by grants to C-H Y from the National Science Council, Taiwan, ROC, grant numbers: NSC96–2752-B-005–007-PAE and NSC 98–2321-B-005–007-MY3. This work was also supported in part by the Ministry of Education, Taiwan, ROC under the ATU plan. We thank Dr Gerco Angenent (Business Unit Bioscience, Plant Research International, 6708PB Wageningen, The Netherlands) for his helpful discussion of the results in this work.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Alexander, M.P. (1969) Differential staining of aborted and nonaborted pollen. Stain Technol. 44, 117122.
  • Bowman, J.L., Drews, G.N. and Meyerowitz, E.M. (1991) Expression of the Arabidopsis floral homeotic gene AGAMOUS is restricted to specific cell types late in flower development. Plant Cell, 3, 749758.
  • Busch, M.A., Bomblies, K. and Weigel, D. (1999) Activation of a floral homeotic gene in Arabidopsis. Science, 285, 585587.
  • Carlsbecker, A., Tandre, K., Johanson, U., Englund, M. and Engström, P. (2004) The MADS box gene DAL1 is a potential mediator of the juvenile-to-adult transition in Norway spruce (Picea abies). Plant J. 40, 546557.
  • Chang, Y.Y., Chiu, Y.F., Wu, J.W. and Yang, C.H. (2009) Four orchid (Oncidium Gower Ramsey) AP1/AGL9-like MADS box genes show novel expression patterns and cause different effects on floral transition and formation in Arabidopsis thaliana. Plant Cell Physiol. 50, 14251438.
  • Chang, Y.Y., Kao, N.H., Li, J.Y., Hsu, W.H., Liang, Y.L., Wu, J.W. and Yang, C.H. (2010) Characterization of the possible roles for B class MADS box genes in regulation of perianth formation in orchid (Oncidium Gower Ramsey). Plant Physiol. 152, 837853.
  • Chen, M.K., Hsu, W.H., Lee, P.F., Thiruvengadam, M., Chen, H.I. and Yang, C.H. (2011) The MADS box gene, FOREVER YOUNG FLOWER, acts as a repressor controlling floral organ senescence and abscission in Arabidopsis. Plant J. 68, 168185.
  • Chou, M.L., Haung, M.D. and Yang, C.H. (2001) EMF genes interact with late-flowering genes in regulating floral initiation genes during shoot development in Arabidopsis thaliana. Plant Cell Physiol. 42, 499507.
  • Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735743.
  • Coen, E.S. and Meyerowitz, E.M. (1991) The war of the whorls: genetic interactions controlling flower development. Nature, 353, 3137.
  • Eklund, D.M., Staldal, V., Valsecchi, I. et al. (2010) The Arabidopsis thaliana STYLISH1 protein acts as a transcriptional activator regulating auxin biosynthesis. Plant Cell, 22, 349363.
  • Favaro, R., Pinyopich, A., Battaglia, R., Kooiker, M., Borghi, L., Ditta, G., Yanofsky, M.F., Kater, M.M. and Colombo, L. (2003) MADS box protein complexes control carpel and ovule development in Arabidopsis. Plant Cell, 15, 26032611.
  • de Folter, S., Immink, R.G., Kieffer, M. et al. (2005) Comprehensive interaction map of the Arabidopsis MADS Box transcription factors. Plant Cell, 17, 14241433.
  • Fujiki, Y., Yoshimoto, K. and Ohsumi, Y. (2007) An Arabidopsis homolog of yeast ATG6/VPS30 is essential for pollen germination. Plant Physiol. 143, 11321139.
  • Gómez-Mena, C., de Folter, S., Costa, M.M., Angenent, G.C. and Sablowski, R. (2005) Transcriptional program controlled by the floral homeotic gene AGAMOUS during early organogenesis. Development, 132, 429438.
  • Guo, Z., Fujioka, S., Blancaflor, E.B., Miao, S., Gou, X. and Li, J. (2010) TCP1 modulates brassinosteroid biosynthesis by regulating the expression of the key biosynthetic gene DWARF4 in Arabidopsis thaliana. Plant Cell, 2, 11611173.
  • Hiratsu, K., Matsui, K., Koyama, T. and Ohme-Takagi, M. (2003) Dominant repression of target genes by chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant J. 34, 733739.
  • Honma, T. and Goto, K. (2001) Complexes of MADS box proteins are sufficient to convert leaves into floral organs. Nature, 409, 525529.
  • Hsu, H.F. and Yang, C.H. (2002) An orchid (Oncidium Gower Ramsey) AP3-like MADS gene regulates floral formation and initiation. Plant Cell Physiol. 43, 11981209.
  • Hsu, H.F., Huang, C.H., Chou, L.T. and Yang, C.H. (2003) Ectopic expression of an orchid (Oncidium Gower Ramsey) AGL6-like gene promotes flowering by activating flowering time genes in Arabidopsis thaliana. Plant Cell Physiol. 44, 783794.
  • Immink, R.G., Tonaco, I.A., de Folter, S., Shchennikova, A., van Dijk, A.D., Busscher-Lange, J., Borst, J.W. and Angenent, G.C. (2009) SEPALLATA3: the ‘glue’ for MADS box transcription factor complex formation. Genome Biol. 10, R24.
  • Ishiguro, S., Nishimori, Y., Yamada, M., Saito, H., Suzuki, T., Nakagawa, T., Miyake, H., Okada, K. and Nakamura, K. (2010) The Arabidopsis FLAKY POLLEN1 gene encodes a 3-hydroxy-3-methylglutaryl-coenzyme A synthase required for development of tapetum-specific organelles and fertility of pollen grains. Plant Cell Physiol. 51, 896911.
  • Ito, T., Wellmer, F., Yu, H., Das, P., Ito, N., Alves-Ferreira, M., Riechmann, J.L. and Meyerowitz, E.M. (2004) The homeotic protein AGAMOUS controls microsporogenesis by regulation of SPOROCYTELESS. Nature, 15, 356360.
  • Ito, T., Nagata, N., Yoshiba, Y., Ohme-Takagi, M., Ma, H. and Shinozaki, K. (2007) Arabidopsis MALE STERILITY1 encodes a PHD-type transcription factor and regulates pollen and tapetum development. Plant Cell, 19, 35493562.
  • Jefferson, R.A., Kavanagh, T.A. and Bevan, M.W. (1987) GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6, 39013907.
  • Kaufmann, K., Melzer, R. and Theissen, G. (2005) MIKC-type MADS-domain proteins: structural modularity, protein interactions and network evolution in land plants. Gene, 347, 183198.
  • Koo, S.C., Bracko, O., Park, M.S. et al. (2010) Control of lateral organ development and flowering time by the Arabidopsis thaliana MADS box Gene AGAMOUS-LIKE6. Plant J. 62, 807816.
  • Koyama, T., Sato, F. and Ohme-Takagi, M. (2012) CRES-T for the functional analysis of transcription factors and modification of morphological traits in plants. Curr. Biotechnol. 1, 2332.
  • Li, H.F., Liang, W.Q., Jia, R.D., Yin, C.S., Zong, J., Kong, H.Z. and Zhang, D.B. (2010) The AGL6-like gene OsMADS6 regulates floral organ and meristem identities in rice. Cell Res. 20, 299313.
  • Li, H.F., Liang, W.Q., Hu, Y., Zhu, L., Yin, C.S., Xu, J., Dreni, L., Kater, M. and Zhang, D.B. (2011) Rice MADS6 interacts with the floral homeotic genes SUPERWOMAN1, MADS3, MADS58, MADS13, and DROOPING LEAF in Specifying floral organ identities and meristem fate. Plant Cell, 23, 25362552.
  • Liu, J., Zhang, Y., Qin, G. et al. (2008) Targeted degradation of the cyclin-dependent kinase inhibitor ICK4/KRP6 by RING-type E3 ligases is essential for mitotic cell cycle progression during Arabidopsis gametogenesis. Plant Cell, 20, 15381554.
  • Ma, H., Yanofsky, M.F. and Meyerowitz, E.M. (1991) AGL1-AGL6, an Arabidopsis gene family with similarity to floral homeotic and transcription factor genes. Genes Dev. 5, 484495.
  • McCormick, S. (2004) Control of male gametophyte development. Plant Cell 16(suppl.), S142S153.
  • Melzer, R. and Theissen, G. (2009) Reconstitution of ‘floral quartets’ in vitro involving class B and class E floral homeotic proteins. Nucleic Acids Res. 37, 27232736.
  • Melzer, R., Verelst, W. and Theissen, G. (2009) The class E floral homeotic protein SEPALLATA3 is sufficient to loop DNA in ‘floral quartet’-like complexes in vitro. Nucleic Acids Res. 37, 144157.
  • Melzer, R., Wang, Y.Q. and Theissen, G. (2010) The naked and the dead: the ABCs of gymnosperm reproduction and the origin of the angiosperm flower. Semin. Cell Dev. Biol. 21, 118128.
  • Mena, M., Mandel, M.A., Lerner, D.R., Yanofsky, M.F. and Schmidt, R.J. (1995) A characterization of the MADS box gene family in maize. Plant J. 8, 845854.
  • Mouradov, A., Glassick, T.V., Hamdorf, B.A., Murphy, L.C., Marla, S.S., Yang, Y. and Teasdale, R.D. (1998) Family of MADS Box genes expressed early in male and female reproductive structures of monterey pine. Plant Physiol. 117, 5562.
  • Murmu, J., Bush, M.J., DeLong, C., Li, S., Xu, M., Khan, M., Malcolmson, C., Fobert, P.R., Zachgo, S. and Hepworth, S.R. (2010) Arabidopsis basic leucine-zipper transcription factors TGA9 and TGA10 interact with floral glutaredoxins ROXY1 and ROXY2 and are redundantly required for anther development. Plant Physiol. 154, 14921504.
  • Pařenicová, L., de Folter, S., Kieffer, M. et al. (2003) Molecular and phylogenetic analyses of the complete MADS box transcription factor family in Arabidopsis: new openings to the MADS world. Plant Cell, 15, 15381551.
  • Peng, Y.J., Shih, C.F., Yang, J.Y., Tan, C.M., Hsu, W.H., Huang, Y.P., Liao, P.C. and Yang, C.H. (2013) A RING-Type E3 ligase controls anther dehiscence by activating the jasmonate biosynthetic pathway gene DEFECTIVE IN ANTHER DEHISCENCE1 in Arabidopsis. Plant J. 74, 310327.
  • Preuss, D., Rhee, S.Y. and Davis, R.W. (1994) Tetrad analysis possible in Arabidopsis with mutation of the QUARTET (QRT) genes. Science, 264, 14581460.
  • Reinheimer, R. and Kellogg, E.A. (2009) Evolution of AGL6-like MADS box genes in grasses (Poaceae): ovule expression is ancient and palea expression is new. Plant Cell, 21, 25912605.
  • Rijpkema, A.S., Zethof, J., Gerats, T. and Vandenbussche, M. (2009) The petunia AGL6 gene has a SEPALLATA-like function in floral patterning. Plant J. 60, 19.
  • Robinson-Beers, K., Pruitt, R.E. and Gasser, C.S. (1992) Ovule development in wild-type Arabidopsis and two female-sterile mutants. Plant Cell, 4, 12371249.
  • Rounsley, S.D., Ditta, G.S. and Yanofsky, M.F. (1995) Diverse roles for MADS box genes in Arabidopsis development. Plant Cell, 7, 12591269.
  • Schauer, S.E., Schlűter, P.M., Baskar, R., Gheyselinck, J., Bolaños, A., Curtis, M.D. and Grossniklaus, U. (2009) Intronic regulatory elements determine the divergent expression patterns of AGAMOUS-LIKE6 subfamily members in Arabidopsis. Plant J. 59, 9871000.
  • Shindo, S., Ito, M., Ueda, K., Kato, M. and Hasebe, M. (1999) Characterization of MADS genes in the gymnosperm Gnetum parvifolium and its implication on the evolution of reproductive organs in seed plants. Evol. Dev. 1, 180190.
  • Smaczniak, C., Immink, R.G.H., Muiño, J.M. et al. (2012) Characterization of MADS-domain transcription factor complexes in Arabidopsis flower development. Proc. Natl Acad. Sci. USA, 109, 15601565.
  • Tandre, K., Albert, V.A., Sundas, A. and Engstrom, P. (1995) Conifer homologues to genes that control floral development in angiosperms. Plant Mol. Biol. 27, 6978.
  • Tejedor-Cano, J., Prieto-Dapena, P., Almoguera, C., Carranco, R., Hiratsu, K., Ohme-Takagi, M. and Jordano, J. (2010) Loss of function of the HSFA9 seed longevity program. Plant, Cell Environ. 33, 11271135.
  • Theissen, G. (2001) Development of floral organ identity: stories from the MADS house. Curr. Opin. Plant Biol. 4, 7585.
  • Theissen, G. and Saedler, H. (2001) Plant biology. Floral quartets. Nature, 409, 469471.
  • Tzeng, T.Y. and Yang, C.H. (2001) A MADS box gene from lily (Lilium longiflorum) is sufficient to generate dominant-negative mutation by interacting with PISTILLATA (PI) in Arabidopsis thaliana. Plant Cell Physiol. 42, 11561168.
  • Webb, M.C. and Gunning, B.E.S. (1990) Embryo sac development in Arabidopsis thaliana. I. Megasporogenesis, including the microtubular cytoskeleton. Sex. Plant Reprod. 3, 244256.
  • Winter, K.U., Becker, A., Münster, T., Kim, J.T., Saedler, H. and Theissen, G. (1999) MADS box genes reveal that gnetophytes are more closely related to conifers than to flowering plants. Proc. Natl Acad. Sci. USA, 96, 73427347.
  • Yang, C., Vizcay-Barrena, G., Conner, K. and Wilson, Z.A. (2007) MALE STERILITY1 is required for tapetal development and pollen wall biosynthesis. Plant Cell, 19, 35303548.
  • Zhang, W., Sun, Y., Timofejeva, L., Chen, C., Grossniklaus, U. and Ma, H. (2006) Regulation of Arabidopsis tapetum development and function by DYSFUNCTIONAL TAPETUM1 (DYT1) encoding a putative bHLH transcription factor. Development, 133, 30853095.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
tpj12363-sup-0001-TableS1-MethodsS1-S9.docWord document117K

Table S1. List of primers used in real-time quantitative RT-PCR in this study.

Methods S1. Plant materials and plant transformation.

Methods S2. Construction of chimeric GUS constructs.

Methods S3. Construction of the 35S:AGL13 RNA interference construct.

Methods S4. Construction of the AGL13 + SRDX and SEP2 + SRDX constructs.

Methods S5. RT-PCR and real-time PCR analysis.

Methods S6. Microscopy and semi-thin sections.

Methods S7. GUS staining, pollen staining and pollen adhesion assay.

Methods S8. Construction of FRET-associated fusion constructs.

Methods S9. The confocal laser scanning microscope imaging and FRET assay.

tpj12363-sup-0002-FigureS1-S7.pdfapplication/PDF6341K

Figure S1. AGL13-GUS reporter gene fusions.

Figure S2. GUS staining patterns in transgenic Arabidopsis plants.

Figure S3. GUS staining patterns in the anther of 13P-1-GUS flower.

Figure S4. GUS staining patterns in carpel of 13P-1-GUS flower.

Figure S5. Confocal laser scanning microscopy (CLSM) and scanning electron micrographs (SEM) of pollen produced in wild-type and 35S:AGL13 RNAi flowers.

Figure S6. Pollen adhesion assay, semi-thin sections of pollen and analysis of ovule formation for wild-type and 35S:AGL13 RNAi plants.

Figure S7. Confocal laser scanning microscopy (CLSM) of ovules in wild-type and 35S:AGL13 RNAi flowers.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.