Molecular Plant Advance Access published online on March 24, 2009
Molecular Plant, doi:10.1093/mp/ssp004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Temporal and Spatial Requirement of EMF1 Activity for Arabidopsis Vegetative and Reproductive Development
a Department of Plant and Microbial Biology, University of California, Berkeley, CA 94720, USA
b Present address: Heidelberg Institute for Plant Science, Biodiversity and Plant Systematics, Heidelberg University, 69120 Heidelberg, Germany
c Present address: Department of Molecular Biology, Pusan National University, 30 Jangjeon-dong, Geumjeong-gu, Busan 609-735, Korea
1 To whom correspondence should be addressed. E-mail zrsung{at}nature.berkeley.edu, fax (510) 642-4995, tel. (510) 642-6966.
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
EMBRYONIC FLOWER (EMF) genes are required to maintain vegetative development via repression of flower homeotic genes in Arabidopsis. Removal of EMF gene function caused plants to flower upon germination, producing abnormal and sterile flowers. The pleiotropic effect of emf1 mutation suggests its requirement for gene programs involved in diverse developmental processes. Transgenic plants harboring EMF1 promoter::glucuronidase (GUS) reporter gene were generated to investigate the temporal and spatial expression pattern of EMF1. These plants displayed differential GUS activity in vegetative and flower tissues, consistent with the role of EMF1 in regulating multiple gene programs. EMF1::GUS expression pattern in emf mutants suggests organ-specific auto-regulation. Sense- and antisense (as) EMF1 cDNA were expressed under the control of stage- and tissue-specific promoters in transgenic plants. Characterization of these transgenic plants showed that EMF1 activity is required in meristematic as well as differentiating tissues to rescue emf mutant phenotype. Temporal removal or reduction of EMF1 activity in the embryo or shoot apex of wild-type seedlings was sufficient to cause early flowering and terminal flower formation in adult plants. Such reproductive cell memory is reflected in the flower MADS-box gene activity expressed prior to flowering in these early flowering plants. However, temporal removal of EMF1 activity in flower meristem did not affect flower development. Our results are consistent with EMF1’s primary role in repressing flowering in order to allow for vegetative growth.
Key Words: EMF1 • stage-specific promoter • early flowering • reproductive cell memory • vegetative/reproductive development • repression of flower MADS-box genes
Received for publication October 23, 2008. Accepted for publication January 5, 2009.
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Plants develop in phases—embryonic, vegetative, reproductive, and flowering, specified by activation and inactivation of specific sets of genes. How the gene program specific for each phase is established and maintained is of central interest to developmental biology. Studies over the last decade have demonstrated that the epigenetic gene regulation plays a crucial role in maintaining stable meristematic and differentiation states, and reprogramming cell fates and memory during phase changes. However, how the different epigenetic mechanisms are integrated and coordinated in each developmental phase is so far unknown.
Mutations in the EMBRYONIC FLOWER (EMF) genes EMF1 and EMF2 cause plants to skip vegetative phase and to transform the embryonic meristem from indeterminate to determinate, producing a terminal flower. emf mutant flowers are abnormal, usually devoid of petals, which can be rescued by the removal of agamous (ag) gene activity (Sung et al., 1992; Yang et al., 1995).
EMF2 encodes a Polycomb group (PcG) protein (Yoshida et al., 2001). PcG proteins are a set of chromatin-modifying proteins that regulate important cellular and developmental processes from mammals to plants (Ringrose and Paro, 2004; Martinez and Cavalli, 2006; Valk-Lingbeek et al., 2004; Pien and Grossniklaus, 2007; Calonje and Sung, 2006; Kohler and Makarevich, 2006; Schmitz and Amasino, 2007). In general, PcG proteins are responsible for maintaining the repressed expression state of their target genes, acting as a part of a cellular memory mechanism that enables cells to remember their chosen fate over many cell divisions (Ringrose and Paro, 2004). Recent efforts have revealed hundreds of potential PcG targets in mammals, insects, and plants. PcG proteins are therefore master regulators of so-called genomic programs, repressing all alternative gene programs that are not needed in a specific differentiation status (Schwartz and Pirrotta, 2007). EMF2 participates in the transcriptional repression of flower homeotic genes via histone 3 lysine 27 trimethylation (Chanvivattana et al., 2004; Schubert et al., 2006) as a component of the putative EMF2 complex that comprises four core proteins—EMF2 (Yoshida et al., 2001), CURLY LEAF (CLF) (Goodrich et al., 1997), FERTILIZATION INDEPENDENT ENDOSPERM (FIE) (Kinoshita et al., 2001), and MULTICOPY SUPPRESSOR OF IRA 1 (MSI1) (Hennig et al., 2003).
EMF1 encodes a transcriptional regulator (Aubert et al., 2001; Calonje et al., 2008). Although EMF1 shares no sequence homology with animal PcG protein genes, it is involved in plant PcG-mediated gene repression mechanism, targeting the flower homeotic genes directly (Calonje et al., 2008).
Both emf1 and emf2 mutants display altered expression pattern of genes required for flower development (Moon et al., 2003), indicating their key role as regulators of this program. Other gene programs affected by the emf mutations include photosynthesis, hormone regulation, and seed maturation (Moon et al., 2003). Some gene programs are more severely affected in the emf1 than emf2 mutants, consistent with EMF2 gene family redundancy (Schubert et al., 2005). Most emf1 alleles display stronger phenotypes than emf2. The extreme and pleiotropic phenotype, in which all lateral organs including cotyledons are converted into carpelloid structures (Chen et al., 1997), may contribute to its greater impact on gene expression pattern.
EMF1 RNA and EMF1 protein are constitutively expressed in Arabidopsis (Aubert et al., 2001; Calonje et al., 2008). PcG genes (EMF2, CLF, MSI1, and FIE) implicated in the repression of the flower program are also expressed constitutively, suggesting that EMF1 together with the EMF2 complex are required throughout development to maintain the epigenetic control of gene expression (Kinoshita et al., 2001; Hennig et al., 2003; Katz et al., 2004). However, it is not clear whether EMF1 is indeed required for all stages of development and whether the differentiation of all tissues and organs depends on EMF1 to maintain the epigenetic control of gene expression. To address this question, we employed stage- and tissue-specific promoters to regulate endogenous EMF1 gene activity. To investigate the spatial expression pattern of EMF1, we employed EMF1 promoter::reporter construct to study EMF1 promoter activity in transgenic plants.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Temporal and Spatial Expression of EMF1
To investigate EMF1 promoter activity during Arabidopsis development, we expressed a reporter gene, beta-glucuronidase (GUS), under the control of EMF1 promoter. Four putative EMF1 promoter::GUS constructs were generated (see Methods) to identify a reasonably short region of promoter sequence containing necessary regulatory elements, and these transgenes were introduced into Arabidopsis. The expression patterns of these four constructs were similar (data not shown). The GUS activity of EMF1::GUS construct consisting of 3.2 Kb EMF1 promoter alone fused to beta-glucuronidase gene was analyzed during Arabidopsis development (Figure 1). The activity was found in mature embryo (Figure 1A), vegetative shoot, and flower. In 7-day-old seedlings, the activity was visible in cotyledon blade (Figure 1B), and shoot apex (Figure 1C). At 14 d, the activity was found in the rosette leaf blade; roots were GUS-negative except for root tips (Figure 1D). After flowering, the stigma and anthers showed a strong GUS activity (Figure 1E). Elongated cells such as hypocotyls and petioles were devoid of GUS activities. These results show that EMF1 promoter is active in most developmental stages, but not in all cell and tissue types. The EMF1::GUS construct was also introduced into emf1-2 mutants. At 9 d after germination, only roots displayed GUS activity (Figure 1F, left). 18 d after germination, cotyledons, which have developed stigmatic papilla (Chen et al., 1997), became GUS-positive (Figure 1F, right).
|
Stage-Specific Activity of EMF1
EMF1::GUS expression pattern shows a spatial and temporal regulated promoter activity. The strong EMF1::GUS activity in shoot apex suggests that rosette fate is determined in shoot apex where EMF1 is expressed to repress flowering; therefore, its expression in shoot apex may be sufficient for rosette development. To test this hypothesis, we expressed EMF1 under the control of a shoot meristem specific promoter, the KNOTTED1-LIKE (KNAT1, Lincoln et al., 1994), and a leaf primordia-specific promoter, LEAFY (LFY, Nilsson et al., 1998b; Blázquez et al., 1997). Similarly, the strong EMF1::GUS activity in mature embryo suggests that the EMF1 protein is needed in the embryo to allow normal seedling development. To test this hypothesis, we generated transgenic plants expressing sense EMF1 cDNA under the control of the promoter of a seed-specific gene, Arabidopsis thaliana 2S ALBUMIN 3 gene (At2S3, Guerche et al., 1990, At4G27160). These three constructs were introduced into plants harboring emf1 mutations and selected by hygromycin resistance in T1s, which were selfed to obtain homozygous emf1 mutants. Based on the rescue of emf1 mutant phenotype by a genomic EMF1 clone described previously by Aubert et al. (2001), we test the ability of each construct generated with sense EMF1 cDNA to rescue emf1 mutant phenotype. It was determined based on the segregation ratio of hygromycin resistance (HygR, Table 1) and emf mutant phenotype on 14-day-old transgenic plants grown on agar medium.
|
KNAT1 promoter is active in the shoot apical meristem (SAM), stronger in peripheral and rib zone than in the central zone. The genetic segregation ratio of KNAT1::EMF1 emf1-2 T2 transgenic plants (Table 1) is close to 9:3:3:1, except for a lower frequency of the HygR emf mutant category, indicating that the expression of EMF1 in shoot apical meristem is insufficient for rescuing the emf mutant phenotype. However, HygR emf1-2 mutants have larger and greener cotyledons than emf1-2, but they are still not able to develop petioles (Figure 1H and 1I). It cannot be considered a partial complementation of emf1 phenotype because these HygR emf1-2 mutants developed terminal flower structures and are indistinguishable from the parental emf mutants.
LFY is highly expressed in flower meristem and flower primordia (Blázquez et al., 1997); however, lower LFY expression can be detected in leaf primordia (Nilsson et al., 1998b) at germination and in embryo (Nilsson et al., 1998a). The genetic segregation ratio of LFY::EMF1 emf1-1 T2 transgenic plants (Table 1) is also close to 9:3:3:1, indicating that the expression in leaf primordia could not rescue emf1 mutant phenotype, and EMF1 in leaf primordia alone is insufficient for rosette growth.
At2S3 encodes a seed storage protein gene, expressed in embryo (Guerche et al., 1990). We expressed GUS under the control of At2S3 promoter in WT Arabidopsis to study its expression after germination. At2S3::GUS activity is found in cotyledons of germinating seedlings, but absent in shoot apex and rosette leaves (Supplemental Figure 1A). The stable glucuronidase enzyme might be responsible for the GUS activity in cotyledons 14 d after germination (DAG). To investigate the promoter activity further, we measured RNA level by RT–PCR and found At2S3 RNA decreased upon germination and no RNA was found in seedlings 5 DAG (Supplemental Figure 1B). The genetic segregation ratio of At2S3::EMF1 emf1-2 T2 transgenic plants (Table 1) is close to 9:3:3:1, indicating that the EMF1 expression in embryo and even in cotyledons cannot rescue rosette development after germination, indicating continuing expression of EMF1 after germination is required to specify rosette development.
In summary, localized expression of EMF1 in just a subset, embryo, or shoot apex of its normal expression domain is not sufficient to rescue emf1 mutant phenotypes, indicating a broader spectrum, temporal and spatial, of EMF1 expression pattern is required to prevent flowering and specify vegetative development.
Stage-Specific Requirement of EMF1
To investigate the requirement of EMF1 during Arabidopsis development, we introduced constructs harboring antisense EMF1 cDNA (asEMF1) under the control of stage- and tissue-specific promoters into WT Arabidopsis.
EMF1 in the SAM Is Required for Vegetative Development
Since expression of EMF1 in shoot apex, either shoot apical meristem (SAM) or leaf primordia, cannot rescue emf1 mutant phenotype, we examined if EMF1 expression in these two tissues is essential for vegetative development. antisense EMF1 (asEMF1) cDNA was expressed under the control of KNAT1 promoter to remove or reduce EMF1 activity in the SAM and LFY promoter to remove EMF1 activity in leaf primordia.
KNAT1::asEMF1 transgenic plants displayed emf phenotype, with short hypocotyls, petiole-less cotyledons that skipped vegetative development and flowered upon germination (Figure 2A). The level of endogenous EMF1 RNA was reduced in these plants (Figure 2B). Failure of rosette development in these transgenic plants that expresses asEMF1 in the SAM shows that EMF1 activity in the SAM is essential to maintain vegetative development.
|
EMF1 in the Leaf Primordia Is Required for Normal Leaf Development and Flowering Time
LFY::asEMF1 transgenic lines (stated as L lines in Figures) germinated like WT; the seedlings show long hypocotyls and cotyledons were round and with petioles. Seedlings produced true leaves, but these leaves were curly, and the plants flowered early and produced terminal flowers (Figure 2C and 2D). LFY promoter is active in leaf primordia at 4 DAG (Blázquez et al., 1997) and in embryo (Nilsson et al., 1998a), but not in the shoot apical meristem. Thus, removing or reducing EMF1 activity from leaf primordia could not prevent rosette leaf growth, indicating that EMF1 in the leaf primordia are not essential for true leaf development; however, it is necessary for completing normal leaf differentiation (see below).
Flowering time was measured as number of rosette leaves produced prior to bolting under long-day (LD) or short-day (SD) conditions (Figure 2E). Fifteen-day-old plants grown on agar plates under SD were transferred to soil and grown under SD or LD conditions. All LFY::asEMF1 lines showed early flowering phenotype in both conditions, with line L#28 displaying the most severe phenotype of curly leaves and earliest flowering time. The phenotype resembled that of the curly leaf (clf) mutant (Goodrich et al., 1997). This result shows that reducing EMF1 activity in leaf primordia affected leaf growth and differentiation, resulting in curly leaf. Although reduced EMF1 activity did not change shoot apical fate immediately, it seems to have a cumulative effect on shoot apical meristem, causing it to change to a floral fate, resulting in early flowering and terminal flower development. That LFY::asEMF1 transgenic plants behave as if it is defective in a member of the PRC2-like complex, CLF, is consistent with EMF1 acting in the PcG-mediated gene silencing mechanism along with PRC2 (Calonje et al., 2008).
Reduced EMF1 Expression in Flowers Does Not Disrupt Flower Development
To study the role of EMF1 activity in flower development, we studied transgenic plants expressing antisense EMF1 under the control of two flower meristem identity genes, namely LFY and APETALA1 (AP1). Both promoters are highly active in flower meristems (Blázquez et al., 1997; Hempel et al., 1997).
LFY::asEMF1 transgenic lines (stated as L lines in Figures) showed abnormal flower development, mainly affecting sepal shape and presence of petals (Figure 2F). This result suggests that EMF1 activity is required not only for vegetative, but also for normal flower development, maybe through EMF1 fine-tuning of the expression of the flower homeotic genes. To investigate this hypothesis further, we expressed antisense EMF1 under the control of APETALA1 (AP1) promoter, which, unlike LFY promoter, is not active before flowering (Hempel et al., 1997).
AP1::asEMF1 transgenic plants (stated as A lines in Figures) did not show any alteration in flowering time (Figure 2E, lower panel) as expected, as AP1 promoter is inactive prior to flower meristem formation. Unlike emf1 mutants and LFY::asEMF1 transgenic lines, AP1::asEMF1 lines had normal flower organs (data not shown). For example, emf1 mutants lack petals, have short filament and large pistil; however, AP1::asEMF1 transgenic plants did not exhibit these defects. Thus, removing EMF1 after flowering did not affect flower organ development as emf1 mutants and LFY::asEMF1 lines did. Since AP1::asEMF1 construct should reduce EMF1 mRNA levels in developing floral organs, normal flower development in these lines may be explained by the pre-existing EMF1 protein, or that EMF1 is not required for flower development. These results suggest that reducing EMF1 activity prior to flowering, such as in emf1 mutants and LFY::asEMF1 plants, has a lasting effect on the development of future flower organs. Thus, EMF1 needs to be expressed throughout vegetative development to prevent secondary effects of ectopic AG expression that could impact subsequent normal flower organ development.
Temporal Knockout of EMF1 Activity during Seed Development Caused Early Flowering
The promoter of a seed storage protein gene, At2S3 (Guerche et al., 1990), was used to direct antisense EMF1 expression in order to knock out or reduce EMF1 activity in the seed. We found that At2S3::asEMF1 seedlings displayed emf mutant phenotype, short hypocotyls and petioles, and oval-shape cotyledons (Figure 3A). Since At2S3 expression decreases rapidly after germination (Supplemental Figure 1B), EMF1 activity would be restored in vegetative tissues. Indeed, at 7 DAG, At2S3::asEMF1 seedlings displayed emf1-like cotyledons (Figure 3B, left). By 14 DAG, some remained emf-like. Others, however, were able to develop two normal rosette leaves with long petioles from the shoot apex (stated as 2S lines; Figure 3B, middle). These plants were transferred to soil and, at 29 DAG, formed normal rosette shoots (Figure 3B, right). However, these At2S3::asEMF1 lines showed reduced inflorescence growth, flowered early (Figure 3C), and produced abnormal and terminal flowers (Figure 3D and 3E). The EMF1 RNA level of the green siliques in At2S3::asEMF1 was greatly reduced relative to that in WT (Figure 3F). These results indicate, despite resumption to normal vegetative rosette growth, reducing EMF1 activity in the embryo and perhaps for a short while after germination imparts a reproductive cell memory, which ultimately affects the timing of vegetative to reproductive transition.
|
Mechanism of Early Flowering
We reasoned that the reduced EMF1 activity in early-flowering transgenic plants harboring At2S3::asEMF1 and LFY::asEMF1 constructs should lead to de-repression of flower MADS-box genes in these plants. To test this hypothesis, we crossed these antisense EMF1 lines with the transgenic plants harboring APETALA3(AP3)::GUS or AGAMOUS (AG)::GUS (Figure 4A; see Methods), and examined temporal and spatial GUS activities in the F2 plants harboring both antisenseEMF1 and GUS transgene driven by AG or AP3 promoter. For controls, we introduced AP3::GUS and AG::GUS into emf1 mutants to show the temporal and spatial expression pattern in plants devoid of EMF1 activity relative to plants with EMF1 activity removed at specific times. Similarly, AP3::GUS and AG::GUS activities in WT plants were studied.
|
For At2S3::asEMF1 transgenic plants, 14-day-old seedlings were WT-like but the shoot apex showed high AP3::GUS and AG::GUS activities (Figure 4B), while the non-transgenic (WT) plants did not show GUS activity at this developmental stage, as expected. The emf1 mutants grew slower; besides the shoot apex, the petiole-less cotyledons and hypocotyls also displayed strong GUS activity (Figure 4B). The AP3::GUS and AG::GUS activity patterns in LFY::asEMF1 transgenic plants were very similar to those of At2S3::asEMF1. The GUS activity was found, in both cases, in the seedling shoot apex. The GUS activity was weaker in LFY::asEMF1 than the At2S3::asEMF1 transgenic plants (Figure 4B). However, AP1::asEMF1 transgenic plants, which did not flower early, showed no GUS activity in the seedlings (Figure 4B).
These results showed that reduced EMF1 activity in the embryo and young seedlings from antisense EMF1 transgenic lines caused ectopic expression of flower MADS-box genes in vegetative shoot apex. The presence of MADS-box proteins reflects a reproductive cell memory of the shoot apex that ultimately caused early flowering.
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
The pleiotropic effect of emf1 mutation and constitutive expression of EMF1 led us to investigate the spatial activity of EMF1 promoter and its stage-specific effect. Stage-specific removal of EMF1 would cause early flowering—a phenomenon that appears in transgenic plants expressing varying amounts of EMF1 RNA (Aubert et al., 2001). However, the pleiotropic effect of emf mutation, which probably results from constitutive expression of the flower homeotic genes, is largely alleviated in transgenic plants temporarily devoid of EMF1 activity.
EMF1 RNA and EMF1 proteins were found in all tissues and organs by RT–PCR (Aubert et al., 2001), GeneChip (Zimmerman et al., 2005), and Western blotting results (Calonje et al., 2008). However, analysis of EMF1::GUS activity in transgenic plants showed a temporal and spatial pattern of EMF1 promoter activity in that, as organs mature, GUS activity disappears from differentiating tissues, such as elongating petiole or pedicel cells. Since flower MADS-box genes are not active in mature cells, they may be maintained in a repressive state or in a heterochromatic state in a manner independent of EMF1 for silencing. Much of the EMF1 RNA and protein found in developing flower bud appear to reside in differentiating anther and stigmatic tissues. In emf1-2 mutants, cotyledons and shoot apex were GUS-negative at 9 DAG, indicating that EMF1 promoter activity in these tissues depends on EMF1. At 18 DAG, emf1 cotyledons are differentiating into carpelloid structures with stigmatic papilla and open ovules (Chen et al., 1997). EMF1::GUS activity in carpelloid tissues of WT and emf1-2 is consistent with the notion of an EMF1-independent EMF1 promoter activity. Furthermore, GUS-positive emf1-2 roots indicate that EMF1 is repressed by itself in WT roots. These results suggest an EMF1 positive auto-regulation in young shoots and negative auto-regulation in seedling roots. The possible organ-specific auto-regulation requires further investigation.
LFY promoter activity is active in leaf primordia as early as 4 DAG (Blázquez et al., 1997), and is gradually up-regulated during vegetative development in leaf primordia, peaking in flower meristem. LFY promoter driven EMF1 expression in the shoot apex is apparently insufficient to rescue emf1-2 mutant phenotype. Similarly, expression of EMF1 in shoot apical meristem by the KNAT1 promoter activity could not rescue emf1-2. Thus, EMF1 is needed not only in shoot meristem, but also in leaf primordia, probably also in differentiating organs to maintain rosette development. However, these promoters likely confer different levels of expression that may complicate these interpretations.
We investigated the effect of removing EMF1 activity in selected tissues and developmental stages on development. KNAT1::antisense EMF1 transgenic plants display emf mutant phenotype, indicating that EMF1 activity is required in shoot apical meristem for maintaining vegetative growth at germination. LFY::antisense EMF1 transgenic plants did not affect rosette development at first, but caused early flowering, terminal flower formation and affected leaf differentiation. The ability of LFY::antisense EMF1 transgenic plants to produce rosette leaves indicates EMF1 activity in the leaf primordia is not required for specifying vegetative growth. However, it is required for the completion of normal leaf differentiation. Furthermore, reducing EMF1 activity in leaf primordia has a lasting effect on flower organ development. As MADS-box proteins interact with one another as well as regulate other MADS-box genes (de Folter et al., 2005), expression of MADS-box genes in wrong time and place could cause secondary effects on subsequent developmental events. Such pleiotropic effects are alleviated in AP1::asEMF1 plants, in which EMF1 activity is not reduced until flower meristem development. In flower meristem, EMF1 apparently no longer represses flowering.
Finally, it is not clear how these transgenic plants produced terminal flowers, which indicates the conversion of shoot apical meristem to a floral fate. LFY promoter is not active in shoot apical meristem; there should not be a reduction in EMF1 activity in the SAM, unless antisense EMF1 RNA or flower MADS-box gene products are diffusible. However, it is worth noting that the ‘florigen’ gene, FLOWERING LOCUS T (FT, Zeevaart, 2008) is de-repressed in emf2 (Yoshida et al., 2001); and constitutive expression of FT caused early flowering and terminal flower formation (Kobayashi et al., 1999; Kardailsky et al., 1999). Since EMF1 is likely to act in the PcG mechanism mediated by the EMF2-containing PRC2 complex, reducing EMF1 activity in leaf primordia could de-repress FT expression and increase the FT transmitted to shoot apical meristem, resulting in the activation of flower homeotic genes in the shoot apical meristem and causing early flowering.
At2S3 is highly expressed in siliques, namely mature embryos. Hence, the germinating seedlings are likely to be devoid of EMF1 RNA for normal development. As a result, they expressed flower MADS-box genes (Figure 4B) and germinated like emf1 mutants with petiole-less, oval-shaped cotyledons and short hypocotyls. Residual At2S3 RNA was found in 3-day-old seedlings; its level significantly reduced afterwards (Supplemental Figure 1B). This is consistent with AP3::GUS activity in 7-day-old At2S3::antisenseEMF1 plants (Supplemental Figure 1C). The At2S3::GUS activity found in older seedlings, especially in leaf blades, could result from residual GUS enzyme activity expressed at earlier stages. However, AP3::GUS activity was found in the shoot apex of 7- and 14-day-old seedlings, not in leaf blades. The spatial difference in AP3 expression and EMF1 reduction can also be explained by the above scenario of FT de-repression in cotyledon and leaf primordia that led to the flower homeotic gene expression in the shoot apex.
However, the AG- and AP3-expressing plants did not flower immediately; the plants continued to produce rosette leaves, but flowered early relative to WT (Figures 2E and 3C). The production of rosette leaves indicates the resumption of EMF1 activity, which apparently can no longer repress these flower MADS-box genes in shoot apex where cells are mitotic. However, EMF1 apparently can repress these genes in differentiating and mature cells of leaves, resulting in rosette leaf development. The role of DNA replication on the maintenance of epigenetic states has been reported. For example, the maintenance of vernalization-induced Flowering Locus C repression is dependent on DNA replication (Finnegan and Dennis, 2008).
Reducing EMF1 activity in the seed does not affect seed development, because EMF1 is only involved in the repression of seed maturation program after germination. Its presence during seed development cannot affect seed maturation, as in flower EMF1 cannot affect flower MADS-box gene expression, probably because it cannot compete with stage-specific transcriptional activators that rendered chromatin active (Ringrose and Paro, 2004).
Flowering is the most prominent phenotype observed whenever EMF1 activity is reduced, overshadowing the effect of altered expression pattern of genes involved in other gene programs; for example, EMF promotes genes involved in photosynthesis and hormone signaling (Moon et al., 2003). It is possible that photosynthesis and hormone signaling are more crucial for normal vegetative than flower development, as the former involves greater variation in cell and organ size to generate varying leaf shape and branching pattern, etc. In fact, early flowering plants, such as LFY::asEMF1 transgenic plants, display dwarf phenotype, such as small rosette and short inflorescence stem (Figure 2D), which might result from altered hormone signaling. Flower organs are not the primary photosynthetic organs, and may require a different set of hormone signaling processes from those for vegetative leaves. Gametophyte and embryo development may also involve diverse hormone signaling genes or is not regulated by EMF1, as in flower. In summary, our results support the hypothesis that EMF1 was evolved to promote vegetative growth by repressing flower and seed development—the two major energy-consuming processes in plant life. At the same time, EMF1 promotes gene programs required for vegetative growth.
|
METHODS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Plant Material and Growth Conditions
Surface-sterilized Arabidopsis thaliana seeds (ecotype Columbia) were plated on agar plates containing Murashige and Skoog Basal Salt Mixture (MS, Sigma), 1.5% sucrose, and 0.8% agar (Murashige and Skoog, 1962). The plates were placed for 2 d at 4°C, and then seedlings were grown under short-day conditions (8 h light/16 h dark) or long-day conditions (16 h light/8 h dark) at 21°C. Fifteen-day-old seedlings grown on agar plates were transferred to soil and grown under long-day conditions or short-day conditions at 21°C in the greenhouse. Flowering time was measured by scoring the number of rosette leaves of at least 20 individuals. Data are expressed as mean ± SE.
Plasmid Constructions
Four beta-glucuronidase (GUS) fusion constructs were generated with 3.2 or 7.6 Kb EMF1 promoter sequences with or without first intron of EMF1 gene (Aubert et al., 2001). The four putative regulatory regions were amplified by PCR using the following primer pairs: PROMF1/PROMR1 for the construct containing 7.6 Kb promoter and the first intron, PROMF2/PROMR1 for the construct containing 3.2 Kb promoter and the first intron, PROMF1/PROMR2 for the construct containing 7.6 Kb promoter alone, and PROMF2/PROMR2 for the construct containing 3.2 Kb promoter alone. Each primer had SalI or SmaI sites (underlined sequence) at the 5’ end:
- PROMF1, 5'-CGCGTCGACGGATCAAGGACAATCATTCTAG-3’;
- PROMF2, 5'-ACGCGTCGACCTG GTGTTCACCCTTAAGCTG-3’;
- PROMR1, 5'-TTTACCCGGGGTCTCGTTGGCAGCACCTGC-3’;
- PROMR2, 5'-TTTACCCGGGTTAATCCGACCAGAAATTAG-3'.
- PROMF2, 5'-ACGCGTCGACCTG GTGTTCACCCTTAAGCTG-3’;
The amplified DNA fragments were cloned into SalI/SmaI sites of pBI101 vector (Clontech, USA) generating the corresponding GUS fusion constructs. EMF1 cDNA (3.3 Kb) was introduced into pCAMBIA1380 vector (Cambia, Australia) in sense (as PstI–SpeI fragment) or antisense (as SpeI-blunt fragment) orientation, generating pE1S and pE1AS plasmids, respectively.
The At2S3, KNAT1, LFY, and AP1 promoters were PCR amplified from Arabidopsis genomic DNA (ecotype Columbia) using specific primers with an ApaI restriction site in the 5’ end of the forward primer. Specific sequences for each primer pair were:
- At2S3p-F, 5'-GAAACCAAATTAACATAG-3’;
- At2S3p-R, 5'-GTTTTGCTATTTGTGTATGTT-3’;
- KNAT1p-F, 5'-GATCTAGAGCCCTAGGAT-3’;
- KNAT1p-R, 5'-ACCCAGATGAGTAAAGATTTGAG-3’;
- LFYp-F, 5'-TTTTTCGCAAAGGAAAGT-3’;
- LFYp-R, 5'-CCGAACTAGTATAATCTATTTTTCTCTC-3’;
- AP1p-F, 5'-TACTATCTTTAGACTGAT-3’;
- AP1p-R, 5'-CCGTACTAGTCTCAGACTTTGGTATGAA-3’;
- At2S3p-R, 5'-GTTTTGCTATTTGTGTATGTT-3’;
All promoters were introduced into pE1S and/or pE1AS plasmids as ApaI-blunt fragments, generating the different promoter fusions with EMF1 cDNA in sense or antisense orientation, respectively. KNAT1 promoter length is described in Lincoln et al., 1994; LFY promoter in Blázquez et al., 1997; AP1 promoter in Hempel et al., 1997; and At2S3 promoter in Kroj et al., 2003.
To generate At2S3::GUS construct, At2S3 promoter (–310 to +35 bp relative to the transcription start, according to Kroj et al., 2003) was introduced into pBI101 vector (Clontech, USA) as a HindIII-blunt fragment.
Plant Transformation
All plasmids were introduced into Agrobacterium tumefaciens strain GV3101 (pMP90) (Koncz and Schell, 1986) and transformed into Arabidopsis wild-type Columbia or emf1 heterozygous plants by floral dip (Clough and Bent, 1998). Seeds from the transformed plants were selected with 50 mg L–1 Kanamycin (pBI101 vector containing plants) or 15 mg L–1 Hygromycin B (pCAMBIA1380 vector containing plants).
RNA Isolation and RT–PCR
Total RNA from seedlings grown in agar plates under short-day conditions (8 h light) was extracted with Trizol (Life Technologies), according to the manufacturer's instructions. For dry seed, green silique, and 2-day-old seedlings samples, total RNA extraction was as follows: 0.5–1 g of tissues were ground in liquid nitrogen and added to 15 ml of Extraction Buffer (2% CTAB, hexadecyltrimethylammonium bromide, 2% PVP, polyvinylpyrrolidinone K30, 100 mM Tris-HCl, pH 8.0, 25 mM EDTA, pH 8.0, 2 M NaCl, 0.5 g L–1 spermidine, 2% β-mercaptoethanol) preheated at 65°C. After two chloroform:isoamyl alcohol (24:1) extractions, 10 M LiCl was added to the aqueous phase containing the RNA and incubated overnight at 4°C. RNA pellet was collected by centrifugation at 10 000 rpm and re-suspended in 500 µl SSTE buffer (1 M NaCl, 0.5% SDS, 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, pH 8.0). After one chloroform:isoamyl alcohol (24:1) extraction, RNA was precipitated with 2 vol. of 100% ethanol at –70°C for 1 h, then collected by centrifugation and re-suspended in DEPC-treated H2O. The integrity of the RNA was verified on a gel before RT–PCR. RNA samples were treated with 2 units of DNaseI for 1 h at 37°C, and then DNaseI was inactivated by treatment with DNaseI Inactivating Reagent (Ambion). This DNA-free RNA (1 µg) was retrotranscribed with oligo(dT) primer and SuperScript II RT (Invitrogen) at 42°C for 1 h. Samples of the first strand cDNA were then used in PCR reactions with the following gene-specific primers:
- EMF1-F, 5'-ACCAGTACGAGAGGTGTCTTC-3’;
- EMF1-R, 5'-CAGAAGGCTGAGTAGATGCA-3’;
- At2S3-F, 5'-CATCCCTTTCTTCCCTCCTT-3’;
- At2S3-R, 5'-GAGCAGCAAGGGTAAGAACATT-3’;
- Actin-F, 5'-GGAAGGATCTGTACGGTAAC-3’;
- Actin-R, 5'-TGTGAACGATTCCTGGACCT-3’;
- GAPc-F, 5'-CACTTGAAGGGTGGTGCCAAG-3’;
- GAPc-R, 5'-AACTCTAGACTCGAGGATGGATGTGTTTACTCATG-3'.
- EMF1-R, 5'-CAGAAGGCTGAGTAGATGCA-3’;
Control reactions were performed using as template non-reverse transcribed RNA to rule out possible amplification from contaminating genomic DNA.
Genetic Crosses
AG::GUS line, pMD200 (Deyholos and Sieburth, 2000), was kindly supplied by Dr L.E. Sieburth (University of Utah, USA). AP3::GUS line, 5D3 (Hill et al., 1998), was kindly supplied by Dr V. Irish (Yale University, USA). To introduce AG::GUS (pMD200) or AP3::GUS (5D3) construct into LFY::asEMF1, AP1::asEMF1, or At2S3::asEMF1 transgenic plants, a homozygous line for each reporter construct was crossed with a homozygous line for each asEMF1 transgenic plant. F1 plants were grown in MS medium containing 50 mg L–1 Kanamycin and 15 mg L–1 Hygromycin B and selfed. F2 plants were selected in MS medium containing both antibiotics, Kanamycin and Hygromycin B, and grown to generate F3 plants, of which homozygous line for both construct was identified by phenotype segregation and drugs resistance.
GUS Activity Assays
GUS activity was detected histochemically using a protocol adapted from Jefferson (1987), with slight modifications. Briefly, the seedlings or tissue was incubated in 2 mM 5-bromo-4-chloro-3-indolyl-β-D-glucuronic acid in 50 mM phosphate buffer, pH 7.0, containing 0.5 mM K3Fe(CN)6 and 0.5 mM K4Fe(CN)6 for 6 h at 37°C, rinsed with 50 mM phosphate buffer, fixed and cleared with ethanol (95%):acetic acid (9:1, v/v) for 4 h at room temperature, and observed and photographed using a Zeiss Lumar.v12 epifluorescence stereoscope equipped with a Q imaging color digital camera.
|
SUPPLEMENTARY DATA |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Supplementary Data are available at Molecular Plant Online.
|
FUNDING |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
This work is supported by NSF grant #IBN 0236399 and USDA grant #03–35301–13244 to Z.R.S. R.S. was supported by a postdoctoral fellowship from the Spanish Ministry of Education and Science.
|
Acknowledgements |
|---|
We thank L. Sieburth (University of Utah, USA) and V. Irish (Yale University, USA) for AG::GUS and AP3::GUS transgenic lines, respectively; L.J. Chen and O. Phan for crossing AG::GUS and AP3::GUS constructs into emf mutants; J. Shin, E. Abbott, C. Heng, C. Tse and K. Loo for their technical support. No conflict of interest declared.
-
Aubert D, Chen L, Moon YH, Martin D, Castle LA, Yang CH, Sung ZR. EMF1, a novel protein involved in the control of shoot architecture and flowering in Arabidopsis. Plant Cell. (2001) 13:1865–1875.
Blázquez MA, Soowal LN, Lee I, Weigel D. LEAFY expression and flower initiation in Arabidopsis. Development (1997) 124:3835–3844.[Abstract]
Calonje M, Sung ZR. Complexity beneath the silence. Curr. Opin. Plant Biol. (2006) 9:530–537.
Calonje M, Sanchez R, Chen L, Sung ZR. EMBRYONIC FLOWER1 participates in polycomb group-mediated AG gene silencing in Arabidopsis. Plant Cell. (2008) 20:277–291.
Chanvivattana Y, Bishopp A, Schubert D, Stock C, Moon YH, Sung ZR, Goodrich J. Interaction of Polycomb-group proteins controlling flowering in Arabidopsis. Development (2004) 131:5263–5276.
Chen L, Cheng JC, Castle L, Sung ZR. EMF genes regulate Arabidopsis inflorescence development. Plant Cell. (1997) 9:2011–2024.[Abstract]
Clough SJ, Bent AF. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J (1998) 16:735–743.[CrossRef][Web of Science][Medline]
de Folter S, et al. Comprehensive interaction map of the Arabidopsis MADS box transcription factors. Plant Cell. (2005) 17:1424–1433.
Deyholos MK, Sieburth LE. Separable whorl-specific expression and negative regulation by enhancer elements within the AGAMOUS second intron. Plant Cell. (2000) 12:1799–1810.
Finnegan E, Dennis E. Vernalization-induced trimethylatin of histone H3 Lysine 27 at FLC is not maintained in mitotically quiescent cells. Current Biol. (2008) 17:1978–1983.[Web of Science]
Goodrich J, Puangsomlee P, Martin M, Long D, Meyerowitz EM, Coupland G. A Polycomb-group gene regulates homeotic gene expression in Arabidopsis. Nature (1997) 386:44–51.[CrossRef][Medline]
Guerche P, Tire C, de Sa FG, De Clercq A, Van Montagu M, Krebbers E. Differential expression of the Arabidopsis 2S Albumin genes and the effect of increasing gene family size. Plant Cell. (1990) 2:469–478.
Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, Feldman LJ, Yanofsky MF. Floral determination and expression of floral regulatory genes in Arabidopsis. Development (1997) 124:3845–3853.[Abstract]
Hennig L, Taranto P, Walser M, Schonrock N, Gruissem W. Arabidopsis MSI1 is required for epigenetic maintenance of reproductive development. Development (2003) 130:2555–2565.
Hill TA, Day CD, Zondlo SC, Thackeray AG, Irish VF. Discrete spatial and temporal cis-acting elements regulate transcription of the Arabidopsis floral homeotic gene APETALA3. Development (1998) 125:1711–1721.[Abstract]
Jefferson RA. Assaying chimeric gene expression in plants the GUS gene fusion system. Plant Mol. Biol. Rep (1987) 5:387–405.[CrossRef]
Kardailsky I, Shukla VK, Ahn JH, Dagenais N, Christensen SK, Nguyen JT, Chory J, Harrison MJ, Weigel D. Activation tagging of the floral inducer FT. Science (1999) 286:1962–1965.
Katz A, Oliva M, Mosquna A, Hakim O, Ohad N. FIE and CURLY LEAF polycomb proteins interact in the regulation of homeobox gene expression during sporophyte development. Plant J (2004) 37:707–719.[CrossRef][Web of Science][Medline]
Kinoshita T, Harada JJ, Goldberg RB, Fischer RL. Polycomb repression of flowering during early plant development. Proc. Natl Acad. Sci. U S A (2001) 98:14156–14161.
Kobayashi Y, Kada H, Goto K, Iwabushi M, Araki T. A pair of related genes with antagonistic roles in mediating flowering signals. Science (1999) 286:1960–1962.
Kohler C, Makarevich G. Epigenetic mechanisms governing seed development in plants. EMBO Rep (2006) 7:1223–1227.[CrossRef][Web of Science][Medline]
Koncz C, Schell J. The promoter of Tl-DNA gene 5 controls the tissue-specific expression of chimeric genes carried by a novel type of Agrobacterium binary vector. Mol. Gen. Genet. (1986) 204:383–396.[CrossRef][Web of Science]
Kroj T, Savino G, Valon C, Giraudat J, Parcy F. Regulation of storage protein gene expression in Arabidopsis. Development (2003) 130:6065–6073.
Lincoln C, Long J, Yamaguchi K, Serikawa K, Hake S. A Knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants. Plant Cell. (1994) 6:1859–1876.
Martinez AM, Cavalli G. The role of polycomb group proteins in cell cycle regulation during development. Cell Cycle (2006) 5:1189–1197.[Web of Science][Medline]
Moon YH, Chen L, Pan RL, Chang HS, Zhu T, Maffeo DM, Sung ZR. EMF genes maintain vegetative development by repressing the flower program in Arabidopsis. Plant Cell. (2003) 15:681–693.
Murashige T, Skoog F. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant (1962) 15:473–497.
Nilsson O, Lee I, Blazquez MA, Weigel D. Flowering-time genes modulate the response to LEAFY activity. Genetics (1998a) 150:403–410.
Nilsson O, Wu E, Wolfe D, Weigel D. Genetic ablation of flowers in transgenic Arabidopsis. Plant J (1998b) 15:799–804.[CrossRef][Web of Science][Medline]
Pien S, Grossniklaus U. Polycomb group and trithorax group proteins in Arabidopsis. Biochim. Biophys. Acta (2007) 1769:375–382.
Ringrose L, Paro R. Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Ann. Rev. Genet. (2004) 38:413–443.[CrossRef][Web of Science][Medline]
Schmitz RJ, Amasino RM. Vernalization: a model for investigating epigenetics and eukaryotic gene regulation in plants. Biochim. Biophys. Acta (2007) 1769:269–275.[Medline]
Schwartz YB, Pirrotta V. Polycomb silencing mechanisms and the management of genomic programmes. Nature Reviews Genetics (2007) 8:9–22.[Web of Science][Medline]
Schubert D, Clarenz O, Goodrich J. Epigenetic control of plant development by Polycomb-group proteins. Curr. Opin. Plant Biol. (2005) 8:553–561.
Schubert D, Primavesi L, Bishopp A, Roberts G, Doonan J, Jenuwein T, Goodrich J. Silencing by plant Polycomb group genes requires dispersed trimethylation of histone H3 at lysine 27. EMBO J (2006) 25:4638–4649.[CrossRef][Web of Science][Medline]
Sung ZR, Belachew A, Bai S, Bertrand-Garcia R. EMF, an Arabidopsis gene required for vegetative shoot development. Science (1992) 258:1645–1647.
Valk-Lingbeek ME, Bruggeman SW, van Lohuizen M. Stem cells and cancer: the polycomb connection. Cell. (2004) 118:409–418.[CrossRef][Web of Science][Medline]
Yang CH, Chen LJ, Sung ZR. Genetic regulation of shoot development in Arabidopsis: role of the EMF genes. Dev. Biol. (1995) 169:421–435.[CrossRef][Web of Science][Medline]
Yoshida N, Yanai Y, Chen L, Kato Y, Hiratsuka J, Miwa T, Sung ZR, Takahashi S. EMBRYONIC FLOWER2, a novel polycomb group protein homolog, mediates shoot development and flowering in Arabidopsis. Plant Cell. (2001) 13:2471–2481.
Zeevaart J. Leaf-produced floral signals. Curr. Opin. Plant Biol. (2008) 11:541–547.
Zimmermann P, Hennig L, Gruissem W. Gene expression analysis and network discovery using Genevestigator. Trends Plant Sci. (2005) 10:407–409.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||