QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 29 September 2004
doi: 10.1242/dev.01400

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chanvivattana, Y. Articles by Goodrich, J. Search for Related Content PubMed PubMed Citation Articles by Chanvivattana, Y. Articles by Goodrich, J. Social Bookmarking                         What's this?

Interaction of Polycomb-group proteins controlling flowering in Arabidopsis

Yindee Chanvivattana^1,*,, Anthony Bishopp^1,, Daniel Schubert¹, Christine Stock¹, Yong-Hwan Moon^2,, Z. Renee Sung² and Justin Goodrich^1,

¹ Institute of Molecular Plant Science, School of Biology, University of Edinburgh, Mayfield Road, Edinburgh EH9 3JH, UK
² Department of Plant and Microbial Biology, University of California, Berkeley, CA 94720, USA

View larger version (72K): [in a new window]   Fig. 1. The moe leaf phenotype. (A) Short-day grown plants of clf-50, emf2-10 and wild-type progenitor (Ws ecotype) after 30 days. The emf2-10 plant is slightly earlier flowering and smaller than clf-50. (B) Fifth rosette leaf of long-day grown plants. The emf2-10 leaf is smaller and narrower than clf-50, but has similar upward curling of leaf margin. (C) Seedlings showing effects of emf2-10 on cotyledon size. (D) Wild-type and emf2-10 flowers at anthesis. (E) emf2-10 flowers showing delayed flower opening (left) resulting in contorted siliques (right). (F) emf2-10 flowers from apex of inflorescence. Arrowheads indicate carpelloid sepals; petals are also stamenoid in shape and have yellow anther-like sectors. Flowers appear terminal because flower buds from shoot apex (between two flowers) aborted early in development and are no longer visible. (G-M) Transgenic plants carrying AG or AP3 reporter genes stained for GUS activity (blue colour). (G) pAG-I::GUS activity in clf-2 seedlings. Expression was also seen in cotyledons at earlier stages. (H) pAG-I::GUS in emf2-10 seedling. (I) pAG-I::GUS in wild-type inflorescence. (J) pAG-I::GUS in clf-2 inflorescence. Arrow indicates expression in the stem. (K) pAG-I::GUS in emf2-10 plant. Arrow indicates expression in inflorescence stem. (L) pAP3::GUS in clf2 seedling. (M) pAP3::GUS in emf2-10 seedling. (N) The emf2-10 ag-2 double mutant (arrow) is shown between ag-2 and emf2-10 parent lines. Note that the double mutant is earlier flowering than ag-2 and smaller. (O) The emf2-10/emf2-3 heterozygote (arrow) is shown between emf2-10 and emf2-3 parents. Its phenotype is intermediate, both with respect to height and overall plant size. Scale bar: 5 mm in A,B; 2 mm in C-O.

View larger version (144K): [in a new window]   Fig. 2. Scanning electron microscopy of Pc-G mutants. (A) Wild-type leaf, abaxial epidermis. (B) emf2-10 leaf, abaxial epidermis. The surface is extremely uneven compared with wild type. (C) Wild-type fifth rosette leaf. The leaf was frozen and fractured to reveal internal anatomy transverse to leaf length axis. The leaf is patterned along dorsal-ventral axis into epidermal cells (e), a palisade mesophyll layer (p), and a spongy mesophyll (s) (D) Fifth rosette leaf of emf2-10 plant of similar age. A similar arrangement of cell types is seen as in wild type, but the cells are smaller. (E) emf2-10 flower showing contorted silique. (F) Wild-type flower showing abaxial epidermi of sepals (se) and petals (pe). The sepal epidermis contains characteristic highly elongated cells (arrow); however, the margin lacks the elongated cells and has smaller, more regularly sized cells (arrowhead). (G) emf2-10 sepal, elongated cells extend to the margin. (H) emf2-10 flower showing carpelloid sepal. The organ has elongated cells typical of sepals but stigmatic papillae (arrowhead) and stylar cells (arrow) characteristic of carpels. (I) Wild-type petal, abaxial surface, note lack of stomates. (J) emf2-10 petal, abaxial surface. Note presence of stomates (arrow) and cell shape, characteristic of stamen epidermis. (K) swn-1 clf-50 inflorescence. The sepals show weak homeotic conversion to carpelloid organs. Arrowheads indicate stigmatic papillae. (L) Radialised organ (arrow) with stigmatic papillae arising from inflorescence stem of swn-1 clf-50 double mutant in position where stipule would normally arise. (M) swn-3 clf-50 double mutant. Organs arise with disorganised phyllotaxy. Note lack of trichomes, cells lack wall thickening and are isodiametric. (N) swn-2 clf-50 plant showing radialised organs. Scale bar: 100 µm throughout.

View larger version (78K): [in a new window]   Fig. 3. Genetic interactions of Pc-G mutants. (A) Comparison of emf2-10 clf-9 double mutant (arrow) and parental single mutants of same age. Note the minute size of the double mutant. (B) emf2-10 clf-9 double mutant showing cotyledons (c), a few sessile leaves (sl) and a terminal flower. Tissue-culture-grown plant. (C) emf2-3 single mutant, grown in tissue culture. Note similarity with B and D. (D) clf-2 emf2-10 double mutant, grown in tissue culture. (E) clf-2 emf2-10 ag-1 triple mutant showing minute plant with several ag– flowers. (F) clf-50 single mutant (left) compared with clf-50 swn-1 double mutant (right) to show enhanced phenotype. swn-1 single mutants (not shown) had wild-type appearance. (G) swn-1 clf-50 double mutant showing small size and reduced inflorescence bolt. (H) swn-3 clf-50 double mutant plants after about 5 week's growth in tissue culture. Arrows indicate root hairs developing from mass of callus-like tissue. (I) swn-3 clf-50 double mutant showing somatic embryo (arrow) formed on callus-like tissue. (J) swn-3 clf-50 double mutant showing green shoot-like tissue developing from primary root (arrow). The arrowhead indicates the main shoot. Scale bar: 1 mm.

View larger version (52K): [in a new window]   Fig. 4. Interaction of EMF2 and CLF in yeast and in vitro. Two-hybrid assays were performed in yeast strain HF7c. For nutritional assays, three independent transformants were stamped onto –LW and –LWH media. Growth on –LW selects for markers carried on the bait and prey plasmids, whereas growth on –LWH also indicates activity of the HIS3 reporter gene. ß-Galactosidase activity was quantified using the assay and units of Miller (Miller, 1972; Miller, 1992). Each value is an average from assays of three independent transformants; the standard error of the mean is also indicated. (A) The EMF2 protein is shown schematically, with the zinc finger motif indicated by the black box and the conserved VEFS domain by the blue box. The uppermost row is a control to show that the CLF bait does not have transcriptional activation activity by itself and cannot interact with an `empty' GAL4-TA prey. The smallest region of EMF2 that was sufficient for interaction with CLF comprised residues 510-631. (B) The CLF protein is shown schematically, with the C5 domain indicated in orange and the CXC region, which precedes the SET domain, shown in turquoise. All prey fusions that contained an intact C5 domain were able to interact with the EMF2-VEFS domain (427-631). The shortest region of CLF sufficient for interaction comprised residues 257-331. (C) Split ubiquitin assay using the system of Kim et al. (Kim et al., 2002). CLF protein, lacking the C-terminal SET domain, was fused to the C-terminal half of ubiquitin (CUB) and the EMF2 VEFS domain was fused to an N-terminal portion of a modified ubiquitin (NUB). The NUB and CUB peptides are unable to interact on their own. Interaction of NUB and CUB fusions reconstitutes ubiquitin activity and results in proteolysis of a URA3 reporter. This allows growth on media containing FOA. Growth on –HW medium selects for the markers on the CUB and NUB constructs. (D) (lane A) In-vitro binding of CLF C5 domain and EMF2 VEFS domain. Bacterial extract containing His6–EMF2 VEFS protein was tested for binding to GST–CLF C5 (lane B) or GST (lane C). Proteins that bound to GST or GST-CLF C5 were separated by SDS–PAGE, transferred to PVDF membrane, and incubated with anti-His6 antibodies. The input lane (lane A) contains 1.5% of the volume of bacterial extract used in the binding assay. The lower band in the input lane corresponds to a His6–EMF2 VEFS degradation product. Note that this is not bound by GST-CLF C5.

View larger version (32K): [in a new window]   Fig. 5. Interaction between Arabidopsis CLF and EMF2 homologues in yeast. Two-hybrid constructs were introduced into the yeast strain AH109, which contains reporter genes that confer histidine and adenine prototrophy. The ADE3 reporter is extremely stringent. Three independent transformants were stamped onto –LW and –LWHA media. –LW selects for markers on the bait and prey constructs, while –LWHA also selects for activation of the HIS3 and ADE2 reporters. (A) Interaction of the VEFS domain of FIS2 (residues 466-692) with full-length MEA protein. (B) Cladogram of plant E(z) homologues. Drosophila E(z) is included as an outlier. Analysis was performed using the PAUP program to align the SET domains of the proteins. The bootstrap values are indicated. (C) Interaction of SWN C5 domain (252-331) with the VEFS domains of EMF2 (510-631), VRN2 (275-440) and FIS2 (394-692). (D) Interaction of CLF C5 domain (257-331) with the VEFS domains of EMF2, VRN2 and FIS2.

View larger version (120K): [in a new window]   Fig. 6. Expression of SWN. Transcript localization by in-situ hybridisation. Tissue appears light blue, whereas the signal appears dark purple/brown. (A-F) were hybridised with an SWN antisense probe generated from a region at the 5' end of the cDNA that lacked similarity with any other Arabidopsis gene. (A) Longitudinal section through an 8-day-old seedling showing expression in the shoot apical meristem (arrowed) and young leaf primordia (P). (B) Seedling section showing expression in vasculature (arrowed) and in older leaf primordia. (C) Longitudinal section through inflorescence showing strong expression in inflorescence meristem (IM), and throughout young stage 1 and 3 floral meristems. (D) Stage 7 flower showing expression in stamens (S), carpels (C) and emerging petal primordium (arrowed). (E) Stage 9 flower showing strong expression in emerging petal primordium (arrow), stamens (S) and carpels (C). (F) Longitudinal section through carpel of stage-12 flower showing mature ovules. Expression is low in the carpel walls (W), but strong throughout the sporophytic tissue of the ovule, particularly in the funiculus (F). Expression is also visible in the embryo sac (arrow). (G) Longitudinal section through seedling hybridised with SWN sense probe. (H) Longitudinal section through inflorescence hybridised with SWN sense probe. (I) Transverse section through 8-day-old seedling hybridised with WUS antisense probe. Signal is confined to the centre of the meristem, as previously described (Mayer et al., 1998).

View larger version (31K): [in a new window]   Fig. 7. Comparison of CLF, MEA and SWN misexpression. Transformed plants (T2 generation) containing 35S::CLF, 35S::MEA and 35S::SWN transgenes in a clf-50 mutant background. 35S::CLF complements clf-50 whereas the other two transgenes do not. At least 23 primary transformants were obtained for each construct. The CLF+ progenitor and clf-50 mutant are shown for comparison.

View larger version (17K): [in a new window]   Fig. 8. Arabidopsis Polycomb-group protein complexes. The core components of the Drosophila PRC2 complex are shown at top. In Arabidopsis, an equivalent ancestral complex is proposed to have diversified into three similar complexes with at least partially discrete functions. The colours indicate homology; so for example, E(z) homologues are coloured red. The contacts indicate interactions; for example, FIE can interact with MEA and MSI1 but not FIS2, whereas FIS2 can interact with MEA but not with other FIS proteins. The target genes shown are not comprehensive; it is likely that all three complexes have many more targets than those shown.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter