Lightregulated largescale reorganization of ... - Wiley Online Library

5 downloads 4435 Views 756KB Size Report
pathways responsive to endogenous and exogenous cues. ..... by COP1, whereas COP1 activity in turn is negatively regulated .... ally processed using ADOBE PHOTOSHOP (Adobe, http://www.adobe.- ... simon.bio.uva.nl/object-image.html).
The Plant Journal (2007) 50, 848–857

doi: 10.1111/j.1365-313X.2007.03093.x

Light-regulated large-scale reorganization of chromatin during the floral transition in Arabidopsis Federico Tessadori, Roeland Kees Schulkes, Roel van Driel and Paul Fransz* Nuclear Organization Group, Swammerdam Institute for Life Sciences, BioCentrum Amsterdam, University of Amsterdam, Kruislaan 318, 1098 SM, Amsterdam, The Netherlands Received 8 November 2006; revised 30 January 2007; accepted 9 February 2007. *For correspondence (fax + 31205257935; e-mail [email protected]).

Summary The floral transition marks the switch from vegetative to reproductive growth, and is controlled by different pathways responsive to endogenous and exogenous cues. The developmental switch is accompanied by local changes in chromatin such as histone modifications. In this study we demonstrate large-scale reorganization of chromatin in rosette leaves during the floral transition. An extensive reduction in chromocenters prior to bolting is followed by a recovery of the heterochromatin domains after elongation of the floral stem. The transient reduction in chromocenters is a result of relocation away from chromocenters of methylated DNA sequences, 5S rDNA and interspersed pericentromeric repeats, but not of 45S rDNA or the 180-bp centromere tandem repeats. Moreover, fluorescence in situ hybridization analysis revealed decondensation of chromatin in gene-rich regions. A mutant analysis indicated that the blue-light photoreceptor CRYPTOCHROME 2 is involved in triggering chromatin decondensation, suggesting a light-signaling pathway towards large-scale chromatin modulation. Keywords: Arabidopsis, floral transition, nucleus, heterochromatin.

Introduction Floral transition is a fundamental stage in plant development, during which the shoot apical meristem switches from generating leaf primordia to production of reproductive organs. A pivotal factor for successful plant reproduction is the timing of the floral transition, as favorable environmental conditions are essential for fertilization and optimal seed production. Plants have developed various strategies to comply with this aim. For example, although facultative long-day plants, such as Arabidopsis, are induced to flower by a rise in temperature and increasing day length, flowering in maize and rice is promoted by shorter days. In the dicotyledonous short-day plant Pharbitis nil, even a single exposure to a 14-h night is floral inductive (Liu et al., 2001; Vince-Prue and Gressel, 1985). A number of different pathways have been described in Arabidopsis that induce the floral transition (Boss et al., 2004; He and Amasino, 2005). Two of these, the vernalization and photoperiod pathways, are dependent on environmental factors and promote flowering after a prolonged cold period and a long-day regime, respectively. In contrast, intrinsic control mechanisms, such as the gibberellin path848

way and the autonomous pathway, are largely dependent on endogenous developmental signals. The external and endogenous cues converge into a complex interaction between flower-inhibiting and -promoting signals, the balance of which determines the timing of the floral transition. Flower-inducing signals are integrated at the downstream end of the flower-promoting pathways by the MADS-box protein SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1) and FLOWERING LOCUS T (FT), of which the biochemical function is unknown. These proteins control the expression of floral meristem identity genes such as LEAFY (LFY) and APETALA 1 (AP1; reviewed in Boss et al., 2004; Putterill et al., 2004). SOC1 and FT themselves are under direct control of CONSTANS (CO), a zinc-finger transcription factor, and FLOWERING LOCUS C (FLC), a MADS-box transcription factor. Flowering time is also controlled by the photoreceptors PHYTOCHROME A (PHYA) and CRYPTOCHROME 2 (CRY2) through the photoperiod pathway, by positively regulating CO activity (Valverde et al., 2004; Yanovsky and Kay, 2002). Under long-day light conditions CO exerts a positive control on floral meristem ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd

Chromatin reorganization during floral transition 849 identity genes, among others through the production of a long-distance systemic florigen signal in the leaves (An et al., 2004; Huang et al., 2005; Martinez-Garcia et al., 2002; Zeevaart, 1976). It was recently demonstrated that the FT gene product is an important component of this signal (Huang et al., 2005). In contrast to CO, FLC has an inhibiting action on the expression of the SOC1 and FT genes (Hepworth et al., 2002). FLC is under the control of several regulators, including the FRIGIDA gene (FRI), which upregulates FLC expression (Johanson et al., 2000; Sheldon et al., 1999). The active FLC locus is associated with transcriptional activation hallmarks, such as methylation at lysine 4 of histone 3 and acetylation of histones (He et al., 2003, 2004). FLC repression is achieved by dimethylation of histone H3 at lysine 9 (H3 K9me2), mediated by VERNALIZATION INSENSITIVE 1 (VIN1), VERNALIZATION 1 (VRN1) and a Polycomblike protein VERNALIZATION 2 (VRN2; Bastow et al., 2004; Gendall et al., 2001; Levy et al., 2002; Sung and Amasino, 2004). Another protein that epigenetically regulates flowering time is the Arabidopsis HP1 homologue LIKE HETEROCHROMATIN PROTEIN 1 (LHP1), also called TERMINAL FLOWER 2 (TFL2) (Gaudin et al., 2001; Kotake et al., 2003; Larsson et al., 1998). LHP1 may participate in repression of the FT gene by counteracting CO (Takada and Goto, 2003). Consequently, lhp1 mutants are early flowering. The mutant appears to be impaired in FLC repression, possibly through interaction with VRN1 (Mylne et al., 2006). VRN1 binds to chromatin, but the binding site is not restricted to the FLC locus. Similar to LHP1, the VRN1 protein decorates euchromatin regions. Whether the function of VRN1 is part of a machinery that affects more genes apart from FLC, like LHP1, (Gaudin et al., 2001; Kotake et al., 2003; Libault et al., 2005) is not known. Floral transition is a major developmental switch that involves physiological processes in the apical meristem, but also in the rosette leaves, which sense the changing environmental conditions such as the photoperiod. The accumulation of light signals triggers a response in the leaves that is transduced to the shoot apical meristem. Considering that several components of the light-signaling pathway operate in the nucleus, we addressed the question of whether the floral transition affects nuclear processes such as chromatin compaction, involving large regions of the Arabidopsis genome. In the Arabidopsis nucleus, the pericentric heterochromatin and the inactive ribosomal genes are generally clustered into compact chromatin domains, so-called chromocenters (Fransz et al., 2002), whereas gene-rich regions, including active and inactive genes, are largely outside the chromocenters (Fransz et al., 2006). Unless stated otherwise, in this study we use the term heterochromatin to indicate compact chromatin or chromocenters. Previous studies have reported changes in the appearance of chromocenters under certain physiological, developmental conditions. For exam-

ple, changes in heterochromatin organization have been observed during seedling differentiation (Mathieu et al., 2003), ageing of rosette leaves (Tessadori et al., 2004), dedifferentiation to protoplasts (Tessadori et al., 2007) and pathogen infection (Pavet et al., 2006). In this study we demonstrate a transient large-scale reduction of compact chromatin over a short period (2–4 days) that corresponds specifically with the floral transition. Mutant analysis of photoperiod pathway components revealed no typical reduction in compact chromatin in plants lacking the bluelight photoreceptor CRY2, suggesting that CRY2 mediates light-signaling to chromatin organization during the floral transition. Results Mesophyll cells of rosette leaves display large-scale reorganization of heterochromatin domains prior to bolting The floral transition is marked by changes in the expression of hundreds of genes (Schmid et al., 2003). We addressed the question of whether this developmental switch involves major changes in large-scale chromatin organization. Parenchyma nuclei from inflorescences display 6–10 conspicuous chromocenters (Fransz et al., 2002). We observed a similar phenotype in nuclei from rosette leaves (Figure 1a). Up to 20% of the nuclei, however, displayed a phenotype with smaller or elongated chromocenters and more heterogeneous euchromatin (Figure 1b). Remarkably, the fraction of nuclei with this aberrant phenotype increased to 50% when visible flower buds appeared at the center of the rosette, which corresponds to stage 5.10 (Boyes et al., 2001). We therefore carried out a quantitative analysis of heterochromatin at different developmental stages around the floral transition. We measured the mean relative heterochromatin fraction (RHF), which is determined by the area and fluorescence intensity of all chromocenters in relation to the area and fluorescence intensity of the entire nucleus (Soppe et al., 2002; Tessadori et al., 2004). Four days before the appearance of the flower buds (stage 1.09) the RHF showed a sharp reduction in compact chromatin content (Figure 1c). We then established the percentage of nuclei with normal chromocenter appearance (i.e. number, shape and size), here denoted as the heterochromatin index (HX). The HX decreased from 70% to about 50% during the same period before bolting (Figure 1d). Apparently, the appearance of flower buds is accompanied by a large-scale decondensation of chromocenters. Moreover, within 3 days after bolting the RHF and the HX increased to normal values, indicating that the reduction of heterochromatin is transient. We addressed the question if the change in nuclear organization is dependent on the flowering time, and examined two accessions that differ from Col-0 in flowering

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

850 Federico Tessadori et al. Figure 1. Morphological changes of mesophyll nuclei and reduction in compact chromatin during the floral transition. (a) and (b) 4¢,6-Diamidino-2-phenylindole stained nuclei before the floral transition (a) and 2-4 days prior to bolting (b). Note the difference in chromocenter morphology. Scale bar = 5 lm. (c) and (d) Quantification of compact chromatin content by means of relative heterochromatin fraction (c) and heterochromatin index (d). Bars represent stages before (white) or after (gray) bolting. For each bar the data are mean  SEM of more than 30 nuclei (c) or more than 68 nuclei (d).

time. Ler flowers earlier (day 17), whereas Cvi-0 flowers more than 30 days after germination. We monitored the RHF value of Col-0, Ler and Cvi-0 plants before and during the floral transition (Figure 2). The results indicate that the

decrease in chromocenters corresponds with the transition to flowering irrespective of the flowering time. Strikingly, although Cvi-0 constitutively shows a lower content of condensed chromatin compared to Col-0 or Ler, it still

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

Chromatin reorganization during floral transition 851

Figure 2. The reduction of heterochromatin in three accessions. The relative heterochromatin fraction was measured in Col-0, Ler and Cvi-0 plants grown under long-day conditions before (white bars) and during the floral transition (gray bars). For each bar the data are mean  SEM of more than 15 nuclei.

displays a decrease in heterochromatin during flower transition (Figure 2). To investigate if heterochromatin decondensation is correlated with the transition to flowering, we delayed the floral transition in Col-0 plants by growing them under a shorter day regime (SD, 12-h light/12-h dark). These plants bolted 19 days later than plants grown under long-day conditions (LD, 16-h light/ 8-h dark). In leaves of SD-grown plants the RHF value decreased after 38 days, which is just before bolting, and recovered after the appearance of the first flower buds (stage 5.10; Figure 3a). The simultaneous delay in both bolting and heterochromatin reduction further indicates a strong correlation between heterochromatin reorganization and floral transition. We finally tested if chromatin reorganization can be promoted by transferring plants from a short-day to a long-day light regime. Such a transfer will induce the transition from a vegetative state to a reproductive state. Col-0 plants were grown under short-day conditions for 37 days and subsequently transferred to a long-day regime. Only 24 h after the transfer the mesophyll nuclei showed a sharp decrease in RHF value (Figure 3b). By 4 days after the transfer the plants produced visible flower buds and the mesophyll nuclei displayed a normal appearance. These observations confirm that the decrease in heterochromatin coincides with the period during which bolting is triggered. Both pericentromeric heterochromatin and gene-rich chromatin show large-scale decondensation during the floral transition The reduction of chromocenters indicates that certain chromosomal regions are no longer in heterochromatin

Figure 3. Heterochromatin fraction under short-day (SD) conditions. (a) Relative heterochromatin fraction (RHF) of plants grown in SD conditions (12-h light/12-h dark). FS: plants have developed a floral stem, but no flower is yet open. (b) RHF of plants grown in SD conditions for 37 days and then transferred to long-day (LD) conditions. Note the decrease in RHF only 24 h after transfer and the bolting after 96 h. Bars represent stages before (white) or after (gray) bolting. For each bar the data are mean  SEM of more than 30 nuclei.

domains. To determine which DNA sequences are involved in the reduction of heterochromatin, we applied fluorescence in situ hybridization (FISH) with probes of dispersed and tandemly arranged repeat sequences, known to localize in chromocenters (Fransz et al., 2002): (i) the 180-bp centromeric tandem repeat, (ii) the pericentromeric BAC F28D6, which contains numerous interspersed repeats, such as transposable elements present on all chromosomes, and (iii) tandemly arranged 5S and 45S rRNA sequences. All repeats are condensed and reside in chromocenters of mesophyll nuclei before the floral transition. During the floral transition the 180-bp tandem repeat and 45S rDNA sequences remained condensed in chromocenters (Figure 4a). In contrast, pericentromeric sequences, targeted by a 5S rDNA probe and BAC F28D6, were decondensed and dispersed. This decondensation is likely to be responsible for the reduction in heterochromatin and the decreased RHF. To find out whether euchromatic regions are also affected during the floral transition, we monitored the compaction state of an approximately 0.3-Mb segment in gene-rich chromatin. We applied FISH with the BAC contig F28A21, F13C5 and T18B16, which map to the long arm of chromosome 4,

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

852 Federico Tessadori et al.

Figure 4. Cytological localization of repetitive sequences, gene-rich regions and DNA methylation. (a) Fluorescence in situ hybridization (FISH) localization of repetitive sequences (right) and 4¢,6-diamidino-2-phenylindole (DAPI) counterstaining (left) in mesophyll nuclei during the floral transition. Note that although the pericentromeric repeat-rich F28D6 BAC (I, green) and 5S rDNA (II, green) are dispersed in nuclei during floral transition, the centromeric 180-bp (I, II and III, red) and the subtelomeric 45S rDNA (III, green) regions remain compact and colocalize with chromocenters. (b) FISH localization of the F28A21-F13C5-T18B16 BAC contig, covering approximately 0.3 Mb. F13C5 appears in red; T18B16 and F28A21 appear in green. The high compaction state prior to floral transition (I; arrows) is lost in nuclei during the 2–4-day period before bolting (II; arrows). Nuclei are counterstained with DAPI. (c) Fraction of nuclei with normal levels of heterochromatin (gray), and condensed FISH signals of BAC F13C5 (red) and T18B16-F28A21 (green) before and during the floral transition. Data are mean  SEM of 29 nuclei. (d) Immunolocalization of 5-methylcytosine (5-mC, right) counterstained with DAPI (left). (I) Prior to floral transition 5-mC signals are clustered at chromocenters. (II) A dispersed pattern of 5-mC is observed in nuclei with low chromocenter content during floral transition. Scale bars = 5 lm.

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

Chromatin reorganization during floral transition 853 and measured the frequency of relatively compact FISH signals before and during the floral transition (Figure 4b). Concomitant to the decrease in chromocenters, the occurrence of condensed BAC signals decreased by about 33% (Figure 4c), indicating that the decompaction of chromatin is not restricted to the heterochromatic sections of the genome. As chromocenters are rich in methylated DNA (Fransz et al., 2002) and because a 5-methyl-cytosine (5-mC) imprint is required to maintain a chromocenter state (Soppe et al., 2002), we examined leaf nuclei for changes in nuclear distribution of 5-mC. Col-0 nuclei displayed a dispersed pattern of 5-mC during the floral transition (Figure 4d), similar to the dispersion of the 5S rDNA and the transposon-rich BAC F28D6. This suggests that heavily methylated DNA sequences are responsible for the decrease of the chromocenters. Furthermore, we found no reduction in DNA methylation at centromeric repeats, indicating that no largescale decrease of DNA methylation has occurred during the floral transition (Figure S1). CO and FT do not affect heterochromatin reorganization, whereas the photoreceptor CRY2 does As the decrease in heterochromatin coincides with the floral transition and its timing is affected by the day length, we inferred that the reorganization of heterochromatin is under control of components of the photoperiod pathway. We therefore examined the RHF in mutants of CO, FT and CRY2. In the photoperiod pathway, the upregulation of CO in rosette leaves stimulates the production of an FT transcript that moves through the phloem to the shoot apex triggering flower development (Abe et al., 2005; Wigge et al., 2005). Both co-2 and ft-1 mutants are in the Ler background and have a late-flowering phenotype in long-day conditions (Bradley et al., 1997; Putterill et al., 1995). They bolt after 22 days when there are 11 rosette leaves (Figure 5b,c). This is 6 days later than in wild-type plants, which start flowering with six rosette leaves (Figure 5a). The RHF in wild-type Ler decreased sharply at stage 1.05 (Boyes et al., 2001), just prior to bolting, and increased during bolting (Figure 5a). Both the co-2 and the ft-1 mutants (Figure 5c) showed a similar sharp reduction in RHF, but 8 days later than in wild type, corresponding with the delay in flowering time. Apparently, CO and FT do not affect the reduction in heterochromatin during the floral transition. However, in comparison with the wild type, a shorter period of the transient heterochromatin reduction (2 days) prior to bolting was observed in the mutants. The delayed decrease of RHF in ft-1 and co-2 further underlines that heterochromatin reduction is related to the floral transition rather than to the age of the plant or the number of rosette leaves. CRY2, a component of the light-signaling pathway, is located in the nucleus (Mas et al., 2000) and controls floral transition by stabilizing CO (Valverde et al., 2004). We

Figure 5. Heterochromatin fraction in co-2 and ft-1 mutants. Relative heterochromatin fraction (RHF) of wild-type Ler (a), co-2 (b) and ft-1 (c) grown in long-day (LD) conditions. Developmental stages (Boyes et al., 2001) and age in days (in brackets) are indicated. Flowering is delayed in the mutants, but the decrease in RHF is still observed 2–4 days before bolting. Bars represent stages before (white) or after (gray) bolting. For each bar the data are mean  SEM of more than 30 nuclei.

examined the nuclear phenotype of the cry2 mutant in a Col-0 (cry2-1) and a Ler (fha-1) background. The two wildtype accessions flowered after 22 and 17 days, respectively, and showed a decrease in RHF during the floral transition. Flowering time for cry2-1 and fha-1 is delayed compared with their wild-type background accession (after 31 and 22 days, respectively). However, in contrast to the wild-type plants, both cry2 mutants displayed no decrease in the HX during floral transition (Figure 6a,b). We conclude that CRY2 is a major component of the light-signaling pathway leading to chromatin decondensation during the floral transition. Taken together with the results obtained in co-2 and ft-1, the data suggest that the heterochromatin reduction is downstream of CRY2, but does not depend on CO and FT activity.

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

854 Federico Tessadori et al.

Figure 6. Nuclear fraction with normal heterochromatin levels in null mutants of CRY2 during the floral transition. Heterochromatin index in cry2 mutant plants in a Col-0 background (a) and a Ler background, fha-1 (b). Plants were grown in long-day (LD) conditions. Bars represent stages before (white) or after (gray) bolting. For each bar the data are mean  SEM of more than 75 nuclei. The age of the plant in days is shown in brackets.

Discussion The floral transition is one of the major developmental switches in plants. Hundreds of genes are up- or downregulated when Arabidopsis is induced to flower (Schmid et al., 2003). Such an extensive change in the gene expression pattern is likely to involve major changes in chromatin state. Our study demonstrates a dramatic decondensation of pericentromeric and gene-rich chromatin in leaf mesophyll nuclei during the floral transition. The decrease in compact chromatin content is under control of a light-signaling pathway that involves the blue-light photoreceptor CRY2. Considering that, for an annual plant such as Arabidopsis, the floral transition is a fundamental and irreversible developmental event, a correctly established transition is crucial. Hence, all processes during the floral transition should be properly timed in order to initiate floral meristem development. The decondensation of large chromatin regions including gene-rich and pericentric regions may facilitate this important developmental switch.

Floral transition is referred to as the switch from vegetative to reproductive growth. However, although this developmental switch results in distinctive morphological features, such as the formation of flower primordia and the elongation of internodes (Hempel and Feldman, 1994; Koornneef and Peeters, 1997), its precise temporal boundaries have not been clearly defined. During vegetative growth the shoot apical meristem produces leaf primordia and associated axillary buds. Under certain environmental conditions that induce flowering, such as extended photoperiods, the shoot apical meristem produces floral primordia upon reception of a florigen signal. This signal is generated in the leaves and transported to the apical meristem. It is reasonable to define floral transition under long-day conditions as the lapse of time between the initiation of the florigen production in the leaves and the bolting of the immature flower buds. Of all possible components that make up the florigen signal so far only the FT gene product (mRNA and/or protein) has been identified (Huang et al., 2005). FT transcription is activated in the vascular tissue of leaves under long-day conditions by the flowering activator CO (An et al., 2004; Samach et al., 2000; Suarez-Lopez et al., 2001; Takada and Goto, 2003). The FT mRNA and/or protein must be transported to induce flowering. Within 6 h of induction in the leaves, significant FT mRNA levels are detectable in the shoot meristem (Huang et al., 2005). The expression of FT mRNA in the leaves is maintained for up to 3 days and leads to upregulation of floral identity genes such as AP1 and LFY. This time frame, between initiation of FT transcript production and the induction of floral identity genes, corresponds remarkably well with the period in which we observe the reduction of pericentric heterochromatin. It suggests that the transient state of reduced heterochromatin may be mechanistically related to the emission of FT gene product. In this study we report decondensation of pericentromeric heterochromatin and gene-rich chromatin during the floral transition. Therefore, we propose that decondensation of chromatin in flowering-induced plants may enhance the accessibility to DNA for the transcription machinery on a genome-wide scale. High CO levels necessary for FT stimulation are established by the coincidence of clock-driven high CO expression in conjunction with CO-stabilizing activities of the photoreceptors CRY2 and PHYA during the late afternoon (Valverde et al., 2004). Both PHYA and CRY2 are involved in the perception of the photoperiod. CRY2 is a nuclear protein that has been shown to associate with chromosomes (Cutler et al., 2000). CRY2 shows homology with DNA photolyase, but does not have photolyase activity. Virtually nothing is known about a direct functional interaction between CRY2 and chromatin. The cry2 mutant shows delayed flowering in LD conditions, whereas overexpression of CRY2 accelerates flowering under SD conditions (Guo et al., 1998). Our study

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

Chromatin reorganization during floral transition 855 has demonstrated that CRY2 is required for chromatin decondensation during the floral transition. The transient reduction of heterochromatin coincides with the floral transition except in the cry2 mutant, where heterochromatin levels remain stable. Notably, the correlation between absence of CRY2 and stable levels of heterochromatin (Fig. 6) is not only observed during the floral transition. We have found a dramatic reduction of heterochromatin in low-light treated wild-type plants, whereas cry2 mutants did not show this reduction (van Zanten et al., unpublished data ). Low light levels trigger a shade-avoidance response, which activates a number of processes, such as hyponastic growth (Ballare, 1999; Millenaar et al., 2005), leading to morphological changes. Apparently, under certain light conditions that promote changes of the nuclear program, CRY2 triggers reorganization of chromatin. The decondensation event is not a prerequisite for flowering, as the cry2 mutant shows high levels of heterochromatin during the floral transition, yet it produces flowerbuds, albeit delayed compared with the wild type. We therefore propose that chromatin decondensation is under control of a light-signaling mechanism involving CRY2, and may facilitate CRY2mediated responses to environmental changes. Heterochromatin reduction coincides with the floral transition in both LD and SD conditions. The perception of LD or SD involves the activity of the photoreceptor CRY2 (Guo et al., 1998). CRY2 stabilizes CO under LD, as its level fluctuates in SD (El-Din El-Assal et al., 2001). Consequently, the floral transition is delayed and under control of the gibberellin pathway (Reeves and Coupland, 2001). Coincidently, the heterochromatin level decreases prior to bolting. The same happens to co and ft plants grown in LD. However, in the absence of CRY2, the floral transition is delayed and under control of the gibberellin pathway, but reduction of heterochromatin does not take place. This implies that chromatin decondensation depends on CRY2 action, but is not affected by CO or FT activity. Conversely, the activities of CO and FT are unlikely to take place downstream of heterochromatin decondensation, as they are expressed before the floral transition. Therefore, we propose that CRY2 promotes two independent processes: CO-mediated flowering and reorganization of chromatin. The two events coincide at the transition from vegetative to reproductive growth. This explains why the reduction in heterochromatin occurs prior to bolting and is absent in the cry2 mutant. How CRY2 may regulate these processes is not clear. It is likely that E3 ligase complexes, such as COP1, are involved. A reciprocal effect has been reported between CRY2 and COP1. CRY2 is unstable and rapidly degraded in blue light by COP1, whereas COP1 activity in turn is negatively regulated by CRY2 (Shalitin et al., 2002; Wang et al., 2001). The latter is carried out via interaction between the WD40 domain of COP1 and the C-terminal domain of CRY2 (Wang et al., 2001). How CO is stabilized by CRY2 is not yet clear.

CO is degraded via a mechanism that involves SPA proteins (Laubinger et al., 2006). Proteins of the SPA family are known to act with COP1 in the degradation of transcription factors of photomorphogenesis. Accordingly, CRY2 might stabilize CO via negatively regulating its degradation by a COP1-like complex. Whether COP1 is required to degrade CO and if the same mechanism controls chromatin organization remains to be elucidated.

Experimental procedures Plant material Arabidopsis accessions Col-0 (N1092); Ler (NW20) and Cvi-0 (N902) were used for this study. The co-2, ft-1, cry2-1 and fha-1 mutants were kindly provided by M. Koornneef (MPIZ, Cologne, Germany). The co-2, ft-1 and fha-1 mutants are in the Ler background. The cry21 mutant is in the Col-0 background. Plants were grown under white fluorescent light (70–90 lmol m)2 sec)1) in photoperiods of 16-h light/8-h dark or 12-h light/12-h dark. Temperature (22C) and humidity (70%) remained constant. For experiments, tissue from young rosette leaves, 5–15 mm in size, was used. The first two juvenile leaves were excluded. The developmental stage of the plants was assessed according to the method described by Boyes et al., 2001.

Fluorescence in situ hybridization (FISH) The DNA fragments used as FISH probes were labeled with either biotin- or digoxigenin-Nick Translation Mixes (Roche, http:// www.roche.com). Plasmid pAL1 (Martinez-Zapater et al., 1986) was used to detect the 180-bp centromeric tandem repeat. BAC F28D6 (DDBJ/EMBL/GenBank accession No. AF147262) in pBeloBAC-Kan vector was obtained from the Nottingham Arabidopsis Stock Centre (NASC, http://nasc.nott.ac.uk) . The AGAMOUS locus was localized using BAC F13C5 (DDBJ/EMBL/GenBank accession No. AL021711; 119.1 kb) and the two adjacent BACs T18B16 (DDBJ/EMBL/GenBank accession No. AL021687; 96.5 kb) and F28A21 (DDBJ/EMBL/GenBank accession No. AL025526; 94.3 kb). FISH experiments were carried out as described in Schubert et al., (2001) and detected with antibodies Avidin-Texas Red (Vector Laboratories, http://www.vectorlabs.com) and goat anti-Avidin (Vector Laboratories) for biotin-labeled probes; mouse anti-digoxigenin (Roche), rabbit anti-mouseFITC (Sigma-Aldrich, http:// www.sigmaaldrich.com) and goat anti-rabbitAlexa488 (Molecular Probes, http://www.invitrogen.com) for the detection of digoxigenin-labeled probes. Prior to observation, slides were counterstained with 4¢,6-diamidino-2-phenylindole (DAPI, 2 lg ml)1; Roche) in Vectashield (Vector Laboratories).

5-Methylcytosine detection Preparations were dried at 60C for 30 min, treated with RNAse (10 lg ml)1 in 2X SSC ; Roche) for 60 min at 37C, rinsed twice for 5 min in 2X SSC and 5 min in PBS (10 mM sodium phosphate, pH 7.0, 143 mM NaCl), fixed for 10 min in 1% formaldehyde in PBS at room temperature, rinsed twice for 5 min in PBS, dehydrated by successive 1-min baths in 70, 90 and 100% ethanol and then air dried. Denaturation was carried out by adding 50 ll HB50 (50% formamide in 1X SSC) and heating at 80C for 2 min. The slides

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

856 Federico Tessadori et al. where then washed in 70% ice-cold ethanol and dehydrated by successive ethanol baths. Blocking of the slides was carried out for 1 h in 1% BSA in PBS and, after three 5-min washes in TNT (1 M Tris/ HCl, pH 8.0, 1 M NaCl, 0.5% Tween 20), incubation with the mouse antibody against 5-methylcytosine (Eurogentec, http://www. eurogentec.be) was carried out at room temperature overnight. Detection of the antibody was carried out with the same antibodies used for FISH with digoxigenin-labeled probes.

Image acquisition and processing Image acquisition was carried out on an Olympus BX60 microscope (Olympus, http://www.olympus-global.com) with filters for DAPI, Texas Red and fluorescein isothiocyanate (FITC). Pictures were captured with a charge-coupled device (CCD) camera (Coolsnap FX; Photometrics, http://www.photomet.com). The images were digitally processed using ADOBE PHOTOSHOP (Adobe, http://www.adobe.com).

Measurement of RHF and HX Greyscale images were analyzed with OBJECTIMAGE freeware (http:// simon.bio.uva.nl/object-image.html). The RHF was calculated from mesophyll nuclei as described by Soppe et al., 2002; Tessadori et al., 2004; and is defined as the area and fluorescence intensity of chromocenters in relation to the area and fluorescence intensity of the entire nucleus. The HX is defined as the percentage of normal nuclei showing a relatively high content of compact chromatin, represented by 6–10 round or oval conspicuous chromocenters (Fransz et al., 2002, 2003). Nuclei containing fewer chromocenters, smaller chromocenters or chromocenters with vague contours are considered abnormal. Only mesophyll nuclei are used to measure RHF or HX. Nuclei from epidermis, vascular tissue or endoreduplicated cells are not included in the measurements.

Acknowledgements We thank Maarten Koornneef (MPIZ, Cologne, Germany) for valuable discussions and Corrie Hanhart (WVR, Wageningen, Netherlands) for the seeds of the co-2, ft-1, cry2-1 and fha-1 mutants used in this work.

Supplementary Material The following supplementary material is available for this article online: Figure S1. Southern blot analysis of DNA methylation. This material is available as part of the online article from http:// www.blackwell-synergy.com

References Abe, M., Kobayashi, Y., Yamamoto, S., Daimon, Y., Yamaguchi, A., Ikeda, Y., Ichinoki, H., Notaguchi, M., Goto, K. and Araki, T. (2005) FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex. Science, 309, 1052–1056. An, H.L., Roussot, C., Suarez-Lopez, P. et al. (2004) CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development, 131, 3615–3626. Ballare, C.L. (1999) Keeping up with the neighbours: phytochrome sensing and other signalling mechanisms. Trends Plant Sci. 4, 201.

Bastow, R., Mylne, J.S., Lister, C., Lippman, Z., Martienssen, R.A. and Dean, C. (2004) Vernalization requires epigenetic silencing of FLC by histone methylation. Nature, 427, 164–167. Boss, P.K., Bastow, R.M., Mylne, J.S. and Dean, C. (2004) Multiple pathways in the decision to flower: enabling, promoting, and resetting. Plant Cell, 16(Suppl.), S18–S31. Boyes, D.C., Zayed, A.M., Ascenzi, R., McCaskill, A.J., Hoffman, N.E., Davis, K.R. and Gorlach, J. (2001) Growth stage-based phenotypic analysis of arabidopsis: a model for high throughput functional genomics in plants. Plant Cell, 13, 1499–1510. Bradley, D., Ratcliffe, O., Vincent, C., Carpenter, R. and Coen, E. (1997) Inflorescence commitment and architecture in Arabidopsis. Science, 275, 80–83. Cutler, S.R., Ehrhardt, D.W., Griffitts, J.S. and Somerville, C.R. (2000) Random GFP::cDNA fusions enable visualization of subcellular structures in cells of Arabidopsis at a high frequency. Proc. Natl Acad. Sci. USA, 97, 3718–3723. El-Din El-Assal, S., Alonso-Blanco, C., Peeters, A.J., Raz, V. and Koornneef, M. (2001) A QTL for flowering time in Arabidopsis reveals a novel allele of CRY2. Nat. Genet. 29, 435–440. Fransz, P., De Jong, J.H., Lysak, M., Castiglione, M.R. and Schubert, I. (2002) Interphase chromosomes in Arabidopsis are organized as well defined chromocenters from which euchromatin loops emanate. Proc. Natl Acad. Sci. USA, 99, 14584–14589. Fransz, P., Soppe, W. and Schubert, I. (2003) Heterochromatin in interphase nuclei of Arabidopsis thaliana. Chromosome Res. 11, 227–240. Fransz, P., ten Hoopen, R. and Tessadori, F. (2006) Composition and formation of heterochromatin in Arabidopsis thaliana. Chromosome Res. 14, 71–82. Gaudin, V., Libault, M., Pouteau, S., Juul, T., Zhao, G., Lefebvre, D. and Grandjean, O. (2001) Mutations in LIKE HETEROCHROMATIN PROTEIN 1 affect flowering time and plant architecture in Arabidopsis. Development, 128, 4847–4858. Gendall, A.R., Levy, Y.Y., Wilson, A. and Dean, C. (2001) The VERNALIZATION 2 gene mediates the epigenetic regulation of vernalization in Arabidopsis. Cell, 107, 525–535. Guo, H., Yang, H., Mockler, T.C. and Lin, C. (1998) Regulation of flowering time by Arabidopsis photoreceptors. Science, 279, 1360–1363. He, Y. and Amasino, R.M. (2005) Role of chromatin modification in flowering-time control. Trends Plant Sci. 10, 30–35. He, Y., Michaels, S.D. and Amasino, R.M. (2003) Regulation of flowering time by histone acetylation in Arabidopsis. Science, 302, 1751–1754. He, Y., Doyle, M.R. and Amasino, R.M. (2004) PAF1-complex-mediated histone methylation of FLOWERING LOCUS C chromatin is required for the vernalization-responsive, winter-annual habit in Arabidopsis. Genes Dev. 18, 2774–2784. Hempel, F.D. and Feldman, L.J. (1994) Bidirectional Inflorescence Development in Arabidopsis thaliana – acropetal Initiation of Flowers and Basipetal Initiation of Paraclades. Planta, 192, 276– 286. Hepworth, S.R., Valverde, F., Ravenscroft, D., Mouradov, A. and Coupland, G. (2002) Antagonistic regulation of flowering-time gene SOC1 by CONSTANS and FLC via separate promoter motifs. EMBO J. 21, 4327–4337. Huang, T., Bohlenius, H., Eriksson, S., Parcy, F. and Nilsson, O. (2005) The mRNA of the Arabidopsis gene FT moves from leaf to shoot apex and induces flowering. Science, 309, 1694–1696. Johanson, U., West, J., Lister, C., Michaels, S., Amasino, R. and Dean, C. (2000) Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science, 290, 344–347.

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857

Chromatin reorganization during floral transition 857 Koornneef, M. and Peeters, A.J.M. (1997) Floral transition mutants in Arabidopsis. Plant Cell Environ. 20, 779–784. Kotake, T., Takada, S., Nakahigashi, K., Ohto, M. and Goto, K. (2003) Arabidopsis TERMINAL FLOWER 2 gene encodes a heterochromatin protein 1 homolog and represses both FLOWERING LOCUS T to regulate flowering time and several floral homeotic genes. Plant Cell Physiol. 44, 555–564. Larsson, A.S., Landberg, K. and Meeks-Wagner, D.R. (1998) The TERMINAL FLOWER2 (TFL2) gene controls the reproductive transition and meristem identity in Arabidopsis thaliana. Genetics, 149, 597–605. Laubinger, S., Marchal, V., Gentilhomme, J., Wenkel, S., Adrian, J., Jang, S., Kulajta, C., Braun, H., Coupland, G. and Hoecker, U. (2006) Arabidopsis SPA proteins regulate photoperiodic flowering and interact with the floral inducer CONSTANS to regulate its stability. Development, 133, 3213–3222. Levy, Y.Y., Mesnage, S., Mylne, J.S., Gendall, A.R. and Dean, C. (2002) Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science, 297, 243–246. Libault, M., Tessadori, F., Germann, S., Snijder, B., Fransz, P. and Gaudin, V. (2005) The Arabidopsis LHP1 protein is a component of euchromatin. Planta, 222, 910–925. Liu, J., Yu, J., McIntosh, L., Kende, H. and Zeevaart, J.A. (2001) Isolation of a CONSTANS ortholog from Pharbitis nil and its role in flowering. Plant Physiol. 125, 1821–1830. Martinez-Garcia, J.F., Virgos-Soler, A. and Prat, S. (2002) Control of photoperiod-regulated tuberization in potato by the Arabidopsis flowering-time gene CONSTANS. Proc. Natl Acad. Sci. USA, 99, 15211–15216. Martinez-Zapater, J.M., Estelle, M.A. and Somerville, C.R. (1986) A high repeated DNA sequence in Arabidopsis thaliana. Mol. Gen. Genet. 204, 417–423. Mas, P., Devlin, P.F., Panda, S. and Kay, S.A. (2000) Functional interaction of phytochrome B and cryptochrome 2. Nature, 408, 207–211. Mathieu, O., Jasencakova, Z., Vaillant, I., Gendrel, A.V., Colot, V., Schubert, I. and Tourmente, S. (2003) Changes in 5S rDNA chromatin organization and transcription during heterochromatin establishment in Arabidopsis. Plant Cell, 15, 2929–2939. Millenaar, F.F., Cox, M.C., van Berkel, Y.E., Welschen, R.A., Pierik, R., Voesenek, L.A. and Peeters, A.J. (2005) Ethylene-induced differential growth of petioles in Arabidopsis. Analyzing natural variation, response kinetics, and regulation. Plant Physiol. 137, 998–1008. Mylne, J.S., Barrett, L., Tessadori, F., Mesnage, S., Johnson, L., Bernatavichute, Y.V., Jacobsen, S.E., Fransz, P. and Dean, C. (2006) LHP1, the Arabidopsis homologue of HETEROCHROMATIN PROTEIN1, is required for epigenetic silencing of FLC. Proc. Natl Acad. Sci. USA, 103, 5012–5017. Pavet, V., Quintero, C., Cecchini, N.M., Rosa, A.L. and Alvarez, M.E. (2006) Arabidopsis displays centromeric DNA hypomethylation and cytological alterations of heterochromatin upon attack by pseudomonas syringae. Mol. Plant Microbe Interact. 19, 577–587. Putterill, J., Robson, F., Lee, K., Simon, R. and Coupland, G. (1995) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell, 80, 847–857. Putterill, J., Laurie, R. and Macknight, R. (2004) It’s time to flower: the genetic control of flowering time. Bioessays, 26, 363–373. Reeves, P.H. and Coupland, G. (2001) Analysis of flowering time control in Arabidopsis by comparison of double and triple mutants. Plant Physiol. 126, 1085–1091.

Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F. and Coupland, G. (2000) Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science, 288, 1613–1616. Schmid, M., Uhlenhaut, N.H., Godard, F., Demar, M., Bressan, R., Weigel, D. and Lohmann, J.U. (2003) Dissection of floral induction pathways using global expression analysis. Development, 130, 6001–6012. Schubert, I., Fransz, P.F., Fuchs, J. and de Jong, J.H. (2001) Chromosome painting in plants. Methods Cell Sci. 23, 57–69. Shalitin, D., Yang, H., Mockler, T.C., Maymon, M., Guo, H., Whitelam, G.C. and Lin, C. (2002) Regulation of Arabidopsis cryptochrome 2 by blue-light-dependent phosphorylation. Nature, 417, 763–767. Sheldon, C.C., Burn, J.E., Perez, P.P., Metzger, J., Edwards, J.A., Peacock, W.J. and Dennis, E.S. (1999) The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. Plant Cell, 11, 445–458. Soppe, W.J., Jasencakova, Z., Houben, A., Kakutani, T., Meister, A., Huang, M.S., Jacobsen, S.E., Schubert, I. and Fransz, P.F. (2002) DNA methylation controls histone H3 lysine 9 methylation and heterochromatin assembly in Arabidopsis. EMBO J. 21, 6549– 6559. Suarez-Lopez, P., Wheatley, K., Robson, F., Onouchi, H., Valverde, F. and Coupland, G. (2001) CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature, 410, 1116–1120. Sung, S. and Amasino, R.M. (2004) Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature, 427, 159–164. Takada, S. and Goto, K. (2003) Terminal flower2, an Arabidopsis homolog of heterochromatin protein1, counteracts the activation of flowering locus T by constans in the vascular tissues of leaves to regulate flowering time. Plant Cell, 15, 2856–2865. Tessadori, F., van Driel, R. and Fransz, P. (2004) Cytogenetics as a tool to study gene regulation. Trends Plant Sci. 9, 147–153. Tessadori, F., Chupeau, M.-C., Chupean, Y., Knip, M., Germann, S., van Driel, R., Fransz, P. and Gaudin, V. (2007) Large-scale dissociation and sequential reassembly of pericentric heterochromatin dedifferentiated Arabidopsis cells. J. Cell Sci, in press. Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach, A. and Coupland, G. (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science, 303, 1003–1006. Vince-Prue, D. and Gressel, J. (1985) Pharbitis nil. in Handbook of Flowering, Vol. IV (Halevy, A., ed.). Boca Raton, FL: CRC Press, pp. 47–81. Wang, H., Ma, L.G., Li, J.M., Zhao, H.Y. and Deng, X.W. (2001) Direct interaction of Arabidopsis cryptochromes with COP1 in light control development. Science, 294, 154–158. Wigge, P.A., Kim, M.C., Jaeger, K.E., Busch, W., Schmid, M., Lohmann, J.U. and Weigel, D. (2005) Integration of spatial and temporal information during floral induction in Arabidopsis. Science, 309, 1056–1059. Yanovsky, M.J. and Kay, S.A. (2002) Molecular basis of seasonal time measurement in Arabidopsis. Nature, 419, 308–312. Zeevaart, J.A.D. (1976) Physiology of flower formation. Annu. Rev. Plant Physiol. 27, 321–348.Accession numbers: co-2, AT5 G15840; seed stock, CS55; ft-1, AT1 G65480; seed stock, CS56; cry2-1, AT1 G04400; seed stock, CS3732; fha-1, AT1 G04400; seed stock, CS108.

ª 2007 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 50, 848–857