This is the accepted version of the article: Martínez-García, Jaime F.; Moreno Romero, Jordi. «Shedding light on the chromatin changes that modulate shade responses». Physiologia Plantarum, Vol. 169, Issue 3 (July 2020), p. 407-417. DOI 10.1111/ppl.13101 This version is avaible at https://ddd.uab.cat/record/231007 under the terms of the license 1 Shedding light on the chromatin changes that modulate shade responses 2 3 Jaime F. Martínez-Garcíaa,b and Jordi Moreno-Romeroa,* 4 5 a 6 Centre for Research in Agricultural Genomics (CRAG), CSIC-IRTA-UAB-UB, Cerdanyola del 7 Vallès, Campus UAB, 08193-Barcelona, Spain. b 8 Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys 23, 9 08010-Barcelona, Spain. 10 11 Correspondence * 12 Corresponding author, 13 e-mail: jordi.moreno@cragenomica.es 14 15 16 Perception of vegetation proximity or plant shade informs of potential competition for resources 17 by the neighboring vegetation. As vegetation proximity impacts on both light quantity and 18 quality, perception of this cue by plant photoreceptors reprograms development to result in 19 responses that allow plants to compete with the neighboring vegetation. Developmental 20 reprogramming involves massive and rapid changes in gene expression, with the concerted 21 action of photoreceptors and downstream transcription factors. Changes in gene expression can 22 be modulated by epigenetic processes that alter chromatin compaction, influencing the 23 accessibility and binding of transcription factors to regulatory elements in the DNA. However, 24 little is known about the epigenetic regulation of plant responses to the proximity of other 25 plants. In this manuscript, we review what is known about plant shade effects on chromatin 26 changes at the cytological level, that is, changes in nuclear morphology and high order 27 chromatin density. We address which are the specific histone post-transcriptional modifications 28 that have been associated with changes in shade-regulated gene expression, such as histone 29 acetylation and histone methylation. Furthermore, we explore the possible mechanisms that 30 integrate shade signaling components and chromatin remodelers to settle epigenetic marks at 31 specific loci. This review aims to be a starting point to understand how a specific environmental 32 cue, plant shade, integrates with chromatin dynamics to implement the proper acclimation 33 responses. 34 35 Abbreviations – B, blue; COP1, CONSTITUTIVE PHOTOMORPHOGENIC 1; cry2, 36 cryptochrome 2; DAPI, 4′,6-diamidino-2-phenylindole; DET1, DEETIOLATED1; FR, far-red; 37 GCN5, GENERAL CONTROL NONDEREPRESSIBLE 5; HAT, histone acetyltransferases; 38 HAF2, HISTONE ACETYLTRANSFERASE OF THE TAFII250 FAMILY 2; HAG4, 39 HISTONE ACETYLTRANSFERASE OF THE GNAT FAMILY 4; HAM1, HISTONE 40 ACETYLTRANSFERASE OF THE MYST FAMILY 1; HDAC, histone deacetylase; HDA6, 41 histone deacetylase 6; HDA15, histone deacetylase 15; HDA19, histone deacetylase 19; HDAC, 42 histone deacetylases; HY5, ELONGATED HYPOCOTYL 5; IAA3, INDOLE-3-ACETIC 43 ACID INDUCIBLE 3; MRG2, MORF RELATED GENE 2; PAR, photosynthetic active 44 radiation; PHYB, PHYTOCHROME B; PIF, PHYTOCHROME INTERACTING FACTOR; R, 45 red; PRE1, PACLOBUTRAZOL RESISTANCE 1; RBCS-1A, RIBULOSE BISPHOSPHATE 46 CARBOXYLASE SMALL CHAIN 1A; SDG8, SET DOMAIN GROUP 8; YUC8, YUCCA8. 47 48 49 50 Light not only provides energy for photosynthesis but also information to adapt plant growth 51 and development to a variable environment. In the high planting densities found in natural and 52 agronomic settings, light availability to drive photosynthesis can be strongly reduced and 53 compromised by the close proximity of other plants. Indeed, photosynthetic tissues absorb 54 specifically blue (B, 400-500 nm) and red light (R, around 660-670 nm), thus photosynthetic 55 active radiation (PAR, from 400 to 700 nm) of sunlight found below a plant canopy (i.e., plant 56 shade) is strongly impoverished in these colors of the spectrum, while most far-red light (FR, 57 around 720-730 nm) is transmitted or reflected. Consequently, the resulting R to FR ratio 58 (R:FR) that impacts plants growing underneath a vegetation canopy is also strongly reduced. In 59 contrast with plant or canopy shade, proximity of vegetation does not decrease the amount of 60 light but specifically reflects FR, hence also lowers the R:FR that impacts in a non-shaded 61 neighboring plant. Therefore, while both conditions result in a decrease in the R:FR, only 62 canopy shade decreases the amount of light. In this review we will refer to both low light and 63 low R:FR as components of plant shade. 64 65 Reduction of the R:FR that accompanies the proximity of vegetation in nature acts as an early 66 and reliable cue of the presence of vegetation that potentially might compete for light and other 67 resources. Early detection of this signal is key to plant survival and triggers acclimation 68 responses to optimize plant development. Among these responses were found the promotion of 69 stem or petiole elongation and hyponastic growth to outgrow neighboring plants. The low R:FR 70 ratio and the reduction of PAR light (low light) mediates the activity (amount, conformation and 71 localization) of the phytochromes, the photoreceptors able to sense R and FR wavelengths. 72 Under low light, the B-sensing cryptochromes also participate in the perception. Both types of 73 photoreceptors mediate rapid changes in gene expression and developmental processes (Yu et 74 al. 2010, Casal 2012, Roig-Villanova and Martinez-Garcia 2016). 75 76 There is a synergistic interaction between low B and low R:FR, as the combination of both 77 factors has more severe effects than the single ones (de Wit et al. 2016). Because of the 78 anticipatory and reliable nature of the low R:FR signal indicating the presence of vegetation 79 proximity, how phytochrome signaling works has received significant attention. Based mostly 80 on studies using Arabidopsis thaliana, it has been established that low R:FR-triggered 81 inactivation of phytochromes decreases their interaction with PHYTOCHROME 82 INTERACTING FACTORs (PIFs), a group of bHLH transcription factors. This results in the 83 accumulation and/or activation of PIFs to regulate the expression of several light-responsive 84 genes. Among the eight PIFs identified, the most relevant in regulating shade responses is PIF7, 85 although PIF4 and PIF5 also participate at a certain extent (Li et al. 2012). Accumulation and/or 86 activation of these PIFs in low R:FR releases the transcription of hundreds of genes that trigger 87 different aspects of the shade developmental program. The best known of these responses is 88 hypocotyl elongation, which is promoted by the action of PIFs, and takes place because of rapid 89 and reversible changes in the expression of dozens of genes. 90 91 Although many of the genetic components that regulate the shade response of the hypocotyl 92 have been elucidated, little is known about how this process is transcriptionally and 93 epigenetically orchestrated. To address this point, it is necessary to understand not only how 94 chromatin organizes the genetic material but also how it is regulated and whether the 95 environmental cues further modulate this regulation. 96 97 Chromatin is a dynamic structure that responds to environmental changes 98 99 Chromatin is a nuclear multicomponent complex composed of DNA, RNA and proteins. The 100 basic unit of chromatin is the nucleosome, a structure conserved in all eukaryotes that is formed 101 by DNA wrapped around an octamer of histone 3 (H3), H4, H2A and H2B, sealed by the H1 102 linker histone. Chromatin can have different degrees of compaction that regulates the 103 accessibility of genes to the transcriptional machinery. How this occurs is basic in order to 104 understand transcriptional regulation. 105 106 Some of the most important mechanisms that regulate changes in chromatin organization 107 involve DNA methylation and histone post-translational modifications and variants, which are 108 highly diverse and dynamic. The various modifications in either DNA or histones are 109 commonly referred to as epigenetic marks. It is assumed that different combinations of marks, 110 or epigenetic status, modulate chromatin compaction impacting either positively or negatively 111 on transcription. When the environment changes, chromatin needs to be modified and 112 restructured to translate the environmental signals into an acclimation response at the 113 transcriptional level (Probst and Mittelsten Scheid 2015). Therefore, the ability of rapidly and 114 reversibly altering the epigenetic status could be a key component to plant plasticity to respond 115 to the environment. 116 117 Chromatin-level regulation has been reported as an essential component of light signaling 118 (Barneche et al. 2014, Kim et al. 2015, Perrella and Kaiserli 2016). Little attention has been 119 given, however, to the chromatin dynamics associated to shade (low light and/or low R:FR) 120 responses. The purpose of this review is to address the current state of the art regarding 121 epigenetic regulation in shade response, with specific focus in the model plant Arabidopsis 122 thaliana. Two different aspects will be considered: chromatin compaction at the nuclear level 123 and histone modifications that affect specific gene expression. 124 125 Nucleus structure changes after low light and low R:FR exposure 126 127 As mentioned, chromatin presents different compaction levels. From this point of view, 128 chromatin can be classified in two major domains: euchromatin and heterochromatin. Each of 129 these chromatin states are enriched in specific histone variants and post-translational 130 modifications to form either transcriptionally active (euchromatin) or silent (heterochromatin) 131 chromatin domains (Fransz et al. 2003). Heterochromatin can be distinguished from 132 euchromatin in Arabidopsis nuclei by analyzing the chromatin density in the interphase nuclei 133 after DNA counterstaining with DAPI (4′,6-diamidino-2-phenylindole), reflecting that they 134 conform large-scale compartments (Fig. 1A). Heterochromatin clusters in highly stained 135 speckles called chromocenters, which are formed by the centromeric and pericentromeric 136 regions, satellite repeats, ribosomal DNA genes and nucleolar organizer regions (NORs) 137 repetitive sequences (Del Prete et al. 2014, Simon et al. 2015). Less condensed loops of active 138 euchromatin emanate from the chromocenters (Fransz et al. 2002), which are enriched in 139 transcriptionally active genes. Euchromatic regions are poorly stained with DAPI (Fig. 1A). 140 141 The two contrasting chromatin states are not static but highly dynamic, with their degree of 142 compaction changing under certain conditions. Quantifying the size and number of 143 chromocenters allows the assessment of wide-range chromatin density that correlates with 144 chromatin state (Tessadori et al. 2007). Using this approach, chromatin decondensation has been 145 detected during developmental switches and also upon environmental changes or stress, 146 including dark-to-light transitions (Exner and Hennig 2008, Bourbousse et al. 2015, Probst and 147 Mittelsten Scheid 2015). Regarding the exposure of seedlings to shade, either low light or low 148 R:FR results in a reduction of chromatin compaction (chromatin decondensation) assessed by a 149 decrease in the amount and size of chromocenters (Tessadori et al. 2009) (Fig. 1B). Maximum 150 decondensation in response to low light is reached at 4 days after the treatment. The reversion of 151 this state once the plant is shifted to normal light levels takes similar times (van Zanten et al. 152 2010). It is unknown, however, if the rate of chromatin decondensation observed in response to 153 low R:FR is similar to the one caused after exposure to low light, but the chromocenter kinetics 154 in other light processes (such as dark-to-light transitions) suggests that they will have a similar 155 behavior. 156 157 To find regulators of chromatin density in different light environment, 21 Arabidopsis 158 accessions from different locations and habitats were used. This study showed that compaction 159 of chromatin appeared to be dependent on light intensity, as the southernmost Arabidopsis 160 accession Cvi-0 presents the lowest chromatin compaction (Tessadori et al. 2009), a finding that 161 agrees with independent observations indicating that accessions closer to the equator are less 162 sensitive to light (Maloof et al. 2001). QTL mapping using the Cvi-0 and Ler accessions, which 163 present different levels of chromatin compaction, identified PHYTOCHROME B (PHYB) and 164 HISTONE DEACETYLASE-6 (HDA6) genes as positive regulators of light-dependent chromatin 165 compaction (Fig. 1B). Accordingly, phyB mutants display lower chromatin density than wild- 166 type plants in similar light conditions (Tessadori et al. 2009). Interestingly, phyB reduces 167 nucleus size (area and perimeter) independent of its positive role in heterochromatin 168 condensation (chromocenter formation and/or maintenance) (Snoek et al. 2017). Therefore, 169 phyB has a central role in the coregulation of the nucleus phenotype in low light. So far, the 170 effect of low R:FR on nuclear size has not been tested. Nevertheless, considering the decrease in 171 chromocenter number induced by low R:FR, and the fact that a clear positive correlation 172 between nucleus size (area and perimeter) and increased chromocenter values has been 173 demonstrated (Snoek et al. 2017), it is very likely that nuclear size is also increased under low 174 R:FR. 175 176 Besides phyB, the blue-light receptor cryptochrome 2 (cry2) has also been found to participate 177 in the low-light reduced chromatin compaction (van Zanten et al. 2010). The mechanisms of 178 action of phyB and cry2 photoreceptors could be direct, as both are nuclear proteins known to 179 interact (Mas et al. 2000), or indirect, as phyB controls cry2 protein levels (van Zanten et al. 180 2010). The latter authors propose a model where phyB and cry2 control a chromatin remodeler 181 complex responsible for chromatin compaction, in which CONSTITUTIVE 182 PHOTOMORPHOGENIC 1 (COP1) might participate (Fig. 1B). COP1 is a member of the E3 183 ligase complex that interacts with cryptochromes and phytochromes (Yi and Deng 2005). 184 COP1, together with DE-ETIOLATED 1 (DET1), maintain chromatin decondensation in 185 etiolated cotyledons (Bourbousse et al. 2015); after light exposure chromatin condensates. The 186 ubiquitination machinery of this protein complex could be important for the cell fate either by 187 inducing chromatin changes, as DET1 binds to H2B (Benvenuto et al. 2002), or reducing the 188 stability of transcription factors. However, there is no information available on whether cop1 or 189 det1 mutants have defective nuclear phenotypes in shade. Another component of this 190 hypothetical chromatin remodeler complex controlled by photoreceptors can be HDA6 (Fig. 191 1B), which was also found in the same QTL analysis as phyB (Tessadori et al. 2009). HDA6 is 192 involved in histone deacetylation, which is expected to compact chromatin (Fig. 2A). 193 Consistently, the compaction of heterochromatin in the hda6 mutant is reduced. Importantly, the 194 hda6 mutant phenotype can be partially restored by higher irradiance (Tessadori et al. 2009), 195 that is sensed by cryptochromes and phytochromes. 196 197 Besides histone acetylation, many other epigenetic processes have been related with large-scale 198 chromatin relaxation. Recently, the linker histone H1 has been also related to regulation of 199 several developmental transitions, e.g., from vegetative to flowering development (Rutowicz et 200 al. 2019). Interestingly, the H1.3 variant is stress inducible. After 3-4 days of low light 201 exposure, H1.3 protein accumulates, and within 1-2 days after moving plants back to standard 202 light conditions, H1.3 expression returns to normal levels, matching the timing with the low- 203 light induced chromatin decondensation. Moreover, the levels of low light-induced H1.3 204 expression in the cry2 and phyB mutants are slightly enhanced (Rutowicz et al. 2015). These 205 results make H1.3 one of the possible candidates linking chromatin condensation and 206 photoreceptor activity (Fig. 1B). 207 208 We still have a lot to understand on chromocenter dynamics and higher order chromatin 209 compaction. In addition, it is unclear how sensing environmental changes through phyB and 210 cryptochromes integrate in these processes. 211 212 Histone marks at specific loci as a mean to control shade-regulated gene expression 213 214 Chromatin changes in specific genes can be studied by analyzing the occurrence of epigenetic 215 marks in determined regions of the genome, such as those corresponding to genes involved in 216 the shade-response regulation. Among the most important epigenetic marks are DNA 217 methylation and histone post-translational modifications. We will discuss in here changes in 218 histone modifications, as so far, only these marks have been related to shade responses. In 219 addition to controlling the higher order chromatin structure, as addressed in the previous 220 section, the presence or absence of specific histone marks at the locus scale can also be directly 221 linked to gene regulation. These epigenetic modifications play an important role in transcription 222 factor binding to DNA regulatory sequences and target gene transcription. When understanding 223 light-regulation of gene expression, several epigenetic marks have been studied (Barneche et al. 224 2014, Kim et al. 2015, Perrella and Kaiserli 2016) but only a few have been related to shade. 225 Below we will describe three examples of histone marks over specific loci and the mechanisms 226 that modulate shade-regulated gene expression. 227 228 Case 1: Histone acetylation of specific genes is induced in response to shade 229 Histone acetylation is a post-translational modification that consists in the transfer of acetyl 230 groups to Lys residues of histones. The addition of the acetyl group (1) neutralizes the Lys 231 positive charge that counteracts the negative charged DNA, eventually provoking the relaxation 232 of the chromatin, and/or (2) provide recognition sites for factors involved in the activation of 233 gene expression (Carrozza et al. 2003). Acetylation is catalyzed by histone acetyltransferases 234 (HATs), whereas histone deacetylation is catalyzed by histone deacetylases (HDACs) (Fig. 2A). 235 These enzymes are encoded by multiple genes with functional redundancy, with at least 12 236 HAT and 18 HDAC genes in Arabidopsis (Pandey et al. 2002). 237 238 In shade, histone acetylation levels play an important role in chromatin compaction at the 239 chromocenter level, as previously discussed (Fig. 1B). The HDAC activity of HDA6 is required 240 for the deacetylation that leads to chromatin compaction. Several studies show that histone 241 acetylation and deacetylation mutants have defects in hypocotyl elongation when grown in 242 monochromatic FR and R conditions, pointing to the importance of acetylation in the phyA- and 243 phyB-mediated responses of hypocotyl elongation. One of the HATs that has been found to play 244 a role in R, FR and B light response pathways is HISTONE ACETYLTRANSFERASE OF 245 THE TAFII250 FAMILY 2 (HAF2, also known as TAF1) (Bertrand et al. 2005). Although the 246 haf2 mutant does not show significant changes in hypocotyl growth in R and FR conditions, 247 when combined with the ELONGATED HYPOCOTYL 5 (HY5) deficient mutant showed an 248 enhancement of the hy5 hypocotyl elongation. HY5 is a light-responsive positive transcription 249 factor that, when mutated, results in hyposensitivity (i.e., long hypocotyls) to several light 250 conditions (Oyama et al. 1997). This enhancement suggests that the HAF2-mediated acetylation 251 interacts with the HY5 pathway downstream of the phytochrome signaling. Another HAT that 252 has a role in the R and FR perception is GENERAL CONTROL NONDEREPRESSIBLE 5 253 (GCN5, also known as HAG1). In contrast with the haf2 single mutant, the single gcn5 mutant 254 shows a long hypocotyl phenotype under different light conditions. Moreover, gcn5 haf2 double 255 mutant shows an additive hypocotyl phenotype. Although both HATs share target genes, the 256 lack of effect on hypocotyl elongation in the gcn5 hy5 double mutant, in contrast with the haf2 257 hy5, suggests that they still regulate different pathways (Fig. 2B) (Benhamed et al. 2006). 258 259 The role of HATs is counterbalanced by the action of HDACs (Fig. 2A). In addition to HDA6, 260 two other HDACs have been reported to regulate hypocotyl growth in R and FR conditions, 261 HDA19 and HDA15. HDA19 (also known as HD1) is a positive regulator of gene expression 262 and hypocotyl growth in FR (and slightly also in R), showing the opposite role than GCN5, as 263 can be concluded from the hda19 mutant short hypocotyl phenotype compared to wild-type 264 (Benhamed et al. 2006). The GCN5/HDA19 pair might act as an antagonistic system to control 265 histone acetylation levels to switch gene expression. But this opposite HAT/HDAC activity is 266 not always as clear. HDA15, which belongs to the same HDAC family than HDA19, plays an 267 opposite role to HDA19 in the regulation of hypocotyl elongation. Under R and FR, hda15 268 hda19 double mutant presents an intermediate hypocotyl phenotype compared to the hda15 269 (long hypocotyls) and hda19 (short hypocotyls) single mutants (Liu et al. 2013). Therefore, 270 different HDACs may have distinct (even opposite) functions in hypocotyl elongation. 271 272 The complexity of the regulation by acetylation can be understood not only by the different 273 HATs or HDACs that participate, but also by (1) the specific genes targeted by the same HAT 274 or HDAC, (2) which histones (H3 or H4) are modified, (3) the particular Lys residues that are 275 acetylated, and (4) the position of the acetylation mark within the gene (e.g., promoter vs. 276 coding region). Some studies have addressed the acetylation levels of specific genes in various 277 light conditions, such as different light wavelengths, low light and low R:FR. In different 278 monochromatic lights and in low light, levels of acetylated H3 in Lys 9 (H3K9ac) rise or fall 279 together with the expression levels of four tested genes in wild-type seedlings: light induces 280 PSII LIGHT HARVESTING COMPLEX GENE 1.4 (LHCB1.4) and RIBULOSE 281 BISPHOSPHATE CARBOXYLASE SMALL CHAIN 1B (RBCS-1B) and represses PEROXIDASE 282 21 and CITRATE SYNTHASE 3 (CSY3) (Guo et al. 2008). Interestingly, changes of acetylation 283 levels were regulated through HY5, which agrees with the long hypocotyl phenotype of hy5 284 haf2 double mutant. When analyzing specific genes, haf2 mutant has lower H3 acetylation 285 levels in the light-induced genes RBCS-1A, CHLOROPHYLL A/B-BINDING PROTEIN 2 286 (CAB2) although does not affect the acetylation in the INDOLE-3-ACETIC ACID INDUCIBLE 287 3 (IAA3) (Bertrand et al. 2005, Benhamed et al. 2006). Acetylation of H3 and H4 in these three 288 genes is also altered in the gcn5 (reduced acetylation) and hda19 (enhanced acetylation) mutants 289 (Benhamed et al. 2006). The acetylated residues differ among these mutants; both GCN5 and 290 HAF2 acetylate Lys 9 and Lys 27 in H3 (H3K9 and H3K27, respectively), but these marks are 291 regulated differently depending on the target gene. Taking this also into consideration, RBCS- 292 1A seems to be regulated by GCN5, HAF2 and HDA19, whereas IAA3 is only dependent on 293 GCN5 and HDA19 (Fig. 2B). To add more complexity, not only the specific target genes differ 294 but also the distribution of the acetylation mark along the gene. The location of the reduced 295 acetylation levels in the gcn5 mutant do not correspond completely with those hyperacetylated 296 in the hda19 mutant. HDA19 is involved the regulation of RBCS-1A and IAA3 with an effect on 297 histone acetylation that seems to be operating in a large range of promoters and both upstream 298 and at the core promoter regions (i.e., the one that contains the TATA box) (Benhamed et al. 299 2006). The specificity of the acetylation location can be explained by the mode of action of the 300 HATs and by the mechanisms that target them to the specific genes: GCN5 binds directly to the 301 RBCS-1A and CAB2 promoters and HAF2 may do it through the transcription factor HY5 (Fig. 302 2B). 303 304 It seems likely that these mechanisms and/or epigenetic components might also have a role in 305 low R:FR regulated gene expression. Indeed, it has been recently demonstrated that histone 306 acetylation also changes in low R:FR ratio (Fig. 2C). Histone acetylation is promoted after just 307 one hour of simulated shade in the auxin biosynthetic gene YUCCA 8 (YUC8). It is reported that 308 shade increases the levels of H3K9ac, H3K27ac and H4K5ac, but not those of H3K14ac or 309 H3K36ac (Peng et al. 2018). Among the different marks, increase in H4K5 acetylation has been 310 proposed to happen through the HAT activity encoded by HISTONE ACETYLTRANSFERASE 311 OF THE MYST FAMILY 1 / HISTONE ACETYLTRANSFERASE OF THE GNAT FAMILY 4 312 (HAM1/HAG4) and HAM2/HAG5 (Earley et al. 2007, Xiao et al. 2013), and the other marks, 313 H3K9ac and H3K27ac, have to be regulated by other HAT, possibly GCN5 and HAF2. 314 315 In summary, the variety of the Lys residue acetylated, the distribution of the mark along the 316 genome, together with the different photomorphogenic hypocotyl phenotypes, exemplifies the 317 complexity of how transcriptional control of the different shade-induced genes is exerted by the 318 epigenetic mark imposed by HATs and HDACs. It seems likely that enhanced acetylation of 319 histones might have an impact on the expression of the specific locus either directly or 320 indirectly: if placed in the promoter region, it might facilitate binding of transcription factors to 321 DNA regulatory sequences; alternatively, if placed in the gene body (the region containing 322 introns and exons), it might also facilitate transcription by reducing compaction of nucleosomes. 323 324 Case 2: Histone methylation is involved in gene regulation in response to low R:FR 325 Methylation of histones consists in the addition of one, two or three methyl groups to a Lys 326 residue. Unlike acetylation, the methyl group is neutral and does not affect the positive charge 327 of the Lys. Methylation marks can be related to transcriptional activation or repression 328 depending on the modified residue. In general, H3 trimethylation of the Lys 4 (H3K4me3) and 329 Lys 36 (H3K36me3) are associated with activation and H3 dimethylation of Lys 9 (H3K9me2) 330 and trimethylation of Lys 27 (H3K27me3) with repression of genes (Pontvianne et al. 2010). 331 332 The H3K4me3 and H3K36me3 activating marks have been described in the 5’-end of the gene 333 body of YUC8 and their presence have an important role in recruiting, by direct binding, the 334 MORF RELATED GENE 2 (MRG2) histone methylation reader (Peng et al. 2018). 335 Accordingly, the reduced levels of those H3 marks in the H3K4-methyltransferase mutant 336 homologue of trithorax 1-2 (atx1-2) and the H3K36-methyltransferase mutant set domain group 337 8-2 (sdg8-2) results in a reduction of MRG2 binding to the modified histones (Peng et al. 2018) 338 (Fig. 2C). Although histone methylation levels of these marks are not affected by simulated 339 shade, their reduction in the atx1-2 and sdg8-2 mutant results in a slightly shorter hypocotyl. 340 This supports a basal role of H3K4me3 and H3K36me3 in the response of seedlings to low 341 R:FR. The methylation reader mrg2 mutant does not exhibit a phenotype, but when the other 342 member of the family MRG1 is also mutated, hypocotyl elongation is drastically reduced in 343 shade (Peng et al. 2018). Interestingly, binding of MRG2 to YUC8, IAA19 and 344 PACLOBUTRAZOL RESISTANCE 1 (PRE1) is enhanced in shade and dependent on PIF7. Peng 345 and collaborators found that PIF7 mediates MRG2 binding to the genes by direct interaction and 346 that the recruitment of MRG2 leads to the acetylation of H4K5 probably by the known 347 interaction of MRG2 with the HAM1/HAM2 HATs (Xu et al. 2014) (Fig. 2C). This case 348 illustrates how different types of marks, histone methylation and acetylation, crosstalk to 349 modulate shade-regulated transcription. 350 351 Case 3. H3K27me3 could modulate gene regulation in response to shade 352 A histone methylation mark associated with gene repression, H3K27me3, has been also 353 indirectly linked to the regulation of the seedling response to shade through the role of LIKE 354 HETEROCHROMATIN PROTEIN 1 (LHP1), also known as TERMINAL FLOWER 2 (TFL2) 355 (Valdés et al. 2012). LHP1 is a chromatin-associated protein recruited to H3K27me3-rich 356 regions that is proposed to be responsible for the stabilization of those repressive regions (Exner 357 et al. 2009). When de-etiolated under R and FR (but not under B), lhp1 mutant seedlings present 358 a strong hypersensitive response as hypocotyl elongation is strongly inhibited. The lhp1 de- 359 etiolation phenotype is dependent on phyA and phyB, as double mutants with lhp1 have the 360 same phenotype of the phyA and phyB single mutants in FR and R, respectively. This indicates 361 that LHP1 acts downstream of phytochromes. Importantly, lhp1 mutant seedlings are almost 362 unresponsive to low R:FR. Because LHP1 has been related to auxin biosynthesis, which is 363 reduced in the lhp1 mutant (Rizzardi et al. 2011), the authors explored the possibility of an 364 impaired auxin-mediated shade response. The addition of external auxin cannot rescue the lhp1 365 phenotype but potentiates the shade-induced gene expression of the Aux/IAA members IAA5 366 and IAA19. As lhp1 is not responsive to auxin, although the Aux/IAAs expression levels are 367 increased, it is expected to act downstream of Aux/IAA, hypothesis reinforced by the direct 368 binding of LHP1 with IAA5, IAA6 and IAA19 proteins (Valdés et al. 2012). This case 369 highlights the role of LHP1 in regulating the seedlings respond to shade probably via the 370 H3K27me3 mark. How LHP1 is acting at the chromatin level and the role of phytochromes and 371 Aux/IAAs in this process is still unknown. 372 373 Future perspectives and open questions 374 375 When the environment changes, chromatin is reshaped. Chromatin relaxation likely allows 376 access to the transcriptional machinery to sustain changes in the expression of a broad range of 377 genes that are instrumental in plant’s acclimation to the new environment. From this 378 perspective, few works have been conducted till date in the shade response field. What aspects 379 need further studies? 380 381 (1) Global changes in nucleus structure are clearly detected after a few days of shade exposure 382 (van Zanten et al. 2010), when changes in gene expression affect a group of late (indirect) target 383 genes belonging to Gene Ontology (GO) categories related to diverse stresses (Leivar et al. 384 2012) (Fig. 3). Consistently, a global chromatin decondensation occurs in a wide variety of 385 abiotic changes and stress situations (Probst and Mittelsten Scheid 2015) and responses to shade 386 are not an exception. How a wide variety of environmental changes converge in a common 387 chromatin decondensation process? Shade also provokes early changes in gene expression 388 within one hour of treatment, which are mostly directed by PIF binding (Kohnen et al. 2016) 389 (Fig. 3). Are nuclear structure changes required for this early transcriptional reprogramming 390 induced by shade? To link chromosome structure and gene expression, higher resolution 391 techniques, like chromosome conformation capture (Kempfer and Pombo 2019), might be 392 needed. 393 394 (2) Chromatin structural changes induced by an environmental cue may persist to make genes 395 more responsive to future changes in the same signal. This “memory” can be useful to allow the 396 plant to give faster acclimation responses (Fig. 3) (Bruce et al. 2007). Addressing the dynamics 397 of shade-induced histone marks might be required to define their possible role in providing a 398 memory. 399 400 (3) So far, shade-triggered epigenetic modifications at specific loci have been studied in a few 401 genes, which makes difficult to propose general mechanisms to regulate shade responses. This 402 can be addressed by using a genome-wide approach of different epigenetic marks in response to 403 shade. 404 405 (4) Current studies have focused only in a limited number of epigenetic marks. When searching 406 for new players in shade-response, we can look into the epigenetic regulation of 407 thermomorphogenesis, a process known to share many phenotypic and regulatory similarities 408 with that regulated by shade (Legris et al. 2017). Two components known to mediate the 409 thermosensory response in plants emerge in this search: the histone variant H2A.Z (Kumar and 410 Wigge 2010), and the ATP-dependent chromatin remodeling factor PICKLE, also shown to 411 negatively control photomorphogenesis (Zha et al. 2017). 412 413 (5) How are the epigenetic and transcriptional mechanisms integrated? We can envisage at least 414 two non-exclusive possibilities: (a) transcription factor binding capacity can be modulated by 415 chromatin structure and (b) the transcription factor itself can act as a pioneer attracting 416 chromatin remodelers to specific loci. Indeed, the latter possibility has been already suggested 417 in a few cases, such as the HDA15-PIF3 (Liu et al. 2013) and the MRG2-PIF7 (Peng et al. 418 2018). 419 420 Author contributions 421 JM-R and JFM-G developed the draft of the review, JM-R wrote the draft of the manuscript, 422 and JM-R and JFM-G reviewed, edited and approved the final manuscript. 423 424 Acknowledgements – We thank Nicolas Bologna and Julia Qüesta (CRAG) for comments on the 425 manuscript. JM-R has received funding from the European Union’s Horizon 2020 research and 426 innovation programme under the Marie Skłodowska-Curie grant agreement No 797473. Our 427 research is supported by grants from FEDER- Spanish Ministry of Science, Innovation and 428 Universities – Agencia Estatal de Investigación (reference BIO2017-85316-R) and AGAUR 429 (2017-SGR710 and Xarba) to JFM-G. We are part of the COST-Action INDEPTH CA16212. 430 We also acknowledge the support of the CERCA Programme /Generalitat de Catalunya and the 431 Ministry of Economy and Competitiveness though the “Severo Ochoa Programme for Centres 432 of Excellence in R&D” 2016-2019 (SEV-2015-0533). 433 434 References 435 436 Barneche F, Malapeira J, Mas P (2014) The impact of chromatin dynamics on plant light 437 responses and circadian clock function. J Exp Bot 65(11): 2895-2913 438 Benhamed M, Bertrand C, Servet C, Zhou DX (2006) Arabidopsis GCN5, HD1, and 439 TAF1/HAF2 interact to regulate histone acetylation required for light-responsive gene 440 expression. Plant Cell 18(11): 2893-2903 441 Benvenuto G, Formiggini F, Laflamme P, Malakhov M, Bowler C (2002) The 442 Photomorphogenesis Regulator DET1 Binds the Amino-Terminal Tail of Histone H2B 443 in a Nucleosome Context. Current Biology 12(17): 1529-1534 444 Bertrand C, Benhamed M, Li YF, Ayadi M, Lemonnier G, Renou JP, Delarue M, Zhou DX 445 (2005) Arabidopsis HAF2 gene encoding TATA-binding protein (TBP)-associated 446 factor TAF1, is required to integrate light signals to regulate gene expression and 447 growth. J Biol Chem 280(2): 1465-1473 448 Bourbousse C, Mestiri I, Zabulon G, Bourge M, Formiggini F, Koini MA, Brown SC, Fransz P, 449 Bowler C, Barneche F (2015) Light signaling controls nuclear architecture 450 reorganization during seedling establishment. Proceedings of the National Academy of 451 Sciences 112(21): E2836 452 Bruce TJA, Matthes MC, Napier JA, Pickett JA (2007) Stressful “memories” of plants: 453 Evidence and possible mechanisms. Plant Science 173(6): 603-608 454 Carrozza MJ, Utley RT, Workman JL, Cote J (2003) The diverse functions of histone 455 acetyltransferase complexes. Trends in genetics : TIG 19(6): 321-329 456 Casal JJ (2012) Shade avoidance. The arabidopsis book 10: e0157 457 de Wit M, Keuskamp DH, Bongers FJ, Hornitschek P, Gommers CMM, Reinen E, Martinez- 458 Ceron C, Fankhauser C, Pierik R (2016) Integration of Phytochrome and Cryptochrome 459 Signals Determines Plant Growth during Competition for Light. Curr Biol 26(24): 460 3320-3326 461 Del Prete S, Arpón J, Sakai K, Andrey P, Gaudin V (2014) Nuclear Architecture and Chromatin 462 Dynamics in Interphase Nuclei of Arabidopsis thaliana. Cytogenetic and Genome 463 Research 143(1-3): 28-50 464 Earley KW, Shook MS, Brower-Toland B, Hicks L, Pikaard CS (2007) In vitro specificities of 465 Arabidopsis co-activator histone acetyltransferases: implications for histone 466 hyperacetylation in gene activation. The Plant Journal 52(4): 615-626 467 Exner V, Hennig L (2008) Chromatin rearrangements in development. Current Opinion in Plant 468 Biology 11(1): 64-69 469 Exner V, Aichinger E, Shu H, Wildhaber T, Alfarano P, Caflisch A, Gruissem W, Köhler C, 470 Hennig L (2009) The chromodomain of LIKE HETEROCHROMATIN PROTEIN 1 is 471 essential for H3K27me3 binding and function during Arabidopsis development. PLoS 472 One 4(4): e5335-e5335 473 Fransz P, Soppe W, Schubert I (2003) Heterochromatin in interphase nuclei of Arabidopsis 474 thaliana. Chromosome research : an international journal on the molecular, 475 supramolecular and evolutionary aspects of chromosome biology 11(3): 227-240 476 Fransz P, de Jong JH, Lysak M, Castiglione MR, Schubert I (2002) Interphase chromosomes in 477 Arabidopsis are organized as well defined chromocenters from which euchromatin 478 loops emanate. Proceedings of the National Academy of Sciences 99(22): 14584 479 Guo L, Zhou J, Elling AA, Charron JB, Deng XW (2008) Histone modifications and expression 480 of light-regulated genes in Arabidopsis are cooperatively influenced by changing light 481 conditions. Plant Physiol 147(4): 2070-2083 482 Kempfer R, Pombo A (2019) Methods for mapping 3D chromosome architecture. Nature 483 Reviews Genetics 484 Kim JM, Sasaki T, Ueda M, Sako K, Seki M (2015) Chromatin changes in response to drought, 485 salinity, heat, and cold stresses in plants. Front Plant Sci 6: 114 486 Kohnen MV, Schmid-Siegert E, Trevisan M, Petrolati LA, Senechal F, Muller-Moule P, Maloof 487 J, Xenarios I, Fankhauser C (2016) Neighbor Detection Induces Organ-Specific 488 Transcriptomes, Revealing Patterns Underlying Hypocotyl-Specific Growth. Plant Cell 489 28(12): 2889-2904 490 Kumar SV, Wigge PA (2010) H2A.Z-Containing Nucleosomes Mediate the Thermosensory 491 Response in Arabidopsis. Cell 140(1): 136-147 492 Legris M, Nieto C, Sellaro R, Prat S, Casal JJ (2017) Perception and signalling of light and 493 temperature cues in plants. Plant J 90(4): 683-697 494 Leivar P, Tepperman JM, Cohn MM, Monte E, Al-Sady B, Erickson E, Quail PH (2012) 495 Dynamic Antagonism between Phytochromes and PIF Family Basic Helix-Loop-Helix 496 Factors Induces Selective Reciprocal Responses to Light and Shade in a Rapidly 497 Responsive Transcriptional Network in Arabidopsis. The Plant Cell 24(4): 1398 498 Li L, Ljung K, Breton G, Schmitz RJ, Pruneda-Paz J, Cowing-Zitron C, Cole BJ, Ivans LJ, 499 Pedmale UV, Jung HS, Ecker JR, Kay SA, Chory J (2012) Linking photoreceptor 500 excitation to changes in plant architecture. Genes Dev 26(8): 785-790 501 Liu X, Chen CY, Wang KC, Luo M, Tai R, Yuan L, Zhao M, Yang S, Tian G, Cui Y, Hsieh 502 HL, Wu K (2013) PHYTOCHROME INTERACTING FACTOR3 associates with the 503 histone deacetylase HDA15 in repression of chlorophyll biosynthesis and 504 photosynthesis in etiolated Arabidopsis seedlings. Plant Cell 25(4): 1258-1273 505 Maloof JN, Borevitz JO, Dabi T, Lutes J, Nehring RB, Redfern JL, Trainer GT, Wilson JM, 506 Asami T, Berry CC, Weigel D, Chory J (2001) Natural variation in light sensitivity of 507 Arabidopsis. Nature Genetics 29(4): 441-446 508 Mas P, Devlin PF, Panda S, Kay SA (2000) Functional interaction of phytochrome B and 509 cryptochrome 2. Nature 408(6809): 207-211 510 Oyama T, Shimura Y, Okada K (1997) The Arabidopsis HY5 gene encodes a bZIP protein that 511 regulates stimulus-induced development of root and hypocotyl. Genes Dev 11(22): 512 2983-2995 513 Pandey R, Muller A, Napoli CA, Selinger DA, Pikaard CS, Richards EJ, Bender J, Mount DW, 514 Jorgensen RA (2002) Analysis of histone acetyltransferase and histone deacetylase 515 families of Arabidopsis thaliana suggests functional diversification of chromatin 516 modification among multicellular eukaryotes. Nucleic Acids Res 30(23): 5036-5055 517 Peng M, Li Z, Zhou N, Ma M, Jiang Y, Dong A, Shen WH, Li L (2018) Linking 518 PHYTOCHROME-INTERACTING FACTOR to Histone Modification in Plant Shade 519 Avoidance. Plant Physiol 176(2): 1341-1351 520 Perrella G, Kaiserli E (2016) Light behind the curtain: photoregulation of nuclear architecture 521 and chromatin dynamics in plants. New Phytologist 212(4): 908-919 522 Pontvianne F, Blevins T, Pikaard CS (2010) Arabidopsis Histone Lysine Methyltransferases. 523 Adv Bot Res 53: 1-22 524 Probst AV, Mittelsten Scheid O (2015) Stress-induced structural changes in plant chromatin. 525 Curr Opin Plant Biol 27: 8-16 526 Rizzardi K, Landberg K, Nilsson L, Ljung K, Sundås-Larsson A (2011) TFL2/LHP1 is involved 527 in auxin biosynthesis through positive regulation of YUCCA genes. The Plant Journal 528 65(6): 897-906 529 Roig-Villanova I, Martinez-Garcia JF (2016) Plant Responses to Vegetation Proximity: A 530 Whole Life Avoiding Shade. Front Plant Sci 7: 236 531 Rutowicz K, Lirski M, Mermaz B, Teano G, Schubert J, Mestiri I, Kroteń MA, Fabrice TN, 532 Fritz S, Grob S, Ringli C, Cherkezyan L, Barneche F, Jerzmanowski A, Baroux C 533 (2019) Linker histones are fine-scale chromatin architects modulating developmental 534 decisions in Arabidopsis. Genome Biology 20(1): 157 535 Rutowicz K, Puzio M, Halibart-Puzio J, Lirski M, Kotliński M, Kroteń MA, Knizewski L, 536 Lange B, Muszewska A, Śniegowska-Świerk K, Kościelniak J, Iwanicka-Nowicka R, 537 Buza K, Janowiak F, Żmuda K, Jõesaar I, Laskowska-Kaszub K, Fogtman A, Kollist H, 538 Zielenkiewicz P, Tiuryn J, Siedlecki P, Swiezewski S, Ginalski K, Koblowska M, 539 Archacki R, Wilczynski B, Rapacz M, Jerzmanowski A (2015) A Specialized Histone 540 H1 Variant Is Required for Adaptive Responses to Complex Abiotic Stress and Related 541 DNA Methylation in Arabidopsis. Plant Physiology 169(3): 2080-2101 542 Simon L, Voisin M, Tatout C, Probst AV (2015) Structure and Function of Centromeric and 543 Pericentromeric Heterochromatin in Arabidopsis thaliana. Front Plant Sci 6: 1049 544 Snoek BL, Pavlova P, Tessadori F, Peeters AJM, Bourbousse C, Barneche F, de Jong H, Fransz 545 PF, van Zanten M (2017) Genetic Dissection of Morphometric Traits Reveals That 546 Phytochrome B Affects Nucleus Size and Heterochromatin Organization in Arabidopsis 547 thaliana. G3 (Bethesda) 7(8): 2519-2531 548 Tessadori F, Chupeau M-C, Chupeau Y, Knip M, Germann S, van Driel R, Fransz P, Gaudin V 549 (2007) Large-scale dissociation and sequential reassembly of pericentric 550 heterochromatin in dedifferentiated Arabidopsis cells. Journal of Cell Science 120(7): 551 1200 552 Tessadori F, van Zanten M, Pavlova P, Clifton R, Pontvianne F, Snoek LB, Millenaar FF, 553 Schulkes RK, van Driel R, Voesenek LA, Spillane C, Pikaard CS, Fransz P, Peeters AJ 554 (2009) Phytochrome B and histone deacetylase 6 control light-induced chromatin 555 compaction in Arabidopsis thaliana. PLoS Genet 5(9): e1000638 556 Valdés AE, Rizzardi K, Johannesson H, Para A, Sundås-Larsson A, Landberg K (2012) 557 Arabidopsis thaliana TERMINAL FLOWER2 is involved in light-controlled signalling 558 during seedling photomorphogenesis. Plant, Cell & Environment 35(6): 1013-1025 559 van Zanten M, Tessadori F, McLoughlin F, Smith R, Millenaar FF, van Driel R, Voesenek LA, 560 Peeters AJ, Fransz P (2010) Photoreceptors CRYTOCHROME2 and phytochrome B 561 control chromatin compaction in Arabidopsis. Plant Physiol 154(4): 1686-1696 562 Xiao J, Zhang H, Xing L, Xu S, Liu H, Chong K, Xu Y (2013) Requirement of histone 563 acetyltransferases HAM1 and HAM2 for epigenetic modification of FLC in regulating 564 flowering in Arabidopsis. Journal of Plant Physiology 170(4): 444-451 565 Xu Y, Gan E-S, Zhou J, Wee W-Y, Zhang X, Ito T (2014) Arabidopsis MRG domain proteins 566 bridge two histone modifications to elevate expression of flowering genes. Nucleic 567 Acids Research 42(17): 10960-10974 568 Yi C, Deng XW (2005) COP1 – from plant photomorphogenesis to mammalian tumorigenesis. 569 Trends in Cell Biology 15(11): 618-625 570 Yu X, Liu H, Klejnot J, Lin C (2010) The Cryptochrome Blue Light Receptors. The arabidopsis 571 book 8: e0135 572 Zha P, Jing Y, Xu G, Lin R (2017) PICKLE chromatin-remodeling factor controls 573 thermosensory hypocotyl growth of Arabidopsis. Plant Cell Environ 40(10): 2426-2436 574 575 576 Data availability statement 577 Data sharing is not applicable to this article as no new data were created or analyzed in this 578 study. 579 580 581 FIGURE LEGENDS 582 583 Fig. 1. Nuclear structure changes in response to low light and low R:FR. (A) Chromatin is 584 heterogeneously distributed in the nucleus. This can be observed after DAPI staining, where 585 nuclei present highly stained speckles called chromocenters (CC). The left panel corresponds to 586 a hypocotyl stained with DAPI and the magnification of a selected nucleus. On the right, a 587 simplified outlined version of this nucleus is represented, highlighting the CC and the 588 surrounding nucleoplasmic area. These CC are formed by densely packed heterochromatin with 589 very low transcriptional activity. From them, chromatin loops emanate occupying the less 590 stained regions of the nucleus. This fraction called euchromatin is less packed (more open) and 591 more accessible to transcriptional machinery to bind, which therefore allows genes to be 592 expressed. Bar corresponds to 50 µm. (B) Shade (low light and/or low R:FR) promotes 593 decondensation of the chromatin, reducing the number and size of CC. Low light also increases 594 nuclear size. So far, it is known that both responses are controlled by the light receptors phyB 595 and cry2. Shade (low R:FR) and low light inhibits phyB activity, that reduces the promotion of 596 CC and increases nuclear size. As phyB, cry2 is a regulator of CC formation. In high light, cry2 597 is degraded and CC is promoted. Low light directly induces cry2 accumulation and indirectly 598 increases its abundance by phyB activity, that results in less induction of CC formation. Both 599 receptors can modulate chromatin structure either directly or indirectly through a chromatin 600 remodeler complex, which can include HDA6, H1.3 (that has been related to chromatin 601 condensation in shade) or members of the ubiquitination machinery (as COP1). * Changes in 602 nuclear size have not been reported in low R:FR. ** The participation of cry2 in CC formation 603 has been reported only in low light conditions, not in low R:FR. 604 605 Fig. 2. Histone acetylation changes in low light and low R:FR. (A) Histone acetylation mediated 606 by HAT promotes chromatin decondensation and allows gene expression, while deacetylation 607 by HDAC provokes chromatin compaction that leads to transcriptional repression. (B) HAT 608 activity targets specific genes. The HAT GCN5 is able to either (1) directly bind to the core 609 promoter (where the TATA element is located) of some genes like IAA3 to acetylate histones 610 (this binding is enhanced when histones are acetylated) and/or (2) indirectly through an 611 unknown protein like in the RBCS-1A gene. This protein could be a transcription factor like in 612 the case of HAF2 (3) that is indirectly guided by HY5 to the regions in the RBCS-1A promoter 613 to be acetylated. HAF2 and GCN5 can act synergistically to control RBCS-1A expression, 614 whereas GCN5 binds alone to promote IAA3 expression. The HDAC activity (4) of HDA19 615 seems to globally de-acetylate many genes. It is still unknown if the binding is direct (for 616 instance, via a TF) or indirect. The participation of HDA6 at the gene level is unknown but 617 likely as it affects nuclei chromatin density. (C) Another mechanism to acetylate histones is 618 linked to the presence of other histone marks, as it occurs in the control of YUC8 expression. 619 The H3K4/K36me mark deposited by the action of SDG8/ATX1 is read by MRG1/2 only when 620 interacts with PIF7. MRG1/2 are responsible to bind the HAT HAM1/2 to acetylate the gene 621 and promote its expression. Acetylation only occurs when PIF7 is activated by low R:FR. 622 623 Fig. 3. Changes in gene expression and epigenetic marks in low light and low R:FR take place 624 at different paces. Gene expression is induced rapidly after low R:FR (within 15 min in some 625 cases) via PIF activity. Many of these genes belong to the auxin biosynthesis pathway or are 626 TFs that trigger the expression of other genes. Increased auxin levels and activation of TFs 627 results in the late activation or repression of PIF-indirect secondary genes. These latter genes 628 belong to GO categories related to different stresses. In addition to gene expression 629 reprogramming, epigenetic changes occur. Some are mid-long term (increase of nuclear size, 630 decrease of chromocenters or the accumulation of the H1.3 variant) and others are faster 631 (increase of acetylation already after one hour of the signal). It is unknown if these rapid 632 changes in acetylation are maintained or not in later time points. 633 634 EUCHROMATIN A chromatin loops transcriptional activation chromocenters (CC) transcriptional repression HETEROCHROMATIN B high light low light high R:FR low R:FR CC formation nuclear size* HDA6 ? ? ? H1.3 COP1 … cry2** phyB light intensity low R:FR Fig.1 A HDAC HAT B IAA3 RBCS-1A TATA low IAA3 light 1 GCN5 ? 2 ? high HDA19 4 light TATA HDA6 ? HAF2 RBCS-1A ? HY5 3 H3/H4 acetylation H3K4/K36 methylation C SDG8 ATX1 YUC8 YUC8 PIF7 MRG1/2 low PIF7 HAM1/2 R:FR Fig.2 low light and/or low R:FR 1h 3h 24h 72h PIFs activation Gene expression Auxin-related genes (YUC, IAA, SAUR…), TF (ATHB, BEE1…), etc. EARLY (PIF direct) targets LATE response to stress GO enriched (indirect) targets Global rearrangements Nuclear size Epigenetic changes Chromo- centers H1.3 Local (loci) rearrangements stable mark (memory) ? Histone removal acetylation Fig.3