This is the accepted version of the article: Leivar, Pablo; Martín, Guiomar; Soy, Judit; [et al.]. «Phytochrome-imposed inhibition of PIF7 activity shapes photoperiodic growth in Arabidopsis together with PIF1 , 3, 4 and 5». Physiologia Plantarum, Vol. 169, issue 3 (July 2020), p. 452-466. DOI 10.1111/ppl.13123 This version is avaible at https://ddd.uab.cat/record/230990 under the terms of the license 1 Phytochrome-imposed inhibition of PIF7 activity shapes photoperiodic growth 2 in Arabidopsis together with PIF1, 3, 4 and 5 3 Pablo Leivar1,2,a,b, Guiomar Martín1, Judit Soy1, Jutta Dalton-Roesler3, Peter H. Quail3, and Elena 4 Monte1,a,b 5 1 6 Department of Molecular Genetics, Centre for Research in Agricultural Genomics (CRAG) 7 CSIC-IRTA-UAB-UB. Campus UAB, Bellaterra, Barcelona, Spain. 2 8 Laboratory of Biochemistry, Institut Químic de Sarrià, Universitat Ramon Llull. Barcelona, 9 Spain. 3 10 Department of Plant and Microbial Biology, University of California-Berkeley, and United 11 States Department of Agriculture, Plant Gene Expression Center, USA. a 12 Equal contribution. b 13 To whom correspondence should be addressed. E-mail: pablo.leivar@iqs.url.edu, 14 elena.monte@cragenomica.es. 15 16 17 Running Title: 18 PIF7 and PIFq regulation of photoperiodic growth 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 1 38 39 ABSTRACT 40 Under photoperiodic conditions, Arabidopsis thaliana seedling growth is inhibited in long days 41 (LD), but promoted under the extended nights of short days (SD). This behavior is partly 42 implemented by phytochrome (phy)-imposed oscillations in the abundance of the growth- 43 promoting, phy-interacting bHLH transcription factors PHY-INTERACTING FACTOR 1 44 (PIF1), PIF3, PIF4 and PIF5 (PIF quartet or PIFq). However, the observation that a pifq mutant 45 is still stimulated to elongate when given a phy-inactivating end-of-day far-red pulse (EODFR), 46 suggests that additional factors are involved in the phy-mediated suppression of growth during 47 the subsequent dark period. Here, by combining growth-analysis of pif7 single- and higher-order 48 mutants with gene expression analysis under SD, LD, SD-EODFR, and LD-EODFR, we show 49 that PIF7 promotes growth during the dark hours of SD, by regulating growth-related gene 50 expression. Interestingly, the relative contribution of PIF7 in promoting growth is stronger under 51 EOD-FR, while PIF3 role is more important under SD, suggesting that PIF7 is a prominent target 52 of phy-suppression. Indeed, we show that in SD, phy imposes phosphorylation and inactivation 53 of PIF7 during the light hours, and prevents full dephosphorylation during the night. This 54 repression can be lifted with an EODFR, which correlates with increased PIF7-mediated gene 55 expression and elongation. In addition, our results suggest that PIF7 function might involve 56 heterodimerization with PIF3. Furthermore, our data indicate that a pifqpif7 quintuple mutant is 57 largely insensitive to photoperiod for hypocotyl elongation. Collectively, the data suggest that 58 PIF7, together with the PIFq, are required for the photoperiodic regulation of seasonal growth. 59 60 61 Abbreviations: 62 EOD-FR, end-of-day far red; FRp, far red pulse; LD, long day; PHY, phytochrome; PIF, phy- 63 interacting factor; PIF7; SD, short day. 64 65 66 67 68 69 70 71 72 73 74 2 75 INTRODUCTION 76 Phytochrome interacting factors (PIFs) are basic helix-loop-helix transcriptional regulators that 77 are promoters of elongation growth in Arabidopsis thaliana (Leivar and Monte 2014). PIFs 78 interact specifically with the active Pfr conformer of phytochrome (phy) photoreceptors. In the 79 dark, the phy inactive red (R) light-absorbing Pr conformer is localized in the cytosol and PIFs 80 accumulate in the nucleus. Upon light exposure, phy Pr converts to the biologically active far red 81 (FR) light-absorbing Pfr form, which translocates into the nucleus and interacts with the PIFs, 82 triggering rapid degradation of the PIF quartet (PIFq) members PIF1, PIF3, PIF4 and PIF5 (with 83 half-lives of 5-20 min) (Pham et al. 2018). Of the five phytochromes in Arabidopsis (phyA-phyE), 84 phyA and phyB dominate the regulation of PIF degradation (Bauer et al. 2004, Al-Sady et al. 85 2006, Shen et al. 2007, 2008). Importantly, thanks to the reversible photoconversion between the 86 Pr and the Pfr phy conformers (Al-Sady et al. 2006, Shen et al. 2007, 2008), PIF levels oscillate 87 in environments with fluctuations in Pr and Pfr levels such as under low Red (R)/Far Red (FR) 88 ratio conditions typical of shade environments, or in diurnal light-dark cycles. 89 Under short-day (SD) photoperiods, growth is rhythmic with maximal hypocotyl 90 elongation rates at the end of the night (Nozue et al. 2007), largely due to the combined actions 91 of the PIFq, which accumulate during the night and promote growth at dawn (Nozue et al. 2007, 92 Niwa et al. 2009, Soy et al. 2012, 2014). Transition to the morning light destabilizes the PIFs and 93 growth rate is rapidly reduced again (Nozue et al. 2007, Soy et al. 2012). Specific dawn-phased 94 PIF accumulation and activity in SD is regulated at several levels (Gommers and Monte 2018). 95 First, photoactivated Pfr imposes a decrease in PIFq proteins during the day and early night, 96 whereas slow progressive Pfr-to-Pr reversion in the dark allows reaccumulation of these PIFs 97 towards the end of the night (Monte et al. 2004, Shen et al. 2005, Nozue et al. 2007, Soy et al. 98 2012, Yamashino et al. 2013). Second, PIF4 and PIF5 expression is regulated by the circadian 99 clock, and their transcripts oscillate to peak at dawn, superimposed on the control of their protein 100 stabilization by light (Nozue et al. 2007, Nusinow et al. 2011, Yamashino et al. 2013). In contrast, 101 PIF1 and PIF3 transcript levels are relatively constant. Moreover, the capacity of PIFs to bind 102 DNA is inhibited by photoactive phyB (Park et al. 2018) and gibberellins/DELLAs (Arana et al. 103 2011). Finally, PIF activity is regulated by circadian clock components like PRRs, ELF3 or 104 GIGANTEA (Soy et al. 2016, Martín et al. 2018, Nieto et al. 2015, Zhang et al. 2020, Nohales et 105 al. 2019), or by the brassinosteroid pathway (Oh et al. 2012, Bernardo-García et al. 2014). 106 Of the PIFq proteins, PIF3 appears to play a more prominent role under SD (Soy et al. 107 2014). Growth promotion by PIF3 and the rest of the PIFq proteins involves direct induction of 108 the growth-related genes PIL1, XTR7 and HFR1 (Hornitschek et al. 2009, Soy et al. 2012, Leivar 109 et al. 2012b), as well as regulation of auxin- and other hormone-related genes (Nozue et al. 2011, 110 Nomoto et al. 2012). Importantly, the observation that pifq mutants still retained elongation 3 111 responses after a phy-inactivating end-of-day (EOD) FR pulse suggests redundancy with 112 additional PIFs (Leivar et al. 2012a, 2012b, Soy et al. 2014). 113 In contrast to PIFq, PIF7 protein is relatively light stable (Leivar et al. 2008a, Li et al. 114 2012), and phy mediates the accumulation of a phosphorylated form of PIF7 in high R/FR. This 115 phosphorylated form of PIF7 is retained in the cytosol by 14-3-3 proteins (Huang et al. 2018) and 116 is thus unable to bind to its target genes (Li et al. 2012). Under the low R/FR conditions of shade, 117 where PIF7 appears to play a major role (Li et al. 2012, de Wit et al. 2015, 2016), inactivation of 118 phy promotes nuclear accumulation of dephosphorylated PIF7 that is able to bind and induce the 119 expression of target genes in part by recruiting chromatin remodeling machinery (Li et al. 2012, 120 Peng et al. 2018). PIF7 is also involved in shade-induced flowering (Galvāo et al. 2019, Zhang et 121 al. 2019). In contrast to shade, the role of PIF7 under photoperiodic growth conditions remains 122 poorly understood. 123 Here, by genetic analysis of pif7 mutants, we show that PIF7 is a prominent target of 124 phyB-mediated suppression of growth at night, when it promotes growth in SD and especially 125 under SD-EODFR conditions. In order to assess the relative contributions of PIFq members and 126 PIF7, we examine the response of multiple pif mutants under different photoperiods and report 127 that the pifqpif7 quintuple mutant, lacking five PIFs (PIF1, 3, 4, 5 and 7), is largely insensitive to 128 the phy-mediated seasonal control of growth. Moreover, we show functional interactions between 129 PIF7 and PIF3 that might involve PIF7-PIF3 heterodimers. Finally, we examine how PIF7 is 130 regulated during the diurnal cycle at the transcriptional and protein level. 131 MATERIALS AND METHODS 132 Plant materials 133 Some of the mutant lines used in this study were described elsewhere including PIF7-FLASH (Li 134 et al. 2012), phyB-9 (Reed et al. 1993), pif3-3 (Monte et al. 2004), pif7-1, pif3pif7, phyBpif7 and 135 pif3pif4 (Leivar et al. 2008a), and pif3pif4pif5 and pif1pif3pif4pif5 (pifq) (Leivar et al. 2008b). 136 New mutant combinations included pif3pif4pif7, which was generated by crossing pif3pif4 and 137 pif3pif7, and pif3pif4pif5pif7 and the quintuple pifqpif7 mutant (Zhang et al. 2020), which were 138 generated by crossing pif3pif4pif7 and pifq. 139 Seedling growth and Measurements 140 Upon sterilization, seeds were plated in GM medium without sucrose and stratified for four days 141 in darkness (D) at 4ºC as described (Monte et al. 2003, Soy et al. 2014). Plates were then 142 transferred to either constant white light (WL, 85 µmol/m2s), long-days (LDs, 16hWL:8hD) or 143 short-days (SDs, 8hWL:16hD), or to LDs or SDs submitted to a 15min saturating FR pulse (30 144 µmol/m2s) at the end of the day period (LD-EODFR and SD-EODFR) before the night hours. 145 Seedlings were arranged horizontally on a plate and photographed using a digital camera (Nikon 146 D80) as described (Soy et al. 2014). Hypocotyl elongation was measured typically from at least 4 147 25 seedlings using NIH Image software (ImageJ, National Institutes of Health). Differences 148 between means were statistically analyzed by one-way analysis of variance using Tukey multiple 149 comparison test (GraphPad Prism6). Statistically significant differences were defined as those 150 with a p value < 0.05. 151 In vitro Co-immunoprecipitaton assay 152 In vitro coimmunoprecipitation experiments were essentially as described (Khanna et al. 2004). 153 Briefly, each protein was expressed from T7 promoters using the TnT in vitro 154 transcription/translation system (Promega). PIF7:GAD and GAD:PIF3 constructs were described 155 previously (Leivar et al. 2008a, Ni et al. 1998), whereas naked PIF7 corresponds to the full-length 156 open reading frame cloned into the pET17b vector using SacI and XhoI sites (Invitrogen, CA). 35 157 Proteins in each binding reaction were cosynthesized as S-Met-labeled products in TnT 158 reactions as specified. Signals were quantified with a Storm 860 PhosphorImager (Molecular 159 Dynamics). 160 Yeast-two hybrid assay 161 GAD alone or fused to full length PIF3 (GAD:PIF3) in pGAD424 (Ni et al. 1998) and GAL4 162 DNA Binding domain (GBD) alone or fused to full length PIF7 (GBD:PIF7) in pGBKT7 were 163 used for yeast two-hybrid interaction assays in yeast strains AH109 and Y187 following the 164 Clontech (Palo Alto, CA) Yeast Protocol Handbook. 165 Electrophoresis mobility shift assay (EMSA) 166 Electromobility shift assays were performed as described (Martínez-García et al. 2000), but with 167 pLUC control plasmid added to the binding reactions (Toledo-Ortiz et al. 2003). All proteins were 168 expressed from T7 promoters using the TnT in vitro transcription/translation system (Promega). 169 Naked PIF3 (Fairchild et al. 2000) and GAD:PIF3 (Ni et al. 1999) constructs were described 170 elsewhere, and naked PIF7 is described above. PIF7 and GAD:PIF3 were synthesized separately 171 (Fig. 3, lanes 4 and 3 respectively) or cosynthesized (Fig. 3 lane 5) in TnT reactions, and 3µL of 172 these TnT mixes were used for DNA binding. A total of 30,000 cpm of labeled G-box probe was 173 used in each lane. The binding conditions were as described (Leivar et al. 2008a). 174 Gene expression analysis 175 For gene expression, RNA extraction, cDNA synthesis and qRT-PCR were performed as 176 described (Sentandreu et al. 2011, Soy et al. 2012, 2014). In the time-course analysis performed 177 here, the gene expression was measured from three technical replicates for each time point and 178 genotype. PP2A (AT1G13320) was used as a normalization control as described (Shin et al. 2007, 179 Leivar et al. 2009). Primers for the gene expression analysis of PIL1 (AT2G46970), XTR7 180 (AT4G14130), HAT2 (AT5G47370), MIDA9 (AT5G02760) and HFR1 (AT1G02340) were 181 described elsewhere (Soy et al. 2012, 2016, Sentandreu et al. 2011). PIF7 (AT5G61270) 182 expression was measured using primers EMP538 (5’-GTTTCAGATGTCGTTGCTTGCA-3’) and 183 EMP539 (5’-TACCCATAGGAGGGACCATCAT-3’). 5 184 Protein Extraction and Immunoblots 185 Total protein extracts were obtained by resuspending grinded tissue samples in extraction buffer. 186 Extraction buffer and protein quantification were done as described (Leivar et al. 2008a, Soy et 187 al. 2016). PIF7-FLASH protein (Li et al. 2012) was detected after separation in a 7.5% SDS- 188 PAGE gel, and immunodetection was performed using mouse monoclonal anti-cMYC (SIGMA). 189 Peroxidase-linked anti-mouse antibody (Amersham Biosciences) were used as secondary 190 antibodies to detect the MYC epitope. Images were captured using a LAS4000 system. 191 Accession numbers 192 Arabidopsis Genome Initiative database accession numbers are: AT2G20180 (PIF1/PIL5), 193 AT1G09530 (PIF3), AT2G43010 (PIF4), AT3G59060 (PIF5/PIL6), AT5G61270 (PIF7), 194 AT2G18790 (phyB). 195 RESULTS 196 phyB suppresses diurnal growth by antagonizing PIF7 197 Analyses of hypocotyl length in Arabidopsis thaliana wild-type (WT) and pif7 mutant (Leivar et 198 al. 2008a) seedlings grown for 4 days in short days (SDs) or in constant white light (WL) (Fig. 199 1A and 1B), showed that the enhanced elongation in SD compared to WL previously reported 200 (Soy et al. 2012) depends on PIF7, since pif7 hypocotyls were clearly shorter than WT in SDs but 201 only marginally shorter in WL (Fig. 1B and 1C). In the phyB mutant background, pif7 mutation 202 partly suppressed the phyB elongated phenotype in both WL and SD (Fig. 1B and 1C). 203 Particularly in SD, quantification showed that the pif7 mutation in phyBpif7 mutants suppressed 204 a significant part of the striking SD-induced elongated phenotype of phyB mutants (3.1 mm vs 205 3.9 mm) compared to WT (1.1 mm) (Fig. 1D). Together, these results are consistent with an 206 antagonistic interplay between phyB and PIF7 in regulating growth under photoperiodic 207 conditions, similar to what was reported for PIF3 and other PIFq members (Soy et al. 2012, 2014), 208 and suggest that phyB suppression of growth in response to the long nights of SDs occurs partly 209 by inhibiting PIF7. 210 Analyses of photoperiodic conditions (SD and LD) supplemented with or without a 15- 211 min phy-inactivating FR-pulse (FRp) at the end of the day (end-of-day-far red (EODFR)) (Fig. 212 1A) revealed that the pif7 mutant displayed a short-hypocotyl phenotype only when the conditions 213 included extended periods under inactive phy Pr conditions, such as SD, LD-EODFR and SD- 214 EODFR (Fig. 1C, 1D). These data are consistent with recent reports that PIF7 induces growth 215 under long-term low R/FR (Li et al. 2012) or in response to prolonged (2-3h) low R/FR exposure 216 at EODFR under diurnal conditions (Mizuno et al. 2015, Jiang et al. 2019). However, in contrast 217 to these reports where the elongation phenotype is mainly dependent on PIF7 (Mizuno et al. 2015, 218 Jiang et al. 2019), in our brief saturating FRp-promoted EOD-FRp in SD schedule, the pif7 mutant 219 still retained significant growth responses, indicating that other PIF members likely contribute to 6 220 growth in our SD-EOD-FR conditions. Consistent with phyB being the major photoreceptor in 221 perceiving the EODFR signal, the elongated hypocotyl phenotype of phyB was strongly 222 attenuated in response to EODFR, especially under SD-EODFR conditions (Fig. 1C, 1D). In all 223 the conditions tested, we observed that the pif7 mutation partly suppressed the phyB mutant 224 phenotype in a quantitatively similar manner (Fig. 1C, 1D), consistent with the view that PIF7 is 225 a target of phyB in suppressing growth in WL and under photoperiodic conditions. 226 Photoperiodic control of growth is explained by the collective action of five PIFs 227 PIF1, PIF3, PIF4, and PIF5 have previously been reported to play a role in promoting growth 228 under diurnal and EODFR conditions, with PIF3 being the main contributor to SD growth (Leivar 229 et al. 2012a, 2012b, Soy et al. 2014, Nozue et al. 2007, Niwa et al. 2009). To assess the function 230 of PIF7 in promoting photoperiodic growth, we first compared pif3, pif7, and pif3pif7 mutants 231 (Monte et al. 2004, Leivar et al. 2008a) under the photoperiodic conditions detailed in Fig. 1A, 232 including EODFR treatments (Fig. 2A and 2B). Interestingly, while pif3 was shorter than pif7 233 under SD, pif7 was clearly shorter than pif3 under SD- and LD-EODFR, conditions where pif3pif7 234 showed additive to synergistic effects (Fig. 2A, 2B, and 2C). Next, to investigate the role of PIF7 235 in the absence of multiple PIFq members, we generated a series of higher order pif mutants lacking 236 PIF7 such as pif3pif4pif7 and pif3pif4pif5pif7, and the quintuple mutant pifqpif7 that lacked all 237 five PIFs. The hypocotyl elongation of these mutants and the corresponding controls (namely 238 pif3pif4, pif3pif4pif7, pif3pif4pif5, pif3pif4pif5pif7, pifq, pifqpif7) was examined under WL, LD, 239 SD, or SD/LD-EODFR (Fig. 1A). Our results showed that they were: (1) only marginally shorter 240 than WT in WL and LD (Fig. 2B and 2D); (2) shorter than WT under the long night of SDs, with 241 no significant growth response compared to WL or LDs (Fig. 2C) and no additive effects as all 242 the higher order pif mutants were similarly short in SDs (Fig. 2B and 2C); (3) shorter than WT 243 under LD-EODFR conditions, with a marginal growth response in pif3pif4, pif3pif4pif5, and pifq 244 mutants that was largely suppressed by the genetic removal of PIF7 in pif3pif4pif7 mutants, 245 pif3pif4pif5pif7, and pifqpif7 mutants (Fig. 2B and 2C). Finally, under SD-EODFR (Fig. 1A), the 246 strong growth of WT seedlings was attenuated to varying degrees in all higher order pif mutants 247 (Fig. 2B and 2D), suggesting collective participation of all PIFs under these conditions. 248 Interestingly, all higher order pif mutants carrying WT PIF7 (pif3pif4, pif3pif4pif5, and pifq) 249 showed a limited but still significant growth response (Fig. 2B and 2C). Genetic removal of PIF7 250 further reduced the growth response (Fig. 2B and 2C), although pif3pif4pif7 and pif3pif4pif5pif7 251 still exhibited a residual growth response to the EODFR treatment indicating that the remaining 252 PIFs (PIF1 or PIF5) are still active. Strikingly, the quintuple pifqpif7 mutant was almost 253 completely insensitive to the EODFR treatment (Fig. 2B and 2C). Together, these data indicate 254 that the relative contribution of PIF7 to diurnal hypocotyl elongation is enhanced by phy 7 255 inactivation, and that the collective action of PIF1, 3, 4, 5, and 7 is sufficient to explain the 256 photoperiodic regulation of seasonal growth. 257 PIF7 can form heterodimers with PIF3 that bind to DNA in vitro 258 PIF7 has been shown to form heterodimers with PIF1 and PIF4 in vitro (Castillon et al. 2007, 259 Kidokoro et al. 2009, Fiorucci et al. 2019, Toledo-ortiz et al. 2003, Bu et al. 2011). To further 260 assess the inter-PIF7 heterodimerization landscape, and given the prominent role of PIF3 in the 261 regulation of growth under diurnal conditions (Soy et al. 2012, 2014) and the observed synergistic 262 contributions of PIF3 and PIF7 to hypocotyl elongation (Fig. 2), we tested possible PIF7-PIF3 263 heterodimerzation. Indeed, in vitro pull-down assays confirmed direct interaction between PIF7 264 and PIF3 (Fig. 3A) that was confirmed in yeast two-hybrid in vivo assays (Fig. 3B). PIF7 was 265 also shown to form homodimers in vitro (Fig. 3A), consistent with previous reports (Kidokoro et 266 al. 2009, Fiorucci et al. 2019). Furthermore, using EMSA assays, we detected binding of PIF7 to 267 G-box DNA elements as homodimer and as a PIF7-PIF3 heterodimer (Fig 3C). These results 268 indicate that in vitro, PIF7 can heterodimerize with PIF3 and that this dimeric PIF7-PIF3 can bind 269 to DNA, suggesting that functional interaction between PIF7 and PIF3 might involve heterodimer 270 formation. 271 Contrasting developmental and temporal growth-promoting activities of PIF3 and PIF7 in 272 SD and SD-EODFR. 273 To further understand the interplay between PIF3 and PIF7 along early growth in SD and SD- 274 EODFR, we performed time-course phenotypic analysis at days 2, 3 and 4 (in contrast to the end- 275 point measurements in 4d-old seedlings shown in Fig. 2). Under SDs, we observed that the short 276 hypocotyl phenotype of pif3 mutants was already established at day 2 (Fig. 4A) and maintained 277 throughout the rest of the treatment, in agreement with previous data (Soy et al. 2012), whereas 278 the short hypocotyl phenotype of pif7 mutants was not manifested until day 4. In contrast, under 279 SD-EODFR, the additive contributions of PIF3 and PIF7 were clearly observed by day 2 and 280 throughout days 3 and 4 (Fig. 4A). Because some of the phenotypes are observed as early as 2 281 days, one possibility is that the participation of PIF3 is exclusively determined during early 282 deetiolation. To test this possibility, we modified the experimental set up so that seedlings were 283 first deetiolated for 2 days in WL, and then were kept for 4 additional days in either WL, SD or 284 SD-EODFR. As shown in Fig. 4B, the observed phenotypes were roughly similar to Fig. 2: (1) 285 WT hypocotyl induction in response to SD and SD-EODFR; (2) A more prominent short 286 hypocotyl phenotype of pif3 mutants in SD compared to pif7; (3) A more prominent short 287 hypocotyl phenotype of pif7 mutants in SD-EODFR compared to pif3; and (4) Additive effects 288 observed in pif3pif7 double mutants, in this case in both SD and SD-EODFR. These data suggest 289 that the interplay between PIF3 and PIF7 under the tested photoperiodic conditions is not 290 exclusively established early during early deetiolation, and instead is sustained in green seedlings. 8 291 The experiments presented above suggest that, similar to what has been reported for PIF3 292 and other PIFq members (Soy et al. 2012, 2014, Nozue et al. 2007, Monte et al. 2004, Shen et al. 293 2005), PIF7 acts to induce growth during the dark hours under diurnal conditions after phy 294 repression is lifted. In order to test this possibility, we performed phenotypic measurements 295 during the night period under SD conditions. First, pif3 and pif7 single and double mutants were 296 grown for 2 days in SD, and during the 3rd day in SD the hypocotyl elongation was measured: (1) 297 at ZT8, the end of the light period (W8); and (2) at the end of the night period (ZT24) in samples 298 that had been (FR24) or not (D24) treated with an EODFR (Fig. 4C, left panel). Consistent with 299 Fig. 4A, WT seedlings grew over the dark period (D24) of the 3rd day in SD in a strong PIF3- 300 dependent manner with marginal to null contributions of PIF7 (Fig. 4C, middle and right panels), 301 whereas EODFR-induced growth (FR24) was additively dependent on both PIF3 and PIF7. 302 PIF7, together with PIF3, acts to induce growth-related gene expression during the dark 303 hours in SDs 304 In order to determine the interplay between PIF3 and PIF7 in regulating gene expression during 305 the night period, we performed time-course experiments of pif3 and pif7 mutants grown in SD 306 and in SD-EODFR following the same experimental design of Fig. 4C. In these experiments, 307 samples were taken at the end of the light period of the 3rd day in SDs (ZT8; W8), and during the 308 night in samples that had (FR) or not (D) been treated with an EODFR at ZT9 (D9, FR9), ZT12 309 (D12, FR12), ZT16 (D16, FR16), ZT20 (D20, FR20), and ZT24 (D24, W24). We also included 310 a time point upon 1h illumination (ZT1, W25) at the 4th day in SDs. Initially we measured the 311 expression of PIL1, HFR1 and XTR7, genes that are directly induced by PIF3 during the night 312 period in SDs (Soy et al. 2012), and in the case of PIL1 and HFR1, they have also been shown to 313 be regulated by PIF7 in response to shade (Li et al. 2012). The data show that under SDs, there is 314 a progressive accumulation of PIL1, HFR1, and XTR7 transcripts during the night in WT 315 seedlings (Fig. 5A), which is strongly attenuated in pif3 mutants as reported (Soy et al. 2012). 316 pif7 mutants also had reduced expression compared to WT, although it was not as strong and was 317 more evident towards the end of the night (D24). Additive effects in pif3pif7 double mutants were 318 only marginally observed for HFR1 (Fig. 5A). In sharp contrast, an EODFR treatment rapidly 319 induced the expression of these genes in WT seedlings to expression values similar or even higher 320 to those at the end of the night in SDs (D24, indicated as a horizontal dotted line) (Fig. 5B). The 321 relative contribution of PIF3 and PIF7 in promoting early EODFR-induced gene expression 322 varied in each case. For PIL1, a minor contribution of PIF7 and a more robust effect of PIF3 was 323 observed, a situation that was reversed in XTR7, where the contribution of PIF7 was more 324 prominent (Fig. 5B). In both cases, small additive effects between both PIFs were observed. For 325 HFR1, the contributions of PIF3 and PIF7 were more similar, and no additive effects were 326 observed. These results were somewhat surprising considering the prominent role of PIF7 in 9 327 promoting growth at least under EODFR conditions (Fig. 4C), and in inducing some of these 328 genes like PIL1 in response to shade (Li et al. 2012), and suggested that under short day PIF7 329 might preferentially target a different set of genes. Therefore we extended this analysis by 330 screening the list of shade-induced PIF7-dependent genes (Li et al. 2012), and identified two 331 candidate genes, the transcription factor HAT2 (AT5G47370) and the protein phosphatase 2C 332 MIDA9 (AT5G02760), which regulate different facets of photomorphogenesis (Sentandreu et al. 333 2011, 2012, Bou-Torrent et al. 2012) and show a diurnal pattern of expression under SDs similar 334 to PIL1, HFR1 and XTR7 (http://diurnal.mocklerlab.org). In agreement, our data showed that 335 under SDs (Fig. 5A), expression of HAT2 and MIDA9 increased towards the end of the night in 336 WT seedlings (ZT24; D24) and decreased upon 1h illumination in the next day (ZT25; W25). 337 Expression analysis showed that the expression levels were strongly attenuated at the end of the 338 night in pif7 mutants, with minor to absent effects in the pif3 mutant (Fig. 5A), in sharp contrast 339 to PIL1, HFR1 and XTR7. Under SD-EODFR conditions, the rapid and transient induction of 340 HAT2 and MIDA9 was also strongly dependent on PIF7, whereas the pif3 mutant was unaffected 341 (Fig. 5B). Whereas the pif3pif7 mutant was indistinguishable from pif7 mutants under SD, in SD- 342 EODFR small additive or synergistic effects were observed at least for MIDA9. Together, these 343 data suggest that the molecular phenotype in SD and SD-EODFR is established by complex 344 functional interactions between PIF3 and PIF7 that vary from gene to gene. Some of these 345 combinatorial activities between PIF3 and PIF7, especially under EODFR conditions, might 346 involve the formation of heterodimers as shown above (Figure 3). Previous data have indicated 347 that PIF3 accumulates progressively along the night to peak at dawn (Soy et al. 2012). phyB 348 activity keeps PIF3 levels low at the beginning of the night, and inactivation of phyB by an EOD- 349 FR induces higher PIF3 accumulation (Soy et al. 2012, 2014). 350 PIF7 protein mobility fluctuates in response to SD and SD-EODFR 351 It has been previously reported that the PIF7 transcript oscillates under diurnal conditions in adult 352 plants (Lee and Thomashow 2012) and in free running conditions (Kidokoro et al. 2009). In 353 agreement with these reports, we also observed that PIF7 transcript oscillates in young seedlings 354 growing under diurnal short day (SD) conditions (Fig. 6A). In addition, it is well known that phy 355 induces the accumulation of an inactive PIF7 phosphorylated form in high R/FR that presents a 356 reduced mobility compared to the dephosphorylated form (Li et al. 2012). These studies show 357 that the PIF7-phosphorylated form accumulates in light, but it dephosphorylates rapidly in 358 response to shade, and this activated form of PIF7 accumulates in the nucleus to induce gene 359 expression (Huang et al. 2018). Here we use the PIF7-flash lines (Li et al. 2012) to study 360 accumulation of PIF7 in SD and in SD-EODFR. At the end of the light period in SD (W8) we 361 observed a band with lower mobility, consistent with the described inactive phosphorylated form 362 (Fig. 6B and 6C). In contrast, during the night, the fast migrating band (the active 10 363 dephosphorylated form) accumulates with a major peak (Niwa et al. 2009) at D24 compared to 364 D9 (Fig. 6B). Results are consistent with gradual accumulation of the dephosphorylated form 365 during the night under SD (Jiang et al. 2019). We also observe a rapid appearance of the slow 366 migrating band upon switching lights on at ZT1 of the following day (W1) (Fig. 6B). In contrast, 367 in response to EODFR, we observe a rapid complete disappearance of the slow migrating band at 368 ZT9 (FR9) compared to SD (D9) (Fig. 6C). This EOD-FR pulse did not affect the overall pattern 369 of accumulation of the PIF7 transcript along the night, although levels were somewhat reduced 370 compared to SD (Fig. 6A), in accordance to previous reports (Mizuno et al. 2015). These results 371 suggest that inactivation of phyB by EODFR rapidly decreases the accumulation of the inactive 372 phosphorylated form of PIF7 to induce growth under these conditions. 373 DISCUSSION 374 Previously, using a combination of SD and SD-EODFR light regimes, we reported the concerted 375 action of PIF1, PIF3, PIF4, and PIF5 (the PIF quartet or PIFq) in promoting and optimizing 376 hypocotyl elongation in diurnal conditions (Soy et al. 2014). However, because the pifq mutant 377 still retained some responsiveness to SD-EODFR, participation of additional factors was 378 suggested. The experiments presented here examine the contribution of PIF7 to seedling growth 379 under diurnal conditions. In response to the long nights of SD photoperiods, PIF7 induces growth 380 together with PIF3 and other PIFq members, whereas PIF7 activity is suppressed by phyB in 381 constant WL or LD. Using a combination of LD, LD-EODFR, SD, and SD-EODFR experiments 382 (Fig. 1A) in higher-order mutant seedlings lacking up to five PIFs, our data indicate that PIF7 is 383 a prominent target of phyB-mediated suppression of growth under diurnal conditions. 384 Furthermore, we establish that the growth-promotion activity of PIF7 together with the PIFq is 385 likely sufficient to explain the photoperiodic control of seasonal growth in Arabidopsis. Finally, 386 we show that diurnal oscillation of PIF7 activity involves increased activation by 387 dephosphorylation during the dark hours. Consistent with these findings, our EODFR 388 experiments establish that residual phy Pfr inhibits full activation of PIF7 during the long nights 389 of SD. 390 The phenotypic and marker gene expression analysis of pif7 single and higher order 391 mutants presented here identify PIF7 as an additional PIF factor promoting growth under diurnal 392 conditions through regulation of growth-related gene expression. PIFq function is antagonistic to 393 the growth suppressing activity of phyB, which targets the PIFq for degradation, not only during 394 the day but also during early night due to the relatively slow Pfr-to-Pr dark reversion rate (Soy et 395 al. 2012, 2014, Sweere et al. 2001, Rausenberger et al. 2010), effectively suppressing growth in 396 non-inductive conditions of LD (Fig. 1 and 2; (Niwa et al. 2009)). Consistently, a phy-inactivating 397 EODFR pulse before the beginning of the 16 h night period in SDs was shown to promote 398 accumulation of PIF3 and possibly other PIF proteins (Soy et al. 2014, Shen et al. 2005), and to 11 399 enhance hypocotyl growth by 3-fold during the night period (Soy et al. 2014). Our observations 400 here suggest that PIF7, similarly to PIFq, promotes growth in response to the long nights of SD 401 (Fig. 1 and 2), also in antagonistic fashion toward phyB, as shown by the partial suppression of 402 the phyB elongated phenotype in the phyBpif7 mutant (Fig. 1). Interestingly, our observation that 403 the contribution of PIF7 to growth is not as strong as PIF3 under SD, but that it appears to 404 contribute more robustly than PIF3 under SD-EODFR (Fig. 2 and 4), suggests that inhibition of 405 PIF7 by phyB during early night is central to suppression of growth under these conditions. 406 The observation that the apparent relative contribution of PIF7 and PIF3 to full induction 407 of growth-related genes under SD varies among genes (Fig. 5) suggests that PIF7 and PIF3 have 408 both shared and specific regulatory functions: whereas PIF3 dominates expression of PIL1, 409 HFR1, and XTR7, PIF7 is a strong contributor to the induction of HAT2 and MIDA9, with non- 410 detectable to minor effects of PIF3 (Fig. 5A). In contrast to SD, examination of marker gene 411 expression under SD-EODFR showed a robust increase with respect to SD in accordance to 412 previous data (Soy et al. 2014), and revealed an enhanced relative contribution of PIF7 compared 413 to PIF3 for some of the growth-related marker genes like XTR7 (Fig. 5B). These results agree 414 with the increased contribution of PIF7 function to hypocotyl growth under SD-EODFR 415 compared to SD, and suggest that PIF7 role might be relatively more important in conditions 416 where it has been suggested that seedlings experience a partial reversion to the etiolated state, like 417 SD-EODFR (Soy et al. 2014) as shown in this work, or in shade-induced responses (Leivar et al. 418 2012b), where PIF7 plays a prominent role (Li et al. 2012, Mizuno et al. 2015, Jiang et al. 2019). 419 Also in agreement, a recent paper reported that several shade-induced genes display remarkable 420 divergence in dependency on the PIF quartet and PIF7 for their shade responsiveness, indicating 421 a spectrum of combinatorial activities toward these genes (Zhang et al. 2020). Part of the observed 422 functional interaction between PIF7 and PIFq could involve heterodimer formation, including the 423 newly described PIF7-PIF3 heterodimers (Fig. 3). 424 The results presented here indicate that sequential removal of PIFs in pif3pif4, 425 pif3pif4pif7, pif3pif4pif5, pif3pif4pif5pif7, pifq and pifqpif7 mutants progressively decreased the 426 sensitivity to longer night periods and the associated growth response (Fig. 2). Our striking 427 finding that a pifqpif7 mutant remains very short in all photoperiodic conditions examined, and is 428 nearly insensitive for hypocotyl elongation to SD-EODFR treatment (Fig. 2), indicates that the 429 collective action of PIF7 and the PIFq can account for the implementation of the phy-regulated 430 control of growth under diurnal conditions, and strongly suggests that the collective contribution 431 of these five PIFs is required for the photoperiodic-regulated seasonal acceleration of growth. A 432 recent observation that the pifqpif7 quintuple mutant is unresponsive to long term exposure to low 433 R/FR (Zhang et al. 2020) is consistent with our results. Together these findings establish that the 434 five PIF-quintet members (PIFs1, -3, -4, -5, and -7) are collectively fully responsible for 12 435 promoting hypocotyl cell elongation in response to environments where there is a fluctuation in 436 Pr/Pfr phy forms such as shade and diurnal conditions. 437 Our observation that PIF7 transcript oscillates in SD (Fig. 6A), suggests that PIF7 438 transcription is probably regulated by the circadian clock. Indeed, Kidokoro et al. (Kidokoro et 439 al. 2009) showed a similar PIF7 oscillatory pattern under free running conditions, a hallmark of 440 regulation by the clock. This is similar to PIF4 and PIF5, which also oscillate in SD under clock 441 regulation, but in contrast to PIF1 and PIF3, which show relatively constant transcript levels 442 across the day and night under SD. Despite these similarities with PIF4 and PIF5, regulation of 443 PIF7 protein stability is unique compared to the PIFq members. Indeed, the rapid light-induced 444 interaction of PIF7 with photoactivated phyB does not lead to robust degradation of PIF7 (Leivar 445 et al. 2008a, Lee and Thomashow 2012, Li et al. 2012). Instead, it triggers accumulation of a 446 cytosolic, phosphorylated form of PIF7 under constant white light, which is rapidly 447 dephosphorylated and transported to the nucleus in response to phy inactivation, such as in FR- 448 enriched shade environments (low R/FR conditions) (Li et al. 2012, Peng et al. 2018). This 449 dephosphorylated PIF7 can bind to genomic target sites and to chromatin remodeling machinery 450 to regulate the expression of its target genes (Li et al. 2012, Huang et al. 2018), and can be rapidly 451 photoreversed to the phosphorylated form when returned to white light (Li et al. 2012). Our results 452 here in alternating day/night conditions, extend these observations to show that phy imposes 453 oscillation of PIF7 activity under SD (Fig. 6): during the day, phy Pfr triggers accumulation of 454 the inactive phosphorylated form of PIF7, whereas in the night reversion to Pr favors 455 accumulation of the dephosphorylated form, which is active in the induction of gene expression 456 (Fig. 5) and in the promotion of hypocotyl elongation, consistent with recent findings (Jiang et al. 457 2019). 458 Collectively, the data presented here indicate that PIF7, together with the PIFq, promote 459 acceleration of hypocotyl elongation in response to photoperiod shortening (i.e. longer nights). 460 The present work suggests that phy-mediated regulation of PIFq stability and PIF7 activity is a 461 central regulatory pathway in conferring day-length sensitivity, in order to ensure growth in the 462 appropriate season. Interestingly, our finding that phytochrome mediates inhibition of growth 463 under SD by suppressing PIF7 activity (as revealed by our EODFR treatments) could represent a 464 safety mechanism to prevent overgrowth while retaining a significant capacity to integrate and 465 respond to situations in SD where additional rapid elongation is needed. Such scenarios could 466 include shading or an increase in temperature. Indeed, high temperature has been shown to trigger 467 fast Pr to Pfr thermal reversion of phyB (Legris et al. 2016, Jung et al. 2016), raising the levels of 468 active PIF7 and increasing PIF4 accumulation, the two main PIFs involved in 469 thermomorphogenesis downstream of phyB (Koini et al. 2009, Franklin et al. 2011, Lee and 470 Thomashow 2012, Fiorucci et al. 2019). Together, the phyB/PIF7 module might equip seedlings 13 471 growing under diurnal conditions with the capacity to rapidly optimize their growth to changes 472 in light quality or temperature. 473 AUTHOR CONTRIBUTIONS 474 P.L., and E.M. conceived and designed the study. P.L., G.M., J.S., J.D-R, P.H.Q., and E.M., 475 acquired, analyzed, and interpreted data. P.L., and E.M. wrote the manuscript. 476 ACKNOWLEDGEMENTS 477 We thank M. Cohn and C. Coates for early work on the project, and J. Chory, Ullas Pedmale and 478 L. Li for PIF7-flash lines. This work was supported by Marie Curie International Reintegration 479 Grant PIRG06-GA-2009-256420 and by funds from Universitat Ramon Llull/Obra Social la 480 Caixa (2016-URL-Internac-019, 2018-LC-05 and 2018-URL-IR2nQ-019) to P.L., by NIH 481 (5R01GM047475-24) and USDA ARS Current Research Information System (2030-21000-051- 482 00D) grants to P.H.Q., and by grants from the Spanish ‘‘Ministerio de Economía y 483 Competitividad’’ (MINECO) (BIO2012-31672 and BIO2015-68460-P), by FEDER / Ministerio 484 de Ciencia, Innovación y Universidades – Agencia Estatal de Investigación (Project Reference 485 BIO2015-68460-P) and from the CERCA Programme / Generalitat de Catalunya (Project 486 References 2014-SGR-1406 and 2017SGR-718) to E.M. We acknowledge financial support by 487 the CERCA programme/Generalitat de Catalunya and from MINECO through the ‘‘Severo 488 Ochoa Programme for Centers of Excellence in R&D’’ 2016-2019 (SEV-2015-0533). 489 490 REFERENCES 491 Al-Sady B, Ni W, Kircher S, Schäfer E, Quail PH (2006) Photoactivated Phytochrome Induces 492 Rapid PIF3 Phosphorylation Prior to Proteasome-Mediated Degradation. Mol Cell 23: 439– 493 446 494 Arana MV, Marín-De La Rosa N, Maloof JN, Blázqueza MA, Alabadí D (2011) Circadian 495 oscillation of gibberellin signaling in Arabidopsis. Proc Natl Acad Sci U S A 108: 9292– 496 9297 497 Bauer D, Viczián A, Kircher S, Nobis T, Nitschke R, Kunkel T, Panigrahi KCS, Ádám É, Fejes 498 E, Schäfer E, Nagy F (2004) Constitutive photomorphogenesis 1 and multiple 499 photoreceptors control degradation of phytochrome interacting factor 3, a transcription 500 factor required for light signaling in Arabidopsis. Plant Cell 16: 1433–1445 501 Bernardo-García S, de Lucas M, Martínez C, Espinosa-Ruiz A, Davière J-M, Prat S (2014) BR- 502 dependent phosphorylation modulates PIF4 transcriptional activity and shapes diurnal 503 hypocotyl growth. Genes Dev 28: 1681–1694 504 Bou-Torrent J, Salla-Martret M, Brandt R, Musielak T, Palauqui J-C, Martínez-García JF, 505 Wenkel S (2012) ATHB4 and HAT3, two class II HD-ZIP transcription factors, control leaf 506 development in Arabidopsis. Plant Signal Behav 7: 1–6 14 507 Bu Q, Castillon A, Chen F, Zhu L, Huq E (2011) Dimerization and blue light regulation of PIF1 508 interacting bHLH proteins in Arabidopsis. Plant Mol Biol 77: 501–511 509 Castillon A, Shen H, Huq E (2007) Phytochrome Interacting Factors: central players in 510 phytochrome-mediated light signaling networks. Trends Plant Sci 12: 514–521 511 Fairchild CD, Schumaker MA, Quail PH (2000) HFR1 encodes an atypical bHLH protein that 512 acts in phytochrome A signal transduction. Genes Dev 14: 2377–2391 513 Fiorucci AS, Costa Galvão V, Ince YÇ, Boccaccini A, Goyal A, Allenbach Petrolati L, Trevisan 514 M, Fankhauser C (2019) PHYTOCHROME INTERACTING FACTOR 7 is important for 515 early responses to elevated temperature in Arabidopsis seedlings. New Phytol: [Epub ahead 516 of print] 517 Franklin KA, Lee SH, Patel D, Kumar SV, Spartz AK, Gu C, Ye S, Yu P, Breen G, Cohen JD, 518 Wigge PA, Gray WM (2011) Phytochrome-Interacting Factor 4 (PIF4) regulates auxin 519 biosynthesis at high temperature. Proc Natl Acad Sci U S A 108: 20231–20235 520 Galvāo VC, Fiorucci AS, Trevisan M, Franco-Zorilla JM, Goyal A, Schmid-Siegert E, Solano R, 521 Fankhauser C (2019) PIF transcription factors link a neighbor threat cue to accelerated 522 reproduction in Arabidopsis. Nat Commun 10: 4005 523 Gommers CMM, Monte E (2018) Seedling establishment: A dimmer switch-regulated process 524 between dark and light signaling. Plant Physiol 176: 1061–1074 525 Hornitschek P, Lorrain S, Zoete V, Michielin O, Fankhauser C (2009) Inhibition of the shade 526 avoidance response by formation of non-DNA binding bHLH heterodimers. EMBO 527 JournalEMBO J 28: 3893–3902 528 Huang X, Zhang Q, Jiang Y, Yang C, Wang Q, Li L (2018) Shade-induced nuclear localization 529 of PIF7 is regulated by phosphorylation and 14-3-3 proteins in arabidopsis. Elife 7: 1–17 530 Jiang Y, Yang C, Huang S, Xie F, Xu Y, Liu C, Li L (2019) The ELF3-PIF7 interaction mediates 531 the circadian gating of the shade response in Arabidopsis. Iscience 22: 288–298 532 Jung JH, Domijan M, Klose C, Biswas S, Ezer D, Gao M, Khattak AK, Box MS, Charoensawan 533 V, Cortijo S, Kumar M, Grant A, Locke JC, Schäfer E, Jaeger KE, Wigge PA (2016) 534 Phytochromes function as thermosensors in Arabidopsis. Science (80- ) 354: 886–889 535 Khanna R, Huq E, Kikis EA, Al-Sady B, Lanzatella C, Quail PH (2004) A novel molecular 536 recognition motif necessary for targeting photoactivated phytochrome signaling to specific 537 basic helix-loop-helix transcription factors. Plant Cell 16: 3033–3044 538 Kidokoro S, Maruyama K, Nakashima K, Imura Y, Narusaka Y, Shinwari ZK, Osakabe Y, Fujita 539 Y, Mizoi J, Shinozaki K, Yamaguchi-Shinozaki K (2009) The phytochrome-interacting 540 factor PIF7 negatively regulates dreb1 expression under circadian control in Arabidopsis. 541 Plant Physiol 151: 2046–2057 542 Koini MA, Alvey L, Allen T, Tilley CA, Harberd NP, Whitelam GC, Franklin KA (2009) High 543 Temperature-Mediated Adaptations in Plant Architecture Require the bHLH Transcription 15 544 Factor PIF4. Curr Biol 19: 408–413 545 Lee CM, Thomashow MF (2012) Photoperiodic regulation of the C-repeat binding factor (CBF) 546 cold acclimation pathway and freezing tolerance in Arabidopsis thaliana. Proc Natl Acad 547 Sci U S A 109: 15054–15059 548 Legris M, Klose C, Burgie ES, Costigliolo C, Neme M, Hiltbrunner A, Wigge PA, Schäfer E, 549 Vierstra RD, Casal JJ (2016) Phytochrome B integrates light and temperature signals in 550 Arabidopsis. Science (80- ) 354: 897–900 551 Leivar P, Monte E (2014) PIFs: systems integrators in plant development. Plant Cell 26: 56–78 552 Leivar P, Monte E, Al-Sady B, Carle C, Storer A, Alonso JM, Ecker JR, Quail PH (2008a) The 553 Arabidopsis phytochrome-interacting factor PIF7, together with PIF3 and PIF4, regulates 554 responses to prolonged red light by modulating phyB levels. Plant Cell 20: 337–352 555 Leivar P, Monte E, Cohn MM, Quail PH (2012a) Phytochrome signaling in green Arabidopsis 556 seedlings: impact assessment of a mutually negative phyB-PIF feedback loop. Mol Plant 5: 557 734–749 558 Leivar P, Monte E, Oka Y, Liu T, Carle C, Castillon A, Huq E, Quail PH (2008b) Multiple 559 Phytochrome-Interacting bHLH Transcription Factors Repress Premature Seedling 560 Photomorphogenesis in Darkness. Curr Biol 18: 1815–1823 561 Leivar P, Tepperman JM, Cohn MM, Monte E, Al-Sady B, Erickson E, Quail PH (2012b) 562 Dynamic antagonism between phytochromes and PIF family basic helix-loop-helix factors 563 induces selective reciprocal responses to light and shade in a rapidly responsive 564 transcriptional network in Arabidopsis. Plant Cell 24: 1398–1419 565 Leivar P, Tepperman JM, Monte E, Calderon RH, Liu TL, Quail PH (2009) Definition of early 566 transcriptional circuitry involved in light-induced reversal of PIF-imposed repression of 567 photomorphogenesis in young Arabidopsis seedlings. Plant Cell 21: 3535–3553 568 Li L, Ljung K, Breton G, Schmitz RJ, Pruneda-Paz J, Cowing-Zitron C, Cole BJ, Ivans LJ, 569 Pedmale UV, Jung HS, Ecker JR, Kay SA, Chory J (2012) Linking photoreceptor excitation 570 to changes in plant architecture. Genes Dev 26: 785–790 571 Martín G, Rovira A, Veciana N, Soy J, Toledo-Ortiz G, Gommers CMM, Boix M, Henriques R, 572 Minguet EG, Alabadí D, Halliday KJ, Leivar P, Monte E (2018) Circadian Waves of 573 Transcriptional Repression Shape PIF-Regulated Photoperiod-Responsive Growth in 574 Arabidopsis. Curr Biol 28: 311-318.e5 575 Martínez-García JF, Huq E, Quail PH (2000) Direct targeting of light signals to a promoter 576 element-bound transcription factor. Science (80- ) 288: 859–863 577 Mizuno T, Oka H, Yoshimura F, Ishida K, Yamashino T (2015) Insight into the mechanism of 578 end-of-day far-red light (EODFR)-induced Shade avoidance responses in Arabidopsis 579 thaliana. Biosci Biotechnol Biochem 79: 1987–1994 580 Monte E, Alonso JM, Ecker JR, Zhang Y, Li X, Young J, Austin-Phillips S, Quail PH (2003) 16 581 Isolation and characterization of phyC mutants in Arabidopsis reveals complex crosstalk 582 between phytochrome signaling pathways. Plant Cell 15: 1962–1980 583 Monte E, Tepperman JM, Al-Sady B, Kaczorowski KA, Alonso JM, Ecker JR, Li X, Zhang Y, 584 Quail PH (2004) The phytochrome-interacting transcription factor, PIF3, acts early, 585 selectively, and positively in light-induced chloroplast development. Proc Natl Acad Sci U 586 S A 101: 16091–16098 587 Ni M, Tepperman JM, Quail PH (1999) Binding of phytochrome B to its nuclear signalling 588 partner PIF3 is reversibly induced by light. Nature 400: 781–784 589 Ni M, Tepperman JM, Quail PH (1998) PIF3, a phytochrome-interacting factor necessary for 590 normal photoinduced signal transduction, is a novel basic helix-loop-helix protein. Cell 95: 591 657–667 592 Nieto C, López-Salmerón V, Davière JM, Prat S (2015) ELF3-PIF4 interaction regulates plant 593 growth independently of the evening complex. Curr Biol 25: 187–193 594 Niwa Y, Yamashino T, Mizuno T (2009) The circadian clock regulates the photoperiodic 595 response of hypocotyl elongation through a coincidence mechanism in arabidopsis thaliana. 596 Plant Cell Physiol 50: 838–854 597 Nohales MA, Liu W, Duffy T, Nozue K, Sawa M, Pruneda-Paz JL, Maloof JN, Jacobsen SE, Kay 598 SA (2019) Multi-level Modulation of Light Signaling by GIGANTEA Regulates Both the 599 Output and Pace of the Circadian Clock. Dev Cell 49: 840-851.e8 600 Nomoto Y, Kubozono S, Yamashino T, Nakamichi N, Mizuno T (2012) Circadian clock-and 601 PIF4-controlled plant growth: A coincidence mechanism directly integrates a hormone 602 signaling network into the photoperiodic control of plant architectures in arabidopsis 603 thaliana. Plant Cell Physiol 53: 1950–1964 604 Nozue K, Covington MF, Duek PD, Lorrain S, Fankhauser C, Harmer SL, Maloof JN (2007) 605 Rhythmic growth explained by coincidence between internal and external cues. Nature 448: 606 358–361 607 Nozue K, Harmer SL, Maloof JN (2011) Genomic analysis of circadian clock-, light-,and growth- 608 correlated genes reveals PHYTOCHROME-INTERACTING FACTOR5 as a modulator of 609 auxin signaling in arabidopsis. Plant Physiol 156: 357–372 610 Nusinow DA, Helfer A, Hamilton EE, King JJ, Imaizumi T, Schultz TF, Farré EM, Kay SA 611 (2011) The ELF4-ELF3-LUX complex links the circadian clock to diurnal control of 612 hypocotyl growth. Nature 475: 398–404 613 Oh E, Zhu JY, Wang Z-Y (2012) Interaction between BZR1 and PIF4 integrates brassinosteroid 614 and environmental responses. Nat Cell Biol 14: 802–809 615 Park E, Kim Y, Choi G (2018) Phytochrome B requires PIF degradation and sequestration to 616 induce light responses across a wide range of light conditions. Plant Cell 30: 1277–1292 617 Peng M, Li Z, Zhou N, Ma M, Jiang Y, Dong A, Shen WH, Li L (2018) Linking phytochrome- 17 618 interacting factor to histone modification in plant shade avoidance. Plant Physiol 176: 1341– 619 1351 620 Pham VN, Kathare PK, Huq E (2018) Phytochromes and phytochrome interacting factors. Plant 621 Physiol 176: 1025–1038 622 Rausenberger J, Hussong A, Kircher S, Kirchenbauer D, Timmer J, Nagy F, Schäfer E, Fleck C 623 (2010) An integrative model for phytochrome B mediated photomorphogenesis: From 624 protein dynamics to physiology. PLoS One 5: e10721 625 Reed JW, Nagpal P, Poole DS, Furuya M, Chory J (1993) Mutations in the gene for the red/far- 626 red light receptor phytochrome B alter cell elongation and physiological responses 627 throughout arabidopsis development. Plant Cell 5: 147–157 628 Sentandreu M, Leivar P, Martín G, Monte E (2012) Branching of the PIF3 regulatory network in 629 Arabidopsis. Plant Signal Behav 7: 510–513 630 Sentandreu M, Martín G, González-Schain N, Leivar P, Soy J, Tepperman JM, Quail PH, Monte 631 E (2011) Functional profiling identifies genes involved in organ-specific branches of the 632 PIF3 regulatory network in Arabidopsis. Plant Cell 23: 3974–3991 633 Shen H, Moon J, Huq E (2005) PIF1 is regulated by light-mediated degradation through the 634 ubiquitin-26S proteasome pathway to optimize photomorphogenesis of seedlings in 635 Arabidopsis. Plant J 44: 1023–1035 636 Shen H, Zhu L, Castillon A, Majee M, Downie B, Huq E (2008) Light-induced phosphorylation 637 and degradation of the negative regulator phytochrome-interacting factor1 from Arabidopsis 638 depend upon its direct physical interactions with photoactivated phytochromes. Plant Cell 639 20: 1586–1602 640 Shen Y, Khanna R, Carle CM, Quail PH (2007) Phytochrome induces rapid PIF5 phosphorylation 641 and degradation in response to red-light activation. Plant Physiol 145: 1043–1051 642 Shin J, Park E, Choi G (2007) PIF3 regulates anthocyanin biosynthesis in an HY5-dependent 643 manner with both factors directly binding anthocyanin biosynthetic gene promoters in 644 Arabidopsis. Plant J 49: 981–994 645 Soy J, Leivar P, González-Schain N, Martín G, Diaz C, Sentandreu M, Al-Sady B, Quail PH, 646 Monte E (2016) Molecular convergence of clock and photosensory pathways through PIF3- 647 TOC1 interaction and co-occupancy of target promoters. Proc Natl Acad Sci U S A 113: 648 4870–4875 649 Soy J, Leivar P, González-Schain N, Sentandreu M, Prat S, Quail PH, Monte E (2012) 650 Phytochrome-imposed oscillations in PIF3 protein abundance regulate hypocotyl growth 651 under diurnal light/dark conditions in Arabidopsis. Plant J 71: 390–401 652 Soy J, Leivar P, Monte E (2014) PIF1 promotes phytochrome-regulated growth under 653 photoperiodic conditions in Arabidopsis together with PIF3, PIF4, and PIF5. J Exp Bot 65: 654 2925–2936 18 655 Sweere U, Eichenberg K, Lohrmann J, Mira-Rodado V, Bäurle I, Kudla J, Nagy F, Schäfer E, 656 Harter K (2001) Interaction of the response regulator ARR4 with phytochrome B in 657 modulating red light signaling. Science (80- ) 294: 1108–1111 658 Toledo-ortiz G, Huq E, Quail PH (2003) The Arabidopsis basic/helix-loop-helix transcription 659 factor family. Plant Cell 15: 1749–1770 660 de Wit M, Keuskamp DH, Bongers FJ, Hornitschek P, Gommers CMM, Reinen E, Martínez- 661 Cerón C, Fankhauser C, Pierik R (2016) Integration of Phytochrome and Cryptochrome 662 Signals Determines Plant Growth during Competition for Light. Curr Biol 26: 3320–3326 663 de Wit M, Ljung K, Fankhauser C (2015) Contrasting growth responses in lamina and petiole 664 during neighbor detection depend on differential auxin responsiveness rather than different 665 auxin levels. New Phytol 208: 198–209 666 Yamashino T, Nomoto Y, Lorrain S, Miyachi M, Ito S, Nakamichi N, Fankhauser C, Mizuno T 667 (2013) Verification at the protein level of the pif4-mediated external coincidence model for 668 the temperature-adaptive photoperiodic control of plant growth in Arabidopsis thaliana. 669 Plant Signal Behav 8: e23390 670 Zhang R, Yang C, Jiang Y, Li L (2019) A PIF7-CONSTANS-Centered Molecular Regulatory 671 Network Underlying Shade-Accelerated Flowering. Mol Plant 12: 1587–1597 672 Zhang Y, Pfeiffer A, Tepperman JM, Dalton-Roesler J, Leivar P, Gonzalez Grandio E, Quail PH 673 (2020) Central clock components modulate plant shade avoidance by directly repressing 674 transcriptional activation activity of PIF proteins. Proc Natl Acad Sci 117: 3261–3269 675 676 DATA AVAILABILITY STATEMENT 677 The data that support the findings of this study are available from the corresponding authors upon 678 reasonable request. 679 FIGURE LEGENDS 680 Fig. 1. PIF7 induces growth under diurnal short day (SD) conditions in antagonistic manner 681 with phyB. 682 (A) Schematic view of the treatments used. Seedlings were grown for 4 days under WL, long day 683 (LD, 16hWL:8hD) and short day (SD, 8hWL:16hD) photoperiods, or under LD and SD 684 conditions submitted to a phy-inactivating far red pulse at the end of the day (LD-EODFR and 685 SD-EODFR). 686 (B) Visible phenotypes of 4d-old seedlings grown in SD or in constant white light (WL). 687 (C) Quantification of hypocotyl elongation of 4d-old seedlings grown under the indicated 688 conditions. Data represent the mean and standard error of at least 25 seedlings. Letters denote 689 statistically significant differences among means by the Tukey’s test (p < 0.05). Independent tests 690 were assayed for each experimental condition. 19 691 (D) Data shown in panel C represented as a growth difference among the indicated treatments. 692 Fig. 2. pifqpif7 mutants are insensitive to the photoperiodic control of seasonal growth. 693 (A) Visible phenotypes of 4d-old seedlings grown in SD and SD-EODFR as detailed in Fig. 1A. 694 (B) Quantification of hypocotyl elongation of 4d-old seedlings grown under the indicated 695 conditions. Data represent the mean and standard error of at least 25 seedlings. Letters denote 696 statistically significant differences among means by the Tukey’s test (p < 0.05). Independent tests 697 were assayed for each experimental condition. 698 (C) Data shown in panel B represented as a growth difference among the indicated treatments. 699 (D) Visible phenotypes of 4d-old seedlings grown in WL, SD and SD-EODFR. 700 Fig. 3. PIF7 is able to heterodimerize with PIF3. 701 A) PIF7 and PIF3 interact in vitro. Coimmunoprecipitation assays using GAL4 Activating 702 Domain (GAD) alone or fused to full-length PIF7 (GAD:PIF7) and PIF3 (GAD:PIF3) as baits, 703 while PIF7 was used as prey. Schematic diagrams on the top show the design of the experiment, 704 and the SDS-PAGE separations of the pellet fractions and the inputs (10%) are shown below. 705 Quantification of the binding expressed as % of bound PIF7 in relation to the initial PIF7 input 706 and normalized to the bait used is shown at the bottom. 707 (B) PIF7 and PIF3 interact in a yeast two-hybrid assay. GAD, GAD:PIF3 and GAL4 DNA 708 Binding domain (GBD) alone or fused to full length PIF7 (GBD:PIF7) were used for yeast two- 709 hybrid interaction assays. Growth assays using restrictive media –LWA (top row) and –LWH 710 (second row from top) were done to test for interaction in the yeast strain AH109. The control 711 using non-restrictive –LW medium is shown in the third row. Interaction assay using qualitative 712 β-galactosidase assay using Y187 yeast strain on plate is shown in the bottom row. The lower 713 panel shows Miller units in a quantitative liquid β-galactosidase assay. 714 (C) PIF7 and PIF3 bind to the G-box DNA motif as heterodimer. Electrophoresis mobility shift 715 assay (EMSA) showing TnT-translated PIF7 and GAD:PIF3 binding to the G-box as homodimers 716 (lanes 4 and 3 respectively) and as a heterodimer (intermediate band in lane 5). TnT-only reactions 717 were included as controls (lanes 2 and 6). FP= free probe. Lane 1 corresponds to FP-only binding 718 reaction, and the asterisks indicate non-specific bands in the TnT-only reactions. 719 Fig. 4. Developmental and temporal regulation of hypocotyl growth by PIF3 and PIF7 under 720 diurnal conditions. 721 (A) Quantification of hypocotyl elongation of seedlings grown for 2, 3 and 4 days in SD or in 722 SD-EODFR. Letters denote statistically significant differences among means by the Tukey’s test 723 (p < 0.05) in SD (letters below lines) and SD-EODFR (letters above lines). Independent tests were 724 assayed for each day and condition. 725 (B) Quantification of hypocotyl elongation of seedlings that were first grown for 2 days in WL 726 (2d-WL) and then grown for 4 additional days in either WL, SD or SD-EODFR. 20 727 (C) Quantification of hypocotyl growth during the night hours of the 3rd day in SD or in SD- 728 EODFR. Seedlings were grown for 2 days in SD (ZT0), and images of the seedlings were taken 729 at the end of the day (ZT8, W8), and at the end of the night (ZT24) with (FR24) or without (D24) 730 an EODFR pulse as indicated in the left panel. Hypocotyl length was measured (middle panel) 731 and data were represented as a growth difference among the indicated time points (right panel). 732 Data in A, B and C represent the mean and standard error of at least 25 seedlings. Letters denote 733 statistically significant differences among means by the Tukey’s test (p < 0.05). Independent tests 734 were assayed for each experimental condition. 735 Fig. 5. PIF7, together with PIF3, acts to induce growth-related gene expression at night. 736 Gene expression was measured during the night hours of the 3rd day in SD (A) or in SD-EODFR 737 (B). Seedlings were grown for 2 days in SD (ZT0) as in Fig. 4C, and samples were harvested at 738 the at the end of the third day (ZT8, W8), and during the night hours at ZT9, ZT12, ZT16, ZT20 739 and ZT24 in seedlings treated with (FR9, FR12, FR16, FR20, FR24; panel B) or without (D9, 740 D12, D16, D20, D24; panel A) an EODFR. An additional sample was harvested upon 1h 741 illumination (ZT25, W25) during the 4th day of growth in SD (panel A). PIL1, HFR1, XTR7, 742 HAT2 and MIDA9 expression relative to PP2A was measured and data are normalized to WT_D24 743 set at one. Data represent the mean and standard deviation of technical triplicates of one biological 744 replicate. 745 Fig. 6. PhyB prevents full de-phosphorylation of PIF7 at night. 746 (A) PIF7 time-course expression in WT seedlings grown under SD or SD-EODFR. Seedlings 747 were grown in SD. During the third day of growth, samples were either harvested at the indicated 748 times (SD), or given an EODFR pulse and collected during the night at the indicated time points 749 (SD-EODFR samples). PIF7 expression relative to PP2A was measured and normalized to the 0 750 time point (ZT0) set at one. Values are means of technical triplicates of one biological replicate. 751 (B) PIF7 protein level in SD at ZT8 (W8), ZT9 (D9), ZT24 (D24) and at 1h of the following day 752 after illumination (W1). 753 (C) PIF7 protein level in SD-EODFR compared to SD at ZT8 (W8), ZT9 (FR9, D9) and at ZT24 754 (FR24, D24). 755 (B, C) Ponceau staining was used as a loading control. Blots correspond to one biological 756 replicate. 21 Fig. 1 A 0 8 16 24 WL LD LD-EODFR SD SD-EODFR B WL SD WT pif7 phyB phyBpif7 WT pif7 phyB phyBpif7 1mm 1mm C 8 c WT c 7 pif7 a Hypocotyl Length (mm) a 6 phyB d phyBpif7 5 b c 4 b 3 b d a c c a b b 2 aa aa 1 0 WL LD LD-EODFR SD SD-EODFR D 4.0 3.5 WT Growth Difference (mm) pif7 3.0 phyB 2.5 phyBpif7 2.0 1.5 1.0 0.5 0.0 [LD] [LD-EODFR] [SD] [SD-EODFR] - [WL] - [LD] - [WL] - [SD] Fig. 2 pif3 pif3 pif3 pif4 A pif3 D pif3 pif4 pif4 pif5 pifq WT pif7 pif3 pif7 pif7 WT pif4 pif7 pif5 pif7 pifq pif7 1mm WL 1mm SD SD 1mm 1mm SD-EODFR SD-EODFR 1mm B 6.5 a 6.0 WT pif7 5.5 pif3 pif3pif7 bc pif3pif4 pif3pif4pif7 5.0 pif3pif4pif5 pif3pif4pif5pif7 Hypocotyl Length (mm) 4.5 pifq pifqpif7 db de 4.0 3.5 3.0 a 2.5 a d f f bceb d eb b 2.0 aab ded ccddd g a abbc c d 1.5 bc c c cc c dee e d dddd c d cccc c eee 1.0 0.5 0.0 WL LD LD-EODFR SD SD-EODFR C 4.0 3.5 WT pif7 Growth Difference (mm) pif3 pif3pif7 3.0 pif3pif4 pif3pif4pif7 2.5 pif3pif4pif5 pif3pif4pif5pif7 pifq pifqpif7 2.0 1.5 1.0 0.5 0.0 [LD] - [WL] [LD-EODFR] - [LD] [SD] - [WL] [SD-EODFR] - [SD] Fig. 4 A B 7 5.0 abcd a WT 4.5 Col a 6 pif7 SD-EODFR abbc pif7 Hypocotyl Length (mm) Hypocotyl Length (mm) pif3 4.0 5 pif3pif7 pif3 3.5 b pif3pif7 4 3.0 c abcd a 2.5 b 3 2.0 c a a a b b d SD 2 1.5 1.0 1 aabc aabb abcc 0.5 0 0.0 2 3 4 WL SD SD-EODFR Days in SD or SD-EODFR 2d-WL + 4d 4.0 2.5 C 3.5 WT a pif7 Hypocotyl length (mm) Growth Difference (mm) b b 2.0 3.0 pif3 ZT a a 0 8 24 pif3pif7 2.5 c 1.5 W8 D24 2.0 a a b b b c 1.0 1.5 +FRp FR24 1.0 0.5 0.5 0.0 0.0 W8 D24 FR24 D24 - W8 FR24 -W8 A SD B SD-EODFR Fig. 5 1.2 1.0 2.5 2.0 PIL1/PP2A WT 0.8 pif7 1.5 0.6 pif3 pif3pif7 1.0 0.4 0.2 0.5 0.0 0.0 6 8 10 12 14 16 18 20 22 24 26 6 8 10 12 14 16 18 20 22 24 26 1.2 1.0 1.5 HFR1/PP2A 0.8 WT pif7 1.0 0.6 pif3 0.4 pif3pif7 0.5 0.2 0.0 0.0 6 8 10 12 14 16 18 20 22 24 26 6 8 10 12 14 16 18 20 22 24 26 1.2 3.5 1.0 3.0 XTR7/PP2A WT 2.5 0.8 pif7 2.0 0.6 pif3 1.5 0.4 pif3pif7 1.0 0.2 0.5 0.0 0.0 6 8 10 12 14 16 18 20 22 24 26 6 8 10 12 14 16 18 20 22 24 26 1.2 3 WT 1 2.5 HAT2/PP2A pif7 0.8 pif3 2 0.6 pif3pif7 1.5 0.4 1 0.2 0.5 0 0 6 8 10 12 14 16 18 20 22 24 26 6 8 10 12 14 16 18 20 22 24 26 1.2 3.5 3 MIDA9/PP2A 1 WT 2.5 0.8 pif7 2 0.6 pif3 pif3pif7 1.5 0.4 1 0.2 0.5 0 0 6 8 10 12 14 16 18 20 22 24 26 6 8 10 12 14 16 18 20 22 24 26 ZT (SD) +FRp ZT (SD-EODFR) Fig. 6 A 20 SD Relative Expression SD-EODFR 15 PIF7/PP2A 10 5 0 0 4 8 12 16 20 24 28 h in SD or SD-EODFR B MW PIF7-Flash WT (kDa) W8 D9 D24 W1 D24 W1 150 100 PIF7-P 75 PIF7 C MW PIF7-Flash WT (kDa) W8 D9 FR9 D24 FR24 FR24 D24 150 100 PIF7-P 75 PIF7