Time series transcriptomics reveals a BBX32-directed control of dynamic acclimation to high light in mature Arabidopsis leaves

The photosynthetic capacity of mature leaves increases after several days’ exposure to constant or intermittent episodes of high light (HL) and is manifested primarily as changes in chloroplast physiology. This is termed dynamic acclimation but how it is initiated and controlled is unknown. From fully expanded Arabidopsis leaves, we determined HL-dependent changes in transcript abundance of 3844 genes in a 0-6h time-series transcriptomics experiment. It was hypothesised that among such genes were those that contribute to the initiation of dynamic acclimation. By focussing on HL differentially expressed transcription (co-)factor (TF) genes and applying dynamic statistical modelling to the temporal transcriptomics data, a gene regulatory network (GRN) of 47 predominantly photoreceptor-regulated (co)-TF genes was inferred. The most connected gene in this network was B-BOX DOMAIN CONTAINING PROTEIN32 (BBX32). Plants over-expressing BBX32 were strongly impaired in dynamic acclimation and displayed perturbed expression of genes involved in its initiation. These observations led to demonstrating that as well as regulation of dynamic acclimation by BBX32, CRYPTOCHROME1, LONG HYPOCOTYL5, CONSTITUTIVELY PHOTOMORPHOGENIC1 and SUPPRESSOR OF PHYA-105 are also important regulators of this process. Additionally, the BBX32-centric GRN provides a view of the transcriptional control of dynamic acclimation distinct from other photoreceptor-regulated processes, such as seedling photomorphogenesis.

Full transcriptome profiles using CATMA microarrays (Sclep et al., 2007) were 158 obtained from leaf 7 of HL-exposed plants along with parallel LL controls (see 159 Methods). Microarray analysis of variance (MAANOVA;Wu et al., 2003;see 160 Methods) was used to extract expression values from each probe for every treatment 161 for each technical and biological replicate. To determine DEGs that showed a 162 significant difference between HL-exposed leaves and the LL controls over all or part 163 of the time period, a Gaussian process two-sample test (GP2S; Stegle et al., 2010) 164 was applied and 4069 probes were selected with a Bayes factor score >10 which 165 corresponded to 3844 DEGs (Supplemental Data Set 1). The full data set is 166 deposited with Gene Expression Omnibus (GEO; GSE78251). 167 To obtain further insight into the overall response to HL at the molecular level, 168 hierarchical co-cluster analysis of the 3844 DEGs was carried out using 169 SplineCluster (Heard et al., 2005). We reasoned that groups of DEGs that display 170 similar temporal patterns of expression could be co-regulated and clustering would 171 be useful in identifying groups of genes for dynamic modelling. On the basis of 172 comparing temporal gene expression patterns in both the HL-exposed and control LL 173 leaves (see Methods), the 3844 DEGs were divided into 43 temporal clusters (Fig. 174 1A; Supplemental Data Set 1). The outcome of this co-clustering was differential The clusters are ordered such that 1-13 show general transcript abundance to be 180 lower in HL vs LL samples and/or displayed a downward pattern over the diel 181 0.002) after 5 daily HL exposures (Fig. 2D). This confirmed that repeated HL 256 exposures did not solely affect stomatal behaviour but brought about an increase in 257 foliar photosynthetic capacity. The changes in CF parameters by day 5 of HL 258 treatments observed in the previous experiments ( Fig. 2A) occurred also in larger 259 older leaves that were required for gas exchange measurements (Supplemental 260 Figure 2C; see Methods). 261 In summary, increased A sat and A max after 5 days of repeated HL exposure (Fig. 2C, 262 D) was accompanied by a highly significant increase in Fq'/Fm' (Fig. 2A, 263 Supplemental Figure 2C; P < 0.0001, ANOVA and Tukey HSD; Supplemental Data 264 Set 3) reflecting an increased photochemical efficiency to support dynamic 265 acclimation. Therefore, a substantial (>40%; typically using the median 800 µmol m -2 266 s -1 actinic PPFD value) change in Fq'/Fm' between days 1 and 5 of repeated HL was 267 subsequently used as a more convenient image-based measurement of the 268 establishment of dynamic acclimation and consequent increased photosynthetic 269 capacity. 270

Dynamic statistical modelling infers a BBX32-centric network of HL-regulated 271 transcription factor genes. 272
The HL time series data were used to infer gene regulatory networks (GRNs) using 273 VBSSM (Beal et al., 2005;Penfold and Wild 2011). We chose VBSSM because it 274 has been demonstrated to infer known GRNs from temporal gene expression data 275 and to infer novel GRNs whose highly connected genes (nodes) have subsequently 276 been shown experimentally to have a novel and important function (Beal et al., 2005;277 Penfold and Wild, 2011;Breeze et al., 2011;Penfold and Buchanan-Wollaston, 278 2014;Windram and Denby 2015;Bechtold et al., 2016). However, due to the limited 279 number of time points, we opted to infer networks for around 100 genes or probes in 280 order to avoid overfitting by constraining the network size (Beal et al., 2005;281 Allahverdeyeva et al., 2015;Windram and Denby, 2015;Bechtold et al., 2016). To 282 accommodate this limitation, we focussed on DEGs coding for transcription (co-) 283 factors (TFs). We reasoned that TF GRNs would control the expression of a wide 284 network of genes and by inferring GRNs this would allow us to identify and focus on 285 the most connected TF genes, often termed hub genes (Windram and Denby, 2015;286 Albihlal et al., 2018). Consequently, we reasoned that upstream hub TF genes 287 would directly and indirectly regulate the expression of a sufficiently large number of 288 genes to influence whole leaf HL responses and acclimation phenotypes. Therefore, 289 the intention was to screen highly connected candidate regulatory hub TF genes 290 directly for their impact upon whole plant dynamic acclimation. 291 It was estimated that there were 371 HL DEGs coding for TFs or transcription co-292 factors (Supplemental Data Set 4). To narrow our selection further, comparisons 293 were made between the 43 HL temporal clusters ( Fig. 1A; Supplemental Data Set 1) 294 and 14 publicly available transcriptomics data sets or meta-analyses of such data for 295 treatments or mutants perturbed in chloroplast-to-nucleus and ROS-mediated 296 signalling (Supplemental Data Set 5). On a cluster-by-cluster basis, the highest 297 number of significant (P < 0.00001) overlaps in clusters 1, 2, 3, 5, 6, 9, 10, 14, 16, 17 298 and 27 were encountered with phyA/phyB DEGs (Supplemental Data Set 5) among 299 which genes involved in chloroplast-to-nucleus (retrograde) signalling had been 300 identified (Shikata et al., 2014). This observation suggested that photoreceptor-301 mediated regulation of HL-responsive genes was highly represented in the time 302 series transcriptomics dataset. Therefore, we examined whether photoreceptor-303 regulated TF and co-transcription factor genes (Shikata et al., 2014;Dong et al., 304 2014) were also over-represented in the HL dataset. This was the case. Ninety-one 305 (91) photoreceptor-and light-regulated TF DEGs were over-represented in the time 306 series transcriptomics data, irrespective of which temporal cluster they were drawn 307 from (P= 1.4E-06; Hypergeometric Distribution Test, Supplemental Data Set 4). The 308 HL time series expression data from these 91 genes were used to infer networks 309 with VBSSM. 310 The first inferred network for HL revealed LATE ELONGATED HYPOCOTYL (LHY) 311 12 This widespread disruption of PhAG transcript levels led us to examine the impact of 353 BBX32 over-expression on other cellular processes. In the RNAseq experiment, of 354 the 2903 genes whose transcript levels were HL responsive (Padj. < 0.05; > 2-fold 355 differentially expressed; Supplemental Data Set 7), BBX32 over-expression 356 perturbed the transcript levels of 32% and 15% of them in LL and HL conditions in the HL time series data (Supplemental Data Sets 1 and 2) may reflect a re-377 distribution of resources towards dynamic acclimation and away from basal immunity 378 (see above and Discussion). The observations here suggest BBX32 may play a 379 regulatory role in these processes (see Discussion) but also reinforces that BBX32 380 influences immediate responses before or during a single exposure to HL. 381

CRY1-and HY5-regulated control of dynamic acclimation 382
BBX32 has been proposed to be a negative regulator of the integration of light 383 signals from phytochromes (PHYs) and cryptochromes (CRYs) during 384 photomorphogenesis (Holtan et al., 2011;Gangappa and Botto, 2014). BBX32-OE 385 seedlings display a long hypocotyl phenotype in the light like photoreceptor mutants 386 and mutations in LONG HYPOCOTYL5 (HY5; Holtan et al., 2011). Notably, HY5 is a 387 member of the BBX32-centric GRN ( Fig. 3; Fig. 5) and along with CRY1, has also 388 been implicated in influencing the expression of HL-inducible gene expression 389 (Kleine et al., 2007;Shaikali et al., 2012;Chen et al., 2013  proteins that had been postulated to be implicated in HL-mediated G protein 400 signaling (Galvez-Valdivieso et al., 2009;Gorecka et al., 2014). However, the one 401 mutant recovered from this screening, was shown subsequently to be deficient in 402 dynamic acclimation due to a disabling second site mutation in CRY1 (see Methods). 403 Since the defective acclimation phenotype was identified prior to knowing the identity 404 of the causal mutation we took this to be forward genetic evidence of the importance 405 of CRY1 in dynamic acclimation in mature leaves. 406 The light environment used to grow plants for this study and subject to HL was 407 enriched for blue wavelengths (Supplemental Figure 7; see Discussion). Therefore, 408 we considered the possibility that a role for phytochromes in dynamic acclimation 409 could be obscured, favouring a predominance of CRY1 under our growth conditions. 410 To test this notion, a mutant harbouring a constitutively active form of PHYB, 411 phyBY276H (YHB) in a Col-0 background (Jones et al., 2015) was tested for 412 dynamic acclimation (Fig. 8C). This mutant exhibited a higher PSII operating 413 efficiency than Col-0 after 1 day of HL exposure. This accelerated acclimation 414 phenotype is in keeping with being a constitutively active positive regulator of 415 dynamic acclimation. 416 Mutants defective in HY5 function were strongly impaired in dynamic acclimation 417 ( Fig. 8D, E) consistent with being a member both of a BBX32-centric GRN and being 418 a positive regulator of CRY1-mediated dynamic acclimation (Fig 8A, B). have been shown also to physically interact independent of COP/DET/FUS 429 (Pedmale et al., 2016;Ma et al., 2016). In the VBSSM-inferred GRN, PIF4 and 430 SPA1 were predicted to have a regulatory connection to BBX32 ( Fig. 3; Fig.5). 431 Significantly, 187 HL time series DEGs overlapped (P = 0.0018; Hypergeometric 432 Distribution Test) with a set of 1120 genes identified as commonly regulated by 433 SPA1, 2, 3, 4, PIF1, 3, 4, 5 and COP1 in de-etiolating and light-exposed seedlings 434 (Pham et al., 2018). Interestingly, the most significant GO Biological Process 435 function coded by these overlapping genes was Photosynthesis (GO: 0015979; FDR 436 = 2.7 E-17; Supplemental Data Set 11). 437 Cop1-4 plants, despite a severely dwarfed shoot morphology (Fig. 9A;Deng and 438 Quail, 1992;Gangappa and Kumar, 2018), displayed a highly elevated PSII 439 operating efficiency (Fq'/Fm') by day 1 of the HL acclimation regime compared with 440 Col-0 ( Fig. 9B) like the HL response of YHB plants (Fig. 8C). In contrast, despite a 441 similar dwarf shoot morphology (Fig. 9A), det1-1 displayed no defect in dynamic 442 acclimation (Fig. 9D). This strongly suggests that the dynamic acclimation response 443 of chloroplasts is independent of shoot size and that these two traits are not coupled. 444 Furthermore, spa1/spa2/spa3 (spa1,2,3) plants also displayed an accelerated 445 acclimation phenotype (Fig. 9C). Therefore, it was concluded that one or more type 446 of COP1/SPA complex (Huang et al., 2014;Hoecker, 2017) are negative regulators 447 of dynamic acclimation and that DET1 plays no role in dynamic acclimation. 448 There is a high degree of redundancy among the PIF TF family and therefore a 449 quadruple null mutant of PIF1,3, 4 and 5 (hereafter called pifq; Leivar et al., 2008) 450 was tested for its capacity for dynamic acclimation. These plants displayed a severe 451 dwarf phenotype as previously described (Leivar et al., 2008), but also a significant 452 inhibition of dynamic acclimation (Fig. 9E). In contrast, the dynamic acclimation of a 453 single mutant allele of PIF4 (pif4-2) was normal (Supplemental Fig. 6D). 454

The time series HL transcriptomics data indicates the initiation of dynamic 456 acclimation processes 457
The exposure to a 7.5-fold increase in PPFD (HL; see Methods) presents both a 458 threat and an opportunity to the plants in this study. The threat comes from the 459 possibility that the PPFD will continue to increase and render the plant susceptible to 460 irreversible photoinhibition. The opportunity comes from enhancing photosynthesis 461 by initiating dynamic acclimation ( Fig. 2A-D). Accompanying enhanced 462 photosynthesis was also a lowering of reliance on the dissipation of excitation energy 463 using NPQ (Supplemental Fig. 2A), which can limit plant productivity (Kromdiijk et 464 al., 2016). 465 The adaptation to a potential increase in photooxidative stress and photoinhibition 466 (see Introduction) is the early (≤ 1h) strong but transient change in transcript 467 abundance of 257 genes in clusters [21][22][23][24][25][26]upon exposure to HL. Clusters 22,23,25 468 and 26 include among them 64 known genes that promote abiotic stress tolerance 469 Data Set 1; Supplemental Data Set 2). The transiently enhanced expression of these 471 genes presumably allows the plant to overcome any potential initial detrimental 472 effects of the HL exposure, as many other studies have reported (e.g. Ball et al., 473 2004;Gadjev et al., 2006;Ramel et al., 2012;2013;Willems et al., 2016;Crisp et al., 474 2017;Huang et al., 2019). 475 Coordinated alteration in specific biological processes was evident in some clusters. 476 Down-regulated clusters include those collectively associated with aspects of basal 477 or innate resistance to pathogens (Underwood, 2012;Piasecka et al., 2015). 478 Examples include genes coding for cell wall modifications and callose deposition 479 (cluster 1), defense response to bacteria (cluster 3) and glucosinolate / glycosinolate 480 biosynthesis (cluster 10). In this study, plants were grown at a PPFD below their light 481 saturated rate of photosynthesis (Asat; Fig. 2C; see Methods). Plants grown under 482 such light-limiting conditions may initially have to re-allocate resources away from 483 some cellular processes in order to begin acclimation and take advantage of a 484 sustained increase in PPFD. Photosynthetically active expanded but not senescing 485 leaves, such as leaf 7 used here (see Methods), have been suggested to maintain a 486 higher degree of poising of immunity to respond to biotic stress compared with 487 abiotic stress (Berens et al., 2019). Therefore, in a converse situation where a 488 potential abiotic stress threat emerges, the diversion of resources from defense 489 against pathogens may be an appropriate response. Meanwhile, among the DEG 490 time series clusters whose transcript levels increased at various points in the 491 experiment, are those that could be preparing the leaf to increase its photosynthetic 492 and metabolic capacity in order to begin acclimation (Eberhard et al., 2008;Dietz, 493 2015). Genes in over-represented GO BP terms included those involved in 494 macromolecule synthesis and especially translation (clusters 41-43) and related 495 metabolic processes such as enhanced amino acid and organic acid biosynthesis 496 (cluster 39). 497 A single exposure to 4h HL is not sufficient to induce dynamic acclimation at the 498 physiological level, requiring, under our conditions, a further 3 daily episodes of 4h 499 HL for this to begin to occur ( Fig. 2A-D). Our experience is consistent with a previous 500 study where dynamic acclimation took around 5 days to be fully manifested and 2-3 501 days to discern any change in photosynthesis rates after a permanent shift from a 502 photoperiod PPFD of 100 µmol m -2 s -1 to 400 µmol m -2 s -1 (Athanasiou et al., 2010). 503 However, it should be noted that the HL regime used did not produce dynamic 504 acclimation for the Col-0 accession but did for others such as Ws-2 and Ler-0 505 (Athanasiou et al., 2010). In contrast, the shorter more intense PPFDs used in this 506 study induced dynamic acclimation in Col-0 and also Ws-0 ( Fig

BBX32 connects a range of cellular processes to dynamic acclimation 509
Of all the comparisons carried out with relevant transcriptomics data sets, the most 510 extensive overlap with time series HL DEGs was with those from dark-germinated 511 phyA/phyB seedlings exposed to red light (Supplemental Data Set 5;Shikata et al., 512 2014). While this was initially surprising because of the very different experimental 513 conditions, earlier studies had shown a strong influence of photoreceptor genes 514 (CRYs and PHYs) on photosynthetic capacity in Arabidopsis grown at a range of 515 PPFDs (Walters et al., 1999) and an impact on the induction of some HL-responsive 516 genes (Kleine et al., 2007;Shaikali et al., 2012;Guo et al., 2017). Our own and the 517 published data above, prompted a selection of 91 light-and PHYA/B-regulated 518 (co)TF genes (Supplemental Data Set 5). The HL time series expression data from 519 these genes was subjected to VBSSM, which after two iterations, inferred a highly 520 interconnected BBX32-centric (co)TF GRN ( Fig. 3 and see Results). In the GRN, 521 >50% of the nodes (genes) were subsequently confirmed by RNAseq to be 522 influenced significantly in their expression by BBX32 ( Fig.3; Fig. 5).  BBX32 overexpression also impacts on a range of cellular processes that can be 531 associated with basal immunity, including multiple GO designations for 532 glucosinolate/glycosinolate metabolism, callose deposition, responses to chitin and 533 to pathogens (Supplemental Data Sets 7, 9 and 10). This observation is consistent 534 with the enrichment of the same processes noted in down-regulated temporal 535 clusters (see above; Supplemental Data Set 2) and supports our suggestion that in 536 wild type plants, down-regulation of basal immunity may be a necessary prerequisite 537 for successful dynamic acclimation (see above). We propose that BBX32 control of 538 aspects of basal immunity is part of its regulation of the initiation of dynamic 539 acclimation. 540 BBX32 showed a greater transcript abundance over LL controls at any point 541 onwards from the 2h HL timepoint. Nevertheless, its transcript abundance was on a 542 downward trend through the diel, paralleling its LL pattern of expression 543 (Supplemental Fig. 5). Interestingly, while BBX32-OE plants displayed a 66-fold 544 elevated BBX32 transcript level in LL, this value reduced to 33-fold after 3.5h HL 545 (Supplemental Data Set 7). The enhanced BBX32 expression in these plants is 546 driven by the CaMV 35S promoter (Holtan et al., 2011), therefore the decline in 547 transcript abundance over a diel could indicate that a temporal post-transcriptional 548 control operates to determine BBX32 transcript levels. 549 The strong negative impact of BBX32 over-expression upon dynamic acclimation 550 suggested that a defective gene ought to confer a converse elevated phenotype. The 551 mutant bbx32-1 (see Results;Holtan et al., 2011), displayed a weakly significant 552 trend of enhanced PSII operating efficiency compared with Col-0 between days 2 553 and 4 of the 5 days of 4h HL exposure (Fig. 4B). This genotype, however, is unlikely 554 to be a null mutant. The mutagenic T-DNA is inserted such that the first 172 amino 555 acid residues of BBX32 would still be produced and a truncated transcript spanning 556 this region has been detected in bbx32-1 seedlings ( functional truncated BBX32 may explain the weak phenotype of bbx32-1 with respect 562 to this acclimation phenotype (Fig. 4B) and also its mild constitutive 563 photomorphogenic phenotype in seedlings (Holtan et al., 2011). 564

Two levels of control of dynamic acclimation 565
The time series data and the VBSSM modelling led us to identify BBX32 and HY5 as 566   Laubinger et al., 2004;Lian et al., 2011, Lau andDeng 2012;Huang et al., 2014;591 Gangappa and Botto, 2016;Hoecker, 2017;Pham et al., 2018;Lau et al., 2019). 592 Consequently, CRY1 would cause the re-direction of HY5 to the activation of 593 dynamic acclimation. However, a further adaptation may be required to slow or 594 accelerate dynamic acclimation. For example, to fine tune the establishment of 595 dynamic acclimation in a fluctuating light environment. We suggest under HL, when 596 HY5 is free of negative regulation by COP1/SPA, that BBX32 is the important 597 additional moderator of the establishment of dynamic acclimation. We speculate in 598 the scheme in Figure 10 how this system may work and provides a basis for further 599 studies. The transcriptional control of HY5 and by extension, other members of the 600 BBX32-centric GRN ( Fig. 3; Fig. 10), could be subject to regulation by additional 601 intracellular signals in HL, such as those from chloroplasts and hormones, serving to 602 coordinate a range of cellular processes for dynamic acclimation (Hardkte et al, 603 2000;Galvez-Valdivieso et al 2009;Estavillo et al., 2011;Ramel et al., 2012;2013;604 Dietz, 2015;Gangappa and Botto, 2016;Guo et al., 2016;Exposito-Rodriguez et al., 605 2017). 606 The opposing regulation of dynamic acclimation by BBX32 and HY5 could mean that 607 some form of genetic interaction between these genes drives its establishment in a 608 manner similar to their negative and positive regulation respectively of 609 photomorphogenesis (Datta et al., 2007;Holtan et al., 2011;Xu et al., 2014;610 Gangappa and Botto, 2016). However, BBX32 does not bind DNA and has been 611 proposed to act as transcription co-factor in complexes with several TFs (Park et al., 612 2011;Holtan et al., 2011;Ganagappa and Botto, 2016;Tripathi et al., 2017). Of 613 relevance here, in a tripartite complex with BBX21, BBX32 has been suggested to 614 diminish the binding of HY5 to its target promoters (Datta et al., 2007;Holtan et al., 615 2011;Xu et al., 2014;Gangappa and Botto, 2016). Therefore, alongside 616 transcriptional control of HY5 by BBX32, there may also be this post-translational 617 control of HY5 action by BBX32 during dynamic acclimation. 618 The proposed need for both a CRY1/COP1/SPA-and a BBX32-mediated control of 619 dynamic acclimation (Fig.10) comes also from considerations about light quality and 620 intensity. First, the fluence of blue light in the HL exposure used in this study would 621 exceed the saturation of CRY1 activation, which is ca. 32 -40 µmol m -2 s -1 blue light 622 (Hoang et al., 2008;Liu et al., 2020). Therefore, while CRY1 signaling would need to 623 be activated (i.e. on) for dynamic acclimation to happen, further signaling input may 624 be required from other sources via BBX32 and its GRN to modulate the degree of 625 response. A second factor is that at high fluence, CRY1 may produce H 2 O 2 in the 626 nucleus (Consentino et al., 2015). H 2 O 2 for HL signaling is primarily synthesized and 627 exported from chloroplasts and is dependent upon an active photosynthetic electron 628 transport chain (Exposito-Rodriguez et al., 2017;Mullineaux et al., 2018). However, 629 this does not exclude the possibility that the HL-dependent accumulation of H 2 O 2 in 630 nuclei may be augmented from other sources such as photo-saturated CRY1, 631 signals from which could be fed into the BBX32-centric GRN. 632 In contrast to Arabidopsis grown at differing PPFDs but using similar fluorescent 633 lighting to this study (Walters et al., 1999;see Methods), no influence of PHYA or 634 PHYB was observed on dynamic acclimation (Supplemental Fig. 6A A further explanation could be that the PHY mutants were altered in leaf 645 development such that this impacted on their photosynthetic properties. Equally, we 646 cannot rule out effects of a similar nature on BBX32-OE, hy5, cry1, cop1 and spa1,2 647 3 plants, but the clear lack of influence of a more severe dwarf shoot morphology on 648 chloroplast level acclimation in det1-1 plants argues against this (Fig. 9A, D). 649

BBX32 over-expressed in Arabidopsis and soybeancontrol of a balance 650 between photosynthetic capacity and leaf longevity? 651
Arabidopsis (At)BBX32 has been over-expressed in transgenic soybean (Glycine The potential negative effect of AtBBX32 over-expression in soybean depressing 673 maximal photosynthetic capacity may not have proved detrimental because of the 674 way the crop is grown commercially. Modern soybean varieties are grown as a row 675 crop to achieve a high canopy coverage that maximises the absorption of light 676 (Shepherd et al., 2018;Koester et al., 2014). Within the canopy, the photosynthetic rate of leaves below Asat may not have been significantly affected by AtBBX32 over-678 expression in soya, as observed in Arabidopsis BBX32-OE plants (Fig. 4D). 679 Furthermore, we speculate that the photosynthetic capability of leaves exposed to 680 full sun in BBX32OE-soya plants, while perhaps being unable to achieve a maximal 681 Asat, benefitted from an enhanced leaf longevity and chloroplast integrity. A further 682 effect could also have been that like their Arabidopsis counterparts, the BBX32OE-683 soya plants had higher NPQ (Supplemental Data Set 6) perhaps linked to enhanced 684 PsbS transcript levels ( Fig. 6 In summary, that a network of TF genes could control dynamic acclimation, 691 encompassing a wide range of cellular processes, implies a complex and extensive 692 regulation that would provide resilience and flexibility in being able to accommodate 693 input from further intracellular and extracellular signaling. At the whole plant level, 694 this would allow for the degree of photosynthetic capacity and acclimation in 695 individual leaves to be adjusted according to their specific micro-environments 696 making this acclimation a truly dynamic process. 697

Growth conditions 699
Plants were grown in an 8 h photoperiod (short day) at a PPFD of 150 (± 10) µmol

Arabidopsis genotypes 706
The following Arabidopsis mutants and transgenic lines, all in a Col-0 background, 707 have been described previously: BBX32-10, BBX32-12, bbx32-1 (Holtan et al., and Discussion), we realized that M32 resembled the phenotype of known cry1 735 mutants. Therefore, we tested if CRY1 was altered in this mutant. CRY1 was 736 amplified from its genomic DNA and the PCR product was Sanger sequenced on 737 both strands. Col-0 CRY1 amplicon was also sequenced. The analysis of the 738 sequence showed that in M32, CRY1 contains a single point mutation (GA), which 739 caused a substitution of Gly 347 Arg mutation in CRY1 (Supplemental Fig. 8F). This 740 mutation was previously identified in a screening of EMS-mutagenized Arabidopsis 741 seedlings (Ahmad et al., 1995) and designated as hy4-15, and affects the domain 742 comprising the photolyase signature sequence. As a consequence, hy4-15 plants 743 produce a wild type amount of full length CRY1, but the protein is not functional. 744 Therefore, we concluded that M32 mutant is in fact a cry1 mutant that we named 745 cry1M32. 746

HL exposures 747
The HL exposure was a PPFD of 1100 (± 100) µmol m -2 s -1 from a white light 748 emitting diode (LED) array (Isolight 4000; Technologica Ltd, Colchester UK) as 749 To elicit dynamic acclimation, plants were subjected to 4h HL, followed by a 775 0.5h dark adaptation and then exposed to a range of actinic PPFDs (over 50 min) to 776 collect CF data (see below). This HL treatment was repeated daily and CF data 777 rosette data were collected. The raw data were fed via Excel into a program in R to 795 calculate, plot and statistically analyse the CF parameters (Gorecka et al., 2014). 796 The fluorimager software produces average data of all leaf pixel values. CF 797 parameters were represented as mean ± SE from a minimum of 4 plants, and 798 statistical significance was estimated with ANOVA followed by a post-hoc TukeyHSD 799 test. 800

Measurement of photosynthesis 801
A was measured on leaf 7 of plants at 49 dpg using an infrared gas exchange 802 system (CIRAS-1, PP Systems, Amesbury, MA, USA). The response of A to 803 changes in the intercellular CO 2 concentration (C i ) was measured under a saturating 804 PPFD, provided by a combination of red and white LEDs (PP Systems, Amesbury, 805 MA, USA). In addition, the response of A to changes in PPFD from saturating to sub-806 saturating levels was measured using the same light source at the current 807 atmospheric CO 2 concentration (390 µmol mol -1 ). All gas analysis was made at a leaf 808 temperature of 20 (±1) °C and a VPD of 1 (±0.2) KPa. Plants were sampled between 809 1 and 4 hours after the beginning of the photoperiod. For each leaf, steady state 810 rates of A at current atmospheric [CO 2 ] were recorded at the beginning of each 811 measurement. 812

RNA extraction, labelling and hybridisation to microarrays 821
For the time series HL experiment, RNA was extracted from leaf 7 samples, labelled 822 Inspection of selected probes from the rank order of likelihood of differential 847 expression was used to identify significant changes in expression with a Bayes' 848 factor cut-off >10 giving 4069 probes corresponding to 3844 DEGs (Supplemental 849 Data Set 1). 850

Clustering of Gene Expression Profiles 851
The expression patterns of the identified DEGs in HL and LL were co-clustered with 852 SplineCluster (Heard et al., 2005), using the mean expression profiles of the 853 biological replicates generated from MAANOVA and a prior precision value of 0.001 854 as described previously (Windram et al., 2012;Bechtold et al., 2016). 855

GO analysis 856
GO annotation analysis was performed using DAVID (Huang et al., 2008) or 857 AGRIGO (Du et al., 2010) with the GO Biological Process (BP) category (Ashburner 858 et al., 2000). Overrepresented GO_BP categories were identified using a 859 hypergeometric test with an FDR threshold of 0.05 compared against the whole 860 annotated genome as the reference set. 861

Comparisons with published transcriptomics data 862
The 3844 HL DEGs were compared on a cluster-by-cluster basis with publicly 863 available transcriptomics data. The references for each dataset are in the main 864 reference section of the paper. Each DEG list from published data was mapped to 865 AGI codes when necessary, cleaned to obtain single AGI codes since in some 866 microarray data, probes mapped to several genes or were listed as "no_match" and 867 were eliminated from the list. Overlaps within each cluster and their statistical 868 significance were determined using a Hypergeometric Distribution Test (phyper 869 function in R (v3.2.1)) in a custom R script, available upon request. When required, 870 Venn diagrams of overlaps between Data Sets were plotted with Venny 871 (http://bioinfogp.cnb.csic.es/tools/venny/index.html) and the significance of the 872 overlaps calculated using the R phyper function.  Fq'/Fm' (PSII operating efficiency) at a 400 µmol m -2 s -1 actinic PPFD.

(B -E)
The plots show the PSII operating efficiencies (Fq'/Fm') determined from CF 1598 images of from 8 plants (24-28 dpg) over 2 experiments (means ± SE). The plants 1599 had been exposed to 4h HL each day for 5 consecutive days (see Methods and 1600 legend of Figure 2). Note that because of the size of the cop1-4, pifq and det1-1 1601 plants, data were collected from whole rosettes rather than from mature leaves. CF to these measurements, the plants were exposed to HL for 4h per day for a total of 5 1676 days and the CF images were collected at day 1 (black lines) and day 5 (red lines) of 1677 the daily HL treatments for lhy21 (dashed lines) and Col-0 (solid lines). The 1678 determination of the parameters was by CF imaging as described in the legend of 1679 The numbers in parentheses is the temporal cluster to which each gene was 1683 assigned ( Fig. 1; Supplemental Data Set 1). The asterisk denotes a significant 1684 difference at that time point between LL and HL samples (P < 0.05; ANOVA).  (A) Plants were exposed daily to 4h HL and Fq'/Fm' determined for mature leaves. After the HL, plants were dark adapted and imaged under increasing actinic PPFD from 200-to-1400 µmol m -2 s -1 in 200 µmol m -2 s -1 increments every 5 min. The data were collected as CF images and processed digitally to collect values from mature leaves. The plants were treated in this way daily for 5 days: day 1 (blue), day 2 (red), day 3 (olive green), day 4 (purple) and day 5 (light blue). The data (mean ± SE) correspond to 38 plants at 24 -28 dpg over 6 experiments and the asterisks show differences in CF parameters between days 1 and 5 were significant (P ≤ 0.001; ANOVA and TukeyHSD). The full statistical data comparing all days of HL exposure are provided in Supplemental Data Set 5. (C) Photosynthesis plotted as CO 2 assimilation rate (A) as a function of actinic PPFD in mature leaf 7 (mean ± SE; n = 8 plants for each treatment; 49 dpg). Measurements were taken the day after 1 (dashed lines) and 5 days (solid lines) of daily 4h HL exposures (blue lines) along with the LL control plants (red lines) not subjected to this treatment.
(D) Photosynthesis plotted as CO 2 assimilation rate (A) as function of leaf internal CO 2 concentration (Ci) in mature leaf 7 (mean ± SE; n = 8 plants for each treatment; 49 dpg). Measurements were taken the day after 5 days of daily 4h HL exposures (blue line) along with the LL control (red line). A was determined by Infra-Red Gas Analysis (see Methods). Asterisks indicate significant differences (P < 0.02; covariant T and two-tailed F tests) between LL and HL-exposed plants.  The network shown was generated from the time series expression data for HL DEGs. The DEGs code for transcription (co)factors that are also light-and/or PHYA/PHYB regulated in de-etiolating seedlings. The network was generated using VBSSM (threshold z-score = 2.33; see Methods) and initially visualised using Cytoscape (v3.3.2;Shannon et al., 2003) but re-drawn manually to improve clarity. The network shown is from the second iteration of the modelling, which omitted expression data for LHY (First iteration; Supplemental Fig. 4A). The genes depicted in rectangular nodes were responsive to BBX32 over-expression in HL and / or LL exposed leaves by showing significantly (P< 0.05; Tukey HSD) higher (+) or lower (+) transcript abundance than Col-0 (see Fig. 5). Locus codes for the network genes can be found in Methods. experiments (means ± SE) which had first been exposed to 4h HL each day for 5 consecutive days (see Methods and legend of Figure 2). CF parameter values were collected at a range of actinic PPFDs (as indicated) at the end of each daily HL exposure. (A) Fq'/Fm' values at day 1 (black lines) and day 5 (red lines) for mutant or OE plants (dashed line) and Col-0 (solid line) of the HL treatments for BBX32-10 and BBX32-12. Asterisks indicate difference between mutant genotype and Col-0 at day 5 (P < 0.01; ANOVA and TukeyHSD). (B) Daily Fq'/Fm' values at 800 µmol m -2 s -1 PPFD actinic light of bbx32-1 compared with Col-0 showing differences that were significant (P <0.01) only between days 2 and 4.
(C) Photosynthesis plotted as CO 2 assimilation rate (A) as a function of incident PPFD in mature leaf 7 of LL-grown BBX32-10 (green line) and BBX32-12 (red line) compared to Col-0 (blue line) plants. Data are the mean ± SE; n = 4 for each genotype at 49 dpg; Asterisk indicates significant differences (P < 0.02; covariant T and two-tailed F tests) between Col-0 and BBX32-10 and BBX32-12 at a given PPFD. Leaf A, as a function of PPFD, was determined by Infra-Red Gas Analysis (see Methods).    Figure 6. BBX32 over-expression in LL and HL-exposed leaves perturbs transcript level of photosynthesis-associated genes. Using RNAseq data, relative cDNA abundance of BBX32-OE compared with Col-0 of photosynthesis-associated genes was determined under LL and 3.5 h HL exposure. The transcripts encoding the above proteins all displayed at >1.45 -fold greater or lesser abundance in fully expanded leaves of BBX32 -OE plants. The values are calculated from mean FPKM values (n=4) and difference between Col-0 LL and Col-0 HL were significant (P adj. < 0.05). In blue are the designated classifications for photosynthesis associated genes (https://www.kegg.jp/dbgetbin/www_bget?pathway+ath00195): AP, Antenna Protein; CBC, Calvin-Benson cycle enzyme; PET, photosynthetic electron transport protein; PSI and PSII, Photosystem I and II component proteins respectively. Most proteins are nuclear encoded but those marked with the suffix "C" are plastid encoded. Locus codes for the genes can be found in Methods.   The above scheme, while was based on this study, incorporates features from schemes published on the photoreceptor-directed control of seedling photomorphogenesis (e.g. Holtan et al., 2011;Hoecker, 2017). Under LL growth conditions to which the plant is acclimated, all or a proportion of cellular CRY1 is not active and consequently COP1/SPA acts to negatively regulate GRN members including HY5 and BBX32. Upon exposure to HL, CRY1 is activated and blocks COP1/SPA, which in turn releases the GRN ultimately leading to the establishment of dynamic acclimation. However, a degree of negative regulation of dynamic acclimation is retained (dotted inverted T) under HL conditions to allow for flexibility in potential fluctuations in the light environment. The plants had been exposed to 4h HL each day for 5 consecutive days (see Methods and legend of Figure 2). Note that because of the size of the cop1-4, pifq and det1-1 plants, data were collected from whole rosettes rather than from mature leaves. CF parameter values were collected at a range of actinic PPFDs (as indicated on the x-axis) at the end of days 1 and 5 of HL. The Fq'/Fm' values at day 1 (black lines) and day 5 (red lines) for mutant plants (dashed line) and Col-0 (solid line) of the HL treatments for (B) cop1-4, (C) spa1,2,3, (D) det1-1 and (E) pifQ. Asterisks (panel E) indicate significant difference between mutant compared with Col-0 at day 5 (P < 0.01; ANOVA and TukeyHSD). Upward arrows (panels B, C) indicate significant difference between mutants and Col-0 at day 1 (P < 0.01; ANOVA and TukeyHSD). The above scheme, while was based on this study, incorporates features from schemes published on the photoreceptor-directed control of seedling photomorphogenesis (e.g. Holtan et al., 2011;Hoecker, 2017). Under LL growth conditions to which the plant is acclimated, all or a proportion of cellular CRY1 is not active and consequently COP1/SPA acts to negatively regulate GRN members including HY5 and BBX32. Upon exposure to HL, CRY1 is activated and blocks COP1/SPA, which in turn releases the GRN ultimately leading to the establishment of dynamic acclimation. However, a degree of negative regulation of dynamic acclimation is retained (dotted inverted T) under HL conditions to allow for flexibility in potential fluctuations in the light environment.