Photoactivated UVR8-COP1 Module Determines Photomorphogenic UV-B Signaling Output in Arabidopsis

In Arabidopsis, ultraviolet (UV)-B-induced photomorphogenesis is initiated by a unique photoreceptor UV RESISTANCE LOCUS 8 (UVR8) which utilizes its tryptophan residues as internal chromophore to sense UV-B. As a result of UV-B light perception, the UVR8 homodimer shaped by its arginine residues undergoes a conformational switch of monomerization. Then UVR8 associates with the CONSTITUTIVELY PHOTOMORPHOGENIC 1-SUPPRESSOR OF PHYA (COP1-SPA) core complex(es) that is released from the CULLIN 4-DAMAGED DNA BINDING PROTEIN 1 (CUL4-DDB1) E3 apparatus. This association, in turn, causes COP1 to convert from a repressor to a promoter of photomorphogenesis. It is not fully understood, however, regarding the biological significance of light-absorbing and dimer-stabilizing residues for UVR8 activity in photomorphogenic UV-B signaling. Here, we take advantage of transgenic UVR8 variants to demonstrate that two light-absorbing tryptophans, W233 and W285, and two dimer-stabilizing arginines, R286 and R338, play pivotal roles in UV-B-induced photomorphogenesis. Mutation of each residue results in alterations in UV-B light perception, UVR8 monomerization and UVR8-COP1 association in response to photomorphogenic UV-B. We also identify and functionally characterize two constitutively active UVR8 variants, UVR8W285A and UVR8R338A, whose photobiological activities are enhanced by the repression of CUL4, a negative regulator in this pathway. Based on our molecular and biochemical evidence, we propose that the UVR8-COP1 affinity in plants critically determines the photomorphogenic UV-B signal transduction coupling with UVR8-mediated UV-B light perception.


Introduction
Light is a critical environmental stimulus that regulates a number of developmental and physiological processes of living organisms. In sessile plants, perception of light is the initial and decisive step in light signaling transduction, and is achieved by several groups of photosensory receptor proteins. Phytochromes sense far-red and red light [1,2]. Cryptochromes and phototropins perceive blue and ultraviolet (UV)-A light [3,4,5,6]. In Arabidopsis thaliana, UV RESISTANCE LOCUS 8 (UVR8) has recently been identified as a photoreceptor that detects UV-B (280 to 320 nm) light [7]. Long-wavelength and low-fluence UV-B induces plant photomorphogenic development that is physically characterized by the inhibition of hypocotyl elongation, flavonoid accumulation, and UV-B stress tolerance [8,9,10,11].
UVR8 was originally isolated as a UV-resistance gene, having been shown to contribute to the UV-B-induced flavonoid accumulation and UV-B protection [12]. Transcriptomic analyses have revealed that UVR8 positively orchestrates UV-B signaling specifically under photomorphogenic UV-B [13]. Later, a series of functional studies have disclosed that UVR8 exhibits a number of features characteristic of photoreceptors, including a broad-range loss of UV-B responsive gene expression in the uvr8 null mutant [13,14], the enrichment of aromatic residues in UVR8 protein and UV-B-induced conformational change of UVR8 [7].
Despite these insights, however, the exact process by which UVR8 mediates UV-B light perception remained unclear until two research groups independently reported the structure of UVR8 [15,16]. Without the presence of UV-B light, UVR8 appears as a symmetric seven-bladed-b-propeller homodimer that is stabilized by arginines primarily Arg 286 and Arg 338. These arginine residues shape intramolecular cation-p interactions with their surrounding tryptophan residues, among which Trp 285 and Trp 233 act as the internal UV-B chromophore. Consequently, unlike phytochromes and cryptochromes, UVR8 is devoid of external small molecules as chromophore. Upon UV-B irradiation, the dark-state dimer of UVR8 is monomerized as a result of the disruption of the intramolecular cation-p interactions and the intermolecular hydrogen bonds mediated by Arg 286 and Arg 338 [7,15,16]. This structural conversion, which takes place in seconds, is a major determinant for UVR8 to sequester CONSTITUTIVELY PHOTOMORPHOGENIC 1-SUPPRESSOR OF PHYA (COP1-SPA) core complex(es) from the CULLIN 4-DAMAGED DNA BINDING PROTEIN 1 (CUL4-DDB1) E3 apparatus. Ultimately, this complex reorganization enables COP1 to act as a positive regulator in the UV-Binduced photomorphogenesis by facilitating the stability and activity of a photomorphogenesis-promoting transcription factor ELONGATED HYPOCOTYL 5 (HY5) [7,17]. Reversibly, upon the elimination of UV-B irradiation, REPRESSOR OF UV-B PHOTOMORPHOGENESIS 1 (RUP1) and RUP2, two UVR8interacting proteins, might disrupt the physical contact of UVR8 and COP1, so that UVR8 dimerization can be regenerated [7,15,16,18,19,20]. However, the exact biological significance of key residues in UVR8 has not been fully determined to date.
Here we take advantage of site-directed mutagenesis to generate UVR8 variant proteins in Arabidopsis, and demonstrate the pivotal roles of two light-absorbing tryptophans, W233 and W285, and two dimer-stabilizing arginines, R286 and R338 in UVR8initiated UV-B-induced photomorphogenesis. We also characterize two constitutively active forms of UVR8, UVR8 W285A and UVR8 R338A , whose photobiological activity is enhanced when the repressor CUL4 is suppressed. Overall, our molecular and biochemical evidence has supported that the intrinsic affinity of UVR8-COP1 critically determines the efficiency of photomorphogenic UV-B signal transduction coupling with UVR8-mediated UV-B light perception.

UVR8 variants differentially interact with COP1 in yeast
In order to assess the roles of key residues in UVR8, we generated six UVR8 variants in two groups based on their functional classification via site-directed mutagenesis. The first group, which included UVR8 W233A , UVR8 W233F , UVR8 W285A and UVR8 W285F , and the second group, which included UVR8 R286A and UVR8 R338A , were designed to interrupt UVR8's perception of UV-B light and dimer stabilization respectively ( Figure S1A). As yeast has been widely used as an efficient system to determine the conformational status of UVR8 and its interaction with other proteins in response to UV-B [7,21,22], we first introduced wild-type and these six mutated UVR8 (mUVR8) into yeast. Using SDS-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblot analyses, we found that UVR8 WT was dimeric under 2UV-B and monomeric under +UV-B ( Figure 1A), while UVR8 W233F , UVR8 W285A and UVR8 W285F were constitutively monomeric, monomeric and dimeric respectively ( Figure 1A). This is consistent with the results reported previously [7]. The other UVR8 variants, UVR8 W233A , UVR8 R286A and UVR8 W338A appeared as monomers irrespective of UV-B treatment ( Figure 1A). These results demonstrate that in yeast, in addition to the dimer-stabilizing arginines, R286 and R338, the light-absorbing tryptophans, W233 and W285, critically contribute to the dimer-to-monomer switch of UVR8 upon UV-B irradiation.
We next performed a series of yeast two-hybrid assays in order to examine the effects of the UVR8 mutations on the interaction between UVR8 and COP1. While COP1 only interacted with UVR8 WT under +UV-B, it interacted with UVR8 W233A , UVR8 W233F , UVR8 W285A , UVR8 R286A and UVR8 R338A under both 2UV-B and +UV-B. The interaction between UVR8 W285F and COP1 was barely observed ( Figure 1B). These results suggest that the UVR8-COP1 interaction in yeast requires UVR8 to be in its monomeric form. Furthermore, previous studies have also proposed that RUP1 and RUP2 interact with UVR8 independent of UV-B irradiation [19]. Specifically, it has been suggested that these two proteins mediate UVR8 redimerization and disrupt UVR8-COP1 interaction, so as to facilitate the inactivation of the photoreceptor [18]. In yeast, RUP1 was found to constitutively interact with UVR8 W233A , UVR8 W233F , UVR8 W285A and UVR8 R286A , but was not observed to interact with UVR8 WT , UVR8 W285F and UVR8 R338A . Differently, RUP2 was observed to interact with the wild-type UVR8 and all the mutant UVR8 proteins ( Figure S1B). These results collectively suggest that light perception and conformational change of UVR8 are both major determinants of its interaction with key players of UV-B-induced photomorphogenesis.

UVR8 variants alter physiological responses to photomorphogenic UV-B in Arabidopsis
To investigate the biological activity of these UVR8 variants in Arabidopsis, we introduced wild-type and mutated UVR8 fused with yellow fluorescent protein (YFP) driven by the native UVR8 promoter into the uvr8 null mutant uvr8-6 background. As expected, proUVR8-YFP-UVR8/uvr8-6 (YFP-UVR8 WT ) was found to express the transgenic UVR8 protein at a level comparable to the endogenous UVR8 protein in Col under 2UV-B and +UV-B ( Figure S2A). It was able to rescue uvr8-6 in UV-B-induced hypocotyl growth and anthocyanin accumulation (Figures 2A, 2B and 2C) which are characteristic physiological responses to photomorphogenic UV-B [14,23]. We next examined these two responses in the six transgenic UVR8 variants expressing YFP-UVR8 proteins at equivalent levels to that in YFP-UVR8 WT ( Figure 2D). The hypocotyl shortening in all the variants failed to reach the shortening degree detected in YFP-UVR8 WT . Interestingly, YFP-UVR8 W285A and YFP-UVR8 R338A displayed shorter hypocotyl than YFP-UVR8 WT under 2UV-B, while YFP-UVR8 R338A instead of YFP-UVR8 W285A showed further shortened hypocotyl under +UV-B (Figures 2A and 2B). Furthermore, compared with YFP-UVR8 WT , reduced anthocyanin accumulation was found in YFP-UVR8 W233A , YFP-UVR8 W233F , YFP-UVR8 W285F and YFP-UVR8 R286A under both 2UV-B and +UV-B. In contrast, enhanced anthocyanin accumulation was detected in YFP-UVR8 W285A and YFP-UVR8 R338A under

Author Summary
Higher plants are able to sense and interpret diverse light signals to modulate their growth. In response to longwavelength and low-intensity ultraviolet-B (UV-B) light, plants establish photomorphogenic development and stress acclimation. UV RESISTANCE LOCUS 8 (UVR8) is a unique UV-B photoreceptor that triggers photomorphogenesis in Arabidopsis thaliana. However, the signaling process following UV-B light perception by plants is not fully understood. In this study, by generating transgenic UVR8 variants in Arabidopsis, we have extensively analyzed the biological significance of key residues in UVR8 for UV-B-induced photomorphogenesis. Furthermore, by engineering and characterizing two constitutively active UVR8 variants, we have provided the biochemical insight that the in vivo association between UVR8 and CONSTITUTIVE-LY PHOTOMORPHOGENIC 1 (COP1) critically determines the photomorphogenic UV-B signaling output.
2UV-B. YFP-UVR8 R338A was even able to accumulate anthocyanin under +UV-B, though at a level lower than that in YFP-UVR8 WT ( Figure 2C).
These observations showed that all the mutations interfered with UV-B-induced photomorphogenesis, though to varying degrees. The mutations in UV-B-absorbing residues resulted in more severe phenotypic defects than those in UVR8-dimerizing residues, suggesting the importance of the sequential action of UV-B light perception before UVR8 monomerization. Meanwhile, a specific activity of UVR8 W285A and UVR8 R338A is indicated by the constitutive short hypocotyl and high anthocyanin content found in YFP-UVR8 W285A and YFP-UVR8 R338A (Figures 2B and  2C).
In addition to these marker genes, key regulators of UV-B specific signaling, including the positive regulators COP1 and ELONGATED HYPOCOTYL 5 (HY5) are also transcriptionally governed by photomorphogenic UV-B [24,25]. We found that the accumulation of COP1 mRNA was diminished in all UVR8 variants with the exception of YFP-UVR8 R338A ( Figure 3D). Similarly, though the accumulation of HY5 mRNA was observed in YFP-UVR8 W233F and YFP-UVR8 R338A , it was limited in YFP-UVR8 W233A , YFP-UVR8 W285A , YFP-UVR8 W285F and YFP-UVR8 R286A ( Figure 3E). At the post-transcriptional level, COP1 protein was clearly induced in YFP-UVR8 WT , YFP-UVR8 W233F and YFP-UVR8 R338A , slightly increased in YFP-UVR8 W233A under +UV-B, and relatively high under both 2UV-B and +UV-B in YFP-UVR8 W285A . However, the accumulation of COP1 protein mediated by photomorphogenic UV-B was scarcely detected in YFP-UVR8 W285F and YFP-UVR8 R286A (Figures 3F and 3G). A similar pattern was observed for the case of HY5 protein accumulation ( Figures 3F and 3G). Taken together, the altered expression profiles of these UV-B responsive genes suggest that these residues responsible for UV-B light perception and UVR8 dimerization are crucial for UVR8 activity in the transcriptional control of UV-B signaling at a molecular level.
Downstream of UVR8-COP1, RUP1 and RUP2 contribute to the negative feedback regulation of UV-B-induced photomorphogenesis. Given that RUP1 and RUP2 are known to be induced by photomorphogenic UV-B dependent on UVR8, COP1 and HY5 [19], it was noteworthy that the accumulation of RUP1 and RUP2 mRNA was apparently reduced in all UVR8 variants except UVR8 R338A (Figures 4A and 4B). As early UV-B responsive genes, RUP1 and RUP2 were activated in a temporal manner in response to photomorphogenic UV-B to balance UV-B specific signaling [19]. Using 4day-old seedlings grown under 2UV-B and then transferred to +UV-B for various periods of time, we found that the temporal induction of RUP1 and RUP2 by photomorphogenic UV-B was apparently observed in YFP-UVR8 WT , but was retarded in all the UVR8 variant lines ( Figures 4C and 4D). It is worth pointing out that the transcript levels of RUP1 and RUP2 were elevated within 1 hours of UV-B irradiation to a peak and then fell back in YFP-UVR8 WT , whereas they continued to rise to a lower peak particularly in YFP-UVR8 W285A , YFP-UVR8 R286A and YFP-UVR8 R338A over the 12-hour UV-B treatment ( Figures 4C and 4D). These results suggest that RUP1 and RUP2 fail to be activated in our UVR8 variants, and thus do not establish the repressive transcriptional modules required for balanced UV-B signaling. Overall, none of the UVR8 variants are functionally equivalent to YFP-UVR8 WT , which was consistent with our phenotypic observations.

UVR8 variants impair UV-B light perception, UVR8 monomerization and UVR8-COP1 association
In order to better understand the biochemical activity of these residues in vivo, we investigated the efficiency of UV-B light perception, UVR8 monomerization and the formation of UVR8containing complex in all of our UVR8 variant plants. By measuring UV-B absorbance at 310 nm, the central wavelength of our photomorphogenic UV-B condition, we found that Col and YFP-UVR8 WT exhibited a strong ability to sense UV-B. In contrast, mutations in either residue of the internal chromophore, W233 or W285, led to a complete loss of UV-B absorption. YFP-UVR8 R286A and YFP-UVR8 R338A retained the ability to sense UV-B, but at significantly diminished levels, which suggests that the disruption of the homodimeric interface has a negative impact on the full activity of UVR8 to perceive UV-B light ( Figure 5A).
We next examined the endogenous association of UVR8 variants and COP1 by using in vivo co-immunoprecipitation (co-IP) assays. In agreement with previous studies [7,14], YFP-UVR8 WT co-immunoprecipitated a high level of COP1 specifically under +UV-B. However, independent of UV-B, YFP-UVR8 W285F and YFP-UVR8 R286A scarcely co-immunoprecipitated COP1, while YFP-UVR8 W233F and YFP-UVR8 W233A co-immunoprecipitated very low levels of COP1. YFP-UVR8 W285A and YFP-UVR8 R338A co-immunoprecipitated medium levels of COP1 under 2UV-B, while the latter was also observed to co-immunoprecipitate more COP1 under +UV-B ( Figure 5C). These results reveal that the monomeric conformation is not sufficient for UVR8 to associate with COP1 in vivo. The relatively close association of YFP-UVR8 W285A and YFP-UVR8 R338A with COP1 without UV-B treatment indicates that a specific activity of UVR8 might be produced by W285A or R338A mutation.

Constitutive activity of UVR8 is dependent on constant UVR8-COP1 interaction
As both cop1-4 and uvr8-6 suffer from diminished gene expression, hypocotyl growth, anthocyanin accumulation and acclimation in response to photomorphogenic UV-B, COP1 and UVR8 share a high degree of functional similarity in photomorphogenic UV-B signaling [14,25]. We found that YFP-UVR8 WT / cop1-4 phenocopied cop1-4 ( Figure 6A), clearly demonstrating that COP1 acts genetically downstream of UVR8. The observation that YFP-UVR8 WT failed to rescue cop1-4 ( Figure 6A) also suggests that the function of UVR8 is dependent on COP1. Though CUL4 works in concert with COP1 in darkness, it functionally disassociates from COP1 and plays a negative role in UV-Binduced photomorphogenesis [17]. Since cul4cs exhibited no obvious defect in hypocotyl growth under UV-B [17], we consistently found that YFP-UVR8 WT /cul4cs phenocopied YFP-UVR8 WT and cul4cs ( Figure 6A).
We have previously proposed that the UVR8-COP1 interaction mediated by UV-B enables a reorganization of COP1 complexes and eventually results in a functional switch of COP1 from a repressor to a promoter of photomorphogenesis [17]. Both the enhanced photomorphogenesis of YFP-UVR8 W285A and YFP-UVR8 R338A under 2UV-B (Figures 2A, 2B and 2C) and the persistent binding of UVR8 W285A and UVR8 R338A (Figures 1B  and 5C) to COP1 prompted us to examine the development of all the UVR8 variants in darkness. In addition to Col, uvr8-6 and YFP-UVR8 WT , YFP-UVR8 W233A , YFP-UVR8 W233F , YFP-UVR8 W285F and YFP-UVR8 R286A all showed typical skotomorphogenic responses, demonstrating long hypocotyls, closed cotyledons and apical hooks. In contrast, YFP-UVR8 W285A and YFP-UVR8 R338A displayed open cotyledons, suggesting that these two UVR8 variants were capable of inducing constitutive photomorphogenesis irrespective of their exposure to light ( Figure 6B). We then examined light-regulated gene expression, and found highly accumulated transcripts of CHS, RIBULOSE BISPHOSPHATE CARBOXYLASE SMALL CHAIN 1A (RBCS1A), CHLOROPHYLL A/B BINDING PROTEIN 3 (CAB3) and ELIP2, in YFP-UVR8 W285A and YFP-UVR8 R338A ( Figure 6C). Though UVR8 W233A , UVR8 W233F and UVR8 R286A had constitutive physical interactions with COP1 in yeast ( Figure 1B), their affinity with COP1 in vivo was much weaker than that of UVR8 W285A and UVR8 R338A (Figure 5C), indicating that conversion of COP1's function might require a threshold level of UVR8-COP1 interaction.

Discussion
Light-absorbing tryptophans and dimer-stabilizing arginines are intrinsically coordinated to fulfill UVR8 function The molecular framework of UV-B-induced photomorphogenesis has been gradually established over the past ten years. For example, both the identification of UVR8 as a UV-B photoreceptor and the subsequent structural analysis of recombinant UVR8 have recently led to a proposed mechanism of UVR8dependent UV-B signaling initiation [7,15,16]. A series of UVR8 variant proteins have been generated to further demonstrate that UVR8 exploits its own light-absorbing tryptophans and dimerstabilizing arginines to perceive light and initiate protein conformational changes, respectively [15,16]. UVR8 W233F , UVR8 W285A and UVR8 W285F are lack of the ability to perceive UV-B in vitro [16]. UVR8 W285A is at least partially monomeric and interacts with COP1. UVR8 W285F , on the other hand, is dimeric and unable to interact with COP1 in yeast, plant and mammalian cells [7,15,16,22,26]. UVR8 R286A and UVR8 R338A are monomeric and display an obviously diminished perception of UV-B light in vitro [16]. In Arabidopsis, mutations in UVR8's tryptophan residues have been pointed out to result in the physical impairment of photomorphogenic UV-B responses [22]. However, the exact mechanism driving this hindered response remains unknown. Moreover, it is far less understood concerning the biological significance of light-absorbing and dimer-stabilizing residues in UVR8, as well as the signaling process that connects UV-B light perception, UVR8 monomerization and subsequent signaling events including the organization of UVR8-COP1-SPA complex(es).
In our study, we generated transgenic plants expressing UVR8 variant proteins under UVR8's native promoter, based on the sitedirected mutagenesis of two light-absorbing tryptophans and two dimer-stabilizing arginines. These variant proteins drove varied levels of UV-B-induced photomorphogenesis. All the variants were observed to impair both UVR8's activity to perceive UV-B light and its ability to undergo a dimer-to-monomer transition ( Figures 5A and 5B), suggesting the reciprocal impacts by these residues involved in the same intramolecular interaction network. More severe phenotypic defects were observed in UV-B-absorbing variants than those in UVR8-dimerizing variants (Figures 2A, 2B and 2C), confirming that UV-B light perception precedes UVR8 monomerization to launch UV-B signaling. In addition, the unchangeable conformational status of the UVR8 variants ( Figure 5B) abolished the dimer-monomer-dimer cycling of UVR8, and further disturbed the balance in UV-B signaling. Molecularly, all the mutations caused an altered hierarchy of UV-B responsive gene expression (Figures 3 and 4). They failed to accurately establish the promotive module formed by the UVR8-COP1-HY5 core pathway and the negative transcriptional feedback mediated by RUP1 and RUP2, leading to inadequate and unbalanced photomorphogenic UV-B responses. Taken together, the roles of light-absorbing tryptophans and dimerstabilizing arginines in UVR8 are intrinsically coordinated for UVR8 activity in UV-B-induced photomorphogenesis.
A threshold UVR8-COP1 interaction plays a critical role in coupling with UV-B light perception to signal transduction The direct interaction between UVR8 and COP1 takes place rapidly following UV-B light perception and UVR8 monomerization [7]. It requires UVR8 to be in its monomeric form [7]  whereas physically monomeric UVR8 is not sufficient for the formation of the UV-B-dependent COP1 complex and to mediate photomorphogenesis in response to UV-B in plants (Figures 2A  and 5C). Though this point has been previously articulated [22], the molecular mechanism underlying this empirical observation is still unknown. This discrepancy serves to indicate that factors downstream of UV-B light perception and UVR8 monomerization might essentially govern the progression of UV-B signaling.
In terms of the transcriptome reprogramming induced by photomorphogenic UV-B, cop1-4 resulted in the loss of transcriptional responses of a broader range of genes than uvr8-6 [14]. While UVR8's ability to mediate UV-B-induced photomorphogenesis was observed to be dependent on COP1 (Figure 6A), COP1 did not appear to significantly influence UVR8 conformation [7,20]. These data collectively indicate that COP1 is at least as essential as UVR8 in photomorphogenic UV-B signaling, if not superior to UVR8, due to its role in the formation of UVR8-COP1-SPA complex(es). By using the native promoter of UVR8 to drive UVR8 variants in plants, we selected those transgenic lines expressing comparable levels of UVR8 variant proteins ( Figure 2D), in order to stringently analyze each variant in parallel, particularly to examine the in vivo association intensity between UVR8 and COP1 ( Figure 5C). As a result, we propose that a threshold level of the in vivo association between UVR8 and COP1 is critically required for photomorphogenic UV-B signaling output, founded on the following evidence. Firstly, UVR8 monomers, rather than UVR8-COP1 interaction, were detected in 2UV-B-treated plant and yeast cells expressing wild-type UVR8 (Figures 1, 5B and 5C). Secondly, the in vivo levels of UVR8-COP1 association correspond with the physiological and  (Figures 2A and 5C). Only those UVR8 variants that possess high affinity with COP1 in plants, namely YFP-UVR8 WT under +UV-B, and YFP-UVR8 W285A and YFP-UVR8 R338A under 2UV-B and +UV-B in our study, result in photobiological activity in specific light contexts ( Figures 5C, 2A and 6B). Thirdly, once the affinity between UVR8 and COP1 is conditionally increased, such as the situation in cul4cs that might release more COP1 for UVR8 to interact with, the photobiological activity of UVR8 is ultimately enhanced ( Figure 6D). In agreement with our hypothesis, a most recent report has presented that comparing with the overexpressed wild-type UVR8 (UVR8-OX), UVR8 W285A leads to improved photomorphogenesis and UV-B tolerance by increased COP1 binding affinity, though UVR8 W285A and UVR8-OX express equivalent UVR8 proteins [27].
Though our UVR8 variants do not demonstrate equivalent patterns of interaction with COP1 in yeast and plants ( Figures 1B  and 5C), this discrepancy can most likely be explained by the fact that the yeast two-hybrid system is devoid of any other factors that might influence the UVR8-COP1 interaction. The variation in the UVR8 variants' affinity for COP1 observed in plants may be due to a wide variety of factors. For example, UVR8 monomer variants might undergo protein folding processes in plants that are distinct from those in yeast, given that each mutated residue locates specifically in the intramolecular interaction network ( Figure S1A). Similarly, the protein levels of endogenous COP1 vary amongst the different UVR8 monomer variants ( Figure 3G). Finally, a number of the other proteins present in vivo might interact with COP1 and/or UVR8 in manner that either enhances or hinders the in vivo contact of UVR8 and COP1. Further research is required, however, in order to fully disentangle and elucidate the impact of the great complexity in vivo.
Constitutive light signaling is mediated by UVR8 W285A and UVR8 R338A Two constitutively active forms of UVR8, UVR8 W285A and UVR8 R338A have been uncovered in our study. Though they differed in the phenotypic features of UV-B-induced photomorphogenesis, they both displayed constitutive interaction with COP1 ( Figure 5C) and photomorphogenic development in darkness ( Figure 6B). It is worth pointing out that a previous report did not find constitutive photomorphogenesis in GFP-UVR8 W285A [22] whereas a recent independent study reached a consensus with ours by showing the constitutive photomorphogenesis in UVR8 W285A [27].
Structure of dimeric UVR8 has revealed that W285, the UV-B chromophore, is located at the center of the strong cation-p and pp interaction network at the dimer interface, while R338, responsible for dimer stabilization, is located at the edge of the interaction network [16]. It is suggested that W285 is more essential to the function of UVR8 than R338. Hence it is reasonable that YFP-UVR8 W285A failed to respond to photomorphogenic UV-B while YFP-UVR8 R338A was still able to, in terms of UV-B light perception, UV-B-induced UVR8-COP1 association and gene expression, and eventually photomorphogenic development (Figures 2A-C, 3A-C, 5A and 5C). On the other hand, in the absence of UV-B, YFP-UVR8 W285A and YFP-UVR8 R338A associated with a sufficient amount of COP1, and thus promoted hypocotyl growth, anthocyanin accumulation and gene activation under 2UV-B (Figures 2A-C, 3A-C and 5C), and even photomorphogenesis in darkness ( Figures 6B and 6C).
As reported, other gain-of-function alleles of photoreceptors were also produced via point mutations in their photosensory regions, such as the GAF domain tyrosine mutants of phytochromes (PHY), PHYA Y242H and PHYB Y276H [28], and the photolyase-related (PHR) domain glycine mutants of cryptochromes (CRY), CRY1 G380R , CRY2 G377R [29]. PHYB Y276H was proposed to interact with COP1 in a light-independent manner to diminish the degradation of HY5 by COP1 [28]. CRY1 G380R was found to co-localize in nucleus with COP1 to promote the nuclear exclusion of COP1 in darkness [29]. In contrast to its role in the repression of photomorphogenesis induced by far-red and visible light [30], COP1 is known to promote UV-B-induced photomorphogenesis [25]. Upon UV-B irradiation, COP1 rapidly interacts with UVR8 in nucleus [7], and switches from degrading to stabilizing HY5 [17,31]. The constitutive photomorphogenesis in YFP-UVR8 W285A and YFP-UVR8 R338A is not resulted from a loss in COP1 protein abundance that was observed in cop1-4 ( Figure S3). Therefore, the way UVR8 modulates COP1 dramatically differs from the activity repression of COP1 by cryptochromes or phytochromes. Furthermore, UVR8 is involved in diverse developmental processes in plants. Photomorphogenic UV-B signaling displays crosstalk with circadian regulation [32], and it also controls leaf morphogenesis [33,34], drought tolerance [35] and plant immune response [36]. Beyond the area of plant researches, UVR8 has also been implemented in optical control of protein interactions [26,37] and multi-chromatic expression regulation [38] in mammalian cells. Thus, the characterization of constitutively active UVR8 variants should serve to elucidate the mechanism of UV-B specific signaling in plants and advance protein engineering pertinent to a variety of medical applications.
The Arabidopsis materials were grown as described previously [24]. The seeds were surface-sterilized and sown on solid Murashige and Skoog medium supplemented with 1% sucrose for biochemical assays or with 0.3% sucrose for phenotypic analysis, and cold treated at 4uC for 4 days.  Added with 46 loading buffer containing 250 mM Tris-HCl (pH 6.8), 2% SDS, 20% b-mercaptoethanol, 40% glycerol, and 0.5% bromophenol blue, the samples were subjected to immunoblot analysis without boiling.
The assay in plants was performed as previously described [17]. Total proteins were extracted from 4-day-old Arabidopsis seedlings grown under 2UV-B or +UV-B in protein extraction buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 10% glycerol, 0.1% Tween 20, 1 mM phenylmethylsulfonyl fluoride (PMSF), and 16 complete protease inhibitor cocktail (Roche). Then the cell extracts were kept on ice under exactly the same condition (2UV-B or +UV-B) as where the seedlings were grown for 30 min. Added with 46 loading buffer containing 250 mM Tris-HCl (pH 6.8), 2% SDS, 20% b-mercaptoethanol, 40% glycerol, and 0.5% bromophenol blue, the samples were subjected to immunoblot analysis without boiling.
Yeast two-hybrid assay

Hypocotyl and anthocyanin measurement
Hypocotyl length was measured as previously described [24]. For each line grown under 2UV-B or +UV-B for 4 days, hypocotyl length was analyzed in three biological replicates. In each replicate, at least 30 Arabidopsis seedlings were measured. The relative hypocotyl length was presented as the percentage of the hypocotyl length under +UV-B with respect to that under 2UV-B (% of 2UV-B). For each line grown in darkness, hypocotyl length was analyzed using at least 30 Arabidopsis seedlings. The quantification of hypocotyl length was performed by ImageJ (http://rsb.info.nih.gov/ij/).
Anthocyanin was extracted and quantified as previously described [42]. Briefly, Arabidopsis seedlings were harvested and placed into extraction solution (18% 1-propanol and 1% HCl), and boiled for 3 minutes. Then the mixture was left in darkness for at least 3 hours at room temperature. After a brief centrifugation to pellet the tissue debris, the supernatant was removed and diluted with the extraction solution. The anthocyanin content was presented as A 535 22(A 650 ) g 21 fresh weight.

Quantitative real-time PCR
Total RNA was extracted from 4-day-old Arabidopsis seedlings grown under 2UV-B or +UV-B using the RNeasy plant mini kit (Qiagen). Reverse transcription was performed using SuperScript II first-strand cDNA synthesis system (Invitrogen) according to the manufacturer's instructions. Real-time qPCR analysis was performed using SYBR Premix Ex Taq (Takara) with Applied Biosystems 7500 Real-Time PCR System. Each experiment was repeated with three independent samples, and RT-PCR reactions were performed in three technical replicates for each sample. The primers are listed in Table S1.

Measurement of UV-B absorbance
Total proteins was extracted from 4-day-old Arabidopsis seedlings grown under 2UV-B and +UV-B in protein extrac-tion buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM PMSF, and 16 complete protease inhibitor cocktail (Roche). Absorbance at 310 nm of plant total proteins adjusted to equal concentration and total amount were measured. Each experiment was repeated with three independent samples.