GIGANTEA recruits the UBP12 and UBP13 deubiquitylases to regulate accumulation of the ZTL photoreceptor complex

Abstract

ZEITLUPE (ZTL), a photoreceptor with E3 ubiquitin ligase activity, communicates end-of-day light conditions to the plant circadian clock. It still remains unclear how ZTL protein accumulates in the light but does not destabilize target proteins before dusk. Two deubiquitylating enzymes, UBIQUITIN-SPECIFIC PROTEASE 12 and 13 (UBP12 and UBP13), which regulate clock period and protein ubiquitylation in a manner opposite to ZTL, associate with the ZTL protein complex. Here we demonstrate that the ZTL interacting partner, GIGANTEA (GI), recruits UBP12 and UBP13 to the ZTL photoreceptor complex. We show that loss of UBP12 and UBP13 reduces ZTL and GI protein levels through a post-transcriptional mechanism. Furthermore, a ZTL target protein is unable to accumulate to normal levels in ubp mutants. This demonstrates that the ZTL photoreceptor complex contains both ubiquitin-conjugating and -deconjugating enzymes, and that these two opposing enzyme types are necessary for circadian clock pacing. This shows that deubiquitylating enzymes are a core element of circadian clocks, conserved from plants to animals.

Fig. 1

GI bridges the interactions between ZTL and UBP12 or UBP13. a Yeast two-hybrid showing interaction between GI and UBP12 or UBP13. The GAL4 DNA-binding domain (GAL4-BD) fused to UBP12 or UBP13 and either ZTL variants (ZTL and ZTL decoy), ZTL targets (TOC1, PRR5, and CHE) or GI fused to GAL4 activation domain (GAL4-AD) were grown on SD-LW medium for autotrophic selection and on SD-LWHA medium to test for interactions. b Bimolecular fluorescence complementation (BiFC) assays to examine the interactions of UBP12 or UBP13 and GI fused to the N- or C-terminus of Venus (YFP) were performed in Arabidopsis protoplasts. The blue arrows indicate the interacting complex forming nuclear foci. The white arrows show fluorescence signal in the cytoplasm. mCherry-VirD2NLS was co-expressed as a nuclear marker, and the scale bar indicates 10 µm. c The protein domains of UBP12 and UBP13 required to interact with GI were tested using yeast two-hybrid assays. The full-length (FL) or truncated UBP12 or UBP13 fragments as diagramed in the lower portion of the panel were fused to GAL4-BD to test for interaction with GAL4-AD-GI. d Scatter plot of proteins identified by IP-MS of ZTL decoys in the Col-0 and gi-2 genotypes. The significance of the interactions were evaluated by SAINTexpress (see Methods and Supplementary Data 1 for complete information) with a false discovery rate (FDR) cutoff < 0.01 and p-value ≤ 5.37E-4 to separate interacting proteins into four groups. Group I: significant interactions with ZTL decoy in the gi-2 but not Col-0. Group II: significant interactions with ZTL decoy in both Col-0 and gi-2. Group III: significant interactions with ZTL decoy in the Col-0 but not gi-2. Group IV: Non-significant interactions with ZTL decoy in both Col-0 and gi-2. The interacting proteins significantly enriched in the gi-2 mutant over Col-0 were labeled along the y-axis, and the proteins enriched in the Col-0 over the gi-2 mutant were labeled along the x-axis. e Co-immunoprecipitation assays showing that UBP12 or UBP13 interact with ZTL in a GI-dependent manner. FLAG-UBP12 or FLAG-UBP13 were co-infiltrated with HA-GI and Myc-ZTL in Nicotiana benthamiana leaves. Anti-FLAG antibody was used to immunoprecipitate FLAG-UBP12 or FLAG-UBP13. Western blotting with anti-FLAG, anti-HA, or anti-Myc was used to detect the presence of FLAG-UBP12, FLAG-UBP13, HA-GI, or Myc-ZTL in the immunoprecipitated samples and inputs. f The diagram depicts the interaction between GI and the MATH domain of UBP12 or UBP13, and between GI and the LOV domain of ZTL. The source data are provided as a Source Data file. Blot images were cropped from their original size, which can be found in Source Data file

Fig. 2

UBP12 and UBP13 regulate the circadian clock through the same pathway as GI and ZTL. a–d The ubp12 and ubp13 mutants have short period phenotypes. a, c The periods of circadian marker pCCA1:Luciferase (pCCA1::LUC) in the wild type (Col-0) (n = 20 for a and n = 19 for c), ubp12-1 (n = 16), ubp12-2w (n = 20), ubp13-1 (n = 15), ubp13-2 (n = 20), and ubp13-3 (n = 14) were measured with bioluminescent assays. Each symbol represents the period from one seedling, and the average period and standard deviation are labeled with gray bars. The significance of period changes between wild type and mutants were analyzed with a two-tailed Welch’s t-test (*** for p-value < 0.001; **** for p-value < 0.0001). Three biological replicates were performed with similar results, and one dataset is presented. b, d The average bioluminescence of the lines displayed in a and c were plotted against time after transfer from 12 h light/12 h dark entrainment conditions to constant light. e, f Circadian expression of CCA1 in Col-0, ubp12-1, ubp13-1, gi-2, gi-2/ubp12-1, and gi-2/ubp13-1 after transferring to constant light for 48 h from the entrainment conditions was measured using qRT-PCR. Subjective dark is colored with light gray. The data represent the average relative expression of CCA1 normalized to IPP2 from three biological replicates, and the error bars are the standard deviation. The same Col-0 and gi-2 data were plotted twice (in e and f) for clarity in the data presentation and for comparison with the other mutant lines. g, h The circadian expression of CCA1 in Col-0, ubp12-1, ubp13-1, ztl-4, ztl-4/ubp12-1, and ztl-4/ubp13-1 after transferring to constant light for 48 h from the entrainment conditions was measured using qRT-PCR. The data analyses and presentation are the same as e–f. The same Col-0 and ztl-4 data were plotted twice (in g and h). i The circadian period of pCCA1::LUC in the wild type (n = 76), ubp12-1 (n = 54), ubp12-1 mutant complemented with pUBP12::UBP12-YFP (n = 32) or deubiquitylating activity-dead pUBP12::UBP12CS-YFP (n = 20). Each symbol represents the period from one seedling, and the black bars indicate the average period and standard deviation. The wild type and ubp12-1 mutants are homogenous populations, and the complementation lines are individual T1 transgenic lines. The presented data are from three independent biological replicates. j Quantitation of the number of lines, from panel i, with periods greater than the average of the ubp12-1 mutant plus one standard deviation. The source data are provided as a Source Data file

Fig. 3

ZTL, GI, and TOC1 protein levels are regulated by UBP12 and UBP13. a, c, e The protein levels of HA-tagged GI driven by native promoter (pGI::GI-HA), ZTL and YFP-tagged TOC1 driven by the TOC1 promoter (TOC1 minigene or TMG) in the wild type (Col-0), ubp12-1, or ubp13-1 mutants under diurnal conditions (12 h light/12 h dark) were detected by immunoblotting. The samples from 0 h to 12 h after dawn were harvested in light, and the samples from 16 h and 20 h after dawn were harvested in the dark (indicated by gray shading). The relative protein levels were quantified by normalization to actin. The Col-0 or ztl-4 samples were used as negative controls for the antibodies. Plots represent the average protein levels from three biological replicates, and the error bars represent standard deviation. Compared to wild type, the levels of ZTL proteins in the ubp12-1 and ubp13-1 were below the linear range for quantification. In the ztl-4 sample, the anti-ZTL antibody recognizes a non-specific band close to the size of endogenous ZTL in the long-exposure blots. b, d, f The relative mRNA levels of GI-HA, ZTL, or TOC1-YFP from the same time course samples were measured by qRT-PCR. The source data are provided as a Source Data file. Blot images were cropped from their original size, which can be found in Source Data file

Fig. 4

The proposed model for UBP12/UBP13 regulation of ZTL. a In the light, GI interacts with ZTL and acts as a co-chaperone, recruiting HSP90 to facilitate folding and maturation of the ZTL protein. Additionally, GI physically bridges an interaction between ZTL and UBP12 or UBP13. UBP12 or UBP13 stabilize the GI-ZTL protein complex before dusk. In the dark, GI dissociates from ZTL, and ZTL mediates ubiquitylation and degradation of the TOC1 protein. b Loss of UBP12 or UBP13 causes instability of ZTL and GI. Interestingly, the TOC1 protein levels are also reduced by loss of UBP12 or UBP13, mimicking the gi loss-of-function mutant