UV-B-responsive association of the Arabidopsis bZIP transcription factor ELONGATED HYPOCOTYL5 with target genes, including its own promoter

Abstract

In plants subjected to UV-B radiation, responses are activated that minimize damage caused by UV-B. The bZIP transcription factor ELONGATED HYPOCOTYL5 (HY5) acts downstream of the UV-B photoreceptor UV RESISTANCE LOCUS8 (UVR8) and promotes UV-B-induced photomorphogenesis and acclimation. Expression of HY5 is induced by UV-B; however, the transcription factor(s) that regulate HY5 transcription in response to UV-B and the impact of UV-B on the association of HY5 with its target promoters are currently unclear. Here, we show that HY5 binding to the promoters of UV-B-responsive genes is enhanced by UV-B in a UVR8-dependent manner in Arabidopsis thaliana. In agreement, overexpression of REPRESSOR OF UV-B PHOTOMORPHOGENESIS2, a negative regulator of UVR8 function, blocks UV-B-responsive HY5 enrichment at target promoters. Moreover, we have identified a T/G-box in the HY5 promoter that is required for its UV-B responsiveness. We show that HY5 and its homolog HYH bind to the T/G(HY5)-box cis-acting element and that they act redundantly in the induction of HY5 expression upon UV-B exposure. Therefore, HY5 is enriched at target promoters in response to UV-B in a UVR8 photoreceptor-dependent manner, and HY5 and HYH interact directly with a T/G-box cis-acting element of the HY5 promoter, mediating the transcriptional activation of HY5 in response to UV-B.

Figure 1.

UV-B Enhances HY5 Binding to the Promoters of Its Target Genes MYB12 and CHS. (A) ChIP of DNA associated with HY5. ChIP-qPCR was performed for the MYB12 and CHS promoters and an intergenic region between the At4g26900 and At4g26910 genes using 10-d-old wild-type plants (Col) or hy5-215 null mutants grown in white light (no UV-B). ChIP was performed with an anti-HY5 antibody or without addition of the antibody (mock). Data shown are representative of five independent experiments. (B) HY5 ChIP-qPCR using 7-d-old wild-type seedlings grown in weak light and treated with narrowband UV-B for 0.5, 1, and 3 h compared with a −UV-B control. (C) HY5 ChIP-qPCR using 7-d-old wild-type seedlings grown in weak light and treated for 2 h with different intensities of narrowband UV-B compared with an untreated control (−UV-B). The numbers of the analyzed DNA fragments indicate the positions of the 5′ base pair of the amplicon relative to the translation start site (referred to as position +1). ChIP efficiency of DNA associated with HY5 is presented as the percentage recovered from the total input DNA (% Input). Error bars represent sd of three technical replicates.

Figure 2.

HY5 Chromatin Association Is Regulated by the UVR8 Photoreceptor Pathway. UV-B-responsive HY5 chromatin association in wild-type plants (Ws and Col) was compared with that in the uvr8-7 mutant (A) and a RUP2 overexpression line (RUP2 Ox) (B). Seedlings were grown for 7 d in a weak light field and exposed for 2 h to narrowband UV-B. ChIP-qPCR was performed for the MYB12 and CHS promoters and an intergenic region between the At4g26900 and At4g26910 genes. The numbers of the analyzed DNA fragments indicate the positions of the 5′ base pair of the amplicon relative to the translation start site (referred to as position +1). ChIP of DNA associated with HY5 is presented as the percentage recovered from the total input DNA (% Input). Data shown are representative of three (A) and two (B) independent biological replicates. Error bars represent sd of three technical replicates.

Figure 3.

HY5 Associates with RUP1, RUP2, COP1, and BBX24 but Not with the UVR8 Promoter. hy5-215, uvr8-6, and wild-type (Col) seedlings were grown for 7 d in a standard growth chamber followed by weak light acclimation for 12 h and narrowband UV-B irradiation for 2 h. ChIP was performed with an anti-HY5 antibody, and copurified DNA was analyzed by qPCR for different primer pairs amplifying parts of the COP1, UVR8, RUP1, RUP2, and BBX24 genomic regions and an intergenic region between genes At4g26900 and At4g26910. The numbers of the analyzed DNA fragments indicate the positions of the 5′ base pair of the amplicon relative to the translation start site (referred to as position +1). ChIP of DNA associated with HY5 is presented as the percentage recovered from the total input DNA (% Input). Data shown are representative of two independent experiments. Error bars represent sd of three technical replicates.

Figure 4.

HY5 and HYH Act Redundantly in Inducing HY5 Gene Expression and Associate with Common Target Genes, Including the HY5 Promoter. (A) and (B) HY5 associates with its own promoter. Seven-day-old wild-type (Col) seedlings grown in weak light were treated for 2 h with different intensities of narrowband UV-B (A), and hy5-215, uvr8-6, and wild-type (Col) seedlings were grown for 7 d in a standard growth chamber (B). ChIP was performed with an anti-HY5 antibody, and copurified DNA was analyzed by qPCR for different primer pairs covering the HY5 genomic locus and an intergenic region between genes At4g26900 and At4g26910. The numbers of the analyzed DNA fragments indicate the positions of the 5′ base pair of the amplicon relative to the translation start site (referred to as position +1). ChIP of DNA associated with HY5 is presented as the percentage recovered from the total input DNA (% Input). Error bars represent sd of three technical replicates. (C) HYH binds to the HY5 promoter in vivo. Wild-type plants (Ws) and null mutant hyh-1 were grown for 7 d in a weak light field and exposed for 3 h to narrowband UV-B. ChIP was performed with an anti-HYH antibody (left) or without the addition of antibody (mock; right). ChIP-qPCR was performed for different primer pairs covering the HY5 genomic locus and an intergenic region between genes At4g26900 and At4g26910. The numbers of the analyzed DNA fragments indicate the positions of the 5′ base pair of the amplicon relative to the translation start site (referred to as position +1). ChIP of DNA associated with HYH is presented as the percentage recovered from the total input DNA (% Input). Data shown are representative of two independent experiments. Error bars represent sd of three technical replicates. (D) and (E) HY5 and HYH play redundant but essential roles in mediating the responsiveness of the HY5 promoter to UV-B. Luciferase activity is shown for transgenic wild-type Arabidopsis (Ws), hy5, hyh, hy5 hyh, and uvr8-7 plants carrying the same copy of the full-length HY5 promoter fused to the LUC reporter gene. Thirty-six individual seedlings were assayed for each genotype. Error bars represent se.

Figure 5.

Identification of Functional cis-Regulatory Elements of the HY5 Promoter. (A) Deletion analysis of the HY5 promoter. A schematic representation of the 5′ truncated derivatives of the HY5 promoter fused to the LUC reporter gene is shown. Constructs 1 to 4 carry the full 5′ untranslated region of HY5, whereas the 5′ untranslated region was replaced by a 35S minimal promoter in construct 5. Nucleotide positions relative to the first base pair of the 5′ untranslated region (transcriptional start site defined as position +1) are shown. Wild-type Arabidopsis (Ws) seedlings carrying the reporter constructs were grown in a standard growth chamber for 1 week. Plants were transferred to continuous white light at 10 µmol m⁻² s⁻¹ fluence rate for 48 h. Luminescence measurements were started 3 to 4 h before application of the light pulse. Plants were irradiated with narrowband UV-B for 30 min and then returned to white light, where luminescence measurement was resumed. Twenty-four to 36 individual seedlings were assayed for each line, and two independent transgenic lines were tested for each construct. Error bars represent se. (B) to (F) Linker scan mutant derivatives of the HY5 promoter. (B) Putative cis-elements (ACG-box, T/G-box, and E-box) are indicated as boxes, and corresponding nucleotides are framed in the sequence panel and shown in boldface in the wild-type sequence. Mutations are underlined and printed in boldface and are shown in the corresponding mutant sequence line. All mutations were generated in the −565 to +192 fragment of the HY5 promoter, but only sequences from −157 to −71 are shown. Nucleotide positions relative to the first base pair of the 5′ untranslated region (transcriptional start site defined as position +1) are shown. (C) Induction profile of ProHY5_{[−565/+192]}:LUC in response to UV-B pulses (plants were treated as described in [A]). Basal activity was calculated from average counts at the three time points just before the light pulse. The highest luminescence value detected after the light pulse is taken as the maximal induced luminescence. Fold induction is the ratio of maximal induced luminescence to basal activity. Data represent averages of 36 individual seedlings for each condition. (D) to (F) Wild-type Arabidopsis (Ws) seedlings carrying the indicated mutant derivatives of the HY5 promoter fused to the LUC reporter gene were grown and assayed as in (C). Twenty-four to 36 individual seedlings were assayed for each line, and two independent transgenic lines were tested for each construct. Basal activity (D), maximal induced luminescence (E), and fold induction (F) were calculated according to (C). Error bars represent se.

Figure 6.

HY5 and HYH Bind to the cis-Regulatory Element T/G^HY5-Box. Biotin-labeled double-stranded probes (40 fmol) were incubated with (+) or without (−) expressed and purified HY5 ([A] and [C]) or HYH ([B] and [D]) protein (400 ng). Binding reactions were resolved on 6% native polyacrylamide gels. WT corresponds to the −117 to −62 fragment of the HY5 promoter, whereas MUT8, MUT10, MUT12, MUT8-10, and MUT8-10-12 carry the corresponding single, double, and triple linker scan mutations in this context. The mE-box probe has mutations identical to those described by Abbas et al. (2014). In (C), reactions were performed as in (A) using HY5 wild-type probe and HY5 protein, but unlabeled HY5 MUT8 or HY5 MUT10 fragments were used as cold competitors at a 200-fold molar excess. In (D), reactions were performed as in (B) using HY5 wild-type probe and HYH protein, but unlabeled HY5 MUT8 or HY5 MUT10 fragments were used as cold competitors at a 200-fold molar excess. Sequences of all probes are provided in Supplemental Table 2. FP indicates free (nonbound) probe. Asterisks mark a nonspecific band that appears independent of the presence of HY5 or HYH protein.

Figure 7.

Working Model of the Transcriptional Regulation of HY5. HY5 and HYH binding to the T/G-box mediates UV-B responsiveness of the HY5 promoter. An ACG-box is postulated to be bound by an unknown repressor protein X that is potentially removed from the HY5 promoter or overruled by the UV-B responsiveness of the T/G-box in response to UV-B (note that the ACG-box requires the T/G-box for its repressive function in –UV-B). The E-box was previously found to be bound by CAM7 (Abbas et al., 2014) and seems to make only a very minor contribution to UV-B responsiveness. The accumulated HY5 protein (combination of new synthesis and stabilization) then binds to and activates multiple downstream target genes, including genes encoding UV-B signaling components (COP1, BBX24, RUP1, and RUP2).