CONSTITUTIVELY PHOTOMORPHOGENIC1 is required for the UV-B response in Arabidopsis

Abstract

CONSTITUTIVELY PHOTOMORPHOGENIC1 (COP1) is a negative regulator of photomorphogenesis in Arabidopsis thaliana. COP1 functions as an E3 ubiquitin ligase, targeting select proteins for proteasomal degradation in plants as well as in mammals. Among its substrates is the basic domain/leucine zipper (bZIP) transcription factor ELONGATED HYPOCOTYL5 (HY5), one of the key regulators of photomorphogenesis under all light qualities, including UV-B responses required for tolerance to this environmental threat. Here, we report that, in contrast with the situation in visible light, COP1 is a critical positive regulator of responses to low levels of UV-B. We show that in the cop1-4 mutant, flavonoid accumulation and genome-wide expression changes in response to UV-B are blocked to a large extent. COP1 is required for HY5 gene activation, and both COP1 and HY5 proteins accumulate in the nucleus under supplementary UV-B. SUPPRESSOR OF PHYTOCHROME A-105 family proteins (SPA1 to SPA4) that are required for COP1 function in dark and visible light are not essential in the response to UV-B. We conclude that COP1 performs a specific and novel role in the plants' photomorphogenic response to UV-B, coordinating HY5-dependent and -independent pathways, which eventually results in UV-B tolerance.

Figure 1.

hy5-1 Is Deficient in UV-B–Activated Gene Expression and UV-B Tolerance. (A) Numbers of genes defined as responding to UV-B in hy5-1 and wild-type Ler. Venn diagram shows the number of genes classified as responding to UV-B in either the wild type only (left; i.e., HY5-dependent genes), the wild type and mutant (center; i.e., HY5-independent genes), or mutant only (right). The corresponding gene list can be found in Supplemental Table 1 online. (B) HY5 is required for UV-B tolerance. Wild-type and hy5-1 plants were grown in white light with or without supplementary UV-B. Plants were photographed after 4 weeks. Under supplementary UV-B, 13 plants died and 35 showed necrotic lesions among the hy5-1 mutants, whereas in the parallel-grown wild-type population, no plant died or was necrotic (n = 50). When control plants from both genotypes were grown without supplementary UV-B, all survived without apparent necrosis.

Figure 2.

COP1 Is Required for UV-B–Responsive Gene Expression, Including HY5 Gene Activation. (A) RNA gel blot analysis of the cop1-4 mutant compared with its wild type (Col). Total RNA was isolated from 7-d-old white light–grown seedlings at 45 min after 15-min irradiation under a UV-B field with different UV spectra (−UV-B under cutoff WG327; +UV-B with decreasing short-wave cutoff under WG305, WG295, or quartz glass [Q]) or left untreated in the standard growth chamber (control [C]). Blots were sequentially hybridized with specific probes for the indicated genes. Ethidium bromide–stained rRNA is shown as loading control. (B) New cop1-4 allele identified in a luciferase-based genetic screen. Normalized data of TopCount luciferase bioluminescence measurements of 10 Pro_HY5:Luc⁺ seedlings each after 15 min of UV-B irradiation. Gray: parental Pro_HY5:Luc⁺ transgenic line (wild type); black: mutant 0650 (due to COP1-4 mutation). Note that the data for each seedling were normalized to their mean (absolute values not shown) to facilitate the comparison of induction properties. An absolutely blind mutant is expected to have a normalized luminescence value of 1 over the time course. (C) Numbers of genes defined as responding to UV-B in the cop1-4 mutant and wild-type Col. Venn diagram shows the number of genes classified as responding to UV-B in either the wild type only (left; i.e., COP1-dependent genes), the wild type and mutant (center; i.e., COP1-independent genes), or mutant only (right). The corresponding gene list can be found in Supplemental Table 2 online. (D) Overlap between COP1- and HY5-independent genes. Analysis limited to genes commonly UV-B induced in both wild types Col and Ler. Venn diagram depicts the relationship of genes induced in cop1-4 and hy5-1. The total number of genes in each category is shown in parentheses. UV-B activation of the 103 genes indicated in the lower left is blocked in both cop1-4 and hy5-1 mutants (see Table 1). The corresponding gene list can be found in Supplemental Table 3 online.

Figure 3.

cop1 Plants Develop Chlorosis under Supplementary UV-B. Wild-type, cop1-4, cop1-4/Pro_35S:YFP-COP1, and det1-1 plants were grown in white light with or without supplementary UV-B. Plants were photographed after 5 weeks. Under supplementary UV-B, 37 plants developed chlorosis among the cop1-4 mutant, whereas among the parallel-grown wild type, cop1-4/Pro_35S:YFP-COP1, and det1-1 mutant, no plant was chlorotic (n = 50). When control plants from all four genotypes were grown without supplementary UV-B, all survived without apparent chlorosis. Bars = 1 cm.

Figure 4.

Functional HY5 and COP1 Are Required for UV-B–Responsive Flavonoid Accumulation, Correlating with HY5 Induction Properties. Seedlings were grown for 4 d under continuous light supplemented with UV-B under a 327-nm cutoff filter (−UV-B) or 305-nm cutoff filter (+UV-B). The 327-nm cutoff filters were exchanged after 4 d for a 305-nm cutoff at 1 h and 6 h before harvesting, where indicated. (A) Reduced hypocotyl growth inhibition by supplementary UV-B in hy5-1. Hypocotyl length was measured from seedlings of phyA-201 phyB-5, hy5-1, cop1^eid6, and cop1-4 mutants, together with their wild types (Ler and Col), after 4 d of growth under white light (WL) or white light supplemented with UV-B (WL+UV-B). Numbers below bars show the relative hypocotyl growth inhibition by UV-B as a percentage. Bars show se of the mean (n = 120). (B) UV-B–induced flavonoid accumulation is impaired in cop1-4 and hy5-1 mutants but not in cop1^eid6. Numbers below bars show the fold enrichment values of supplemental +UV-B compared with −UV-B. Bars show se of the mean (n = 9). (C) Scheme for UV-B treatment. The time under a WG327 cutoff filter (−UV-B) is represented by a white bar and time under a WG305 cutoff filter (+UV-B) by a gray bar. Time is indicated in hours. (D) Corresponding RNA gel blot showing UV-B–responsive CHS and HY5 gene expression. (E) UV-B–induced HY5 protein accumulation is impaired in cop1-4 but not in cop1^eid6. The protein gel blot was sequentially probed with anti-HY5 and anti-actin antibodies.

Figure 5.

UV-B–Mediated HY5 Gene Activation and HY5 Protein Accumulation Correlates with Induction of Putative HY5 Target Genes. (A) RNA gel blot analysis of the wild type (Col). Total RNA was isolated from 4-d-old wild-type seedlings (Col) that were irradiated with UV-B for the indicated times before harvesting. Blots were sequentially hybridized with specific probes for the indicated genes. Ethidium bromide–stained rRNA is shown as loading control. (B) Corresponding UV-B–responsive accumulation of HY5 and CHS protein in the Col wild type. The protein gel blot was sequentially probed with anti-HY5, anti-CHS, and anti-actin antibodies.

Figure 6.

YFP-COP1 Localizes to the Nucleus in Response to UV-B, Not Affecting HY5 Accumulation. (A) The YFP-COP1 fusion protein is functional and complements the cop1-4 mutant phenotype. Top: after 6 h of light treatment to synchronize germination, the seedlings were grown for 4 d in the dark before the photograph was taken. Middle: 4-d-old seedlings grown under constant light were either kept in light (L) or transferred to dark (D) for 24 h. The protein gel blot was sequentially analyzed with anti-HY5 and anti-actin antibodies. Bottom: RNA gel blot. Total RNA was isolated from 7-d-old white light–grown seedlings at 45 min after 15-min irradiation under a UV-B field with different UV spectra (−UV-B under cutoff WG327; +UV-B with decreasing short-wave cutoff under WG305 or quartz glass [Q]). Ethidium bromide–stained rRNA is shown as loading control. (B) to (D) Seedlings were grown for 4 d under continuous light supplemented with UV-B under a 327-nm cutoff filter (−UV-B) or 305-nm cutoff (+UV-B). The 327-nm cutoff filters were exchanged after 4 d for a 305-nm cutoff at 1 h and 6 h before harvesting, as indicated (C). (B) UV-B–responsive accumulation of YFP-COP1 in the nucleus. Nuclear accumulation is confirmed by Hoechst 33342 nuclear stain (bottom panel). Bars = 20 μm. (C) Accumulation of YFP-COP1 still allows UV-B–induced HY5 protein accumulation. The protein gel blot was sequentially analyzed with anti-HY5, anti-YFP, and anti-actin antibodies. (D) UV-B–responsive accumulation of HY5-YFP in the nucleus. Bars = 20 μm.

Figure 7.

COP1 Function in the UV-B Response Is Independent of SPA Proteins. RNA gel blot analysis of the spa1234 quadruple mutant and the cop1-1 single mutant in comparison with the wild type (Col). Total RNA was isolated from 7-d-old white light–grown seedlings at 45 min after 15-min irradiation under a UV-B field with different UV spectra (−UV-B under cutoff WG327; +UV-B with decreasing short-wave cutoff under WG305 or quartz glass [Q]) or left untreated in the standard growth chamber (control [C]). Blots were sequentially hybridized with specific probes for the indicated genes. Ethidium bromide–stained rRNA is shown as loading control.

Figure 8.

Working Model of an Early Function of COP1 and HY5 in UV-B Signaling. We propose that UV-B perceived by the postulated UV-B photoreceptor activates COP1 to degrade a repressor of HY5 and other UV-B–responsive genes, including yet unidentified transcription factors X and Y. These transcription factors will act in concert to regulate gene expression, leading to a photomorphogenic UV-B response. As determined by microarray analysis, the 103 COP1- and HY5-dependent and 137 COP1-dependent but HY5-independent genes are indicated (Figure 2D; see Supplemental Table 3 online).