The bZIP transcription factor HY5 interacts with the promoter of the monoterpene synthase gene QH6 in modulating its rhythmic expression

Abstract

The Artemisia annua L. β-pinene synthase QH6 was previously determined to be circadian-regulated at the transcriptional level, showing a rhythmic fluctuation of steady-state transcript abundances. Here we isolated both the genomic sequence and upstream promoter region of QH6. Different regulatory elements, such as G-box (TGACACGTGGCA, -421 bp from the translation initiation site) which might have effects on rhythmic gene expression, were found. Using the yeast one-hybrid and electrophoretic mobility shift assay (EMSA), we confirmed that the bZIP transcription factor HY5 binds to this motif of QH6. Studies with promoter truncations before and after this motif suggested that this G-box was important for the diurnal fluctuation of the transgenic β-glucuronidase gene (GUS) transcript abundance in Arabidopsis thaliana. GUS gene driven by the promoter region immediately after G-box showed an arrhythmic expression in both light/dark (LD) and constant dark (DD) conditions, whereas the control with G-box retained its fluctuation in both LD and DD. We further transformed A. thaliana with the luciferase gene (LUC) driven by an 1400 bp fragment upstream QH6 with its G-box intact or mutated, respectively. The luciferase activity assay showed that a peak in the early morning disappeared in the mutant. Gene expression analysis also demonstrated that the rhythmic expression of LUC was abolished in the hy5-1 mutant.

Keywords: Artemisia annua; G-box; HY5; QH6; circadian rhythm; monoterpene synthase.

Figure 1

QH6 gene structure showing sites and sizes of exons and introns, predicted conserved domains, and positions of designed primers. M, methionine as a translation start signal; RR(X)₈W, conserved tandem arginine as a transit peptide signal (R⁴⁸R⁴⁹), and a conserved tryptone (W⁵⁸) at the 9th residue after RR. The conserved RXR domain (R³⁰⁰D³⁰¹R³⁰²) used to direct the diphosphate anion away from the reactive carbocation after ionization and the DDXXD (D³³⁷D³³⁸V³³⁹Y³⁴⁰D³⁴¹) domain for divalent cation (typically Mg²⁺ or Mn²⁺)-assisted substrate binding were both found at the 4th exon. Sequences of primers are listed in Supplementary Table S1.

Figure 2

Sequence analysis of the QH6 promoter and the construction of different promoter truncations driving GUS expression in Arabidopsis thaliana. The sequence from −800 bp to the QH6 translation initiation site was searched for known regulatory elements in PLACE and NSITE-PL databases to identify any known motifs, as described in Materials and Methods Section.

Figure 3

Rhythmic expression of GUS driven by the upstream fragment of the QH6 promoter immediately before (GB+) or after (GB−) the G-box motif. Arabidopsis thaliana plants were transformed by GB+ or GB− constructs. The steady-state level of GUS transcripts was quantified by real-time PCR, using Actin8 as a reference gene. Leaves were sampled in triplicate for two continuous days, with the first day in standard 12h/12h light/dark cycle and the second day in constant dark. ZT is Zeitgeber time and here represents time from light on at dawn. Filled and open bars indicate subjective days and nights. Data represent means ± SD. The differences of expression levels were assessed by unpaired Student's t-test. ^*p < 0.05 and ^**p < 0.01 in GB+ vs. GB− transgenic Col-0 plants.

Figure 4

Diurnal expression of the luciferase gene (LUC) driven by the 1400 bp upstream fragment of the QH6 promoter, with the G-box intacted (1400::LUC) or mutated (1400M::LUC). Arabidopsis thaliana Col-0 plants were transformed by 1400::LUC and 1400M::LUC constructs. The activity of luciferase expressed was quantified by measuring the radiance emitted and captured in 4 min by IVIS Lumina XR Imaging System (Caliper Life Sciences). Plants were growing in standard 12h/12h light/dark conditions. ZT is Zeitgeber time and here represents time from light on at dawn. Filled and open bars indicate subjective days and nights. Data represent means ± SD. The differences of expression levels were assessed by unpaired Student's t-test. ^*p < 0.05 and ^**p < 0.01 in 1400::LUC vs. 1400M::LUC transgenic Col-0 plants.

Figure 5

Rhythmic expression of the luciferase gene (LUC) driven by the 1400 bp upstream fragment of the QH6 promoter in transgenic Arabidopsis thaliana Ler-0 and hy5-1 mutant lines. The steady-state level of LUC transcripts was quantified by real-time PCR, using Actin8 as a reference gene. Leaves were sampled in triplicate for two continuous days, with the first day in standard 12h/12h light/dark cycle and the second day in constant dark. ZT is Zeitgeber time and here represents time from light on at dawn. Filled and open bars indicate subjective days and nights. Data represent means ± SD. The differences of expression levels were assessed by unpaired Student's t-test. ^*p < 0.05 and ^**p < 0.01 in 1400::LUC transgenic Ler-0 vs. 1400::LUC transgenic hy5-1.

Figure 6

The yeast one-hybrid showed that the QH6 promoter region could bind HY5 in vivo. The QH6 promoter fragment from −258 to −425 bp upstream of the translation initiation site, including the G-box, was used as the bait (+GB), and the same fragment with the G-box removed (-GB) or mutated (GBM) was used as the control. A sequence containing four consecutive repeats of the G-box motif (4 GB) was also synthesized to test enhanced interactions. pHIS2.1 harboring +GB, -GB, GBM, or 4GB was co-transformed with pGADT7-HY5. Cells were grown in liquid medium to OD₆₀₀ = 0.1 (10⁻¹) and diluted in a 10 × series (10⁻² to 10⁻⁵). At each dilution, 5 μ L was spotted on media selecting for the existence of both plasmids (-Leu/-Trp double drop-out) or the interaction (-Leu/-Trp/-His triple drop-out).

Figure 7

EMSA confirmed the interaction between HY5 and the QH6 G-box motif in vitro. GST and GST-HY5 proteins were expressed and purified as described in the Materials and Methods Section. For each binding reaction, 5 μ g of recombinant protein was used. A double-stranded probe was generated by annealing synthesized single strand primers. The −391 to −442 bp fragment of the QH6 promoter was biotin-labeled as 391-442, while M391-442 represents a mutant probe. In each lane, 20 fmol of biotin-labeled probe was used. In lane 1, only 20 fmol of the wild-type probe was used, while in lane 2, 20 fmol of the wild-type probe was used along with 5 μ g GST-HY5 protein. An unlabeled probe was used as a competitor as marked on the left. In lanes 3–6, unlabeled competitors (0.5 pmol, 1 pmol, 2 pmol, 4 pmol, respectively) were added to the binding assay with the GST-HY5 protein and the 391-442 biotin-labeled probe. The relative concentration ratio was marked as in the competitor row. Negative controls were 20 fmol mutant probe together with 5 μ g GST-HY5 protein (Lane 7) and 20 fmol wild-type probe with 5 μ g GST protein (Lane 8).