A loss-of-function mutation in the DWARF4/ PETANKO5 gene enhances the late-flowering and semi-dwarf phenotypes of the Arabidopsis clock mutant lhy-12;cca1-101 under continuous light without affecting FLC expression

Abstract

The circadian clock plays important roles in the control of photoperiodic flowering in Arabidopsis. Mutations in the LATE ELONGATED HYPOCOTYL (LHY) and CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) genes (lhy;cca1) accelerate flowering under short days, whereas lhy;cca1 delays flowering under continuous light (LL). The lhy;cca1 mutant also exhibits short hypocotyls and petioles under LL. However, the molecular mechanisms underlying the regulation of both flowering time and organ lengths in the LHY/CCA1-dependent pathway are not fully understood. To address these questions, we performed EMS mutagenesis of the lhy-12;cca1-101 line and screened for mutations that enhance the lhy;cca1 phenotypes under LL. In this screen, we identified a novel allele of dwarf4 (dwf4) and named it petanko 5 (pta5). A similar level of enhancement of the delay in flowering was observed in these two dwf4 mutants when combined with the lhy;cca1 mutations. The lhy;cca1 and dwf4 mutations did not significantly affect the expression level of the floral repressor gene FLC under LL. Our results suggest that a defect in brassinosteroid (BR) signaling delayed flowering independent of the FLC expression level, at least in plants with the lhy;cca1 mutation grown under LL. The dwf4/pta5 mutation did not enhance the late-flowering phenotype of plants overexpressing SVP under LL, suggesting that SVP and BR function in a common pathway that controls flowering time. Our results suggest that the lhy;cca1 mutant exhibits delayed flowering due to both the BR signaling-dependent and -independent pathways under LL.

Keywords: CCA1; DWF4; LHY; brassinosteroid; flowering time.

Figure 1. Effects of an enhancer mutation on organ elongation in lhy;cca1 in LL. (A) Images of wild-type (WT, Ler), lhy-12;cca1-101, pta5, and lhy-12;cca1-101;pta5 grown under LL for 2 weeks. Scale bar=1 cm. (B) Lengths of leaves of WT (Ler), lhy-12;cca1-101, pta5, and lhy-12;cca1-101;pta5 grown under LL for 4 weeks. Single and double asterisks denote statistical significance in comparison with the WT (Ler) and lhy-12;cca1-101, respectively (Student’s t-test, p<0.001). Means±SE are shown (n≧10). (C) Lengths of hypocotyls of WT (Ler), lhy-12;cca1-101, pta5, and lhy-12;cca1-101;pta5 grown under LL for 2 weeks; data are means of 10 plants±SE Single and double asterisks denote statistical significance in comparison with the values for WT (Ler) and lhy-12;cca1-101, respectively (Student’s t-test, p<0.001). Each experiment was performed at least twice with similar results.

Figure 2. Identification of pta5 as a new missense allele of dwf4. (A) Allelism test between dwf4-101 (Ws) and pta5 (Ler). Asterisks denote statistical significance in comparison with the values for WT (Ler)×WT (Ws) (Student’s t-test, p<0.01). (B) Gene and protein structure of DWF4. A C-to-T substitution point mutation occurred in the eighth exon of DWF4. (C) Flowering times of WT (Ler), pta5, lhy-12;cca1-101, and lhy-12;cca1-101;pta5 plants grown under LL. (D) Flowering times of WT (Ws), dwf4-101, lhy-21;cca1-11, and lhy-21;cca1-11;dwf4-101 plants grown under LL. Flowering times were scored by counting the total numbers of rosette and cauline leaves on the main stem after bolting. Means are shown±SE (n≧10). Each experiment was performed at least twice with similar results. Single and double asterisks denote statistical significance in comparison with the WT (Ler or Ws) and lhy;cca1 (Ler or Ws) (Student’s t-test, p<0.01).

Figure 3. Expression levels of the floral activator genes FT and SOC1 and the repressor gene FLC in plants grown under LL for 14, 21, or 28 days. The levels of the FT, SOC1, and FLC mRNAs in WT (Ler), pta5, lhy-12;cca1-101, and lhy-12;cca1-101;pta5 relative to TUB2 were measured by semi-quantitative (A) and quantitative (B) RT-PCR. Numbers below the bands in (A) indicate relative expression levels compared to TUB2. For quantitative RT-PCR (B), cDNA prepared from plants grown under LL for 28 days was used. The FT, FLC, and SOC1 transcript levels were normalized to the expression of TUB2 measured in the same RNA samples. Data are the mean±SD of three independent RNA samples.

Figure 4. Genetic analysis of 35S:SVP;pta5 and svp-3;pta5, and a hypothetical model. (A) Flowering times of WT (Ler), pta5, 35S:SVP, and 35S:SVP;pta5 grown under LL. Single and double asterisks denote statistical significance in comparison with the WT (Ler) and 35S:SVP, respectively (Student’s t-test, p<0.01). (B) Flowering times of WT (Ler), pta5, svp-3, and svp-3;pta5 grown under LL. Single and double asterisks denote statistical significance in comparison with the WT (Ler) and svp-3, respectively (Student’s t-test, p<0.01). Means are shown±SE (n=10). Each experiment was performed at least twice with similar results. (C) A hypothetical model of the regulation of organ elongation and flowering by LHY, CCA1, SVP, and BR.