Abscisic acid induces CBF gene transcription and subsequent induction of cold-regulated genes via the CRT promoter element

Abstract

Many cold-regulated genes of Arabidopsis are inducible by abscisic acid (ABA) as well as by cold. This has been thought to occur via two separate signaling pathways, with ABA acting via ABA-responsive promoter elements and low temperature activating the C-repeat element (CRT; dehydration-responsive) promoter element via CBF (DREB1) transcription factors. We show here that ABA is also capable of activating the CRT promoter element. Although the more recently discovered ABA-inducible CBF4 transcription factor might have accounted for this, we show here that CBF1-3 transcript levels also increase in response to elevated ABA levels. This increase in CBF1-3 transcript levels appears to be at least in part due to increased activity of the CBF promoters in response to ABA. A total of 125 bp of the CBF2 promoter, which has previously been shown to be sufficient for cold-, mechanical-, and cycloheximide-induced expression, was also sufficient for ABA-induced expression. However, the ABA-responsive promoter element-like motif within this region is not needed for ABA-induced expression. An observed increase in CBF protein levels after ABA treatment, together with previous data showing that increased CBF levels are sufficient for cold-regulated gene induction, suggests that ABA-induced increases in CBF1-3 transcript levels do have the potential to activate the CRT. Our data indicate therefore that activation of the CRT may also occur via a novel ABA-inducible signaling pathway using the normally cold-inducible CBFs.

Figure 1.

Induction of CRT::LUC and COR gene expression by ABA or low temperature treatment. Plants were treated for (A) 1, 2, 3, or 6 h with 100 μm ABA (+) or 0.1% ethanol control (−) or for (B) 0, 2, 6, 12, or 24 h at 5°C before harvesting tissue for northern analysis. Samples were loaded on the same gel and transferred to a filter, which was probed with a LUC probe and reprobed to detect levels of the COR genes LTI78 and KIN1/2. rRNA stained with ethidium bromide was used to compare loading between lanes and the blot was also reprobed with a β-TUBULIN probe to confirm equal loading between lanes.

Figure 2.

Induction of CBF expression by ABA or low temperature treatment. The filter used for Figure 1 was reprobed with a probe for CBF. A, Samples from plants treated for 1, 2, 3, or 6 h with 100 μm ABA (+) or 0.1% ethanol control (−). B, Samples from plants treated for 0, 2, 6, 12, or 24 h at 5°C before harvesting tissue for northern analysis. rRNA stained with ethidium bromide was used to compare loading between lanes.

Figure 3.

Analysis of ABA-inducibility of individual CBF genes. Samples were extracted from plants treated for 1, 2, 4, or 24 h with 100 μm ABA (+) or 0.1% ethanol control (−) and transcript levels of CBF1, CBF2, and CBF3 detected by northern blot using gene-specific probes. rRNA stained with ethidium bromide was used to compare loading between lanes.

Figure 4.

Activation of CBF1, 2, and 3 promoters in response to ABA. Samples were extracted from plants expressing CBF promoter GUS fusions and treated for 1, 2, 4, or 24 h with 100 μm ABA (+) or 0.1% ethanol control (−) and GUS transcript levels detected by northern blot. rRNA stained with ethidium bromide was used to compare loading between lanes.

Figure 5.

Activation of the wild type and mutated −189 deleted CBF2 promoter in response to ABA. Samples were extracted from plants expressing the GUS gene fused to various full-length and deletion versions of the CBF2 promoter, in response to a 1.5-h treatment with with 100 μm (+) or 0.1% ethanol control (−) and GUS transcript levels detected by northern blot. rRNA stained with ethidium bromide was used to compare loading between lanes. A, Comparison of GUS expression driven by either the full-length CBF2 promoter (CBF2pro) or by a 5′ deletion of this promoter to −189 bp upstream of the start codon. B, Comparison of GUS expression driven by either the full-length promoter (CBF2pro) or the −189/−65 sequence from the CBF2 promoter, as a dimer fused to the CaMV minimal promoter (Zarka et al., 2003). The −189/−65 sequence is shown in the inset. C, Comparison of GUS expression driven by either the full-length CBF2 promoter (CBF2pro) or the mutated version of the −189 deletion construct (m2) in which a putative ABA-responsive element/bZIP binding site sequence CACGTG has been changed to TCTAGA. Two independent transgenic lines (m2 1-6 and m2 2-1) are shown.

Figure 6.

Accumulation of CBF-AEQUORIN fusion protein in response to cold and ABA. Seven-day-old plants were used, expressing a CBF-AEQUORIN fusion protein. Aequorin activity was measured by an in vitro assay. The values shown are means of three replicates from 5 plants/sample, after the subtraction of background luminescence. Five seedlings were harvested per replicate, and results are typical of those obtained with three different transgenic lines. Error bars represent SEM. Plants were treated for 1, 3, or 6 h with 100 μm ABA or 0.1% ethanol (control, A), or for 0, 2, 6, or 24 h at 5°C (B).