Cold induction of Arabidopsis CBF genes involves multiple ICE (inducer of CBF expression) promoter elements and a cold-regulatory circuit that is desensitized by low temperature

Abstract

The Arabidopsis CBF1, 2, and 3 genes (also known as DREB1b, c, and a, respectively) encode transcriptional activators that have a central role in cold tolerance. CBF1-3 are rapidly induced upon exposing plants to low temperature, followed by expression of CBF-targeted genes, the CBF regulon, resulting in an increase in plant freezing tolerance. At present, little is known about the cold-sensing mechanism that controls CBF expression. Results presented here indicate that this mechanism does not require a cold shock to bring about the accumulation of CBF transcripts, but instead, absolute temperature is monitored with a greater degree of input, i.e. lower temperature, resulting in a greater output, i.e. higher levels of CBF transcripts. Temperature-shift experiments also indicate that the cold-sensing mechanism becomes desensitized to a given low temperature, such as 4 degrees C, and that resensitization to that temperature requires between 8 and 24 h at warm temperature. Gene fusion experiments identified a 125-bp section of the CBF2 promoter that is sufficient to impart cold-responsive gene expression. Mutational analysis of this cold-responsive region identified two promoter segments that work in concert to impart robust cold-regulated gene expression. These sequences, designated ICEr1 and ICEr2 (induction of CBF expression region 1 or 2), were also shown to stimulate transcription in response to mechanical agitation and the protein synthesis inhibitor, cycloheximide.

Figure 1.

Accumulation of CBF transcripts in response to low temperatures. A, Plants were abruptly transferred from 22°C to 4°C for the indicated periods and CBF transcript levels were determined by RNA-blot hybridization. Gene-specific probes were used to detect, independently, CBF1, 2, and 3 transcripts. B, Plants were abruptly transferred from 22°C to the indicated cold temperatures for 2 h and CBF transcript levels were determined by RNA-blot hybridization. The probe used, a full-length cDNA of CBF1, cross-hybridized with CBF1, 2, and 3 transcripts. C, Plants grown at warm temperatures were subjected to a cold shock or to a gradual temperature decrease, and CBF and COR15a transcript levels were determined by RNA-blot hybridization. The full-length cDNA for CBF1 was used to detect CBF1, 2, and 3 transcripts. In each experiment, rRNA stained with ethidium bromide was used to compare loading.

Figure 2.

Desensitization and resensitization of the regulatory circuitry controlling accumulation of CBF transcripts. A, Plants were subjected to repeated transfer between 20°C (W) and 4°C (C), and CBF and COR15a transcript levels were determined by RNA-blot hybridization. B, CBF transcript levels of in plants that were cold acclimated at 4°C (C) for 14 d, returned to warm temperatures for the indicated times, and then subjected to a cold shock (WCS) at 4°C. CBF transcript levels were also determined in cold-acclimated plants (14 d) that were transferred (Xfer) to 0°C or –5°C 2 h. C, CBF and COR15a transcripts levels in plants cooled at 2°C h^–1. Plants were held at 10°C for 48 h before cooling to 4°C. The probe used hybridized with CBF1, 2, and 3 transcripts. rRNA stained with ethidium bromide was used to compare loading in each experiment.

Figure 3.

Estimation of CBF transcript half-life at warm temperatures. Plants that had been exposed to a low temperature (4°C) for 2 h were returned to a warm temperature (24°C) and CBF transcript levels were determined at 10-min intervals by RNA-blot hybridization. Air and agar temperatures in a plate were recorded during the experiment. Transcript levels for eif-4A did not change significantly during the time course and were used as a reference for loading. The probe for CBF hybridized with CBF1, 2, and 3 transcripts.

Figure 4.

Activation of the CBF1, 2, and 3 promoters in response to low temperatures. Representative transgenic lines of Arabidopsis carrying the CBF1, 2, or 3 promoter fused to the GUS reporter gene were transferred from 22°C (W) to 4°C (C) for 2 h. The transcript levels of the gene fusions were determine using a probe for the GUS sequences.

Figure 5.

CBF2 promoter fragments used to identify cold-regulatory elements. Each fragment was inserted into a binary vector upstream of the GUS reporter gene. Fragments that were deleted from the 3′ end were inserted into a binary vector upstream of the cauliflower mosaic virus (cauliflower mosaic virus) 35S minimal promoter (–46 minimal promoter) fused to GUS. A, Promoter fragments with deletions from the 5′ end. B, Promoter fragments with deletions from the 3′ end. C, Dimer constructs. D, Sequence of the 125-bp region of the CBF2 promoter (positions –189 to –65) that imparts cold-regulated gene expression. The locations of the conserved boxes IV, V, and VI are indicated and the E-box/ABRE-like sequences are underlined. Relative levels of cold induction (data from Figs. 6, 7, 8) are indicated as “+” (strong) and “–” (weak).

Figure 6.

Activities of CBF2::GUS promoter constructs. RNA isolated from plants grown at 22°C (W) or grown at 22°C and transferred to 4°C for 2 h (C) was analyzed for GUS transcript levels by RNA-blot hybridization. A and B, Plants containing 5′ deletion constructs. C, Plants containing 3′ deletion constructs. D, Plants containing dimer constructs. rRNA stained with ethidium bromide was used to compare loading in each experiment.

Figure 7.

Mutational analysis identifying regions of the CBF2 promoter involved in cold-regulated gene expression. A, CBF2::GUS promoter constructs with mutations in boxes IV (mboxIV), V (m2), and VI (m3 and m4) were tested for cold responsiveness by transferring plants from 22°C (W) to 4°C (C) for 2 h and determining GUS transcript levels by RNA-blot hybridization. rRNA stained with ethidium bromide was used to compare loading. B, Sequence corresponding to wild-type and mutated (m2, m3, m4, and mboxIV) CBF2::GUS promoter constructs. Regions identified as required for cold induction are underlined and designated ICEr1 and ICEr2.

Figure 8.

Sequences of the CBF2 promoter involved in cold, cycloheximide, and mechanical responsiveness. Plants containing CBF2 promoter constructs were grown at 22°C (W) and then exposed to 4°C (C) for 2 h or treated with cycloheximide (X) or mechanical agitation (M) as described in “Materials and Methods.” RNA was isolated and GUS transcript levels were determined by RNA-blot hybridization. A, Plants containing the 5′ deletion to –189 (5′Del) or the –189/–65 dimer (Dimer) construct. B, Plants containing the m2, m3, m4, and mboxIV mutations. rRNA stained with ethidium bromide was used to compare loading in each experiment.