Ca²⁺-Responsive cis-Elements in Plants

Abstract

External physical and chemical stimuli are transduced via second messengers, following primary interaction with specific membrane or soluble receptors. Ca²⁺ is an important second messenger in plants as in other eukaryotes, mediating responses to numerous environmental stimuli and affecting a multitude of cellular processes including gene expression. However, there is yet very little information concerning the cis-elements that mediate Ca²⁺-responsive gene expression. In this article we discuss a recent investigation combining bioinformatics with experimental data, revealing DNA regulatory elements that convey specific cytosolic Ca²⁺ transients to the transcription machinery.

In response to environmental stimuli, including abiotic (cold, heat, salt, drought, light, touch) and biotic stresses, Ca²⁺ concentrations are transiently elevated, via an increased Ca²⁺ influx.¹ Ca²⁺ transients are transduced by Ca²⁺-binding proteins, many of which contain an ‘EF-hand’ motif, which is a helix-loop-helix structure with high specificity for Ca²⁺ binding.² Ca²⁺ transducers include calmodulin (CaM) and CaM-like proteins (CMLs),^3,4 Ca²⁺-dependent protein kinases,⁵ calcineurin-B-like proteins (CBL),⁶ potassium⁷ and Ca²⁺-channels.⁸ These affect numerous downstream targets and cellular processes including metabolism,⁹ cellular structures¹⁰ and gene expression.^11,12

The apparent complexity of these responses raises questions concerning the mechanisms underlying stimulus-response specificity and cross-talk with other cellular processes.^13–15 Studies of Ca²⁺ signaling in plants and animals suggest that the intracellular Ca²⁺ signals carry different elements of information including duration of the signals, amplitude, and frequency of oscillations.^16–19 These signals need to be decoded¹⁷ and translated to the appropriate cellular responses. However, in spite of the obvious importance of Ca²⁺ signaling, little is known about the mechanisms mediating Ca²⁺-responsive gene expression in plants. In the discussed paper, Ca²⁺-responsive genes and cis-elements in Arabidopsis were characterized. The implications and open questions emerging from these studies are discussed.

The complexity of the Ca²⁺ signaling machinery and stress-induced cellular signaling makes it difficult to distinguish the effects of the Ca²⁺-response per se from other effects evoked by the same stimuli that induce the Ca²⁺ signals. For example, ABA activates both Ca²⁺-dependent and -independent pathways.^{15, 20} In addition, subcellular compartmentalization of Ca²⁺ signaling is also a complex issue that needs to be addressed.²¹ One approach to identify Ca²⁺-specific responses is by induction of artificial Ca²⁺ transients with or without Ca²⁺-signaling inhibitors (e.g., channel blockers, protein inhibitors). Artificial cytosolic Ca²⁺ transients may be achieved by repeatedly changing the ionic composition of the extra-cellular environment from a depolarizing to hyperpolarizing buffer.¹⁶ This type of artificial manipulation revealed the physiological importance of the frequency of Ca²⁺ oscillations.¹⁶ In addition, because several plant Ca²⁺ transporters are regulated by CaM,⁴ exposure of plants to CaM antagonists may also alter their Ca²⁺ homeostasis. Indeed, using aequorin-based luminometry and photon imaging it was shown in the discussed study that CaM antagonists including TFP, W7, calmidazolium and SKF-7171, induced rapid transient changes in cytosolic Ca²⁺ in intact Arabidopsis and tobacco seedlings. These Ca²⁺ bursts peaked 30–100 sec after the addition of the CaM antagonists, lasting 5–10 min, with an elevation of at least 10-fold of cytosolic Ca²⁺ over the measured basal levels. The Ca²⁺ signals triggered by these antagonists were completely inhibited by the Ca²⁺-channel blockers, lanthanum and gadolinium, but not by potassium channel blockers.

Analysis of transcriptome changes 1-hr post stimulus revealed 230 Ca²⁺-responsive genes, of which 162 were upregulated and 68 downregulated. Considering that the chip used in this study represented only 25% of Arabidopsis genome, the total number of upregulated genes might be ∼650, comprising some 2.3% of the whole genome, and a total of 280 downregulated genes, comprising 1% of the genome, for this specific Ca²⁺ response. The upregulated genes contained more TFs, post-transcriptional regulatory proteins, and signaling-related genes, whereas the downregulated genes included more transporters, specifically aquaporins, and defense-related genes, particularly peroxidases. Thus, although the induced cytosolic Ca²⁺ burst was artificial, it allowed the identification of genes responding to a cytosolic Ca²⁺ burst, in the absence of an applied environmental stress. However, many of the upregulated genes were found to be known early stress-responsive genes, such as the touch-responsive genes TCH2 (CML24), TCH3 (CML12) and TCH4-like, the cold-induced genes COR47 and KIN2 and the dehydration early-responsive genes ERD1, ERD6, ERD10, ERD13, ERD14 and ERD15.

Bioinformatic analysis of the 5′ upstream regions of 162 of the upregulated genes revealed a DNA element associated with Ca²⁺-responsive upregulated gene expression (p < 10⁻¹³). The sequence motif [C/A)ACG(T/C)G(T/G/C)] includes the ABRE consensus (ACGTG) and the ABRE-CE core (ACGCG), reported as a functional equivalent of the classical ABREs.²² The analysis also showed that overrepresentation of the ABRE-related motifs occurred exclusively in the upregulated genes. Subsequently, to verify the bioinformatic predictions, a tetramer of the classical ABRE cis-element was tested and found sufficient to confer transcriptional activation in response to cytosolic Ca²⁺ transients, suggesting that ABREs function as Ca²⁺-responsive cis-elements at least in some promoter combinations and Ca²⁺ signaling pathways. However, this does not imply that every ABRE-related promoter sequence is a Ca²⁺-responsive cis-element, or that every cytosolic Ca²⁺ signal is transduced by an ABRE-related regulatory element. Ca²⁺ signals are also associated with the rapid oxidative burst leading to the formation of ROS,²³ which plays a role in both biotic and abiotic signaling.²⁴ In this context it is interesting to note that CaM antagonists (e.g., W7), which induced the cytosolic Ca²⁺ transients in the discussed manuscript, also induced rapid and transient ROS (Fluhr R, personal communication).

Therefore, the results in the discussed paper raise several questions; How Ca²⁺ signals are transduced to the transcription machinery at the ABRE cis-elements? Are these Ca²⁺ signals transduced directly to TFs containing EF-hands, or through TFs that respond to a Ca²⁺-binding protein (e.g., CaM and CMLs), or through other Ca²⁺-responsive proteins like kinases, or phosphatases? Do these Ca²⁺ transduction pathways involve components of ABA-signaling, and if not, how do these Ca²⁺-transduction pathways interact with ABA-signaling pathways operating at the same DNA regulatory elements? How do these Ca²⁺ signaling pathways interact with ROS-signaling in relation to ABA responses? Are these Ca²⁺ signals transduced by downstream ROS signals? Figure 1 depicts some of these possibilities and open questions.

Figure 1.

Possible transduction pathways of stress signals to Ca²⁺-responsive and/or ABA-responsive cis elements. (A) ABA signaling, independent of Ca²⁺ (dotted arrows); (B) Ca²⁺ signaling induced by ABA (black arrows), or other stress induced factors, independent of ABA (white arrows), coinciding at ABRE-related cis-elements; (C) Stress-induced Ca²⁺ signaling acting at cis-elements different from ABREs (stripped arrows); Gray arrows: other signaling pathways not involving ABA or Ca²⁺; Question marks indicate unknown signaling intermediates.

Independently, a novel family of Ca²⁺-dependent CaM-binding TFs, designated CAMTAs (or AtSRs) has been characterized in plants^12,25,26 and in other multicellular organisms.^27,28 In plants, these TFs may function as a link between Ca²⁺ signaling and ABRE-related cis-elements. The DNA-binding specificity of CAMTAs was shown to match both the ABRE-CE core sequence (ACGCGT/G/C) and the classical ABRE (ACGTGT), though with somewhat lower affinity¹² (Finkler A, Fromm H, unpublished results). Interestingly, Choi et al.¹² showed that CAMTA-dependent activation of gene expresssion via a synthetic CAMTA-binding site is inhibited by Ca²⁺/CaM. Putative candidates for activation or repression by CAMTAs via ABRE-related cis-elements are members of the DREB1 family.²⁹ Analysis of DREB1 promoters revealed a number of motifs that correspond to classical ABRE (ACGTG), ABRE-CE motif (CCGCGT or ACGCGG), and ICEr1 motif (CACATG). DREB1A, B and C—all contain in their promoters the ABRE core sequence (ACGTG) with various flanking sequences. The variations in the flanking sequences may impose different affinities for CAMTA binding (Finkler A, Fromm H, unpublished results), with possible implications on fine- tuning of DREB1 expression.

In summary, different approaches may reveal other Ca²⁺-responsive cis-elements and Ca²⁺-responsive TFs in plants. In this context, it should be mentioned that a CaM-binding MYB transcription factor was found to enhance salt tolerance in Arabidopsis,¹¹ and Arabidopsis WRKY group IId TFs also bind CaM.³⁰ Thus, certain Ca²⁺-transduction pathways in plants operate through CaM-binding TFs to modulate gene expression. We found that the Ca²⁺/CaM-responsive CAMTAs bind to the same Ca²⁺-responsive cis-elements that were identified in the discussed study, suggesting a link between Ca²⁺-responsive TFs and ABRE-related cis-elements. This link merits further investigations in the context of ABA, ROS signaling and stress responses.

Abbreviations

CaM

calmodulin

TF

transcription factor

ABA

abscisic acid

ABRE

ABA-responsive element

Addendum to: Kaplan B, Davydov O, Knight H, Galon Y, Knight MR, Fluhr R, Fromm H. Rapid Transcriptome Changes Induced by Cytosolic Ca2+ Transients Reveal ABRE-Related Sequences as Ca2+-Responsive cis Elements in Arabidopsis. Plant Cell. 2006;18:2733–2748. doi: 10.1105/tpc.106.042713.