Background

The calcium ion (Ca2+) functions as a ubiquitous second messenger in plants, and participates in signal transduction in response to various environmental and developmental stimuli. The transient Ca2+ signal in the cytoplasm of plant cells is captured by a variety of Ca2+ sensors including calmodulins, calmodulin-like proteins (CML), calcium-dependent protein kinases (CDPKs), and calcineurin B-like (CBL) proteins [1–3]. These Ca2+ sensors transfer the environmental and developmental signals to downstream components, causing a series of physiological and biochemical changes.

CBLs, which harbor a core region consisting of four conserved EF hands, capture the Ca2+ signal, and interact with CBL-interacting protein kinases (CIPKs) to phosphorylate their target proteins [4]. The CIPKs with the NAF motif for interaction with CBL belong to the SnRK3 protein family, whose members have a specific Ser/Thr protein kinase domain [5]. Recently, CIPK proteins were shown to phosphorylate their interacting CBL proteins at N-terminal conservative Ser residue (AtCBL1-Ser201, AtCBL2-Ser216, AtCBL4-Ser205 and AtCBL10-Ser237) [6–8] and this proper phosphorylation of CBL was proved to be absolutely required for the in vivo activation of the AKT1 K+ channel by CBL1-CIPK23 and CBL9-CIPK23 complexes in oocytes [6]. The CBL and CIPK gene families have been described in Arabidopsis thaliana, Oryza sativa, Zea mays, and Brassica napus [9–12].

In previous studies, CBL–CIPK complexes were found to play very important roles in responses to external stimuli. The salt overly sensitive (SOS) pathway was the first CBL-CIPK network to be identified, and this CBL-CIPK complex comprised CBL4 (SOS3) and CIPK24 (SOS2). The CBL4 protein was shown to interact with CIPK24 and recruit it to the cytoplasmic membrane, where it activated the H+/Na+ (SOS1) antiporter to enhance salt tolerance [13]. This result indicated that the CBL-CIPK complex functions in regulating sodium ion homeostasis. Meanwhile, V-ATPase proteins were shown to directly interact with SOS2 to rescue the salt-defective phenotype of sos2 mutant plants [14]. In recent years, many CBLs and CIPKs have been shown to promote stress tolerance by balancing the intracellular ion concentration in plants. The sodium concentration was markedly lower in the Arabidopsis cbl10 mutant than in control plants, in both normal and salt-stress conditions [15]. The CBL10-overexpressing lines exhibited a K+-sensitive phenotype, like that of the akt1 mutant under low K+ conditions. The interaction between CBL10 and AKT1 was verified in bimolecular fluorescence complementation (BiFC), yeast two-hybrid (Y2H), and co-immunoprecipitation experiments, and was shown to impair AKT1-mediated inward potassium (K+) currents [16]. AtCIPK23 could directly interact with CBL1 to promote K+ uptake under low K+ conditions by phosphorylating and activating the K+ channel (AKT1) in Arabidopsis and rice [17, 18]. However, a recent study suggested that CBL1/CBL9 might interact with AKT1 independently of CIPK23 [19]. Moreover, in response to low-K+ conditions, CIPK9 was recruited to the tonoplast by CBL3 to regulate K+ homeostasis in Arabidopsis [20].

Over-expression and mutant analyses of CBLs and CIPKs have greatly enriched our understanding of their functions. The cbl2/cbl3 double mutant showed defects in growth and ion homeostasis, possibly as a result of disrupted vacuolar H+-ATPase activity [21]. In another study, CBL2 and CBL3 were shown to affect seed development and morphology [22]. The ectopic overexpression of ZmCIPK16 in the Arabidopsi sos2 mutant enhanced the expression of SOS1 and improved salt tolerance under saline conditions [12]. AtCIPK3 was induced by abscisic acid (ABA) and stress conditions and established a molecular connection between stress- and ABA-induced Ca2+ signals [23]. CBL10 and CIPK6 were shown to positively regulate both immunity- and disease-associated programmed cell death in tomato and tobacco [24]. The oscipk31 insertion mutant was hypersensitive to ABA and stresses (salt, mannitol, and glucose), and showed altered expression levels of several stress-responsive genes when it was exposed to these abiotic stresses during the seed germination and seedling stages [25]. In a recent study, a few AtCBL members were also found to target to another group of enzymes 5'-methylthioadenosine nucleosidases by yeast two-hybrid system, indicting an additional level of complexity on the CBL-mediated signaling networks [26]. However, with respect to the larger families of CBLs and CIPKs, few CBL-CIPK complexes have been functionally characterized, especially in bread wheat (Triticum aestivum L.), which is one of the most important food crops. Bread wheat has an allohexaploid genome with a size of approximately 17 GB, making it one of the largest and most complex plant genomes. In recent years, the genome sequencing of T. aestivum, Triticum urartu, and Aegilops tauschii has promoted research on bread wheat, especially studies on agronomically important gene families related to disease resistance, abiotic stress endurance, and grain quality [27–29].

In our previous work, we demonstrated that transgenic tobacco overexpressing wheat TaCIPK14 and TaCIPK29 showed increased tolerance to salinity and drought stress, respectively [30, 31]. In this study, seven TaCBL and 20 TaCIPK genes in wheat were identified and cloned by a genome-wide analysis combined with expressed sequence tag (EST) assembly. Here, we present the results of analyses of expression profiles, the specific interactions between TaCBLs and TaCIPKs, and the stress tolerance of transgenic Arabidopsis plants expressing TaCIPK24.

Results and discussion

Identification of TaCBL and TaCIPK gene families in wheat

The bread wheat genome is thought to have formed via fusion of three ancestral genomes, which are believed to be related to T. urartu, Aegilops speltoides, and A. tauschii. These fusion events are thought to have occurred several hundred thousand years ago [32]. We searched the wheat genome and found 24 CBL (7, 9, and 8 loci in sub-genomes A, B, and D, respectively) and 79 CIPK (21, 34, and 24 loci in sub-genomes A, B, and D, respectively) loci on wheat chromosomes (Table 1). The relative positions of these loci on chromosomes were identified on genetic maps (Additional file 1). The protein and DNA sequences of T. urartu and A. tauschii (the B sub-genome was not analyzed in this work) were retrieved from NCBI and all the CBL and CIPK nucleotide sequences from these three species (T. aestivum, T. urartu, and A. tauschii) were clustered against each other (Fig. 1). The closest matching genes were regarded as equivalent partners, as shown in Table 1. Ten and nine genes of T. urartu and A. tauschii, respectively, had no equivalents in T. aestivum, implying the existence of potential un-identified TaCBL/TaCIPK genes. Additionally, TaCIPK15-A, TaCIPK26-A, TaCIPK8-D, and TaCIPK22-D2 (wheat specific genes) had no equivalents in T. urartu and A. tauschii (Table 1). Considering that the physical maps of current wheat genome do not cover all the chromosome arms, it is difficult to know whether these potential un-identified TaCBL/TaCIPK genes are resulted from gene retention, duplication and whether these wheat specific genes are the results of the loss of CBL and CIPK genes after polyploidization. In the future, a more detailed wheat physical map may resolve this question.

Table 1 Identification of TaCBL and TaCIPK gene families in wheat genome
Full size table
Fig. 1
figure 1

Phylogenetic tree of CBL a and CIPK b proteins of T. aestivum (Ta), A. tauschii (Aet), T. urartu (Tu), A. thaliana (At) and O. sativa (Os). Multiple sequence alignment of CBL and CIPK proteins was performed using ClustalW. The wheat proteins were indicated with “rhombus” symbol. The genes belonging to A, B and D sub-genomes were marked with letters suffix and genes cloned in this study were colored with red

Full size image

Among the identified TaCBL/TaCIPK genes, the cDNAs of seven TaCBLs and 20 TaCIPKs were successfully PCR amplified from hexaploid bread wheat (T. aestivum) using gene-specific primers (Additional file 2) from a cDNA mixture prepared from wheat cv. Chinese Spring as the template. The genomic DNA and cDNA of all identified TaCBL and TaCIPK are provided in Additional file 3. To explore the evolutionary relationship of the CBL/CIPKs, two phylogenetic trees were generated based on all transcribed CBL/CIPKs from T. aestivum, T. urartu, A. tauschii, A. thaliana, and O. sativa. As expected, the genes from T. aestivum, T. urartu, and A. tauschii had the closest evolutionary relationships, and these genes had closer phylogenetic relationships with rice (a monocot) than with Arabidopsis (a dicot). Thus, the names assigned to the identified genes in wheat followed their homology to rice CBLs and CIPKs (Fig. 1). To analyze the structural characteristics of TaCBL and TaCIPK genes, the gene structures including exons and introns were mapped based on the wheat genome sequence (Additional file 4). Some genes' structures were partly characterized with exons/intros, because of lack adequate genome sequences. Only the structures covering the ORF were regarded as complete structures and used for further analysis. All the TaCBLs were intron-rich, with five to eight introns. Among the TaCIPKs, 26 TaCIPKs had no introns, and the other 15 TaCIPKs had 5–14 introns. For most genes, there were no obvious differences in gene structure among genomes A, B, and D (e.g., TaCIPK3, TaCIPK10 and TaCIPK23). Some genes (like TaCIPK31, TaCBL9) had a different intron length between the structures in different genome locations.

The CBL proteins containing EF-hand motifs are known as Ca2+-binding proteins, and show similarities to CML and CaM proteins. Each of the seven TaCBL proteins harbored four EF-hand motifs. Although these motifs in CBLs did not shared high similarity compared with CaM and CML proteins in other species (Arabidopsis, rice, C.elegans, yeast, zebrafish, mouse, and human), they were conserved among the CBLs in Arabidopsis, rice, and wheat (Additional file 5). As for the TaCIPK proteins, all 20 TaCIPKs (the genes were amplified by PCR) had domain structures similar to that of AtSOS2; that is, these 20 TaCIPKs harbored a protein kinase catalytic domain (PKC) and the NAF/FISL motif. Considering that CIPK proteins belong to the SnRK superfamily, we selected some representative SnRK superfamily proteins for comparisons of their structural features. As shown in Additional file 6, the PKC domains in SnRK family proteins were conserved and had some highly consistent sites, especially in the activation loop, while the NAF/ FISL motif was unique to CIPK proteins.

Identification and validation of TaCBL–TaCIPK interactions

We analyzed the physical interactions between TaCBL and TaCIPK proteins in wheat using the Y2H method. As shown in Table 2 (Additional file 7), the interactions between TaCBLs and TaCIPKs showed different strengths and specificities. TaCBL1 strongly interacted with five CIPKs (TaCIPK3, 5, 14, 15, and 25). TaCBL2 strongly interacted with 11 CIPKs (TaCIPK3, 5, 11, 14, 15, 17, 21, 25, 26, 27, and 32). TaCBL3 showed strong interactions with 10 TaCIPKs (TaCIPK3, 8, 11, 14, 15, 17, 21, 26, 27, and 31) out of 19 TaCIPKs. TaCBL4, the ortholog of AtSOS3 (CBL4), strongly interacted with six TaCIPKs (TaCIPK3, 5, 14, 15, 26, and 31). TaCBL6 strongly interacted with five TaCIPKs (TaCIPK5, 11, 15, 21, and 27), and TaCBL7 interacted with seven TaCIPKs (TaCIPK5, 14, 15, 21, 22, 26, and 31). TaCBL9 strongly interacted with only two TaCIPK proteins (TaCIPK11 and TaCIPK31). To view the interactions from a different perspective, we found that TaCIPK7, 10, 19, 22, 24, 28, and 29 did not interact with any of the seven TaCBL proteins assayed, implying that these CIPK proteins might perceive signals from other unidentified CBL proteins in wheat. We selected nine of the interactions (CBL1–CIPK15, CBL2–CIPK31, CBL3–CIPK15, CBL3–CIPK27, CBL4–CIPK15, CBL4–CIPK31, CIPK6–CIPK15, CBL7–CIPK27, and CBL9–CIPK31) detected between CBL and CIPK proteins using the Y2H assay for further analyses in planta using the BiFC method (Additional file 8). For all of the assayed TaCBL–TaCIPK interaction complexes, yellow fluorescence signals were observed when TaCBLs and TaCIPKs were co-expressed in epidermal cells of tobacco leaves. Notably, it seemed that the observed preferential interactions were inconsistent with the close phylogenetic relationships of the TaCIPKs, namely, the closely phylogenetic related CIPK pairs do not display same interaction profile. For example, TaCIPK14 and TaCIPK15, which had 81 % similarity, showed similar interactions with five TaCBLs (TaCBL1, 2, 3, 4 and 7) but different interactions with TaCBL6 and 9. Compared with TaCIPK14/15, other TaCIPKs showed lower amino acid sequence similarity and more diverse interaction specificity. Thus, sequence similarity or phylogenetic relationships are not sufficient to predict the result of CBL-CIPK interactions in the Y2H assay.

Table 2 Interactions of TaCIPKs with TaCBLs in yeast two-hybrid assay
Full size table

Previous studies have shown that the NAF/FISL motif located in the C-terminal regulatory domain and the kinase activity of the PKC domain in the N-terminal region of CIPKs are necessary and sufficient for mediating interactions with CBLs [5, 6, 33]. The roles of the redundant parts (those other than the NAF/FISL motif) in the C-terminal region of CIPKs have not yet been fully studied. Here, the TaCIPK11 protein, which interacted with CBL2, 3, 6, and 9, contained a large C-terminal region (amino acids 311–507) that was not present in the other CIPKs. This led us to speculate about the role of this large fragment in regulating the specific interactions of this protein. Analyses of a series of C-terminal deletion mutants of TaCIPK11 (CIPK11–M1, M2, M3, and M4; M1–M3 containing the complete NAF/FISL motif, and M4 with a truncated NAF/FISL motif) showed that there were diverse interaction patterns between TaCIPK11 mutants and TaCBLs (Fig. 2a). CBL3, 6, and 9 could not interact with any of the TaCIPK11 deletion mutants, while TaCBL1/TaCBL4 physically interacted with some or all of the mutants. TaCBL2 lost its specific interaction with the minimal mutant (CIPK11–M4). TaCBL7 did not interact with TaCIPK11 or any of its mutants. It seems that TaCBL3, 6, and 9 require the last 86 amino acid for interaction with TaCIPK11 and also inhibits the binding of TaCBL1 and TaCBL4. These changes in the patterns of interaction may result from changes in protein structure after the deletions. In terms of this deletion assay, we supposed that the interaction between CBL and CIPK followed a “concave–convex” model as shown in Fig. 2d, similar to the model for the structures of AtSOS3-AtSOS2 and AtCBL2-AtCIPK14 complexes proposed by Sánchez et al. [34]. In this model, the spatial structures of CBLs/CIPKs have an important role for their interactions, and a consummate physical interaction depends on whether the NAF/FISL motif properly reach the CBL cavity. It is believed that the diversified spatial structures resulted from divergent C-terminal region of CIPKs explain the molecular basis of the selectivity of certain CBLs towards particular CIPKs. This led to another intriguing question; that is, do the nonsynonymous sequence variations of alleles from different wheat sub-genomes result in aberrant spatial structures that affect the patterns of interaction? To answer this question, we amplified TaCIPK17-A, which encodes a protein with many amino acid variations in the C-terminal regulatory region compared with the homologous TaCIPK17-B protein (Fig. 2b). Interestingly, TaCIPK17-A not only interacted with TaCBL2 and TaCBL3, but also with TaCBL6 in Y2H experiments (Fig. 2c). Despite two mutant sites (G23 to A23, G48 to D48) were observed at PKC domain, it is possible that several amino acid variations in the C-terminal regulatory region would confer different patterns of interaction. Such changes in interactions may render CIPKs capable of perceiving signals from other CBLs. These results indicate that the TaCBL-TaCIPK signaling pathway in wheat is very complex because of its allohexaploid genome.

Fig. 2
figure 2

Deletions or mutations in C-terminal ends of TaCIPKs dramatically changes its ability to interact with TaCBLs. a series deletions in C-terminal ends of TaCIPK11 leads to changing interacting patterns with TaCBLs. The boxes indicate the PKC and NAF domains, and numbers beside the lines represent the beginning and the ending positions of TaCIPK11 protein fragments. b, C-terminal ends alignment between TaCIPK17-A and TaCIPK17-B. Residues with grey background indicates mutant amino acid sites. c, interactions of TaCIPK17-A and TaCIPK17-B with TaCBLs. Yeast was monitored on the selection medium and indicated as growth (+) and no growth (−). d, Schematic representation of the “concave-convex” model

Full size image

Expression patterns of TaCBL and TaCIPK genes at different developmental stages in wheat

To investigate the spatial expression patterns of TaCBLs and TaCIPKs in wheat, we detected their transcriptional levels using RT-PCR in 10 representative tissues, i.e., coleoptile, root (seedling and flowering stage), stem (seedling and flowering stage), leaf (seedling and flowering stage), flag leaf (flowering stage), pistil, and stamen (Fig. 3d). The results showed that the majority of TaCBLs and TaCIPKs were expressed at different levels in all tissues tested. Most TaCBL genes were constitutively expressed in all organs and at all developmental stages, although the transcript levels of certain genes were very low in some tissues. Some CIPK genes were expressed abundantly in certain tissues. For example, TaCIPK22 was specifically expressed in the root at the seedling stage, indicating that it may play an important role in this organ. Moreover, the gene transcript levels differed between vegetative (seedling) and flowering stages. There were higher transcript levels of TaCBL3 in seedlings than in plants at the flowering stage (roots, stems, and leaves). The flag leaf, as the last leaf of wheat, delivers the largest proportion of photosynthate to fill wheat grains [35]. Among the genes expressed in the flag leaf, TaCBL7 showed lower transcript levels in leaves of seedlings and plants at the flowering stage, and higher transcript levels in the flag leaf, implying that it might play roles in regulating photosynthesis and/or metabolism in the flag leaf. To study the transcription levels of TaCBLs and TaCIPKs during the whole life cycle of wheat, we analyzed microarray data that were obtained from tissues at various developmental stages; these data were obtained from publically accessible databases. The genes formed two groups according to their expression patterns (Fig. 3a). Group I consisted of the genes with high transcript abundance in nearly all tissues (TaCBL1, 2, and 6 and TaCIPK2, 8, 9, 15, 16, 23, 25, and 27). Group II genes consisted of 4 TaCBLs and 15 TaCIPKs that showed different expression patterns in different tissues and stages. Some of them showed low transcript levels in all tissues. For most of the TaCBLs/TaCIPKs, their expression profiles in the microarray data matched to the RT-PCR results, but a few genes showed inconsistencies. For example, high transcript levels of TaCBL4 were detected in the stamen in RT-PCR analyses, but low expression levels in anthers (before anthesis) were indicated in the microarray data. This inconsistency may have resulted from differences in the growth stages of plants between experiments, or from the different methods used.

Fig. 3
figure 3

Expression profile analyses of TaCBL and TaCIPK in wheat tissues with microarray data a, b and c) and RT-PCR (d and e). a gene expression analyses in wheat whole life cycle using public microarray data. b, gene expression analyses during seeds germination process after immersed with water. c, comparative transcriptomic analyses between dormant and after-ripened seeds in both imbibed and ABA application states. The value (b and c) represents relative transcript levels compared to control (before immersion; 0 h). The up panels in a b and c represent log2 transformed values. d, gene expression analyses during wheat development using RT-PCR method. “S” and “F” represent seedlings and flowering stages, respectively. e, the induced genes in seeds germination (b) were confirmed by RT-PCR

Full size image

To further clarify the detailed roles of TaCIPK and TaCBL in seed germination, we analyzed public microarray data representing the transcription patterns during seed germination. As shown in Fig. 3b, TaCIPK6, 12, 14, 15, 16, 23, 25, and 27, and TaCBL2 were up-regulated (log > 1, P < 0.05) after 12 h of imbibition, and then down-regulated (log < 1, P < 0.05). TaCIPK12, TaCIPK31, TaCIPK32, and TaCBL3 were down-regulated during imbibition (log > 1, P < 0.05). Semi-quantitative RT-PCR analyses were conducted to confirm the TaCBL and TaCIPK transcription patterns at five stages of germination (0, 12, 24, 36, and 48 h). All of the selected up-regulated genes were induced during seed germination (the EST of TaCIPK12 was unavailable, and the probe sequence of CIPK12 was not amplified), but the timing of induction differed (Fig. 3e). TaCIPK16, TaCIPK25, and TaCIPK27 showed peak transcript levels after 12 h of imbibition, while TaCIPK14 and TaCIPK23 showed peak transcript levels at 48 h and lower transcript levels at other time points. These transcriptional patterns indicated that the genes played roles at different stages of germination. High transcript levels of TaCIPK15 were detected only at 36 h, when the radicle of seeds emerged and the coleoptile became visible. There were high transcript levels of TaCIPK16 at the initial stage of germination (12 h). For down-regulated genes (TaCBL3, TaCIPK31 and TaCIPK32) (Fig. 3b), our data showed that these genes were actually up-regulated or showed no significantly changes in transcription. Moreover, despite some up-regulated genes were all validated in RT-PCR and microarray data, they displayed peak transcript levels at different time point. For example, TaCIPK23 reached peak levels at 48 h in RT-PCR analysis, while it was up-regulated at 12 h followed with decreasing transcript levels in microarray analysis. The inconsistencies in the expression levels of genes may have been because the data were obtained using different methods, from different sub-species, or from seeds at slightly different stages of germination. We analyzed microarray data from another experiment, which focused on seed dormancy and after-ripening in bread wheat [36], and found that four genes (TaCIPK7, 10, 17, and 21) corresponded to germination and that TaCIPK31 was up-regulated gene during germination (Fig. 3c). Interestingly, TaCIPK32 was down-regulated in dormant seeds, but up-regulated in after-ripened seeds. These data provided the preliminary information about the roles of TaCBL–TaCIPK in seed germination.

Expression profiles of TaCBL and TaCIPK in response to abscisic acid and abiotic stresses

There is considerable evidence that CBL and CIPK genes play important roles in the responses to phytohormones and abiotic stresses [10, 11, 37]. Among the stress hormones, ABA is a well-known signaling molecule in biotic or abiotic stress responses [38]. Cold, drought, and salinity are crucial factors affecting wheat growth and crop yields, and these stresses often cause oxidative stress.

To investigate the expression profiles of TaCBL and TaCIPK genes under stress conditions, we conducted real-time quantitative RT-PCR analyses. We analyzed the transcript levels of two TaCBLs and five TaCIPKs in wheat seedling roots and leaves under salt (NaCl), H2O2, drought (PEG), and cold (4 °C) stress treatments, and in response to ABA application (Fig. 4). The transcript levels of all tested TaCBL and TaCIPK genes changed in response to many of the treatments. The genes that did not show significant changes in transcript levels (significant criterion: log > 2, up-regulated; log < 0.5, down-regulated) were TaCBL9 in the leaf (ABA/PEG treatments) and in the root (cold treatment), TaCBL7 in the root (NaCl treatment), TaCIPK15 in the root (H2O2 treatment), and TaCIPK31 in the leaf (ABA treatment). Among the analyzed genes, TaCIPK31 exhibited remarkable ABA-induced up-regulation in roots (350-fold increase compared with that in the control at 3 h), but not in the leaves. Under salt stress, TaCIPK24 was up-regulated in roots and leaves. Therefore, it was selected for further functional characterization. The expression patterns of the analyzed TaCBLs and TaCIPKs were not completely consistent between seedling roots and leaves. For example, TaCBL4 was up-regulated in roots and down-regulated in leaves in response to ABA application. Several genes responded to the same stress; for example, TaCBL9, TCIPK7, 15, 24, and 32 were induced by H2O2. Among the five treatments, cold treatment resulted in the largest number of up-regulated genes, and none of the assayed genes were down-regulated by cold in roots or leaves. In brief, the expression patterns of the assayed genes differed among different treatments and tissues, indicating that various TaCBLs or TaCIPKs may participate in the signaling response to the same stress, and that a single TaCBL or TaCIPK might function in multiple stress responses.

Fig. 4
figure 4

Expression analyses of TaCBL and TaCIPK in leaves and roots responding to various treatments, including ABA (10 μM), NaCl (200 mM), H2O2(10 mM), PEG (20 %) and cold (4 °C). Five time points (0 h, 1 h, 3 h, 6 h, 9 h, 12 h and 24 h) were used to detect the gene expressions. The vertical axis represents gene relative expression levels. Data is the mean of three biological replicates ± S.E

Full size image

In previous studies, monocot and dicot species were shown to have divergent gene expression patterns in response to abiotic stresses or ABA application [11, 39]. Thus, we conducted a comparative expression analysis of CBLs/CIPKs among wheat and two other important monocot crops (rice and maize). Microarray datasets of rice CBL/CIPK expression patterns in response to abiotic stresses were collected by Zhang et al. [11], and a dataset of maize CIPK expression patterns was obtained by semi-quantitative-PCR by Chen et al. [10]. The comparative analyses showed that several orthologous genes had similar transcription patterns. For example, the transcript level of CIPK32 increased in leaves and decreased in roots under salinity treatment in wheat and maize. However, some orthologous genes exhibited divergent expression patterns. For example, TaCBL4 showed increased transcript levels in roots and leaves under cold stress, while the transcript levels of OsCBL4 decreased under cold stress. Thus, there are both conserved and divergent expression models for orthologous genes among wheat, maize, and rice.

Next, we focused on stress-responsive cis-elements to decode the CBL/CIPK expression models. These elements included the ABA responsive element (ABRE), the dehydration-responsive element (DRE)/C-repeat, and the low-temperature-responsive element (LTRE) [40–43]. Sequence analyses of 1-kbp sequences upstream from the start codon of the stress-inducible genes (TaCBL4, TaCBL9, TaCIPK7, TaCIPK15, and TaCIPK24) revealed that these genes contained putative ABRE, DRE, and LTRE elements (Additional file 9). These elements at least partly explained the stress-inducible expression patterns of these genes. However, even though TaCIPK31 and TaCIPK32 were induced by ABA, PEG, and cold, they contained only the ABRE element in their promoter regions (Additional file 9). This result is not so surprising, since we only searched for three cis-elements in this study, and there may be other unidentified elements that are important for stress-responsive gene expression. Generally, the putative stress-responsive cis-elements in promoter regions are crucial for stress-responsive expression, and activation of the cis-elements depends on the tissue, the development stage, and/or the genetic background. Therefore, there are many explanations for the diversified expression patterns of orthologous genes among different species.

Over-expression of TaCIPK24 improves salt tolerance in Arabidopsis

Previous studies showed that AtCIPK24 (SOS2) and BnaCIPK24 (homolog of TaCIPK24) participated in the salt stress response in transgenic Arabidopsis [11, 44, 45]. In our expression analysis, TaCIPK24 was induced in roots and leaves under salt stress (Fig. 4). Therefore, we generated transgenic Arabidopsis plants over-expressing TaCIPK24 (or harboring pBI121 as the empty-vector control). In total, 28 transgenic lines were obtained. Three independent T3 homozygous lines (designated as OE-2, OE-7, and OE-9) were selected for further analyses. The transcript levels of TaCIPK24 were detected by RT-PCR (Fig. 5a). In plants grown on Murashige and Skoog (MS) medium, root growth did not differ significantly among the transformants, the empty-vector control, and wild type (Fig. 5b and d). However, the 35S:TaCIPK24 transgenic plants showed greater root growth than those of controls after treatment with 150 mM NaCl (Fig. 5c and e; p < 0.05). There were no significant differences in the phenotypes of aerial parts between transgenic lines and controls at the seedling stage under normal and salt-stress conditions. Therefore, our data indicated that TaCIPK24 promoted root elongation in transgenic plants under salt stress.

Fig. 5
figure 5

Overexpression of TaCIPK24 in Arabidopsis. a TaCIPK24 expression levels in selected transgenic lines, vector control and wild type. b, c the plants grown on control (MS medium) and stress condition (MS + 150 mM NaCl). d, e statistical analysis of the changing root length between MS control and salt stress condition (p < 0.05)

Full size image

To further confirm the salt tolerance of TaCIPK24, seedlings of 10-day-old, soil-grown, wild-type and transgenic lines were irrigated with 200 mM NaCl for 25 d. As shown in Fig. 6a, the transgenic lines grew better than did the control plants. Furthermore, Na+ accumulated to lower levels in leaves of transgenic lines than in leaves of the empty-vector control and wild type. The K+ contents were not significantly different between transgenic lines and the empty-vector control/wild type (Fig. 6b), indicating that the improved salt tolerance may involve Na+ efflux. In addition, excess Na+ would generate reactive oxygen species (ROS), such as peroxide, superoxide, and hydrogen peroxide (H2O2), causing oxidative stress in plants. The accumulation of H2O2 was reduced in transgenic lines compared with that in the control plants (Fig. 6c). This result indicated that decreased Na+ contents led to lower ROS production (at least H2O2) and thus, the plants were able to grow well under salt stress conditions. Antioxidant enzymes are crucial for ROS scavenging and play roles in the salt stress response in plants. Therefore, we measured the activities of three important antioxidant enzymes (catalase, CAT; peroxidase, POD; and superoxide dismutase, SOD) in transgenic plants and controls under salt stress and normal conditions. The activities of CAT, POD, and SOD were higher in transgenic lines than in wild-type and empty-vector controls (Fig. 6c). These results showed that the ectopic expression of TaCIPK24 enhanced salt tolerance through facilitating Na+ efflux and ROS scavenging.

Fig. 6
figure 6

Phenotype of TaCIPK24 over-expression lines in Arabidopsis. a phenotypic comparison of transgenic (OE-4, OE-7, and OE-9) and control (WT) plants under 200 mM salt stress and water control. b Na+/K+ concentration in seedlings under stress conditions. c contents of H2O2, oxidants malondialdehyde (MDA), enzymatic activities of peroxidase (POD), catalase (CAT) and superoxide dismutase (SOD) in control and transgenic plants

Full size image

Conclusions

CBL and CIPK proteins play important roles in the Ca2+ signaling pathway and affect plant development, as well as participating in biotic and abiotic stress responses. CBLs and CIPKs have been identified and functionally characterized in Arabidopsis, rice, maize, and canola [10–12, 46, 47], but less is known about the CBLs and CIPKs in wheat. In this study, we identified seven TaCBL and 29 TaCIPK genes from wheat, and cloned cDNAs of seven TaCBLs and 20 TaCIPKs. Analyses of TaCBL–TaCIPK identified 24 TaCBL and 79 TaCIPK loci in the wheat genome, indicating that there were large numbers of these gene loci in the hexaploid wheat. Blast analyses suggested that four TaCBL and 55 TaCIPK loci were derived from the T. urartu and A. tauschii genomes. We studied the preferential interactions between TaCBL and TaCIPK proteins and found 44 CBL–CIPK interaction complexes. The proteins TaCIPK7, 10, 19, 22, 24, 28, and 29 did not interact with any of the seven TaCBL proteins assayed in these experiments. Expression analyses revealed that the majority of TaCBLs and TaCIPKs were expressed at varying levels in all tested tissues, although the transcript levels of some of these genes were very low in some tissues. We conducted RT-PCR analyses to monitor the transcript levels of seven TaCBL and 17 TaCIPK genes under abiotic stress and hormone treatments. Most were distinctly regulated by at least one treatment. Among them, TaCIPK24, the ortholog of Arabidopsis SOS2, was up-regulated by salt stress in root and leaf tissues. We generated transgenic Arabidopsis lines over-expressing TaCIPK24, and all of them showed enhanced salt tolerance. Overall, this study lays a solid foundation for further exploration of the roles of TaCBL and TaCIPK proteins in abiotic stress responses, and enhances our understanding the functions of the wheat TaCBL and TaCIPK gene families. These results also provide information that will be useful for the genetic manipulation of wheat to improve stress tolerance.

Methods

Identification of TaCBL and TaCIPK genes

The reported CBL and CIPK protein sequences from rice and Arabidopsis were used to search the NCBI UniGene database. All the selected EST sequences were collected to form a local database and were assembled into longer cDNA sequences. To further validate the integrity of TaCBL and TaCIPK gene sequences, these sequences were used to search the T. aestivum cv. Chinese Spring draft genome using the BLASTn method at Gramene (http://www.gramene.org/) and IWGSC (http://www.wheatgenome.org/). Rigorous criteria (E < 10−5, identity > 90 %) were used to ensure the reliability of the nucleotide sequences.

The gene-specific primers used for gene cloning were designed using Primer Premier 6 software (http://www.premierbiosoft.com/). TaCBLs and TaCIPKs were amplified by RT-PCR from cDNA mixtures of prepared from wheat (cv. Chinese Spring) that included cDNAs from the coleoptile, root, stem, leaf, flag, leaf, pistil, anthers, and from seedlings treated with polyethylene glycol, NaCl, cold, and ABA.

The structures of TaCBL and TaCIPK genes were illustrated using the Gene Structure Display Server (GSDS, http://gsds.cbi.pku.edu.cn/). Briefly, the cDNA sequences of T. aestivum, T. urartu, and A. tauschii were used to search the related genomic DNA sequences of T. aestivum following the instruction of GSDS. Multiple sequence alignments were executed using Clustal W (http://www.clustal.org) with default parameters. An un-rooted neighbor-joining phylogenetic tree was constructed using MEGA 5.2.2 software (http://www.megasoftware.net). Genome-based sequences were used to confirm the affiliations of genes in all 21 chromosomes. Chromosomal locations were identified on genetic maps with Genome Zipper V5 at IWGSC and visualized by MapDraw. To analyze putative cis-elements in the promoters of TaCBL/TaCIPK genes, 1-kbp regions upstream of the CDS were extracted from wheat genomic sequences and subjected to a cis-element search using PLACE (http://www.dna.affrc.go.jp/PLACE/signalup.html).

Gene expression analyses

Wheat (T. aestivum L. cv. Chinese Spring) seeds were surface sterilized and then germinated in distilled water in a greenhouse at 25 °C. The germinated seeds were collected at five different time points (0 h, 12 h, 24 h, 36 h, and 48 h) to analyze gene expression during seed germination. After 2 weeks, young seedlings were subjected to ABA (10 μM), H2O2 (10 mM), cold (4 °C), NaCl (200 mM), and PEG (20 % v/v) treatments for 24 h. Samples were collected at 0 h, 1 h, 3 h, 6 h, 9 h, 12 h, and 24 h after the stress treatments for analyses. Leaf samples for analyses of CAT, POD, and SOD activities were collected at 25 d after the stress treatments. All the samples were harvested, frozen in liquid nitrogen, and stored at −80 °C until use. Total RNA was isolated from frozen tissue using a Plant total RNA Extraction kit (ZOMANBIO, http://www.zomanbio.com). The reverse-transcription reactions were performed using a FastQuant RT kit (TIANGEN, http://www.tiangen.com/).

The primers used for expression analyses are listed in Additional file 2. The microarray datasets were processed using R 3.0.1 with the “Affy” package. Expression intensities were first normalized using the arithmetic RMA normalization method and then used for expression analyses. TaCBL and TaCIPK DNA sequences were used to probe the Affymetrix microarray datasets with blastn in PLEXdb (http://www.plexdb.org/index.php). The microarray expression data and gene probes used for these analyses are listed in Additional file 10.

Yeast two-hybrid (Y2H) and bimolecular fluorescence complementation (BiFC) assays

The Y2H assays were conducted using the MatchMaker yeast two-hybrid system (www.clontech.com). The ORF regions of TaCBL and TaCIPK were respectively sub-cloned into pGBKT7 and pGADT7 vectors and co-transformed into the Y187 yeast strain. The transformants were grown on double-dropout medium (DDO: SD/-Trp/-Leu) and selected on triple-dropout medium (TDO: SD/-Trp/-Leu/-His) containing 10 mM 3-amino-1, 2, 4-triazole (3-AT). For BiFC assays, the ORF regions of TaCBL and TaCIPK were sub-cloned into 35S-SPYCE and 35S-SPYNE vectors, respectively. After confirmation by sequencing, these constructs were separately transformed into Agrobacterium tumefaciens GV3101, then into tobacco leaves by Agrobacterium infiltration. Freshly transformed Agrobacterium cell cultures were re-suspended in suspension medium (10 mM MES-KOH (pH 5.6), 10 mM MgCl2, and 0.1 mM acetosyringone), adjusted to an OD600 of 0.5-0.8, and left at room temperature for 3 h before infiltration into tobacco leaves. Infiltrated leaf discs were collected 3–5 d later for observation under a confocal microscope. The TaCBLs and TaCIPKs primers used for vector construction are listed in Additional file 2: Table S1.

Arabidopsis transformation and treatments

The full-length coding sequence of TaCIPK24 was inserted into the pBI121 overexpression vector. Arabidopsis plants were transformed using the floral-dip method with A. tumefaciens strain EHA105 [48]. The assayed plants (transgenic Arabidopsis lines and wild type) were grown on MS medium until they reached the four-leaf stage, then transferred onto MS medium containing NaCl (150 mM) for salt stress analysis. The seedling root lengths were measured 7 days later. For salt tolerance analysis, 10-day-old soil-grown plants were irrigated with 200 mM NaCl at 5 d intervals for 25 d.

Ion accumulation, H2O2 content, and enzyme activity measurements

To determine ion contents, samples were analyzed by atomic absorption spectrometry. Briefly, leaves were dried at 80 °C for 3 d, mixed with 30 % (v/v) H2O2, heated for 15 min at 180 °C, and then digested with concentrated HNO3 overnight. The samples were then analyzed using an atomic absorption spectrometer (AA-6300, Shimadzu Corporation). The H2O2 content and activities of CAT, POD, and SOD were measured by spectrophotometric methods using commercial detection kits (A064, A007, A084, and A001, Nanjing Jiancheng, China).

Availability of supporting data

All the supporting data are included as additional files.