The Arabidopsis (Arabidopsis thaliana) cDNA expression libraries (CD4-10 and CD4-22) were obtained from the Arabidopsis Biological Resource Center (ABRC) and used in the yeast two-hybrid screening, which was performed basically according to Oh et al. (2008). For yeast two-hybrid assays, full-length open reading frames (ORFs) for genes of interest were amplified from total RNA of Arabidopsis seedlings by reverse transcription (RT)-polymerase chain reaction (PCR) using gene-specific primer sets and subsequently cloned into either the activation domain (pGAD.GH) or the DNA binding domain (pGBT9.BS) vectors. These constructs were then introduced into yeast strain Y190 by the lithium acetate method (Schiestl and Gietz, 1989). Yeast transformants carrying both plasmids were selected on the synthetic medium lacking Leu and Trp (SC-LW) for 3–5 days at 30°C. The yeast cells were subsequently streaked on the synthetic complete medium lacking His, Leu and Trp (SC-HLW) plate to determine the expression of the HIS3 nutritional reporter gene.

Arabidopsis [ecotype Columbia (Col-0)] plants were grown in a growth chamber at 23°C under long-day (LD) conditions (16-h-light/8-h-dark cycle) at a light intensity of 120 μmol m⁻² s⁻¹. The total RNA was isolated from a variety of tissues using the plant RNA mini kit (Qiagen, Germany) according to the manufacturer's instructions. RNA quality was determined with a Nanodrop ND-2000 spectrophotometer (Nanodrop Technologies, USA), and only high quality RNA samples (A₂₆₀/A₂₃₀ > 2.0 and A₂₆₀/A₂₈₀ > 1.8) were used for subsequent experiments. To remove possible genomic DNA contamination, RNA samples were treated with DNaseI for 60 min at 37°C. RNA (1 μg) was used for complementary DNA (cDNA) synthesis, in accordance with the manufacturer's instructions (New England Biolabs). Quantitative real-time reverse transcription PCR (qRT-PCR) was carried out using a Rotor-Gene Real-Time Centrifugal DNA Amplification system (Corbett Research). PCR reactions were performed using the QuantiTect SYBR Green PCR Master MIX following the manufacturer's instructions (Qiagen). Data analysis was performed with Rotor-Gene software and relative amounts of mRNA were determined based on the comparative threshold cycle method. The housekeeping gene Actin2 was used as a reference gene to normalize the relative expression of target genes, according to “The eleven golden rules for quantitative RT-PCR” (Udvardi et al., 2008). All qRT–PCR experiments were carried out in two biological replicates (independently harvested samples) with three technical replicates. Oligonucleotide sequences used for the expression analysis are provided below: CBL10-F (5′-TTCATTGAGCGAGAAGAGGTGCA-3′), CBL10-R (5′- GGAATGCTGTCGTCACAT CCTTT-3′), TOC34-F (5′- TGCTGCAGTTAGTACTTTCCAGTCT-3′), TOC34-R (5′- TAT AGTCATGTTGAGGAGAAATCGT-3′), Actin2-F (5′- TGAGGATATTCAGCCACTTGT CTG-3′), and Actin2-R (5′- GATTGGATACTTCAG AGTGAGGAT-3′).

For subcellular localization and BiFC analyses in Arabidopsis protoplasts, different plasmids were transformed into Arabidopsis mesophyll cells by a PEG-mediated transfection procedure (Yoo et al., 2007). After the transfected Arabidopsis protoplasts were incubated at 23°C for 18 h, fluorescence signals were analyzed with a confocal laser scanning microscope (LSM 510 META, Carl Zeiss). For BiFC analysis in tobacco (Nicotiana benthamiana) leaves, the Agrobacterium tumefaciens strain GV3101 harboring the various combinations of constructs was infiltrated into the abaxial sides of 3-week-old tobacco plants. Subsequently, epidermal cells of infiltrated tobacco leaves were examined for fluorescence using the confocal laser scanning microscope. The detailed procedure has been previously reported (Ok et al., 2015).

GST-fusion proteins such as GST-CBL4, GST-CBL10-cMyc, and GST-TOC34 were purified basically according to the protocols described earlier (Ok et al., 2015). Briefly, E. coli BL21 cells possessing a GST fusion construct were cultured at 37°C overnight and were subcultured until the OD₆₀₀ reached 0.5–0.6. Following 3-h induction with 0.3 mM Isopropyl-β-D-thiogalactopyranoside at 20°C, the cells were lysed in ice-cold buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 1 mM PMSF, 5 mM DTT, 5 mM EDTA, and 1 mM EGTA). Glutathione-Sepharose 4B beads were used to retrieve the GST fusion protein. Ice-cold washing buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl) was used to wash the beads. Protein concentration was determined according to Bradford (1976).

Pull-down assay was performed as described previously (Ok et al., 2015). Briefly, GST fusion proteins attached to the glutathione-Sepharose 4B beads were incubated at 4°C with prey proteins lacking the GST protein in the binding buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 0.05% Tween 20, and 1 mM PMSF) supplemented with either 0.2 mM CaCl₂ or 1 mM EGTA. Pull-down samples were resolved by SDS-PAGE, transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, USA), and detected by immunoblot analysis as described previously (Shi et al., 1999).

To investigate the GTPase activity of TOC34, we performed GTPase assay using a Quantichrom™ATPase/GTPase assay kit (BioAssay Systems, USA). The GTP hydrolysis activity of the TOC34 protein purified from E. coli was determined according to manufacturer's instructions. Briefly, the assays were initiated by adding 10 μL of TOC34 (2.5 μg) into 30 μL reaction samples containing 20 mM Tris (pH 7.0), 40 mM NaCl, 4 mM MgAc₂, 0.5 mM EDTA, 4 mM GTP, and 5 mM CaCl₂ or 2 mM EGTA. Depending on reaction conditions, either 2.1 μg of CBL10 or 1.8 μg of CBL4 was also included. After incubation at 27°C for 30 min, free phosphates generated were quantified according to the manufacturer's protocols using a VersaMax ELISA microplate reader (Molecular Devices, USA).

The following plasmids were constructed as described previously (Halfter et al., 2000; Ok et al., 2015); pGBT·CBL1, pGBT·CBL2, pGBT·CBL3, pGBT·CBL4, pGBT·CBL5, pGBT·CBL6, pGBT·CBL7, pGBT·CBL8, pGBT·CBL9, pGBT·CBL10, pGEX·CBL4, pUC-SPYNE-bZIP63, and pUC-SPYCE-bZIP63. To make pGAD·CBL10, the coding region of the CBL10 gene was PCR amplified with a primer set of CBL10-1/CBL10-2 using pGBT·CBL10 as template. Following digestion with BamHl/SalI, the PCR product was subsequently ligated into the pGAD.GH vector. To create the pGBT·TOC34 and pGAD·TOC34 plasmids, the coding region of the TOC34 cDNA was first amplified with primers TOC34-1 and TOC34-2. Then, the resulting PCR product was digested with BamHI/SalI and ligated into pGAD.GH and pGBT9.BS, respectively. The plasmid pGAD·TOC34-N was constructed by cloning the PCR product amplified with TOC34-2 and TOC34-7 primers into the BamHI/SalI sites of the pGAD.GH plasmid. Similarly, the pGAD·TOC34-C1 and pGAD·TOC34-C2 plasmids were generated using primer sets TOC34-8/TOC34-1 and TOC34-9/TOC34-1, respectively.

The plasmid pGEX·TOC34 was constructed by cloning the PCR product amplified with TOC34-5/TOC34-6 primer set into the BamHI/SalI sites of the pGEX.4T-3 plasmid (GE Healthcare Life Sciences). For construction of the pGEX·CBL10 and pGEX·CBL10-c-myc plasmids, each of the PCR products amplified with primer sets (CBL10-3/CBL10-4 and CBL10-3/cMYC-1) were cloned into the pGEX-4T-3 vector, which was digested with BamHI/SalI and BamHI/NotI, respectively. For creation of the TOC34-GFP chimeric construct (pMD·TOC34), primers TOC34-3 and TOC34-4 were used to PCR amplify the TOC34 coding region without a stop codon. Following digestion with XbaI/BamHI, the PCR product was cloned into the pMD1 binary vector that contains a GFP reporter gene (Sheen et al., 1995). For BiFC assays in Arabidopsis protoplasts, the pUC-SPYNE-TOC34 (pUC-TOC34-YFP^N) and pUC-SPYCE-CBL10 (pUC-CBL10-YFP^C) plasmids were created by cloning each coding region of TOC34 and CBL10, which were PCR amplified with TOC34-6/TOC34-7 and CBL10-3/CBL10-4 primer sets, into the BamHI/SalI sites of the pUC-SPYNE and pUC-SPYCE vectors, respectively. In addition, the TOC34:cMyc-YFP^N and CBL10:HA-YFP^C regions were PCR amplified from the pUC-TOC34- YFP^N and pUC-CBL10- YFP^C constructs with the primer sets (TOC34-3/YFPN-1 and CBL10-5/YFPC-1, respectively) and then each of them were cloned into the XbaI/BamHI sites of the pCAM35S and pBI121ΔGUS binary vectors, thereby creating TOC34-YFP^N and CBL10-YFP^C. These two constructs were used to perform BiFC assays in tobacco plants. All the PCRs were carried out using Pfu DNA polymerase (Stratagene) to enhance fidelity. All the constructs above were verified by DNA sequencing.

Primers used in this study were listed below, with restriction enzyme sites underlined. Three additional bases, which were chosen randomly by considering their effect on melting temperature and on dimer and stem-loop formation, were included at the 5′ end of the primers for efficient digestion by restriction enzymes:

To identify a novel interaction partner(s) of CBL10, we extensively screened the Arabidopsis cDNA expression libraries CD4-10 and CD4-22 obtained from the Arabidopsis Biological Resource Center via a yeast two-hybrid system using CBL10 as a bait. The CBL10 bait (pGBT·CBL10 or BD·CBL10) was made by cloning the open reading frame (ORF) of CBL10 cDNA into the GAL4 DNA-binding domain vector pGBT9.BS (BD). We obtained a total of 38 positive clones from this screening and found out that most of them derived from the CIPK members such as CIPK6 and CIPK24, which are previously known CBL10 interactors (Kim et al., 2007; de la Torre et al., 2013). Isolation of these genes indicated that the yeast two-hybrid screening had been performed efficiently and successfully. Among the positive clones, we were also able to discover a novel gene that does not belong to the CIPK family. Sequence analysis via GenBank (http://www.ncbi.nlm.nih.gov) showed that this gene is TOC34 (Translocon of the Outer membrane of the Chloroplasts; At5g05000) encoding a polypeptide with an estimated molecular mass of 34.7 kDa. It was reported to act as a GTP-dependent receptor at the chloroplast surface (Andrès et al., 2010).

To test whether the full-length TOC34 protein interacts with CBL10 in yeast cells, we constructed the pGAD·TOC34 (AD·TOC34) plasmid by cloning the ORF of TOC34 cDNA into the yeast expression vector pGAD·GH containing the GAL4 activation domain (AD). As shown in Figure 1, the Y190 yeast cells carrying both AD·TOC34 and BD·CBL10 grew well on the selection medium (SC-HLW) and exhibited a blue color in the filter-lift assay, indicating expression of the HIS3 and LacZ reporter genes, respectively. However, a vector-swapping analysis showed that the yeast cells harboring AD·CBL10 and BD·TOC34 did not express the two reporter genes, suggesting that the interaction between TOC34 and CBL10 occurs in a vector-dependent manner. Meanwhile, the yeast cells cotransformed with the empty vectors (either with BD·TOC34 and AD, or with BD and AD·TOC34), which were used as negative controls, failed to express the reporter genes. Together, these results clearly indicated that TOC34 interacts with CBL10 in the yeast two-hybrid system.

Because the CBL family members consist of 10 genes in Arabidopsis genome (Luan et al., 2002; Kolukisaoglu et al., 2004), we also investigated whether or not the TOC34 protein could interact with other members of the CBL family using the yeast two-hybrid system. To do this, we first cloned each of the 10 CBL ORFs into the BD vector (Ok et al., 2015) and then introduced them into the Y190 yeast cells harboring AD·TOC34. As shown in Figure 2, TOC34 was able to also interact with CBL7 besides CBL10, but not with other CBL family members. However, it should be noted that TOC34 interacted with CBL7 at much lower strength than with CBL10, indicating that TOC34 possesses different binding interaction affinities toward the two CBL members. This finding strongly suggests that each CBL family member can have distinct interacting partners in addition to the previously reported targets, CIPK and AtMTAN proteins (Oh et al., 2008; Luan, 2009; Ok et al., 2015).

To delimit the TOC34 region necessary for the interaction with CBL10, we created a series of deletion constructs by cloning TOC34 fragments into the AD vector. These constructs were then introduced into the Y190 yeast cells carrying either BD or BD·CBL10. Interactions were determined by monitoring whether or not the transformed yeast cells grow well on the selection medium (SC-HLW). As shown in Figure 3, removal of the last 54 amino acids from the C-terminal end of TOC34 (AD·TOC34-N) completely disrupted its interaction with CBL10, whereas the N-terminal deletion down to 260th amino acid residue containing small GTP-binding domain of TOC34 (AD·TOC34-C1) maintained the interaction with CBL10. However, a further deletion down to 283th amino acid residue (AD·TOC34-C2) containing a transmembrane domain (267 to 283 amino acid residues) resulted in abolishing the interaction. These results indicated that the 23-amino acid residues of TOC34, spanning from 261 to 283, are critical and necessary for the interaction with CBL10.

We performed pull-down assays in order to verify in vitro the CBL10-TOC34 interaction demonstrated in the yeast two-hybrid system (Figures 1, 2). To do this, we expressed and purified both TOC34 and CBL10-cMyc proteins from E. coli using the glutathione S-transferase (GST) gene fusion system. Figure 4A shows an approximately 35 kDa TOC34 protein band that was originally purified as a GST-fusion form and subsequently digested with thrombin to remove the GST protein. In the same way, the CBL10-cMyc protein was also prepared (Figure 4A). For the pull-down assays, we incubated GST-TOC34 (bait) with CBL10-cMyc (prey) in the presence or absence of Ca²⁺ and examined whether GST beads pulled down the CBL10-cMyc prey protein with the immunoblot analyses using anti-cMyc antibody as a probe. As shown in Figure 4B, the GST-TOC34 bait protein successfully retrieved CBL10-cMyc in the presence and absence of Ca²⁺, indicating that Ca²⁺ does not have significant influence at least on the interaction strength between CBL10 and TOC34. In contrast, GST protein alone used as a negative control failed to pull down CBL10-cMyc in both conditions. These in vitro protein-protein interaction assays demonstrated that CBL10 and TOC34 physically interact with each other regardless of Ca²⁺.

We also carried out the bimolecular fluorescence complementation (BiFC) assay to further confirm the CBL10-TOC34 interaction in vivo (Walter et al., 2004). To do this, we first created TOC34-YFP^N and CBL10-YFP^C chimeric constructs by fusing TOC34 and CBL10 to the N-terminal YFP fragment (YFP^N) and the C-terminal YFP fragment (YFP^C) in the binary vectors, respectively. Subsequently, these constructs were transiently expressed in the tobacco leaves via an Agrobacterium-infiltration method. As shown in the Figure 5A fluorescence images, YFP signals were observed strongly at the outer membrane of chloroplasts and weakly at the plasma membrane of the transformed tobacco cells, indicating a physical association between CBL10 and TOC34 in a living tobacco leaf cell. The fluorescent marker dye FM4-64 was used to visualize the plasma membrane. However, no yellow fluorescence was observed from the tobacco cells expressing either TOC34-YFP^N/YFP^C or YFP^N/CBL10-YFP^C (a negative control).

Furthermore, we also verified the CBL10-TOC34 interaction in Arabidopsis protoplasts transfected with pUC-TOC34-YFP^N and pUC-CBL10-YFP^C chimeric constructs. Similar to the result obtained from the tobacco leaf cells, strong YFP signals were detected around the chloroplast (Figure 5B). The basic leucine zipper (bZIP) transcription factor 63, bZIP63 (At5g28770), was used as a positive control, because it is known to form a homodimer in the nucleus (Sibéril et al., 2001). Taken together, these results clearly indicated that the physical interaction of CBL10 with TOC34 occurs primarily at the chloroplast in vivo.

Our results above clearly indicated that CBL10 associates with TOC34 when they exist in the same space. Therefore, it is very important to investigate the spatial expression patterns of the CBL10 and TOC34 genes in order to determine whether the CBL10-TOC34 association indeed occurs in Arabidopsis plant cells. We performed qRT–PCR analysis using total RNA prepared from the various organs of 6-week-old Arabidopsis plants [ecotype Columbia (Col-0)], which include leaves, stems, flowers, and roots. As shown in Figure 6A, both TOC34 and CBL10 genes were broadly and significantly expressed in the vegetative and reproductive organs with some differences in their expression patterns and levels. TOC34 transcripts were most abundantly expressed in roots, and to a lesser extent, in flowers, stems, and leaves, which coincides with the published result (Gutensohn et al., 2004). In the case of CBL10, however, the highest expression level was observed in the leaves although the other organs also accumulated the transcripts at fairly high levels. This was also in good agreement with the previous findings (Kim et al., 2007; Quan et al., 2007). It should be noted that CBL10 was more strongly expressed in all organs tested in this study than TOC34. Anyway, TOC34 and CBL10 clearly displayed overlapping spatial expression patterns in Arabidopsis plants, and therefore it seems reasonable to conclude that the two proteins actually associate with each other in Arabidopsis cells.

We further examined the subcellular localization of TOC34 in Arabidopsis cells, although our BiFC results indicated that TOC34 can be localized at the outer membrane of the chloroplasts and the plasma membrane when co-expressed with CBL10 in plant cells. To this end, we first fused GFP to the C-terminal end of TOC34 and created a TOC34-GFP chimeric gene, which is driven by the cauliflower mosaic virus 35S promoter (pMD·TOC34). TOC34-GFP fusion protein was then transiently expressed in Arabidopsis protoplasts and visualized with confocal laser scanning microscopy. According to the fluorescence images (Figure 6B), TOC34-GFP appeared to be predominantly localized at the chloroplasts (probably, the outer envelope membrane) and to a much lesser extent at the plasma membrane, which is generally consistent with the previous report (Dhanoa et al., 2010). The GFP control, however, exhibited similar intensities of fluorescence throughout the cytoplasm.

Because TOC34 acts as a GTP-dependent receptor at the outer membrane of the chloroplast (Jarvis et al., 1998), we attempted to empirically determine whether the recombinant GST-TOC34 protein prepared from E. coli is able to hydrolyze GTP into GDP and inorganic phosphate. As shown in Figure 7A, GST-TOC34 successfully yielded free phosphate (~10 U/μg), whereas GST alone did not. These enzyme assays clearly suggested that the TOC34 protein indeed possesses the GTPase activity in vitro. Because TOC34 can form a complex with the Ca²⁺-binding protein CBL10, we also examined the effect of CBL10 on the GTPase activity of the TOC34 protein in the presence or absence of Ca²⁺. As shown in Figure 7B, the GTPase activity of TOC34 was most significantly decreased when both CBL10 and Ca²⁺ were present together. Without either Ca²⁺ or CBL10, however, such a strong inhibitory effect was not detected. Furthermore, it is also important to noted that CBL4, which is another member of the CBL family and does not associate with TOC34, did not exert any substantial effect on the TOC34 enzymatic activity regardless of Ca²⁺. Together, these results clearly indicate that the GTPase activity of TOC34 can be specifically inhibited by the association with Ca²⁺-bound CBL10.

Our extensive yeast two-hybrid screening revealed that CBL10 can interact not only with the CIPK family members but also with TOC34. Using the pull-down assays and BiFC analyses (Figures 4B, 5), we verified that the CBL10-TOC34 interaction detected in the yeast two-hybrid system takes place in vitro as well as in vivo. Furthermore, we found that although CBL10 does not need Ca²⁺ to physically interact with TOC34, it still requires Ca²⁺ to inhibit the GTPase activity of TOC34 (Figures 4B, 7B). It is obvious that the interaction alone is not enough to modulate the TOC34 enzyme activity. This type of Ca²⁺ dependency was also observed in other CBL members: CBL2 and CBL4 interacted with CIPK14 and CIPK24, respectively, regardless of Ca²⁺, and yet Ca²⁺ binding was essential for the CBL members to activate their target kinases (Halfter et al., 2000; Ishitani et al., 2000; Akaboshi et al., 2008). Crystal structure analyses of CBL4 and CIPK24 provided a useful insight into the molecular mechanism underlying CBL4-mediated activation of CIPK24 (Sánchez-Barrena et al., 2005, 2007). Upon Ca²⁺ binding, CBL4 undergoes a conformational change to trigger structural alteration in the interacting partner CIPK24, resulting in activation of the kinase activity. Therefore, it is likely that Ca²⁺-bound CBL10 also exerts its inhibitory effect on TOC34 in a way similar to the CBL4-CIPK23 case. However, further investigation is required to unravel the molecular mechanism by which CBL10 decreases the GTPase activity of TOC34 in a Ca²⁺-dependent manner.

In order for the CBL10-TOC34 association to occur in Arabidopsis plant cells, it is prerequisite for both proteins to be present in the same space. Therefore, we examined whether CBL10 and TOC34 display some overlaps in their expression patterns and subcellular localizations. First, our qRT-PCR analysis (Figure 6A) along with the previous works (Gutensohn et al., 2004; Kim et al., 2007; Quan et al., 2007) clearly indicated that both CBL10 and TOC34 can exist together in the same tissues of Arabidopsis, albeit their relative amounts are significantly different. Second, TOC34 carries a transit peptide and mainly localizes to the chloroplasts (Dhanoa et al., 2010), which was further verified by our fluorescence image analysis of the TOC34-GFP fusion protein in this study (Figure 6B). Although TOC34-GFP was also detected at the plasma membrane at a lower level, it could be an artifact due to the 35S promoter driving overexpression of the TOC34-GFP protein in Arabidopsis cell. In the case of CBL10, it was reported to mainly localize at the cell membranes such as the plasma membrane and the tonoplast (Kim et al., 2007; Quan et al., 2007). However, our recent analysis using the ChloroP server (http://www.cbs.dtu.dk/services/ChloroP) predicted that CBL10 can be also targeted to the chloroplast, because it harbors a transit peptide sequence (42 amino acid residues) in the N terminus. In this aspect, it makes sense that the BiFC assays (Figure 5) displayed the strong YFP signals around the chloroplast, where both CBL10 and TOC34 are located. Taken together, these results strongly suggest that the CBL10-TOC34 association indeed occurs in vivo; therefore, CBL10 can relay the cytosolic Ca²⁺ signals to the chloroplast-localized TOC34 protein in addition to the previously known CIPK family members in Arabidopsis.

We learned through GenBank search that the Arabidopsis genome possesses an additional TOC34-like gene, TOC33 (At1g02280), which encodes a polypeptide similar to TOC34 (55% identity; 72% similarity). Recently, we isolated a full-length cDNA of the gene and found that TOC33 also interacts only with CBL10 but not with the other CBL members in the yeast two-hybrid system (data not shown). It seems that both members of the TOC34 family, TOC34 and TOC33, can be modulated by CBL10. These findings, together with our previous reports in which we showed that CBL3 targets the AtMTAN family, strongly suggest that each member of the CBL family can target various proteins with different biochemical properties like other Ca²⁺ sensor relays such as CaMs and CMLs. Such target diversity not only dramatically increases the level of complexity in the CBL-mediated Ca²⁺-signaling pathways but also allows CBLs to control a much wider range of cellular and physiological processes in response to a variety of Ca²⁺-eliciting stimuli in plants. In this context, it is critical to uncover novel interaction partners for other CBL members in order to fully understand the CBL-mediated signaling network in plants.

As stated above, CBL10 associates with and activates CIPK24/SOS2 in a Ca²⁺-dependent manner (Kim et al., 2007; Quan et al., 2007). Contrary to this activator role, we found in the present study that CBL10 acts as inhibitor when it forms a complex with the TOC34 protein; the GTPase activity of TOC34 was significantly decreased by Ca²⁺-bound CBL10 (Figure 7B). Therefore, it appears that CBL10, like CBL3, can also serve as either activator or inhibitor depending on its interaction partner proteins (Oh et al., 2008; Ok et al., 2015). How can Ca²⁺-bound CBL10 inhibit the GTPase activity of TOC34? Although the underlying mechanism is currently unavailable and waits to be investigated, we speculate that it has something to do with the fact that TOC34 forms a homodimer, causing a 1.5-fold increase in the GTPase activity (Reddick et al., 2007; Yeh et al., 2007; Koenig et al., 2008). In this aspect, it is interesting to note that CBL10 decreases the TOC34 GTPase activity by about 0.5-fold in the presence of Ca²⁺ (Figure 7B). Based on these facts, it is conceivable that Ca²⁺-bound CBL10 may disrupt or destabilizing the TOC34 homodimer, which is crucial for the optimal enzyme activity.

A large number of chloroplast proteins are nuclear encoded and many of them are imported into the plastid through two hetero-oligomeric protein apparatuses, the TOC and TIC complexes (Translocon at the Outer/Inner envelope Chloroplasts) (reviewed in Oreb et al., 2008; Schleiff and Becker, 2011). TOC34 is one of the core components constituting the TOC complex and plays a critical role in the protein import process by recognizing plastid-destined precursor proteins and presenting them to the translocation channel TOC75 in cooperation with TOC159. Because the GTPase activity of TOC34 is tightly linked to the translocation efficiency, modulation of the TOC34 enzymatic activity is considered a major regulatory step in the translocation process (Paila et al., 2015). In addition, it has been known that the import of a certain subset of chloroplast proteins is regulated by Ca²⁺ (Chigri et al., 2005). In this context, our present finding may provide a novel insight into a molecular mechanism how Ca²⁺-signals elicited by environmental stresses modulate the TOC34 GTPase activity and thereby regulate protein import into the chloroplast. The physical interaction between CBL10 and TOC34 in the presence of Ca²⁺ could be the actual regulatory mechanism occurring in plants to control the activity of the TOC complex. Anyway, considering the important roles of chloroplasts in photosynthesis and other metabolic pathways in plant cells, it is not surprising that this organelle should be integrated into the Ca²⁺-signaling network and be regulated to accommodate environmental changes. Since the CBL10 and TOC34 genes are expressed in both green and non-green tissues of Arabidopsis (Figure 6A; Gutensohn et al., 2004; Kim et al., 2007; Quan et al., 2007), it appears that the CBL10-TOC34 complex functions not only in the chloroplast but also in other plastid types.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.