The magnesium-chelatase H subunit binds abscisic acid and functions in abscisic acid signaling: new evidence in Arabidopsis.

Using a newly developed abscisic acid (ABA)-affinity chromatography technique, we showed that the magnesium-chelatase H subunit ABAR/CHLH (for putative abscisic acid receptor/chelatase H subunit) specifically binds ABA through the C-terminal half but not the N-terminal half. A set of potential agonists/antagonists to ABA, including 2-trans,4-trans-ABA, gibberellin, cytokinin-like regulator 6-benzylaminopurine, auxin indole-3-acetic acid, auxin-like substance naphthalene acetic acid, and jasmonic acid methyl ester, did not bind ABAR/CHLH. A C-terminal C370 truncated ABAR with 369 amino acid residues (631-999) was shown to bind ABA, which may be a core of the ABA-binding domain in the C-terminal half. Consistently, expression of the ABAR/CHLH C-terminal half truncated proteins fused with green fluorescent protein (GFP) in wild-type plants conferred ABA hypersensitivity in all major ABA responses, including seed germination, postgermination growth, and stomatal movement, and the expression of the same truncated proteins fused with GFP in an ABA-insensitive cch mutant of the ABAR/CHLH gene restored the ABA sensitivity of the mutant in all of the ABA responses. However, the effect of expression of the ABAR N-terminal half fused with GFP in the wild-type plants was limited to seedling growth, and the restoring effect of the ABA sensitivity of the cch mutant was limited to seed germination. In addition, we identified two new mutant alleles of ABAR/CHLH from the mutant pool in the Arabidopsis Biological Resource Center via Arabidopsis (Arabidopsis thaliana) Targeting-Induced Local Lesions in Genomes. The abar-2 mutant has a point mutation resulting in the N-terminal Leu-348-->Phe, and the abar-3 mutant has a point mutation resulting in the N-terminal Ser-183-->Phe. The two mutants show altered ABA-related phenotypes in seed germination and postgermination growth but not in stomatal movement. These findings support the idea that ABAR/CHLH is an ABA receptor and reveal that the C-terminal half of ABAR/CHLH plays a central role in ABA signaling, which is consistent with its ABA-binding ability, but the N-terminal half is also functionally required, likely through a regulatory action on the C-terminal half.

The phytohormone abscisic acid (ABA) regulates many aspects of plant growth and development, such as seed maturation, germination, and seedling growth, and is a central hormone in the control of plant adaptation to environmental challenges, including drought, salt, and cold stresses, by regulating stomatal aperture and expression of stress-responsive genes (for review, see Koornneef et al., 1998;Leung and Giraudat, 1998;. ABA functions through a complex network of signaling pathways, where ABA signal perception by ABA receptors is the primary event that triggers downstream signaling cascades to induce the final physiological responses. Numerous cellular components that modulate ABA responses downstream of ABA receptors have been identified (for review, see Himmelbach et al., 2003;Fan et al., 2004), leading to considerable progress in understanding ABA signaling pathways. ABA signal perception, which has been considered to be mediated by multiple receptors, including plasma membrane and intracellular receptors (for review, see Assmann, 1994;, has attracted much attention. In recent years, two classes of plasma membrane ABA receptors, an unconventional G-protein-coupled receptor (GPCR), GCR2, and a novel class of GPCR, GTG1 and GTG2, that regulate the major ABA responses in seed germination, seedling growth, and stomatal movement (Johnston et al., 2007;Liu et al., 2007aLiu et al., , 2007bPandey et al., 2009), have been reported. However, whether GCR2 regulates ABA-mediated seed germination and postgermination growth is controversial, because the ABA-related phenotypes are weak to absent in gcr2 mutants Guo et al., 2008). GTGs are positive regulators of ABA signaling and interact with the sole Arabidopsis (Arabidopsis thaliana) G-protein a-subunit, GPA1, that may negatively regulate ABA signaling by inhibiting the activity of GTG-ABA binding (Pandey et al., 2009). Most recently, a PYR/PYL/RCAR family of START proteins was reported to function as a cytosolic ABA receptor by inhibiting directly type 2C protein phosphatases, which is suggested to trigger a reversible protein phosphorylation cascade to mediate ABA responses (Ma et al., 2009;Park et al., 2009).
We previously reported the magnesium-protoporphyrin IX (ProtoIX) chelatase large subunit (Mgchelatase H subunit; CHLH) as an ABA receptor residing in chloroplasts and mediating ABA responses in seed germination, postgermination growth, and stomatal movement (Shen et al., 2006). CHLH has multiple functions in plant cells. As a subunit of the Mg-chelatase, CHLH catalyzes the introduction of Mg to ProtoIX, a key regulatory step of chlorophyll biosynthesis. In addition, CHLH plays a key role in mediating plastid-to-nucleus retrograde signaling (for review, see Nott et al., 2006). We found that a broad bean (Vicia faba) ABA-specific binding protein, ABAR (for putative abscisic acid receptor), is encoded by the CHLH gene (Zhang et al., 2002;Shen et al., 2006). In the reference plant Arabidopsis, we showed that ABAR/ CHLH specifically binds ABA and mediates ABA signaling as a positive regulator, indicating that ABAR is an intracellular ABA receptor. We showed also that ABAR-mediated ABA signaling is distinct from chlorophyll biosynthesis and plastid-to-nucleus retrograde signaling (Shen et al., 2006). However, in barley (Hordeum vulgare), a recent report showed that the barley chlorophyll-deficient mutants with mutations in the XanF gene (the same large subunit of Mgchelatase as the CHLH in Arabidopsis) showed no ABA-related phenotypes in seed germination, postgermination growth, and stomatal movement, and the ABA-binding activity of XanF was not detected using the system of the authors, which led them to question the ABA-receptor nature of ABAR/CHLH, at least in barley (Mü ller and Hansson, 2009). Here, we report, using a newly developed ABA-affinity chromatography technique and a [ 3 H]ABA-binding assay, that ABAR/CHLH was shown to specifically bind ABA via the C-terminal half but not the N-terminal half. Expression of the ABAR C-terminal half truncated protein in wild-type plants and in an ABA-insensitive cch mutant of ABAR/CHLH induced much stronger ABA-related phenotypes than expression of the N-terminal half truncated protein did, revealing a critical role of the C-terminal half in ABA signaling. Characterization of two new mutant alleles with point mutations in the N-terminal half of ABAR/CHLH, abar-2 and abar-3, suggests that the N-terminal half may cover a regulatory domain involved in modulating C-terminal function. These data support the idea that the Mg-chelatase H subunit is an ABA receptor.

Affinity Chromatography Shows That ABAR Specifically Binds ABA
We developed an ABA-affinity chromatography technique to detect ABA binding to a known protein essentially based on our ABA-binding protein purification procedures (Zhang et al., 2002). An ABA-linked Sepharose 4B was used to assess qualitatively whether a given protein binds ABA. The Escherichia coliexpressed full-length ABAR was loaded onto the ABA-linked Sepharose 4B column and first eluted by 150 mM NaCl of total volume 36 mL at 3 mL each for 12 times (Fig. 1A, lanes 1-12) to eliminate completely unspecifically bound or loosely bound protein by the column. It is noteworthy that this eliminated portion by the 150 mM NaCl elution should also include a disassociated portion of an ABA-specific binding protein in the reversible binding process. After this elution, the column was further eluted by 100 or 150 mM NaCl plus 4 mM (6)ABA. This ABA-affinity competitive elution released efficiently the ABA-specific binding portion of the ABAR protein (Fig. 1A, lanes 13-15). In contrast, elution with the inactive isomer 2-trans,4trans-ABA could not substantially recover the ABAR protein (Fig. 1B, lanes 13-15). As a negative control, the Sepharose 4B column that was not coupled with ABA did not bind tightly to the ABAR protein. The loosely bound protein was completely removed from the non-ABA-linked column with 150 mM NaCl of total volume about 15 mL at 3 mL each for five times (Fig. 1C, lanes 1-5), and a subsequent elution with 100 mM NaCl plus 4 mM (6)ABA did not recover any detectable ABAR protein (Fig. 1C,. It is noteworthy that the protein eluted by 150 mM NaCl from the ABA-linked column needed, in most cases, about 30 mL of elution with 3 mL each for nine to 10 times (Fig. 1, A and B, lanes 1-10), showing that the ABA-linked column bound the ABAR protein specifically and tightly. The 4 mM ABA elution with either 100 or 150 mM NaCl showed no difference in efficiency in recovering ABAR protein from the ABA-affinity column (Fig. 1, A and D), indicating that this ABA-specific elution was independent of NaCl concentrations in the range from 100 to 150 mM.
To further test the specificity of binding of ABAR protein to the ABA-affinity column, the ABAR protein was incubated with 50 or 500 nM (+)ABA in order to compete with the Sepharose 4B-linked ABA for free (+)ABA. The ABA concentration of 50 nM is slightly lower than the saturation concentration of ABA binding to ABAR protein (disassociation constant [K d   . Affinity chromatography shows that ABAR binds ABA via the C-terminal half but not the N-terminal half. A, The E. coliexpressed, purified, full-length ABAR was loaded onto an ABA-linked Sepharose 4B chromatography column and eluted by 150 mM NaCl of total volume 36 mL at 3 mL each for 12 times (lanes 1-12) to remove unspecifically bound or loosely bound protein by the column. The eluate was subjected to 10% SDS-PAGE, and the ABAR protein was detected by the anti-full-length ABAR serum with immunoblotting. Note that the protein eluted by 150 mM NaCl from the ABA-linked column needed, in most cases, about 30 mL of elution with 3 mL each for nine to 10 times (lanes 1-10). The column was eluted, after the elution with 150 mM NaCl, by 100 mM NaCl plus 4 mM (6)ABA, which released the ABA-specific binding portion of the ABAR protein (lanes 13-15). The left panel shows molecular mass markers (kD). B, The ABA-specific binding portion of the ABAR protein could not be eluted substantially from the column by the ABA isomer 2-trans,4-trans-ABA (lanes 13-15) under the same elution conditions as used in A: only a trace of the ABAR protein was detected (lane 13). C, A negative control. The Sepharose 4B column that was not coupled with ABA did not tightly bind the ABAR protein. The loosely bound protein was completely removed from the non-ABA-linked column with 150 mM NaCl of total volume about 15 mL at 3 mL each for five times (lanes 1-5), and a subsequent elution with 100 mM NaCl plus 4 mM (6)ABA did not recover any detectable ABAR protein (lanes 13-15). D, The same assays were performed as in A, but for the ABA-specific elution the combination of 150 mM NaCl plus 4 mM (6)ABA was used instead of 100 mM NaCl plus 4 mM (6)ABA. Note that the same result was observed as in A. E, Competition assays of ABA binding to the ABA-affinity column. The protein sample was preincubated with 0, 50, or 500 nM (+)ABA (top panels) before being loaded onto the column, of the Sepharose 4B-linked (6)ABA was about 500 nM in the binding-incubation system (see "Materials and Methods"), equivalent to 250 nM (+)ABA. The incubation of ABAR protein with 50 nM free (+)ABA efficiently decreased the amount of ABAR protein eluted from the column, and the immunosignal of the ABAR protein was undetectable with the 500 nM (+)ABA incubation (Fig. 1E). This competition assay showed that the ABAR binding to the ABA-affinity column is specific. The competition efficiency of the free (+)ABA with Sepharose-bound ABA for ABAR protein was high, likely because the important carboxyl group at the C1 site of ABA is blocked by the Sepharose coupling, which may decrease the ability of binding to ABAR protein in comparison with free (+)ABA.
In addition to trans-ABA, we used a set of potential agonists/antagonists to ABA, gibberellin GA 3 , cytokininlike regulator 6-benzylaminopurine (6-BA), auxin indole-3-acetic acid (IAA), and jasmonic acid methyl ester (MeJA), to elute ABAR protein from the ABAaffinity column in place of (6)ABA. However, elution with any of these substances did not recover detectable ABAR protein (Fig. 1F,, showing that these substances are not able to bind the ABAR protein. To get further evidence that the Arabidopsis natural ABAR binds ABA, the Arabidopsis total protein was loaded on the ABA-linked Sepharose 4B, and the column was subjected to the same elution procedures mentioned above. The ABA-affinity competitive elution recovered efficiently the ABAR protein from the ABA-linked Sepharose 4B column but not from the ABA-free column (Fig. 1G). A competition assay, performed with the same procedure as described for the E. coli-expressed ABAR protein (Fig. 1E), showed that preincubation of the Arabidopsis total protein with 50 nM (+)ABA decreased significantly the amount of ABAR protein eluted from the column, and the 500 nM (+)ABA preincubation completely eliminated the ABAR immunosignal (Fig. 1G). These results, consistent with those of E. coli-expressed ABAR protein, showed that the Arabidopsis natural ABAR protein is able to bind ABA.
Additionally, using this ABA-affinity column, we showed that the E. coli-expressed barley XanF binds ABA and that, in the barley natural proteins, both the full-length and a truncated XanF (if it exists naturally) bind ABA (Supplemental Fig. S2). In rice, the two natural CHLHs both bind ABA (Supplemental Figs. S2 and S3). To assess ABA-binding domains in the ABAR molecule, three truncated ABAR proteins were used to perform the ABA-affinity chromatography assays. The truncated C-terminal C751 protein (amino acid resi- Figure 1. (Continued.) and the column was subjected to the same elution procedures as in D. Note that the 50 nM ABA treatment decreased significantly the amount of ABAR protein eluted from the column, and the immunosignal of the ABAR protein was undetectable with the 500 nM ABA treatment. The protein samples displayed in lanes 13, 14, and 15 [ABAR protein eluted by 150 mM NaCl plus 4 mM (6)ABA] were further assayed by immunoblotting on the same gel (bottom panel, red arrow), and the relative amounts of the immunodetected ABAR protein were estimated by scanning the bands with the values given above the bands. F, The potential agonists/antagonists, other phytohormones, or plant growth regulators GA 3 , cytokinin-like regulator 6-BA, auxin IAA, and MeJA are not able to bind the ABAR protein. The ABA-linked Sepharose 4B affinity column was eluted by 100 mM NaCl plus 4 mM of the substances used separately after 150 mM NaCl elution of 36 mL at 3 mL each for 12 times under the same conditions as in A. Elution with any of these substances did not recover detectable ABAR protein (lanes 13-15). The symbol X indicates the different substances. G, The Arabidopsis natural ABAR protein binds ABA. The Arabidopsis total protein was loaded on the ABA-linked Sepharose 4B, and the column was subjected to the same elution procedures as described in A. The 100 mM NaCl plus 4 mM (6)ABA elution after the 150 mM NaCl elution recovered the ABAR protein from the ABA-linked column (top panel, Sepharose 4B-ABA) but not from the ABA-free Sepharose 4B column (Sepharose 4B). An incubation of the protein sample with 50 nM (+)ABA (50 nM ABA/Sepharose 4B-ABA) decreased significantly the ABAR protein bound to the ABA-affinity column, and incubation with 500 nM (+)ABA completely abolished the ABAR binding to the column: no immunosignal of ABAR protein was detected with the ABA-specific elution (500 nM ABA/Sepharose 4B-ABA). A control was performed with 0 nM (+)ABA but the same amount of ethanol for solubilizing 500 nM ABA (0 nM ABA/Sepharose 4B-ABA). Lanes 1, 2, and 3 denote immunosignals of the first and last elution with 150 mM NaCl and the first elution with 100 mM NaCl plus 4 mM (6)ABA, respectively. H, The truncated C-terminal C751 protein (amino acid residues 631-1,381; a) and C370 protein (amino acid residues 631-999; b) bind ABA, but the N-terminal truncated N772 protein (amino acid residues 1-772; c) does not. The C751, C370, and N772 proteins were loaded on the ABA-linked Sepharose 4B column that was subjected to the same elution procedures as in A. The 100 mM NaCl plus 4 mM (6)ABA elution after the 150 mM NaCl elution recovered the C751 (a) and C370 (b) proteins from the ABA-linked column (lanes 13 and 14), but the N772 protein was removed quickly from the column by 150 mM NaCl elution (c) and was not recovered by the 100 mM NaCl plus 4 mM (6)ABA elution (lanes 13 and 14). I, The C751 truncated protein was not recovered from the ABA-linked column by an elution with 100 mM NaCl plus 4 mM IAA-like substance NAA (lanes 13 and 14). The elution procedures were the same as in H. All of the experiments shown here were replicated three times with similar results. J, Saturable ABA-specific binding to the pure E. coli-expressed full-length ABAR and truncated C751 and C370 proteins. Asterisks indicate significant differences at P , 0.05 (Student's test) when comparing values (means 6 SD, n = 3) within the same [ 3 H]ABA concentration. The arrow indicates the unspecific binding data for the three proteins, which are all lower than 10% of total binding. K, Scatchard plot of binding data in J. The parameters of the curves are as follows: for ABAR, K d = 32 nM, B max = 1.27 mol mol 21 , r 2 = 0.97; for C751, K d = 40 nM, B max = 1.09 mol mol 21 , r 2 = 0.96; for C370, K d = 41 nM, B max = 0.89 mol mol 21 , r 2 = 0.96. dues 631-1,381; Fig. 2A), and a fragment located in the C751 portion, C370 truncated protein (amino acid residues 631-999; Fig. 2A), were shown to bind ABA (Fig. 1H, a and b). The specificity of ABA binding to the C751 truncated protein was assayed using a set of potential agonists/antagonists (including all of the substances used above and additionally the auxin-like substance naphthalene acetic acid [NAA]) as for ABAR full-length protein (Fig. 1F), and the results showed that these potential agonists/antagonists were not able to compete with ABA for binding to the C751 truncated protein, revealing that the C751-ABA binding is specific ( Fig. 1I; data not shown). However, the N-terminal N772 truncated protein (amino acid residues 1-772; Fig. 2A) was removed quickly from the ABA-linked column by 150 mM NaCl elution, of which the immunoblotting band disappeared almost completely in the fourth elution (total volume, 12 mL) by 150 mM NaCl (Fig. 1H, c, lanes 1-4), while the immunoblotting bands of the C751 and C370 truncated proteins were still retained in the ninth or 10th elution by 150 mM NaCl (total volume, 27-30 mL; Fig. 1H, a and b, lanes 9 and 10). Importantly, the N772 truncated protein was not recovered by the ABA-affinity competitive elution (Fig. 1H, c, lanes 13 and 14). All of these findings showed that the N-terminal N772 truncated protein did not bind ABA.
The ABA-binding parameters of the C751 and C370 truncated ABAR proteins were assayed by a [ 3 H]ABA in-buffer binding system. The C751 truncated protein binds ABA with a K d of 40 nM and maximum binding (B max ) of 1.09 mol ABA mol 21 protein, and the C370 truncated protein binds ABA with a K d of 41 nM and B max of 0.89 mol mol 21 (Fig. 1, J and K; Supplemental Table S1). The mutations of C751 and C370 affected the two ABA-binding parameters compared with the fulllength ABAR (K d = 32 nM, B max = 1.27 mol mol 21 ; Shen et al., 2006), but both the truncated ABAR proteins could still perceive ABA at the nanomolar level.

Expression of the C-Terminal Half of ABAR in the Wild-Type Plants Confers ABA Hypersensitivity in All of the Major ABA Responses, But the Effect of Expression of the N-Terminal Half Was Limited to Seedling Growth
To test the functional domains of ABAR/CHLH based on the findings from the ABA-binding assays, we first created transgenic lines in the wild-type ecotype Columbia (Col-0) background using four different constructs of ABAR/CHLH cDNA fragments. We previously found that the expression of the fulllength ABAR open reading frame often caused cosuppression of the ABAR/CHLH gene, which made it difficult to obtain transgenic lines. This was also observed by another group (Strand et al., 2003). We hypothesized that there might be a sequence responsible for the protein/mRNA stability of ABAR/CHLH and tried to use a truncated ABAR with the C-terminal 78 amino acid residues deleted (amino acid residues 1-1,303, called ABARn; Fig. 2A), which indeed made it much easier to obtain transgenic lines (see below). We also tried to use a truncated ABAR with N-terminal deletion of 100 to 200 amino acid residues covering the chloroplast transit peptide, but we failed to obtain transgenic lines; therefore, we used an N-terminal 310amino acid-deleted ABAR truncation (amino acid residues 311-1,381, named ABARc) for transgenic manipulation ( Fig. 2A). The other three constructs included the truncated ABARs used in the abovementioned ABA-binding assays (i.e. C751, C370 [linked to the N-terminal chloroplast transit peptide, amino acid residues 1-120], and N772; Fig. 2A). These truncated ABARs were fused with GFP and shown to correctly express in the transgenic lines ( Fig. 2B) and localized in chloroplasts except for ABARc and C751, which localized in the cytosol because of lack of the chloroplast transit peptide (Fig. 2D). Expression of any of the truncated ABARs in the wild-type background did not affect the contents of Mg-ProtoIX, ProtoIX, and chlorophyll of the transgenic lines, showing that the chlorophyll biosynthesis was not altered substantially (Fig. 2,E and F). In contrast, the ABA sensitivity of these transgenic plants was . The transgenic plants expressing the four truncated ABARs ABARn, ABARc, C751, and C370 all showed significantly the ABA-hypersensitive phenotypes in seed germination, postgermination growth (Fig. 3), and ABA-induced stomatal closure and ABA-inhibited stomatal opening (Fig. 5A). However, the transgenic plants expressing the N-terminal N772 truncated protein showed substantially wild-type phenotypes in seed germination (Fig. 3A) and in postgermination growth when the seeds were planted directly in the ABA-containing medium (Fig. 3B), while the ABAhypersensitivity phenotype in seedling growth was observed in the N772 transgenic plants when the seedlings were transferred to the ABA-containing medium from the ABA-free medium 48 h after stratification ( Fig. 3, C and D), which suggests the complexity of the mechanisms involved in the ABAR-mediated ABA signaling. In ABA-induced stomatal closure and ABA-inhibited stomatal opening, the N772 transgenic plants showed wild-type phenotypes (Fig. 5A). It is noteworthy that the transformation of the plants with empty vector (control vector harboring GFP) did not induce any chlorophyll-or ABA-related phenotypes (Supplemental Fig. S1). We created transgenic lines in the ABA-insensitive cch mutant background using the same constructs of ABAR/CHLH cDNA fragments as in wild-type Col-0. Molecular analysis showed that the constructs were correctly expressed in the transgenic plants without ABARn and ABARc, The truncated ABAR with the C-terminal 1,304-to-1,381 sequence (78 amino acid residues) and the N-terminal 1-to-310 sequence (310 amino acid residues) deleted, respectively. ABAR, Full-length ABAR. The cch mutation site in the ABAR protein is indicated. C751, The truncated ABAR with the N-terminal 1-to-630 sequence deleted. C370, The truncated ABAR with the N-terminal 121-to-630 sequence and the C-terminal 1,000-to-1,381 sequence deleted but fused with the 1-to-120 sequence covering the transit peptide. N772, The ABAR N-terminal 1-to-772 sequence only. B, Immunoblot analysis of the transgenic lines with both the anti-full-length ABAR and anti-Actin sera. The left panels show the molecular mass markers (kD). Col, Wild-type plants; 1, 2, and 3 indicate three independent transgenic lines. In all panels, ABAR indicates the natural ABAR protein and Actin indicates the loading control. a, ABARn transgenic lines. ABARn indicates the GFP-tagged ABARn transgenic protein. altering the cch mutation ( Fig. 2C; data not shown). Expression of two truncated ABARs, ABARn and C370, linked to the chloroplast transit peptide enhanced the contents of Mg-ProtoIX and ProtoIX of the transgenic lines, although the contents were still lower than in the wild-type plants (Fig. 2E). However, the chlorophyll contents of the transgenic lines were comparable to those in the wild-type plants (Fig. 2F). Contrarily, expression of the other three truncated proteins, ABARc, C751, and N772, did not restore the levels of Mg-ProtoIX, ProtoIX, and chlorophyll (Fig. 2,E and F). For the ABA-related phenotypes, expression of C751 and N772 truncated ABAR restored the ABA sensitivity in seed germination of the cch mutant, while expression of the ABARn, ABARc, and C370 truncated proteins made the cch transgenic seeds hypersensitive to ABA compared with the wild-type plants (Fig. 4A). However, the expression of the N772 A 257-bp PCR-amplified fragment with an introduced HaeIII restriction site can be cut into a smaller sequence for the wild-type ABAR (ABAR-f) but cannot be cut for the cch mutation (cch-f), so that the transgenic lines in the cch mutant background harbor two bands of the PCR products. Col, Wild-type plants; cch, cch mutant; ABARn, ABARc, C370, N772, and C752 indicate ABARn, ABARc, C370, N772, and C751 transgenic lines, respectively. D, Transient expression of the GFP-tagged ABARn, ABARc, C751, C370, and N772 constructs in Arabidopsis protoplasts. GFP, Fluorescence of the GFP-tagged truncated ABAR protein; Auto-, chlorophyll autofluorescence; Bright, bright field; Merged, overlay of the three images. Note that ABARn was predominantly localized to the envelope of chloroplasts, ABARc and C751 were localized to cytosol, and C370 and N772 were localized to the chloroplast stroma. E, The concentrations of Mg-ProtoIX and ProtoIX in leaves of the transgenic plants. F, The concentrations of chlorophyll in leaves of the transgenic plants. Chla, Chlb, and Total indicate chlorophyll a, chlorophyll b, and total amounts of two kinds of chlorophyll, respectively. Each value in E and F is the mean 6 SE of three biological determinations, and different letters indicate significant differences at P , 0.05 (Student's test) when comparing values within the same group. Other symbols in E and F are as in C. . Phenotypic analysis of transgenic lines of the wild-type background in seed germination and postgermination growth. A, Seed germination rate of the wild-type plants (Col) and different transgenic lines in the ABA-free medium (0 mM ABA) and ABA-containing medium (0.5 mM) from 24 to 60 h after stratification. B, Postgermination growth of the wild-type plants (Col) and different transgenic lines in the ABA-free medium (0 mM ABA) and 0.5 mM ABA-containing medium 12 d after stratification. The seeds were directly planted in the ABA-free or ABA-containing medium to investigate the response to ABA after germination. C and D, Seedling growth of the wild-type plants (Col) and different transgenic lines in the ABA-free medium (0 mM ABA) and ABA-containing medium 12 d after seedling transfer. Seeds were germinated after stratification on common MS medium and about 48 h later transferred to the ABA-containing medium in the vertical position. Seedling growth was investigated 12 d after the transfer (D), and the length of primary roots was measured (C). Each value in A and C is the mean 6 SE of three biological determinations, and asterisks indicate significant differences at P , 0.05 (Student's test) when comparing values within the same point of time (A) or the same ABA concentration (C).
truncated ABAR had no effect on postgermination growth and stomatal movement, while the expression of the other four truncated ABARs, ABARn, ABARc, C751, and C370, restored the ABA sensitivity in both of the ABA major responses (Figs. 4, B-D, and 5B). It is noteworthy that the transformation of the cch mutant plants with empty vector (a control) did not induce any phenotypic change (Supplemental Fig. S1).

Characterization of Two Mutant Alleles in the ABAR Gene Reveals That the N-Terminal Half Is Required to Regulate ABA Signaling
We identified two new mutant alleles of the ABAR/ CHLH gene, abar-2 and abar-3, from the mutant pool in the Arabidopsis Biological Resource Center (ABRC) via the Arabidopsis Targeting Induced Local Lesions in Genomes (TILLING) project (Henikoff et al., 2004). The abar-2 mutation had a C-to-T change at nucleotide 1,042, and abar-3 had a C-to-T substitution at nucleotide 548, resulting in Leu-348/Phe and Ser-183/Phe mutations, respectively, in ABAR protein (Fig. 6A), both of which localize in the N-terminal half of the ABAR/CHLH protein. Neither of the mutations affects Mg-ProtoIX and chlorophyll production (Fig.  6, B and C), but both of the mutations alter the ABA sensitivity in germination and postgermination growth (Fig. 6, D-G). The abar-2 mutant has an ABAinsensitive phenotype in seed germination, like the cch mutant (Fig. 6D), while the abar-3 mutant has an ABAhypersensitive phenotype in seed germination (Fig.  6D), contrary to abar-2. The abar-3 plants showed a stronger ABA-insensitive phenotype in postgermination growth, like the cch mutant, when the seeds were planted directly in the ABA-containing medium (Fig.  6, E-G), but the abar-2 plants showed a weaker ABAinsensitive phenotype in postgermination growth only when postgermination growth was prolonged ( Fig. 6E; 1 mM ABA with seedling growth for 30 d). However, both abar mutants have substantially no ABA-related phenotypes in ABA-induced stomatal closure and ABA-inhibited stomatal opening (data not shown).  The newly developed ABA-affinity chromatography technique allowed us to assay the ABAR/CHLH ABA binding with a different system from those we used previously (Shen et al., 2006) and further to assess the ABA-binding domain in this protein. Unspecific elution by 150 mM NaCl showed that the ABA-affinity column bound tightly the ABAR/CHLH protein compared with the Sepharose 4B column uncoupled with ABA, but it bound the non-ABA-binding ABAR truncated N772 protein loosely (Fig. 1). The longer retained ABAR/CHLH protein in the ABA-affinity column during the unspecific elution process (Fig. 1) may likely be an ABA-specific binding portion of ABAR/ CHLH that was reversibly disassociated from the affinity column during the known receptor-ligand reversible binding process. Importantly, the tightly and specifically bound portion of ABAR/CHLH to the ABA-affinity column was efficiently eluted by (6)ABA but not by a set of potential agonists/antagonists, including 2-trans,4-trans-ABA, GA 3 , 6-BA, IAA, NAA, and MeJA (Fig. 1). Furthermore, this ABA-affinity column allowed us to pull down the natural ABAR/ CHLH protein, but the Sepharose 4B column uncoupled with ABA could not (Fig. 1). Free (+)ABA at 50 nM competed efficiently with the Sepharose-linked ABA for both E. coli-expressed and natural ABAR protein (Fig. 1), indicating that ABAR bound ABA at the nanomolar level. These findings showed that the ABA-affinity chromatography technique is specific and reliable for detecting ABA-binding proteins, on the one hand, and that ABAR/CHLH directly binds ABA specifically, on the other hand. Using this ABAaffinity column, we showed that the C-terminal half of ABAR/CHLH, but not the N-terminal half, binds ABA (Fig. 1). A C-terminal C370 truncated ABAR with 369 amino acid residues (631-999) was shown to bind ABA, which may be a core of the ABA-binding domain in the C-terminal half. With the in-buffer [ 3 H]ABAbinding assays we conducted previously (Shen et al., 2006), we further showed that the two C-terminal truncated ABAR proteins bound ABA at the nanomolar level, which allows plant cells to perceive physiological concentrations of ABA, although their ABA-binding ability decreases compared with fulllength ABAR (Fig. 1).
Using both the in-buffer radiolabeled ABA-binding system and a pull-down assay, we previously showed that ABAR/CHLH binds ABA with high affinity (K d = 32 nM), high stereospecificity [with (-)ABA and trans-ABA unable to compete with (+)ABA in the ABA binding], and in a saturable, reversible manner and that an ABAR/CHLH protein can bind maximally 1.28 ABA molecules (Shen et al., 2006), which meets the primary criteria for receptor-ligand binding. Consistently, this experiment, using a newly developed system, provides, to our knowledge, new evidence that ABAR/CHLH is an ABA-specific binding protein.

The C-Terminal Half of ABAR Plays a Central Role in ABA Signaling, and the N-Terminal Half Is Also Functionally Required
The transgenic manipulation showed that expression of the three C-terminal truncated ABARs, ABARc,  , and total chlorophyll (Total Chl; C) in wild-type plants (Col) and cch, abar-2, and abar-3 mutants. D, Seed germination rates of the wild-type plants and the three mutants in the medium supplemented with different concentrations of (6)ABA (0, 0.5, 1, and 3 mM) from 24 to 84 h after stratification. E, Postgermination growth of the wild-type plants (Col) and the three mutants in the medium supplemented with 0.5, 1, or 3 mM (6)ABA at different periods (14, 17, or 20 d) after stratification. Seeds were directly planted in the ABA-containing medium to investigate the response of seedling growth to ABA after germination. F and G, Seedling growth (estimated by root length) of the wild-type plants (Col) and the three mutants in the ABA-free medium (0 mM ABA) and ABA-containing medium (0.5, 1, and 3 mM) 30 d after stratification. Seeds were directly planted on plates containing medium supplemented with ABA, which C751, and C370, in wild-type Col-0 plants conferred ABA hypersensitivity and, in the ABA-insensitive cch mutant, restored ABA sensitivity in all of the ABA responses, including ABA-inhibited seed germination, postgermination growth, ABA-induced stomatal closure, and ABA-inhibited stomatal opening (Figs. 3-5). The effects induced by expression of these three truncated ABARs were comparable to those induced by the substantially full-length ABAR (ABARn; Figs. 3-5), although the ABARc and C751 truncated proteins uncoupled with the chloroplast transit peptide were expressed in cytosol but not in chloroplast (Fig. 1). All of these findings provide evidence for a central role of the C-terminal half of ABAR/CHLH in ABA signaling.
The underlying mechanism of ABAR-mediated ABA signaling across the chloroplast envelope remains an open question. The ABARc and C751 transgenically induced effects suggest that the C-terminal half of ABAR/CHLH may function in cytosol to mediate ABA signaling. The function of the truncated ABARs in the cytosolic side allows a hypothetical model that the entire ABAR molecule may move, through the chloroplast envelope, into the cytosolic space in response to ABA stimulation or that, structurally, ABAR may have transmembrane domains (e.g. C or N terminus) across the chloroplast envelope into the cytosolic side, with which ABAR mediates cytosolic signaling events. The work to test this hypothesis is ongoing in our laboratory. A recent report showed that a chloroplast protein, NRIP1, is recruited to the cytoplasm and nucleus by the tobacco mosaic virus p50 effector to mediate pathogen recognition processes in plant cells (Caplan et al., 2008), supporting the idea of chloroplast protein trafficking through the chloroplast envelope.
It is noteworthy that the expression of the four truncated ABARs (ABARn, ABARc, C751, and C370) in the wild-type plants did not significantly change the chlorophyll biosynthesis when altering ABA responses, and the expression of the ABARc and C751 truncated proteins uncoupled with the chloroplast transit peptide in the cch mutant affected ABA responses but did not restore the low chlorophyll level of the yellow mutant, which reveals that the ABAR/ CHLH-mediated ABA signaling is distinct from chlorophyll biosynthesis, consistent with our previous observation (Shen et al., 2006). It is also noteworthy that the C370 truncated protein linked to the N-terminal chloroplast transit peptide restored chlorophyll biosynthesis in the cch mutant as the substantially fulllength ABAR (ABARn) did (Fig. 2), suggesting that this C-terminal fragment is crucial to both ABA signaling and chlorophyll biosynthesis.
The expression of the N-terminal truncated ABAR N772 protein in the wild-type plants did not affect, and in the cch mutant did not restore, chlorophyll biosynthesis. However, the N772 expression partially altered the ABA-responsive phenotypes in seedling growth of the wild-type plants and in seed germination of the cch mutant (Figs. 3 and 4) but not in stomatal response to ABA in either transgenic wildtype or cch mutant plants (Fig. 5). Two mutant alleles with point mutations in the N-terminal half of ABAR/ CHLH, abar-2 and abar-3 (Fig. 6), have chlorophyll levels comparable to the wild-type plants but significantly altered ABA responses in seed germination and postgermination growth (Fig. 6), whereas neither of the two point mutations significantly affects stomatal responses to ABA (data not shown), contrary to the cch mutant, which has a strong ABA-insensitive phenotype in stomatal movement (Shen et al., 2006;Fig. 5). These findings support the idea that the N-terminal half of ABAR/CHLH plays a secondary role compared with the C-terminal half, which is consistent with the findings that the N-terminal half does not bind ABA (Fig. 1) but the N-terminal half is still functionally required, at least partly, in the modulation of ABA signaling. Distinct from its regulatory role in ABA signaling, the N-terminal half may have no function, or at least no important function, in chlorophyll biosynthesis, consistent once again with the idea that the ABAR-mediated signaling is independent of chlorophyll biosynthesis.
We previously showed that down-and up-regulation of ABAR/CHLH expression resulted in strong ABA insensitivity and hypersensitivity, respectively, in all of the major ABA responses in Arabidopsis (Shen et al., 2006). The experiment described here provides new, additional evidence for the involvement of ABAR/ CHLH in ABA signaling through both transgenic manipulation and creation of new point mutations in the ABAR/CHLH gene via TILLING. Taken together with the ABA-binding data both previously reported (Shen et al., 2006) and presented in this experiment (Fig. 1), all of the findings support the idea that ABAR/CHLH is an intracellular receptor for ABA.
In this regard, however, we understand the current controversy over the ABA receptor nature of ABAR/ CHLH. Although all of the data we have for ABAR/ CHLH to date are consistent with the essential criteria of an ABA receptor, there still exist a series of crucial questions to be answered in relation to the ABA receptor identity of ABAR/CHLH. For example, al- though ProtoIX was previously shown to be unable to compete with ABA for binding to ABAR/CHLH (Shen et al., 2006), does ABA partly share a binding site with ProtoIX in the ABAR/CHLH molecule because the structure of ABA is similar to that of ProtoIX, which also binds CHLH? This possible unspecific binding, even if trace, would lead to an overestimation or underestimation of the number of ABA-binding sites.
The most important open question is how ABAR/ CHLH transmits ABA signal to downstream regulators to mediate the ABA response, particularly if or how it could transduce the signal to genetically defined, currently well-characterized ABA signaling regulators. To answer these questions is essential to define a bona fide ABA receptor.
Nevertheless, the experimental data obtained here allow us to postulate a working model of the functional domains in the ABAR/CHLH molecule in the regulation of ABA signaling (Fig. 6H). The C-terminal half is likely to play central role in ABA signal perception and signal output, covering the ABA-binding domain and possibly the domains interacting with downstream players to relay the ABA signal. The extreme C terminus is most likely to be a domain responsible for protein/mRNA degradation. The N-terminal half may cover, in addition to a transit peptide, a regulatory center that may be involved in regulation of the functions of the C-terminal half. To test this postulation in the future will provide new insight into the ABAR-mediated ABA perception and downstream signaling.

Is the Large Subunit of Mg-Chelatase Also a Candidate Receptor for ABA in Monocotyledonous Plants?
In contrast to the previous report, in which the ABAbinding activity of the barley XanF was not detected (Mü ller and Hansson, 2009), ABA-affinity chromatography allowed us to reveal that both rice (Oryza sativa) CHLH (OsCHLH) and barley XanF are ABA-binding proteins (Supplemental Figs. S2 and S3).
The ability of ABA binding to XanF and OsCHLH suggests that the two proteins may be involved in ABA signaling, although the determination of the ABA-binding kinetics will be necessary with an inbuffer radiolabeled ABA-binding system optimized for barley and rice to assess if the binding properties meet the essential criteria of ligand-receptor binding. Genetic approaches including transgenic manipulations will aid in testing if the two large subunits of Mgchelatase XanF and OsCHLH modulate ABA signaling in monocotyledonous plants. First, however, it is noteworthy that a possible functional redundancy of XanF genes in barley and OsCHLH in rice may occur, which is different from Arabidopsis, which harbors a single copy of CHLH. Rice has two copies of the CHLH gene, one coding for a full-length CHLH and another for a C-terminal truncated CHLH (Supplemental Fig. S3). We showed that the two OsCHLH proteins both bind ABA and uncovered the possibility that barley may have also two XanFs, as in rice (Supplemental Fig. S2). The Arabidopsis C-terminal half of ABAR/CHLH binds ABA and regulates ABA signaling (Figs. 1-6), which suggests the potential ABA signaling functionality of the smaller CHLH in rice and also possibly in barley. In this case, mutations in only one copy of the CHLH/XanF genes may not be able to alter ABA responsibility. Second, as it was observed in Arabidopsis that the ABAR/CHLH-mediated ABA signaling is independent of chlorophyll biosynthesis, the chlorophyll-deficient mutants do not necessarily have defects in ABA responses. This may be one possible explanation for why the two chlorophyll-deficient mutants of the barley XanF gene showed wild-type ABA responses (Mü ller and Hansson, 2009). Third, we observed that the ABA-insensitive phenotypes of the RNA interference (RNAi) lines of the ABAR/CHLH gene became weaker when the mRNA of ABAR/CHLH decreased to a very low level (in the RNAi mutants with pale yellow leaves), suggesting that a strong feedback effect is involved in the ABAR/CHLHmediated ABA signaling (Shen et al., 2006). This may be explained by a possible up-regulation of other ABA signaling pathways (mediated by other receptors for ABA) when the ABAR/CHLH-mediated signaling pathway is down-regulated to a certain low threshold level. Stomatal response, however, did not show this low-threshold-related phenomenon and displayed a strong ABA insensitivity that was negatively correlated with the ABAR/CHLH mRNA level in these ABAR/CHLH-RNAi lines (Shen et al., 2006). So the mutants of barley with low XanF protein and pale leaves may be subjected to a similar feedback phenomenon as in the severe ABAR/CHLH-RNAi lines of Arabidopsis. Lastly, we do not exclude the possibility that CHLH/XanF may not function in ABA signaling in monocotyledonous plants, probably due to the occurrence of possibly different signaling networks between the monocotyledonous and dicotyledonous plants. An RNAi manipulation resulting in downregulation of the full-length CHLH/XanF and smaller CHLH/XanF may be necessary in barley and rice, respectively, and will aid in clarifying whether the two proteins are involved in ABA signaling.

Plant Materials and Generation of Transgenic Plants
Arabidopsis (Arabidopsis thaliana Col-0) and the cch mutant of the ABAR/ CHLH gene were used in the generation of truncated ABAR/CHLH transgenic plants. Truncated ABAR genes were amplified by PCR from Col-0 cDNA with KOD-plus DNA polymerase (Toyabo) and cloned into the binary vector pCAMBIA1300, which contains the cauliflower mosaic virus 35S promoter and a C-terminal GFP flag. The ABAR cDNA fragments were isolated by PCR using the forward primer 5#-GGACTAGTATGGCTTCGCTTGTGTATT-CTC-3# and reverse primer 5#-GGGGTACCACTCCATCCCACAGTGTT-GGA-3# for the ABARn fragment (corresponding to the 1-1,303 amino acid residue-encoding cDNA sequence); the forward primer 5#-GGACTAGTATG-GACACCAATGACTCACTCAAG-3# and reverse primer 5#-GGGGTACCTC-GATCGATCCCTTCGATCTTG-3# for the ABARc fragment (corresponding to the 311-1,381 amino acid residue-encoding cDNA sequence); the forward primer 5#-GGACTAGTATGGCTTCGCTTGTGTATTCTC-3# and reverse primer 5#-GGGGTACCATCAAGATTACATTGCTTAGC-3# for the N772 fragment (corresponding to the 1-772 amino acid residue-encoding cDNA sequence); the forward primer 5#-GGACTAGTATGCCCATGAGGCT-GCTTTTCTCC-3# and reverse primer 5#-GGGGTACCTCGATCGATCCC-TTCGATCTTG-3# for the C751 fragment (corresponding to the 631-1,381 amino acid residue-encoding cDNA sequence); and the forward primer 5#-GGACTAGTATGCCCATGAGGCTGCTTTTCTCC-3# and reverse primer 5#-GGGGTACCTGTTGTGGGAATAGCCTGAGG-3# for the C370 fragment (corresponding to the 631-999 amino acid residue-encoding cDNA sequence). The C370 fragment was then ligated with a chloroplast transit peptide fragment that was isolated using the forward primer 5#-GGACTAGTATGG-CTTCGCTTGTGTATTCTC-3# and reverse primer 5#-CGGGGCCCTCTA-AGCTCCTCGACCAAGTA-3#. All constructs were confirmed by sequencing. These constructs were introduced into the GV3101 strain of Agrobacterium tumefaciens and transformed into plants by floral infiltration. The homozygous T3 seeds of the transgenic plants were used for analysis. At least five transgenic lines were obtained for each construct, and all of the lines had similar ABArelated phenotypes. The results from one representative line (or three in some cases, as indicated) are presented in this report. The cch mutant was a generous gift from Dr. J. Chory (Salk Institute). The seeds of the abar-2 and abar-3 mutants were obtained from the ABRC as mentioned below. All of the mutants were isolated from ecotype Col-0.
Plants were grown in a growth chamber at 19°C to 20°C on Murashige and Skoog (MS) medium (Sigma) at about 80 mmol photons m 22 s 21 or in compost soil at about 120 mmol photons m 22 s 21 over a 16-h photoperiod. It is particularly noteworthy that the cch mutant was originally identified as a light-sensitive mutant deficient in chlorophyll (Mochizuki et al., 2001) and later shown to be insensitive to ABA in stomatal movement (Shen et al., 2006). So the cch mutant is more susceptible to both strong light and water deficiency than its wild-type background Col-0, and mild-stress conditions to Col-0 are probably severe to the cch mutant. We observed that the good growth status of the cch mutant parental plants is of critical importance to their progeny to display the ABA-related phenotypes: the stressed mutant parental plants had progeny with seeds whose insensitive phenotypes to ABA became weaker in germination and postgermination growth, but the ABA-insensitive phenotype in stomatal movement was not affected. This may be due to a possible upregulation of ABA-responsive mechanisms independent of ABAR-mediated signaling, which may be induced by environmental stresses imposed on this mutant.

Identification of the cch Mutation in the cch Background Transgenic Lines
The single-nucleotide cch mutation in the cch background transgenic lines was identified according to a procedure described previously (Neff et al., 1998). A 257-bp fragment was amplified by PCR, and a HaeIII restriction site was introduced into the wild-type sequence corresponding to the cch mutation site, while the cch mutation in the fragment results in abolishment of this restriction site. The forward primer used was 5#-AGGCTGCTTTTCTC-CAAGTCAGCAAGGC-3# and the reverse primer was 5#-TTGGCATAA-CTTCTCCTCTTTG-3#.
Identification of Two ABAR Mutants, abar-2 and abar-3, from TILLING Lines The abar-2 and abar-3 mutants were identified via the Arabidopsis TILL-ING project (Henikoff et al., 2004). The tilled M3 seeds with predicted deleterious mutations in the ABAR/CHLH gene were provided by the ABRC. Two lines, CS89100 and CS92346 (ABRC stock numbers), were identified as the abar alleles with the altered phenotypes of ABA sensitivity and are designated abar-2 and abar-3, respectively. CS89100 (abar-2) had a C-to-T change at nucleotide 1,042. CS92346 (abar-3) had a C-to-T substitution at nucleotide 548. Homozygous mutant plants were identified by PCR using allele-specific primers designed as described by Konieczny and Ausubel (1993) via restriction site analysis. A 544-bp fragment was amplified by PCR using the forward primer 5#-GCTTGTTAGGACTTTGCCTAAG-3# and reverse primer 5#-TCCACCAACAAGAGCAAAAC-3# for analyzing the abar-2 mutant. An AluI restriction site is present in the wild-type fragment that can be degraded into 150-and 394-bp fragments, while the abar-2 mutation abolished this restriction site in the fragment that cannot be cut. Similarly, a 330-bp fragment was amplified using the forward primer 5#-GAG-GAATTGGCGATTAAAGT-3# and reverse primer 5#-CTGAAGATTATCAG-GAGAGCCTC-3# for analyzing the abar-3 mutant. An MboI restriction site is present in the wild-type fragment that can be cut into 109-and 221-bp fragments, while this restriction site was abolished by the abar-3 mutation. The two alleles were back-crossed three times to remove the erecta allele that was present in the parental background and additional mutations induced by ethyl methanesulfonate.

Expression of ABAR and Truncated ABAR in Escherichia coli
The cDNAs encoding the full-length ABAR and three ABAR fragments were amplified by PCR with the following primers: the forward primer 5#-GGAATTCTATGGCTTCGCTTGTGTATTCTC-3# and reverse primer 5#-ACGCGTCGACTTATCGATCGATCCCTTCGATCTTG-3# for full-length ABAR; the forward primer 5#-GGAATTCTATGGCTTCGCTTGTGTATTCTC-3# and reverse primer 5#-ACGCGTCGACTTAATCAAGATTACATTGCTTAGC-3# for N772; the forward primer 5#-GGAATTCTCCCATGAGGCTGCTTTTC-TCC-3# and reverse primer 5#-ACGCGTCGACTTATCGATCGATCCCTTCGA-TCTTG-3# for C751; and the forward primer 5#-GGAATTCTCCCATGAGG-CTGCTTTTCTCC-3# and reverse primer 5#-ACGCGTCGACTTATGTTGTGG-GAATAGCCTGAGG-3# for C370 (without the sequence encoding the chloroplast transit peptide). The forward primers introduced an EcoRI restriction site, and the reverse primers introduced a SalI restriction site into the fragments. The PCR products were then digested and cloned directly into pET48b between the EcoRI and SalI sites. The fragments in the plasmids were sequenced to check for errors. The recombinant ABAR and ABAR fragments were expressed in E. coli Rosetta gami 2 (DE3; Novagen) strains as a 63His-ABAR (or truncated ABAR) fusion protein. The E. coli strains containing the expression plasmids were grown at 37°C in 1 L of Luria-Bertani medium containing 20 mg mL 21 kanamycin until the optical density at 600 nm of the cultures was 0.6 to 0.8. Protein expression was induced by the addition of isopropyl b-D-thiogalactopyranoside to a final concentration of 1 mM at 16°C and 150 rpm. After 16 h, the cells were lysed and proteins were purified on a Ni 2+ -chelating column as described in the pET system manual.

Anti-Full-Length ABAR Serum Production and Immunobloting
A standard immunization protocol was used to immunize female rabbits. The E. coli-expressed, purified, 63His full-length ABAR fusion protein (2 mg) was injected five times at intervals of 2 weeks. The antiserum was affinity purified. The immunoblotting of the ABAR and truncated ABAR proteins with the anti-ABAR serum was done essentially according to previously described procedures (Shen et al., 2006). Proteins were separated by SDS-PAGE on 10% polyacrylamide gels, and the polypeptides were transferred to nitrocellulose membranes (0.45 mm; Amersham Life Science) in a medium consisting of 25 mM Tris-HCl (pH 8.3), 192 mM Gly, and 20% (v/v) methanol. After rinsing in Tris-buffered saline (TBS) containing 10 mM Tris-HCl (pH 7.5) and 150 mM NaCl, the blotted membranes were preincubated for 3 h in a blocking buffer containing 3% (w/v) bovine serum albumin dissolved in TBS supplemented by 0.05% (v/v) Tween 20 (TBST1) and then incubated with gentle shaking for 2 h at room temperature in appropriate antibodies (diluted 1:2,000 in the blocking buffer). Following extensive washes by TBST1, the membranes were incubated with goat anti-rabbit IgG conjugated with alkaline phosphatase (diluted 1:500 in TBST1) at room temperature for 1 h and then washed with TBST2 (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.1% [v/v] Tween 20) and TBS. The locations of antigenic proteins were visualized by incubating the membranes with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate.
To test the specificity of the purified anti-ABAR serum, we preincubated the antiserum (diluted 1:2,000 in the blocking buffer as described above) with 0.5% (w/v) purified ABAR fusion protein at 0°C for 30 min before use to incubate the protein sample as described, and as a control, 5% (w/v) bovine serum albumin was used in the same preincubation instead of the ABAR protein. This preincubation of the purified anti-ABAR serum with the antigen ABAR protein completely abolished the ability of the antiserum to recognize either E. coli-expressed, purified, ABAR protein or natural ABAR protein from Arabidopsis total protein, but the preincubation of the antiserum with bovine serum albumin did not affect this antiserum-antigen recognition (Supplemental Fig. S4). This test showed that the anti-ABAR serum is specific to ABAR protein.

Preparation of the Arabidopsis Total Protein
The leaves of Arabidopsis were harvested from 3-week-old plants and ground in liquid nitrogen. The sample was then transferred into an Eppendorf tube containing ice-cold extraction buffer (1 mL g 21 sample) consisting of, for ABA-affinity chromatography, 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100, 10% glycerol, 2 mM 1,4-dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 5 mg mL 21 leupeptin, 5 mg mL 21 pepstatin A, and 5 mg mL 21 aprotinin. The buffer for immunoblotting was composed of 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100, 10% glycerol, and 13 protease inhibitor cocktail (Roche). The sample was extracted for 3 h in ice. The extracts were centrifuged for 20 min at 16,000g, the supernatant was transferred to a new Eppendorf tube and centrifuged again at 16,000g for 20 min, and then the concentration of the supernatant was detected by Coomassie Brilliant Blue G-250 (Amresco). The samples were either kept at 0°C for immediate use or frozen and stored at 280°C until use.

ABA Binding Assay with Affinity Chromatography
ABA was linked to EAH-Sepharose 4B via C1 (-COOH) of the ABA molecule according to previously described procedures (Zhang et al., 2002). (6)ABA (1 g; Sigma) dissolved in 60 mL of 50% (w/v) dimethylformamide (Sigma) solution was mixed with 50 mL of drained EAH-Sepharose 4B (GE Healthcare). 1-Ethyle-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (4 g; Sigma) was added into the ABA-EAH-Sepharose 4B solution, the pH of which was adjusted to 8.0 with 1 N NaOH. The ABA-EAH-Sepharose 4B solution was shaken for 20 h at 4°C in the dark. The gel was then washed with 50% (w/v) dimethylformamide and then with both washing buffer I consisting of 0.5 M NaCl and 0.1 M Tris-HCl (pH 8.3) and washing buffer II composed of 0.5 M NaCl and 0.1 M sodium acetate-acetic acid (pH 4.0). The gel was washed again with double distilled water. The coupling amount of ABA to EAH-Sepharose 4B is determined essentially according to Nilsson and Mosbach (1984). ABA-EAH-Sepharose 4B (40 mg) was dissolved in 80% (v/v) glycerol, and then the UVA 252 of the solution was measured with a UV-Photometer (UV-240; Shimadzu) using 80% (v/v) glycerol as the control. The amount of the coupled ABA was calculated according to the standard UV absorbance per millimolar ABA at 252 nm. The tested coupling efficiency, which was the ratio of the coupling amount of ABA to the total amount of the amino groups conjugated to the gel, was approximately 60% to 70%.
All of the procedures described below were done at 4°C. The ABA-linked EAH Sepharose 4B gel (1 mL) was first equilibrated with buffer A solution consisting of 10 mM MES-NaOH (pH 6.5), 150 mM NaCl, 2 mM MgCl 2 , 2 mM CaCl 2 , and 5 mM KCl. The equilibrated ABA-linked Sepharose 4B was then incubated with 0.5 mL of pure E. coli-expressed ABAR protein (2 mg mL 21 ) or 2 mL of Arabidopsis total protein (1 mg mL 21 ) for 60 min in 5 mL of buffer A. The final volume was 6.5 mL. For the competition assays of ABAR binding to the ABA-affinity column, (+)ABA (Sigma) at 50 or 500 nM concentration was added into buffer A for this incubation of the ABA-linked Sepharose 4B gel with ABAR protein. The same amount of ethanol for solubilizing 50 or 500 nM (+)ABA was used instead of (+)ABA as a control. The ABA-linked Sepharose 4B gel mixed with ABAR protein (or with ABA in the competition assays) was then packed into a column of 1.6 3 10 cm for affinity chromatography. The column was first eluted with 150 mM NaCl in buffer A (36 mL, divided into 12 times with 3 mL each) to remove the unspecific bound proteins. ABA-binding proteins were then eluted with the same buffer A containing 100 or 150 mM NaCl and 4 mM (6)ABA (or the same concentration of other potential agonists/antagonists as indicated). The eluting solutions were assayed for immunoblotting with the anti-ABAR serum.

H-Labeled ABA in-Buffer Binding Assay
3 H-labeled ABA binding was performed essentially as described previously (Zhang et al., 2002;Shen et al., 2006) with modifications. We replaced the Dextran T70-coated charcoal with a glass fiber filter to remove free [ 3 H]ABA from the binding medium. We previously used the ABA-binding technique with a filter (Zhang et al., 1999(Zhang et al., , 2001, and this technique was recently used to assay the ABA receptors GTG1 and GTG2 (Pandey et al., 2009).
[ 3 H](+)ABA was made by American Radiolabeled Chemicals (2.37 3 10 12 Bq mmol 21 ; purity, 98.4%). The binding medium was composed of 50 mM MES-NaOH (pH 7.0), 2 mM MgCl 2 , 1 mM CaCl 2 , 1 mM 1,4-dithiothreitol, 250 mM mannitol, and 10 mg mL 21 protease inhibitor cocktail (Sigma). Each binding assay contained 10, 20, 40, or 60 nM [ 3 H]ABA with or without a 1,000-fold molar excess of unlabeled (6)ABA (Sigma), 2 mg of purified E. coliexpressed ABAR/CHLH (or truncated ABAR/CHLH protein as indicated), and binding buffer in a 200-mL total volume. The mixtures were incubated at 25°C for 60 min. The bound and free [ 3 H]ABA were separated, as mentioned above, by filtering the mixture through a GF/F glass fiber filter (Whatman) and washing with 3 mL of ice-cold ABA-binding buffer. The [ 3 H]ABA bound to ABAR protein retained by the filter was then quantified by scintillation counting. The specific binding was determined by subtracting the binding in the presence of a 1,000-fold molar excess of unlabeled (6)ABA (unspecific binding) from total binding [the binding in the absence of unlabeled (6)ABA]. It is noteworthy that the radioactivity retained by the glass fiber filter (the control in the absence of ABAR protein) should be subtracted from the total and unspecific binding. The unspecific binding in all of the assays was less than 10% of the total binding. The ABA-binding activity was expressed as moles of [ 3 H]ABA per mole of protein.

Transient Expression in Arabidopsis Protoplasts
For observation of the subcellular localization of the truncated ABAR (ABARn, C751, the transit peptide-linked C370, and N772), the corresponding cDNA fragments, driven by the cauliflower mosaic virus 35S promoter and downstream tagged by GFP, were obtained by enzymatic degradation from the above-mentioned binary vector pCAMBIA1300-221, which was created for generating transgenic plants and harbors these cDNA fragments linked to the 35S promoter in the 5# end and a C-terminal GFP flag in the 3# end. Each of the 35S promoter-driven and GFP-tagged cDNA fragments was fused to the pMD-19-T vector (Takara, Dalian Division) with the SphI (5# end) and EcoRI (3# end) sites. Protoplasts were isolated from the leaves of 3-to 4-week old plants of Arabidopsis (Col-0) and transiently transformed using polyethylene glycol essentially according to an established protocol (http://genetics.mgh. harvard.edu/sheenweb/). Fluorescence of GFP was observed with a confocal laser scanning microscope (Zeiss; LSM 510 META) after incubation at 23°C for 16 h.

Phenotypic Analysis
Phenotypic analysis was done essentially as described previously (Shen et al., 2006). For the germination assay, approximately 100 seeds each from the wild type (Col-0), mutants, or transgenic mutants were sterilized and planted in triplicate on MS medium (Sigma; product no. M5524; full-strength MS). The medium contained 3% Suc and 0.8% agar (pH 5.9) and was supplemented with or without different concentrations of (6)ABA. The seeds were incubated at 4°C for 3 d before being placed at 20°C under light conditions, and germination (emergence of radicles) was scored at the indicated times.
For the seedling growth experiment, seeds were germinated after stratification on common MS medium and transferred to MS medium supplemented with different concentrations of (6)ABA in the vertical position. The time for transfer was about 45 to 48 h after stratification for the transgenic plants and 48 to 50 h for the assays of the abar-2, abar-3, and cch mutants. This narrow ABA-responsive window of around 48 h in seedling growth was observed for all of the ABAR-related mutants (including transgenic lines), which may be partly associated with ABI5 regulation (Shen et al., 2006). Seedling growth was investigated at the indicated times after the transfer, and the lengths of primary roots were measured using a ruler. Seedling growth was also assessed by directly planting the seeds in ABA-containing MS medium to investigate the response of seedling growth to ABA after germination.
For stomatal aperture assays, 3-week-old leaves were used. To observe ABA-induced stomatal closure, leaves were floated in the buffer containing 50 mM KCl and 10 mM MES-Tris (pH 6.15) under a halogen cold light source (Colo-Parmer) at 200 mmol m 22 s 21 for 2.5 h followed by the addition of different concentrations of (6)ABA. Apertures were recorded on epidermal strips after 2.5 h of further incubation to estimate ABA-induced closure. To study ABA-inhibited stomatal opening, leaves were floated on the same buffer in the dark for 2.5 h before they were transferred to the cold light for 2.5 h in the presence of ABA, and then apertures were determined.

Chlorophyll and Porphyrin Measurements
The contents of chlorophyll, ProtoIX, and Mg-ProtoIX were assayed essentially according to previously described procedures (Mochizuki et al., 2001).