Phosphorylation of serine residue modulates cotton Di19-1 and Di19-2 activities for responding to high salinity stress and abscisic acid signaling

Di19 (drought-induced protein 19) family is a novel type of Cys2/His2 zinc-finger proteins. In this study, we demonstrated that cotton Di19-1 and Di19-2 (GhDi19-1/-2) proteins could be phosphorylated in vitro by the calcium-dependent protein kinase (CDPK). Mutation of Ser to Ala in N-terminus of GhDi19-1/-2 led to the altered subcellular localization of the two proteins, but the constitutively activated form (Ser was mutated to Asp) of GhDi19-1/-2 still showed the nuclear localization. GhDi19-1/-2 overexpression transgenic Arabidopsis seedlings displayed the hypersensitivity to high salinity and abscisic acid (ABA). However, Ser site-mutated GhDi19-1(S116A) and GhDi19-2(S114A), and Ser and Thr double sites-mutated GhDi19-1(S/T-A/A) and GhDi19-2(S/T-A/A) transgenic Arabidopsis did not show the salt- and ABA-hypersensitive phenotypes. In contrast, overexpression of Thr site-mutated GhDi19-1(T114A) and GhDi19-2(T112A) in Arabidopsis still resulted in salt- and ABA-hypersensitivity phenotypes, like GhDi19-1/-2 transgenic lines. Overexpression of GhDi19-1/-2 and their constitutively activated forms in Atcpk11 background could recover the salt- and ABA-insensitive phenotype of the mutant. Thus, our results demonstrated that Ser phosphorylation (not Thr phosphorylation) is crucial for functionally activating GhDi19-1/-2 in response to salt stress and ABA signaling during early plant development, and GhDi19-1/-2 proteins may be downstream targets of CDPKs in ABA signal pathway.


GhDi19-1/-2 is phosphorylated by calcium-dependent protein kinase (CDPK) in a Ca 2+dependent manner.
Sequence analysis showed that both GhDi19-1 and GhDi19-2 contain Ser/Thr kinase phosphorylation sites in the conserved nuclear localization signal region next to the two zinc finger domains (GhDi19-1: Thr114 and Ser116; GhDi19-2: Thr112 and Ser114), and the predicted Ser phosphorylation is potentially stronger than the Thr phosphorylation (http://myhits.isb-sib.ch/cgi-bin/motif_scan). Then, we employed an in vitro kinase assay to determine whether GhDi19-1 and GhDi19-2 are phosphorylated by a kinase (such as AtCPK11). As shown in Fig. 1a, the S 116 of GhDi19-1 was mutated to A 116 [GhDi19-1(S116A)], and the S 114 of GhDi19-2 was mutated to A 114 [GhDi19-2(S114A)]. The in vitro kinase assay revealed that the recombinant AtCPK11 phosphorylated GhDi19-1 and GhDi19-2 in the presence of Ca 2+ . Additionally, weaker phosphorylation of GhDi19-1(S116A) was detected in the presence of Ca 2+ , and no phosphorylation of GhDi19-2(S114A) was found with or without Ca 2+ (Fig. 1b). The above results indicated that GhDi19-1/-2 is phosphorylated in vitro by a kinase (such as AtCPK11) in a Ca 2+ -dependent manner, and the 116th Ser of GhDi19-1 or the 114th Ser of GhDi19-2 is required for the phosphorylation.
Ser site is crucial for activating GhDi19-1 and GhDi19-2 proteins in response to salt stress and ABA signaling. Our previous study indicated that overexpression of the GhDi19-1 and GhDi19-2 genes in Arabidopsis enhances plant sensitivity to salt stress in ABA-dependent manner 34 . To further investigate the mechanism of GhDi19-1 and GhDi19-2 involving in response to salt stress and ABA signaling, the S 116 of GhDi19-1 was mutated to A 116 , and the S 114 of GhDi19-2 was mutated to A 114 . The coding sequences of GhDi19-1(S116A) and GhDi19-2(S114A) were inserted into plant expression vector pBI121 driven by CaMV 35S promoter, and introduced into Arabidopsis, respectively. Homozygotes of T3 generation of GhDi19-1 overexpression transgenic lines (L4 and L19), GhDi19-1(S116A) overexpression transgenic lines (L2 and L8), GhDi19-2 overexpression transgenic lines (L3 and L6) and GhDi19-2(S114A) overexpression transgenic lines (L1 and L4) were analyzed by PCR to ensure the presence of the respective transgene and by quantitative RT-PCR to confirm transgene expression (Fig. 3a).
Seeds of GhDi19-1, GhDi19-1(S116A), GhDi19-2 and GhDi19-2(S114A) overexpression transgenic lines and wild type were sowed on MS medium with or without NaCl and ABA, respectively. On MS medium, all the seeds from wild type and GhDi19-1/-2 variants were able to fully germinate (≤ 4 days) after sowing, and there was no significant difference in seed germination rate between each other. In the presence of 150 mM NaCl, however, the GhDi19-1 and GhDi19-2 transgenic seeds germinated much later than those of wild type. After 4 day, approximately 75% of wild type seeds germinated, but only about 35%-40% of GhDi19-1 and GhDi19-2 transgenic seeds germinated. After 6 days, germination rate of wild type seeds reached to about 90%, but both GhDi19-1 and GhDi19-2 transgenic seeds reached to only approximately 60%. In contrast, seed germination rates of GhDi19-1(S116A) and GhDi19-2(S114A) transgenic lines were similar to those of wild type with NaCl treatment. On MS medium supplemented with 0.8 μ M ABA, seed germination rate of both GhDi19-1 and GhDi19-2 transgenic lines was significantly lower than that of wild-type. About 60% of wild type seeds, but only 40% of GhDi19-1 and 30% of GhDi19-2 overexpression transgenic seeds germinated after 4 days. On the contrary, seed germination rate of GhDi19-1(S116A) and GhDi19-2(S114A) overexpression transgenic lines was as same as those of wild type under ABA treatment (Fig. 4a). To assess effects of Ser site on GhDi19-1 and GhDi19-2 involving in response to salt and ABA during early seedling development, we used two approaches in the experiments. One way is that seeds were directly planted on MS medium containing 150 mM NaCl or 1 μ M ABA for determining growth status of the seedlings after germination (Fig. 4b). Another is that seeds were germinated on MS medium for 48 h after stratification and then transferred onto MS medium containing 150 mM NaCl and 5 μ M ABA in the vertical position (Fig. 4d). The results obtained with these two approaches were similar. There was no significant difference among the GhDi19-1/-2 and their site-mutated transgenic lines and wild type when seedlings grew on MS medium. However, the GhDi19-1 and GhDi19-2 overexpression seedlings were more sensitive to salt and ABA than wild type, but growth status of GhDi19-1(S116A) and GhDi19-2(S114A) transgenic seedlings was almost as same as those of wild type (Fig. 4b).

Thr site is inessential for GhDi19-1 and GhDi19-2 in response to salt stress and ABA signaling.
To understand whether Thr phosphorylation site also play an important role in GhDi19-1 and GhDi19-2 involving in response to salt and ABA signaling, the T 114 of GhDi19-1 was mutated to A 114 , the T 112 of GhDi19-2 was mutated to A 112 , both T 114 and S 116 of GhDi19-1 were mutated to A, and both T 112 and S 114 of GhDi19-2 were mutated to A. The coding sequences of GhDi19-1(T114A), GhDi19-1(S/T-A/A), GhDi19-2(T112A) and GhDi19-2(S/T-A/A) were inserted into pBI121 vector, and introduced into Arabidopsis, respectively. Homozygotes of T3 generation of the respective overexpression transgenic lines were analyzed by PCR to ensure confirm the presence of the respective transgene and by quantitative RT-PCR to confirm that the transgene was expressed (Fig. 3a). Two transgenic lines (L3 and L5) with higher GhDi19-1(T114A) expression, two lines (L1 and L6) with higher GhDi19-1(S/T-A/A) expression, two lines (L1 and L4) with higher GhDi19-2(T112A) expression and two lines (L2 and L7) with higher GhDi19-2(S/T-A/A) expression were selected for analyzing their phenotypes under different stresses.
When seeds sowed on MS medium, there was no significant difference in seed germination between all the transgenic lines and wild type under normal conditions (Fig. 5a,d,g, j). When sowing on MS medium containing 150 mM NaCl or 0.8 μ M ABA, however, the GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seeds germinated much later than those of wild type, and showed salt-and ABA-sensitivity. In the presence of 150 mM NaCl, only about 35%-40% of the GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seeds germinated, while approximately 65% of wild type germinated after 3 days. And germination rate of the transgenic seeds only reached to 60%, while wild type was about 80% after five days of germination (Fig. 5b,h). In addition, seed germination rate of the GhDi19-1(T114A) and GhDi19-2(T112A) transgenic lines was reduced much more than that of wild type under ABA treatment. When treated with 0.8 μ M ABA, the seed germination rates of GhDi19-1, GhDi19-2, GhDi19-1(T114A) and GhDi19-2(T112A) transgenic lines were decreased to approximately 40%, whereas wild type retained 75% seed germination rate after 4 days (Fig. 5c,i). On the other hand, there was no significant difference in seed germination rate between GhDi19-1(S/T-A/A) and GhDi19-2(S/T-A/A) transgenic lines and wild type under NaCl and ABA treatments (Fig. 5e,k,f,l).
We also calculated cotyledon greening of the transgenic seedlings and wild type under salt stress. As shown in Fig. 6a, both GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seedlings growing on MS containing 150 mM NaCl were more sensitive than wild type. Cotyledon greening of GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seedlings was drastically inhibited by NaCl treatment, compared with that of wild type. Similarly, cotyledon greening of GhDi19-1(T114A) and GhDi19-2(T112A) transgenic plants was also inhibited more than that of wild type. In the presence of 1 μ M ABA, approximately 20% and 25% of the GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seedlings with expanded and turned-green cotyledons was observed, respectively, while wild type reached to 60% cotyledon greening rate. In contrast, GhDi19-1(S/T-A/A) and GhDi19-2(S/T-A/A) transgenic plants showed the similar phenotype to wild-type under NaCl and ABA treatments (Fig. 6b).
Additionally, seeds of all the transgenic Arabidopsis lines and wild type germinated on MS medium for 48 h, and then were transferred onto NaCl-and ABA-containing MS medium for investigating growth of the seedlings responding to high salinity and ABA (Fig. 6c). When transferred onto MS plates, root growth of all the transgenic seedlings was almost as same as that of wild type. However, when the seedlings were transferred onto MS medium supplemented with 150 mM NaCl or 5 μ M ABA for several days, primary roots growth of the GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seedlings was more significantly inhibited than that of wild type by NaCl or ABA. The roots of GhDi19-1(T114A) and GhDi19-2(T112A) transgenic seedlings were shorter than those of wild type under NaCl and ABA treatments. However, there was no significant difference in root length between GhDi19-1(S/T-A/A) and GhDi19-2(S/T-A/A) transgenic lines and wild type under NaCl and ABA treatments (Fig. 6c). These results demonstrated that overexpression of the mutated GhDi19-1(T114A) and GhDi19-2(T112A) genes in Arabidopsis still resulted in the transgenic plants salt-and ABA-hypersensitivity, like GhDi19-1/-2 transgenic lines, suggesting that the Thr site is inessential for GhDi19-1 and GhDi19-2 proteins involving in response to salt stress and ABA signaling during seed germination and early seedling development.
During early seedling development, no significant difference was observed among the transgenic variants and wild type when seedlings grew on MS medium. However, the seedlings of Atcpk11-2 mutant grew better than those of wild type on ABA-and NaCl-containing medium, but the GhDi19-1(S116D)/cpk11 and GhDi19-2(S114D)/cpk11 transgenic seedlings were more sensitive to ABA and salt than wild type, while growth status of GhDi19-1/cpk11 and GhDi19-2/cpk11 transgenic seedlings was similar to wild type under ABA and NaCl treatments (Fig. 7b). Under ABA and NaCl treatments, cotyledon greening rate of Atcpk11-2 mutant was significantly higher than that of wild type, while cotyledon greening rate of GhDi19-1/cpk11 and GhDi19-2/cpk11 transgenic lines was similar to wild type. In contrast, cotyledon greening of GhDi19-1(S116D)/cpk11 and GhDi19-2(S114D)/ cpk11 transgenic lines was inhibited much more by ABA and NaCl, compared with wild type (Fig. 7c).
When the seedlings were transferred onto MS plates in the vertical position without any stress treatments, the root growth of all transgenic variants was almost as same as those of wild type. However, when the seedlings were transferred onto MS medium supplemented with 5 μ M ABA or 150 mM NaCl for several days, primary root growth of Atcpk11-2 mutant was much better than that of wild type, and root growth of GhDi19-1/cpk11 and GhDi19-2/cpk11 transgenic lines was similar to wild type, whereas root growth of the GhDi19-1(S116D)/cpk11 and GhDi19-2(S114D)/cpk11 transgenic seedlings was inhibited much more than that of wild type (Fig. 7d). Roots of Atcpk11-2 mutant were longer than those of wild type, and root length of the GhDi19-1/cpk11 and GhDi19-2/ cpk11 transgenic lines was similar to wild type, but roots of GhDi19-1(S116D)/cpk11 and GhDi19-2(S114D)/cpk11 transgenic seedlings were shorter than those of wild type under ABA and NaCl treatments (Fig. 7e). These results indicated that overexpression of GhDi19-1/-2 in Atcpk11 background could recover the salt-and ABA-insensitive phenotype of the mutant, implying the other CDPK (besides CPK11) may also activate GhDi19-1/-2 in the transgenic Arabidopsis plants, and Ser phosphorylation is important for functionally activating GhDi19-1/-2 in response to salt stress and ABA signaling during seed germination and early seedling development.

Discussion
A large number of genes encoding receptors, kinases, transcription factors and other signal molecules in plants are induced after exposure to various abiotic stresses 35,36 . These genes function in various ways to confer plants stress tolerance [37][38][39] . Due to much low similarity of the different Di19-related proteins outside the zinc finger domain, Di19 proteins may play diverse roles in plant development and in response to abiotic stresses 32 . Previous studies revealed that GhDi19-1 and GhDi19-2 are involved in plant response to salt stress and ABA signaling 34 . In this study, we further investigated the mechanism of GhDi19-1 and GhDi19-2 involving in response to salt stress and ABA signaling, and found the Ser site is essential for GhDi19-1 and GhDi19-2 proteins involving in plant response to salt and ABA during seed germination and early seedling development.
Modification of phosphorylation by kinases and of dephosphorylation by phosphatases is an important physiological process as plants respond to abiotic stresses and ABA signaling. Some ABA/stress signaling regulators are modulated at the post-translational level by changing their phosphorylation states [40][41][42][43][44] . Calcium-dependent protein kinases (CDPKs) are unique serine/threonine kinases in plants responding in abiotic stresses 45 . Their multifunctionality and signaling specificity may be conferred by their ability to phosphorylate different substrates. Arabidopsis CPK4 and CPK11 positively regulate ABA signaling via phosphorylating downstream ABA-responsive transcription factors in plants 11 . Furthermore, previous studies revealed that most of seven Arabidopsis AtDi19s are phosphorylated by CPK3 and CPK11 in vitro in a Ca +2 -dependent manner 33 . Likewise, both GhDi19-1 and GhDi19-2 contain a conserved nuclear localization signal (NLS), in which two putative kinase phosphorylation sites (Thr114 and Ser116 in GhDi19-1, and Thr112 and Ser114 in GhDi19-2, respectively) are located 34 . In this study, our data indicated that GhDi19-1/-2 could be phosphorylated in vitro by CPK11 in a Ca 2+ -dependent manner, and the Ser site of GhDi19-1/-2 is required for phosphorylation and normal subcellular localization of GhDi19-1 and GhDi19-2 proteins. Previous study reported that CPK11 phosphorylates Ser104 and Ser107 sites within the AtDi19-1 bipartite NLS probably, but effects of Ser phosphorylation at different sites on Di19 proteins involving in plant response to salt and ABA are unknown so far 33 . In this study, our data demonstrated the Ser site (but not Thr site) is crucial for functionally activation of cotton Di19-1/-2 in response to salt stress and ABA signaling.
AtCPK11 interacts with AtDi19-1 protein in the cell nucleus, and increases AtDi19-1 transactivation of PR1, PR2, and PR5 expression 31 . Furthermore, AtCPK4 and AtCPK11 overexpression transgenic plants display the hypersensitivity to ABA during seeds germination and seedlings growth 11 . Similar phenotype was also observed in the GhDi19-1/-2 overexpression transgenic Arabidopsis 34 . These findings indicated that there is a specific functional relationship of cotton Di19s and CDPKs. Additionally, overexpression of GhDi19-1/-2 in Atcpk11 could recover the salt-and ABA-insensitive phenotype of the mutant, implying the other kinases (besides CPK11) may also activate GhDi19-1/-2 in the transgenic Arabidopsis plants.
Based on the data presented in this study, we thought that Di19 proteins may be potential downstream targets of CDPKs in ABA signal pathway during early plant growth and development. When plants are exposed to abiotic stress stimuli (such as drought and high salinity etc.), concentrations of intracellular ABA and Ca 2+ may be increased in plants for responding the environmental stress signaling. Subsequently, Ser/Thr kinases (e.g. CDPK) are activated by ABA and in turn the activated kinases phosphorylate Di19 proteins at Ser site in the cell nucleus. Finally, the activated Di19 proteins transduce the signals to downstream ABA-and stress-responsive genes, thereby promoting plants response to abiotic stresses.  thaliana (Columbia ecotype), including wild type, mutant and transgenic lines, were surface-sterilized with 75% ethanol for 1 min and 10% NaClO for 5 min, followed by washing with sterile water. The sterilized Arabidopsis seeds were plated on Murashige and Skoog (MS) medium. After stratification at 4 °C for 2 days, the plates were transferred to a plant growth incubator (Sanyo, Osaka, Japan) for seed germination (16 h light/8 h dark at 22 °C) ten days later, seedlings were transplanted in soil and grown in a growth room under the conditions of 16 h light/8 h dark cycle at 22-24 °C. Tissues were derived from these seedlings for RNA extraction.

Generation of transgenic plant and mutant lines. The coding sequences of cotton GhDi19-1 and
GhDi19-2 genes, amplified from its cDNA by PCR with the proofreading pfu DNA polymerase, were cloned into pBI121 vector with BamH I/Sac I sites to replace the GUS gene, respectively 34 .
All generated binary vectors were transformed into Agrobacterium turmefaciens strain GV3101. Arabidopsis transformation was performed by the floral dip method 46 , and transformants were identified on selective medium with kanamycin or hygromycin.
Phenotypic analysis of the transgenic Arabidopsis plants. Homozygous plants (T3 and T4 generations) of the GhDi19-1/-2 transgenic lines and their site-mutations were used for phenotypic analysis, employing wild type, mutant and the transgenic line harboring the "empty vector" (i.e. pBI121-eGFP) as controls. We first tested a series of ABA and NaCl concentrations in the pre-experiments, and determined which concentration of abscisic acid (ABA) or NaCl is suitable to the respective experiments. Then, seeds of wild type and independent transgenic lines, mutants, or transgenic mutants were germinated on MS medium supplemented without or with 150 mM NaCl, 0.8 and l μ M ABA, respectively. The seeds were incubated at 4 °C for 2 days and then transferred into a plant growth incubator at 22 °C conditions (16 h light/8 h dark). Seeds were considered successfully germinated when radicles completely penetrated the seed coats. Germination rate and proportion of seedlings with opened green cotyledons were expressed as a percentage of the total number of seeds plated.
The seedling growth experiments were performed as described previously 11 . Seeds were germinated after stratification on MS medium for 48 h and then transferred onto MS medium containing 150 mM NaCl and 5 μ M ABA in the vertical position. The growth status of seedling was recorded for 10 days and the length of seedling primary roots was measured at tenth day after the transfer. In addition, seedling growth was also assessed by directly planting the seeds on NaCl-or ABA-containing MS medium to investigate the response of seedling growth to salt or ABA for 10 days.