- © 2017 American Society of Plant Biologists. All Rights Reserved.
Abstract
Brassinosteroids (BRs) are a class of steroid hormones regulating multiple aspects of plant growth, development, and adaptation. Compared with extensive studies in Arabidopsis (Arabidopsis thaliana), the mechanism of BR signaling in rice (Oryza sativa) is less understood. Here, we identified OsWRKY53, a transcription factor involved in defense responses, as an important regulator of rice BR signaling. Phenotypic analyses showed that OsWRKY53 overexpression led to enlarged leaf angles and increased grain size, in contrast to the erect leaves and smaller seeds in oswrky53 mutant. In addition, the oswrky53 exhibited decreased BR sensitivity, whereas OsWRKY53 overexpression plants were hypersensitive to BR, suggesting that OsWRKY53 positively regulates rice BR signaling. Moreover, we show that OsWRKY53 can interact with and be phosphorylated by the OsMAPKK4-OsMAPK6 cascade, and the phosphorylation is required for the biological function of OsWRKY53 in regulating BR responses. Furthermore, we found that BR promotes OsWRKY53 protein accumulation but represses OsWRKY53 transcript level. Taken together, this study revealed the novel role of OsWRKY53 as a regulator of rice BR signaling and also suggested a potential role of OsWRKY53 in mediating the cross talk between the hormone and other signaling pathways.
Brassinosteroids (BRs) are a group of plant-specific steroidal hormones that play various roles in plant growth, development, and stress responses. In the past decades, extensive studies have identified numerous BR-signaling components to establish a signaling network and further provided a global view of BR function in the model plant Arabidopsis (Arabidopsis thaliana; Kim and Wang, 2010). In brief, BR is recognized by the membrane-localized receptor BR insensitive1 (BRI1) and its coreceptor BRI1-associated receptor kinase1 (BAK1) and forms the BRI1-BR-BAK1 complex (Li and Chory, 1997; Li et al., 2002; Sun et al., 2013). Transphosphorylation between BRI1 and BAK1 activates BRI1, which then phosphorylates cytoplasmic kinase BSKs (BR signaling kinases) and Constitutive differential growth1, and activation of BSKs/Constitutive differential growth1 leads to phosphorylation and activation of the protein phosphatase bri1-suppressor1 (Tang et al., 2008; Wang et al., 2008; Kim et al., 2011). bri1-suppressor1 dephosphorylates and inactivates the GSK3/Shaggy-like kinase BR insensitive2 (BIN2; Kim et al., 2009). Therefore, in the presence of BR, BIN2 is inhibited, which allows Brassinazole-resistant1 (BZR1) and BRI1 EMS suppressor1 (BES1) to be dephosphorylated by Protein phosphatase 2A (Tang et al., 2011). Finally, dephosphorylated BZR1 and BES1 regulate the expression of numerous BR-responsive genes through binding to a BR response element or E-box cis-element (Yin et al., 2005; Sun et al., 2010; Yu et al., 2011).
In contrast to the tremendous progress of BR signaling in Arabidopsis, relatively fewer components have been characterized in rice (Zhang et al., 2014). So far, many known rice BR signaling components (e.g. OsBRI1, OsBAK1, OsGSK2, OsBZR1, and OsBSK) have orthologs in Arabidopsis and served conserved functions (Yamamuro et al., 2000; Bai et al., 2007; Li et al., 2009; Tong et al., 2012; Zhang et al., 2016). However, some components, including Oryza sativa Dwarf and low tillering (OsDLT), Taihu dwarf1, and Leaf and tiller angle increased controller have no counterpart identified in Arabidopsis, implying that a rice-specific BR-signaling pathway might exist (Tong et al., 2009; Zhang et al., 2012; Hu et al., 2013). In addition, helix-loop-helix proteins such as Oryza sativa Brassinsteroid upregulated1 (OsBU1), Brassinsteroid upregulated like1, and Increased laminar inclination1 can positively regulate rice BR signaling (Tanaka et al., 2009; Zhang et al., 2009; Jang et al., 2017). Reduced leaf angle1 (RLA1)/Small organ size1 was characterized as a positive regulator of BR signaling, which can form a complex with OsBZR1 and OsDLT to coregulate the expression of downstream genes (Hirano et al., 2017; Qiao et al., 2017). Very recently, Enhanced leaf inclination and tiller number1, a receptor-like protein, was shown to promote BR signaling through interacting with and suppressing the degradation of OsBRI1 (Yang et al., 2017).
Plants WRKY transcription factors contain one or two conserved WRKYGQK sequences followed by a C2H2 or C2HC zinc finger motif. Accumulating evidence revealed that the WRKY proteins play diverse roles in responses to biotic and abiotic stresses and are involved in various processes of plant growth and development by regulating the expression of target genes via binding to the W-box cis-element (Rushton et al., 2010). In rice, WRKY family has at least 102 members, and only a few members have been functionally characterized (Xie et al., 2005; Sun et al., 2014). Most of the identified rice WRKY members are involved in plant biotic stress response, including OsWRKY70, OsWRKY53, OsWRKY13, OsWRKY45, and OsWRKY28 (Chujo et al., 2007; Qiu et al., 2007; Shimono et al., 2007; Tao et al., 2009; Chujo et al., 2013; Hu et al., 2015; Ma et al., 2015). Recently, OsWRKY70 and OsWRKY53 were proposed to function in tradeoff mechanisms between biotic stress response and growth; however, the mechanism by which they regulate plant growth remains elusive (Hu et al., 2015; Ma et al., 2015).
MAPK cascade is comprised of three components, MAPK kinase kinases, MAPK kinases, and MAPKs, and has pivotal roles in plant innate immunity (Ishihama et al., 2011; Adachi et al., 2015). Several group Ia members of WRKY proteins have been shown as direct downstream targets of MAPK cascade. MAPK can interact with and phosphorylate group Ia WRKYs via five conserved Ser in SP clusters (Clustered Prodirected Ser residues), and phosphorylation of WRKY proteins by MAPK can enhance the DNA binding activity or transcriptional activity of WRKY proteins (Qiu et al., 2008; Ishihama et al., 2011; Shen et al., 2012; Chujo et al., 2014; Yoo et al., 2014). In rice, the mutation of either OsMAPKK4 or OsMAPK6 leads to BR-deficient phenotypes (dwarfism, erect leaf, and smaller grain size), decreased BR sensitivity, and disrupted expression of BR-related genes, which suggested that the OsMAPKK4-OsMAPK6 cascade is involved in rice BR signaling (Duan et al., 2014; Liu et al., 2015). However, their downstream targets remain elusive.
Here, we show that OsWRKY53 plays a positive role in rice BR signaling and acts downstream of the OsMAPKK4-MAPK6 cascade. Overexpression and knockout of OsWRKY53 led to contrasting BR-related phenotypes. Genetic analysis showed that OsWRKY53 acts downstream of OsBRI1 receptor. In addition, we show that the phosphorylation of OsWRKY53 by OsMAPKK4-MAPK6 is critical for its biological function. Moreover, OsWRKY53 protein is enhanced by BR and can repress its own expression. Taken together, our results suggest that OsWRKY53 is a novel fine tuner of rice BR signaling.
RESULTS
Characterization of OsWRKY53 Overexpression Lines
We previously generated a rice mutant library overexpressing different transcription factors directed by the maize (Zea mays) Ubiquitin promoter. A number of lines carrying OsWRKY53 (Os05g0343400) showed obviously increased leaf angles (described in detail below) and were chosen for further study.
Generally, OsWRKY53 overexpression lines (OsWRKY53-OEs) showed greatly enlarged leaf angles and dwarfism (Fig. 1, A–C). At the three-leaf stage, the angles of the first leaf in the wild type were about 30 degrees, while the angles of the parallel leaf in OsWRKY53-OE reached almost 60 degrees (Supplemental Fig. S1, A and B). With the growth of plant, the larger leaf angles of OsWRKY53-OE appeared to be more evident. At heading stage, the angles of flag leaf in OsWRKY53-OE were about −120 degrees, which is significantly larger than that of wild type at the same stage (Fig. 1, B and C). It has been shown that enlarged leaf inclination is mainly associated with abnormal development of lamina joint (Sun et al., 2015). Morphological observation revealed that the collar length of adaxial surface in OsWRKY53-OE is significantly increased compared with that in wild type (Supplemental Fig. S1C). Scanning electron microscopic observation of lamina joint showed that the adaxial surface in OsWRKY53-OE is not smooth and contains some protuberance compared with that in wild type (Supplemental Fig. S1D). Further cytological observation showed that the cell length of adaxial surface in OsWRKY53-OE is markedly longer than that in wild type (Supplemental Fig. S1, E and F), which contributed to the enlarged leaf angles of OsWRKY53-OE.
Overexpression of OsWRKY53 leads to enlarged leaf angle and increased grain size. A, The gross morphology of OsWRKY53-OE and wild type (WT) at heading stage. B, The lamina joint of flag leaf in WT and OsWRKY53-OE. C, Quantification of lamina angle of the flag leaf. Data are means ± se (n = 20). (D) The grain phenotype of OsWRKY53-OE and WT (scale bar = 5 mm). E and F, Quantification of grain length (E) and grain width (F), respectively. Data are means ± se (n = 50). G, Scanning electron microscopic observation of spikelet of OsWRKY53-OE and WT. The squares are positions for observation. The palea and lemma of OsWRKY53-OE and WT were observed, respectively (scale bar = 50 μm). H and I, Quantification of cell length (H) and cell width (I) in G were shown as means ± se (n = 50). P values were calculated by Student’s t test; *P < 0.05, and **P < 0.01.
In addition, compared with wild type, both the grain length and grain width of OsWRKY53-OE increased significantly (Fig. 1, D–F; Supplemental Fig. S11, A–C). Scanning electron microscopic observation of the spikelet hull showed that the epidermal cells of both palea and lemma in the OsWRKY53-OE are much longer than that in wild type, indicating that increased grain size is mainly attributed to the cell enlargement (Fig. 1, G–I). In agreement with this result, a number of genes associated with cell expansion were up-regulated in the panicles of the OsWRKY53-OE (Supplemental Fig. S2). Moreover, compared with wild type, the leaves of OsWRKY53-OE appeared to be pale green (Supplemental Fig. S3), and the plant height was also reduced (Supplemental Fig. S4D).
Most of the independent OsWRKY53-OEs showed similar phenotypes as described above (Supplemental Fig. S4, A and B), thus excluding the possibility that the phenotypes are caused by mutations of other endogenous genes coming from T-DNA insertion or tissue culture. We also showed the expression of OsWRKY53 is indeed overexpressed in three representative OsWRKY53-OE plants (Supplemental Fig. S4B). Further immunoblotting analyses confirmed the accumulation of OsWRKY53 proteins in these OsWRKY53-OEs (Supplemental Fig. S4C). Taken together, we concluded that the overexpression of OsWRKY53 leads to enlarged leaf angles and increased grain size in rice.
OsWRKY53 Positively Regulates BR Signaling in Rice
The increased leaf angles and enlarged grain size in OsWRKY53-OEs resembled many of the typical enhanced-BR-signaling mutants, such as osbzr1-d, GSK2-RNAi, OsBU1-OE, and mRLA1-OE (Tanaka et al., 2009; Tong et al., 2012; Qiao et al., 2017). Therefore, we hypothesized that OsWRKY53 might be involved in rice BR signaling. We then performed the BR-induced lamina inclination assay to test the BR sensitivity of OsWRKY53-OE in response to different concentrations of 24-epibrassinolide (24-epiBL), an active form of BRs (Tanabe et al., 2005). The results showed that lamina bending of OsWRKY53-OE was obviously much more sensitive than that of wild type (Fig. 2, A and B). After incubation in 10 nm 24-epiBL for 3 d, leaf angles in OsWRKY53-OE reached around ∼140 degrees, dramatically higher than those in wild type, which were about ∼70 degrees (Fig. 2, A and B). This result strongly suggested that overexpression of OsWRKY53 leads to enhanced BR responses.
OsWRKY53-OE is hypersensitive to 24-epiBL treatment. A The leaf inclination of OsWRKY53-OE and wild type (WT) in the presence of indicated concentration of 24-epiBL. B, Statistical analysis of leaf inclination in A, data are means ± se (n = 20). C, The relative expression of BR biosynthesis genes D2, OsDWF4, and D11 in WT and OsWRKY53-OE. The expression level in WT was set as “1”; data were shown as means ± se (n = 3). D, The relative expression of BR-signaling pathway genes OsBU1 and OsXTR1 in WT and OsWRKY53-OE. The expression level in WT was set as “1”; data were shown as means ± se (n = 3).
It is well known that expressions of BR biosynthesis genes are usually negatively feedback regulated by BR signaling in both rice and Arabidopsis (Tong et al., 2012; Zhang et al., 2016; Qiao et al., 2017). The enhancement of BR signaling in OsWRKY53-OE prompted us to investigate whether OsWRKY53 is involved in this process. We analyzed the expression of BR biosynthesis genes, including D2, OsDWF4, and D11, and found all of them have significantly decreased expression in OsWRKY53-OE plants compared with wild type (Fig. 2C), indicating that OsWRKY53 is involved in feedback inhibition of BR biosynthesis genes. In addition, we checked the expression of several BR-responsive genes and found that the expression of OsBU1 and OsXTR1 was indeed increased in OsWRKY53-OE plants (Fig. 2D). Taken together, these results provided strong evidence that OsWRKY53 positively regulates BR signaling in rice.
Characterization of oswrky53 Mutant
To further confirm the function of OsWRKY53 in regulating BR signaling, we generated oswrky53 mutant via CRISPR/Cas9-mediated genome-editing technology. Two independent oswrky53 mutant alleles, oswrky53-1 and oswrky53-2, were identified, and the mutation sites were characterized by DNA sequencing (Fig. 3A; Supplemental Fig. S5). In contrast to OsWRKY53-OE, the leaves of oswrky53 were more erect than those of wild type (Fig. 3, B–D). The plant height of oswrky53 was also slightly decreased (Fig. 3E). In addition, the mutants produced obviously smaller grains, with reduced seed length and seed width (Fig. 3, F–H, Supplemental Fig. S11, D–F). Moreover, oswrky53 were dark green compared with wild type (Supplemental Fig. S6). These phenotypes of oswrky53 were similar to those of typical BR-deficient or BR-defective mutants, such as d11, d61, and GSK2-OE (Yamamuro et al., 2000; Tanabe et al., 2005; Tong et al., 2009; Tong et al., 2012; Qiao et al., 2017). Furthermore, a lamina-bending assay showed that oswrky53 is hyposensitive to BR, which is opposite to OsWRKY53-OE (Fig. 3, I and J). Together, oswrky53 exhibited BR-deficient phenotypes and decreased BR response, which further supported our hypothesis that OsWRKY53 plays positive roles in regulating BR signaling in rice.
oswrky53 mutant show decreased sensitive to BR. A, Identification of oswrky53 mutants. ATG and TGA are initiation codon and stop codon, respectively. B, The gross morphology of oswrky53 mutants and wild type (WT) at heading stage. C and D, The phenotype (C) and statistical analysis (D) of leaf angle; the flag leaf was the first leaf. Data were shown as means ± se (n = 20). E, The plant height of oswrky53 mutants and WT (scale bar = 10 cm). Data were shown as means ± se (n = 20). F, The grain phenotype of oswrky53 mutants and WT (scale bar = 5 mm). G and H, Quantification of grain length (G) and grain width (H) of oswrky53 mutants and WT. Data were shown as means ± se (n = 50). I, The leaf inclination of oswrky53 and WT in the presence of indicated concentration of 24-epiBL. J, Statistical analysis of leaf inclination in I. Data are means ± se (n = 20). P values were calculated by Student’s t test; *P < 0.05, and **P < 0.01).
OsWRKY53 Acts Downstream of OsBRI1
To analyze whether OsWRKY53 functions in the primary BR-signaling pathway, we crossed OsWRKY53-OE with d61-2, a weak allele of the BR receptor OsBRI1 mutant (Yamamuro et al., 2000), and identified the double mutant (Fig. 4A). Compared with wild type, OsWRKY53-OE exhibited larger leaf angles and increased seed size, while d61-2 showed smaller leaf angles and decreased seed size (Fig. 1A; Yamamuro et al., 2000). However, for both leaf angles and seed size, d61-2 OsWRKY53-OE double mutant was remarkably larger than d61-2 (Fig. 4, B–F; Supplemental Figure S11, G–I), suggesting that overexpression of OsWRKY53 can largely rescue BR signaling deficiency phenotypes of d61-2. This result suggested that OsWRKY53 is involved in BR signaling and might act downstream of BR receptor.
OsWRKY53-OE can partially rescue phenotype of d61-2. A, The gross phenotype of OsWRKY53-OE, d61-2, and d61-2 OsWRKY53-OE. CK is WT control that separated from the F2 population. B and C, The leaf angle phenotype (B) and statistical analysis of OsWRKY53-OE, d61-2, and d61-2 OsWRKY53-OE. Data are means ± se (n = 20). D to F, The grain phenotype (scale bar = 5 mm). D, and the statistical analysis (E and F) of OsWRKY53-OE, d61-2, and d61-2 OsWRKY53-OE. Data are means ± se (n = 50). P values were calculated by Student’s t test; *P < 0.05, and **P < 0.01.
OsWRKY53 Interacts with and Is Phosphorylated by OsMAPK6
OsWRKY53 belongs to the group Ia subset of WRKY family and contains two WRKY domains followed by a C2H2-type zinc finger motif (Xie et al., 2005; Chujo et al., 2007; Hu et al., 2015). It has been shown that group Ia WRKY proteins can be phosphorylated by the MAPK cascade, and the phosphorylation can affect the DNA binding activity or transcription activity of WRKY proteins (Ishihama et al., 2011; Chujo et al., 2014; Adachi et al., 2015). Interestingly, previous studies also showed that OsWRKY53 can interact with and be phosphorylated by OsMAPK3 and OsMAPK6, and the phosphorylated sites were located in the SP cluster at the amino terminal domain of OsWRKY53 (Chujo et al., 2014; Hu et al., 2015). Consistent with these results, we confirmed that OsWRKY53 can interact with OsMAPK6 in yeast two-hybrid assays (Fig. 5A). Further bimolecular fluorescence complementation (BiFC) and LUC complementation imaging (LCI) assays confirmed that OsWRKY53 can interact with the OsMAPK6 in planta system (Fig. 5, B and C). In addition, we showed that the OsWRKY53 was weakly phosphorylated by OsMAPK6, whereas this phosphorylation was greatly enhanced in the presence of constitutively active form of OsMAPKK4 (Fig. 5E). However, the phosphorylation of OsWRKY53 by OsMAPKK4-OsMAPK6 was markedly decreased when the five conserved Ser residues in the SP cluster were mutated to Ala, indicating that the five conserved Ser residues are the critical phosphorylation sites of OsWRKY53 by OsMAPKK4-OsMAPK6 cascade (Fig. 5, D and E).
OsMAPK6 can interact with and phosphorylate OsWRKY53. A, Interactions between OsMAPK6 and OsWRKY53 in yeast two-hybrid assays. B, Interaction between OsMAPK6 and OsWRKY53 in BiFC assays. C, Interaction between OsMAPK6 and OsWRKY53 in LCI assays. D, Schematic diagram of OsWRKY53. Five Ser amino acids in conserved SP cluster and WRKY domain were indicated, respectively. E, In vitro phosphorylation of OsWRKY53 by OsMAPK6. Top, western blot detected by Phostag Biotin BTL-104. Bottom, the equal protein loading in top was monitored by Coomassie blue staining.
To examine the effect of OsWRKY53 phosphorylation by OsMAPK6, EMSA was performed, and the result indicated that phosphorylation of OsWRKY53 by the OsMAPKK4-OsMAPK6 cascade can markedly enhance the DNA binding activity of OsWRKY53 protein to the W-box containing the DNA sequence (Supplemental Fig. S7A). In addition, we coexpressed OsWRKY53 and the OsMAPKK4-OsMAPK6 cascade in rice protoplast transient-expression assay and found that the addition of OsMAPKK4 and OsMAPK6 does not change the protein level of OsWRKY53, implying that phosphorylation by the OsMAPKK4-OsMAPK6 cascade cannot affect OsWRKY53 stability (Supplemental Fig. S7B).
Phosphorylation of OsWRKY53 by the OsMAPKK4-OsMAPK6 Cascade Is Critical for the Function of OsWRKY53 in Regulating BR Response
To test biological significance of phosphorylation of OsWRKY53 by the OsMAPKK4-OsMAPK6 cascade, we mutated the five conserved Ser in SP cluster to Ala (generating OsWRKY53 [SA]) or Asp (generating OsWRKY53 [SD]) to mimic the inactive or active phosphorylated forms of OsWRKY53, respectively (Fig. 5D). We then generated the transgenic lines OsWRKY53 (SA)-OEs and OsWRKY53 (SD)-OEs in which OsWRKY53 (SA) and OsWRKY53 (SD) were overexpressed, respectively. Interestingly, we found that OsWRKY53 (SD)-OEs showed much stronger phenotypes than OsWRKY53-OEs, including larger leaf angles, increased grain size, and dwarfism (Fig. 6, A–E; Supplemental Figure S11, J and K). In addition, OsWRKY53 (SD)-OEs were more hypersensitive to BR than OsWRKY53-OEs, as shown by lamina-bending assay (Fig. 6, F and G). Furthermore, compared with that in OsWRKY53-OEs, expression of the three BR biosynthesis genes including D2, OsDWF4, and D11 was decreased more significantly in OsWRKY53 (SD)-OEs (Fig. 6H). Together, these results suggested that constitutive phosphorylation of OsWRKY53 by the OsMAPKK4-OsMAPK6 cascade can enhance the function of OsWRKY53 in BR signaling. In contrast, OsWRKY53 (SA)-OEs failed to produce any obvious phenotypes; even the expression level of OsWRKY53 was much higher than that in OsWRKY53-OE (Supplemental Figs. S8 and S11, L and M), suggesting that these phosphorylation sites are critical for OsWRKY53 functions in regulating BR-related phenotypes. Collectively, these results indicated that OsMAPKK4-OsMAPK6-mediated phosphorylation of OsWRKY53 is indispensable for function of OsWRKY53 in regulating BR signaling.
OsMAPK6 mediated phosphorylation of OsWRKY53 is required for BR-related function of OsWRKY53. A, The gross morphology of OsWRKY53-OE-1 and OsWRKY53(sd)-OE-3. The picture at top right is phenotype of the corresponding flag leaf angle. B, The relative expression of OsWRKY53 in OsWRKY53-OE-1, OsWRKY53(sd)-OE-3, and wild type (WT). The expression level in WT was set as “1”; data were shown as means ± se (n = 3). C, The statistical analysis of leaf angles in OsWRKY53-OE-1 and OsWRKY53(sd)-OE-3. Data are means ± se (n = 20). D and E, The grain phenotype (scale bar = 5 mm; D) and the quantification of grain length and grain width of OsWRKY53-OE-1, OsWRKY53(sd)-OE-3, and WT (E). Data are means ± se (n = 50). F, The leaf inclination of OsWRKY53-OE-1, OsWRKY53(sd)-OE-3, and WT in the presence of indicated concentration of 24-epiBL. G, Statistical analysis of leaf inclination in F; data are means ± se (n = 20). H, The relative expression of D2, OsDWF4, and D11 in WT, OsWRKY53-OE-1, and OsWRKY53(sd)-OE-3. The expression level in WT was set as “1.” Data were shown as means ± se (n = 3). P values were calculated by Student’s t test; *P < 0.05, and **P < 0.01.
OsWRKY53 Is Promoted by BR and Negatively Feedback Regulated by Itself
The fact that OsWRKY53 is involved in BR responses prompted us to test whether expression of OsWRKY53 is regulated by BR. For this purpose, 2-week-old seedlings of wild type were treated with 1 μm 24epi-BL, and then the transcript levels of OsWRKY53 were examined by reverse transcription (RT)-qPCR assay. As shown in Figure 7A, OsWRKY53 transcripts were decreased upon BR treatment, similar to the expression of OsDWF4 gene. We also investigated the BR effects on OsWRKY53 protein stability using MYC-OsWRKY53-OE plants in which OsWRKY53 fused with MYC tag was overexpressed. MYC-OsWRKY53-OE lines were characterized and show similar phenotypes to OsWRKY53-OE (Supplemental Fig. S9). Interestingly, unlike OsWRKY53 transcript, OsWRKY53 protein level was enhanced by BR treatment (Fig. 7B). By contrast, when the MYC-OsWRKY53-OE plants were grown on medium containing brassinozole (BRZ), a BR biosynthesis inhibitor, the MYC-OsWRKY53 protein level decreased (Fig. 7C), whereas the RNA level of OsWRKY53 did not change (Fig. 7D). These results indicated that BR can promote OsWRKY53 protein stability. It has been reported that the transcript and protein levels of several key components in BR signaling (e.g. DLT, BZR1, RLA1) also show a similar BR-responsive pattern, which tends to represent a common negative feedback regulation mechanism in fine-tuning BR signaling (Tong et al., 2009; Qiao et al., 2017).
BR promotes the accumulation of OsWRKY53 protein but represses its transcription level. A Time course of relative expression of OsWRKY53 in response to BR treatment. The expression level before treatment was set as “1”; data were shown as means ± se (n = 3). The expression of OsDWF4 was examined as control. B, Protein gel blot of MYC-OsWRKY53-OE treated with 1 μm BR. The heat shock protein was used as internal control. C, Protein gel blot of MYC-OsWRKY53-OE treated with 5 μm BRZ. The HSP was used as internal control. D, The relative expression of OsWRKY53 in response to BRZ treatment. The expression level in dimethyl sulphoxide treatment was set as “1”; data were shown as means ± se (n = 3). E, The schematic diagrams of OsWRKY53. The position of PCR primers (P1, P2, and P3) used for detecting the relative expression of internal and ectopic OsWRKY53 was indicated. DNA fragments (a and b) were used for ChIP; the black square is the conserved W-box region. F, Relative expression levels of internal OsWRKY53 in oswrky53 mutant plants. The expression level in wild type (WT) was set as “1”; data were shown as means ± se (n = 3). G, Relative expression levels of internal and ectopic OsWRKY53 in OsWRKY53-OE plants. The expression level in WT was set as “1”; data were shown as means ± se (n = 3). H, Schematic diagrams of the effector and reporter plasmids used in the transient assay in rice protoplasts. REN, Renilla luciferase; LUC, firefly luciferase. I, The relative LUC activity expressed with reporter 35S:REN-OsWRKY53Pro:LUC together with control (empty vector) or 35Spro-OsWRKY53 effector. Data were shown as means ± se (n = 5). J, OsWRKY53 binds to the conserved W-box of the OsWRKY53 promoter. The W-box close to ATG was shown as reprehensive. The W-box mutated to AAAAAA was used as competitive probes; maltose-binding protein was used as a negative control. K, ChIP assays showing that OsWRKY53 binds to the promoter of OsWRKY53 in vivo. Immunoprecipitation was performed with anti-OsWRKY53 antibody. Immunoprecipitated chromatin was analyzed by RT-qPCR using primers indicated in (D). RT-qPCR enrichment was calculated by normalizing to actin and to the total input of each sample. Data were shown as means ± se (n = 3). P values were calculated by Student’s t test; *P < 0.05, and **P < 0.01.
Considering that BR oppositely regulates the expression pattern of OsWRKY53 transcript and OsWRKY53 protein level (Fig. 7, A and B), we asked if OsWRKY53 can regulate its own expression. RT-qPCR assay was performed to check the expression of native OsWRKY53 in oswrky53 mutants. We showed that the expression of OsWRKY53 in both of these two mutants was increased (Fig. 7F). Then, we checked the expression of native OsWRKY53 in OsWRKY53-OE lines using the 5′-UTR- and 3′-UTR-specific primers, respectively (Fig. 7E). As shown in Figure 7G, in OsWRKY53-OE lines with increased expression of ectopic OsWRKY53, the native OsWRKY53 expression was reduced. This result was further supported by the transient expression assay in rice protoplast, in which 35Spro:OsWRKY53 can suppress the expression of LUC fused with the native OsWRKY53 promoter (Fig. 7, H and I). There are two W-box elements in OsWRKY53 promoter that could be bound by WRKY transcription factors (Fig. 7E). We then performed EMSA to test if OsWRKY53 binds to its native promoter. As shown in Figure 7J, the MBP-OsWRKY53 fusion protein bound to its own promoter in a W-box-dependent manner. Furthermore, chromatin immunoprecipitation (ChIP) assay also indicated that OsWRKY53 was associated with the W-box region of its own promoter (Fig. 7K). Taken together, these results suggested that OsWRKY53 protein can directly suppress its own expression, which partially explained the negative feedback regulation of OsWRKY53 and implied that OsWRKY53 might be a fine tuner in rice BR-signaling pathway.
In addition, we analyzed the expression of OsWRKY53 in different tissues by RT-qPCR. We showed that OsWRKY53 transcripts can be detected in various tissues and have the highest expression in lamina joint (Supplemental Fig. S10A). In addition, we also generated OsWRKY53 pro:GUS transgenic plants. In agreement with RT-qPCR results, GUS staining of transgenic plants showed that activity of the promoter could be detected in different organs, preferentially higher in lamina joint compared with other tissues (Supplemental Fig. S10B). Altogether, the spatial expression pattern of OsWRKY53 was correlated with one of its physiological functions that controlled the leaf angles.
DISCUSSION
Function of OsWRKY53 in Regulating BR Signaling
In this study, we provide substantial physiological, genetic, and biochemical evidence for the involvement of OsWRKY53 in rice BR-signaling pathway. First, OsWRKY53-OE showed enhanced BR-signaling phenotypes, including enlarged leaf angles and increased seed size (Fig. 1; Supplemental Figs. S1–S4), whereas oswrky53 mutant exhibited BR-deficient phenotypes, such as erect leaves, smaller seed size, and dark green leaves (Fig. 3; Supplemental Fig. S6). Second, BR-sensitivity assay of lamina bending in OsWRKY53-OE and oswrky53 demonstrated that OsWRKY53 is a positive regulator of BR signaling (Figs. 2 and 3). Third, genetic analysis revealed that OsWRKY53 overexpression can largely rescue BR-deficiency phenotypes of BR receptor mutant d61-2 (Fig. 4). Forth, OsWRKY53 interacts with and is phosphorylated by OsMAPK6, and the phosphorylation is required for BR-related function of OsWRKY53 (Figs. 5 and 6). Finally, we showed that the protein level of OsWRKY53 is promoted by BR treatment, and the transcript level of OsWRKY53 is suppressed by BR, possibly through a negative autofeedback regulation (Fig. 7). Taken together, these results demonstrated an important biological role of OsWRKY53 in regulating rice BR signaling (Supplemental Fig. S12).
Functional Diversity of OsWRKY53
WRKY family transcription factors play a variety of developmental and physiological roles in plants. Rice genome contains more than 102 WRKY genes that were divided into four subgroups (Xie et al., 2005; Sun et al., 2014). Up to now, several rice WRKY genes have been functionally characterized, and most of them were involved in stress responses. For instance, OsWRKY31, OsWRKY33, and OsWRKY53 can positively regulate pathogen-infection response (Chujo et al., 2007; Zhang et al., 2008; Koo et al., 2009), while OsWRKY53 and OsWRKY70 negatively regulate herbivore resistance (Hu et al., 2015; Li et al., 2015). Besides, OsWRKY30 and OsWRKY42 play roles in drought response and senescence process, respectively (Shen et al., 2012; Han et al., 2014). By contrast, a role of WRKY genes in controlling growth and development is less known. Similar to OsWRKY53-OE plants, overexpression of OsWRKY70 led to a severe reduction in plant height (Li et al., 2015). In addition, a decreased expression of OsWRKY78 resulted in semidwarfism and small grain due to the reduced cell length (Zhang et al., 2011), which is similar to the oswrky53 mutant. Interestingly, OsWRKY53, OsWRKY70, and OsWRKY78 belong to the same subgroup Ia of WRKY genes and show high sequence similarity (Xie et al., 2005), implying that this subgroup of WRKY genes may play common roles in regulating growth and development. However, the underlying mechanisms might be diverse. OsWRKY70 overexpression lines showed strong dwarfism, but no observable effect on leaf angles and seed size, while OsWRKY70 RNAi lines were similar to wild type, and OsWRKY70 was suggested as growth suppressor by inhibiting GA biosynthesis (Li et al., 2015). In contrast, in this study, both OsWRKY53-OE and oswrky53 showed opposite BR-related phenotypes and BR sensitivities. Therefore, we proposed that OsWRKY53 is a positive regulator of rice BR signaling. It is worthy to note that two previous studies also generated the OsWRKY53 overexpression lines; however, while one study showed that OsWRKY53-OE plants have dwarfism and larger leaf angles (Hu et al., 2015), another study showed the normal growth of the OsWRKY53-OE plants (Chujo et al., 2007). We speculated that these differences may either result from different rice varieties used for the analyses or different expression levels of OsWRKY53 in transgenic lines.
Interestingly, it had been shown that OsWRKY53 can enhance the defense response to pathogen (Chujo et al., 2007; Chujo et al., 2014), and our data show that OsWRKY53 positively regulates BR signaling as well. It is reasonable to speculate that active OsWRKY53 can positively regulate both BR signaling and defense response, which greatly support the previous conclusion that BR can enhance the pathogen response in rice (Nakashita et al., 2003) and also suggest that OsWRKY53 might be a new node mediating the crosstalk between the growth versus defense response.
Moreover, OsWRKY53 transcription is induced by pathogen infection, wounding, and herbivore attack (Chujo et al., 2007; Hu et al., 2015), whereas is repressed by BR (this study). Nevertheless, OsWRKY53 positively regulated BR (this study) and pathogen response (Chujo et al., 2007, 2014), but negatively regulated herbivore-induced defenses (Hu et al., 2015), suggesting that OsWRKY53 might mediate cross talk among diverse signaling pathways and play diverse roles in context-dependent conditions (Supplemental Fig. S12).
MAPK Module in Regulating Rice BR Signaling
Plant MAPK plays crucial roles in multiple signal-transduction pathways, such as plant immunity response and hormone signaling (Tena et al., 2001; Ishihama et al., 2011; Shen et al., 2012; Chujo et al., 2014; Adachi et al., 2015). Rice dwarf and small grains1 and small grains1 were caused by mutations in OsMAPK6 and OsMAPKK4, respectively, and both showed small-grain phenotype (Duan et al., 2014; Liu et al., 2015). OsMAPK6 interacts with OsMAPKK4, implying that the OsMAPKK4-OsMAPK6 cascade was also involved in regulating seed development (Liu et al., 2015). Further studies indicated that the mechanism of OsMAPK6 and OsMKK4 in controlling seed size is through regulating BR response and expression of BR-related genes (Duan et al., 2014; Liu et al., 2015). However, the downstream target of the OsMAPKK4-OsMAPK6 cascade involved in controlling seed size and BR signaling has not been identified. In rice, OsMAPK6 interacts with and can phosphorylate OsWRKY53, and the OsMAPKK4-OsMAPK6-OsWRKY53 module plays key roles in pathogen, wounding, and herbivore-induced defense response (Chujo et al., 2007, 2014; Yoo et al., 2014; Hu et al., 2015). In this study, we showed that OsMAPK6 can phosphorylate OsWRKY53 in an OsMAPKK4-dependent manner, and this phosphorylation is required for the biological function of OsWRKY53 in regulating BR-related responses. Collectively, we proposed that OsWRKY53 is one of the downstream targets of the OsMAPKK4-OsMAPK6 cascade in mediating BR signaling as well (Supplemental Fig. S12). Very interestingly, during preparation of this manuscript, it was reported that the Arabidopsis WRKY46/54/70 are positively involved in BR-regulated growth and negatively involved in drought responses; WRKY54 can be phosphorylated by BIN2 and interact with BES1 to cooperatively regulate the expression of target genes (Chen et al., 2017). These findings in Arabidopsis greatly support our conclusion that OsWRKY53 can positively regulate BR signaling in rice, which implied that the some WRKY family members may play conserve roles in regulating plant BR signaling to some extent. However, their function mechanisms might not be similar. First, AtWRKY46/54/70 are not the OsWRKY53’s homologs with the highest sequence similarity. AtWRKY46/54/70 belong to group III WRKY proteins, which contains one WRKY domain, whereas OsWRKY53 contains two WRKY domains and belongs to group I WRKY proteins. Second, we showed that phosphorylation of OsWRKY53 by OsMAPKK4/OsMAPK6 is required for function of OsWRKY53, and the five Ser residues in the SP cluster of OsWRKY53 are the critical phosphorylation sites. But there is no SP cluster in AtWRKY46/54/70, and whether AtWRKY46/54/70 can be phosphorylated by MAPK is not known. Third, we need to test the physical and genetic interaction of OsWRKY53 with rice BR signaling components (OsBZR1 and OsGSK2) to further investigate the function mechanism of OsWRKY53 in rice BR-signaling pathway. Collectively, there is still much work to be done to prove whether function mechanisms of OsWRKY53 and AtWRKY46/54/70 are similar or not.
WRKY family genes function as transcriptional activator or repressor by binding W-box motif in promoters of target genes (Han et al., 2014; Adachi et al., 2015). Previously, it was reported that OsWRKY53 can function as transcriptional activator in pathogen response (Chujo et al., 2007, 2014), but as a suppressor in herbivore response (Hu et al., 2015), although the direct target genes and the underlying mechanism remained unknown. In this study, we showed that OsWRKY53 could function as a transcription repressor to inhibit its own expression via directly binding to W-box elements in promoter. Considering the functional diversity of OsWRKY53, it is reasonable to speculate that OsWRKY53 might regulate diverse target genes in response to differential environmental stimuli, and it is of interest to identify the direct target genes that are involved in specific functions of OsWRKY53.
MATERIALS AND METHODS
Plant Materials and Growth Conditions
Rice (Oryza sativa) cultivar Longjing 11 (Oryza sativa ssp. japonica) was used for generate OsWRKY53 transgenic plants. d61-2 (Yamamuro et al., 2000) was used to develop double mutant. Plants were grown in the field (natural long-day conditions) or growth chamber at 30°C for 14 h (day) and 24°C for 10 h (night). For transient induction analysis, 2-week-old seedlings were sprayed with 1 μm 24-epiBL (Sigma, E1641), and then whole plants were harvested at various time points. For BRZ treatment, the 3-d-old plants were moved to medium with 5 μm BRZ for 7 d (Sigma, SML 1406). For BR-related gene expression, 2-week-old seedlings were sampled. Materials were collected and frozen in liquid nitrogen immediately or stored at −80°C for RNA extraction.
Lamina Joint Assay
The lamina joint assay using excised leaf segments was performed as described previously (Tong et al., 2009). Simply, uniform germinated seeds were selected and grown in the dark for 8 d at 30°C. The entire segments comprising 1 cm of the second leaf blade, the lamina joint, and 1 cm of the leaf sheath were incubated in various concentrations of 24-epiBL for 48 h in dark. The angles of lamina joint bending were measured using ImageJ software (http://rsbweb.nih.gov/ij/).
Generation of Transgenic Plants and Double Mutant
To generate the OsWRKY53 overexpression plant, the coding regions of OsWRKY53 (Os05g0343400) was amplified by PCR and cloned into p1390U, and Ubiquitinpro: OsWRKY53 construct in pCAMBIA1300 backbone was made. For creating 35Spro: MYC-OsWRKY53, OsWRKY53 in pENTRY vector was cloned into pGWB18 through LR recombination reaction (Nakagawa et al., 2007). For the generation of oswrky53 mutant, the target sequence (Supplemental Table S1) was synthesized and ligated with respective sgRNA catastases and then were sequentially ligated into the CRISPR/Cas9 binary vectors pYLCRISPR/Cas9Pubi-H as described (Ma et al., 2015). To generate OsWRKY53(SA) and OsWRKY53(SD), the mutation was introduced by point mutation kit following the manufacturer’s instructions. Primers used to generate the above-mentioned constructs are listed in Supplemental Table S1, and all of the constructs were confirmed by sequencing. The constructs were introduced into Agrobacterium tumefaciens strain EHA105. Cultivar Longjing11 was used as the recipients for Agrobacterium-mediated transformation as described previously (Hiei et al., 1994). Homozygous T2 transgenic rice seedlings were used for the phenotype analysis.
To generate d61-2 OsWRKY53-OE, we crossed d61-2 (Yamamuro et al., 2000) with OsWRKY53-OE#1 (Fig. 1A). The mutation site and expression level of corresponding genes was examined by DNA sequence and RT-qPCR.
Total RNA Isolation and RT-qPCR Analysis
Total RNA was extracted using TRIzol (Invitrogen) and treated with DNaseI. cDNA was synthesized from 2 µg of total RNA using SuperscriptII Reverse Transcriptase (Invitrogen). Real-time PCR was performed with SYBR Green PCR master mix (Takara). Data were collected using Bio-Rad chromo 4 real-time PCR detector. All expressions were normalized against the Actin gene (Os03g0718100). The primers used are listed in Supplemental Table S1. Three biological repeats were performed for each analysis. Values are means ± se of three biological repeats.
Yeast Two-Hybrid Assay
The full-length coding sequence of OsWRKY53 and the OsMAPK6 were cloned into pGADT7 and pGBKT7, respectively. These two vectors and corresponding empty vectors were cotransformed into the yeast strain Y2H gold. The yeast two-hybrid assays was performed as per the kit’s instructions.
BIFC Assay
For BiFC assays, the OsMAPK6 and OsWRKY53 were fused with partial GFP and generated nGFP-OsMAPK6 and cGFP-OsWRKY53. These vectors were transformed into Agrobacterium strain GV3101 and coinjected into young leaves of Nicotiana benthamiana. The fluorescence was observed by confocal microscopy (Leica) after 2 d growth.
LCI Assay
The LCI assays for the interaction between OsMAPK6 and OsWRKY53 were performed in N. benthamiana leaves. The full-length OsMAPK6 and OsWRKY53 coding region were fused with the N-terminal part and C-terminal part of the luciferase reporter gene, respectively. Agrobacteria harboring nLUC-OsMAPK6 and cLUC-OsWRKY53 constructs were coinfiltrated into N. benthamiana, and the infiltrated leaves were analyzed for LUC activity 48 h after infiltration using Chemiluminescence imaging (Tanon 5200).
EMSA
Full-length coding sequence of OsWRKY53 was cloned into PVP13 vector via an LR recombination reaction to generate MBP-OsWRKY53 fusion protein and transformed into Escherichia coli strain BL21 (DE3). The recombinant proteins were affinity purified using Amylose Resin (New England Biobabs, cat. no. E8021S). About 40-bp length oligonucleotide probes containing wide-type W-box (TTGACC) or mutated W-box (AAAAAA) motifs were synthesized and labeled with biotin using EMSA Probe Biotin Labeling Kit (Beyotime, cat. no. GS008). For nonlabeled probe competition, nonlabeled probe was added to the reactions. EMSA was performed using a Chemiluminescent EMSA kit (Beyotime, cat. no. GS009). Probe sequences are shown in Supplemental Table S1.
Transient Transcription Dual-Luciferase Assay
Full-length coding sequence of OsWRKY53 was cloned into KpnI and BamHI site of pRT107, and effector (35Spro:OsWRKY53) was generated. The promoter regions (upstream of the ATG) of OsWRKY53 were amplified and cloned into SalI and BamHI site of pGreenII 0800-LUC vector and used as reporter (Hellens et al., 2005). The resulting effector and reporter constructs were cotransfected into protoplasts prepared from 2-week-old rice seedlings. Protoplast isolation and polyethylene glycol transformation was carried out as described (Huang et al., 2015). The Renilla luciferase (REN) gene directed by 35S promoter in the pGreenII 0800-LUC vector was used as an internal control. Firefly LUC and REN activities were measured with a Dual-Luciferase reporter assay kit using a GloMax 20/20 luminometer (Promega). The LUC activity was normalized to REN activity, and LUC/REN ratios were calculated. For each plasmid combination, five independent transformations were performed. Values are means ± se of five biological repeats. Primers used for these constructs are listed in Supplemental Table S1.
ChIP Assay
OsWRKY53-OE was used for ChIP assay as previously described (Zhu et al., 2012). In brief, approximately 2 g of rice seedlings was cross linked in 1% formaldehyde under a vacuum, and cross-linking was stopped with 0.125 m Gly. The sample was ground to a powder in liquid nitrogen and used to isolate nuclei. Anti-OsWRKY53 (1:150 dilutions) was used to immunoprecipitate the protein-DNA complex, and the precipitated DNA was recovered and analyzed by quantitative PCR. Chromatin precipitated without antibody was used as a control. The data are presented as means ± se of three biological repeats. Anti-OsWRKY53 (AbP80050-A-SE) was ordered from Beijing Protein innovation. Primers used for ChIP-qPCR are listed in Supplemental Table S1.
In Vitro Kinase Assay
The full-length OsMAPK6 and OsMAPKK4DD was cloned into pDEST15 vector via an LR recombination reaction and transformed into E. coli strain BL21 (DE3). OsMAPKK4DD was generated as described (Chujo et al., 2014). Primers used for these constructs are listed in Supplemental Table S1. The recombinant proteins were affinity purified using Glutathione Sepharose 4B beads (GE Healthcare). For each kinase assay, 0.5 μg MBP-OsWRKY53 or mutated fusion protein MBP-OsWRKY53 (SA) and 0.3 μg glutathione S-transferase (GST)-OsMAPK6, GST-OsMAPKK4DD, or GST were incubated with phosphorylation buffer (25 mm Tris, pH 7.4, 12 mm MgCl2, 1 mm dithiothreitol, and 1 mm ATP). The reactions were incubated at 30°C for 45 min and boiled with 5× sodium dodecyl sulfate (SDS) loading buffer then separated by SDS-polyacrylamide gel electrophoresis. The protein was transferred on polyvinylidene difluoride membrane (Millipore), and phosphorylation signal was detected by Phos-tag Biotin BTL-104 (Wako) according to the manufacturer’s instruction.
Transient Expression Assay in Rice Protoplast
The coding region of OsMAPKK4 DD, OsMAPK6, and OsWRKY53 fused with MYC tag were ligated into the pRT107 vector to generate the 35Spro: OsMAPKK4 DD, 35Spro: OsMAPK6, and 35Spro: MYC-OsWRKY53 constructs. Rice protoplast was isolated from stem and sheath tissues of wild-type young seedlings as described previously (Chen et al., 2006). Different combinations of plasmid DNAs (about 10 μg DNA of each construct) were transiently expressed in protoplasts by PEG-mediated transfection procedure.
After overnight incubation in dark at 28°C, total proteins were isolated from rice protoplasts with extraction buffer (50 mm Tris/HCl, pH 7.5, 150 mm NaCl, 0.2% NP-40, 0.1% Triton X-100, 5 mm ethylenediaminetetraacetic acid disodium salt, complete protease inhibitor cocktail, and 50 μm MG132), and 5× SDS buffer was added and boiled for 5 min. The samples were loaded on a 10% SDS-polyacrylamide gel electrophoresis for immunoblotting with anti-MYC antibody (Abmart, M20002L) and anti-HSP antibody (BGI Tech, AbM51099), respectively.
Accession Numbers
Sequence data from this article can be found in the GenBank/EMBL database under the following accession numbers: OsWRKY53 (Os05g0343400); OsMAPKK4 (Os02g0787300); OsMAPK6 (Os06g0154500).
Supplemental Data
The following supplemental materials are available.
Supplemental Figure S1. Enlarged leaf angle of OsWRKY53 overexpression lines.
Supplemental Figure S2. Expression of cell-enlargement-related gene in OsWRKY53-OE.
Supplemental Figure S3. Pale green phenotype of OsWRKY53-OE plant.
Supplemental Figure S4. Identification of OsWRKY53 overexpression lines.
Supplemental Figure S5. Identification of oswrky53 mutant.
Supplemental Figure S6. The leaf of oswrky53 is dark green.
Supplemental Figure S7. OsWRKY53 phosphorylation by OsMAPK6 enhances the DNA binding activity of OsWRKY53 protein without affecting its stability.
Supplemental Figure S8. Phenotypic analysis of OsWRKY53(SA)-OE.
Supplemental Figure S9. Characterization of MYC-OsWRKY53-OE.
Supplemental Figure S10. Spatial expression pattern of OsWRKY53.
Supplemental Figure S11. Phenotype and quantification of unhusked grain of diverse lines.
Supplemental Figure S12. Schematic diagram of working model of OsWRKY53.
Supplemental Table S1. List of primers.
Acknowledgments
The authors thank our laboratory members for their helpful comments and discussions during the article preparation. They thank Prof. Chengcai Chu, Prof. Wenqiang Tang, and Dr. Jiuyou Tang for their comments on manuscript and helpful suggestions. They thank Prof. Enamul Huq for his critical reading and revision of the manuscript.
Footnotes
The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Qingyun Bu (buqingyun{at}iga.ac.cn).
Q.B. conceived and supervised the project; X.T. performed most of experiments; X.L., W.Z., Y.R., Z.W., Z.L., and J.T. assisted the experiments; Q.B. and X.T. analyzed the data and wrote the article with the contributions from H.T. and J.F.
↵1 This study was supported by National Natural Science Foundation of China (grant no. 31671653), the Strategic Priority Research Program of Chinese Academy of Sciences (grant no. XDA08040101), the Natural Science Foundation of Heilongjiang (grant no. ZD2015005), and the Hundred-Talent-Program of Chinese Academy of Sciences to Q.B.
- Received July 13, 2017.
- Accepted September 6, 2017.
- Published September 11, 2017.