FtbZIP12 Positively Regulates Responses to Osmotic Stress in Tartary Buckwheat

ABFs play a key role in regulating plant osmotic stress. However, in Tartary buckwheat, data on the role of ABF genes in osmotic stress remain limited and its associated mechanism in osmoregulation remain nebulous. Herein, a novel ABF family in Tartary buckwheat, FtbZIP12, was cloned and characterized. FtbZIP12 is a transcriptional activator located in the nucleus; its expression is induced by NaCl, mannitol, and abscisic acid (ABA). Atopic expression of FtbZIP12 in Arabidopsis promoted seed germination, reduced damage to primary roots, and improved the tolerance of seedlings to osmotic stress. The quantitative realtime polymerase chain reaction (RT-qPCR) results showed that the expressions of the typical genes related to stress, the SOS pathway, and the proline synthesis pathway in Arabidopsis were significantly (p < 0.05) upregulated under osmotic stress. FtbZIP12 improved the osmotic pressure resistance by reducing the damage caused by reactive oxygen species to plants and maintained plant homeostasis by upregulating the expression of genes related to stress, osmotic regulation, and ion homeostasis. This study identified a key candidate gene for understanding the mechanism underlying osmotic-stress-regulated function in Tartary buckwheat, thereby providing a theoretical basis for improving its yield and quality.


Introduction
Stress tolerance of plants is influenced by the expression of stress-induced genes regulated by specific transcription factors (TFs) [1]. TFs directly or indirectly regulate the expression of downstream stress-related target genes via cis-acting elements [1,2]. The bZIP transcription factor is one of the key regulators of plant stress resistance involved in regulating the defense against pathogens, light tolerance, and stress-signal transduction [3]. The domain of bZIP is conserved in eukaryotes and consists of 6080 amino acids. Its N-terminus contains a N-x7-R/K-x9 motif and a C-terminal leucine zipper region that mediates homologous and/or heterodimerization of the bZIP protein [4,5]. Therefore, mining bZIP TFs related to abiotic stress is of great significance for understanding their regulatory mechanisms during stress.
Among the bZIP TFs family, the ABF subfamily is involved in various pathways such as growth and development and abiotic stress. In Arabidopsis thaliana, 13 members of the ABF subgroup have been identified, most of which are induced by abiotic stress, such as AREB1/ABF2, AREB2/ABF4, ABF1, and ABF3 [6,7]. These genes not only improve the resistance of Arabidopsis to abiotic stress alone, but also regulate plant resistance in the form of homodimers [7,8]. The regulation of ABF subfamily genes in the abscisic acid (ABA) signaling pathway has been extensively studied. ABF1-4 gene expression and accumulation of their endogenous proteins are induced by ABA and play a negative regulatory role in the ABA signal-transduction pathway [9]. During ABA induction, ABF regulates the expression of SAG29 and other SAG12, leading to the senescence and death

Isolation and Characterization of FtbZIP12
To study its functional properties and role in Tartary buckwheat, FtbZIP12 was cloned and analyzed via homologous cloning. FtbZIP12 contains a complete open reading frame with a 1299 bp sequence comprising 432 amino acids with a molecular weight of 46,612.39 kDa, a theoretical pl of 9.22, and a formula of C2003H3240N606O643S16 (Expasy ProtParam tool).
A phylogenetic analysis of the Arabidopsis bZIP family revealed that FtbZIP12 (FtPinG0201993700. 01. T 01) clustered with ABH04550.1 (unknown), AEE86309.1 (ABF3), Q9M7Q2.1 (ABF2), Q9M7Q5.1 (ABF1), and Q9LES3.1 (ABI5) in the ABF subfamily ( Figure S1). Among these, FtbZIP12 was the closest to ABH04550.1, although the functional properties of ABH04550.1 remain to be determined. To better understand the characteristics of FtbZIP12, a phylogenetic analysis and a multiple sequence alignment analysis were performed with ABFs from other species with known functions ( Figure 1A,B). The results showed that the subfamily of AREB/ABF TFs with known functions were divided into three subfamilies in different species, among which FtbZIP12 was closely related to Arabidopsis, grapevine, and tobacco ( Figure 1A). FtbZIP12 was most similar to the ABFs in Arabidopsis (NP849777 and BT026443), whereas NP849777 was an osmotic-stress-related TF ( Figure 1A). The multiple sequence analysis of ABF2 revealed that all had the same bZIP conserved domain with four phosphorylation sites (C1-C4), suggesting that FtbZIP12 was likely have the same functional characteristics ( Figure 1B). We subsequently conducted a more detailed analysis of the properties of FtbZIP12 in Tartary buckwheat to determine its mechanism of action.

FtbZIP12 Expression Is Induced by Abiotic Stress Stimuli
The above bioinformatics analysis suggested that FtbZIP12 may be related to abiotic stress; therefore, we analyzed the expression pattern of FtbZIP12 in Tartary buckwheat following NaCl, mannitol, and ABA treatment using RT-qPCR. FtbZIP12 expression was significantly upregulated by NaCl, mannitol, and ABA in Tartary buckwheat (p < 0.05) and its expression trends differed. Under NaCl treatment, with a peak at 3 h, its expres-

FtbZIP12 Expression Is Induced by Abiotic Stress Stimuli
The above bioinformatics analysis suggested that FtbZIP12 may be related to abiotic stress; therefore, we analyzed the expression pattern of FtbZIP12 in Tartary buckwheat following NaCl, mannitol, and ABA treatment using RT-qPCR. FtbZIP12 expression was significantly upregulated by NaCl, mannitol, and ABA in Tartary buckwheat (p < 0.05) and its expression trends differed. Under NaCl treatment, with a peak at 3 h, its expression decreased over time ( Figure 2A). However, the reverse expression pattern was observed under mannitol treatment, which was more than 10-fold higher at 24 h than 0 h ( Figure 2B). Following ABA stimulation, the expression level first increased then decreased, peaking at 3 h ( Figure 2C). The stimulated response of FtbZIP12 to abiotic stress described above showed that the mechanism of FtbZIP12 in abiotic stress response is relatively complex. The expression levels of FtbZIP12 in roots, stems, flowers, seeds, and leaves at different developmental stages while taking the seed as the control. Values are expressed as means ± SD (p < 0.05); each data value is from three replicate experiments.

Tissue-Specific Analysis of FtbZIP12
RT-qPCR was used to analyze the tissue specificity in FtbZIP12 in Tartary buckwheat. The results showed that FtbZIP12 was expressed in various tissues, although at different levels ( Figure 2D). Taking the seed as the control, the relative expression was the highest in old leaves and stems followed by young leaves, and was relatively low in flowers and roots ( Figure 2D).

FtbZIP12 Has Transactivation Activity
To further understand the characteristics of FtbZIP12, its subcellular localization was predicted using the plant PLOC server (SJTU. Edu. CN). The results showed that FtbZIP12 was located in the nucleus. To determine whether FtbZIP12 had self-activation ability, the complete CDS of FtbZIP12 was inserted into pGBKT7 and pGADT7 vectors and FtbZIP12-pGBKT7 + pGADT7/FtbZIP12-pGADT7 + pGBKT7 were co-transformed into Y2HGold cells (Saccharomyces cerevisiae). The results showed that the reporter gene could be activated by the two constructs; however, the activation ability of the The expression levels of FtbZIP12 in roots, stems, flowers, seeds, and leaves at different developmental stages while taking the seed as the control. Values are expressed as means ± SD (p < 0.05); each data value is from three replicate experiments.

Tissue-Specific Analysis of FtbZIP12
RT-qPCR was used to analyze the tissue specificity in FtbZIP12 in Tartary buckwheat. The results showed that FtbZIP12 was expressed in various tissues, although at different levels ( Figure 2D). Taking the seed as the control, the relative expression was the highest in old leaves and stems followed by young leaves, and was relatively low in flowers and roots ( Figure 2D).

FtbZIP12 Has Transactivation Activity
To further understand the characteristics of FtbZIP12, its subcellular localization was predicted using the plant PLOC server (SJTU. Edu. CN). The results showed that FtbZIP12 was located in the nucleus. To determine whether FtbZIP12 had self-activation ability, the complete CDS of FtbZIP12 was inserted into pGBKT7 and pGADT7 vectors and FtbZIP12-pGBKT7 + pGADT7/FtbZIP12-pGADT7 + pGBKT7 were co-transformed into Y2HGold cells (Saccharomyces cerevisiae). The results showed that the reporter gene could be activated by the two constructs; however, the activation ability of the FtbZIP12-pGADT7 construct was weaker than that of FtbZIP12-pGBKT7, indicating that FtbZIP12 was a transcriptional activator (Figure 3). FtbZIP12-pGBKT7 + pGADT7/FtbZIP12-pGADT7 + pGBKT7 were co-transformed into Y2HGold cells and grown on SD-LW or SD-ALWH plates to check their growth status. PGADT7 + PGBKT7 was used as the negative control. SD-LW: -trp and -leu; SD-ALWH: -trp, -leu, and -his. The experiment was repeated three times.

FtbZIP12 Confers Osmotic Stress Tolerance to Arabidopsis
The overexpression plants of the FtbZIP12 (35S: FtbZIP12) construct were obtained to study their function using gain-of-function methods. Transgenes F1, F2, and F3 were positively screened using 100 mg/L of hygromycin and the overexpressing plants were selected using RT-qPCR. Among them, the T3 generation lines with high (OE#18) and moderate (OE#06) expression levels were selected for subsequent experiments ( Figure  S2).
To verify the resistance of FtbZIP12 and determine whether it could improve the stress tolerance of seeds, T3 generation seeds were sown on a half-strength MS solid medium containing 125 mM of NaCl and 225 mM of mannitol; the germination rate was calculated from the second day. As shown in Figure 4, the seed germination rate and growth status of all lines were consistent on the normal medium ( Figure 4A,B). Under salt stress, some wild-type seeds did not germinate and the germination rate reached approximately 85% on day 6; the germination rates of transgenic plants (97%) were significantly higher than those of the wild-type plants ( Figure 4C,D). Under mannitol-simulated drought conditions, the germination rate of the wild-type plants was only 58%, whereas that of the transgenic lines was >98%, which was significantly higher than that of the wild-type plants ( Figure 4E,F). This indicated that FtbZIP12 may significantly reduce damage to seed germination under osmotic stress. FtbZIP12-pGBKT7 + pGADT7/FtbZIP12-pGADT7 + pGBKT7 were co-transformed into Y2HGold cells and grown on SD-LW or SD-ALWH plates to check their growth status. PGADT7 + PGBKT7 was used as the negative control. SD-LW: -trp and -leu; SD-ALWH: -trp, -leu, and -his. The experiment was repeated three times.

FtbZIP12 Confers Osmotic Stress Tolerance to Arabidopsis
The overexpression plants of the FtbZIP12 (35S: FtbZIP12) construct were obtained to study their function using gain-of-function methods. Transgenes F1, F2, and F3 were positively screened using 100 mg/L of hygromycin and the overexpressing plants were selected using RT-qPCR. Among them, the T3 generation lines with high (OE#18) and moderate (OE#06) expression levels were selected for subsequent experiments ( Figure S2).
To verify the resistance of FtbZIP12 and determine whether it could improve the stress tolerance of seeds, T3 generation seeds were sown on a half-strength MS solid medium containing 125 mM of NaCl and 225 mM of mannitol; the germination rate was calculated from the second day. As shown in Figure 4, the seed germination rate and growth status of all lines were consistent on the normal medium ( Figure 4A,B). Under salt stress, some wild-type seeds did not germinate and the germination rate reached approximately 85% on day 6; the germination rates of transgenic plants (97%) were significantly higher than those of the wild-type plants ( Figure 4C,D). Under mannitol-simulated drought conditions, the germination rate of the wild-type plants was only 58%, whereas that of the transgenic lines was >98%, which was significantly higher than that of the wild-type plants ( Figure 4E,F). This indicated that FtbZIP12 may significantly reduce damage to seed germination under osmotic stress. We next determined whether FtbZIP12 could improve plant resistance at the seedling stage. The 3-day-old seedlings growing in the normal medium were selected and transplanted to the medium containing 125 mM NaCl and 225 mM mannitol; their root lengths and fresh weights were measured on day 7. As shown in Figure 5A, under normal growth conditions, the roots of all lines reached the bottom of the culture plate. The leaves and stems became larger and longer, whereas the fresh weights and root lengths remained almost unchanged ( Figure 5A). Under salt stress, the growth and development of all lines were inhibited, with the wild-type plants showing the most evident damage; their root lengths were only approximately 2.96 cm, which was much lower than that of the transgenic plants (>5 cm) ( Figure 5B,D). The fresh weight was only 14.67 mg, which We next determined whether FtbZIP12 could improve plant resistance at the seedling stage. The 3-day-old seedlings growing in the normal medium were selected and transplanted to the medium containing 125 mM NaCl and 225 mM mannitol; their root lengths and fresh weights were measured on day 7. As shown in Figure 5A, under normal growth conditions, the roots of all lines reached the bottom of the culture plate. The leaves and stems became larger and longer, whereas the fresh weights and root lengths remained almost unchanged ( Figure 5A). Under salt stress, the growth and development of all lines were inhibited, with the wild-type plants showing the most evident damage; their root lengths were only approximately 2.96 cm, which was much lower than that of the transgenic plants (>5 cm) ( Figure 5B,D). The fresh weight was only 14.67 mg, which was significantly lower than that of overexpression lines 6 (OE#06) (19.67 mg) and 18 (OE#18) (21 mg) ( Figure 5B,D). Under drought stress, the root lengths of transgenic plants (3.45 cm) were also higher than those of the wild-type plants (2.75 cm) (* p < 0.05) ( Figure 5C,D). The fresh weight of the wild-type plants was only 6.66 mg, whereas those of OE#06 and OE#18 were 15.5 and 21 mg, respectively (** p < 0.01) ( Figure 5C,E). These results suggested that FtbZIP12 may reduce the sensitivity of Arabidopsis seedlings to osmotic stress, thereby improving plant resistance. was significantly lower than that of overexpression lines 6 (OE#06) (19.67 mg) and 18 (OE#18) (21 mg) ( Figure 5B,D). Under drought stress, the root lengths of transgenic plants (3.45 cm) were also higher than those of the wild-type plants (2.75 cm) (* p < 0.05) ( Figure 5C,D). The fresh weight of the wild-type plants was only 6.66 mg, whereas those of OE#06 and OE#18 were 15.5 and 21 mg, respectively (** p < 0.01) ( Figure 5C,E). These results suggested that FtbZIP12 may reduce the sensitivity of Arabidopsis seedlings to osmotic stress, thereby improving plant resistance. Finally, to assess the physiological and biochemical effects of osmotic stress on the overexpression plant seedlings, soil-cultured seedlings were subjected to NaCl and drought stress and their physiological and biochemical indicators were tested. The results shown in Figure 6A illustrate that under regular watering, the growth vigor and leaf color of the wild-type and transgenic lines did not differ. The growth of all lines was seriously hindered under drought stress, indicating that the plants were short and the Finally, to assess the physiological and biochemical effects of osmotic stress on the overexpression plant seedlings, soil-cultured seedlings were subjected to NaCl and drought stress and their physiological and biochemical indicators were tested. The results shown in Figure 6A illustrate that under regular watering, the growth vigor and leaf color of the wild-type and transgenic lines did not differ. The growth of all lines was seriously hindered under drought stress, indicating that the plants were short and the leaves were withered. The wild-type plants were the most severely damaged ( Figure 6A). Under salt stress, the leaves of the wild-type plants turned yellow or even withered, whereas some new leaves of the overexpressing plants remained green ( Figure 6A). 6A). Under salt stress, the leaves of the wild-type plants turned yellow or even withered, whereas some new leaves of the overexpressing plants remained green ( Figure 6A).
By detecting the content of malondialdehyde (MDA) and the enzymatic activity of proline (PRO) and catalase (CAT), we explored the accumulation of reactive oxygen species (ROS) and the degree of membrane oxidative damage in the transgenic lines under osmotic stress, as shown in Figure 6B-D. Under normal watering conditions, the MDA, CAT, and PRO levels were similar in all plants (p > 0.05). After three weeks of NaCl solution application and two weeks of no watering, the OE lines exhibited the highest CAT and PRO contents ( Figure 6B,C), whereas the MDA levels were the highest in the wild-type plants and more than three-fold higher than that in the transgenic plants ( Figure 6D). These results indicated that under drought and NaCl stress, ROS rapidly increased in the plants. The overexpression of FtbZIP12 could reduce the damage attributed to ROS by increasing the accumulation of PRO and CAT.

Analysis of the Changes in the Expression Levels of Genes Downstream of FtbZIP12 Related to Stress
To determine the response mechanism of FtbZIP12 under osmotic stress, we selected genes from A. thaliana that may be involved in osmotic stress, including ion transporter genes (SOS1, SOS2, and SOS3), abundant late embryogenesis genes (LEA6 and LEA7), PRO synthetic genes (P5CS1), RD29B, and ERD1. The expression level of these By detecting the content of malondialdehyde (MDA) and the enzymatic activity of proline (PRO) and catalase (CAT), we explored the accumulation of reactive oxygen species (ROS) and the degree of membrane oxidative damage in the transgenic lines under osmotic stress, as shown in Figure 6B-D. Under normal watering conditions, the MDA, CAT, and PRO levels were similar in all plants (p > 0.05). After three weeks of NaCl solution application and two weeks of no watering, the OE lines exhibited the highest CAT and PRO contents ( Figure 6B,C), whereas the MDA levels were the highest in the wild-type plants and more than three-fold higher than that in the transgenic plants ( Figure 6D). These results indicated that under drought and NaCl stress, ROS rapidly increased in the plants.
The overexpression of FtbZIP12 could reduce the damage attributed to ROS by increasing the accumulation of PRO and CAT.

Analysis of the Changes in the Expression Levels of Genes Downstream of FtbZIP12 Related to Stress
To determine the response mechanism of FtbZIP12 under osmotic stress, we selected genes from A. thaliana that may be involved in osmotic stress, including ion transporter genes (SOS1, SOS2, and SOS3), abundant late embryogenesis genes (LEA6 and LEA7), PRO synthetic genes (P5CS1), RD29B, and ERD1. The expression level of these downstream genes under osmotic stress was detected using RT-qPCR. Under osmotic stress, cells lose water, thereby lowering the osmotic pressure. Ions are disordered in vivo; however, SOS3, SOS2, and SOS1 in the SOS pathway maintain ion stability [28]. Our results showed that SOS3, SOS2, and SOS1 were significantly induced in FtbZIP12-overexpressing lines; these genes were increased several-fold to dozens-fold under stress ( Figure 7A-C), suggesting that FtbZIP12 can activate the SOS pathway. After stress treatment, the relative expression of AtRD29B, AtLEA6, AtLEA7, AtP5CS1, and AtERD1 was also induced and the transgenic lines were significantly higher ( Figure 7D-H). These results indicated that the enhanced stress tolerance of transgenic lines may be linked to the expression of downstream stress genes induced by FtbZIP12. downstream genes under osmotic stress was detected using RT-qPCR. Under osmotic stress, cells lose water, thereby lowering the osmotic pressure. Ions are disordered in vivo; however, SOS3, SOS2, and SOS1 in the SOS pathway maintain ion stability [28].
Our results showed that SOS3, SOS2, and SOS1 were significantly induced in FtbZIP12-overexpressing lines; these genes were increased several-fold to dozens-fold under stress ( Figure 7A-C), suggesting that FtbZIP12 can activate the SOS pathway. After stress treatment, the relative expression of AtRD29B, AtLEA6, AtLEA7, AtP5CS1, and AtERD1 was also induced and the transgenic lines were significantly higher ( Figure 7D-H). These results indicated that the enhanced stress tolerance of transgenic lines may be linked to the expression of downstream stress genes induced by FtbZIP12.

Discussion
With higher levels of pollution, the size of saline-alkali soil and arid areas have been increasing annually, which seriously threatens the yield and quality of crops. The bZIP TFs in rice, soybean, grape, and other crops have different mechanisms for biological and abiotic stresses [29][30][31][32]. In this study, FtbZIP12 was identified and isolated. Based on the results of a phylogenetic tree analysis and a multiple sequence analysis, FtbZIP12 was presumed to be a member of the ABF subfamily (Figures 1 and S1). However, the function of ABH04550.1, which was closest to FtbZIP12, has not been identified, hence the specific functional characteristics of FtbZIP12 warrant further studies ( Figure S1). According to the results of the multiple sequence analysis of the ABF gene in each species, FtbZIP12 was closely related to A. thaliana, grape, and tobacco ( Figure 1A) with the same conserved domain ( Figure 1B). In addition, the subcellular localization prediction results for FtbZIP12 and its transcriptional activity in yeast were the same as those of Arabidopsis ABF, both of which have transcriptional activators located in the nucleus ( Figures 2D and 3) [30]. This indicated that FtbZIP12 is likely to be a member of the ABFs. In this study, a high expression of FtbZIP12 in old leaves and stems of Tartary buckwheat was detected ( Figure 2D). We speculated that FtbZIP12 was related to leaf senescence, stress response, and stress signal transmission. Under abiotic stress, old

Discussion
With higher levels of pollution, the size of saline-alkali soil and arid areas have been increasing annually, which seriously threatens the yield and quality of crops. The bZIP TFs in rice, soybean, grape, and other crops have different mechanisms for biological and abiotic stresses [29][30][31][32]. In this study, FtbZIP12 was identified and isolated. Based on the results of a phylogenetic tree analysis and a multiple sequence analysis, FtbZIP12 was presumed to be a member of the ABF subfamily (Figure 1 and Figure S1). However, the function of ABH04550.1, which was closest to FtbZIP12, has not been identified, hence the specific functional characteristics of FtbZIP12 warrant further studies ( Figure S1). According to the results of the multiple sequence analysis of the ABF gene in each species, FtbZIP12 was closely related to A. thaliana, grape, and tobacco ( Figure 1A) with the same conserved domain ( Figure 1B). In addition, the subcellular localization prediction results for FtbZIP12 and its transcriptional activity in yeast were the same as those of Arabidopsis ABF, both of which have transcriptional activators located in the nucleus (Figures 2D and 3) [30]. This indicated that FtbZIP12 is likely to be a member of the ABFs. In this study, a high expression of FtbZIP12 in old leaves and stems of Tartary buckwheat was detected ( Figure 2D). We speculated that FtbZIP12 was related to leaf senescence, stress response, and stress signal transmission. Under abiotic stress, old leaves are usually the first to show stress symptoms, following which a series of signals are transmitted through the stalk to various plant parts in response as part of the defense mechanism. Therefore, whether FtbZIP12 participates in the reception and transmission of abiotic stimulus signals requires further investigations.
In Arabidopsis, the ABF subfamily genes ABF2, ABF3, and ABF4 were induced under high salinity, drought, freezing, and cold conditions, which improved the ability of the plants to resist external stresses [10,33,34]. Even though ABF1 was induced at a low expression level, it played a key role in drought stress [35]. The expression of FtbZIP12 was induced by ABA, NaCl, and mannitol; however, the RT-qPCR results showed that changes in the FtbZIP12 expression after 0-24 h of mannitol induction were different from those under NaCl stress (Figure 2A,B). This expression pattern was different from that of FtbZIP83 and FtbZIP5, the relative expression of which first increased then decreased under ABA, NaCl, and drought stress [26,27]. We concluded that although FtbZIP12 belongs to the same subfamily as FtbZIP83 and FtbZIP5, they are not only different in sequence, but also have different roles under osmotic stress. Therefore, we speculated that the regulatory mechanism of FtbZIP12 in abiotic stress is different from that of FtbZIP83 and FtbZIP5. At present, there is no mature and stable genetic transformation system for Tartary buckwheat, hence it is difficult to obtain stable silencing and overexpression in buckwheat plants. The perfect transformation system of A. thaliana provides the possibility of the functional verification of Tartary buckwheat stress-resistance genes. In this study, FtbZIP12-overexpressing T3 generation of stably expressing plants was obtained using the floral dip method to verify the regulatory mechanism of FtbZIP12 in osmotic stress. Previously, several ABF functions were verified in A. thaliana; ABFs can significantly improve plant resistance to external environment. In A. thaliana, overexpression of ABI5 and ABI3 increased the seed germination rate and reduced post-germination growth retardation via the ABA signaling pathway under osmotic stress [36]. In Arabidopsis mutants, the deletion of AtABF1 increased the sensitivity of primary roots to abiotic stress and seriously affected the expression of dehydration-related genes [34]. Similar to previous results, the ability of Arabidopsis with ectopic FtbZIP12 expression to resist osmotic stress was significantly increased. In terms of the phenotype, the germination rate of seeds was increased, the inhibition of primary root growth was reduced, and seedlings were slightly damaged (Figures 4-6).
Under osmotic stress, excessive accumulation of ROS can damage the structure of plant cell membranes and be toxic to plants; the degree of plant damage can be expressed by the MDA content [37]. To resist these poisons, plants produce a series of response mechanisms. CAT is a superoxide that scavenges ROS; increasing its activity helps in such scavenging [38]. PRO is an important osmotic regulator; the increased accumulation of PRO under stress conditions not only helps to regulate intracellular and extracellular osmotic pressure, but also participates in the regulation of different metabolic processes [39]. CAT, MDA, and PRO are key indicators of osmotic stress. Studies have shown that the ABF subfamily can participate in plant defense responses to salt stress by regulating ROS, antioxidant enzymes, and PRO [40,41], which is consistent with the results of this study. Here, the transgenic plants were less injured by stress than the wild-type plants, possibly because FtbZIP12 promoted CAT and PRO accumulation in vivo, reduced MDA production, and decreased cell membrane damage ( Figure 6B-D).
In addition, ABFs have been shown to improve plant stress tolerance by acting on downstream stress genes and promoting their expression [41]. RD29B, LEA6, LEA7, and ERD1 are typical stress genes; their promoters contain ABRE, which can effectively bind to ABFs under osmotic stress [42][43][44]. Under normal conditions, the expression of these stress genes in transgenic plants did not differ from that in wild-type plants. After 6 h of osmotic stress, all plants had an upward trend, whereas the expression of wild-type plants was significantly lower than that of transgenic plants ( Figure 7D-F,H). Transgenic plants may upregulate these stress genes, which may protect cells from dehydration. The SOS pathway plays a crucial role in osmotic stress. Indeed, SOS3 received the Ca 2+ ion signal, formed a complex with SOS2, and stimulated the downstream SOS1, ultimately removing excess Na + from the cell and reducing the harm of ions to cells under osmotic stress [28,[45][46][47][48]. In this study, the expression levels of SOS1, SOS2, and SOS3 related to osmotic regulation in transgenic lines were significantly higher than those of the wild-type plants after 6 h of osmotic stress (Figure 7A-C,G). Therefore, FtbZIP12 may also improve the resistance of plants to osmotic stress by regulating the SOS pathway to stabilize the ion balance in plants.
Previous studies have shown that ABFs are one of the main participants in P5CS1-driven proline synthesis under extreme osmotic stress [44]. However, the role of ABF transcription factors (TFs) in regulating proline accumulation and proline-mediated osmotic stress adaptation remains unknown. The transcription of P5CS1 is a prerequisite for proline accumulation under osmotic stress [49,50]. In this study, although the expression pattern of P5CS1 under osmotic regulation was the same as that of other genes (RD29B and ERD1), the proline accumulation pathway of P5CS1 mediated by FtbZIP12 was complex. The relative expression of P5CS1 in overexpression lines was significantly higher than that in wild-type plants after 6 h of osmotic stress; after 3 days, the accumulation of proline was much higher than that of the wild-type plants (Figures 6C and 7A). This suggested that the expression of P5CS1 was regulated by FtbZIP12, thus increasing the accumulation of proline under osmotic stress. There are different views on the pathway of P5CS1-driven proline synthesis. Under low permeability treatment, the proline accumulation of ABA-deficient mutant (abi1) plants was less than that of wild-type plants [49,50], whereas the proline accumulation in ABA-deficient mutant (aba2) plants under simulated drought conditions was much higher than that under salt stress [51]. In this study, the transcriptions of P5CS1 and proline content under salt stress were much higher than those under simulated drought. Interestingly, FtbZIP12 was also more sensitive to NaCl under the induction of mannitol and NaCl in Tartary buckwheat (Figure 2), possibly because the regulation mechanism of proline varies greatly under different osmotic regulation pathways. FtbZIP12 may also be involved in the osmoregulatory pathway. It has also been reported that under extreme drought conditions, the proline accumulation of the four mutants (abf1, abf2, abf3, and abf4) was much slower than that of the wild-type plants in the early stage. Although proline accumulation can increase in the late stage, mutant plants have shown evident oxidative and membrane damage [44]. This also showed that proline plays an important role in the repair of plants at the initial stage of osmotic adjustment. Therefore, it is necessary to determine how FtbZIP12 drives P5CS1 to regulate proline accumulation and the mechanism underlying proline accumulation under different osmotic conditions.

Plant Growth and Treatments
The "Chuanqiao-2" seeds were germinated in clear water and then transplanted into pots with a perlite/soil (2:1 v/v) mixture. For the Arabidopsis culture, all seed lines were first treated with 75% ethanol for 15 min then washed numerous times with absolute ethanol, sown on half-strength MS medium for 2 weeks, and transplanted into small pots with a perlite/soil (2:1 v/v) mixture. All plants were grown normally in the greenhouse. The greenhouse environment was maintained at 70-75% relative humidity, a 16 h light/8 h dark light cycle, and 22 • C.

Vector Construction and Plant Transformation
Full-length FtbZIP12 coding sequences (CDS) were isolated from "Chuanqiao-2" and inserted into pCAMBIA1307 (35S: FtbZIP12-Myc), pGADT7, and pGBKT7 via homologous recombination. For 35S, FtbZIP12-Myc-overexpressing plants were generated using the floral dip method [52]. Positive plant selection using 100 mg/L of hygromycin and the pure T3 generation plants were used in subsequent experiments.

Analysis of the Transactivation of FtbZIP12
The combination of plasmids FtbZIP12-pGBKT7 + pGADT7/FtbZIP12-pGADT7 + pGBKT7 was co-transferred into yeast AH109; pGADT7 + pGBKT7 was negative. A single yeast colony was selected and cultured overnight in SD-LW liquid medium. The shaken yeast colonies were then diluted to 10 −1 , 10 −2 , and 10 −3 with 0.9% NaCl and grown on SD-LW and SD-ALWH plates to examine their growth. The plates were examined after 3 days of incubation at 30 • C and 3-6 clones were selected and tested for each experiment. This experiment was repeated three times. SD-LW: -trp and -leu; SD-ALWH: -trp, -leu, -his, and -ade.

Stress-Induced Treatment of Buckwheat and Arabidopsis
Two-week-old Tartary buckwheat was treated with 300 mM of NaCl and 375 mM of mannitol solution and its tissues were collected at 0, 3, 12, and 48 h following treatment. At the same time, treatment with 100 µM of ABA was conducted and tissues were harvested at 0, 1, 3, 6, and 12 h following treatment; the relative expression of FtbZIP12 was detected. The treatment of 0 h was used as the control. There were 30 plants for each treatment and five plants for each period. The Arabidopsis plants were treated with 300 mM of NaCl and 375 mM of mannitol solution and harvested at 0 and 6 h following treatment.
Each line had nine pots for a total of 27 pots, with six plants in each pot and three pots of each line for each treatment. The wild-type material was treated for 0 h as a control. The expression levels of AtRD29B, AtLEA6, AtLEA7, AtERD1, AtP5CS1, AtSOS1, AtSOS2, and AtSOS3 were investigated using RT-qPCR. The treatment methods were conducted as previously described [2]. Three biological replicates were used to detect changes in gene expression.

Real-Time PCR Analysis
Approximately 0.1 g of Tartary buckwheat or Arabidopsis leaves was completely ground into powder in liquid nitrogen and the total RNA was extracted using the RNAprep Pure Plant Plus kit (polysaccharide-and polyphenol-rich) (TIANGEN BIOTECH Co., Ltd., Beijing, China). The integrity of the extracted RNA was checked using 1% agarose gel electrophoresis and the RNA purity and concentration were measured using a spectrophotometer (Beijing Kaiao Technology Development Co., Ltd., Beijing, China). The RNA was reverse transcribed into cDNA using a HiScript ® III 1st Strand cDNA Synthesis Kit (+gDNA wiper) (Vazyme Biotech Co., Ltd., Nanjing, China) in a 20 µL reaction system. Primer-BLAST was used to design specific primers for the qPCR with FtH3 and Atactin 2 as the internal control gene (Table 1).

Homologous recombination
The quantitative real-time PCR (RT-qPCR) was performed using Cham Q Universal SYBR qPCR Master Mix (Vazyme Biotech Co., Ltd., Nanjing, China), 1.0 µL of template cDNA, 10.0 µL of 2×SYBR mix, 0.4 µL of each primer, and ddH 2 O to top off the reaction volume to 20 µL. The qPCR procedure was conducted as follows: 95 • C for 30 s followed by 40 cycles of 95 • C for 10 s, 56 • C for 30 s, and 95 • C for 15 s; then 60 • C for 60 s and 95 • C for 15 s (Bio-Rad, Hercules, CA, USA). The relative expression levels of the target genes compared to those of the internal control gene were calculated using the 2 −∆∆Ct formula [53]. Three biological replicates and three technical replicates were set up in this experiment.
All procedures were performed in accordance with the manufacturer's instructions. Primer Premier 5.0 software (Premier Corporation, Vancouver, British Columbia, Canada) was used to design specific primers for the PCR. All primers used for the PCR, vector construction, and RT-qPCR analysis are listed in Supplementary Table S1.

Stress Tolerance Assays of Transgenic Arabidopsis
To examine the effect of FtbZIP12 on the seed germination rate, seeds of all lines were sown on half-strength MS intensity agar plates with or without 125 mM of NaCl and 225 mM of mannitol. The seed sterilization procedure was the same as that given in Section 4.1. Each treatment was conducted in five replicates with 30 seeds per line. The germination numbers were counted daily for 7 days.
To determine the effect of osmotic stress on transgenic seedlings, we used the same training method for plants that grew normally for 3 days on half-strength MS medium. Plants with the same growth vigor were selected and transplanted onto plates with or without 125 mM of NaCl and 225 mM of mannitol and cultured upright for 7 days. There were three lines in each treatment and nine plants in total. The experiment was repeated three times. After 7 days, the root length of each plant was measured and the fresh weight of each line (three plants) were weighed.
A. thaliana was grown on half-strength MS medium for 2 weeks; we then selected plants with the same growth vigor, transplanted them into the soil, and left them to grow for 2 weeks. Seedlings were treated with 300 mM of NaCl for 3 weeks or no watering for 2 weeks. For plants under the drought treatment, we stopped watering 1 week in advance. Each treatment had six pots, totaling 54 pots, with six plants per pot. The seedlings treated for 10 days were used to measure physiological indicators; the amount of PRO, CAT, and MDA in all lines were detected using Boxbio kits (Beijing Boxbio Science & Technology Co., Ltd., Beijing, China) according to the manufacturer's instructions. Photos were taken two weeks and three weeks later to record the status.

Bioinformatics Analysis
The physical and chemical parameters of FtbZIP12 were predicted using the Expasy website (accessed on 25 April 2021 at https://web.expasy.org/protparam). The plant PLOC server (accessed on 6 July 2021 at http://www.csbio.sjtu.edu.cn/bioinf/plant/) was used to predict subcellular localization. We used MEGA 11 for the multiple sequence analysis and alignment as well as the phylogenetic tree analysis [54]. The N-J method (1000 times) was used to construct the phylogenetic tree.

Statistical Analysis
Each treatment had 3-5 independent biological repeats. Each repeat contained 3-6 plants. The experiment was conducted at least three times independently. Data statistics and calculations were conducted using Microsoft Excel (Microsoft, Redmond, WA, USA). To identify significant differences between two sets of data, a Student's t-test was performed using Excel. The error bars denote ±SDs. Significant differences are denoted by * p < 0.05 and ** p < 0.01. For multiple comparisons, one-way ANOVA with Duncan's multiple range test was performed using IBM SPSS Statistics 26.0 software (International Business Machines Corporation, New York, NY, USA). Significant differences between treatments are indicated by different letters. Line charts and column charts were drawn using GraphPad Prism 7.0 software (GraphPad Software, LLC, San Diego, CA, USA).

Conclusions
In this study, we identified that FtbZIP12, an ABF gene, was induced and expressed by ABA, NaCl, and mannitol. Heterotopic expression of FtbZIP12 in A. thaliana significantly improved the resistance of plants to osmotic stress due to the following possible reasons: (a) inducing P5CS1 expression and P5CS1-driven proline accumulation; (b) inducing the expression of genes related to the SOS pathway and starting SOS pathway; and (c) directly regulating the expression of stress-related genes ( Figure 8). However, further studies are warranted to determine the relationship between FtbZIP12 and the proline synthesis and SOS pathways. The preliminary results of this study on the function and osmotic regulation mechanism of FtbZIP12 provided novel insights into the osmotic stress regulation mechanism of Tartary buckwheat and will facilitate further research on ABFs.
Business Machines Corporation, New York, NY, USA). Significant differences between treatments are indicated by different letters. Line charts and column charts were drawn using GraphPad Prism 7.0 software (GraphPad Software, LLC, San Diego, CA, USA).

Conclusions
In this study, we identified that FtbZIP12, an ABF gene, was induced and expressed by ABA, NaCl, and mannitol. Heterotopic expression of FtbZIP12 in A. thaliana significantly improved the resistance of plants to osmotic stress due to the following possible reasons: (a) inducing P5CS1 expression and P5CS1-driven proline accumulation; (b) inducing the expression of genes related to the SOS pathway and starting SOS pathway; and (c) directly regulating the expression of stress-related genes ( Figure 8). However, further studies are warranted to determine the relationship between FtbZIP12 and the proline synthesis and SOS pathways. The preliminary results of this study on the function and osmotic regulation mechanism of FtbZIP12 provided novel insights into the osmotic stress regulation mechanism of Tartary buckwheat and will facilitate further research on ABFs.   FtbZIP12 may improve Tartary buckwheat stress resistance by regulating related downstream stress genes, the SOS pathway gene, and the P5CS1-driven proline synthesis pathway. Solid arrows indicate proven pathways whereas the dashed arrows show speculative pathways that remain to be confirmed.