Plants are continuously exposed to a variety of environmental challenges, including temperature fluctuations. Exposure to cold temperatures is one of the most serious risk factors for plant growth and development, and survival. In response to cold stress, plants have evolved a sophisticated signaling pathway to perceive, transduce, and coordinate their cellular and physiological responses. This complex network, known as the plant cold signaling pathway, is essential for plants to adapt and survive in cold environments. Understanding the mechanisms that underlie this signaling pathway is not only of fundamental importance in plant biology but also holds practical significance in agriculture, where cold stress can have detrimental effects on crop yield and quality.

The family of transcription factors, C-REPEAT BINDING FACTOR (CBFs), plays a crucial role in the cold response regulatory network, allowing plants to adapt to low temperatures. The expression of CBF gene family in plants is highly stimulated after 15-30 min of cold exposure, peaks at 1-3 h, and then rapidly decreased (Novillo et al. 2004). This temporal pattern is critical for promoting low-temperature tolerance. To promote low temperature tolerance, a number of cold regulatory (COR) genes are directly activated by CBF transcription factors, which play crucial roles in enhancing cold tolerance (Jia et al. 2016; Stockinger et al. 1997; Zhao et al. 2016). The expression of CBF genes in response to cold stress is regulated by a group of transcription factors (Shi et al. 2018). INDUCER OF CBF EXPRESSION1 (ICE1) and CALMODULIN BINDING TRANSCRIPTION ACTIVATORS (CAMTA) are among the positive regulators of CBFs that are active upstream (Chinnusamy et al. 2003; Doherty et al. 2009; Kim et al. 2013). The negative regulators of CBF expression play a significant role in fine-tuning the cold response of plants, which are ETHYLENE INSENSITIVE 3, MYB15, PHYTOCHROME-INTERACTING FACTOR 3 (PIF3), PIF4, and PIF7 (Agarwal et al. 2006; Jiang et al. 2017; Lee and Thomashow 2012; Shi et al. 2012).

The ICE1-mediated cold response regulatory network is a crucial component of the plant’s response to cold stress, and it plays a central role in regulating the expression of cold-responsive genes (Chinnusamy et al. 2003). When plants are exposed to low temperatures, ICE1 is activated in response to an influx of calcium and other early cold- sensing mechanisms. ICE1 encodes a MYC-like basic- helix-loop-helix transcription factor capable of binding to canonical MYC cis-elements (CANNTG) in CBF gene promoters, making it a pivotal regulatory factor of CBF genes (Chinnusamy et al. 2003; Ding et al. 2015). These ICE1-CBF modules hold significance in the plant’s response to low temperatures and environmental stress (Kim et al. 2015). The stabilization and activation of ICE1 are critical steps in initiating the plant’s response to cold. SnRK2.6/ OST1 (Ser/Thr protein kinase 2.6/OPEN STOMATA 1) interacts with and phosphorylates ICE1, leading to the stabilization and increased transcriptional activity of ICE1 (Ding et al. 2015). After accumulating, the mitogen-activated protein kinases MPK3 and MPK6, along with BRASSINOSTEROID-INSENSITIVE 2 (BIN2), phosphorylate ICE1 (Li et al. 2017; Ye et al. 2019). This phosphorylation leads to degradation of ICE1 via the 26S proteasome pathway, facilitated by the RING-type ubiquitin E3 ligase gene, high expression of the osmotically responsive gene 1 (HOS1) (Dong et al. 2006). Recent studies have confirmed this model as they demonstrate that, despite being a positive regulator of cold signaling, ICE1 is degraded in response to long-term cold treatment (Ding et al. 2015; Dong et al. 2006). To maintain the balance between cold tolerance and plant growth, it necessitates accurate regulation of ICE1 phosphorylation and CBF genes expression. However, the signaling pathway genes responsible for the response of P. ginseng plants to cold remain largely unknown.

Here, we aimed to discover the existence of evolutionarily conserved elements related to cold response, especially PgICE1 and PgCBFs, within the P. ginseng genome. These genes are integral to the plant’s response to low temperatures. In this study, we explored and evaluated the functional genomics of the components used in the cold signaling pathway of P. ginseng. Our investigation confirms that overexpression of the cold-inducible gene PgCBF3 in the atcbf1/2/3 triple mutant compensated for COR15A deficiency and enhanced salt stress tolerance. Furthermore, it has been discovered that PgICE1, which is located within the nucleus, contains evolutionary conserved sites for phosphorylation that can be influenced by the PgBIN2 and PgSnRK3 proteins. In a yeast two-hybrid assay, it was observed that these proteins have physical interactions with each other.

Plant materials and transgenic plants

A. thaliana Col-0 and cbf1/2/3 triple knockout plants were served as the genetic background and wild-type control in this study. The cbf1/2/3 triple knockout seeds were kindly provided by Prof. Byeong-ha Lee at Sogang University. The surface-sterilized seeds were placed on 1/2 MS agar plates and grown in a greenhouse under long-day conditions (16-h light/8-h dark cycles) at 22°C. To generate transgenic plants that overexpress HA-tagged PgCBF3 in cbf1/2/3 triple knockout plant backgrounds and FLAG-tagged PgICE1 (S90A, S264D) in Col-0 background. The pCB302ES (HA) and pBI121 (FLAG) plant expression binary vectors were used to clone the full-length cDNAs of PgCBF3, and PgICE1 (S90A, S264D) as described previously (Hong and Ryu 2022; Ryu et al. 2007). All phenotypic studies were performed on homozygous T3 plants, and the Agrobacterium-mediated floral dip method with the GV3101 strain was used to generate Arabidopsis overexpressing plants. To investigate the effect of salt stress on root growth, plants were cultivated on half-strength Gamborg B5 plates for three days before being transferred to a 100 mM NaCl plate for five days.

Protein sequence alignments

The protein sequences of AtCBF1 and AtCBF3-related genes, including PgCBF1 (Pg_S1369.3) and PgCBF3 (ISO_ 033347), and AtICE1-related genes, including PgICE1 (Pg_S0055.2) were selected. The protein sequence was aligned using SMS, an online program available at http://www.bioinformatics.org. For identity or similarity coloring to be added, the percentage of sequences that must agree is 70%.

qRT-PCR analysis

Total RNA was extracted using the easy-spin Total RNA Extraction Kit (iNtRON) to measure transcript expression levels. Following this, 1 µg of RNA was used to create double-stranded cDNA with TOPscript RT DryMIX (Enzynomics). The cDNA was subsequently analyzed via real-time quantitative PCR, utilizing an Applied Biosystems Quant Studio 3 device and SYBR Green Real-time PCR Master Mix (Applied Biosystems). Primer lists are followed: PgACT1, 5’-TGGCATCACTTTCTACAACG-3’ and 5’-TTTGTGTCATCTTCTCCCTGTT-3’; AtACT2, 5’-CAGTGTCTGGATCGGAGGAT-3’ and 5’-TGAACAATCGATGGACCTGA-3’; PgCBF1, ; 5’-TGACGGAGGAGAAGATAGGAGTTG-3’ and 5’-TCAATTAAACCCGGCATGTTAA-3’ ,; PgCBF3, 5’- TGCCCGGGTTGATTACAAGT -3’,; 5’-TCGGCTCCAAATTCCATGTC -3’. The gene-specific primer sets for AtCOR15A and AtRD29A were previously reported (Li et al. 2017).

Protoplast transient expression and yeast two hybrid assay

The complete cDNA sequences of PgICE1 were inserted into a plant expression vector containing GFP tags at the C-terminus, driven by the 35S:C4PPDK promoter, using previously established methods (Ryu et al. 2007). For the protoplast transient expression assays, approximately 4 × 10⁴ protoplasts were transfected with 20 µg of plasmid DNA, followed by incubation under constant light conditions at 20°C for 6 hours. For subcellular localization, GFP- tagged constructs were transfected into the protoplasts. AtARR2-RFP (Ryu et al. 2007) served as a nuclear marker. The fluorescence of GFP and RFP was observed via a fluorescence microscope (Olympus, BX53). To identify physical interactions between PgICE1 and PgBIN2; PgBIL1; PgBIL2; PgSnRK1; PgSnRK2; PgSnRK3, AH109 yeast strains were co-transfected with pGADT7-PgICE1 and pGBKT7- PgBIN2; PgBIL1; PgBIL2; PgSnRK1; PgSnRK2; PgSnRK3 (Hong et al. 2018; Hwang et al. 2020). Clones exhibiting favorable interactions were selected on synthetic medium containing 1 mM 3-aminotriazole (3-AT) without Leu, Trp, and His.

In the model plant Arabidopsis, three C-repeat binding proteins, including CBF1, CBF2, and CBF3 have been characterized as key regulators for cold acclimation (Gilmour et al. 2004; Medina et al. 1999; Zhao et al. 2016). To identify the cold-response related CBFs in P. ginseng genome, we initially performed a phylogenetic analysis. Several CBF-like sequences were identified from three genomic data sets of P. ginseng, which were integrated with our previous study (2017) as well as studies conducted by Waminal et al. (2018) and Xu et al. (2017). The protein sequences of PgCBF1 (Pg_S1369.3, Xu et al. 2017) and PgCBF3 (ISO_ 033347, Jo et al. 2017) were analyzed and aligned with the protein sequences of AtCBF1 and AtCBF3, respectively (Fig. 1A and B). Two PgCBFs were found to have more than 60% similarity in their amino acid sequences with AtCBFs. Additionally, these two PgCBFs displayed evolutionary conservation of the AP2 (APETALA2) DNA binding domains (Fig. 1A and B). The AtCBF genes exhibited a rapid and transient induction, typically attaining their peak expression levels within 1 to 2 hours of exposure to cold exposure (Gilmour et al. 1998; Novillo et al. 2004; Zarka et al. 2003). To validate the cold responsiveness of the identified PgCBFs, we assessed the expression levels of them in shoot tissues of P. ginseng that were exposed to cold treatment. As shown in Fig. 1C, the expression of PgCBF1 was increased by about 10-fold and PgCBF3 by about 120-fold by 2 hours of exposure at 4°C. These results suggest that the evolutionary conserved PgCBF1 and 3 would play a critical role in cold acclimation of P. ginseng.

Figure 1. Identification of cold-inducible PgCBF1 and PgCBF3. (A, B) Protein-sequence alignment of PgCBF1 (A) and PgCBF3 (B) with their orthologs AtCBF1 and AtCBF3, respectively. The yellow box indicates the evolutionarily conserved AP2/ERF domains. (C) Cold treatment enhanced the expression levels of PgCBF1 and PgCBF3. The relative transcript levels of PgCBF1 and PgCBF3 in 2-year-old P. ginseng seedlings treated with or without 4°C for 2 h were determined by qRT-PCR. PgACT1 was used as an internal control. The error bars indicate the S.E. (n = 3), and student’s t-test was performed (**, P < 0.01). (D) PgCBF3 overexpression in the cbf1/2/3 knock-out mutant background considerably activates the expression levels of cold-inducible COR15A. Two independent 35S:Pg:CBF3 #1 and #2 were subjected to cold treatment for 2 h. AtACT2 was used as an internal control. Error bars indicate the S.E. (n = 3), and student’s t-test was performed (*; P < 0.05, **; P < 0.01, ns: non-significant)

To determine the physiological role of PgCBF in the cold stress response, we generated PgCBF3 overexpressing plants in a cbf1/2/3 triple knockout Arabidopsis mutant background (Zhao et al. 2016). CBF-mediated upregulation of CORs, including COR15A and RD29A, is critical for cold acclimation and freezing stress tolerance (Provart et al. 2003; Steponkus et al. 1998). As previously evaluated (Zhao et al. 2016), the cold stress-dependent induction of COR15A was completely abolished in the cbf1/2/3 triple mutant (Fig. 1D). However, in two independent 35s:PgCBF3 cbf1/2/3 transgenic plants, the COR15A expression was dramatically enhanced, almost independent of the cold treatment (Fig. 1D). These findings provide evidence that PgCBF3 plays a crucial role in the expression of genes associated with cold acclimation and likely acts as a major regulator for enhancing cold stress tolerance in P. ginseng.

We next investigated the upstream regulator of PgCBFs. Cold stress-induced CBFs expression is directly regulated by ICE1, a MYC-type bHLH transcription factor (Chinnusamy et al. 2003). A protein sequence alignment revealed that a putative PgICE1 exhibited sequence similarities with AtICE1 (Fig. 2A). We validated the presence of a well- conserved bHLH DNA binding domain and two independent phosphorylation sites, Ser 90 and Ser 264, which were controlled by BIN2 (indicated by a red asterisk) and SnRK2.6 (OST1; indicated by a green asterisk), respectively (Fig. 2A, ref). Direct phosphorylation of ICE1 by the stress signaling-related protein kinases such as MAPKs, BIN2, and SnRK2.6 is an important mechanism in fine- tuning ICE1 activity for freezing tolerance (Ding et al. 2015; Ye et al. 2019). PgICE1 phosphorylation residue conservation assumes a protein-protein interaction between PgICE1 and its binding protein kinases. To test this possibility, we conducted a yeast-two hybrid assay using PgICE1 in combination with previously reported PgBIN2s and PgSnRKs (Hong et al. 2018; Hwang et al. 2020). Although there is functional redundancy of PgGSK3s in the brassinosteroid response in P. ginseng (Hwang et al. 2020), it is noteworthy that only PgBIN2 exhibited a physical interaction with PgICE1, but not other PgGSKs (Fig. 2B). The protein interaction analysis of ABA-signaling related PgSnRKs revealed that only the interaction between PgSnRK3 and PgICE1 could be successfully determined (Fig. 2C). However, the interaction between PgSnRK1 and PgSnRK2 could not be assessed due to their auto transcriptional activity in yeast cells (Fig. 2C). Theese findings suggest that the activity of PgICE1 is likely influenced by regulation through multiple signaling pathways similar to a model Arabidopsis plant.

Figure 2. PgICE1 interacts with PgBIN2 and PgSnRK3. (A) Protein-sequence alignment of PgICE1 with AtICE1. The red box indicates the evolutionarily conserved bHLH domain. (B, C) PgICE1 physically interacted with PgBIN2 (B) and PgSnRK3 (C) in a yeast two-hybrid assay. The yeast strains were selected on synthetic medium lacking Leu, Trp, and His (-LTH) containing 1 mM 3-AT or medium lacking Leu and Trp (-LT)

We then tested whether the PgICE1 is localized in the nucleus for its transcription factor activity. GFP-tagged PgICE1 was co-expressed with a nuclear localized AtARR2- RFP in Arabidopsis mesophyll protoplasts. As presented in Fig. 3A, the PgICE1-GFP signal was colocalized with a nuclear AtARR2-RFP, indicating that PgICE1 functions in the nucleus. To validate the PgICE1 functions in cold stress signaling pathways, we generated transgenic plants overexpressing PgICE1 and gain-of-functional phosphorylation mutants including PgICE1 S90A and PgICE1 S264D. In 35S:PgICE1 transgenic plants, cold-responsive COR15A and RD29A expression was enhanced compared to wild- type control plants (Fig. 3B). Consistently, gain-of-function PgICE1 S90A and PgICE1 S264D overexpressing plants exhibited higher COR gene expression levels than wild- type PgICE1 overexpressing plants (Fig. 3B). These data suggest that phosphorylation-mediated regulation of PgICE1 activity is important for the activation of cold-responsive gene expression.

Figure 3. Nuclear-localized PgICE1 regulates cold-responsive gene expression. (A) PgICE1 is localized in the nucleus. 35S-PgICE1-GFP was co-transfected into Arabidopsis mesophyll protoplasts with 35S-AtARR2-RFP as a nuclear marker. (B) PgICE1 overexpression and gain-of-function phosphorylation mutant forms (PgICE1 S90A and PgICE1 S264A) in Arabidopsis enhanced the expression of cold-responsive COR15A and RD29A. The transcript levels in the overexpression lines were determined by qRT-PCR. Error bars indicate the S.E. (n = 3). Different lowercase letters indicate statistically significant differences P < 0.05; one-way analysis of variance (ANOVA), followed by Tukey’s multiple range test

Finally, we evaluated whether PgCBF3 and PgICE1, which are involved in cold signaling, can enhance general abiotic stress tolerance. Since cold stress signaling pathways are highly correlated with salt stress tolerance (Teige et al. 2004; Xiong et al. 2002), we tested the effectiveness of overcoming salt stress-mediated root growth inhibition by the COLD signaling genes in P. ginseng. The cbf1/2/3 triple mutant was highly sensitive to salt stress in root growth, but the growth inhibition was completely recovered by PgCBF3 overexpression (Fig. 4). Similarly, all PgICE1 and its gain-of-function mutant overexpression transgenic plants showed a higher salt stress tolerance phenotype than wild- type Col-0 control plants (Fig. 4). Taken together, our results identify evolutionary conserved PgCBFs and PgICE1 in P. ginseng genome and demonstrate their role in enhancing abiotic stress tolerance.

Figure 4. PgCBF3 and PgICE1 overexpression enhances abiotic salt stress tolerance. Three-day-old seedlings of Col-0, cbf1/2/3, 35S:PgCBF3/ cbf1/2/3, 35S:PgICE1, 35S:PgICE1 S90A, and 35S:PgICE1 S264D were transferred onto 100 mM NaCl containing MS agar medium plates. The root length was measured after another 5 days, and relative growth was compared with that of seedlings grown on MS medium (error bars indicate S.E. n = 15). Student’s t-test was performed (*; P < 0.05, **; P < 0.01, ns: non-significant)

In summary, our data shed light on the functional roles of PgICE1 and PgCBFs in the P. ginseng. The analyzed cold signaling components of P. ginseng have been shown to be functionally significant in the plant’s responses to cold and abiotic stress. Understanding these mechanisms not only adds to our knowledge of this valuable medicinal plant but also offers insights into strategies for enhancing its resilience in the face of environmental challenges, bridging the gap between fundamental plant science and practical applications in agriculture and ecology.