The deduced amino acid sequence of PsbHLH1 was analyzed via the Simple Modular Architecture Research Tool (SMART) (Schultz et al., 1998). Multiple alignment of bHLH amino acid sequences was performed by ClustalW (Higgins et al., 1994). A phylogenetic tree was constructed by the neighbor-joining method using a bootstrap test with 1,000 replications in Phylip (http://evolution.gs.washington.edu/phylip.html). The phylogenetic tree was visualized in MEGA5 (Tamura et al., 2011). The analysis of the domain architecture of the PsbHLH1 protein was performed by SMART (Letunic et al., 2021). The TF binding motifs in the promoter and 5′ untranslated (UTR) regions of phenylalanine ammonia-lyase (PAL), 4-coumarate-CoA ligase (4CL), pinosylvin synthase (STS), pinosylvin O-methyltransferase (PMT), and acetyl-CoA carboxylase (ACC) in Korean pine (Pinus koraiensis) or other pine species were analyzed using the Softberry NSITE Program (version 5.2013, http://www.softberry.com).

To isolate the genes encoding pinosylvin synthase (PsSTS), pinosylvin methyltransferase (PsPMT), and TF (PsbHLH1) in P. strobus, mRNA was extracted from cultured cells using the RNeasy® Plant Mini Kit (Qiagen, Germany) and cDNA was prepared using reverse transcriptase. The primers used to obtain the full-length cDNA of these three genes were prepared as shown in Supplementary Table S1 , and then the cDNA was subsequently amplified via PCR using the template to obtain the full-length cDNA.

A total of three binary vectors were constructed: a vector for the PsbHLH1 gene; a vector containing two genes, PsSTS and PsPMT, for coexpression; and a vector containing three genes, PsSTS, PsPMT, and PsbHLH1, for coexpression. The first vector used the hygromycin phosphotransferase (HPT) gene as the selection marker, and the second and third vectors used the Basta-resistant gene (BAR) as the selection marker. These three binary vectors were mobilized into Agrobacterium tumefaciens GV3101 for plant transformation.

To obtain transgenic calli, mature zygotes of Korean pine (Pinus koraiensis) seeds were used. Mature zygotes were isolated from seeds following the methods of Kim et al. (2022). The culture conditions for inducing calli from the culture of zygotic embryos of P. koraiensis were the same as those in Kim et al. (2022). Mature zygotic embryos were soaked in Agrobacterium suspended in 1/2 LV medium (Litvay et al., 1985) supplemented with 50 μg/ml acetosyringone for 10 min and then subsequently placed on sterilized filter paper. After approximately 1 h, the zygotic embryos were transferred to 1/2 LV medium supplemented with 0.5 mg/L 2,4-D and 0.5 mg/L BA and cocultured for 3 days. The embryos were then transferred to the same medium supplemented with 500 mg/L cefotaxime and cultured for 2 weeks, followed by 2 passages at 2-week intervals to eliminate Agrobacterium. Transgenic calli induced from zygotic embryos were then selected by transfer to medium supplemented with 10 mg/L hygromycin and subculturing on the same medium at 2-week intervals. When a hygromycin-resistant callus was obtained, a callus from each zygotic embryo line was selected as an independent line. The same protocol for the construction of transgenic calli was used for the other two vectors containing the BAR gene marker, and 20 mg/L Basta was added for the selection of transgenic calli.

Genomic polymerase chain reaction (PCR), reverse transcription PCR (RT-PCR), and reverse transcription quantitative PCR (RT-qPCR) were performed to investigate the introduction and expression of the genes in the transgenic pine callus and transgenic tobacco plants. After the cultured cells were transformed with the pine PsbHLH1 gene, genomic DNA was isolated from the transgenic calli using the Genomic DNA Extraction Kit Mini (RBC bioscience, New Taipei City, Taiwan). Genomic PCR was performed using PsbHLH1 and HPT gene primers as shown in Table S1 . The selected lines were subjected to RT-PCR and RT-qPCR using cDNA obtained from the transgenic lines as a template to investigate the expression of the introduced genes. The RT-PCR and RT-qPCR information for the amplification of the PsbHLH1 and HPT genes are shown in Table S1 . For RT-qPCR, the pine β-actin gene was used as a reference gene.

A. tumefaciens harboring two binary vectors, as shown in Figures 1A, B , was harvested and prepared for infiltration. The bacterial suspension was infiltrated into the intercellular spaces of the leaf tissue of tobacco (Nicotiana tabacum cv Xanthi) using a needleless syringe. After infiltration, the infiltrated tobacco plants were usually kept in a growth chamber for 2 days. The infiltrated leaf tissues were harvested and subjected to analysis for the production of pinosylvin stilbenes.

Calli from the untransformed control and the three transgenic lines were sampled for GC analysis, then dried in a 50°C dry oven and ground into powder. One gram of each powder was placed in 100% MeOH, sonicated for 30 min at 40°C, and centrifuged at 15000 ×g. The supernatant was filtered through a membrane filter. For GC analysis of tobacco leaf tissues, control and infiltrated tobacco leaves were dried in a 50°C dry oven and ground into powder. One hundred milligrams of each powder were placed in 100% MeOH, sonicated for 30 min at 40°C, and centrifuged at 15000 ×g, after which the supernatant was filtered through a membrane filter.

The filtered aliquots were analyzed by GC (Agilent 7890A) with an Agilent 5975C MSD system equipped with an HP-5MS capillary column (30 m × 0.25 mm, film thickness 0.25 mm). The injection temperature was 250°C. The column temperature was as follows: 70°C for 4 min, 220°C at a rate of 5°C min^-1, heating at 4°C min^-1 up to 320°C, and holding at 320°C for 5 min. The carrier gas was He. The interface temperature was 300°C, and split/splitless injection (10:1) was used. The ionization chamber temperature was 250°C, and ionization was performed by electron impact at 70 eV. The standards of PME and DPME used in GC-MS analysis were purchased from Sigma–Aldrich Co. The experiment was performed in triplicate and repeated three times.

Quantitative data are expressed as the mean values ± standard error (SE). All the statistical analysis were performed with SPSS software (SPSS Science, Chicago, IL, USA). Statistical differences among means were calculated using one-way ANOVA followed by Duncan’s post hoc analysis.

The PsbHLH1 gene was cloned from P. strobus and deposited in the NCBI GenBank database under accession number GIIE01106737.1. The open reading frame (ORF) of PsbHLH1 is 1479 bp in length and encodes a bHLH family protein of 492 amino acids, with a calculated molecular weight of 54.31 kDa. BLAST analysis revealed that PsbHLH1 shares the highest identity with the functionally unknown Pinus massoniana bHLH15 and 55% identity with the Arabidopsis thaliana AtbHLH041, which belongs to the group IVd subfamily. Domain architecture analysis via SMART (Letunic et al., 2021) revealed that the PsbHLH1 protein belongs to the MYC-type bHLH domain ( Figure 2A ). The bHLH domain comprises two alpha helices connected by a nonconserved loop region and is used for dimerization, and the basic domain is used for DNA binding (Jones, 2004). Multiple sequence alignment revealed that PsbHLH1 contains a conserved bHLH domain that has four conserved regions, namely, the basic region, two helixes, and a loop region ( Figure 2B ).

Since the genomic sequence of Pinus strobus has not been published, we analyzed the promoters of the PAL, 4CL, STS, PME, and ACC genes in the Pinus lambertiana whole-genome shotgun sequence (GenBank: LMTP000000000.1) using NSITE-PL (version 5.2013, at http://www.softberry.com). The promoter regions of the PAL, 4CL, STS, and PME genes in Pinus lambertiana lacked a G-box motif for MYC-type bHLH protein binding. However, only the ACC gene had a MYC2 binding G-BOX motif (CACGTG) in the 5-UTR, not in the promoter region of the ACC gene. The regulatory elements in the untranslated regions (UTRs) of the ACC genes of all 5 selected Pinus species (P. koraiensis, P. strobus, Pinus sylvestris, Pinus taeda, and Pinus flexilis) were also examined. All of these genes had a MYC2-binding G-BOX motif (CACGTG) in the 5’ -UTR ( Supplementary Figure S1 ).

Phylogenetic analysis based on the neighbor-joining algorithm revealed that PsbHLH1 was first clustered with a functionally unknown bHLH15 gene of P. massoniana and was also related to a subgroup (IVd) that included the functionally unknown AtbHLH041 gene ( Figure 3 ) (Heim et al., 2003). AtbHLH041 is a transcriptional repressor of root stem cell factors during the establishment of auxin-induced callus pluripotency (Xu et al., 2024). However, PsbHLH1 is an independent group in which Plagiochasma appendiculatum bHLH1 protein regulates bis-bibenzyl synthesis (Wu et al., 2018), Solanum tuberosum bHLH1 protein regulates phenylpropanoid metabolism (Payyavula et al., 2013), Chrysanthemum x morifolium bHLH2 is involved in anthocyanin biosynthesis (Xiang et al., 2015), and Malus domestica bHLH3 activates anthocyanin biosynthesis (Xie et al., 2017).

A vector for overexpressing of the PsbHLH1 gene with the hygromycin selection marker gene (HPT) was constructed for the transformation of P. koraiensis ( Figure 4A ). Hygromycin-resistant calli were obtained by the subculture of calli derived from zygotic embryos after cocultivation with Agrobacterium harboring the PsbHLH1 gene ( Figures 4B, C ). The final selection of transgenic callus lines was obtained after more than five times subcultures on a hygromycin-containing medium. Three hygromycin-resistant calli were selected from different zygotic lines.

To confirm the gene introduction in these transgenic lines, primers for the HPT and PsbHLH1 genes were prepared, and PCR was performed on genomic DNA from the control untransformed callus and transgenic lines. The PsbHLH1 gene and the selective marker HPT gene were not amplified from the control, but in the transgenic line, the HPT and PsbHLH1 genes were amplified ( Figure 4D ). We also extracted mRNA from the calli of the control and transgenic lines and performed RT-qPCR using cDNA as a template; the results revealed that the expression level of the PsbHLH1 gene was strongly increased in all three transgenic lines ( Figure 4E ).

Three lines of transgenic calli overexpressing PsbHLH1 and wild-type control calli were analyzed via GC/MS. In the control, untransformed callus, a small PME peak was detected at a retention time of 36.8 min, and a very small peak of DPME was identified at a retention time of 32 min ( Figure 5A ). The calli of the transgenic lines were characterized by clearly enhanced PME and DPME peaks in the GC chromatogram ( Figures 5B–D ), and quantification of these two substances revealed a more than 32-fold increase in PME and a 10-fold increase in DPME compared to those in the control ( Figure 5H ). The two substances were identified by comparing the retention times of the DPME and PME standards ( Figure 5E ) and by comparing the mass fraction patterns of these two substances ( Figures 5F, G ).

The genes in the PME and DPME biosynthesis pathway included the PAL, 4CL, STS, PME, and ACC genes, as shown in Figure 6 . Pinosylvin is produced by combining one molecule of the cinnamoyl-CoA gene and three molecules of malonyl-CoA, and the ACC gene is involved in the synthesis of malonyl-CoA. The STS gene makes the backbone of pinosylvin, and the PMT gene attaches the methyl group to pinosylvin. In a transgenic P. koraiensis callus line (line 3) in which the PsbHLH1 gene was overexpressed, the expression of the PAL and 4CL genes was unchanged in the transgenic line compared to the control ( Figure 6 ). On the other hand, the STS and PMT genes of P. koraiensis were more highly expressed in the transgenic lines than in the control. In addition, the expression of the ACC gene involved in malonyl-CoA synthesis in of P. koraiensis was significantly greater than that in the control ( Figure 6 ).

A vector for the overexpression of both the PsSTS and PsPMT genes with the BAR selection marker gene (BAR) was constructed for the transformation of P. koraiensis ( Figure 7A ). Basta-resistant calli were obtained using a protocol similar to that used for PsbHLH1 overexpression. The final selection of three Basta-resistant callus lines was confirmed by RT-PCR of the introduced genes. The expression of the PsSTS and PsPMT genes in P. koraiensis calli was detected in all three transgenic lines but not in the wild-type control ( Figure 7B ).

Three lines of transgenic calli lines overexpressing the PsSTS and PsPMT genes and the wild-type control were analyzed via GC-MS. The PME and DPME peaks in the GC chromatogram were not different between the transgenic lines and the untransformed control ( Figures 7C–G ), and quantification of these two substances revealed no significant difference between them ( Figure 7H ).

To analyze the production of PME and DPME in transgenic tobacco leaves by transient expression, the two vectors used are described above ( Figures 4A , 7A ), and the vector coexpressing three genes, PsbHLH1, PsSTS, and PsPMT ( Figure 1A ), was used. Tobacco leaves were infiltrated with Agrobacterium solution using a transient expression system with these two vectors, after which then the leaves were dried and analyzed via GC/MS after 2 days. No pinosylvin derivatives (PME or DPME) were detected in wild-type tobacco leaves ( Figure 1B ). In the transient expression of PsbHLH1 ( Figure 1C ), or PsSTS and PsPMT genes ( Figure 1D ), no pinosylvin derivatives (PME or DPME) were detected. On the other hand, transient expression of the three PsbHLH1, PsSTS, and PsPMT genes in tobacco leaves resulted in the production of PME ( Figure 1E ), which was confirmed by comparing the retention time ( Figure 1F ) and mass fraction pattern ( Figure 1G ) of the standard PME compound.

Transcriptome analysis following PWN infection of Pinus strobus revealed that the PsbHLH1 TF gene had the highest expression value upon PWN infection among all the TF genes analyzed, and RT-qPCR analysis after PWN infection also revealed PsbHLH1 as the most promising TF for responding to PWN infection (Hwang et al., 2021). In the present study, we investigated whether this gene regulates the biosynthesis of pinosylvin stilbenes in PWN-susceptible Korean pine (P. koraiensis). In transgenic P. koraiensis calli overexpressing the PsbHLH1 gene, the production of PME and DPME significantly increased. This result suggested that the PsbHLH1 gene is an important TF for the regulation of pinosylvin stilbene biosynthesis. To analyze whether the overexpression of the PsbHLH1 gene affects the genes involved in pinosylvin biosynthesis in P. koraiensis, the expression of P. koraiensis genes in transgenic P. koraiensis callus lines was analyzed via RT-qPCR. The analysis revealed that the expression of the PkACC gene of P. koraiensis, which is involved in the synthesis of malonyl-CoA, an important precursor in the synthesis of pinosylvin, was significantly increased. Moreover, the expression of the PkSTS and PkPMT genes of P. koraiensis significantly increased, while the expression of the PkPAL and Pk4CL genes of P. koraiensis did not change. These results suggest that the ectopic expression of the PsbHLH1 gene in transgenic P. koraiensis calli affects the expression of the PkACC, PkSTS, and PkPMT genes, which results in increased production of pinosylvin derivatives in P. koraiensis calli.

The MYC2 bHLH TF is known to be a master regulator that activates defense gene expression and responds to JA and has been reported to be involved in the regulation of specialized metabolites (Du et al., 2017; López-Vidriero et al., 2021; Luo et al., 2023). PsbHLH1 is a MYC-type bHLH. PsbHLH1 shares the highest identity with Arabidopsis thaliana AtbHLH041, which belongs to the group IVd subfamily. The role of subgroup IVd within the bHLH TF family in the biosynthesis of secondary compounds has not been characterized.

We analysed the promoter and UTR regions of the PAL, 4CL, STS, PME, and ACC genes in Pinus lambertiana and found that the promoter regions of the PAL, 4CL, STS, and PME genes lacked a bHLH binding domain. On the other hand, the ACC gene has a bHLH binding G-box motif (CACGTG) in the 5-UTR. This G-box motif was commonly present in the UTRs of all five Pinus species ( Supplementary Figure S1 ). MYC-type bHLH TFs bind to the G-box (5′-CACGTG-3′), a DNA motif typically recognized by bHLH proteins from plants, yeast, and animals (López-Vidriero et al., 2021). This result suggested that the PsbHLH1 TF may bind to the UTR of the ACC gene and regulate its expression. ACC is an enzyme that catalyzes the conversion of acetyl-CoA to malonyl-CoA which is an essential building block for the biosynthesis of fatty acids, polyketides, stilbenoids, and flavonoids (Wu et al., 2017). Indeed, malonyl-CoA is an important precursor in the synthesis of resveratrol stilbene (Xu et al., 2011; Yang et al., 2015). In the metabolic engineering of Escherichia coli for the synthesis of the pinosylvin stilbene, malonyl-CoA availability in the heterologous host is thought to be the main bottleneck for pinosylvin production (van Summeren-Wesenhagen and Marienhagen, 2015).

No information is available on the transcriptional regulators involved in pinosylvin stilbene synthesis in plant species. There is some information on the TFs involved in resveratrol synthesis in grapes. Two R2R3-type MYB genes, MYB14 and MYB15, have been identified as transcriptional regulators of resveratrol stilbene biosynthesis in Vitis vinifera (Höll et al., 2013). WRKY (VviWRKY03) and MYB TF (VviMYB14) in Vitis vinifera have been reported to combinatorically regulate the stilbene synthase gene (Vannozzi et al., 2018). In particular, VviWRKY24 seems to act as a singular effector in the activation of the stilbene synthase (STS) promoter, while VviWRKY03 acts through a combinatorial effect with VviMYB14 (Vannozzi et al., 2018). Overexpression of VaMyb40 and VaMyb60 resulted in stilbene biosynthesis in cell cultures of Vitis amurensis (Ananev et al., 2022). Although the STS genes for pinosylvin synthesis in pine plants and resveratrol synthesis in dicots are holomorphic, they utilize different precursors to synthesize the two stilbenes. Pinosylvin stilbene is synthesized by the coupling of one molecule of cinnamoyl-CoA and three molecules of malonyl-CoA, whereas resveratrol is synthesized by the coupling of coumaroyl-CoA and malonyl-CoA (Valletta et al., 2021).

In transgenic P. koraiensis calli overexpressing the PsSTS and PsPMT genes, the production of PME and DPME in transgenic calli was unaffected. Similarly, the overexpression of the Pinus sylvestris STS gene in poplar did not result in the synthesis of pinosylvin or its derivatives (Seppänen et al., 2004). Malonyl-CoA is a key precursor of pinosylvin bioproduction (Liang et al., 2016). In the metabolic engineering of E. coli, malonyl-CoA availability is thought to be the main bottleneck for pinosylvin production (van Summeren-Wesenhagen and Marienhagen, 2015). ACC is an important enzyme that catalyzes the production of malonyl-CoA (Wu et al., 2017). Interestingly, the PsbHLH1 TF may bind to the MYC2-binding G-box motif (CACGTG) in the UTR of the ACC gene. Thus, a transcriptional regulator, PsbHLH1, may be essentially needed for the malonyl-CoA precursor supply in the biosynthesis of pinosylvin stilbenes. Our study is the first to report the production of pinosylvin stilbenes by heterologous expression of genes in plants.

In our experiment, the transient expression of the PsSTS and PsPMT genes in tobacco resulted in no production of pinosylvin stilbenes. The same result was obtained for the transient expression of the PsbHLH1 in tobacco. Like in our study, no pinosylvin stilbene was produced when the Pinus sylvestris STS gene was overexpressed in poplar (Seppänen et al., 2004). Seppänen et al. (2004) suggested that the failure of stilbene synthesis by heterologous expression of the P. sylvestris STS gene in poplar may occur by restriction of the cinnamoyl-CoA supply or metabolic channeling. Moreover, the overexpression of pinosylvin synthase genes (PjSTS1a, PjSTS2, or PjSTS3) derived from the coniferous tree (Picea jezoensis) in transgenic calli of grape plants resulted in the production of resveratrol-related stilbenes but not pinosylvin stilbenes (Suprun et al., 2020). Tobacco plants do not produce stilbenes such as pinosylvin or resveratrol, and may not harbor stilbene synthase genes. Thus, the lack of pinosylvin stilbenes during transient expression of PsbHLH1 in tobacco may be expected. However, in our study, coexpression of the PsbHLH1 gene together with the PsSTS and PsPMT genes in tobacco leaves resulted in the production of pinosylvin derivatives. The requirement of the PsbHLH1 gene for the production of PME in tobacco leaves by transient expression indicated that PsbHLH1 may play an important regulatory role in the biosynthesis of pinosylvin stilbenes. Analysis of the molecular mechanism of the PsbHLH1 protein may shed light on its regulatory role in PME and DPME biosynthesis.

In conclusion, although the two genes, PsSTS and PsPMT, are known to be key players in pinosylvin stilbene biosynthesis, the overexpression of these two genes did not result in either PME in either P. koraiensis callus or tobacco leaves, but additional expression of PsbHLH1 successfully induced pinosylvin stilbene production. These results indicate that PsbHLH1 is an important TF for pinosylvin stilbene biosynthesis, and our study is the first to report the production of pinosylvin stilbene using plant genetic engineering technology. These findings suggest that introducing the PsbHLH1 gene into PWN-susceptible pine species will increase the amount of pinosylvin stilbene compounds, which are highly toxic to PWN, and thus increase its resistance to PWN.