Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2025 Dec 23;16:2620. doi: 10.1038/s41598-025-33501-w

Red light-driven enhancement of calycosin biosynthesis in Astragalus membranaceus via functional characterization of AmI3’H

Ji Won Seo 1, Won Hyeok Choi 1, Suho Lim 2, Eun Soo Seong 2,✉
PMCID: PMC12820219  PMID: 41430428

Abstract

Astragalus membranaceus is a well-known medicinal plant rich in bioactive compounds, particularly the isoflavonoid calycosin, which exhibits diverse pharmacological properties. However, there is limited research on the influence of light conditions on calycosin biosynthesis and the underlying molecular mechanisms. In this study, we exposed in vitro plantlets of A. membranaceus to different LED light sources (white, red, and blue) and found that red light treatment significantly enhanced both biomass accumulation and calycosin content. Through transcriptome analysis combined with real-time PCR we identified several key genes involved in calycosin biosynthesis, among which AmI3’H showed the most significant upregulation under red light. Structural analysis and molecular docking indicated that AmI3’H, a member of the CYP81 clan of cytochrome P450 enzymes, interacts with formononetin, a precursor of calycosin. To verify its functional role, transient overexpression of AmI3’H was performed in Nicotiana benthamiana using Agrobacterium-mediated infiltration along with exogenous formononetin. The co-infiltration resulted in calycosin production exclusively at 24 h post-infiltration, accompanied by elevated antioxidant activity and total flavonoid content. Additionally, transcriptome-based prediction identified AmbHLH30 as a potential transcription factor (TF) regulating calycosin biosynthesis, showing the highest expression under red light conditions. These results suggest that red light not only promotes calycosin accumulation but also modulates the expression of biosynthetic genes and transcriptional regulators, providing valuable insights for the metabolic engineering of isoflavonoid biosynthesis in A. membranaceus.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-33501-w.

Keywords: Astragalus membranaceus, Calycosin biosynthesis, Led light treatment, AmI3’H, Transient overexpression, Antioxidant activity

Subject terms: Biochemistry, Biotechnology, Computational biology and bioinformatics, Molecular biology, Plant sciences

Introduction

Astragalus membranaceus (Fabaceae) has a long history of use as a traditional medicine in China, Japan, Korea, and other regions in Asia. The dried root of A. membranaceus, known as “Huangqi”, is one of the most important medicinal herbs in traditional Chinese medicine1. A. membranaceus is known for its liver-protective and diuretic effects2, as well as its anti-hyperglycemic3, anti-inflammatory4, immune-modulating5,6, antioxidant7,8, and antiviral activities9. To date, over 100 compounds, including isoflavonoids, saponins, polysaccharides, and amino acids, have been identified in A. membranaceus, and various biological activities of these compounds have been reported10,11.

Isoflavonoids are one of the main active components of A. membranaceus12. Among these isoflavonoids, calycosin exhibits a wide range of pharmacological activities, including anticancer13, anti-inflammatory14, anti-osteoporosis15, neuroprotective16, and hepatoprotective effects17. Previous studies have shown that calycosin and its related isoflavonoids accumulate predominantly in the roots, which constitute the medicinal part of A. membranaceus, whereas their levels in aerial tissues are relatively low18. The biosynthetic pathway of calycosin in A. membranaceus was proposed based on results found in other leguminous plants. This pathway begins with phenylalanine ammonia-lyase (PAL)19. PAL converts L-phenylalanine into trans-cinnamic acid, which is then converted into p-coumarate CoA ester by cinnamate-4-hydroxylase (C4H) and 4-coumarate: coenzyme A-lyase (4CL), respectively20. Next, p-coumaroyl-CoA condenses with three malonyl-CoA molecules and is catalyzed by chalcone synthase (CHS) to synthesize chalcone. Through an isomerization reaction catalyzed by chalcone isomerase (CHI), naringenin (5,7,4′-trihydroxyflavanone) is produced. In species that synthesize isoflavones, isoflavone synthase (IFS) converts naringenin to genistein and catalyzes daidzein21. Daidzein is methylated by isoflavone O-methyltransferase (IOMT) to form formononetin22, and, lastly, isoflavone 3′-hydroxylase (I3’H) hydroxylates daidzein to form calycosin23. Many transcription factors (TFs), including MYB, basic helix-loop-helix (bHLH), and WD40 proteins, regulate secondary metabolites biosynthesis24.

The quality (wavelength), quantity (intensity), circadian rhythm (duration), and direction of light are each important components of light conditions25,26. Light-emitting diodes (LEDs) are artificial light sources that emit visible and ultraviolet light in a narrow spectrum, enabling the use of specific wavelength (color) light27. Compared to conventional lighting, LEDs are a promising technology in plant photobiology research28,29. Recent studies have shown that specific LED lighting has a significant impact on plant morphogenesis and that LED lighting at certain wavelengths greatly affects the yield of primary metabolites and species-specific secondary metabolites in plants30. For example, UV-B was found to increase the flavonoid and anthocyanin content in Artemisia annua31 and Withania somnifera32,33. In Heteropogon contortus, it increased tannin and total phenolic content34, and in Perilla frutescens, rosmarinic acid content increased35. LEDs set to the blue spectrum of visible light increased the chlorogenic acid, plumbagin, and total phenolic content in Dracocephalum forrestii36, Drosera indica37, and Ocimum basilicum38. When treated with red light, the salidroside and cucurbitacin contents of Rhodiola imbricata39 and Aquilaria agallocha40 were increased, respectively. In the case of R. imbricata, when treated with blue instead of red light, the total phenolic and total flavonoid contents increased39.

The Agrobacterium-mediated transient gene expression system is an effective method for rapidly analyzing gene expression in plants and has the advantage of producing various heterologous proteins. Additionally, this method is useful because it can be applied to species that are difficult to generate transgenic plants41–43. There are two main methods for introducing Agrobacterium into plants: vacuum infiltration and syringe infiltration. With vacuum infiltration it is possible to treat entire leaves or entire plants simultaneously; whereas syringe infiltration is simple to perform and widely used for small-scale protein production experiments44. In particular, syringe transfection is useful in that genes are expressed individually or in combination from a single leaf45. Agrobacterium infiltration is often applied to Nicotiana benthamiana because it has a short growth cycle and relatively large, easily penetrable leaves, enabling high levels of recombinant protein production. N. benthamiana is a suitable model plant for heterologous protein expression studies given that most Agrobacterium strains do not induce necrosis44. Furthermore, agroinfiltration technology allows for the direct introduction of specific metabolic pathways into N. benthamiana leaves, similar to microbial-based systems, making it a powerful tool for metabolic engineering and protein production research46.

Recently, various studies of the legume A. membranaceus have been conducted; however, most of these studies have focused on the underground parts of the plant, and there have been no reports on the calycosin content and expression levels of biosynthetic-related genes in the entire plant, and little is known about the effect of light sources. In a previous study, we performed transcriptomic analysis on whole plants of A. membranaceus seedlings treated with artificial light (https://www.ncbi.nlm.nih.gov/sra?linkname=bioproject_sra_all%26from_uid=865476)47. We compared the calycosin content and the expression levels of calycosin biosynthesis-related genes obtained through transcriptomic analysis in A. membranaceus seedlings cultured under various light conditions. Building on these initial results, we carried out the present study to experimentally confirm that the AmI3’H gene is a key gene involved in calycosin biosynthesis by transforming the AmI3’H gene into Agrobacterium and analyzing the antioxidant activity, total flavonoid content, and HPLC analysis of N. benthamiana leaves transiently overexpressing the gene.

Results

Comparison of growth characteristics and calycosin content of in vitro-grown sprouts of A. membranaceus under different light sources

A. membranaceus seedlings grown in vitro were treated with three different light sources for six weeks. The length of the plants was highest under blue light at 21.83 ± 2.08 cm, and there was no significant difference under white light and red light conditions (Fig. 1c). However, fresh weight and dry weight were highest under red light treatment, at 0.22 ± 0.05 g and 0.05 ± 0.006 g, respectively (Fig. 1d and Fig. 1e). Additionally, HPLC analysis was conducted to compare the calycosin content in A. membranaceus seedlings grown under different light sources. The results showed that the calycosin content was highest in plants treated with red light at 44.07 µg/g dry weight, followed by white light and blue light at 34.13 µg/g dry weight and 32.33 µg/g dry weight, respectively (Fig. 2).

Fig. 1.

Fig. 1

Open in a new tab

Effects of different LED light sources on the growth of in vitro cultured A. membranaceus sprouts. (a) Photographic images of A. membranaceus sprouts cultured under various LED light treatments; Wavelengths of LED light sources. (b) Growth characteristics of in vitro cultured A. membranaceus. (c) Plant height, (d) Fresh weight, and (e) dry weight. Error bars represent standard deviations of biological replicates. Statistical differences were determined using Duncan’s Multiple Range Test (DMRT, p < 0.05). Different letters indicate statistically significant differences.

Fig. 2.

Fig. 2

Open in a new tab

Quantitative analysis of calycosin content in A. membranaceus using HPLC. HPLC chromatograms of extracts from A. membranaceus sprouts in vitro-cultured under different LED light sources, (a) white, (b) blue, and (c) red. (d) Calycosin content was expressed as a percentage of dry weight.

Identification of calycosin biosynthetic genes and expression level verification

Using transcriptomic data obtained from previous studies, we identified genes involved in calycosin biosynthesis (Fig. 3). These include 8 PAL genes, 1 C4H gene, 12 4CL genes, 10 CHS genes, 4 CHR genes, 1 CHI gene, 1 IFS gene, 2 IOMT genes, and 2 I3’H genes. To confirm gene expression levels at the mRNA level through real-time PCR analysis, the log2 (FPKM) values of the genes were compared, and the genes with the highest values for each gene were selected as candidate genes. Using the pI/Mw program to confirm the molecular weight and pI of the selected candidate genes, AmPAL was found to have the highest molecular weight (78.36 kDa) and a pI of 6.2. In contrast, the gene with the lowest molecular weight was AmCHR (14.27 kDa), with a pI of 7.85. Additionally, using the WoLF PSORT program to predict the intracellular localization of candidate genes, most genes were found to be present not only in chloroplasts and plasma membranes but also in various cellular organelles. However, AmIFS was confirmed to be present exclusively in chloroplasts (Table S2). Real-time PCR analysis of gene expression levels revealed that, interestingly, genes from AmPAL to AmCHR did not show a significant trend (Fig. 4a-d), but starting with the AmCHS gene, all genes including AmCHI, AmIFS, AmIOMT, and AmI3’H exhibited the highest expression levels in the red light-treated group (Fig. 4e-i). Notably, AmI3’H, the last gene expressed in the calycosin biosynthesis pathway, showed more than fivefold higher expression levels compared to the blue light-treated control group, and the p-value was also confirmed to be statistically significant at 0.001 or below.

Fig. 3.

Fig. 3

Open in a new tab

Expression profile of calycosin biosynthetic genes based on RNA-Seq and related biosynthetic pathway. The accompanying heatmap represents the FPKM values of genes encoding pathway enzymes across different light treatments. Color gradients indicate relative expression levels from low (blue) to high (red).

Fig. 4.

Fig. 4

Open in a new tab

Validation of calycosin biosynthetic gene expression in in vitro cultured A. membranaceus sprouts under different LED light sources. Relative gene expression levels were normalized to an housekeeping gene. The analyzed genes include (a) AmPAL (Gene_135530T), (b) AmC4H (Gene_159510T), (c) Am4CL (Gene_118780T), (d) AmCHR (Gene_050470T), (e) AmCHS (Gene_430150T), (f) AmCHI (Gene_339900T), (g) AmIFS (Gene_071270T), (h) AmIOMT (Gene_244070T), and (i) AmI3’H (Gene_046760T). Error bars represent standard deviations of biological replicates. Statistical significance was indicated by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001).

Analysis of AmI3’H gene structure and identification of formononetin binding sites through molecular docking

Based on the real-time PCR analysis results, the AmI3’H gene, which is expected to have the greatest influence on calycosin biosynthesis, was selected as a candidate gene. The gene structure and conserved motifs were confirmed using the MEME program and InterPro online tool through comparative analysis of the amino acid sequences of the selected gene with those of other plants. Because the amino acid sequence of AmI3’H obtained through RNA-Seq analysis was only confirmed as a partial CDS, motif analysis was conducted using the complete CDS of AmI3’H registered in NCBI (Gene Bank ID: AFA89978.1). As a result, motifs 1, 3, 4, 5, and 7 were found to share common functional roles (oxidoreductase activity, acting on paired donors, with incorporation or reduction of molecular oxygen (GO:0,016,705), monooxygenase activity (GO:0,004,497), iron ion binding (GO:0,005,506), heme binding (GO:0,020,037)), while the remaining motifs were expected to have low functional relevance. Analysis of the protein family of these motifs revealed that they belong to the cytochrome P450 family, specifically the CYP81 clan (Fig. 5a, Table S3). Additionally, sequence alignment was performed to confirm the sequence similarity with I3’H genes from other plants. The results showed the highest sequence similarity with TpI3’H (Fig. 5a), and the amino acid sequence of the conserved motif domain performing functional roles was identified (Fig. 5b). To determine where formononetin, the precursor of calycosin, binds to AmI3’H, molecular docking analysis was performed. Using the analysis program, protein–ligand docking was conducted, and a total of nine potential binding sites were identified in order of affinity. Among these, models with an RMSD l.b value below 2.0 were considered reliable. Based on these criteria, model 1 was excluded as it did not dock, and analysis was ultimately conducted using model 2 (Table S4). The results confirmed that the H atom in the A-ring of formononetin binds to the leucine residue at position 308 of AmI3’H (Fig. 6).

Fig. 5.

Fig. 5

Open in a new tab

Conserved motif analysis and sequence alignment of AmI3’H protein. (a) Conserved motifs within AmI3’H and homologous proteins were identified using MEME Suite and visualized by motif distribution across amino acid sequences. (b) Multiple sequence alignment of AmI3’H with related cytochrome P450 family members highlights conserved domains and functionally important residues. Functionally conserved domains were highlighted in orange boxes.

Fig. 6.

Fig. 6

Open in a new tab

Molecular docking of formononetin with AmI3’H protein. Structural model showing docking interactions between formononetin and AmI3’H. The ligand is bound near the predicted active site with key interacting amino acid residues labeled. Both A-ring and B-ring orientations of formononetin are shown, with hydrogen bonding and hydrophobic interactions indicated.

Confirmation of AmI3’H gene expression, physiological activity, and calycosin production in N. benthamiana leaves

We aimed to confirm the expression pattern of the AmI3’H gene in N. benthamiana leaves inoculated with Agrobacterium containing AmI3’H or formononetin, a precursor of calycosin, using real-time PCR. The treatment groups included leaves treated with the empty vector pMBP1, formononetin, and AmI3’H alone, as well as leaves harvested at different time points after co-infiltration with formononetin and AmI3’H (Fig. 7b–g). Real-time PCR was performed using leaves inoculated with pMBP1 as a control. The results showed that the expression level of the AmI3’H gene was lowest in leaves treated with formononetin alone and highest in leaves treated with AmI3’H alone. The expression pattern of the AmI3’H gene in leaves co-infiltrated with formononetin and AmI3’H was interesting, showing a decreasing trend in AmI3’H gene expression as the harvest time was delayed (from 24 to 72 h) (Fig. 7h). Additionally, to confirm whether formononetin co-infiltrated with AmI3’H affects not only AmI3’H but also the expression levels of genes present in N. benthamiana leaves, the expression levels of N. benthamiana leaf phenylalanine pathway genes NbPAL, NbC4H, Nb4CL, NbCHS, and NbCHI, which are part of the phenylalanine pathway in N. benthamiana leaves, were also examined using real-time PCR. The results showed that the expression patterns of these genes differed slightly in leaves treated with formononetin and AmI3’H alone, but in leaves co-infiltrated with both, the expression levels decreased as the harvest time was delayed, similar to AmI3’H (Fig. 8).

Fig. 7.

Fig. 7

Open in a new tab

Transient expression of AmI3’H in N. benthamiana. (a) Schematic map of the cloning vector used for AmI3’H expression. (b–g) Photographs of N. benthamiana leaves after Agrobacterium-mediated transient expression with (b) vector, (c) formononetin, (d) AmI3’H, and (e–g) AmI3’H + formononetin 24 h, 48 h, and 72 h. (h) Validation of AmI3’H gene expression in transient expressed N. benthamiana leaf tissues. Error bars represent standard deviations of biological replicates. Statistical significance was indicated by Student’s t-test (**p < 0.01; ***p < 0.001).

Fig. 8.

Fig. 8

Open in a new tab

Expression analysis of phenylpropanoid pathway genes in transiently expressed N. benthamiana. Relative gene expression levels were normalized to an housekeeping gene. The analyzed genes include (a) NbPAL, (b) NbC4H, (c) Nb4CL, (d) NbCHS, (e) NbCHI. Error bars represent standard deviations of biological replicates. Statistical significance was indicated by Student’s t-test (*p < 0.05; **p < 0.01).

To confirm the correlation between gene expression levels and physiological activity, antioxidant activity and total flavonoid content were examined. In the DPPH radical scavenging experiment, the antioxidant activity of tobacco leaves was highest at 52.90% when AmI3’H and formononetin were co-infiltrated and 24 h had elapsed (Fig. 9a). When total flavonoid content was measured using quercetin as an indicator compound, the same treatment group also showed the highest content at 10.39 mg·QE/g dry weight (Fig. 9b).

Fig. 9.

Fig. 9

Open in a new tab

Assessment of antioxidant activity and total flavonoid content in transient expressed N. benthamiana. (a) DPPH radical scavenging activity and (b) total flavonoid content were measured in leaf extracts after transient expression of AmI3’H and/or formononetin treatment. Error bars represent standard deviations of biological replicates. Statistical differences were determined using Duncan’s Multiple Range Test (DMRT, p < 0.05). Different letters indicate statistically significant differences.

In addition, HPLC analysis was performed to confirm whether calycosin was synthesized when the AmI3’H gene was infiltrated together with the calycosin precursor formononetin. The results showed that calycosin was not produced in N. benthamiana leaves treated with the vector, formononetin, or with AmI3’H alone (Fig. 10a–e). Interestingly, among the N. benthamiana leaves co-infiltrated with AmI3’H and formononetin, calycosin was not synthesized in leaves harvested at 48 and 72 h, but only in leaves harvested at 24 h, resulting in 2.2 µg/mL dry weight of calycosin. Additionally, the formononetin content was most preserved in the 24-h treatment group (Fig. 10f–h). This confirmed that formononetin, the precursor, is necessary for calycosin synthesis and that calycosin can be synthesized when AmI3’H is expressed at a certain level or higher.

Fig. 10.

Fig. 10

Open in a new tab

HPLC analysis of calycosin production in N. benthamiana following transient expression. Chromatographic profiles of (a) calycosin standard, (b) formononetin standard, and leaf extracts from various treatment groups: (c) vector, (d) formononetin, (e) AmI3’H, and (f–h) AmI3’H + formononetin at 24 h, 48 h, and 72 h. Calycosin was detected only in the AmI3’H + formononetin co-treatment group at 24 h.

Prediction of TFs related to calycosin biosynthesis and verification of expression levels by light source

Using transcriptomic analysis data obtained from previous studies, we predicted the interactions between TFs and calycosin biosynthesis genes in in vitro-grown A. membranaceus seedlings treated with different light sources using the string program and compared the expression levels of selected TFs by light source using real-time PCR. The results indicated that the AmbHLH30 gene was predicted to be associated with three calycosin biosynthesis genes (AmPAL, AmC4H, AmIFS) and three TFs (AmMYB184, AmMYB16, AmMYB41) (Fig. 11a). When the degree of association was quantified, the AmbHLH30 gene was confirmed to play a central role in the response (Table S5). The log2(FPKM) value also showed that the AmbHLH30 gene had a higher value than other TFs, with the maximum value observed in the red light treatment group (Fig. 11b). Additionally, when comparing the expression levels of the gene by light source using real-time PCR, the highest expression level was observed in the red light treatment group (Fig. 11c).

Fig. 11.

Fig. 11

Open in a new tab

Gene regulatory network and expression validation of TFs related to calycosin biosynthesis. (a) Protein–protein interaction network among calycosin biosynthesis-related genes and TFs constructed using STRING database. (b) FPKM-based expression comparison of key TFs under different light conditions. (c) Validation of selected AmbHLH30 (Gene_242840T) TF genes. Error bars represent standard deviations of biological replicates. Statistical significance was indicated by Student’s t-test (**p < 0.01).

Discussion

Effects of LED light on growth and calycosin accumulation

LED light treatment is widely used in controlled cultivation systems to improve crop yield, nutrition, and phytochemical levels due to the fact that this technique is considered environmentally friendly and economically advantageous48. In particular, light sources are important abiotic factors that directly or indirectly influence secondary metabolites biosynthesis in plants and are applied to various plant species. Among these, red and blue light have been reported to efficiently absorb light wavelengths, promoting the accumulation of major secondary metabolites and phytochemical production, and are therefore primarily used in smart farming facilities49,50. In this study, we aimed to compare the growth and calycosin content, one of the major indicator compounds, by treating in vitro-cultured sprouts of A. membranaceus with artificial light sources of different wavelengths. Maximum values of both dry weight and fresh weight were obtained from plants treated with red light, with growth in the underground parts being more pronounced than in the aboveground parts. Conversely, blue light, which resulted in the longest shoot length, promoted active growth of the aboveground part. When treated with white light and red light, multiple stems developed from a single hypocotyl, whereas under blue light, a single elongated stem was produced (Fig. 1a). Blue light not only promotes stem growth when applied alone51 but also performs various important photomorphogenic roles in plants28, including stomatal control52, which is known to influence water relations, CO₂ exchange53, and phototropism54. Red light strongly affects phytochrome and inhibits plant growth. When blue light and red light are applied simultaneously, plant growth is inhibited more than when each light source is applied separately51,55. However, these light source characteristics may vary depending on the plant species. For example, when blue light was applied to Chrysanthemum grown in vitro, plant growth was inhibited, but dry weight and photosynthetic pigment content increased56. Similarly, when blue light was applied to wheat, plant growth was most inhibited, and both tiller number and yield were the lowest57. The content of isoflavones also varies among plant species depending on the light wavelength. In a study comparing isoflavone content in soybean seedlings treated with red and blue light, changes in content were observed over time. Malonyl daidzin and malonyl genistin increased in content until 36 h of red light exposure but then decreased until 120 h, whereas blue light exposure resulted in time-dependent increases, reaching maximum content at 120 h. Genistein content increased when treated with red light for a short period, while blue light increased it when treated for a long period58. In a study comparing the content of isoflavonoid and triterpenoid compounds in A. membranaceus hairy root cultures (AMHRCs) treated with LED light sources, the content of calycosin, formononetin, and astragaloside increased the most when treated with blue light48. In the present study, calycosin content was maximized when red light was applied to in vitro-cultured A. membranaceus (Fig. 2), indicating that even within the same plant species, changes in useful substance content may vary depending on the plant organ exposed to the light source. Recent studies have consistently reported that growth and yield are enhanced when plants are treated with mixed light forms or UV-B by adjusting the ratio of red and blue light59–61. The results we report here also suggest that further experiments using various combinations of light wavelengths are necessary to establish optimal growth conditions.

Red light-responsive expression and structural features of AmI3’H

We examined the expression levels of calycosin biosynthetic genes in in vitro-cultured A. membranaceus sprouts and found that the expression level of the AmI3’H gene was highest when treated with red light (Fig. 4i). We selected this gene as a key gene, as we believed it to be associated with calycosin content. The I3’H gene belongs to the cytochrome P450 81E subfamily, and little is known about it to date. The post-isoflavone stage of the isoflavonoid pathway in leguminous plants begins with C-2’ or C-3’ hydroxylation of the B-ring62. This reaction was confirmed to be catalyzed by P450 enzymes in experiments using chickpea extracts63,64, and subsequently, the I2’H gene discovered in licorice was classified into the CYP81E subfamily65. I3’H was discovered in Medicago truncatula, and it was revealed that I2’H is encoded by CYP81E7 and I3’H by CYP81E966. To analyze the structural characteristics of the AmI3’H gene selected in this study, conserved motif analysis and phylogenetic tree analysis were performed. As a result, AmI3’H was found to belong to the CYP81 clan, and motifs 1, 3, 4, 5, and 7 were identified as common conserved motifs (Fig. 5 and Table S3). All five conserved motifs were found to perform the same function, and characteristics such as monooxygenase activity (GO:0,004,497) and heme binding (GO:0,020,037) are known to be typical features of cytochrome P45067. Molecular docking analysis was conducted to identify the binding site between formononetin, the precursor of calycosin, and the AmI3’H gene. As a result, leucine 309 in motif 3 was identified as the region predicted to directly bind formononetin (Fig. 5b and 6). The results of this predictive model suggest that AmI3’H may be involved in the calycosin biosynthetic pathway by hydroxylating formononetin to produce calycosin. The structural characteristics of the AmI3’H gene provide clues to the calycosin biosynthesis mechanism in A. membranaceus seedlings cultured in vitro, as well as suggesting an association between light source-dependent expression levels and increased calycosin accumulation under red light treatment. However, since this study was based solely on in silico analysis, further functional validation of the enzymatic activity of the AmI3’H gene is required in the future.

Functional validation of AmI3’H in a heterologous system and metabolic implications

The transient expression system using Agrobacterium is a useful tool for rapidly evaluating gene expression in plants and is efficient for producing high-value compounds, with the potential for industrial-scale applications42,46. Additionally, transient expression offers the advantage of achieving high yields per unit of biomass through a rapid process while minimizing biological safety and regulatory issues associated with field cultivation68,69. In this study, we aimed to determine whether the expression level of AmI3’H influences calycosin biosynthesis using the transient expression system. After co-infiltrating N. benthamiana leaves with Agrobacterium containing the calycosin precursor formononetin and the AmI3’H gene, we harvested the leaves at different time points and compared the expression levels of the AmI3’H gene. The results showed that the highest expression level was observed 24 h after harvest, and the expression level decreased over time (Fig. 7h). Additionally, leaves infiltrated with empty vector, formononetin, or AmI3’H gene-containing Agrobacterium alone showed no or mild chlorosis, but when formononetin and the gene were co-infiltrated, chlorosis worsened over time (Fig. 7b–g). These results suggest that the metabolic pathway is activated when AmI3’H and its substrate formononetin are present simultaneously and that formononetin is converted into calycosin, a downstream metabolite.

Regulatory insights and future applications in metabolic engineering

AmI3’H—an enzyme belonging to the cytochrome P450 monooxygenase family (Fig. 5 and Table S3)—consumes NADPH and performs reactions through electron transfer processes, during which reactive oxygen species (ROS) may be generated as a byproduct70. Excessive ROS production can cause various problems such as lipid peroxidation, protein modification, chlorophyll decay, and cell death71–73, which may be associated with chlorosis in tobacco leaves. These results suggest that AmI3’H gene expression and substrate supply induce the activation of the target metabolic pathway in N. benthamiana while potentially accompanying physiological changes due to metabolic imbalance in the heterologous plant system. The expression levels of the phenylalanine pathway genes NbPAL, NbC4H, Nb4CL, NbCHS, and NbCHI in N. benthamiana also decreased as the infiltration time progressed (Fig. 8), which may be related to metabolic regulation mechanisms associated with metabolic channeling. Metabolic channeling plays a role in maintaining efficient metabolic flow by minimizing the diffusion of intermediates and protecting unstable metabolites through enzyme complexes74. However, when intermediate products such as formononetin are introduced from the outside, their conversion efficiency may be limited because they cannot effectively access the intrinsic enzyme complexes in N. benthamiana, which lacks the basic isoflavonoid biosynthetic pathway. In the present study, calycosin was synthesized in the leaves only after co-infiltration of formononetin and AmI3’H for 24 h (Fig. 10). This result reflects the characteristics of the metabolic channeling system. Furthermore, the accumulation of formononetin in N. benthamiana may have transmitted feedback inhibitory signals not only to the AmI3’H gene but also to upstream pathway genes, potentially leading to a general reduction in metabolic flux and inducing cellular stress responses such as ROS production, which may have contributed to chlorosis. This suggests a close interrelation between the phenylpropanoid and isoflavonoid metabolic pathways, demonstrating that the accumulation of intermediates in one pathway can affect the upstream pathway. Although the co-expression of AmI3’H with precursor feeding and the subsequent detection of calycosin provide highly compelling support for its catalytic function, the absence of direct in vitro enzyme activity assays remains a limitation. Recently, there have been a number of studies on enhancing useful substances in tobacco using multigene vectors46,75,76. Similarly, to address the limitations of the present study, further research is needed to develop a multiple expression cassette including the AmIFS and AmIOMT genes using a multigene vector to enhance the efficiency of metabolic channeling in N. benthamiana.

TFs are involved in the synthesis of secondary metabolites by various biotic and abiotic factors. For example, isoflavonoid biosynthesis is known to be regulated by three TF complexes: R2R3-MYB, bHLH, and WD-repeat77,78. TFs related to isoflavonoid biosynthesis have been investigated in soyabean (Glycine max) and in legumes, such as as A. membranaceus; as a result, the genes GmMYB176, GmMYBJ3, and GmMYB29 have been found to be involved in isoflavonoid biosynthesis79–81. In this study, we predicted the possibility that R2R3-MYB and bHLH family TFs derived from A. membranaceus, which were identified through transcriptomic analysis in previous studies, may be involved in interactions with genes related to calycosin biosynthesis47. We propose, based on the findings reported here, that the AmbHLH30 gene may act as a key regulator of this pathway, and in light-condition-specific in vitro culture experiments, the expression level of AmbHLH30 was highest in the red light treatment group (Fig. 11 and Table S5). This trend aligns with the finding that calycosin content was highest under red light conditions, suggesting that the AmbHLH30 gene may be functionally associated with calycosin biosynthesis in A. membranaceus. However, no studies have yet been conducted on the role and mechanism of the AmbHLH30 gene. The focus of our study was the correlation between gene expression levels and metabolite accumulation; thus, further functional analyses such as dual-luciferase reporter assays, ChIP-qPCR, and knockout/overexpression in planta are required to elucidate the target genes and transcriptional regulatory mechanisms of the AmbHLH30 gene.

In this study, we investigated the effects of artificial light sources on calycosin biosynthesis and related gene expression in in vitro-grown A. membranaceus sprouts and identified key genes in the calycosin biosynthesis pathway. Under red light treatment conditions, plant growth was enhanced, and the highest accumulation of calycosin was observed. Through transcriptomic analysis and real-time PCR, it was confirmed that the AmI3’H gene in the calycosin biosynthesis pathway exhibited a significant increase in expression under red light conditions. Structural analysis of the gene and molecular docking analysis were conducted to predict its binding characteristics with formononetin. Furthermore, using a transient expression system with N. benthamiana, it was experimentally verified that the expression of the AmI3’H gene directly contributes to calycosin synthesis. In particular, when formononetin was provided as a precursor and AmI3’H was overexpressed, not only was the expression level of the gene highest after 24 h but calycosin was also synthesized, confirming that the expression level of the AmI3’H gene affects calycosin biosynthesis. Additionally, interactions between the AmbHLH30 TF and calycosin biosynthesis genes were predicted. The results of this study serve as foundational data for studies on transcriptional regulation mechanisms. and highlight the possibility of a metabolic engineering approach based on light source control and genetic manipulation to enhance the production of useful substances in in vitro cultured A. membranaceus sprouts. Further investigation of multigene vector systems is warranted to optimize metabolic flux and verify TF function.

Materials and methods

A. membranaceus sample preparation and LED treatment

The A. membranaceus seeds used in this study were purchased from KS Seed Co., Ltd. (Incheon, Republic of Korea). To obtain sterile plants, A. membranaceus seeds were shaken in 70% EtOH for 1 min and in 3% NaClO with Tween-20 added for 5 min. After washing three times with sterile distilled water, the plants were cultured on MS medium (MS 4.4 g/L + sucrose 30 g/L + plant agar 8 g/L) for 6 weeks (Fig. 1a). The LEDs used were white light (continuous spectrum), red light (550 ~ 650 nm), and blue light (450 ~ 550 nm). The photoperiod was set to 16 h of light and 8 h of darkness, and the wavelength of each light source was measured using a PG200N lux meter (United Power Research Technology Co., Zhunan Township, Taiwan) (Fig. 1b). The cultivated plants were carefully removed from the medium without damaging the roots. Samples for molecular experiments were immediately placed in liquid nitrogen and stored at −80 °C, and samples for HPLC analysis and physiological activity analysis were freeze-dried and used in the experiments.

HPLC analysis of calycosin and formononetin contents in in vitro-grown sprouts of A. membranaceus

To analyze the calycosin content according to the light source, freeze-dried A. membranaceus samples were crushed, 100% MeOH was added, and extraction was performed using a shaker for 24 h. The extract was then filtered using 250 mm filter paper (Huyndai Micro Co., Ltd., Seoul, Republic of Korea) and concentrated using a rotary vacuum concentrator. The vacuum-concentrated sample was dissolved by adding 100% MeOH and filtered using a 0.45-µm filter (Hyundai Micro Co. Ltd., Seoul, Republic of Korea). Calycosin content analysis was performed using a high-performance liquid chromatography (HPLC) Agilent 1260 series instrument (Agilent Technologies Inc., Santa Clara, California, USA) and an HC-C18 column (Agilent Technologies Inc., USA), using water-containing 0.1% acetic acid (solvent A) and acetonitrile (solvent B) as the mobile phase. The mobile phase gradient conditions were as follows: 90% A: 10% B (0 ~ 4 min), 70% A: 25% B (5 ~ 19 min), 60% A: 40% B (20 ~ 29 min), 10% A: 90% B (30 ~ 35 min), 90% A: 10% B (36 ~ 45 min). HPLC was performed with a sample injection volume of 10 µL, a flow rate of 0.8 mL/min, a temperature of 30 °C, and a wavelength of 260 nm. The calibration curve was prepared using calycosin indicator material diluted with 100% MeOH, and the calycosin content in the sample was converted to the content relative to the dry weight.

Quantitative real-time PCR analysis

Total RNA was extracted from in vitro-cultured sprouts of A. membranaceus and transiently overexpressed in N. benthamiana using Trizol reagent (Invitrogen Scientific, Inc., USA). The purity of the extracted total RNA was confirmed using a Microvolume Spectrophotometer (Keen Innovative Solutions, Daejeon, Korea). cDNA was synthesized using PrimeScript™ RT MasterRMix (Perfect Real Time) (Takara Korea Biomedical Inc., Seoul, Republic of Korea). real-time PCR was carried out with a 25-µL mixture prepared using TB Green® Premix Ex Taq™ II (Tli RNaseH Plus) (Takara Korea Biomedical Inc., Seoul, Republic of Korea), and the analysis was performed using the CronoSTAR™ 96 Real-Time PCR System (Takara Korea Biomedical Inc., Seoul, Republic of Korea). The analysis conditions were set as follows: initial denaturation at 95 °C for 30 s, followed by two-step amplification (denaturation at 95 °C for 5 s, annealing at 60 °C for 30 s) for 40 cycles. The melting curve was confirmed using a melting step at 95 °C for 1 min, 60 °C for 15 s, and 98 °C for 5 s. The calycosin biosynthesis genes and TF used in the analysis were selected based on the FPKM values from transcriptomic data registered in NCBI in previous studies, and primers were designed using the Primer3Plus program (https://www.primer3plus.com/)82. The expression levels of each reference gene were determined using Ct values and calculated using the 2-△△ct method. 18S and NbActin genes were used as housekeeping genes, respectively. The primer sequences used are shown in Table S1.

Bioinformatics analysis of key genes involved in calycosin biosynthesis

Multiple sequence alignment was performed using the Genedoc 2.7 program83, and the phylogenetic tree was constructed using the neighbor-joining algorithm in MEGA 11 software84. The conserved motifs of AmI3’H proteins were analyzed using the MEME program (https://meme-suite.org/meme/tools/meme) with the following parameter conditions: minimum motif width of 6, maximum motif width of 50, and maximum motif number of 10. Conserved domain analysis was performed using InterPro online tools (https://www.ebi.ac.uk/interpro/). The molecular weight (MW) and isoelectric point (pI) of the calycosin biosynthesis genes were analyzed using compute pI/Mw (https://web.expasy.org/compute_pi/), and the subcellular localization prediction for each gene was performed using WoLF PSORT (https://www.genscript.com/wolf-psort.html). Lastly, the generated phylogenetic tree, motif patterns, and domain distributions were visualized using TBtools85.

Protein–protein interaction and molecular docking analysis

Protein–protein interaction networks were created using the Search Tool for the Retrieval of Interaction Genes/Proteins (STRING) online tool (https://string-db.org/cgi/input?sessionId=bJBNP8WcwEJg%26input_page_show_search=on), selecting Glycine max as the organism and a confidence score of 0.400, and visualized using Cytoscape software v3.10.386. The three-dimensional structure of formononetin was downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/), and the protein structure of AmI3’H was obtained by using the alphafold structure of AmI3’H registered in NCBI, which has the same sequence as the protein sequence obtained from the transcriptome analysis (Gene bank ID: AFA89978.1). Every structure was visualized using Pymol software87, and protein–ligand docking models were created using Autodocktools-1.5.7.

Cloning and Agrobacterium transformation of reference genes

The full-length AmI3’H gene was amplified by PCR using gene-specific primers containing restriction enzyme sites at the 5’ end of the coding region. PCR products were extracted using a PCR cleanup kit (INFUSION TECH, Anyang, Republic of Korea) and treated with restriction enzymes such as XbaI and SacI. The cut fragments were recombined into the restriction enzyme-treated pMBP1 vector and transformed into Escherichia coli DH5α cells. The transformed colonies were confirmed by PCR, and the plasmid was extracted using the PURE™ Plasmid Miniprep Kit (INFUSION TECH, Anyang, Republic of Korea). The recombinant plasmid DNA was transformed into Agrobacterium tumefaciens LBA4404 cells to generate 35S-AmI3’H (Fig. 7a). The transformed Agrobacterium was cultured in YEP medium with antibiotics, and the plasmid was extracted. The extracted plasmid was confirmed by PCR, purified, and finally selected by sequencing to identify the transformed colony.

Agrobacterium-mediated transient overexpression in N. benthamiana

AmI3’H genes were transformed into A. tumefaciens strain LBA4404 using the freeze–thaw method88. Transformed Agrobacterium cells were inoculated into liquid YEP medium containing antibiotics (25 µg/mL kanamycin and rifampicin) and grown for 24 h at 28 °C to a cell concentration of OD600 1.0. Cultured cells were harvested and resuspended in infiltration medium (10 mM MES, 10 mM MgCl2, and 200 mM acetosyringone) to OD600 of 0.8. After 2 h of incubation in the dark, the Agrobacterium suspension was infiltrated into the abaxial surfaces of N. benthamiana leaves using a needleless 1-mL syringe. Leaves infiltrated with pMBP1 and formononetin, the precursor of calycosin, were harvested after 4 days. For leaves infiltrated with both formononetin and AmI3’H, AmI3’H was infiltrated first, followed by a 24 h recovery period, after which formononetin was infiltrated, and the leaves were harvested at 24, 48, and 72 h. The harvested leaves were stored in liquid nitrogen at −80 °C for RNA extraction or freeze-dried for bioactivity analysis. As a negative control, a medium containing only the empty vector (pMBP1) was infiltrated into N. benthamiana leaves.

Extraction of transient expressed N. benthamiana leaves

N. benthamiana leaf samples subjected to transient expression were freeze-dried, ground, and extracted by adding 100% MeOH and sonicating twice at 30 °C for 30 min. During the filtration process, the supernatant was separated by centrifugation at 13,000 rpm for 10 min. The filtrate was then concentrated under reduced pressure using a rotary vacuum concentrator (EYELA N-1000, Tokyo Rikakikai Co., Ltd., Tokyo, Japan), and the concentrated sample was further freeze-dried to produce a powder.

DPPH radical scavenging activity analysis

The DPPH radical scavenging activity was evaluated using a modified method described by Sharma and Bhat (2009)89. First, 100 µL of 0.2 mM DPPH (2,2-diphenyl-1-picrylhydrazyl) (Biozoa Co., Ltd., Seoul, Republic of Korea) and 100 µL of N. benthamiana leaf extract diluted to an appropriate concentration were added to a 96-well plate and reacted under dark conditions for 30 min. Subsequently, the absorbance was measured at 519 nm using a UV/Vis spectrophotometer (Multiskan FC Microplate Photometer, Thermo Fisher Scientific Inc., Waltham, MA, USA). The radical scavenging ability of each treatment group was expressed as the relative absorbance of the sample treatment group compared to the untreated control group, expressed as a percentage.

Total flavonoid content analysis

The total flavonoid content was measured using the method described by Moreno et al. (2000)90. A 100-µL volume of 1 M potassium acetate (Mallinckrodt Co., Ltd., Tokyo, Japan), 10% aluminum nitrate (Yakuri Co., Ltd., Shizuoka, Japan), and 500 µL of the sample diluted to a concentration of 1,000 µg/mL were added to a 96-well plate. The mixture was reacted at room temperature for 40 min, and the absorbance was measured at 414 nm using a UV/Vis spectrophotometer (Multiskan FC Microplate Photometer, Thermo Fisher Scientific Inc., Waltham, MA, USA). The total flavonoid content was determined using quercetin as the standard to create a standard curve, and the values were expressed as quercetin equivalents (mg·QE/g).

Statistical analysis

All experiments were conducted in triplicate, and IBM SPSS Statistics v29 software (SPSS, International Business Machines Co. Ltd., USA) was used for significance testing. Student’s t-test was applied when comparing two groups, whereas one-way ANOVA followed by Duncan’s multiple range test was used for comparisons among more than three treatments. The choice of statistical test was based on the number of groups and the purpose of each analysis. A p-value of 0.05 or less was considered statistically significant.

Core

This study demonstrates that red LED light significantly enhances calycosin biosynthesis in A. membranaceus by upregulating the AmI3’H gene, a cytochrome P450 enzyme. Functional validation using N. benthamiana confirmed that AmI3’H catalyzes the conversion of formononetin to calycosin. In summary, this study experimentally verifies the key regulatory factor of calycosin biosynthesis induced by red light. It also provides important scientific evidence that can be utilized for future optimization of commercial cultivation of A. membranaceus, development of functional plant materials, and establishment of a mass production platform. Future efforts, including multigene pathway recombination, functional characterization of AmbHLH30, and development of smart cultivation technologies combining diverse light environments, are expected to contribute to maximizing calycosin production industrially.

Genes and accession numbers

The genetic information used in this study can be accessed in the Genebank data library with the following accession numbers: AmPAL (Gene_135530T), AmC4H (Gene_159510T), Am4CL (Gene_118780T), AmCHR (Gene_050470T), AmCHS (Gene_430150T), AmCHI (Gene_339900T), AmIFS (Gene_071270T), AmIOMT (Gene_244070T), AmMYB16 (Gene_250860), AmMYB184 (Gene_273460T), AmMYB41 (Gene_360960T), AmbHLH130 (Gene_177260T), and AmI3’H (Gene_046760T), and AmbHLH30 (Gene_242840T). The transcriptome data are available in the NCBI Sequence Read Archive (SRA) database with the accession number PRJNA865476.

Supplementary Information

Supplementary Information. (20.2KB, xlsx)

Abbreviations

A. membranaceus

Astragalus membranaceus

AmI3’H

Astragalus membranaceus isoflavone 3′-hydroxylase

CHS

Chalcone synthase

CHI

Chalcone isomerase

C4H

Cinnamate-4-hydroxylase

CL

Confidence level

DPPH

2,2-diphenyl-1-picrylhydrazyl

HPLC

High-performance liquid chromatography

IFS

Isoflavone synthase

IOMT

Isoflavone O-methyltransferase

LED

Light-emitting diode

MS

Murashige and Skoog medium

OD

Optical density

PAL

Phenylalanine ammonia-lyase

QE

Quercetin equivalent

RNA-Seq

RNA sequencing

ROS

Reactive oxygen species

TF

Transcription factor

UV

Ultraviolet

Author contributions

J.W.S: formal analysis and writing-original draft; W.H.C and S.L: formal analysis; E.S.S: investigation, supervision and writing-review and editing. All authors have read and agreed to the published version of the manuscript.

Data availability

All data generated or analyzed during are available in the present article.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Tuan, P. A. et al. Transcriptional profiling and molecular characterization of astragalosides, calycosin, and calycosin-7-O-β-D-glucoside biosynthesis in the hairy roots of Astragalus membranaceus in response to methyl jasmonate. J. Agric. Food Chem.63, 6231–6240 (2015). [DOI] [PubMed] [Google Scholar]
  • 2.Zhang, X. et al. Research progress on the treatment of related diseases with astragalus. Drug Des. Devel. Ther.19, 2845–2862 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Chan, J. Y. W., Lam, F. C., Leung, P. C., Che, C. T. & Fung, K. P. Antihyperglycemic and antioxidative effects of a herbal formulation of Radix astragali, Radix codonopsis and Cortex lycii in a mouse model of type 2 diabetes mellitus. Phytother. Res.23, 658–665 (2009). [DOI] [PubMed] [Google Scholar]
  • 4.Choi, H. S. et al. Quality characteristic of Hwangki (Astragalus membranaceus) Chungkukjang during fermentation. Food Sci. Preserv.14, 356–363 (2007). [Google Scholar]
  • 5.Bedir, E., Pugh, N., Calis, I., Pasco, D. S. & Khan, I. A. Immunostimulatory effects of cycloartane-type triterpene glycosides from Astragalus species. Biol. Pharm. Bull.23, 834–837 (2000). [DOI] [PubMed] [Google Scholar]
  • 6.Wang, T. et al. Enhancement of non-specific immune response in sea cucumber (Apostichopus japonicus) by Astragalus membranaceus and its polysaccharides. Fish Shellfish Immunol.27, 757–762 (2009). [DOI] [PubMed] [Google Scholar]
  • 7.Yu, D. H., Bao, Y. M., Wei, C. L. & An, L. J. Studies of chemical constituents and their antioxidant activities from Astragalus mongholicus Bunge. Biomed. Environ. Sci.18, 297 (2005). [PubMed] [Google Scholar]
  • 8.Li, R., Chen, W., Wang, W., Tian, W. & Zhang, X. Antioxidant activity of astragalus polysaccharides and antitumour activity of the polysaccharides and siRNA. Carbohydr. Polym.82, 240–244 (2010). [Google Scholar]
  • 9.Zhu, H. et al. In vivo and in vitro antiviral activities of calycosin-7-O-β-D-glucopyranoside against coxsackie virus B3. Biol. Pharm. Bull.32, 68–73 (2009). [DOI] [PubMed] [Google Scholar]
  • 10.Xu, F. et al. Absorption and metabolism of Astragali radix decoction: In silico, in vitro, and a case study in vivo. Drug Metab. Dispos.34, 913–924 (2006). [DOI] [PubMed] [Google Scholar]
  • 11.Chu, C. et al. Radix astragali (Astragalus): Latest advancements and trends in chemistry, analysis, pharmacology and pharmacokinetics. Curr. Org. Chem.14, 1792–1807 (2010). [Google Scholar]
  • 12.Auyeung, K. K., Han, Q. B. & Ko, J. K. Astragalus membranaceus: A review of its protection against inflammation and gastrointestinal cancers. Am. J. Chin. Med.44, 1–22 (2016). [DOI] [PubMed] [Google Scholar]
  • 13.Sun, H., Yin, M., Qian, W. & Yin, H. Calycosin, a phytoestrogen isoflavone, induces apoptosis of estrogen receptor-positive MG-63 osteosarcoma cells via the phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathway. Med. Sci. Monit.24, 6178 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Cui, L. K., Han, Y. M. & Ma, Y. K. Calycosin ameliorates inflammatory paw edema in mice via inhibiting NF-kappa B activation. Int. J. Pharmacol.15, 744–751 (2019). [Google Scholar]
  • 15.Fang, Y. et al. Calycosin stimulates the osteogenic differentiation of rat calvarial osteoblasts by activating the IGF1R/PI3K/Akt signaling pathway. Cell Biol. Int.43, 323–332 (2019). [DOI] [PubMed] [Google Scholar]
  • 16.Yang, J., Jia, M., Zhang, X. & Wang, P. Calycosin attenuates MPTP-induced Parkinson’s disease by suppressing the activation of TLR/NF-κB and MAPK pathways. Phytother. Res.33, 309–318 (2019). [DOI] [PubMed] [Google Scholar]
  • 17.Deng, T. et al. Inhibition effect of phytoestrogen calycosin on TGF-β1-induced hepatic stellate cell activation, proliferation, and migration via estrogen receptor β. Can. J. Physiol. Pharmacol.96, 1268–1275 (2018). [DOI] [PubMed] [Google Scholar]
  • 18.Wang, E. B. et al. Comparison of aerial parts of Astragalus membranaceus and Astragali Radix based on chemical constituents and pharmacological effects. Food Agric. Immunol.30, 1046–1066 (2019). [Google Scholar]
  • 19.Xu, R. Y. et al. Ultraviolet irradiation induces accumulation of isoflavonoids and transcription of genes of enzymes involved in the calycosin-7-O-β-d-glucoside pathway in Astragalus membranaceus Bge. var. mongholicus (Bge.) Hsiao. Physiol. Plant.142, 265–273 (2011). [DOI] [PubMed] [Google Scholar]
  • 20.Russell, D. W. The Metabolism of aromatic compounds in higher plants: X. properties of the cinnamic acid 4-hydroxylase of pea seedlings and some aspects of its metabolic and developmental control. J. Biol. Chem.246, 3870–3878 (1971). [PubMed] [Google Scholar]
  • 21.Yu, O. & McGonigle, B. Metabolic engineering of isoflavone biosynthesis. Adv. Agron.86, 147–190 (2005). [Google Scholar]
  • 22.Akashi, T. et al. cDNA cloning and biochemical characterization of S-adenosyl-L-methionine: 2, 7, 4′-trihydroxyisoflavanone 4′-O-methyltransferase, a critical enzyme of the legume isoflavonoid phytoalexin pathway. Plant Cell Physiol.44, 103–112 (2003). [DOI] [PubMed] [Google Scholar]
  • 23.Lee, J., Kim, Y. B., Uddin, M. R., Cho, J. W. & Park, S. U. Terpenoid and flavonoid biosynthesis in Astragalus membranaceus. Life Sci. J.10, 970–973 (2013). [Google Scholar]
  • 24.Hichri, I. et al. Recent advances in the transcriptional regulation of the flavonoid biosynthetic pathway. J. Exp. Bot.62, 2465–2483 (2011). [DOI] [PubMed] [Google Scholar]
  • 25.Ding, Z. et al. Light-mediated polarization of the PIN3 auxin transporter for the phototropic response in Arabidopsis. Nat. Cell Biol.13, 447–452 (2011). [DOI] [PubMed] [Google Scholar]
  • 26.Kami, C., Lorrain, S., Hornitschek, P. & Fankhauser, C. Light-regulated plant growth and development. Curr. Top. Dev. Biol.91, 29–66 (2010). [DOI] [PubMed] [Google Scholar]
  • 27.Darko, E., Heydarizadeh, P., Schoefs, B. & Sabzalian, M. R. Photosynthesis under artificial light: The shift in primary and secondary metabolism. Phil. Trans. R. Soc. B369, 20130243 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Massa, G. D., Kim, H. H., Wheeler, R. M. & Mitchell, C. A. Plant productivity in response to LED lighting. HortScience43, 1951–1956 (2008). [Google Scholar]
  • 29.Morrow, R. C. LED lighting in horticulture. HortScience43, 1947–1950 (2008). [Google Scholar]
  • 30.Xie, D. et al. Transcriptomic and metabolomic profiling reveals the effect of LED light quality on morphological traits, and phenylpropanoid-derived compounds accumulation in Sarcandra glabra seedlings. BMC Plant Biol.20, 1–18 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Pandey, N. & Pandey-Rai, S. Short term UV-B radiation-mediated transcriptional responses and altered secondary metabolism of in vitro propagated plantlets of Artemisia annua L. Plant Cell Tissue Organ Cult.116, 371–385 (2014). [Google Scholar]
  • 32.Takshak, S. & Agrawal, S. B. Secondary metabolites and phenylpropanoid pathway enzymes as influenced under supplemental ultraviolet-B radiation in Withania somnifera Dunal, an indigenous medicinal plant. J. Photochem. Photobiol. B140, 332–343 (2014). [DOI] [PubMed] [Google Scholar]
  • 33.Takshak, S. & Agrawal, S. B. Defence strategies adopted by the medicinal plant Coleus forskohlii against supplemental ultraviolet-B radiation: Augmentation of secondary metabolites and antioxidants. Plant Physiol. Biochem.97, 124–138 (2015). [DOI] [PubMed] [Google Scholar]
  • 34.Hamid, A., Singh, S., Agrawal, M. & Agrawal, S. B. Heteropogon contortus BL-1 (Pilli grass) and elevated UV-B radiation: The role of growth, physiological, and biochemical traits in determining forage productivity and quality. Photochem. Photobiol.95, 572–580 (2019). [DOI] [PubMed] [Google Scholar]
  • 35.Yoshida, H., Nishikawa, T., Hikosaka, S. & Goto, E. Effects of nocturnal UV-B irradiation on growth, flowering, and phytochemical concentration in leaves of greenhouse-grown red perilla. Plants10, 1252 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Weremczuk-Jeżyna, I., Hnatuszko-Konka, K., Lebelt, L. & Grzegorczyk-Karolak, I. The protective function and modification of secondary metabolite accumulation in response to light stress in Dracocephalum forrestii shoots. Int. J. Mol. Sci.22, 7965 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Boonsnongcheep, P. et al. Artificial color light sources and precursor feeding enhance plumbagin production of the carnivorous plants Drosera burmannii and Drosera indica. J. Photochem. Photobiol. B199, 111628 (2019). [DOI] [PubMed] [Google Scholar]
  • 38.Nazir, M. et al. Light-mediated biosynthesis of phenylpropanoid metabolites and antioxidant potential in callus cultures of purple basil (Ocimumbasilicum L. var purpurascens). Plant Cell Tissue Organ Cult.142, 107–120 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Kapoor, S. et al. Influence of light quality on growth, secondary metabolites production and antioxidant activity in callus culture of Rhodiola imbricata Edgew. J. Photochem. Photobiol. B183, 258–265 (2018). [DOI] [PubMed] [Google Scholar]
  • 40.Kuo, T. C. Y. et al. The effect of red light and far-red light conditions on secondary metabolism in Agarwood. BMC Plant Biol.15, 1–10 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Fischer, R., Stoger, E., Schillberg, S., Christou, P. & Twyman, R. M. Plant-based production of biopharmaceuticals. Curr. Opin. Plant Biol.7, 152–158 (2004). [DOI] [PubMed] [Google Scholar]
  • 42.Fischer, R. et al. Towards molecular farming in the future: Transient protein expression in plants. Biotechnol. Appl. Biochem.30, 113–116 (1999). [PubMed] [Google Scholar]
  • 43.Horn, M., Woodard, S. & Howard, J. Plant molecular farming: Systems and products. Plant Cell Rep.22, 711–720 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.D’Aoust, M. A. et al. Transient expression of antibodies in plants using syringe agroinfiltration. In Recombinant Proteins from Plants: Methods and Protocols (eds Faye, L. & Gomord, V.) 41–50 (Humana Press, 2009). [DOI] [PubMed] [Google Scholar]
  • 45.Ma, L., Lukasik, E., Gawehns, F. & Takken, F. L. W. The use of agroinfiltration for transient expression of plant resistance and fungal effector proteins in Nicotianabenthamiana leaves. In Plant Fungal Pathogens: Methods and Protocols (eds Bolton, M. D. & Thoma, B. P. H. J.) 61–74 (Humana Press, 2012). [DOI] [PubMed] [Google Scholar]
  • 46.Lee, S. B. et al. Engineering Nicotiana benthamiana for chrysoeriol production using synthetic biology approaches. Front. Plant Sci.15, 1458916 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Seo, J. W. et al. Cellular morphology and transcriptome comparative analysis of astragalus membranaceus bunge sprouts cultured in vitro under different LED light. Plants12, 1914 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Gai, Q. Y. et al. Blue LED light promoting the growth, accumulation of high-value isoflavonoids and astragalosides, antioxidant response, and biosynthesis gene expression in Astragalus membranaceus (Fisch.) Bunge hairy root cultures. Plant Cell Tissue Organ Cult.153, 511–523 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Thoma, F., Somborn-Schulz, A., Schlehuber, D., Keuter, V. & Deerberg, G. Effects of light on secondary metabolites in selected leafy greens: A review. Front. Plant Sci.11, 497 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Li, Y. et al. Effects of red and blue light on leaf anatomy, CO2 assimilation and the photosynthetic electron transport capacity of sweet pepper (Capsicumannuum L.) seedlings. BMC Plant Biol.20, 1–16 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Jishi, T. LED lighting technique to control plant growth and morphology. In Smart Plant Factory: The Next Generation Indoor Vertical Farms (ed. Kozai, T.) 211–222 (Springer, 2018). [Google Scholar]
  • 52.Schwartz, A. & Zeiger, E. Metabolic energy for stomatal opening. Roles of photophosphorylation and oxidative phosphorylation. Planta161, 129–136 (1984). [DOI] [PubMed] [Google Scholar]
  • 53.Cosgrove, D. J. Rapid suppression of growth by blue light: Occurrence, time course, and general characteristics. Plant Physiol.67, 584–590 (1981). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Blaauw, O. & Blaauw-Jansen, G. The phototropic responses of Avena coleoptiles. Acta Bot. Neerl.19, 755–763 (1970). [Google Scholar]
  • 55.Craver, J. K. & Lopez, R. G. Control of morphology by manipulating light quality and daily light integral using LEDs. In LED Lighting for Urban Agriculture (eds Kozai, T. et al.) 203–217 (Springer, 2016). [Google Scholar]
  • 56.Kurilčik, A. et al. In vitro culture of Chrysanthemum plantlets using light-emitting diodes. Cent. Eur. J. Biol.3, 161–167 (2008). [Google Scholar]
  • 57.Monostori, I. et al. LED lighting–modification of growth, metabolism, yield and flour composition in wheat by spectral quality and intensity. Front. Plant Sci.9, 605 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Lim, Y. J., Kwon, S. J. & Eom, S. H. Red and blue light-specific metabolic changes in soybean seedlings. Front. Plant Sci.14, 1128001 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Choi, J., Kim, J., Yoon, H. I. & Son, J. E. Effect of far-red and UV-B light on the growth and ginsenoside content of ginseng (Panax ginseng CA Meyer) sprouts aeroponically grown in plant factories. Hortic. Environ. Biotechnol.63, 77–87 (2022). [Google Scholar]
  • 60.Flores, M., Urrestarazu, M., Amorós, A. & Escalona, V. High intensity and red enriched LED lights increased the growth of lettuce and endive. Ital. J. Agron.17, 1915 (2022). [Google Scholar]
  • 61.Zhang, D., Liu, Y., Ni, C. & Chen, J. Effect of LED light on the growth and physiological indices of blueberry. Agron. J.114, 2105–2112 (2022). [Google Scholar]
  • 62.Ayabe, S. & Akashi, T. Cytochrome P450s in flavonoid metabolism. Phytochem. Rev.5, 271–282 (2006). [Google Scholar]
  • 63.Hinderer, W., Flentje, U. & Barz, W. Microsomal isoflavone 2′-and 3′-hydroxylases from chickpea (Cicerarietinum L.) cell suspensions induced for pterocarpan phytoalexin formation. FEBS Lett.214, 101–106 (1987). [Google Scholar]
  • 64.Gunia, W., Hinderer, W., Wittkampf, U. & Barz, W. Elicitor induction of cytochrome P-450 monooxygenases in cell suspension cultures of chickpea (Cicerarietinum L.) and their involvement in pterocarpan phytoalexin biosynthesis. Z. Naturforsch. C46, 58–66 (1991). [Google Scholar]
  • 65.Akashi, T., Aoki, T. & Ayabe, S. CYP81E1, a cytochrome P450 cDNA of licorice (Glycyrrhizaechinata L.), encodes isoflavone 2′-hydroxylase. Biochem. Biophys. Res. Commun.251, 67–70 (1998). [DOI] [PubMed] [Google Scholar]
  • 66.Liu, C. J., Huhman, D., Sumner, L. W. & Dixon, R. A. Regiospecific hydroxylation of isoflavones by cytochrome p450 81E enzymes from Medicago truncatula. Plant J.36, 471–484 (2003). [DOI] [PubMed] [Google Scholar]
  • 67.Munro, A. W., McLean, K. J., Grant, J. L. & Makris, T. M. Structure and function of the cytochrome P450 peroxygenase enzymes. Biochem. Soc. Trans.46, 183–196 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Kaur, M. et al. Agroinfiltration mediated scalable transient gene expression in genome edited crop plants. Int. J. Mol. Sci.22, 10882 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Owen, C., Patron, N. J., Huang, A. & Osbourn, A. Harnessing plant metabolic diversity. Curr. Opin. Chem. Biol.40, 24–30 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Veith, A. & Moorthy, B. Role of cytochrome P450s in the generation and metabolism of reactive oxygen species. Curr. Opin. Toxicol.7, 44–51 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Halliwell, B. & Gutteridge, J. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J.219, 1 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Baker, C. J. & Orlandi, E. W. Active oxygen in plant pathogenesis. Annu. Rev. Phytopathol.33, 299–321 (1995). [DOI] [PubMed] [Google Scholar]
  • 73.Giardi, M. T., Masojídek, J. & Godde, D. Effects of abiotic stresses on the turnover of the D1 reaction centre II protein. Physiol. Plant.101, 635–642 (1997). [Google Scholar]
  • 74.Dixon, R. A. et al. The phenylpropanoid pathway and plant defence—A genomics perspective. Mol. Plant Pathol.3, 371–390 (2002). [DOI] [PubMed] [Google Scholar]
  • 75.Pan, Z. et al. In vivo assembly of the sorgoleone biosynthetic pathway and its impact on agroinfiltrated leaves of Nicotiana benthamiana. New Phytol.230, 683–697 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Park, S., Mani, V., Kim, J. A., Lee, S. I. & Lee, K. Combinatorial transient gene expression strategies to enhance terpenoid production in plants. Front. Plant Sci.13, 1034893 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Patra, B., Schluttenhofer, C., Wu, Y., Pattanaik, S. & Yuan, L. Transcriptional regulation of secondary metabolite biosynthesis in plants. Biochim. Biophys. Acta Gene Regul. Mech.1829, 1236–1247 (2013). [DOI] [PubMed] [Google Scholar]
  • 78.Ramsay, N. A. & Glover, B. J. MYB–bHLH–WD40 protein complex and the evolution of cellular diversity. Trends Plant Sci.10, 63–70 (2005). [DOI] [PubMed] [Google Scholar]
  • 79.Yi, J. et al. A single-repeat MYB transcription factor, GmMYB176, regulates CHS8 gene expression and affects isoflavonoid biosynthesis in soybean. Plant J.62, 1019–1034 (2010). [DOI] [PubMed] [Google Scholar]
  • 80.Zhao, M. et al. Isolation and characterization of GmMYBJ3, an R2R3-MYB transcription factor that affects isoflavonoids biosynthesis in soybean. PLoS ONE12, e0179990 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.Chu, S. et al. An R2R3-type MYB transcription factor, GmMYB29, regulates isoflavone biosynthesis in soybean. PLoS Genet.13, e1006770 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82.Untergasser, A. et al. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Res.35, W71–W74 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.Nichloas, K. B. GeneDoc: Analysis and visualization of genetic variation. EMBnet J.4, 1–4 (1997). [Google Scholar]
  • 84.Tamura, K., Stecher, G. & Kumar, S. MEGA11: Molecular evolutionary genetics analysis version 11. Mol. Biol. Evol.38, 3022–3027 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 85.Chen, C. et al. TBtools: An integrative toolkit developed for interactive analyses of big biological data. Mol. Plant.13, 1194–1202 (2020). [DOI] [PubMed] [Google Scholar]
  • 86.Shannon, P. et al. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res.13, 2498–2504 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 87.DeLano, W. L. Pymol: an open-source molecular graphics tool. CCP4 Newsl. Protein Crystallogr.40, 82–92 (2002). [Google Scholar]
  • 88.Weigel, D. & Glazebrook, J. Transformation of agrobacterium using the freeze-thaw method. CSH Protoc.2006, pdb-prot4666 (2006). [DOI] [PubMed] [Google Scholar]
  • 89.Sharma, O. P. & Bhat, T. K. DPPH antioxidant assay revisited. Food Chem.113, 1202–1205 (2009). [Google Scholar]
  • 90.Moreno, M. I. N., Isla, M. I., Sampietro, A. R. & Vattuone, M. A. Comparison of the free radical-scavenging activity of propolis from several regions of Argentina. J. Ethnopharmacol.71, 109–114 (2000). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information. (20.2KB, xlsx)

Data Availability Statement

All data generated or analyzed during are available in the present article.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES