Evaluation of cytokinin types and LED light spectra for enhanced production of diarylheptanoids in Alnus incana subsp. incana

Abstract

This study aimed to develop an efficient in vitro propagation system for Alnus incana subsp. incana that enables both vigorous shoot proliferation and enhanced production of bioactive diarylheptanoids. Among the tested cytokinins, 5 µM 6-benzylaminopurine (BA) was most effective for promoting shoot proliferation(1.311 ± 0.070), whereas 10 µM zeatin moderately induced shoots but significantly increased secondary metabolites, with oregonin being 1.82-fold, hirsutenone 1.39-fold, and hirsutanonol 1.25-fold higher than BA 5 µM treatment. In terms of light quality, blue LED irradiation (450 nm) led to the highest antioxidant activity, with 2,2-diphenyl-1-picrylhydrazyl(DPPH) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)(ABTS) radical scavenging values of 48.67 ± 27.33% and 92.78 ± 0.35%, respectively. In contrast, white LED (broad-spectrum) supported both balanced growth (1.133 ± 0.051 shoots per explant under 10 µM zeatin) and elevated metabolite production, including oregonin (108.089 µM) and hirsutanonol (13.203 µM). The combination of 10 µM zeatin and white LED thus provided an optimal compromise between shoot proliferation and secondary metabolite accumulation, offering a sustainable strategy for producing antioxidant-rich A. incana plantlets in vitro.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-20160-0.

Introduction

Alnus incana subsp. incana is a fast-growing deciduous tree of the Betulaceae family, widely distributed across the Northern Hemisphere, including Korea, Japan, China and the Russian Far East¹. Its adaptability to nutrient-poor soils has made it a preferred species for afforestation and ecological restoration projects ². In recent years, A. incana has attracted increasing attention for its biotechnological potential, particularly as a source of bioactive secondary metabolites with applications in functional foods and botanical therapeutics³.

Species within the genus Alnus are known to produce diverse secondary metabolites, including flavonoids, triterpenes, sterols, tannins, and phenolic compounds. Among these, diarylheptanoids represent a distinctive class, with oregonin being the first identified and most abundant compound in Alnus species^4–8. Oregonin, along with its hydrolysis product hirsutanonol and the dehydrated derivative hirsutenone, exhibits diverse biological activities (Fig. 1). Oregonin and its derivatives, hirsutanonol and hirsutenone, exhibit a broad range of pharmacological properties, anti-inflammatory⁹, including antioxidant¹⁰, anticancer¹¹, hepatoprotective¹², anti-alopecia¹³, anti-atopic¹⁴, and anti-sarcopenic effects¹⁵. These compounds are believed to exert their effects through phenolic hydroxyl groups, which contribute to their potent antioxidant capacity and their role in alleviating oxidative stress-related disorders such as aging, inflammation, and degenerative diseases^16,17. Similar bioactivities have also been observed in hirsutanonol and hirsutenone^18–20.

Fig. 1.

Chemical structures of major diarylheptanoids found in Alnus incana : oregonin, hirsutanonol, and hirsutenone.

Plant tissue culture provides a viable biotechnological platform for the standardized production of plant biomass and target metabolites under controlled conditions. It enables the propagation of elite genotypes and the manipulation of biosynthetic pathways through culture medium composition, plant growth regulators (PGRs), and environmental factors²¹. Among the commonly used PGRs, 6-benzylaminopurine (BA) and zeatin are key cytokinins known to influence shoot regeneration and metabolite accumulation. BA promotes rapid cell division and callus formation, whereas zeatin supports more organized development and sustained morphogenic responses^22–24. Thidiazuron (TDZ), a synthetic phenylurea derivative with strong cytokinin-like activity, has been reported to effectively stimulate shoot proliferation and enhance morphogenesis in various plant species²⁵. However, its application at higher concentrations may lead to morphological abnormalities such as hyperhydricity and fasciation^25,26. Similarly, 2-isopentenyladenine (2iP), a naturally occurring cytokinin, is known for its relatively mild effects and has been used to promote balanced shoot development and metabolite accumulation²⁶.

Although several tissue cultures studies have been conducted on some Alnus species^27,28, research on A. incana remains in its early stages. To date, there have been no reports on the extraction or utilization of secondary metabolites, including diarylheptanoids, from in vitro-derived tissues of A. incana²⁹.

Light quality is another critical factor in regulating both growth and secondary metabolism. Light-emitting diodes (LEDs) have gained prominence in in vitro culture systems due to their spectral specificity and energy efficiency. They have been shown to influence gene expression, chlorophyll activity, and stress-responsive metabolic pathways, often stimulating the accumulation of defense-related secondary metabolites under specific wavelengths^27–29. Red and blue LED light, in particular, are known to enhance metabolite biosynthesis in many species, including phenolic and flavonoid compounds^30,31. Moreover, LEDs have been reported to significantly induce phenylpropanoid/flavonoid pathways mediated by type III polyketide synthases (e.g., CHS), resulting in 1.2–2.9-fold increases in polyketide metabolites such as rosmarinic acid and tilianin under blue (450–470 nm) and red (660 nm) light conditions^32,33.

Despite increasing interest in A. incana as a medicinal resource, few studies have addressed its micropropagation or secondary metabolite production under in vitro conditions. While some reports exist for other Alnus species^28,34, no prior research has investigated the accumulation of diarylheptanoids from in vitro-derived A. incana tissues².

The objective of this study was to optimize an in vitro culture system for A. incana that enables both efficient plantlet propagation and enhanced production of oregonin and related diarylheptanoids. Specifically, we evaluated the effects of BA and zeatin at various concentrations on shoot proliferation, and assessed the influence of different LED light spectra on metabolite accumulation and antioxidant activity. This research aims to develop a sustainable strategy for producing high-value phytochemicals from A. incana for future use in horticultural biotechnology and health-oriented applications.

Materials and methods

Plant material and culture initiation

Branches of Alnus incana subsp. incana were collected from a natural population in Chuncheon, Republic of Korea. Explants were derived from in vitro-established shoot cultures. Nodal segments (0.2 cm) from these cultures were excised and used as initial explants for all experiments. The basal medium consisted of Woody Plant Medium (WPM; Duchefa Biochemie, Netherlands), supplemented with 30 g/L sucrose, 4 g/L agar, and 1 g/L gelrite. The pH was adjusted to 5.8 prior to autoclaving, and 8 mL of medium was dispensed into each culture tube. The plant material (Alnus incana subsp. incana) was identified by Dr. Hee Kyu Kim at Gangwon State Forest Science Institute. The collection was conducted with permission from relevant national authorities, as the research was funded and approved by the Korea Forest Service (KFS). The voucher specimen (voucher number: KWNU103494) has been deposited in the Kangwon National University Herbarium (KWNU) for future reference.

Screening of cytokinin types

To evaluate the effects of different plant growth regulators on shoot induction, explants were cultured on WPM containing 1 μM of either BA, zeatin, N⁶-(2-isopentenyl)adenine(2iP), thidiazuron(TDZ), or gibberellic acid (GA₃). A hormone-free (HF) medium served as the control. Cultures were maintained at 22 ± 2 °C under a 16/8 h (light/dark) photoperiod with fluorescent lighting. Each treatment consisted of 3 replicates with 15 explants per replicate (n = 45). Growth was assessed after four weeks.

Cytokinin concentration optimization

To determine optimal concentrations, BA and zeatin were applied at 1, 5, 10, and 15 μM. HF was again included as a control. Culture conditions, explant preparation, and assessment criteria were identical to the previous experiment.

LED light quality treatment

Explants cultured on media containing 5 μM BA or 10 μM zeatin (identified as optimal in the previous trial) were exposed to five light sources: fluorescent light (control) and LEDs emitting white (broad spectrum), green (peak: 518 nm; FWHM: 36 nm), red (636 nm; 19 nm), and blue (450 nm; 23 nm) light. The spectral bandwidth (full width at half maximum, FWHM) was determined using a SpectraPen spectroradiometer (Fig. 2). The photosynthetic photon flux density (PPFD) was calibrated and maintained at 80 μmol m⁻² s⁻¹, with a 25 cm distance between light source and culture surface. The white LED exhibited the broadest spectrum (FWHM: 77 nm), while the red LED had the narrowest peak.

Fig. 2.

Relative spectral distribution of the LEDs and fluorescent light.

Growth parameter evaluation

The following growth parameters were measured after four weeks of culture: Fresh weight (g) Number of shoots Shoot length (cm) Induced shoot length (cm) Number of leaves.

Chlorophyll fluorescence analysis

Chlorophyll fluorescence was measured to assess photosynthetic performance under different light conditions. Explants from HF, 5 μM BA, and 10 μM zeatin treatments were exposed to all five light conditions for four weeks. Dark-adapted (20 min) samples were measured using a FluorPen FP-100 fluorometer (Photon Systems Instruments, Czech Republic) in O–J–I–P (OJIP) transient mode under 1500 μmol m⁻² s⁻¹ saturating pulse. Parameters included Fo, Fm, Fv/Fm, Pi_abs, ABS/RC, TRo/RC, ETo/RC, and DIo/RC. Each treatment consisted of three replicates with 15 explants per replicate, and the mean value for each treatment was calculated based on data from a total of 45 explants.

Measured parameters included:

• Fo (Minimal Fluorescence): Minimum fluorescence after dark adaptation.

• Fm (Maximal Fluorescence): Maximum fluorescence after dark adaptation.

• Fv/Fm (Maximum Quantum Yield of PSII): Calculated as (Fm − Fo)/Fm.

• Pi_Abs (Performance Index on Absorption Basis): Reflects overall photosynthetic performance.

• ABS/RC (Absorbed Energy per Reaction Center): Total light energy absorbed by PSII reaction centers.

• TRo/RC (Trapped Energy per Reaction Center): Energy trapped and used in photochemistry.

• ETo/RC (Electron Transport per Reaction Center): Energy used for electron transport.

• DIo/RC (Dissipated Energy per Reaction Center): Energy dissipated as heat.

Extraction and HPLC analysis of diarylheptanoids

The Alnus incana culture was extracted at room temperature for 1 week using 100% ethanol at a ratio of 10 times the fresh weight (1 g/10 mL), followed by filtration through filter paper. The extraction solutions were concentrated using a rotary evaporator at 45 °C.

The standard compounds—oregonin, hirsutanonol, and hirsutenone—were isolated and purified from Alnus incana in the Department of Forest Biomaterials Engineering at the College of Forest and Environmental Sciences, Kangwon National University.

High-performance liquid chromatography (HPLC) analysis was performed using a Waters 2695 Separation Module and a Waters 2487 Dual-λ Absorbance Detector (Waters, Massachusetts, USA). Chromatographic separation was carried out on a Hector C18 analytical column (250 × 4.6 mm, 5 μm), which was fitted with a Phenomenex KJ0-4282 guard column. The column temperature was maintained at 25 ℃. The mobile phase consisted of 1% acetic acid in water (A) and acetonitrile (B). The flow rate was set at 1 mL/min, the injection volume was 20 μL, and a detection wavelength set at 280 nm (Supplementray Table S1). An elution gradient was employed as follows: 0 min at 10% B, 0–12 min at 25% B, 12–24 min at 40% B, 24–25 min at 10% B, and 25–40 min at 10% B (Supplementray Table S2).

For quantitative analysis, each standard compound—oregonin, hirsutanonol, and hirsutenone—was dissolved in methanol and then diluted to a range of 1–100 μg/mL. Calibration plots for oregonin, hirsutanonol, and hirsutenone were constructed using the peak areas (y) obtained from 6 different concentration solutions (x) to establish each standard curve: oregonin (y = 17223x − 20,314, R² = 0.9976), hirsutanonol (y = 13167x − 385.69, R² = 1), hirsutenone (y = 15625x + 2161.2, R² = 0.9999) (Supplementray Fig. S1). Based on these result, the calculated limit of detection (LOD) and limit of quantitation (LOQ) were as follows: LOD 1.00 and LOQ 3.02 for oregonin; LOD 1.06 and LOQ 3.21 for hirsutanonol; and LOD 1.95 and LOQ 5.91 for hirsutenone (Supplementray Fig. S1). The formula used to calculate the LOD and LOQ are as follows:

1

2

‘SE’ = Standard Error of the y-intercept, ‘N’ = The number of tests, ‘S’ = Slope of the calibration curve.

All analytical samples were dissolved in methanol at a concentration of 1000 μg/mL and filtered through a 0.2 μm syringe filter to prepare the test solutions. The analysis was repeated three times to ensure reliability. The identification of each compound was conducted by comparing its retention time to that of a standard compound. Quantitative analysis was conducted using the regression equation derived from the standard calibration curve. This analysis was subsequently validated using LOD, which indicates the lowest concentration at which the presence of the target compound can be clearly identified, and LOQ, which indicates the lowest concentration that ensures the reliability of the results.

Antioxidant analysis

DPPH radical scavenging assay

2,2-diphenyl-1-picrylhydrazyl(DPPH) radical scavenging activity was measured by modifying Hatano`s method³⁵. A stock solution of the DPPH radical was prepared by dissolving 0.04 g of DPPH (2,2-diphenyl-1-picrylhydrazyl) (Sigma-Aldrich, St Louis, MO, USA) in 500 mL of ethanol. The working solution was obtained by diluting the stock solution with ethanol to achieve an absorbance of 0.95–0.99 at a wavelength of 517 nm. Absorbance was measured using an INNO microplate spectrophotometer (TEK, Seongnam, Gyeonggido, South Korea). The sample was prepared by dissolving it in methanol at a concentration of 1000 μg/mL and then diluting it to 6 concentrations: 1000, 500, 250, 125, 62.5, and 31.25 μg/mL. L( +)-ascorbic acid (Vit.C), a strong reducing agent, was used as a positive control at the same concentrations as the samples. Methanol was used as a negative control (blank). 20uL of each sample concentration was reacted with 180μL of the DPPH working solution in a 96-well microplate for 30 min at room temperature in the dark. Subsequently, the absorbance was measured at a wavelength of 517 nm. A lower absorbance of the reaction mixture indicates a higher free radical scavenging activity. The DPPH free radical scavenging activity was calculated as a percentage using the following formula:

‘OD_sample’ = The absorbance of the sample reaction mixture. ‘OD_CON’ = The absorbance of the working solution. ‘OD_Blank’ = The absorbance of the blank reaction mixture.

The IC₅₀ value, which represents the concentration of a sample required to reduce free radicals by 50%, was calculated by linear regression of inhibition percentage and concentration data using Microsoft Office Ecel FORECAST function. Lower IC₅₀ value indicates a higher free radical scavenging activity. The experiments were conducted in triplicate.

ABTS radical scavenging assay

2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid(ABTS) radical scavenging activity was measured by modifying Re`s method³⁶. The stock solution of the ABTS radical was prepared by reacting ABTS (2,2azinobis-(3-ethylbenzo-6-sulphonate)) (Sigma-Aldrich, St Louis, MO, USA), diluted in water to 7.0 mM, with potassium persulfate (Samchun Chemical Corporation, Gyeonggi, Korea), diluted in water to 2.45 mM, in a 1:1 ratio at room temperature for 12–16 h in the dark. The working solution was obtained by diluting the stock solution with PBS buffer (pH 7.4) to achieve an absorbance of 0.68–0.72 at a wavelength of 750 nm. The samples and controls were identical to those used in the DPPH assay. 20uL of each concentration of the sample was reacted with 180μL of the ABTS working solution in a 96-well microplate for 15 min at room temperature in the dark. Subsequently, the absorbance was measured at a wavelength of 750 nm. A lower absorbance of the reaction mixture indicates higher free radical scavenging activity. The ABTS free radical scavenging activity was expressed as a percentage and IC₅₀ value, consistent with the DPPH assay.

Composite index calculation and data standardization

In this study, the effects of varying concentrations of zeatin and BA on the growth and chemical composition of Alnus incana were evaluated. A range of treatment groups was established to assess multiple physiological and biochemical responses. Data were collected in three main categories: (1) growth parameters, including fresh weight, number of shoots, shoot length, length of induced shoots, and number of leaves; (2) antioxidant activity, assessed via IC₅₀ values obtained from DPPH and ABTS radical scavenging assays; and (3) secondary metabolite content, focusing on the quantification of oregonin, hirsutanonol, and hirsutenone. The number of replicates varied depending on the data type, with growth-related parameters measured in 45 replicates and biochemical assays conducted in triplicate.

To integrate these heterogeneous datasets, two complementary analytical approaches were employed: the efficiency index and the T-score. First, all indicator values were normalized to a 0–1 scale using min–max normalization, wherein each value was divided by the maximum observed value within the dataset:

3

The unweighted efficiency index was then calculated as the average of all normalized values:

4

To account for the relative contribution of each indicator, principal component analysis (PCA) was applied. Weights were derived from the proportion of variance explained by each indicator in the PCA results:

5

The weighted efficiency index was then calculated as:

6

In parallel, a T-score-based standardization was conducted. Each raw value was first converted into a Z-score:

7

The Z-score was then transformed into a T-score using the following equation:

8

Finally, the PCA-derived weights were applied to compute the weighted T-score:

9

Given the discrepancy in sample sizes among datasets (e.g., growth data: n = 45 vs. biochemical data: n = 3), adjustments were made to ensure balanced representation across variables. After normalization and adjustment for sample size, PCA was recalculated using the integrated dataset. Final efficiency indices and weighted T-scores were used to comparatively evaluate the relative performance of each treatment condition in terms of promoting growth, antioxidant capacity, and secondary metabolite accumulation.

Statistics analysis

Statistical analyses were conducted using R software (v4.2.2). One-way and two-way analysis of variance (ANOVA) and Pearson correlation were performed. Tukey’s honestly significant difference HSD was used for post hoc comparisons. Results are presented as mean ± SE, with significance set at p < 0.05.

Results

Effect of plant growth regulator (PGR) type on in vitro growth and secondary metabolite production

One-way ANOVA revealed that 1 μM BA significantly enhanced all measured growth parameters, including shoot number, shoot length, fresh weight, and number of leaves (Table 1, Supplementray Fig. S2). In contrast, GA₃ treatment resulted in the poorest growth performance across all indicators.

Table 1.

Effect of plant growth regulator (PGR) type on growth parameters of A. incana explants after 4 weeks of culture.

Treatment (µM)	Fresh weight(g)	No. of shoot induced	Shoot length(cm)	Shoot induced length(cm)	No. of leaves
HF	0.007 ± 0.001C	1.020 ± 0.020B	0.488 ± 0.029AB	0.002 ± 0.002B	2.100 ± 0.170C
2iP 1	0.010 ± 0.001BC	1.120 ± 0.046AB	0.510 ± 0.029A	0.028 ± 0.011AB	2.640 ± 0.195BC
Zeatin 1	0.013 ± 0.001AB	1.100 ± 0.052AB	0.558 ± 0.046A	0.080 ± 0.021B	3.460 ± 0.244AB
BA 1	0.017 ± 0.001A	1.280 ± 0.070A	0.558 ± 0.041A	0.024 ± 0.013A	4.360 ± 0.425A
TDZ 1	0.016 ± 0.001A	1.180 ± 0.062AB	0.486 ± 0.024AB	0.054 ± 0.019AB	2.020 ± 0.163C
GA3 1	0.006 ± 0.000C	1.000 ± 0.000B	0.358 ± 0.021B	0.000 ± 0.000B	0.720 ± 0.128D

Values are means ± standard error (n = 45). Different letters within each column indicate significant differences (p < 0.05) according to Tukey’s HSD test.

HPLC analysis showed that 1 μM zeatin significantly increased the accumulation of all three diarylheptanoids—oregonin, hirsutanonol, and hirsutenone—compared to other PGR treatments (Table 2, Supplementray Fig. S3). While BA treatment promoted the second-highest level of hirsutenone, it showed relatively low levels of oregonin and hirsutanonol.

Table 2.

Concentration of oregonin, hirsutanonol, and hirsutenone (μg/mL) in A. incana explants cultured under different PGR treatments.

Treatment(uM)	Oregonin(ug/mL)	Hirsutanonol(ug/mL)	Hirsutenone(ug/mL)
HF	96.980 ± 2.370BC	10.083 ± 0.913B	N.D
2iP 1	108.096 ± 0.657B	8.042 ± 0.187BC	7.095 ± 0.062D
Zeatin 1	189.414 ± 5.265A	15.020 ± 0.692A	20.291 ± 1.232A
BA 1	90.370 ± 1.662C	7.652 ± 0.018CD	17.434 ± 0.039B
TDZ 1	47.340 ± 2.555D	15.858 ± 0.269A	14.527 ± 0.260C
GA3 1	47.596 ± 1.233D	5.522 ± 0.113D	N.D

Values are means ± standard error (n = 3). Different letters indicate statistically significant differences (p < 0.05).

Antioxidant activity under different PGR treatments

Antioxidant activity was assessed using ABTS and DPPH radical scavenging assays, and both assays showed a concentration-dependent increase in radical scavenging capacity, confirming the reliability of the measurements (Fig. 3). In the ABTS assay (Fig. 3A), the 10 μM zeatin treatment exhibited the highest overall antioxidant activity, with scavenging levels comparable to the positive control (ascorbic acid(vitamin C)) across most concentrations. The 5μM zeatin treatment also demonstrated strong scavenging activity, showing significantly higher values than other treatments in the 12.5–50μg/mL concentration range.

Fig. 3.

Comparison of ABTS (A) and DPPH (B) radical scavenging activity (%) at various cytokinins in A. incana in vitro cultures. Bars represent means (N = 3), and error bars indicate standard errors.

In contrast, treatments with BA and GA₃ exhibited relatively low antioxidant activities. Specifically, 1 μM GA₃ and 1 μM BA treatments showed poor scavenging capacity at concentrations below 50μg/mL.

Similar trends were observed in the DPPH assay (Fig. 3B). The 10 μM zeatin treatment consistently exhibited the highest DPPH radical scavenging activity across all concentrations, followed closely by the 5μM zeatin group. BA and GA₃ treatments showed lower activity at lower concentrations (3.1–12.5μg/mL), although their activity increased at higher concentrations (50–100μg/mL). Notably, the 1μM TDZ treatment demonstrated intermediate antioxidant activity and was more effective than both BA and GA₃ treatments.

Overall, both ABTS and DPPH assays identified 10μM zeatin as the most effective treatment for enhancing antioxidant activity, followed by 5 μM zeatin and 1 μM TDZ. In contrast, BA and GA₃ treatments resulted in the lowest scavenging activities, particularly at lower concentrations.

To validate our antioxidant assay, ascorbic acid was used as a positive control. The IC₅₀ values for ascorbic acid were determined to be 3.96 ± 0.22 µg/mL in ABTS assay and 5.53 ± 0.19 µg/mL in DPPH assay. This result is consistent with previously reported values for ascorbic acid in range of 3.9–8.4 µg/mL in ABTS assay and of 4.9–8.4 µg/mL in DPPH assay under similar experimental conditions. IC₅₀ values supported these findings, with the 10 μM zeatin treatment showing the lowest IC₅₀ values for both ABTS (4.36 ± 0.20 μg/mL) and DPPH (18.98 ± 2.38 μg/mL), indicating the highest antioxidant potential (Table 3). Conversely, the GA₃ treatment showed the highest IC₅₀ values in both ABTS (15.34 ± 0.95 μg/mL) and DPPH (56.20 ± 8.81 μg/mL) assays, indicating the lowest antioxidant activity.

Table 3.

Antioxidant activity (IC₅₀ values in μg/mL) of A. incana extracts under different PGR treatments, measured by ABTS and DPPH assays.

Treatment (uM)	HF	2iP 1	Zeatin 1	BA 1	TDZ 1	GA3 1	VIT.C
ABTS IC50	8.15 ± 0.34	7.58 ± 0.18	4.36 ± 0.20	13.51 ± 0.40	11.74 ± 0.73	15.34 ± 0.95	3.96 ± 0.22
DPPH IC50	27.81 ± 4.74	23.42 ± 1.25	18.98 ± 2.38	33.02 ± 3.73	40.18 ± 5.36	56.20 ± 8.81	5.53 ± 0.19

Values are means ± standard error (n = 3). Lower values denote higher antioxidant capacity. Different letters indicate significant differences (p < 0.05).

Optimization of BA and zeatin concentrations for in vitro growth

BA at 5 μM significantly improved all growth parameters, including fresh weight, shoot number, and leaf number (Table 4). Zeatin at 10 μM produced the second-best growth results. At 15 μM, both BA and zeatin reduced growth performance, indicating a threshold beyond which higher cytokinin concentrations become inhibitory.

Table 4.

Growth performance of A. incana explants cultured with varying concentrations of BA and zeatin.

Teatment (µM)	Fresh weight(g)	No. of shoot induced	Shoot length(cm)	Shoot induced length(cm)	No. of leaves
HF	0.007 ± 0.000CD	1.067 ± 0.038B	0.471 ± 0.026ABC	0.011 ± 0.007C	2.422 ± 0.147B
Zeatin 1	0.008 ± 0.001BCD	1.044 ± 0.031B	0.484 ± 0.049ABC	0.007 ± 0.005C	2.511 ± 0.179B
Zeatin 5	0.010 ± 0.001ABCD	1.111 ± 0.047AB	0.449 ± 0.030BC	0.029 ± 0.014BC	2.867 ± 0.247AB
Zeatin 10	0.012 ± 0.001AB	1.267 ± 0.067AB	0.529 ± 0.031AB	0.089 ± 0.026AB	2.889 ± 0.251AB
Zeatin 15	0.007 ± 0.001D	1.067 ± 0.038B	0.358 ± 0.023C	0.009 ± 0.005C	2.089 ± 0.193BC
BA 1	0.010 ± 0.001ABCD	1.111 ± 0.047AB	0.387 ± 0.022C	0.033 ± 0.017BC	3.600 ± 0.150A
BA 5	0.013 ± 0.001A	1.311 ± 0.070A	0.580 ± 0.030A	0.109 ± 0.027A	3.644 ± 0.268A
BA 10	0.011 ± 0.001ABC	1.111 ± 0.047AB	0.371 ± 0.020C	0.016 ± 0.007C	2.178 ± 0.253BC
BA 15	0.011 ± 0.001BCD	1.267 ± 0.067AB	0.449 ± 0.023BC	0.062 ± 0.016ABC	1.267 ± 0.178C

Values are means ± standard error (n = 45). Significant differences within each parameter are indicated by different letters (p < 0.05).

Effect of BA and zeatin concentration on diarylheptanoid accumulation

HPLC analysis revealed that all three diarylheptanoids—oregonin, hirsutanonol, and hirsutenone—were most abundant in the 15 μM zeatin treatment group (Table 5, Supplementray Fig. S5). In general, zeatin treatments showed a concentration-dependent increase in secondary metabolite accumulation, with higher zeatin concentrations corresponding to greater compound levels.

Table 5.

Diarylheptanoid content (μg/mL) in A. incana explants under different concentrations of BA and zeatin.

Treatment(uM)	Oregonin(ug/mL)	Hirsutanonol(ug/mL)	Hirsutenone(ug/mL)
HF	136.963 ± 1.182D	5.666 ± 0.070CD	8.361 ± 0.081C
Zeatin 1	85.409 ± 0.323F	4.873 ± 0.057D	< LOQ
Zeatin 5	145.687 ± 0.560C	8.619 ± 0.023BCD	15.223 ± 0.763B
Zeatin 10	161.991 ± 0.545B	11.140 ± 0.217B	16.271 ± 0.068AB
Zeatin 15	247.340 ± 0.488A	16.691 ± 0.010A	20.868 ± 0.015A
BA 1	107.567 ± 1.482E	4.998 ± 2.833D	13.671 ± 0.268BC
BA 5	57.447 ± 0.921F	< LOQ	< LOQ
BA 10	104.126 ± 0.119E	12.879 ± 0.207AB	10.941 ± 0.884BC
BA 15	66.047 ± 0.128G	9.986 ± 0.011BC	8.518 ± 2.683C

Values are means ± standard error (n = 3). Asterisks indicate significant differences relative to the HF control.

In contrast, BA treatments resulted in overall lower levels of the target compounds. Notably, in the 5 μM BA treatment group, which showed the most favorable growth performance, only oregonin was detected, while the other two compounds were below detectable levels.

Antioxidant capacity at varying cytokinin concentrations

Antioxidant activity measured via ABTS and DPPH assays showed a clear concentration-dependent increase across all treatments (Fig. 4). In the ABTS radical scavenging assay (Fig. 4A), the 10 μM and 15 μM zeatin treatments exhibited significantly higher scavenging activity than other treatments in the 3.1–12.5 μg/mL concentration range. At higher concentrations (25–100 μg/mL), most treatments showed more than 90% scavenging activity, indicating a plateau in antioxidant performance. As expected, the positive control (vitamin C) exhibited the highest antioxidant activity across all concentrations, while the hormone-free (HF) group showed relatively low activity throughout.

Fig. 4.

Comparison of ABTS (A) and DPPH (B) radical scavenging activity (%) at various BA and Zeatin concentrations in A. incana in vitro cultures. Bars represent means (N = 3), and error bars indicate standard errors.

A similar trend was observed in the DPPH assay (Fig. 4B). In the 3.1–12.5 μg/mL range, both 10 μM and 15 μM zeatin treatments demonstrated significantly higher scavenging activity compared to BA-treated and HF groups. At concentrations above 25 μg/mL, zeatin treatments at 5, 10, and 15 μM consistently maintained higher DPPH scavenging activity than BA and HF treatments.

The IC₅₀ values of ascorbic acid were 3.007 ± 0.172 µg/mL in the ABTS assay and 4.992 ± 0.325 µg/mL in the DPPH assay, consistent with values reported in the literature and supporting the accuracy of this antioxidant assay. IC₅₀ values further supported these observations. The 15 μM zeatin treatment recorded the lowest IC₅₀ values for both ABTS (4.226 ± 0.229 μg/mL) and DPPH (17.375 ± 1.169 μg/mL), indicating the strongest antioxidant capacity (Table 6). In contrast, the 1 μM zeatin treatment exhibited the highest IC₅₀ values—22.211 ± 0.739 μg/mL for ABTS and 66.332 ± 2.056 μg/mL for DPPH—suggesting the weakest activity among all zeatin concentrations.

Table 6.

Antioxidant activity (IC₅₀) of A. incana extracts under varying concentrations of BA and zeatin.

Treatment(uM)	ABTS IC50 (μg/mL)	DPPH IC50 (μg/mL)
HF	8.474 ± 0.121	27.603 ± 1.679
Zeatin 1	22.211 ± 0.739	66.332 ± 2.056
Zeatin 5	13.708 ± 0.216	42.422 ± 0.499
Zeatin 10	8.661 ± 0.227	30.514 ± 0.157
Zeatin 15	4.226 ± 0.229	17.375 ± 1.169
BA 1	9.697 ± 0.300	35.224 ± 0.598
BA 5	11.070 ± 0.454	43.299 ± 0.193
BA 10	10.572 ± 0.092	45.529 ± 0.843
BA 15	11.440 ± 0.320	45.970 ± 1.237
VIT.C	3.007 ± 0.172	4.992 ± 0.325

Results expressed as means ± standard error (n = 3). Different letters indicate significant differences (p < 0.05).

Overall, zeatin treatments displayed an increasing trend in antioxidant activity with rising concentration, whereas BA treatments showed the opposite tendency, with higher antioxidant activity at lower concentrations.

Integrated analysis of growth and metabolite accumulation: T-score evaluation

T-score analysis integrating growth, metabolite, and antioxidant data identified 10 μM zeatin as the most effective treatment overall (average T-score: 57.38), followed by 5 μM BA (T-score: 53.35) (Table 7). These two conditions, along with the HF control, were selected for subsequent LED light quality experiments.

Table 7.

Efficiency and weighted index scores derived from normalized growth, metabolite, and antioxidant data for each treatment condition.

Treatment (μM)	Fresh weight	No. of shoot induced	Shoot length	Shoot induced length	No. of leaves	Oregonin	Hirsutanonol	Hirsutenone	DPPH (IC₅₀)	ABTS (IC₅₀)	Average T-score
HF	36.55	41.78	52.43	42.13	47.53	52.29	43.92	43.69	55.85	52.05	46.82
Zeatin 1	41.2	39.52	54.2	41.06	48.71	43.45	39.21	41.97	38.88	38.39	42.66
Zeatin 5	50.52	46.1	49.44	46.92	53.46	53.78	56.81	50.13	45.7	43.61	49.65
Zeatin 10	59.83	61.43	60.31	62.9	53.75	56.58	58.78	55.62	53.08	51.57	57.38
Zeatin 15	36.55	41.78	37.08	41.6	43.09	71.2	67.42	67.71	72.97	74.26	55.37
BA 1	50.52	46.1	41.02	47.99	63.23	47.25	53.9	42.24	49.56	49.26	49.11
BA 5	64.49	65.75	67.24	68.23	63.82	38.66	36.98	36.14	45.31	46.87	53.35
BA 10	55.17	46.1	38.84	43.46	44.27	46.66	48.77	59.41	44.41	47.67	47.48
BA 15	55.17	61.43	49.44	55.71	32.13	40.13	44.22	53.1	44.24	46.32	48.19

T-scores used for integrative evaluation of optimal in vitro conditions.

Interaction effects of LED light quality and PGR on growth performance

Two-way ANOVA was conducted to evaluate the effects of plant growth regulator (PGR) treatment, LED light condition, and their interaction on plant growth parameters (Table 8, Supplementray Fig. S5). The main effect of treatment was statistically significant for all growth indicators, with very large effect sizes for fresh weight (η² = 0.904) and number of leaves (η² = 0.916), indicating that PGR treatment explained the vast majority of variance in these traits. The main effect of LED light was significant only for shoot length and number of leaves. A significant treatment × LED interaction effect was observed for fresh weight, shoot length, and number of leaves, whereas no significant effects were found for the number and length of induced shoots with respect to either the LED main effect or the interaction. These interaction effects highlight that optimal growth responses cannot be achieved by PGR or LED treatment alone, but rather require specific combinations of both factors.

Table 8.

Growth parameters of A. incana explants subjected to combinations of plant growth regulator (PGR) and LED light treatments.

Teatment (µM)	LED	Fresh weight(g)	No. of shoot induced	Shoot length(cm)	Shoot induced length(cm)	No. of leaves
HF	Control	0.005 ± 0.000Ca	1.044 ± 0.031Ba	0.449 ± 0.021Ba	0.013 ± 0.009Ba	1.200 ± 0.151Ca
	White	0.005 ± 0.000Cab	1.022 ± 0.022Ba	0.358 ± 0.016Bc	0.007 ± 0.007Ba	1.133 ± 0.173Cab
	Green	0.004 ± 0.000Cab	1.000 ± 0.000Ba	0.431 ± 0.015Bab	0.000 ± 0.000Ba	0.644 ± 0.115Cb
	Red	0.004 ± 0.000Cb	1.022 ± 0.022Ba	0.379 ± 0.014Bbc	0.002 ± 0.002Ba	0.800 ± 0.108Cab
	Blue	0.004 ± 0.000Cab	1.000 ± 0.000Ba	0.347 ± 0.014Bc	0.000 ± 0.000Ba	0.889 ± 0.111Cab
Zeatin 10	Control	0.008 ± 0.001Ab	1.067 ± 0.038Aa	0.493 ± 0.035Abc	0.020 ± 0.013Aa	2.244 ± 0.214Ab
	White	0.014 ± 0.001Aa	1.133 ± 0.051Aa	0.704 ± 0.041Aa	0.064 ± 0.028Aa	3.822 ± 0.183Aa
	Green	0.012 ± 0.001Aa	1.133 ± 0.051Aa	0.627 ± 0.055Aab	0.036 ± 0.015Aa	3.178 ± 0.204Aa
	Red	0.012 ± 0.001Aa	1.067 ± 0.038Aa	0.576 ± 0.060Aabc	0.058 ± 0.032Aa	3.200 ± 0.226Aa
	Blue	0.011 ± 0.000Aa	1.178 ± 0.058Aa	0.411 ± 0.022Ac	0.033 ± 0.011Aa	3.178 ± 0.228Aa
BA 5	Control	0.010 ± 0.001Ba	1.090 ± 0.010Aa	0.391 ± 0.003Bab	0.022 ± 0.002Aa	2.290 ± 0.037Ba
	White	0.008 ± 0.001Bab	1.110 ± 0.010Aa	0.389 ± 0.003Bab	0.027 ± 0.002Aa	1.160 ± 0.041Bbc
	Green	0.008 ± 0.001Bb	1.160 ± 0.011Aa	0.356 ± 0.003Bab	0.033 ± 0.002Aa	0.622 ± 0.028Bc
	Red	0.009 ± 0.001Bab	1.160 ± 0.011Aa	0.422 ± 0.004Ba	0.049 ± 0.002Aa	1.360 ± 0.043Bb
	Blue	0.009 ± 0.001Bab	1.111 ± 0.010Aa	0.331 ± 0.003Bb	0.013 ± 0.001Aa	1.510 ± 0.034Bb
Significance of two-way ANOVA
Treatment		***	***	***	**	***
LED		N.S	N.S	***	N.S	**
Treatment × LED interaction		***	N.S	***	N.S	***
Treatment	df	1	1	1	1	1
	F-value	329.0	17.9	63.4	14.5	382
	p-value	***	***	***	***	***
	η2	0.904	0.338	0.644	0.293	0.916
LED	df	4	4	4	4	4
	F-value	4.29	0.46	6.69	0.64	4.94
	p-value	**	N.S	***	N.S	***
	η2	0.329	0.050	0.433	0.068	0.361
Treatment × LED interaction	df	4	4	4	4	4
	F-value	7.76	1.53	6.53	0.87	0.69
	p-value	***	N.S	***	N.S	***
	η2	0.470	0.149	0.427	0.090	0.073

A, B, C: Differences among treatments (p < 0.05). Values are presented as mean ± standard error(n = 45). Different letters within a column indicate significant differences at p < 0.05 according to Tukey’s HSD test. a, b, c: Differences among LEDs (p < 0.05). Significance levels: ***p < 0.001, *p < 0.01, *p < 0.05, N.S = not significant.

Among the combinations tested, the zeatin + white LED treatment produced the highest values across all growth indicators—including fresh weight, shoot length, induced shoot length, and number of leaves—demonstrating its effectiveness in promoting overall plant growth. The zeatin + blue LED combination showed the highest number of induced shoots, indicating its specific effect in enhancing shoot induction. In contrast, the BA + green LED treatment resulted in lower values for fresh weight and number of leaves, suggesting a limited growth-promoting effect. The HF (hormone-free) group combined with either blue or green LED showed the lowest performance in terms of fresh weight, shoot length, and number of leaves, indicating minimal influence on growth without hormonal supplementation.

While the zeatin + green LED treatment produced moderate results in some growth indicators, its overall effectiveness was inferior compared to the zeatin + white or zeatin + blue LED combinations. These findings indicate that white LED is particularly effective for promoting biomass accumulation, while blue LED is more favorable for shoot induction. The synergistic effect observed with zeatin suggests that the appropriate combination of PGR and light quality plays a critical role in maximizing in vitro plant growth responses.

LED light and PGR effects on secondary metabolite accumulation

Two-way ANOVA revealed that both the main effects (treatment and LED condition) and their interaction had statistically significant impacts on the concentrations of oregonin, hirsutanonol, and hirsutenone (Table 9, Supplementary Fig. S6). Effect size (η²) analysis confirmed that treatment accounted for nearly all variance in oregonin (0.996), hirsutanonol (0.980), and hirsutenone (0.943), underscoring the predominant role of PGRs. LED effects, while smaller, still explained a substantial proportion of variance (η² = 0.632–0.991), particularly for oregonin.

Table 9.

Accumulation of oregonin, hirsutanonol, and hirsutenone (μg/mL) in A. incana explants under combined LED and PGR treatments.

Treatment(μM)	LED	Oregonin	Hirsutanonol	Hirsutenone
HF	Control	56.399 ± 1.103Ae	4.340 ± 0.087ABd	< LOQ
	White	137.209 ± 2.700Aa	13.109 ± 0.120ABa	< LOQ
	Green	85.513 ± 1.987Ac	5.726 ± 0.263ABc	< LOQ
	Red	69.158 ± 1.804Ad	6.145 ± 1.101ABc	< LOQ
	Blue	96.825 ± 0.588Ab	7.669 ± 0.094ABb	< LOQ
Zeatin 10	Control	59.309 ± 0.397Ad	6.691 ± 0.025Ad	5.986 ± 0.121Ac
	White	108.089 ± 0.538Ab	13.203 ± 0.007Aa	9.883 ± 0.045Aa
	Green	55.041 ± 0.292Ae	6.514 ± 0.017Ad	< LOQ
	Red	86.135 ± 0.681Ac	7.938 ± 0.018Ac	7.075 ± 0.072Ac
	Blue	115.79 ± 0.112Aa	9.667 ± 0.193Ab	8.791 ± 0.503Ab
BA 5	Control	26.435 ± 0.094Bd	3.923 ± 0.017Be	< LOQ
	White	35.895 ± 0.170Bb	5.143 ± 0.042Bc	< LOQ
	Green	58.676 ± 0.129Ba	6.501 ± 0.037Bb	< LOQ
	Red	30.073 ± 0.038Bc	7.050 ± 0.014Ba	< LOQ
	Blue	19.226 ± 0.064Be	4.167 ± 0.088Bd	< LOQ
Significance of two-way ANOVA
Treatment		***	***	***
LED		***	***	***
Treatment × LED interaction		***	***	***
Treatment	df	2	2	2
	F-value	2292.968	725.735	249.261
	p-value	***	***	***
	η2	0.996	0.980	0.943
LED	df	4	4	4
	F-value	197.834	180.027	12.907
	p-value	***	***	***
	η2	0.991	0.960	0.632
Treatment × LED interaction	df	8	8	8
	F-value	458.340	406.909	43.290
	p-value	***	***	***
	η2	0.992	0.991	0.920

Values are presented as mean ± standard error(n = 3). Different letters within a column indicate significant differences at p < 0.05 according to Tukey’s HSD test. a, b, c: Differences among LEDs (p < 0.05). Significance levels: ***p < 0.001, **p < 0.01, *p < 0.05.

A significant treatment × LED interaction effect was observed for all three compounds, with very high η² values (0.920–0.992), demonstrating that metabolite accumulation was determined by the synergistic interplay between hormones and light rather than by either factor alone. For oregonin, the highest concentration was observed in HF + white LED (137.209 μM), while zeatin 10 μM under white and blue LEDs also produced elevated levels. Hirsutenone was exclusively detected under zeatin treatments with control and white/red LEDs, highlighting the specificity of cytokinin-light interactions. By contrast, BA consistently yielded lower metabolite concentrations across all light treatments, further reinforcing the critical role of zeatin in diarylheptanoid biosynthesis.

Antioxidant activity under combined LED and PGR treatments

The effects of LED light conditions and plant growth regulator (PGR) treatments on antioxidant activity were analyzed using ABTS and DPPH radical scavenging assays (Fig. 5). A strong dose-dependent response was observed in both assays, with radical scavenging activity increasing consistently with higher sample concentrations. Distinct differences among treatment groups were most clearly observed at 50 and 100 µg/mL concentrations.

Fig. 5.

Comparison of ABTS (A) and DPPH (B) radical scavenging activity (%) at various HF, Z 10, B 5, and VIT.C concentrations in A. incana in vitro cultures under different LED conditions. Bars represent means (N = 3), and error bars indicate standard errors.

Among the LED conditions, white LED was the most effective in differentiating antioxidant activity between treatment and control groups. In particular, combinations of white LED with zeatin 10 μM and BA 5 μM treatments showed high radical scavenging activity, approaching that of the positive control (vitamin C). Blue LED also enhanced antioxidant activity in the zeatin 10 μM and BA 5 μM groups, ranking second only to white LED in efficacy.

In contrast, green LED and the fluorescent control light conditions had relatively limited effects on enhancing antioxidant activity. While zeatin 10 μM and BA 5 μM treatments exhibited lower antioxidant activity compared to the positive control, both treatments showed improved performance under white and blue LED conditions. The hormone-free (HF) group consistently recorded the lowest radical scavenging activity under all LED conditions, indicating negligible antioxidant effects.

These results suggest that white LED is the most effective condition for maximizing antioxidant activity, particularly when combined with zeatin or BA treatments.

The IC₅₀ values for ABTS and DPPH radical scavenging activity were analyzed across various treatment and LED light conditions (Table 10). The IC₅₀ values of ascorbic acid were 3.887 ± 0.098 µg/mL in the ABTS assay and 8.963 ± 0.179 µg/mL in the DPPH assay, consistent with values reported in the literature and supporting the accuracy of this antioxidant assay. The hormone-free (HF) group exhibited low antioxidant activity under the control (fluorescent light) condition; however, its activity increased significantly under white and blue LED conditions. The zeatin 10 μM treatment showed the highest antioxidant activity under both blue and white LED lights, with blue LED yielding the most pronounced enhancement.

Table 10.

Antioxidant activity (IC₅₀ values) of A. incana extracts under combined PGR and LED light treatments.

Treatment(μM)	LED	ABTS IC50	DPPH IC50
HF	Control	25.796 ± 0.914	69.752 ± 1.272
	White	10.960 ± 0.283	33.194 ± 1.297
	Green	50.296 ± 3.607	44.466 ± 1.833
	Red	18.259 ± 0.338	49.555 ± 1.366
	Blue	11.949 ± 0.744	32.030 ± 0.515
Zeatin 10	Control	16.392 ± 0.342	45.478 ± 1.348
	White	9.673 ± 0.192	34.241 ± 0.432
	Green	13.196 ± 0.585	46.849 ± 1.156
	Red	14.444 ± 0.255	37.714 ± 0.302
	Blue	9.134 ± 0.177	22.254 ± 0.210
BA 5	Control	23.465 ± 0.581	76.379 ± 3.899
	White	18.130 ± 0.508	65.008 ± 1.723
	Green	11.803 ± 0.357	31.059 ± 1.759
	Red	17.483 ± 0.473	56.837 ± 1.965
	Blue	14.644 ± 0.243	65.512 ± 0.743
VIT.C		3.887 ± 0.098	8.963 ± 0.179

Results are means ± standard error (n = 3). Different letters indicate significant differences (p < 0.05).

The BA 5 μM treatment demonstrated favorable antioxidant activity under green, white, and blue LED conditions, but its performance under the control condition was relatively low. Among the different LED conditions, white LED was the most effective in enhancing antioxidant activity across most treatments. Blue LED also showed strong efficacy, particularly in the zeatin 10 μM treatment group.

In contrast, control and green LED conditions resulted in comparatively limited antioxidant enhancement. These results indicate that the combination of LED light quality and PGR treatment significantly influences antioxidant activity. Among the tested conditions, white and blue LEDs were identified as the most effective for enhancing radical scavenging capacity.

Photosynthetic efficiency under LED × PGR conditions (chlorophyll fluorescence)

Chlorophyll fluorescence parameters ABS/RC, DIo/RC, and TRo/RC showed similar trends across treatments(Fig. 6). In general, Zeatin 10 μM treatments exhibited lower ABS/RC values compared to the other two hormone treatments under all light conditions except green and red LEDs. Within the Zeatin 10 μM group, green and red LEDs resulted in relatively higher ABS/RC values. DIo/RC followed a pattern similar to that of ABS/RC, with Zeatin 10 μM showing lower values under most light conditions compared to BA 5 μM and hormone-free (HF) treatments, except under green and red LEDs. The highest DIo/RC value within the Zeatin 10 μM group was observed under red LED.

Fig. 6.

Chlorophyll fluorescence parameters of A. incana explants under combinations of PGR treatment (HF, 5 µM BA, 10 µM zeatin) and light spectrum (control, white, red, green, blue LED). Measured parameters include Fv/Fm, Pi_ABS, ABS/RC, ETo/RC, and DIo/RC. Bars represent means (N = 45), and error bars indicate standard errors. Asterisks denote significance levels (*p < 0.05; **p < 0.01; ***p < 0.001).

TRo/RC was also lower in the Zeatin 10 μM treatment than in BA 5 μM and HF under all light conditions except green and red LEDs. Under blue LED, TRo/RC in the Zeatin 10 μM treatment was lower than HF and showed no significant difference compared to BA 5 μM. Within the Zeatin 10 μM group, green and red LEDs yielded higher TRo/RC values, consistent with the patterns observed in ABS/RC. In contrast, ETo/RC values were consistently lower in the BA 5 μM treatment than in the other hormone treatments under all LED conditions. Within the BA 5 μM group, the highest ETo/RC was observed under red LED, a pattern also found in the Zeatin 10 μM group.

Fv/Fm and Pi_ABS exhibited comparable trends. Zeatin 10 μM treatment showed higher Fv/Fm values than BA 5 μM and HF under all light conditions except green and red LEDs. Pi_ABS showed the same trend, with Zeatin 10 μM exhibiting significantly higher values, especially under white and blue LEDs. Interestingly, the HF group also exhibited a relatively high Pi_ABS value under green LED. As for Fm and Fo, the combination of Zeatin 10 μM with green LED resulted in the highest values among all hormone × light treatment groups.

Correlation analysis among physiological, growth, and biochemical parameters

Correlation analysis revealed that photosynthetic efficiency and antioxidant activity were generally negatively correlated, while specific diarylheptanoids—particularly oregonin and hirsutenone—showed positive correlations with antioxidant capacity and certain growth traits under optimized conditions(Fig. 7).

Fig. 7.

Pearson correlation heatmap of growth, photosynthetic, biochemical, and antioxidant variables in A. incana explants across treatments. (A) HF, (B) Zeatin 10 μM, (C) BA 5 μM, (D) Control, (E) White, (F) Green, (G) Red, (H) Blue. Statistical significance is denoted as follows: p ≤ 0.05 (*), p ≤ 0.01 (**), and p ≤ 0.001 (***). Data are based on three biological replicates (N = 3).

In HF and BA-treated groups, ABS/RC and TRo/RC negatively correlated with DPPH (r ≈ − 0.66 to − 0.70, p < 0.01), whereas Fv/Fm was positively associated with antioxidant activity (r ≈ 0.67, p < 0.01). Zeatin 10 μM treatment consistently enhanced correlations between shoot length and metabolite content (r > 0.57, p < 0.05). Under white and blue LEDs, ETo/RC, Fm, and Pi_ABS showed strong positive associations with oregonin and hirsutenone (r = 0.70–0.87, p < 0.05). Green LED enhanced oregonin accumulation (r = 0.91, p < 0.001), while red LED was more linked to shoot induction.

Notably, oregonin displayed consistent negative correlations with ABTS and DPPH across light types (r < − 0.70, p < 0.05), reinforcing its role in antioxidant responses. These patterns highlight the synergy between LED spectrum and hormone treatment in enhancing bioactive metabolite production in vitro.

Principal component analysis (PCA)

Principal Component Analysis (PCA) was performed to assess the effects of LED light conditions (Control, White, Green, Red, Blue) and plant growth regulator (PGR) treatments (BA 5 μM, HF, Zeatin 10 μM) on plant growth and physiological characteristics (Fig. 8). In the HF group, a strong dependency on LED treatment conditions was observed (Fig. 8A). White and Red LEDs, which were primarily influenced by secondary metabolite accumulation, formed a similar cluster, clearly distinguished from Control and Green LED groups. Red LED, in particular, was associated with growth parameters.

Fig. 8.

Principal component analysis (PCA) biplot showing clustering of treatment groups based on integrated growth, metabolite, and antioxidant traits. Arrows indicate variable loadings. Treatments are grouped by cytokinin type (BA, zeatin, HF) and LED spectrum. (A) HF, (B) Zeatin 10 µM, (C) BA 5 µM, (D) Control LED, (E) White LED, (F) Green LED, (G) Red LED, (H) Blue LED. Data are based on three biological replicates (N = 3).

With Zeatin 10 μM treatment, a distinct clustering pattern emerged due to the combined effects of PGR and LED conditions (Fig. 8B). White LED was primarily influenced by enhanced growth parameters and secondary metabolite accumulation, whereas Red LED was associated with both secondary metabolite accumulation and chlorophyll fluorescence responses.

In contrast, in the BA 5 μM treatment, Blue LED had the most distinct influence on secondary metabolite accumulation, forming a separate cluster from other LED conditions (Fig. 8C). However, unlike HF White LED, BA 5 μM resulted in the lowest metabolite accumulation, which likely contributed to its classification. Under the Control LED condition, BA exhibited greater variability in response compared to HF and Zeatin 10 μM (Fig. 8D). Zeatin 10 μM formed a clearly distinct cluster from HF and BA 5 μM.

Across LED light conditions, various parameters exhibited complex interactions, with PCA revealing a clear separation among the three PGR treatments (Fig. 8E–H). These results indicate that the combination of hormone treatments and LED light conditions exerts a multifaceted influence on growth, metabolite production, and physiological responses.

Discussion

Comparison to wild material

This study aimed to establish optimized culture and light conditions for the efficient production of diarylheptanoid-type secondary metabolites—specifically oregonin, hirsutenone, and hirsutanonol—through plant tissue culture. While various prior studies have investigated the application of plant growth regulators to callus or suspension cultures for secondary metabolite production in different plant species^37–39, there is a notable lack of research focusing on shoot or root explants. Furthermore, studies specifically targeting phenolic compounds in the diarylheptanoid class remain limited.

Although oregonin and other diarylheptanoids have been previously extracted from naturally growing Alnus species in forest ecosystems, to date, no research has demonstrated the continuous and optimized production of these compounds using in vitro cultured A. incana. Traditional methods for acquiring oregonin involve harvesting branches and bark from mature trees in natural habitats³⁹, a process that not only risks ecological disturbance but also presents challenges in ensuring a stable and scalable raw material supply. In contrast, plant tissue culture enables the aseptic propagation of plant tissues in controlled environments, allowing for sustainable and reproducible production without the need to fell trees. This method eliminates the long growth periods associated with field cultivation and offers significant advantages in terms of time, yield, and resource efficiency.

According to previous studies, the oregonin content obtained via natural extraction from wild-harvested Alnus species ranged from 123.09 ± 5.21 to 497.37 ± 10.49 µg/mL⁸. In the present study, in vitro-cultured explants treated with 10 μM zeatin + white LED consistently yielded oregonin concentrations of 108.089 ± 0.538 µg/mL. Although slightly lower than the highest values observed in wild-harvested materials, these concentrations are within the reported natural range, suggesting that tissue culture-based production may achieve comparable metabolite yields. These findings indicate that in vitro production offers an environmentally sustainable and industrially viable alternative to traditional harvesting methods.

Mechanisms

High concentrations (15 μM) of both BA and zeatin were found to reduce growth-promoting effects in this study. This suggests that excessive levels of cytokinins may have inhibitory effects on plant growth, which may emphasize the importance of applying optimal concentrations. The growth suppression observed at higher cytokinin levels may be related to elevated ethylene production, which has been reported to inhibit shoot elongation or interfere with organogenic redifferentiation processes, as noted in previous studies⁴⁰.

In the case of zeatin, a concentration-dependent increase in oregonin accumulation was observed, whereas BA treatments generally resulted in lower levels of oregonin. These findings align with previous reports suggesting that zeatin may contribute to the biosynthesis of secondary metabolites. For instance, reported that zeatin contributed not only to improved plant growth but also to enhanced levels of antioxidants and secondary metabolites in cadmium-stressed wheat plants⁴¹. Our findings that zeatin was associated with enhanced accumulation of diarylheptanoids under specific LED conditions are also consistent with recent reviews emphasizing that cytokinins modulate not only cell division but also secondary metabolite biosynthetic pathways through transcriptional regulation⁴².

Conversely, while the 5 μM BA treatment yielded the most robust growth response, only oregonin was detected in the BA-treated explants. This suggests that BA appears to primarily as a growth-promoting cytokinin, with limited influence on secondary metabolite production. Such a distinction may be attributed to the specific mode of action of BA, which is known to primarily stimulate cell division and shoot proliferation, rather than metabolic regulation, thereby distinguishing its effects from those of zeatin.

In this study, GA₃ treatment resulted in the lowest growth, antioxidant activity, and secondary metabolite accumulation in A. incana cultures. Although GA₃ is generally associated with promoting stem elongation, its physiological effects vary widely depending on plant species, concentration, and cultivation conditions. While some in situ studies, such as in Vitis vinifera, reported increased flavonoid and anthocyanin levels following GA₃ application⁴³, others observed reduced phenolic content and inhibited growth⁴⁴. These discrepancies underscore the necessity of species- and condition-dependent assessments when evaluating phytohormonal effects. Our findings suggest that GA₃ appears to exert a suppressive effect in in vitro A. incana, underscoring the importance of optimizing hormone selection and concentration for efficient production of bioactive compounds in tissue culture systems.

The highest antioxidant activity was observed in the combination of blue LED irradiation and zeatin 10 μM treatment. Previous studies have reported that blue light induces the generation of reactive oxygen species (ROS), which lead to increased oxidative stress within plant cells. In response, plants activate their antioxidant defense systems, including the production of various secondary metabolites with ROS-scavenging properties⁴⁵. This mechanism may explain how blue light contributes to the enhancement of antioxidant activity by stimulating secondary metabolite biosynthesis⁴⁶.

Furthermore, the observed differences in metabolite accumulation under blue and white LED light are supported by studies demonstrating that light quality strongly influences plant morphology and secondary metabolite production. For instance, Trivellini⁴⁷ review the roles of blue and red light in orchestrating secondary metabolites and nutrient transport, while Zhang⁴⁸ summarize how light manipulation affects secondary metabolite biosynthesis via photosynthetic regulation.

In Rehmannia glutinosa cultures, blue LED light was found to enhance ROS scavenging capacity and promote the accumulation of phenolic compounds⁴⁹. Similarly, blue light has been shown to activate the phenylpropanoid pathway, leading to increased synthesis of phenolics and flavonoids^46,49. In Scutellaria lateriflora cultures, blue LED treatment enhanced the accumulation of specific phenolic compounds such as baicalin, wogonoside, and verbascoside, further supporting the role of blue light in stimulating antioxidant compound production⁵⁰.

In parallel, zeatin treatment resulted in the highest antioxidant capacity among the plant growth regulators tested, as evidenced by both DPPH and ABTS assays. Zeatin, as a representative cytokinin, contributes to ROS homeostasis by enhancing antioxidant defense pathways, supporting the maintenance of cellular redox balance⁵¹. This occurs partly through the accumulation of compatible solutes and the activation of non-enzymatic ROS detoxification mechanisms, which help mitigate oxidative stress⁵².

These observations suggest that zeatin may play a dual role—both as a growth regulator and as a modulator of oxidative stress responses—potentially by activating biosynthetic pathways involved in the production of antioxidant secondary metabolites such as oregonin. Further studies are warranted to elucidate the underlying biosynthetic mechanisms and confirm the link between zeatin signaling, ROS suppression, and oregonin biosynthesis.

In the zeatin-treated group, all three diarylheptanoid compounds—oregonin, hirsutanonol, and hirsutenone—were consistently detected, whereas in the BA treatment group, only oregonin was reliably present. Hirsutanonol and hirsutenone were either absent or exhibited inconsistent accumulation patterns, a phenomenon rarely addressed in previous studies on phenolic compound profiling^53,54. These findings suggest that the biosynthesis and metabolic regulation of diarylheptanoid compounds can be significantly influenced by the specific type of plant growth regulator applied, as also implied by hormone-dependent regulation patterns reported in earlier research^55,56.

Zeatin is known to activate the phenylpropanoid pathway, which is closely associated with the biosynthesis of oregonin and its derivatives. This aligns with the results of the present study, where zeatin treatment led to the stable detection of all three target compounds. Previous research has similarly demonstrated that zeatin can enhance the production of phenolic secondary metabolites by stimulating key enzymatic steps in the phenylpropanoid biosynthetic^57–59.

Cytokinins such as zeatin modulate phenylpropanoid biosynthesis by regulating phenylalanine ammonia-lyase (PAL) activity and maintaining ROS homeostasis, thereby enhancing the accumulation of phenolic compounds and lignin in plant tissues⁶⁰. Moreover, light quality directly affects this pathway; blue LED irradiation activates ROS-mediated signaling, leading to the upregulation of key structural genes in the phenylpropanoid pathway and enhanced secondary metabolite production, whereas red LED often exhibits a suppressive effect⁵⁸.

Correlation and PCA-based observations

The 10 μM zeatin treatment, which produced the highest T-score for overall growth performance, was generally associated with reduced physiological stress. However, chlorophyll fluorescence analysis revealed that under green and red LED lighting, plants experienced higher levels of light-induced stress⁶¹. Notably, stress levels under red LED were higher than those under blue LED, which aligns with previous findings in Melissa officinalis L.⁶², where red light was found to reduce antioxidant enzyme activity and downregulate related genes. This pattern was also observed in our study. This is likely mediated by light-dependent regulation of photosynthetic efficiency and chlorophyll fluorescence dynamics, which in turn affect metabolic flux toward secondary pathways⁶³.

On the other hand, in the BA 5 μM treatment group, green LED resulted in superior outcomes for Number of shoots and Induced shoot length, as well as elevated oregonin and antioxidant levels compared to blue LED. This may be attributed to the fact that green light can penetrate deeper into leaf tissue, potentially enhancing photosynthetic efficiency⁶¹. Supporting this, a study on Brassica oleracea L. var. italica (broccoli) also reported that green LED increased both chlorophyll content and phenolic compound accumulation⁶⁴. In our study, green LED treatment had a favorable effect on specific growth indicators, suggesting that green light may confer physiological advantages under particular conditions.

Synthesizing these findings, it becomes evident that the interaction between light quality and plant growth regulators—rather than light quality alone—is a more decisive factor in modulating antioxidant accumulation. This also implies that the responses observed in A. incana may not be universally applicable to other species. Future research should aim to elucidate the molecular mechanisms underlying light spectrum-specific stress responses and investigate gene-level regulation to better understand how the interplay between light and hormones influences secondary metabolite biosynthesis.

In this study, the relationships among photosynthetic parameters, antioxidant activity, and diarylheptanoid-type secondary metabolites (oregonin, hirsutenone, and hirsutanonol) were analyzed under various plant growth regulator treatments (HF, zeatin 10 μM, and BA 5 μM) and LED light conditions (Control, White, Green, Red, and Blue). The results revealed distinct correlations between photosynthetic efficiency, antioxidant activity, and secondary metabolite accumulation depending on treatment conditions. A negative correlation between photosynthetic efficiency and antioxidant activity was observed in the HF group (ABS/RC), and in both zeatin 10 μM and BA 5 μM groups (ABS/RC, DIo/RC, TRo/RC), as well as under control (Fv/Fm) and red LED (Fm, Fo) conditions. These findings suggest that increased photosynthetic efficiency may be associated with enhanced activation of antioxidant defense mechanisms to mitigate ROS accumulation. Specific LED light qualities—particularly white, green, and blue—were found to promote oregonin biosynthesis, with white and green LEDs contributing to a more balanced relationship between diarylheptanoid accumulation and growth parameters⁶⁵. In contrast, red LED treatment tended to enhance diarylheptanoid production but appeared to suppress overall growth. Notably, in the zeatin and BA treatment groups, increased photosynthetic efficiency was associated with reduced oregonin accumulation, indicating a possible interaction between cytokinin-induced metabolic regulation and photosynthetic activity. This interaction may be associated with a metabolic trade-off or hormonal modulation of carbon flux toward growth versus secondary metabolism.

Limitations & future work

This study demonstrated that the combination of specific plant growth regulators and LED light spectra can significantly influence in vitro growth, antioxidant capacity, and diarylheptanoid accumulation in A. incana. Significantly, Zeatin 10 μM combined with blue LED treatment maximized antioxidant activity and secondary metabolite production, while Oregonin and Hirsutenone showed strong correlations with antioxidant responses. These findings suggest there may be a resource allocation trade-off between photosynthetic efficiency and secondary metabolism, highlighting the importance of selecting optimal culture conditions.

While this study elucidated the physiological and biochemical responses of A. incana subsp. hirsuta to different plant growth regulators and LED light spectra, it did not investigate the underlying molecular mechanisms. In particular, the transcriptional regulation and pathway-specific gene expression involved in diarylheptanoid biosynthesis remain unexplored. Future studies incorporating transcriptomic or targeted gene expression analyses will be essential to clarify the regulatory networks linking cytokinin signaling, light spectrum responses, and the biosynthesis of oregonin, hirsutenone, and hirsutanonol. Such molecular-level insights will facilitate the precise optimization of in vitro culture systems for functional metabolite production.

In addition, the present experiments were conducted with a relatively small sample size (n = 3), which may limit the statistical power and generalizability of the findings. Furthermore, the optimized culture and light protocols have only been tested under laboratory-scale conditions. Scaling up these protocols to pilot- or industrial-scale bioreactor systems will be essential to validate their feasibility for commercial applications, ensure production stability, and assess cost-effectiveness under non-laboratory environments.

These findings provide valuable insights into the optimization of light conditions and hormonal treatments for maximizing diarylheptanoid production in in vitro cultures. The results may serve as a foundational reference for future strategies aimed at enhancing the biosynthesis of functional secondary metabolites through tissue culture approaches. Compared to previous studies focusing solely on light spectrum effects on plant growth or antioxidant capacity, this study demonstrates a more integrated approach by simultaneously evaluating plant growth, secondary metabolite accumulation, and photosynthetic responses under various combinations of plant growth regulators (PGRs) and LED spectra. Notably, the comprehensive use of PCA-based efficiency indices and T-score analysis enabled a multidimensional evaluation of treatment effectiveness, which is rarely addressed in similar in vitro studies. The findings highlight the potential of zeatin combined with white or blue LED light to maximize both growth and bioactive compound production in A. incana. These results not only expand the understanding of hormone-light interactions but also offer a practical basis for optimizing in vitro propagation systems for functional metabolite production.

Conclusion

This study demonstrated that the combined manipulation of cytokinins and LED light spectra significantly influences in vitro growth, antioxidant activity, and diarylheptanoid production in Alnus incana. BA primarily promoted shoot proliferation and biomass accumulation, whereas zeatin enhanced secondary metabolite biosynthesis, particularly oregonin and hirsutenone, and improved antioxidant capacity. The integration of zeatin treatment with specific LED spectra further optimized both growth and metabolite production, highlighting the importance of balancing hormonal and environmental cues. These findings provide a scalable and sustainable strategy for producing functional secondary metabolites and may serve as a model for propagation systems in other woody medicinal species. Future studies incorporating transcriptomic validation and pilot-scale bioreactor testing will be essential to confirm feasibility and ensure industrial applicability. Overall, the optimized protocol highlights a novel and sustainable approach to metabolite production in woody species, addressing both conservation and industrial needs.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SK, EJC and SC: Conceived and designed the study. SK and HL: Performed the experiments and collected the data. SK and HL: Conducted antioxidant activity assays. SK, UJ and TJ: Performed statistical analysis and prepared the original draft. EJC and SC: Revised and finalized the manuscript. All authors have read and approved the final manuscript. Correspondence should be addressed to EJC and SC.

Funding

This work was supported by the R&D Program for Forest Science Technology (Project No. 2023469A00-2325-EE01), provided by the Korea Forest Service and the Korea Forestry Promotion Institute.

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.

Research involving plants

The present study complied with the international and national guidelines for the use of experimental plant material.