De Novo Regeneration of Cannabis sativa cv. Cheungsam and Evaluation of Secondary Metabolites of Its Callus

Abstract

Cannabis sativa L. cv. ‘Cheungsam’ is an industrial hemp plant of Republic of Korea origin, primarily cultivated for fiber and seed production. In vitro seed germination and tissue culture are valuable tools for developing various biotechnological techniques. In the present study, we aimed to develop a tissue culture process for hemp plants using Cheungsam as a model plant and examine the secondary metabolites produced from its callus. We also developed a method to prepare pathogen-free seedlings from field-derived seeds using hydrogen peroxide (H₂O₂) solution as a liquid germination medium. Treating seedlings with removed seed coat in 3% H₂O₂ significantly reduced the contamination rate. Callus formation and de novo organogenesis of shoots and roots from callus were successfully achieved using cotyledon and leaf tissues prepared from the pathogen-free seedlings. The most effective in vitro regeneration results were obtained using the Murashige and Skoog (MS) medium supplemented with certain targeted growth regulators. An optimal combination of 0.5 mg/L thidiazuron (TDZ) and 1.0 mg/L 1-naphthalene acetic acid proved highly effective for callus induction. The addition of 0.5 mg/L TDZ in the MS medium significantly stimulated shoot proliferation, while robust root development was best supported by MS medium supplemented with 2.5 mg/L indole-3-butyric acid for both cotyledon and leaf explants. Finally, gas chromatography–mass spectrometry (GC–MS) analysis of ethanol extract from Cheungsam leaf callus revealed the presence of different secondary metabolites, including 9-octadecenamide, methyl salicylate, dodecane, tetradecane, and phenol, 2,4-bis-(1,1-dimethylethyl). This study provides a comprehensive de novo regeneration protocol for Cheungsam plants and insight into the secondary metabolite profiles of its callus.

1. Introduction

Cannabis (e.g., Cannabis sativa L.) is one of the world’s oldest cultivated plants, valued for its wide range of applications. Typically, it is classified and recognized as either industrial hemp or marijuana, a distinction based primarily on the concentration of Δ9-tetrahydrocannabinol (THC) contents, the principal psychoactive compound [1]. Cannabis plants are gaining increasing attention due to their medicinal effects (anti-cancer, anti-inflammatory, anti-spastic, anti-pruritic, anticonvulsant, antimicrobial, immunomodulatory, neuro-protective, and anti-psychotic effects), which are derived from various secondary metabolites, such as phytocannabinoids (THC, CBD, CBN, CBC, CBG, THCV, CBDV, etc.), terpenoids, flavonoids, fatty acids, and more [1,2,3,4,5,6,7,8,9]. In addition to its medicinal value, Cannabis plants are also widely cultivated to produce various industrial products, such as fiber, oil seed, animal feed, biodegradable plastics, construction materials, biofuel, and textiles [1,2,3,4,5,6,7,8]. Thus, Cannabis plants provide numerous benefits across different industries, making them an important crop worldwide.

Explant sources, especially juvenile tissues from in vitro-grown seedlings, play a fundamental role in in vitro tissue culture. Under natural conditions, endophytic microbes within seeds provide plants with enhanced nutrient availability, beneficial metabolites, and enzymes that protect them from different stress factors [3,10,11,12,13,14,15,16,17,18,19,20]. However, endophytic microbes can also create serious contamination problems under in vitro culture conditions [3,10,11,12,13,14,15,16,17,18,19,20]. Notably, the generation of pathogen-free seedlings in vitro from seeds having a thick seed coat, such as Cannabis, is highly challenging, and surface sterilization alone is not usually enough [3,10,11,12,13,14,15,16,17,18,19,20]. Recently, the dual roles of H₂O₂ as a disinfectant and a germination-promoting signaling molecule have led to remarkable success in generating pathogen-free seedlings in vitro [10]. However, varying concentrations of H₂O₂ showed conflicting roles in the survival of seeds; thus, its use (or concentration) must be critically managed and optimized in relation to contamination frequency, germination rate, and seed viability [10].

Due to the totipotency of plants, plant tissues cultured in vitro can be regenerated into complete, fertile, and morphologically and physiologically true-to-type plants from explants derived from various tissues under appropriate cultivation conditions [21,22,23,24]. This process is called de novo shoot/root organogenesis (DNSO/DNRO). DNSO/DNRO can be induced in vitro by targeted manipulation of cytokinin (CK) and auxin signaling. By modifying the CK or auxin signaling pathways, callogenesis, shooting, and root regeneration can be considerably enhanced, even in recalcitrant tissues [22,23,24]. Plant tissue culture offers a controlled environment for the propagation of plants and serves as a basis for the genetic modification of plants. This technique allows for the mass production of disease-free, genetically uniform plants, therefore extensively used in research, agriculture, horticulture, and plant secondary metabolite production [22,23,24]. However, the tissue culture of Cannabis plants poses unique challenges due to their high recalcitrance, with significant variation among explant types and varieties creating a bottleneck for the application of in vitro tissue culture to the improvement of C. sativa [22,23,24]. Thus, a variety-specific optimization of tissue culture protocols is required to successfully regenerate specific genotypes (or cultivars) of Cannabis plants.

Beyond its pivotal role as a propagation method, plant tissue culture has garnered substantial industrial interest for its broader application in plant improvement and the large-scale production of valuable secondary metabolites [25]. Tissue culture techniques have become invaluable not only for the mass propagation of plants and the production of plant-based products at an industrial scale through different bioreactor systems and genetic manipulation but also as a tool for advancing core insights in biochemistry, plant physiology, embryology, cytology, and molecular biology [25]. Recently, plant biotechnology has increasingly utilized callus culture as a prominent approach for the in vitro synthesis of secondary metabolites and large-scale bioactive compounds compared with other existing techniques, primarily due to the efficiency of in vitro callus induction as a direct and rapid cell multiplication system [26,27]. Plant cells can be readily cultivated to produce secondary metabolites sustainably under controlled conditions, allowing their direct extraction from the callus without the need to sacrifice the whole plant [26,27].

Metabolomics is a powerful analytical technique for identifying a vast array of metabolites at specific time points, and it has become widely applicable in food and nutritional science research [28]. Among these techniques, GC–MS stands out as a rapid, reliable, and highly influential analytical method, particularly suited for analyzing volatile and thermally stable compounds and primary metabolites, such as alcohols, sugars, organic acids, amino acids, volatile oil, esters, fatty acids, amino acids, and nitro compounds. It is also instrumental in tailoring biosynthetic pathway precursors across various plant species, including Cannabis sativa [25,28,29,30,31,32,33,34]. Consequently, it has been used to describe the response of carbon (C) and nitrogen (N) primary metabolism to major environmental cues, for example, herbivores, CO₂ mole fraction, drought, nutrient conditions, or stress/elicitor combinations [35]. Recently, numerous non-targeted metabolome studies employing GC–MS [36] have been conducted across diverse plant species, demonstrating the technique’s capability to identify metabolite-level variations in various physiological and metabolic processes [28].

Cheungsam is an industrial hemp cultivar widely used in South Korea and recently studied for its richness in beneficial metabolites [2,4,5,6]. Several reports suggest that its extracts exhibit antioxidant, anti-inflammatory, anti-obesity, and anti-diabetic activities, thereby offering significant potential as a raw material in the food and medicinal industries [2,4,5,6]. Recently, one study established a micropropagation protocol for the Cheungsam variety, enabling the direct production of normal plantlets from in vitro-grown seedlings [37]. However, an optimized protocol for callogenesis and in vitro organogenesis from the Cheungsam plant is not yet available. Thus, in this study, we aimed to: (i) develop optimized conditions for callus formation from leaf and cotyledon explants, (ii) achieve in vitro regeneration from callus, and (iii) investigate the metabolite profile of the callus.

2. Materials and Methods

2.1. Optimization of Sterilization Methods for the Generation of Pathogen-Free Seedlings

A large-seeded variety of Cannabis sativa L., named Cheungsam, was collected from the field and used for this study. Seeds were surface sterilized with 70% ethanol for 3 min and 10% sodium hypochlorite containing 0.1% Tween 20 for 10 min and then washed with sterile distilled water until all bubbles were completely removed [9]. Surface-sterilized seeds were directly submerged in conical tubes containing different volumes (10, 15, 20, and 25 mL) of liquid germination medium (LGM: 1% H₂O₂). Sterile distilled water was used as a negative control in the LGM experiments. Fresh LGM was replaced twice daily, with the old LGM removed each time, for 4 days. Experiments were performed with three replicates per treatment group, and 25 seeds were used for each replicate. H₂O₂-treated seeds were stored at room temperature under dark conditions. The germination rate for each treatment was assessed by observing radicle emergence, while the survival rate was evaluated based on the development of light-green cotyledons.

As LGM alone proved insufficient to prevent contamination of in vitro seedlings, further sterilization was applied to the seedlings after their seed coats were removed. Germinated seedlings in LGM were detached from seed coats using sterilized forceps and submerged in 1% or 3% H₂O₂ solutions. For 1% H₂O₂ treatment, seedlings with removed seed coats were submerged for 1 day, with the solution changed twice daily. For 3% H₂O₂ treatment, seedlings with removed seed coat were submerged for different durations (1, 2, 3, and 4 h). Sterilized seedlings were washed with 1% sodium hypochlorite (1 min), 70% ethanol (1 min), and sterile distilled water (3 times) and then transferred to Murashige and Skoog (MS) agar medium to promote further growth [37]. Seedlings grown on MS medium were transferred to a growth chamber set at 24 °C with an 18 h light/6 h dark cycle and a light intensity of 200 µmol·m⁻²·s⁻¹ for two weeks to assess contamination and survival rates. To prepare 1 L of MS medium, 4.43 g of MS salts (including vitamins) and 20 g of sucrose were dissolved in 1 L of distilled water. The pH of the medium was adjusted to 5.7 using KOH, followed by the addition of 8 g of agar before autoclaving [10]. The contamination rate was calculated by dividing the number of contaminated seedlings by the total number of seedlings and expressing the result as a percentage.

2.2. Callus, Shoot, and Root Induction from Cotyledon and Leaf Explants

Plant hormones and chemicals used in tissue culture study were purchased from Duchefa Biochemia (Haarlem, The Netherlands). To identify the optimal conditions for inducing callus formation from cotyledon and leaf explant, callus induction medium (CIM) containing full-strength MS medium with different concentrations of thiadizuron (TDZ; 0 and 1 mg/L) and naphthalene-acetic acid (NAA; 0, 0.5, 1.0, 1.5, and 2.0 mg/L) were tested. Each explant was precisely sectioned into 1 cm² using a sterile scalpel (Duchefa Biochemie, Haarlem, The Netherlands) and was obtained from the central axis along the mid-rib tissues, excluding petioles or apex areas from leaves and cotyledons. For optimal callus formation, the abaxial side of one explant was gently placed onto the CIM and carefully tapped to maximize surface contact, reducing any unintended movement during the handling process. Experiments were performed with three replicates, and each replicate included 10 explants of leaf or cotyledon tissues. The capped jars were securely double-wrapped with sealing tape (Kisan Bio, Seoul, Republic of Korea) and randomly arranged within a controlled growth chamber set to 24 °C with an 18 h light/6 h dark cycle and a light intensity of 200 µmol·m⁻² s⁻¹. Leaf and cotyledon explants were cultivated under these precisely regulated conditions to support callus induction over 8 and 3 weeks, respectively, with weekly evaluations conducted to monitor callus development. To avoid nutrient and hormone depletion, each explant and callus was transferred to fresh CIM with the same concentrations of hormones every week. After 8 weeks of culture on the CIM, the callus mass (g) and response percentage were recorded.

Callus derived from leaves and cotyledons were transferred to the shoot induction medium (SIM) after 4 and 3 weeks of incubation on the CIM, respectively. To determine the optimal SIM, full-strength MS medium was supplemented with varying concentrations of TDZ (0, 0.5, 1.0, 2.5, 5.0, and 10.0 mg/L) and evaluated. Eight-week-old calli previously cultured on the CIM were transferred into plastic containers containing 50 mL of SIM and kept under the same conditions as previously described. Calli cultures were maintained on SIM until shoots reached a height of over 2.5 cm; at this point, they were transferred to the root induction medium (RIM). Non-responsive calli, including those that had ceased growing or died, were discarded after a four-month cultivation period. Finally, to assess the de novo root organogenesis, RIM composed of full-strength MS medium was prepared with varying concentrations of IBA (0, 0.5, 1.0, 2.5, 5.0, and 10.0 mg/L). De Novo organogenesis of roots was induced in plastic jars containing 50 mL of RIM and maintained under the same conditions described above.

2.3. GC–MS Measurement of Secondary Metabolites in Callus Cultures

One-month-aged calli were transferred on fresh medium and subcultured every 3 weeks. Then, callus was separated from the medium. A 0.1 g sample of crushed callus was dried in an oven at 150 °C for 30 min to extract secondary metabolites. The dried callus was ground using a Silamat S6 device (Ivolar Vivadent Inc., NY, USA) and glass beads, mixed with GC-grade EtOH (10 mL EtOH added per 0.1 g sample), thoroughly mixed by vortexing for 1 min, and incubated at room temperature for 1 h under vigorous shaking. Undissolved plant debris for each sample was removed by centrifugation at 12,000× g for 5 min. The supernatant of each sample was transferred to a new tube, filtered with a syringe filter (pore size of 0.2 µm, Pall Corporation, NY, USA), and used for the GC–MS analysis.

A GC/MSD system (5977A Series, Agilent Technologies, Santa Clara, CA, USA) and DB-5 ms columns (Agilent Technologies, Santa Clara, CA, USA) were used for the GC–MS analysis. The oven temperature was held at 80 °C for 1 min and then increased to 240 °C at a rate of 10 °C·min⁻¹, from 240 °C to 260 °C at a rate of 5 °C·min⁻¹, and from 260 °C to 300 °C at a rate of 20 °C·min⁻¹, and was finally held at 300 °C for 10 min. The injector and mass interface temperature was 300 °C. Helium was used as a carrier gas at a flow rate of 1 mL ·min⁻¹. The injection mode was split, and the scanned mass range was from 40 to 450 m/z. Two µL of each sample was injected for the analysis. The result was analyzed using the National Institute of Standards and Technology version 11 (NIST 11 spectral library).

2.4. Data Analysis

The average seed germination rate across various liquid germination media, contamination rate in the medium multiplied, and survival percentages on MS medium were determined and reported as means ± standard errors (SEs). Each experiment was performed in triplicate and provided consistent results, with 25 seeds in each replicate. Statistical analysis (One-way ANOVA) was performed using Minitab (Version 21.3.1), with significance tested at p < 0.05 using the post-hoc Tukey Honestly Significant Difference (HSD) test. GraphPad Prism (version 8.1, GraphPad Software, San Diego, CA, USA) was used for graphical visualization.

3. Results

3.1. Preparation of Pathogen-Free Seedlings

In the previous study, various hemp plant seeds showed the highest germination rate when a 1% H₂O₂ solution was used as liquid germination medium (LGM) [10]. To examine the optimal volume of H₂O₂ solution for the seed germination of Cheungsam plants, the same numbers of seeds (25 seeds) were incubated with different volumes of LGM (10 mL water for control and 10, 15, 20, and 25 mL of 1% H₂O₂). As expected, seeds treated with a 1% H₂O₂ solution as LGM showed a higher germination rate than those germinated in water (Figure 1A). Interestingly, the seeds incubated with 15 mL H₂O₂ showed significantly higher germination rates than the other conditions. This suggests that using the optimal volume of 1% H₂O₂ solution as LGM can maximize the germination of Cannabis seeds.

Unexpectedly, we observed a significant contamination in the seedlings treated with a 1% H₂O₂ solution as LGM (Figure 1B). This could be due to the seeds used in this study having more severe endophytic contamination compared to the previous study [10]. To further reduce the contamination rate, we performed additional sterilization of seedlings after removing the seed coat using different volumes of 1% H₂O₂ for 24 h. Although additional sterilization of seedlings with a removed seed coat with 25 mL of 1% H₂O₂ reduced the contamination rate to 40.0 ± 2.36%, it was not enough to fully remove the endophytic contamination (Figure 1C). Thus, we used a higher concentration of H₂O₂ for seedling sterilization for different time periods. Notably, sterilization of seedlings with a removed seed coat with 25 mL of 3% H₂O₂ further reduced the contamination rate of seedlings in a sterilization duration time-dependent manner (Figure 1D). Sterilization of seedlings with removed seed coat with 25 mL of 3% H₂O₂ for 4 h successfully reduced the contamination rate to 0.00%. However, as shown in Figure 1E, prolonged 3% H₂O₂ treatment induced seedling growth retardation. This suggests that the additional sterilization of seedlings with a removed seed coat using 3% H₂O₂ can be applied to generate pathogen-free seedlings from heavily contaminated seeds; however, the protocol needs to be optimized considering the seed contamination rate under the premise of normal growth conditions.

3.2. In Vitro Callogenesis from Leaf and Cotyledon Explants

To generate callus, excised leaf and cotyledon tissues were placed on the surface of the callus induction medium (CIM) containing a fixed concentration of TDZ (1.0 mg/L) and varying concentrations of NAA to determine the optimal auxin–cytokinin ratio for callus induction. Among the different combinations of TDZ and NAA, both leaf and cotyledon showed the highest callogenesis response on the CIM containing 1.0 mg/L TDZ and 0.5 mg/L NAA (

Table 1. Effect of different concentrations of TDZ and NAA on callogenesis from leaf and cotyledon explants of Cheungsam.

Tissue Used	TDZ (mg/L)	NAA (mg/L)	Callogenesis Response (%)	Amount of Callus (g)
Leaf	0.0	0.0	0.00 ± 0.00 d	0.00 ± 0.00 d
	1.0	0.5	92.80 ± 1.46 a	6.87 ± 0.12 a
	1.0	1.0	80.03 ± 0.90 b	4.90 ± 0.61 b
	1.0	1.5	73.67 ± 1.18 c	3.23 ± 0.30 c
	1.0	2.0	66.40 ± 5.73 c	1.73 ± 0.32 c
Cotyledon	0.0	0.0	0.00 ± 0.00 d	0.00 ± 0.00 c
	1.0	0.5	88.37 ± 1.77 a	4.87 ± 0.45 a
	1.0	1.0	78.67 ± 1.45 b	2.93 ± 0.30 b
	1.0	1.5	71.43 ± 4.77 b	1.67 ± 0.44 b
	1.0	2.0	57.93 ± 4.97 c	1.50 ± 0.12 b

TDZ: thidiazuron; NAA: naphthalene-acetic acid. Data were obtained 4 weeks after the incubation of leaf explants on CIM with the indicated concentrations of TDZ and NAA. Different letters indicate statistically significant differences at p < 0.05 according to the Tukey HSD test.

). As expected, no callogenesis was observed from either leaf or cotyledon explants when hormone-free MS medium was used as CIM. Leaf tissue excided from the contamination-free seedlings started to show initial callus formation 14 days after placing it on CIM (Figure 2A,B). Four weeks after incubation, the callus formed a green nodular photosynthetic and friable callus (Figure 2C,D). Leaf explants generated the highest amount of callus (6.87 ± 0.12 g) and showed the highest callogenesis response (92.80 ± 1.46%) on CIM containing 1.0 mg/L TDZ and 0.5 mg/L NAA (Table 1). Increasing concentrations of NAA (≥1.0 mg/L) negatively affected both the callogenesis response and the amount of callus generated. For example, leaf explants on CIM containing 1.0 mg/L TDZ and 2.0 mg/L NAA showed a 74.8% reduction in the amount of callus generated and a 28.4% reduction in callogenesis response compared to those on CIM containing 1.0 mg/L TDZ and 0.5 mg/L NAA.

Cotyledon tissues started to show initial callus formation 5–7 days after placing them on CIM (Figure 3A,B). Similarly, cotyledon explants on CIM containing 1 mg/L TDZ and 0.5 mg/L NAA generated the highest amount of callus (4.87 ± 0.45 g) and showed the highest callus induction rate (88.37 ± 1.77%) (Table 1). Cotyledon explants on CIM containing more NAA (≥1.0 mg/L) showed reduced callus and callogenesis responses compared to those on CIM containing 1.0 mg/L TDZ and 0.5 mg/L NAA. Notably, cotyledon explants generated green, nodular, photosynthetic, and friable callus faster than leaf explants (within 3 weeks; Figure 3B).

3.3. De Novo Shoot Morphogenesis from Callus

To induce de novo shoot morphogenesis from the callus, the fresh and green photosynthetic components of the callus masses were carefully excised with a sharp knife and placed on a shoot induction medium (SIM) containing different concentrations of TDZ. Among the SIMs with varying TDZ concentrations, the SIM with 0.5 mg/L TDZ elicited the highest de novo shoot morphogenesis responses from the calli derived from both leaf and cotyledon tissues (

Table 2. Effect of different concentrations of TDZ on de novo shoot morphogenesis from the calli derived from leaf or cotyledon explants of Cheungsam.

Origin of Callus	TDZ (mg/L)	Response (%)	Average Number of Shoots	Average Length of Shoots (cm)
Leaf	0.0	0.00 ± 0.00 e	0.00 ± 0.00 d	0.00 ± 0.00 e
	0.5	15.67 ± 0.27 a	11.60 ± 0.59 a	9.03 ± 0.39 a
	1.0	12.33 ± 0.98 b	7.90 ± 0.42 b	6.13 ± 0.84 b
	2.5	9.67 ± 0.98 c	7.47 ± 0.49 b	4.63 ± 0.44 b,c
	5.0	8.17 ± 0.49 c	4.33 ± 0.24 c	3.30 ± 0.26 c,d
	10.0	4.97 ± 0.80 d	2.80 ± 0.17 c	2.43 ± 0.29 d
Cotyledon	0.0	0.00 ± 0.00 f	0.00 ± 0.00 d	0.00 ± 0.00 d
	0.5	22.67 ± 1.19 a	8.33 ± 1.20 a	6.53 ± 0.48 a
	1.0	10.67 ± 0.98 b	5.67 ± 0.88 a,b	4.57 ± 0.50 a,b
	2.5	8.37 ± 0.62 c	3.33 ± 0.88 b,c	3.73 ± 0.39 b,c
	5.0	6.03 ± 0.78 d	2.83 ± 0.17 b,c	3.17 ± 0.69 b,c
	10.0	4.03 ± 0.42 e	2.00 ± 0.29 c	2.37 ± 0.19 c

Data were obtained 4 weeks after incubation of callus on SIM with indicated concentrations of TDZ. Different letters indicate statistically significant difference at p < 0.05 according to the Tukey HSD test.

). De Novo shoot morphogenesis from leaf-derived photosynthetic callus was first observed 2 weeks after incubation on SIM. Leaf explant-induced callus generated the highest shooting response (15.67 ± 0.27%) with the highest number of shoots (11.60 ± 0.59) and the most extended shoot length (9.03 ± 0.39 cm) after subculturing for another 2 weeks on fresh SIM containing 0.5 mg/L TDZ. Similarly, cotyledon-derived callus showed the highest de novo shoot morphogenesis response (22.67 ± 1.19%) on SIM with 0.5 mg/L TDZ (Table 2; Figure 3C,D). Increasing concentrations of TDZ above 0.5 mg/L reduced de novo shoot morphogenesis responses from both leaf- and cotyledon-derived calluses.

3.4. De Novo Root Morphogenesis from Callus-Derived Shoots

To induce de novo root morphogenesis, shoots with callus portions reaching a height of more than 5 cm were placed on the root induction medium (RIM) with different concentrations of IBA. Among the RIMs with varying IBA concentrations, the RIM with 2.5 mg/L IBA induced the highest de novo root morphogenesis responses from the callus-derived shoots of both leaf and cotyledon explant tissues (

Table 3. Effect of different concentrations of IBA on de novo root morphogenesis from the calli derived from leaf or cotyledon explants of Cheungsam.

Origin of Callus	IBA (mg/L)	Response (%)	Average Number of Roots	Average Length of Roots (cm)
Leaf	0.0	0.00 ± 0.00 d	0.00 ± 0.00 d	0.00 ± 0.00 c
	0.5	86.27 ± 0.44 b	5.70 ± 0.92 b,c	4.83 ± 0.32 b
	1.0	82.50 ± 0.64 b	6.70 ± 0.59 b,c	5.23 ± 0.32 b
	2.5	95.30 ± 1.14 a	12.77 ± 1.43 a	7.20 ± 0.64 a
	5.0	77.10 ± 0.99 c	8.70 ± 0.42 b	5.33 ± 0.60 a,b
	10.0	71.97 ± 0.42 c	4.57 ± 0.29 c	4.23 ± 0.12 b
Cotyledon	0.0	0.00 ± 0.00 d	0.00 ± 0.00 d	0.00 ± 0.00 d
	0.5	68.07 ± 2.87 c	4.47 ± 0.55 c	4.60 ± 0.95 a,b
	1.0	75.43 ± 1.53 b	7.03 ± 0.62 b	4.83 ± 0.49 a,b
	2.5	92.83 ± 2.31 a	9.70 ± 0.59 a	7.47 ± 0.71 a
	5.0	65.37 ± 1.98 c	4.70 ± 0.21 c	4.17 ± 1.05 b
	10.0	69.70 ± 4.14 c	3.70 ± 0.42 c	2.80 ± 0.17 b,c

IBA: indole-3-butyric acid. Data were obtained 4 weeks after the incubation of callus on RIM with the indicated concentrations of IBA. Different letters indicate statistically significant differences at p < 0.05 according to the Tukey HSD test.

; Figure 2G and Figure 3E). De Novo root morphogenesis from leaf-derived callus was first observed 2 weeks after incubation on RIM. After subculturing for another 2 weeks on fresh RIM with 2.5 mg/L IBA, the highest rooting response (95.30 ± 1.14%) was achieved, with the highest number of roots (12.77 ± 1.43) and the longest root length (7.20 ± 0.64 cm). Similarly, shoots from cotyledon-derived calli showed the highest de novo root morphogenesis response (92.83 ± 2.31%) on RIM with 2.5 mg/L IBA. Increasing concentrations of IBA above 2.5 mg/L significantly reduced de novo root morphogenesis responses from both leaf- and cotyledon-derived calluses.

3.5. Gas Chromatography–Mass Spectrometry (GC–MS) Analysis

For the analysis of secondary metabolites in the leaf-derived callus cultures of Cheungsam, ethanol was used as the extraction solvent. GC–MS analysis revealed distinct compound profiles, with nineteen unique peaks detected in the ethanol extract. The total ion chromatograph (TIC) corresponding to the ethanol-extracted compounds from the callus culture is shown in Figure 4. Among the compounds detected, the top five most abundant chemical compounds are listed in

Table 4. List of five of the most abundant chemical compounds found in leaf-derived callus cultures of Cheungsam by gas chromatography–mass spectrometry (GC–MS) analysis.

No.	Compound	Structure	Peak No.	Formula	Molecular Weight	Retention Time (min)
1	9-Octadecenamide, (Z)-		15	C18H35NO	281.4766	22.497
2	Methyl salicylate		3	C8H8O3	152.15	4.706
3	Dodecane		2	C12H26	170.33	4.644
4	Tetradecane		5	C14H30	198.39	6.401
5	2,4-bis(1,1-dimethylethyl)phenol		6	C17H30O	278.5	7.325

. From the analysis, 9-Octadecenamide (Z) is identified as the most abundant compound. However, none of the cannabinoids are detected.

4. Discussion

In this study, we successfully developed optimized conditions for the in vitro regeneration of Cheungsam. The findings of the study are as follows: (i) the additional sterilization of seedlings with removed seed coats using H₂O₂ can further reduce the contamination rate of in vitro seedlings germinated from heavily contaminated seeds; (ii) optimal hormone concentrations that can be used for the preparation of Cheungsam-specific CIM, SIM, and RIM were determined; and (iii) secondary metabolites in callus derived from Cheungsam leaf explants were identified through GC–MS analysis. As previously reported, using a 1% H₂O₂ solution as an LGM significantly improved seed germination rates. However, unlike our previous report [10], this treatment still resulted in considerable contamination, likely due to severe endophytic contamination in the seeds. Subsequent sterilization of seedlings with removed seed coats using higher concentrations of H₂O₂ effectively reduced the contamination rates. However, the observed growth retardation caused by prolonged H₂O₂ exposure highlights the need for the precise optimization of both H₂O₂ concentration and exposure time during sterilization to ensure seedling health [10,38,39,40,41,42].

In plants, de novo organogenesis refers to the regeneration of new organs (e.g., roots or shoots) from wounded or detached explants through the formation of callus under in vitro conditions. When plants experience wounding stress, they activate responses to repair the damage. This includes forming an unorganized callus mass and, in some cases, generating new organs through vascular differentiation and de novo organogenesis [23,24,43]. These processes involve complex hormonal interactions between auxin and cytokinin, where (i) intermediate auxin-to-cytokinin ratio induces callus formation, (ii) high auxin-to-cytokinin ratio promotes root formation, and (iii) low auxin-to-cytokinin ratio promotes shoot regeneration. These hormonal signals are regulated by intricate gene networks, including transcription factors, hormonal transport/metabolism genes, and cell cycle-related regulators, ensuring successful organogenesis.

The selection of appropriate explants is critical for the success of in vitro de novo organogenesis. In dicotyledonous plants, leaves and cotyledons act as natural sources of auxin, with auxin transport occurring from the tip toward the stem/hypocotyl and roots [43,44]. This characteristic makes leaves and cotyledons particularly suitable for inducing cell division and serving as effective explants for de novo organogenesis. Interestingly, cotyledon explants often exhibit a higher callogenesis response compared to leaf explants in many dicotyledonous species, including Cannabis [45]. This is likely due to their greater capacity for auxin biosynthesis. In our experiments, the optimal callus induction medium (CIM) for leaf and cotyledon explants contained 1.0 mg/L TDZ and 0.5 mg/L NAA. Under these conditions, cotyledon explants demonstrated faster callus formation than leaf explants. However, higher concentrations of synthetic auxins, such as NAA (≥1.0 mg/L), are well-known for their phytotoxic effects [45,46,47]. These auxins, often referred to as “auxinic herbicides”, can disrupt the auxin homeostasis, leading to the overproduction of ethylene, abscisic acid, reactive oxygen species (ROS), and other harmful metabolites. In our study, higher doses of NAA negatively impacted both the callogenesis response and callus production, underscoring the critical importance of maintaining hormonal balance in plant tissue culture.

Both leaf- and cotyledon-derived calluses exhibited optimal de novo shoot morphogenesis on shoot induction medium (SIM) containing 0.5 mg/L TDZ. TDZ, a potent plant growth regulator, is known to stimulate cytokinin-associated pathways by interacting with cytokinin receptors [48,49]. It also modulates the activities of enzymes involved in auxin and cytokinin regulation, enabling the induction of both the auxin and cytokinin signaling pathways. This dual hormonal regulation makes TDZ particularly effective in inducing morphogenesis, even in recalcitrant species such as Cannabis [48,49]. However, TDZ-driven morphogenesis is highly sensitive to its concentration and the duration of exposure. Excessive concentrations can inhibit growth and development, likely due to the disruption of hormonal balance. In our study, while callus induction medium (CIM) produced satisfactory results for both leaf and cotyledon explants, the morphogenic response on SIM was suboptimal for both explant types. This reduced response might be attributed to the highly recalcitrant nature of the tissues, as reported in earlier studies [23,24,43,45,46,47]. Additionally, shoot morphogenesis decreased significantly at higher TDZ concentrations (≥1.0 mg/L). This emphasizes the importance of precise hormonal regulation in plant tissue culture. Higher concentrations of TDZ have been associated with adverse effects, including tissue necrosis, loss of rooting ability, hyperhydricity (excessive water retention), and cytogenetic abnormalities, as highlighted in the recent reviews [50]. These findings further underscore the need to carefully optimize TDZ concentrations to balance effective morphogenesis with minimal phytotoxic effects.

Rooting from de novo-formed shoots is a crucial step in plant tissue culture to complete the micropropagation process [45]. Physiologically, roots play a vital role in channeling the shoot-derived auxin, which is essential for coordinated plant growth and development. Successful rooting depends on three key factors [23,45]: (i) High auxin production in specific xylem-associated cells within the stem. (ii) Effective delivery of auxin to the target tissues. (iii) Optimal hormonal balance to induce root formation without causing phytotoxicity. Several de novo root induction protocols have been developed, including treatments with relatively high auxin concentrations, such as IBA [23,24,44,51,52,53,54]. In our study, the best de novo root morphogenesis was observed on RIM containing 2.5 mg/L IBA. Both leaf- and cotyledon-derived calluses showed excellent responses, with high rooting percentages (>92%), significant root numbers (>9.7 roots per explant), and substantial root lengths (>7.2 cm). However, a reduction in root morphogenesis was observed at both lower and higher IBA concentrations, highlighting the importance of maintaining a precise balance in auxin application for successful root induction.

Our GC–MS analysis identified 19 different compounds; however, unexpectedly, no cannabinoids or their related compounds were detected. Instead, 9-octadecenamide (Z), also known as oleamide, was identified as the most abundant compound. This bioactive lipid belongs to the fatty acid amides (FAAs) class, a group of molecules found in both the central nervous system (CNS) and peripheral tissues [55]. FAAs, which are composed primarily of “orphan” ligands, often interact with the “orphan” or unidentified receptors, playing key roles in various biological processes [56]. Within this lipid class, oleamide functions as a chemical signaling molecule and has been shown to interact with cannabinoid receptors [55,56,57]. Through this interaction, oleamide can induce diverse biological effects, such as promoting sleep and modulating immune responses. Additionally, oleamide acts as a neurotransmitter and has been implicated in processes such as thermoregulation and pain modulation (antinociception). These findings highlight its potential role as a multifunctional signaling molecule with a broad spectrum of physiological effects.

Methyl salicylate (MeSA or methyl 2-hydroxybenzoate) was identified as the second most abundant compound in the GC–MS analysis. MeSA is a volatile organic compound (VOC) and serves as a methyl ester derivative of salicylic acid (SA). MeSA is widely emitted by various plant species, and it functions as a signal-inducing systemic acquired resistance (SAR), helping plants defend against pests, pathogens, and competing antagonists [58,59,60,61,62]. In plants, MeSA is synthesized from SA via the isochorismate and the phenylalanine ammonia-lyase pathways. However, beyond its role in plant defense, recent studies have highlighted the pharmacological significance of MeSA for human health [62]. Naturally occurring salicylates, including MeSA, are known to target multiple regulatory pathways, contributing to beneficial health effects. These include anti-inflammatory, anticancer, neuroprotective, and anti-diabetic properties, making MeSA a molecule of interest in both plant and medical sciences.

Our GC–MS analysis identified two specific alkanes, dodecane and tetradecane, as the third- and fourth-most abundant compounds, respectively. Alkanes, also referred to as paraffins (from Latin, meaning “with little chemical affinity”), are non-polar, stable, and minimally reactive organic compounds [63]. Structurally, they are linear, saturated hydrocarbons with the general formula C_nH_2n₊₂ and are widely used as lipid biomarkers for characterizing essential and volatile oils [63,64,65,66]. In plants, alkanes are synthesized as part of the protective wax layer on leaf surfaces, serving critical physiological and ecological roles, such as reducing water loss, providing UV protection, and acting as a barrier against pests and pathogens [67,68,69,70]. Long-chain alkanes are also considered intermediate VOCs and may function as precursors for the synthesis of secondary metabolites, such as terpenes [71]. Terpenes, prominent secondary metabolites, exhibit potent bioactivity and have therapeutic potential for treating infectious and chronic diseases. Their properties include anti-inflammatory, antioxidant, antimicrobial, antitumor, and antidiabetic effects. Terpenes are synthesized from acetyl-CoA via the mevalonic acid pathway in the cytoplasm. Additionally, integrative genomic analyses have revealed co-expression networks of genes involved in the biosynthesis of both cannabinoids and terpenoids, highlighting the shared origin of the cannabinoid and terpenoid biosynthetic pathways [72]. Furthermore, the degradation and oxidation of alkanes can yield fatty acids and acetyl-CoA, which are essential intermediates for various biosynthetic pathways, including secondary metabolite production [67,68,69,70,73]. These findings underscore the multifaceted role of alkanes in the biosynthesis of bioactive compounds with significant ecological and therapeutic implications.

Finally, 2,4-bis(1,1-dimethylethyl)phenol (2,4-DTBP) was identified as the fifth-most abundant compound, highlighting its prominence among the detected metabolites. Also known as 2,4-Di-tert-butylphenol, 2,4-DTBP is a secondary metabolite produced by a wide range of organisms, including plants, fungi, and bacteria [74]. 2,4-DTBP is frequently detected as a major constituent in volatile and essential oils and is well-known for its pronounced toxicity against various organisms, including its own producers. This toxicity is mediated through multiple mechanisms, such as enzyme inhibition, DNA structure alterations, disruption of ribosomal translation, and cellular membrane destabilization [74,75].

Beyond its toxicity, 2,4-DTBP displays a wide range of bioactivities, including antimicrobial, herbicidal, antioxidant, insecticidal, nematicidal, and cytotoxic effects [74,75]. These properties make it a compound of interest in various fields, including agriculture, medicine, and environmental science.

5. Conclusions

This study successfully optimized the in vitro regeneration conditions for the Cheungsam cultivar, providing a robust framework for the mass propagation of pathogen-free, genetically uniform Cannabis plants. Through de novo organogenesis, this study demonstrates the importance of hormonal balance, particularly the auxin-to-cytokinin ratio, in regulating callus formation, root development, and shoot regeneration. Moreover, the GC–MS analysis revealed a diverse range of secondary metabolites, providing valuable insights into the chemical composition of Cheungsam-derived calluses. Key compounds such as oleamide, MeSA, dodecane, tetradecane, and 2,4-DTBP were identified, each possessing significant ecological and therapeutic potential. Collectively, this study lays the groundwork for future research into optimizing tissue culture techniques and exploring the molecular mechanisms underlying in vitro regeneration and secondary metabolite biosynthesis. The integration of advanced omics approaches, including transcriptomics, proteomics, and metabolomics, could further unravel the genetic and biochemical pathways involved, paving the way for metabolic engineering and the development of high-value compounds for agricultural and pharmaceutical applications.