Strain- and System-Specific Enhancement of Artemisinin in Artemisia annua Composite Plants Grown in Hydroponic and Aeroponic Systems

Abstract

Efficient production of artemisinin, a valuable secondary metabolite from Artemisia annua, remains a challenge for pharmaceutical applications. This study investigated the use of ex vitro composite plants—generated by inoculation with Agrobacterium rhizogenes strains 2659 and 1523—under hydroponic and aeroponic conditions to enhance artemisinin and phenolic compound accumulation. In leaves, artemisinin content increased in a cultivation-specific, strain-dependent manner: strain 2659 was effective under aeroponics (+36%), while strain 1523 enhanced accumulation under hydroponics (+32%). In roots, strain 2659 led to higher artemisinin accumulation than strain 1523 under both systems, with increases of up to 145% in hydroponics and 75% in aeroponics. Strain 1523 strongly promoted artemisinin exudation, especially in hydroponics, suggesting active regulation of artemisinin export. Aeroponic cultivation increased total phenolic content (TPC) in roots, while strain 1523 reduced TPC in leaves. Although total biomass was unaffected, A. rhizogenes altered assimilate partitioning, decreasing the shoot-to-root ratio and enhancing root metabolism. These findings demonstrate that ex vitro composite plants, combined with optimized soilless cultivation, represent a flexible tool to boost accumulation and secretion of high-value compounds in A. annua. The strain and environment-specific responses emphasize the importance of selecting appropriate bacterial strain–cultivation combinations for scalable production systems.

1. Introduction

Controlled agricultural systems, such as hydroponics and aeroponics, offer a promising platform for cultivating high-value medicinal plants [1]. These systems enable efficient harvesting of both above- and below-ground biomass, as well as continuous, non-invasive collection of root exudates throughout the plant’s life cycle. Recovery of bioactive compounds directly from nutrient solutions avoids the need for complex tissue extraction, providing a less invasive and more sustainable approach. In controlled environments, minimized external variability ensures standardized production of herbal medicines, addressing a critical challenge in pharmaceutical applications where batch-to-batch consistency of active ingredients is essential [2]. However, widespread adoption of these systems remains limited, partly due to incomplete understanding of how to optimize cultivation for maximizing both biomass and secondary metabolite yields.

Despite the advantages of hydroponic and aeroponic systems, realizing their full potential for medicinal plant production remains challenging. Successful cultivation requires not only maximizing biomass yield but also enhancing the accumulation of bioactive secondary metabolites, two goals that are often physiologically antagonistic. Specifically, the trade-off between growth and defense mechanisms must be carefully managed: stress conditions that limit growth can simultaneously stimulate secondary metabolite biosynthesis [3]. Achieving an optimal balance requires understanding the plant’s response to various stressors, often framed as a competition for carbon resources between growth processes (such as cell division and photosynthesis) and defense mechanisms, including secondary metabolism [4].

Secondary metabolites are crucial for plant adaptation to stress and are the main contributors to the therapeutic properties of many medicinal plants. Therefore, overcoming the growth–defense dilemma is essential for advancing sustainable agricultural practices and for the cultivation of plants rich in beneficial compounds. Previous research has demonstrated the possibility of decoupling growth and secondary metabolite production through specific treatments, such as jasmonates and far-red light [5]. Beyond hormonal or light-based treatments, hydroponic and aeroponic cultivation systems have emerged as promising alternatives for enhancing secondary compound accumulation under controlled environmental conditions without negatively impacting plant growth.

Another approach to boosting secondary metabolite production involves the use of genetically transformed roots, known as hairy root cultures (HRCs). Hairy roots, induced by Agrobacterium rhizogenes infection, are characterized by rapid growth, high genetic stability, and the ability to produce secondary metabolites at levels comparable to or exceeding those found in intact plants [6]. However, their application remains largely confined to laboratory-scale production due to technical challenges in scaling up. To overcome these limitations, the concept of ex vitro composite plants—combining wild-type shoots with genetically transformed roots—has emerged as a promising strategy. This system retains the metabolic advantages of transformed roots while enabling cultivation under scalable hydroponic and aeroponic systems.

Examples from Datura innoxia and Glycyrrhiza glabra demonstrate that composite plants can significantly enhance secondary metabolism in both roots and shoots, with increases in alkaloid and flavonoid accumulation [7,8]. These findings are consistent with the well-established properties of hairy roots and support the potential of composite plants as a scalable platform for bioactive compound production [9].

The interaction of transformed roots with different root-zone environments, particularly in hydroponic and aeroponic systems, remains insufficiently examined. These systems allow refined nutrient management, which may influence root architecture, nutrient assimilation, and secondary metabolism. Among them, aeroponics is a particularly promising system due to its advantages in root-zone management, superior nutrient absorption, and oxygenation, leading to improved growth and enhanced accumulation of phenolics, flavonoids, antioxidants, and vitamins compared to soil cultivation [10,11,12].

Within medicinal plants, the Artemisia genus stands out for producing potent bioactive compounds, most notably artemisinin, the most effective antimalarial drug currently known [13]. Artemisinin is synthesized in glandular trichomes of leaves, where its precursors are secreted into the subcuticular cavity for final conversion and storage, making leaves the primary site of production [14]. There is growing interest in cultivating Artemisia under controlled systems, including hydroponics, to improve both yield and biomass quality. For example, A. afra showed optimal growth and chlorophyll content at near-neutral pH in hydroponics [15]. In addition to leaf trichomes, root-based production has been explored through hairy root cultures of A. annua, which can significantly enhance artemisinin accumulation [16]. However, these cultures remain confined to laboratory scale because their maintenance requires sterile conditions and costly in vitro media, making large-scale production economically impractical. At the same time, it remains unclear whether composite plants can also enhance artemisinin accumulation under scalable hydroponic or aeroponic cultivation.

In this study, we aimed to evaluate the influence of inoculation with two A. rhizogenes strains (2659 and 1523), combined with cultivation method (hydroponics vs. aeroponics), on biomass allocation, secondary metabolite accumulation (total phenolics and artemisinin), and root exudation in ex vitro composite A. annua plants. By integrating genetic and environmental strategies, we sought to identify approaches to enhance artemisinin production and improve the biosynthetic performance of A. annua under scalable, controlled cultivation systems.

2. Materials and Methods

2.1. Growth and Experimental Setup for Artemisia annua

A. annua seeds were obtained from Anamed International and sterilized prior to sowing. The sterilization procedure involved immersion in 70% ethanol for 2 min, followed by treatment with 2% sodium hypochlorite (equivalent to commercial bleach) for 10 min. Seeds were then thoroughly rinsed with deionized water. For germination, seeds were placed between two irrigation mats and supplied with a modified half-strength nutrient solution (NS) based on the formulation of Hoagland and Arnon [17]. The full-strength NS contained: 5 mM Ca(NO₃)₂, 0.25 mM KH₂PO₄, 1.25 mM MgSO₄, 1.75 mM K₂SO₄, and the following micronutrients: 20 µM Fe, 1.25 µM Mn, 1.5 µM Zn, 25 µM B, 0.5 µM Cu, and 0.175 µM Mo. To counteract autotoxic and growth-inhibiting compounds naturally released by A. annua, a continuous circular flow of nutrient solution was applied from above via drippers positioned between the plates.

2.2. Agrobacterium rhizogenes Strains and Transformation

This study examined the effects of two A. rhizogenes strains, representing distinct opine types, on artemisinin accumulation in different organs and on root exudation in A. annua. Strain NCPPB 2659 (K599, cucumopine type), obtained from the laboratory of Csaba Koncz at the Max Planck Institute for Plant Breeding Research (Cologne, Germany), carries a cucumopine-type Ri plasmid with contiguous T-DNA and is widely used for its high transformation efficiency and robust hairy root induction. Strain NCPPB 1523 (ATCC 15834), obtained from Fera Science Ltd. (York, UK), harbors an agropine-type Ri plasmid with non-contiguous T-DNA, auxin biosynthetic genes, and broad host range; it is one of the earliest and most frequently studied wild-type strains. These strains were selected to represent two different opine classes (cucumopine vs. agropine), allowing comparison of strain-specific effects on metabolite production in composite plants [18].

For transformation, seedlings were used 30 days after sowing (DAS), when the primary roots reached a length of 4–5 cm. Roots were excised under sterile conditions, and the hypocotyl region was inoculated for 10 min with bacterial suspensions (OD600 = 0.4). The bacterial cultures were grown in LB medium at 27 °C with shaking at 160 rpm in darkness for 20 h. Control shoots were treated identically but incubated in sterile LB medium.

Following inoculation, shoots were inserted upright into autoclaved rock wool cubes (Grodan, Roermond, The Netherlands; 2 cm × 2 cm × 2 cm) saturated with 4 mL of the bacterial suspension. Cubes were placed in trays covered with transparent lids to maintain >90% humidity. The trays were maintained overnight under low ambient fluorescent lighting (~50 µmol m⁻²s⁻¹) to allow acclimation. On the second day, lids were removed, often causing temporary wilting; recovery was supported by adding 2–5 mL sterile water to each cube and temporarily placing the lid back on to restore humidity. This watering cycle was repeated for up to 4 days. From day 5, cubes were irrigated every 1–2 days with 4–5 mL of 10% nutrient solution supplemented with 1 mM KNO₃. Plants were maintained in covered cultivation boxes to preserve humidity, with shoots reduced to two leaves to minimize transpiration. Conditions were 16 h light/8 h dark at 22 °C/18 °C (day/night), ~80 µmol m⁻² s⁻¹.

Once both non-transformed controls and transformed lines developed multiple adventitious roots overgrowing the cubes, the composite plants (wild-type shoots with transgenic roots) and controls were transferred at 41 DAS to hydroponic or low-pressure aeroponic systems. Plants were cultivated in 2.4 L pots containing 1.8 L half-strength nutrient solution supplemented with 1.5 mM KNO₃, maintained at pH 5.5–6.0, and renewed weekly. Growth chambers were set to 16 h light/8 h dark at 22 °C/18 °C (day/night), relative humidity of 60/80% day/night, and a light intensity of ~190 µmol m⁻² s⁻¹. After a one-week adaptation, plants received full-strength nutrient solution (pH 5.5–6.3). Each treatment included four biological replicates. In hydroponics, nutrient solution was continuously aerated, whereas in the aeroponic system, the roots were sprayed continuously with nutrient mist via nozzles.

2.3. Exudate Collection and Root Exudate Extraction

Eighteen days after transferring plants to aeroponic or hydroponic systems, each plant was placed in a foil-covered, light-impermeable pot containing 400 mL of continuously aerated distilled water (pH 6.2) for 17–23 h. Following this collection period, plant tissues were harvested. Artemisinin was extracted from the root exudates as described in [19]. The collected eluates were stored at −20 °C until further analysis.

2.4. Final Sampling of A. annua

At 59 DAS, plants were harvested for evaluation. Leaves, shoots, and roots were weighed, and leaf area was measured. Samples were dried at 40 °C for dry weight determination. The following traits were calculated: total biomass, leaf dry matter percentage (LDM%), root dry matter percentage (RDM%), stem dry matter percentage (SDM%), leaf weight ratio (LWR), and specific shoot area (SSA; cm²/g dry weight). Approximately 1 g each of root and leaf tissue per treatment was frozen in liquid nitrogen and stored at −80 °C. An additional 1 g of fresh root material was preserved in 50% ethanol for root hair analysis, following the method described by Paponov et al. [20]. For microscopic analysis, twenty first-order adventitious root segments (~1 cm each) per plant were preserved in 50% ethanol and visualized using a TopView stereomicroscope at 64× magnification. Images were analyzed with ImageJ software (National Institutes of Health, Bethesda, MD, USA; version 1.52a) to quantify root hair length and density per segment.

2.5. Sample Preparation for UHPLC, Extraction and Quantification of Artemisinin

Lyophilized and powdered leaf and root samples of A. annua (50 mg each) were ground at 29 Hz for 3 min using a Starbeater (VWR, Radnor, PA, USA) in 2 mL Eppendorf tubes with a 5 mm steel bead. The powder was extracted with 1.5 mL of 80% methanol by vortexing for 20 min at maximum speed. The extract was then centrifuged at 17,000× g for 7 min (MicroStar 17R, VWR, Radnor, PA, USA), and the supernatant was collected and centrifuged again to remove residual sediment. The clarified extract was stored at −20 °C until UHPLC analysis. The exact protocol for quantitative determination of artemisinin has been described previously [20].

2.6. Total Phenolic Content (TPC) Extraction and Assay

In addition to artemisinin, we quantified total phenolic compounds (TPCs) because A. annua is a rich source of phenolics such as chlorogenic and caffeic acid, which are potent antioxidants strongly linked to stress responses and ROS regulation [21]. Although arising from a different biosynthetic pathway, phenolics—like artemisinin—are stress-inducible metabolites, and their analysis provided complementary insight into secondary metabolism and defense physiology in composite plants.

Total phenolic content (TPC) in leaves, roots, and exudates was quantified using the Folin–Ciocalteu assay [22]. Freeze-dried samples of 100 mg dry mass (DM) from leaves and 20–100 mg DM from roots were extracted in 1.5 mL of 80% methanol in darkness for 24 h at 35 °C, followed by centrifugation at 13,000× g for 5 min. The supernatant was diluted accordingly (leaves: 50×; roots: 25×). Root exudate extracts were diluted 1:2 (v/v).

For the assay, a 200 µL aliquot of each extract was mixed with 200 µL of 10% Folin–Ciocalteu reagent (Merck, Darmstadt, Germany), followed by 800 µL of 700 mM Na₂CO₃. Samples were incubated for 2 h at room temperature in darkness, and absorbance was measured at 765 nm using a Multiscan GO plate reader. Standard curves were generated using gallic acid (50 µM–2.5 mM in 80% MeOH), and TPC was expressed as µg gallic acid equivalents (GAE) per g dry weight (DW). The exudation rate was calculated relative to root fresh weight (FW) and duration of exudation.

2.7. Statistical Analysis

All data were analyzed using two-way analysis of variance (ANOVA). Each treatment included four biological replicates. Where significant differences were detected, Fisher’s protected Least Significant Difference (LSD) test was applied. Statistical analyses were performed using Statistica version 13 (StatSoft, Palo Alto, CA, USA).

3. Results

Neither inoculation with A. rhizogenes strains 2659 and 1523 nor cultivation method (hydroponics vs. aeroponics) significantly altered total plant biomass compared to non-inoculated control plants (Figure 1A). Inoculation with strains 2659 and 1523 either decreased or tended to decrease leaf dry matter content (LDM%) compared to non-inoculated plants (Figure 1B). Aeroponic cultivation increased LDM% in control plants (non-inoculated cuttings). However, in inoculated plants, no significant differences in LDM% were observed between hydroponics and aeroponics.

The two-way ANOVA indicated that aeroponic cultivation significantly increased stem dry matter content (StemDM%) overall (Table S1, Figure 1C). Inoculation with strain 2659 strongly reduced StemDM%, while strain 1523 caused only a slight reduction. For root dry matter content (RootDM%), a significant interaction between cultivation method and inoculation was found (Table S1, Figure 1D). Aeroponics increased RootDM% in both control and 2659-inoculated plants, but not in 1523-inoculated plants. Notably, under hydroponic conditions, RootDM% in 1523-inoculated plants was higher than in control and 2659-inoculated plants.

For the leaf weight ratio (LWR), a significant interaction between cultivation method and inoculation was also found (Figure 1E). In control plants, aeroponics increased LWR, whereas in plants inoculated with strains 2659 and 1523, aeroponics decreased or tended to decrease LWR. This suggests that inoculation impairs dry matter allocation between shoots and roots, possibly related to oxygen supply to roots. Specific shoot area (SSA) was not affected by cultivation method. Inoculation, however, affected SSA differently: strain 2659 had no significant effect, whereas strain 1523 significantly reduced SSA (Figure 1F).

As expected, aeroponic cultivation enhanced both root hair length (Figure 1G) and, more markedly, root hair density (Figure 1H). Inoculation with A. rhizogenes strains 2659 and 1523 did not significantly affect either root hair length or density.

Because phenolics are potent antioxidants and stress-inducible metabolites, we quantified total phenolic content (TPC) as a complementary indicator of secondary metabolism. In the leaves, inoculation with strain 2659 did not significantly alter total phenolic content (TPC) under either cultivation method, whereas inoculation with strain 1523 significantly decreased TPC in aeroponics and showed a tendency to decrease in hydroponics (Figure 2A). In the roots, inoculation with either strain did not significantly change TPC in hydroponics (Figure 2B). Aeroponics increased root TPC in control plants and in plants inoculated with strain 2659, whereas strain 1523 showed no significant effect. For exudates, inoculation with strain 2659 increased the TPC exudation rate under aeroponics but not under hydroponics, whereas strain 1523 had no significant effect (Figure 2C).

Inoculation with strain 2659 increased artemisinin concentration in leaves by approximately 18% in hydroponic culture and 36% in aeroponic culture (Figure 3A). Inoculation with strain 1523 increased leaf artemisinin by about 32% in hydroponics, but no significant effect was observed in aeroponics. The type of cultivation (hydroponics vs. aeroponics) did not significantly affect artemisinin accumulation in roots (Figure 3B). Inoculation with strain 2659 markedly increased root artemisinin content by 145% in hydroponics (significant) and 75% in aeroponics (non-significant). Inoculation with strain 1523 showed a non-significant tendency to increase root artemisinin in hydroponics but had little effect in aeroponics.

Despite the higher accumulation of artemisinin in roots, inoculation with strain 2659 did not affect the rate of artemisinin exudation compared to control plants (Figure 3C). In contrast, inoculation with strain 1523 caused a strong increase in exudation under hydroponics (~50-fold, significant) and showed a tendency toward increased exudation in aeroponics (~8-fold), although not statistically significant. While the absolute amounts remained small compared to leaf accumulation, the significant increase in hydroponics highlights the potential of composite plants to stimulate metabolite release into nutrient solutions.

4. Discussion

4.1. Enhancing Artemisinin Biosynthesis in Leaves via Composite Plants

Our findings demonstrate that ex vitro composite plants, generated through A. rhizogenes inoculation, represent a flexible and scalable platform for enhancing artemisinin production in A. annua cultivated in soilless systems. Leaf artemisinin levels increased in a cultivation-dependent, strain-specific manner: strain 2659 was effective only in aeroponics, while strain 1523 enhanced accumulation exclusively in hydroponics (Figure 3A). Importantly, these improvements occurred without affecting total biomass (Figure 1A), although some growth-related parameters such as leaf and stem dry matter percentage and specific shoot area (Figure 1B,C,F) were significantly altered. The absence of biomass reduction highlights the practical utility of composite plants as a low-cost strategy for metabolic enhancement. This aligns with broader observations that hydroponic systems can improve secondary metabolite content in a species- and organ-specific fashion, making them ideal for precise optimization in medicinal plant production [1].

Similar results were reported in Glycyrrhiza glabra, where ex vitro composite plants grown hydroponically showed significant increases in flavonoid and triterpenoid levels, further supporting the efficiency and scalability of this approach for high-value compound production [8]. The observed enhancement of artemisinin in A. annua composite plants is likely linked to the activity of rol genes introduced via A. rhizogenes. In Rubia cordifolia callus cultures, transformation with rolB and rolC significantly boosted anthraquinone accumulation—shikimate-derived secondary metabolites—relative to non-transformed controls [23]. The response of these transformed cultures to methyl jasmonate and salicylic acid provides further evidence for the involvement of defense-related signaling pathways. Notably, rolB, which encodes a protein with tyrosine phosphatase activity, may influence phosphorylation-dependent regulatory cascades, as shown by the specific induction of anthraquinone biosynthesis upon treatment with the phosphatase inhibitor cantharidin. In contrast, rolA appears to act via an auxin-dependent mechanism that modulates ROS homeostasis, suppressing ROS accumulation and downregulating related genes in rolA-expressing cells [24]. Importantly, direct evidence for such regulation in A. annua comes from hairy root cultures. In these systems, rolB expression has been associated with strong increases in artemisinin accumulation, likely through transcriptional activation of key pathway genes such as ADS, CYP71AV1, DBR2, and ALDH1, combined with alterations in auxin and ROS signaling [25]. This suggests that the strain-specific increases in leaf artemisinin observed in our composite plants could similarly arise from transcriptional and hormonal reprogramming triggered by transformation, even in untransformed shoots. Taken together, these studies highlight the capacity of rol genes to reprogram plant metabolic networks through distinct signaling routes—even in composite systems where only the roots are genetically modified.

The observed strain × environment interaction (Table S1) likely reflects how root-zone factors—such as oxygenation, moisture, and nutrient availability—modulate the systemic signals triggered by root transformation. Although only the roots were genetically transformed in composite plants, significant metabolic changes were also observed in the untransformed aerial tissues (Figure 3A). This supports the idea that A. rhizogenes-mediated transformation initiates long-distance signaling capable of reprogramming shoot metabolism. The systemic nature of this response has been further confirmed in other species, reinforcing the broader relevance of composite plant systems for whole-plant metabolic enhancement [8]. However, this systemic effect is context-dependent: different strains were effective only under specific cultivation systems. These findings suggest that the outcome of transformation is not dictated solely by the bacterial genotype but also by environmental parameters that shape signal perception, transduction, and potentially metabolite transport. Thus, optimizing composite plant systems for enhanced metabolite production requires strategic matching of the bacterial strain with the cultivation environment. A better understanding of how transformation-induced signaling integrates with environmental modulation is essential for realizing the full potential of composite plants as metabolic biofactories.

The differential effects of A. rhizogenes strains on metabolite accumulation are likely driven by multiple factors—including differences in T-DNA copy number and integration sites, variation in virulence gene expression and transfer efficiency, co-transfer of accessory genes, and strain-specific modulation of host signaling pathways—despite conservation of rol gene sequences [26].

4.2. Root Artemisinin Accumulation and Exudation: A Neglected but Significant Target

Although artemisinin research has focused primarily on leaves due to their much higher baseline content—over 100-fold greater than in roots [19]—our results show that transformation with strain 2659 significantly increased artemisinin accumulation in roots: by 145% in hydroponics and 75% in aeroponics (Figure 3B). Conversely, strain 1523 induced only a weak tendency to increase root artemisinin content but caused a dramatic increase in artemisinin exudation under hydroponics (~50-fold compared to controls, significant) and a tendency toward higher exudation under aeroponics (~8-fold), although not statistically significant (Figure 3C). Although the absolute amounts exuded were still small compared to accumulation in leaves, the significant effect in hydroponics together with the trend in aeroponics indicates the potential of composite plants to enhance metabolite release into the nutrient solution. This provides an indication that root exudation could be harnessed in controlled cultivation systems as a complementary route for continuous, non-destructive compound recovery.

The observed decoupling between artemisinin concentration in roots and its exudation suggests that export is under active regulation rather than simple diffusion. This interpretation is supported by findings that artemisinin exudation in A. annua responds to nutrient signals independently of root hair formation and internal content [19]. In shoots, secretion from glandular trichomes depends on dedicated transport systems, such as the ABC transporter AaPDR2 and the lipid-transfer protein AaLTP3, which together facilitate the movement of artemisinin precursors into the subcuticular cavity [27]. By analogy, related transporter systems may also operate in roots, mediating the release of artemisinin or its precursors into the surrounding medium. It is possible that transformation with different A. rhizogenes strains affects the regulation of such transport processes, with some strains primarily enhancing biosynthesis and others preferentially stimulating export.

Our results demonstrate that hydroponic systems can enable the collection of artemisinin exuded from roots in a non-destructive manner, highlighting their potential for continuous compound recovery in medicinal plants [19]. In composite plants, this trait may potentially be exploited in recirculating hydroponics or nutrient-film systems that allow media capture, providing a conceptual alternative to destructive harvesting for secondary metabolite recovery.

These findings highlight the value of root transformation and hydroponic systems for enhancing and harnessing root-based artemisinin production in composite plants. To our knowledge, this is the first study to systematically address both root accumulation and exudation of artemisinin in composite plants under defined cultivation systems, offering a dual-target strategy for optimizing compound yield and recovery.

4.3. Tissue- and Compound-Specific Regulation of Phenolics

Beyond changes in metabolite profiles, transformation also affected whole-plant physiology, particularly the partitioning of biomass between shoots and roots. While total biomass remained unchanged, clear differences emerged in dry matter allocation. Transformation with either bacterial strain consistently reduced the leaf weight ratio (LWR), indicating a shift in resource allocation toward the roots (Figure 1E). This response is consistent with the characteristic phenotype of rol-transformed roots, which display enhanced sink strength and altered hormonal regulation. Hairy roots are easily distinguished by their rapid growth, extensive lateral branching, and plagiotropic development—even in hormone-free media—features first described in vitro [28].

Aeroponic cultivation also influenced LWR, but in the opposite direction: in control plants, LWR increased under aeroponics, reflecting a reduced investment in root biomass due to improved oxygen availability. However, this aeroponic effect was diminished in transformed plants. Specifically, LWR decreased further in plants inoculated with strain 2659 and remained unchanged in those inoculated with strain 1523. This suggests that the hormonal and metabolic reprogramming induced by transformation can override or counteract the typical aeroponic effect on shoot–root allocation, reinforcing root investment even under conditions that would normally favor shoot growth. Notably, enhanced oxygenation in aeroponics appears to further promote carbon allocation to transformed roots, potentially reinforcing their sink strength and metabolic activity. This observation contrasts with earlier reports from in vitro hairy root cultures, where roots grown in oxygen-rich gas-phase reactors showed reduced biomass accumulation [29]. In our intact composite plants, however, transformed roots maintained or even enhanced their allocation under aeroponic conditions, while non-transformed roots showed reduced investment. This discrepancy suggests that root responses to oxygen availability are not solely governed by local conditions, but are modulated by systemic shoot–root communication.

Phenolic compounds, especially those with ortho-dihydroxy and quinoid structures, play well-established roles as antioxidants, radical scavengers, and structural stabilizers during stress responses [30]. While their accumulation is typically induced under abiotic stresses such as drought, salt stress, heavy metals, UV radiation, and temperature [31], none of these stress factors were present in our aeroponic conditions. Instead, the increase in root phenolic content (Figure 2B) is more likely linked to enhanced oxygen availability, which may stimulate aerobic metabolism and moderate ROS production. This interpretation is supported by findings from waterlogged Arabidopsis roots, where oxygen deficiency led to the reduced accumulation of major phenolic classes, including flavonoids, lignans, and terpenoids—consistent with a metabolic suppression under hypoxia [32].

Phenolic exudation remained largely unchanged across treatments (Figure 2C). In control plants, no difference was observed between hydroponic and aeroponic cultivation, despite significantly higher root TPC under aeroponics (Figure 2B)—indicating that tissue content alone does not determine exudation levels. A tendency toward increased exudation was observed only in 2659-inoculated plants, which may be explained by diffusion. However, the discrepancy in control plants suggests that additional regulatory mechanisms—beyond passive diffusion—likely influence phenolic exudation, possibly involving changes in root membrane permeability, transporter activity, or compartmentalization.

Together, these findings highlight the compound- and tissue-specific nature of metabolic responses in composed plants and reinforce the importance of profiling multiple metabolites when evaluating transformation-based strategies for secondary metabolite enhancement.

4.4. Physiological Trade-Offs and Biomass Allocation

The observed shift in biomass allocation toward the roots in transformed plants (Figure 1, Table S1) likely reflects deeper physiological reprogramming beyond compound-specific responses. This reallocation is consistent with the known role of rol genes—particularly rolB—in modulating signal transduction pathways and redirecting developmental priorities in hairy roots. rolB has been shown to possess tyrosine phosphatase activity and to interact with key regulatory proteins such as 14-3-3s, which control cellular growth and differentiation [33]. These systemic effects likely account for the stable reinforcement of root sink strength observed in composite plants, even though only the roots are genetically transformed. In our experiments, this was evident from the increased root dry matter percentage and altered leaf weight ratio in transformed plants (Figure 1D,E).

Moreover, [33] emphasized that hairy root cultures function as self-inducing systems for secondary metabolism, eliminating the need for exogenous hormones that would otherwise interfere with developmental balance. This intrinsic reprogramming capacity offers a powerful tool for redirecting plant resources toward compound synthesis at the expense of structural growth. In line with this, we observed increased phenolic accumulation in roots and enhanced exudation of phenolics under aeroponics after inoculation with strain 2659 (Figure 2B,C)

Such trade-offs—reduced shoot development in favor of enhanced root metabolism and compound production—must be carefully considered when optimizing composite plant systems for commercial applications. In production settings where root harvest, rhizosecretion, or metabolite enrichment is the primary goal, such reallocation may be advantageous. However, when shoot biomass remains a key target or yield determinant, balancing metabolite output with architectural traits becomes essential.

4.5. A Scalable and Ecologically Relevant Production Platform

The ex vitro transformation method [34] provides a non-sterile, rapid, and low-cost strategy to generate composite plants with transgenic roots and wild-type shoots. This system bypasses the need for tissue culture, allowing high-throughput generation of transformed material using simple materials like rockwool and standard greenhouse infrastructure. Composite plants generated in this way retain full shoot development while enabling genetic manipulation of roots, offering a practical and scalable method for root-specific metabolic engineering under non-sterile conditions.

Our study confirms that this modular setup is particularly suited for controlled-environment agriculture, where spatial efficiency, easy root access, and controlled input systems are essential. As demonstrated in Glycyrrhiza glabra, hydroponically grown composite plants showed enhanced root metabolite production, with additional systemic effects observed in aerial tissues [8]. This reinforces the potential of composite plant systems as a scalable and production-oriented strategy for root-targeted metabolic engineering.

By targeting root-based pathways and transport processes, this approach also opens opportunities for rhizosphere engineering—e.g., enhancing beneficial microbe interactions or suppressing pathogens via targeted exudates. Recent findings indicate that root-exuded artemisinin contributes to A. annua’s defense against pathogens and herbivores, regulates ROS-mediated responses, and modulates microbial interactions through feedback signaling [19]. Enhancing or redirecting such exudation patterns in composite systems could therefore serve not only production goals but also promote sustainable cultivation through natural biocontrol and microbiome management—further underscoring the strategic value of composite plants for rhizosphere-level engineering.

As demonstrated in Datura innoxia co-cultivated with A. rhizogenes in hydroponics, the inoculation of the nutrient solution led to significant increases in both plant biomass and alkaloid production in roots and shoots, with partial root transformation resulting in chimeric, semi-composite plants [7]. This strategy—based on in situ transformation without regeneration—offers a promising route for scalable metabolite production in greenhouse settings, even under non-sterile conditions.

In vertical farming systems, where spatial efficiency, root access, and controlled inputs are critical, composite plants offer a scalable solution for producing high-value bioactive compounds such as artemisinin. Recent reviews highlight that hydroponics and vertical farming systems not only improve spatial and input efficiency but can also enhance metabolite yields and allow year-round production of medicinal species [1].

However, for widespread application, biosafety must be addressed. Residual A. rhizogenes in tissues or nutrient solutions could affect microbiome composition or contaminate recirculating systems. Future applications should employ self-limiting strains and include post-inoculation decontamination protocols to meet regulatory and ecological standards. Taken together, the ex vitro composite plant system represents a powerful and adaptable tool for both research and production, bridging the gap between gene function studies and commercial-scale bioactive compound manufacturing.

5. Conclusions

This study demonstrates that ex vitro composite A. annua plants, generated by A. rhizogenes transformation, provide a practical tool for enhancing secondary metabolite accumulation under controlled cultivation. In leaves, artemisinin accumulation was increased in a cultivation-dependent, strain-specific manner: strain 2659 was effective under aeroponics, while strain 1523 enhanced accumulation under hydroponics. In roots, strain 2659 significantly increased artemisinin in hydroponics and tended to increase it in aeroponics, whereas strain 1523 showed weaker effects. Importantly, strain 1523 also stimulated artemisinin exudation, particularly in hydroponics, indicating the potential of composite plants to promote metabolite release into cultivation media.

Overall, composite plants represent a scalable and low-cost approach to redirect metabolic activity toward valuable secondary compounds while maintaining total biomass. Their application in hydroponic and aeroponic cultivation, particularly within vertical farming systems, offers promising opportunities both for sustainable enhancement of tissue-based metabolite yields and for continuous, non-destructive recovery of compounds through root exudation.