CO₂ Enrichment Alters the Phytochemical Composition of Centella asiatica: GC-MS Analysis

Abstract

Centella asiatica (L.) Urban is a medicinal herb containing valuable bioactive compounds widely used in pharmaceutical, cosmetic, and traditional medicine applications. This study investigated the effects of elevated CO₂ levels (1000, 800, and 600 ppm compared to ambient ~420 ppm) on secondary metabolite composition in C. asiatica using GC-MS analysis of ethyl acetate extracts. Significant treatment effects (p < 0.001) were observed across nine identified compounds, with α-copaene showing the most pronounced response. At 1000 ppm CO₂, sesquiterpene hydrocarbons, including α-copaene (10.60%) and trans-caryophyllene (8.97%), reached their highest concentrations, representing 232% and 413% increases over ambient conditions, respectively. Germacrene D demonstrated optimal synthesis at 800 ppm (8.12%) while remaining undetectable under ambient conditions. In contrast, the diterpene neophytadiene (16.84%) and the oxygenated sesquiterpene caryophyllene oxide (11.27%) exhibited maximum concentrations under ambient conditions. Principal Component Analysis confirmed distinct metabolic profiles, with the first two components explaining 84.38% of the total variance. Correlation analysis revealed strong positive relationships (r > 0.85, p < 0.001) between structurally related sesquiterpenes. These findings establish a foundation for optimizing cultivation conditions to enhance specific bioactive compound production in C. asiatica, with potential applications in pharmaceutical production systems targeting sesquiterpene-derived medicines. The research demonstrates that atmospheric CO₂ modulation offers a promising strategy for targeted enhancement of secondary metabolite synthesis, though further investigation of molecular mechanisms and environmental interactions is necessary for commercial implementation.

1. Introduction

Centella asiatica (L.) Urban, commonly known as Gotu kola, is a medicinal herb belonging to the Apiaceae family. The plant is indigenous to tropical regions of Southeast Asia, including India, Sri Lanka, China, Indonesia, and Malaysia, as well as parts of South Africa and Madagascar [1]. This perennial herb thrives in moist environments and synthesizes valuable triterpenoid compounds with established wound-healing properties, notably asiatic acid, madecassic acid, and asiaticoside [2]. These compounds have been utilized in traditional Asian medicine for centuries, demonstrating significant antioxidant properties that mitigate oxidative stress and reduce chronic disease risk, particularly through bioactive compounds, such as madecassoside [3].

Global demand for plant-derived bioactive compounds continues to increase, driven by sustainability concerns, health consciousness, and environmental considerations [4]. The therapeutic profile of C. asiatica extends to antimicrobial, neuroprotective, and anti-inflammatory properties, enhancing its pharmaceutical significance [5,6,7]. Despite extensive phytochemical and therapeutic characterization of C. asiatica, research examining the effects of carbon dioxide enrichment on its secondary metabolite production, particularly regarding volatile compounds, remains limited [8]. The escalating demand for plant biomass in both traditional and contemporary medicine necessitates the development of sustainable cultivation practices, including environmental carbon optimization to enhance both growth parameters and bioactive metabolite synthesis [9].

Carbon dioxide (CO₂) supplementation represents an essential strategy in greenhouse cultivation systems, addressing daytime atmospheric CO₂ deficiencies that limit plant productivity. Emerging sustainable approaches, including agro-industrial symbiosis systems and carbon capture and utilization (CCU) technologies, offer promising solutions for optimizing agricultural carbon utilization [10]. Recent innovations, such as modified polydimethylsiloxane/polysulfone (PDMS/PSF) composite membranes, have demonstrated significant enhancements in plant growth parameters under controlled CO₂ conditions [11]. Furthermore, integrated control strategies in greenhouse environments effectively optimize light and atmospheric conditions, improving energy efficiency and crop yield [12].

Significant knowledge gaps persist regarding the mechanisms through which carbon dioxide modulation affects essential oil biosynthesis in medicinal plants like C. asiatica, particularly the long-term effects of elevated carbon levels across various environmental stresses and their underlying biochemical pathways [13].

Variations in carbon dioxide concentrations can stimulate the synthesis of secondary metabolites, including triterpenoids, flavonoids, and phenolic compounds, potentially increasing bioactive constituent concentrations under enriched atmospheric conditions [8,14]. Doubling carbon dioxide levels can enhance photosynthetic efficiency by 25–75%, facilitating improved growth parameters and biomass accumulation. Recent advances in LED technology and photoselective films have revolutionized plant cultivation through spectral control, significantly influencing growth, metabolism, and ecosystem dynamics [15,16].

C. asiatica is significantly influenced by environmental factors, including geographical location and climatic conditions. Regional variations in key compounds, such as asiaticosides and madecassosides, indicate that environmental parameters, including carbon dioxide concentrations, play a crucial role in determining the plant’s chemical composition [17,18]. C. asiatica contains diverse bioactive compounds, including phenolic acids, flavonoids, and terpenoids, which contribute to its medicinal properties and demonstrate responsiveness to environmental stress factors, affecting the plant’s overall phytochemical content [19].

Our selection of GC-MS-based metabolomic profiling for this study offers significant methodological advantages. While previous research on C. asiatica has predominantly focused on triterpene saponin extraction [20], our GC-MS approach provides superior separation efficiency for volatile and semi-volatile sesquiterpenes that contribute significantly to the plant’s bioactive profile. The high-resolution capabilities of modern GC-MS systems facilitate the detection of trace compounds that may be responsive to environmental perturbations, yielding a more comprehensive metabolic fingerprint.

This methodology presents analytical constraints that require acknowledgment. GC-MS inherently favors thermally stable, non-polar compounds, potentially underrepresenting polar metabolites such as asiaticoside and madecassoside that constitute important bioactive components in C. asiatica. Furthermore, our relative quantification approach based on peak area percentages, while methodologically robust for comparative analysis across treatments, does not provide absolute quantification of individual compounds—a limitation that could be addressed in subsequent targeted analyses. The prioritization of comprehensive metabolic profiling over targeted quantification was a deliberate methodological decision aligned with the exploratory objectives of this investigation, establishing a foundation for subsequent targeted analyses of CO₂-responsive compounds.

This study employs Gas Chromatography–Mass Spectrometry (GC-MS) analysis to investigate the effects of controlled carbon dioxide levels (1000, 800, and 600 ppm compared to ambient conditions) on secondary metabolite composition in C. asiatica. We hypothesize that elevated CO₂ levels will significantly alter the production of key bioactive compounds, particularly sesquiterpenes and related secondary metabolites, in a concentration-dependent manner. The specific aims are to (1) characterize the chemical profile of C. asiatica extract under varying CO₂ conditions, (2) identify which compounds show enhanced synthesis at specific CO₂ concentrations, and (3) determine the optimal cultivation conditions for maximizing the production of target bioactive compounds. This approach aims to provide detailed insights into metabolite modulation under different environmental conditions, supporting the development of optimized cultivation strategies for enhanced bioactive compound production.

2. Materials and Methods

2.1. Plant Material and Experimental Design

Centella asiatica (L.) Urban plants were propagated through vegetative reproduction using runners with young shoots and roots. Plant material was collected from an indigenous variety in northeastern Thailand (Nong Bua Lam Phu province, 17°05′60.00″ N, 102°29′59.99″ E) characterized by dark green ovate leaves with crenate margins and thin petioles. This local variety was selected based on its established phytochemical profile and genetic stability. The experiment was conducted at the Agricultural Machinery Research Center and Post-Harvest Technology, Khon Kaen University, during February–April 2024.

Eight-week-old plants were transplanted into plastic pots (15 cm diameter) and arranged according to a Completely Randomized Block Design (CRBD) (Figure 1). This experimental design was selected to control environmental gradients in controlled environment studies while minimizing experimental error. The 60-day study period was established based on previous research that demonstrated that this duration represents the optimal vegetative stage for secondary metabolite accumulation in C. asiatica without confounding senescence effects.

The experiment compared four CO₂ levels, 1000, 800, and 600 ppm, with ambient conditions (approximately 415–420 ppm during the experimental period) serving as the control. These concentration ranges were selected based on previous studies that demonstrated significant metabolic responses in medicinal plants within this range [21]. Each treatment comprised 25 plants divided into five blocks of five plants each, yielding 100 total experimental units. This replication level was determined through power analysis (β = 0.85), ensuring adequate statistical power for detecting treatment effects while minimizing resource requirements.

2.2. Environmental Control System Design and Operation

The study was conducted in custom-built sealed chambers (2 × 2 m) constructed with ISO wall panels to ensure standardized insulation properties and minimize environmental fluctuations. Plants were cultivated on elevated tables 80 cm above the floor, with dimensions of 1 × 1 m to eliminate potential confounding effects from vertical CO₂ stratification.

Internal air circulation was maintained using small fans positioned at table corners operating at 1 m/s velocity. This specific configuration was determined through preliminary computational fluid dynamics modeling [22] to be the most efficient for ensuring uniform gas distribution while preventing mechanical stress to the plants, which could induce confounding metabolic responses.

The system, controlled by a Mitsubishi FX3U PLC (Mitsubishi Electric, Bangkok, Thailand), monitored CO₂ levels at 5 min intervals using calibrated non-dispersive infrared (NDIR) sensors (accuracy ± 2%, precision ± 5 ppm). When concentrations fell below target thresholds, the system automatically activated gas valves for 5 s intervals, followed by mixing periods until desired ranges were achieved. This dynamic control approach prevented overshooting target concentrations while maintaining stable atmospheric conditions. CO₂ treatments were administered during the 14 h photoperiod to coincide with active photosynthesis, a protocol established for optimizing plant metabolic responses to carbon enrichment.

To minimize potential position effects, a recognized source of experimental bias in controlled environment studies, plants were systematically rotated within blocks every three days. Additionally, to address potential chamber effects, environmental parameters were continuously monitored and cross-validated among chambers using calibrated sensors, with variation coefficients maintained below 5% throughout the experimental period.

2.3. Growth Conditions and Substrate Properties

The controlled environment chambers were maintained at 27 °C with 75% relative humidity (RH). These parameters were selected based on previous research that identified this temperature and humidity range as optimal for secondary metabolite production in C. asiatica. Photosynthetic photon flux density (PPFD) was set at 300 µmol m⁻²·s⁻¹ with a 14 h photoperiod following established recommendations for optimizing photosynthetic efficiency while avoiding photooxidative stress that could confound metabolic responses to CO₂ treatments.

Plants were grown in loam soil with the following characteristics: pH 6.87, electrical conductivity (EC) 0.22 dS/m, and soil organic matter (SOM) 2.13%. The soil’s pH was selected based on research that demonstrated optimal bioactive compound synthesis in C. asiatica within the pH range of 6.5–7.0. The soil’s nutrient composition consisted of 0.05 mg/kg nitrogen (N), 4.26 mg/kg phosphorus (P), and 63.32 mg/kg potassium (K). This relatively low nitrogen level was deliberately maintained following previous findings that showed that elevated nitrogen can mask plant metabolic responses to CO₂ enrichment [23].

To eliminate water stress as a confounding variable, plants were irrigated daily with tap water (EC 0.18 dS/m, pH 7.2) until reaching field capacity, as determined through gravimetric analysis. Irrigation timing was standardized to morning application (0800–0900 h) to minimize diurnal variations in water relations.

2.4. Sample Preparation and Extraction Method

Plant samples were systematically collected from the basal region upward, accounting for potential variations in metabolite concentrations along the plant axis. All sampling was conducted between 0900 and 1100 h to minimize diurnal variations in metabolite concentrations.

Harvested plant material was immediately flash-frozen in liquid nitrogen to prevent enzymatic degradation of volatile compounds and then lyophilized at −50 °C and 0.05 mbar for 48 h. This preservation method was selected based on comparative analysis, which demonstrated superior retention of volatile components compared to oven-drying methods.

Dried C. asiatica leaf powder (130 g) was extracted with ethyl acetate (500 mL) using mechanical agitation for 3 h. The selection of ethyl acetate as the extraction solvent was based on previous research that demonstrated its superior efficiency for extracting semi-polar secondary metabolites, particularly sesquiterpenes and related compounds, from medicinal plants. This extraction method yields a complex mixture of plant secondary metabolites, including terpenoids, phenolics, and flavonoids, rather than isolating specific essential oil components.

The solution was filtered through Whatman No. 1 filter paper, and the solvent was removed through rotary evaporation (40 °C, 175 mbar) to dryness. These specific evaporation parameters were selected to minimize thermal degradation of heat-sensitive compounds while ensuring complete solvent removal. The dried extract was reconstituted in 1 mL ethyl acetate to increase the concentration. Subsequently, 100 µL of the concentrated extract was diluted with 900 µL ethyl acetate (10-fold dilution).

The necessity of derivatization prior to GC-MS analysis stems from the inherent properties of many plant secondary metabolites found in C. asiatica. Due to the polar nature of these compounds, chromatographic analysis without derivatization is generally unsatisfactory owing to adsorption and decomposition on the column, resulting in peak tailing and losses. Therefore, derivatization reactions are employed to reduce polarity, improve chromatographic separation of analytes, and increase the selectivity and sensitivity of detection [24]. Previous research has demonstrated that the low volatility and high molecular weight of triterpenes require a derivatization step prior to GC analysis, as derivatization (methylation, esterification, silylation, etc.) modifies the molecular ion M+ and influences chromatographic behavior [25]. Consequently, direct injection of ethyl acetate extracts into GC-MS without derivatization would not be suitable for comprehensive analysis of the polar secondary metabolites present in C. asiatica.

Silylation represents the most widely employed derivatization technique for analyzing plant secondary metabolites due to its effectiveness in replacing active hydrogen atoms in hydroxyl, carboxyl, and amine groups with trimethylsilyl (TMS) groups. This chemical modification significantly reduces molecular polarity, increases thermal stability, and improves chromatographic separation. The derivatization procedure followed optimized protocols established for plant terpene analysis [26,27]. Specifically, 100 μL aliquots of the diluted extract were evaporated to dryness under a gentle nitrogen stream. The dried residue was then treated with 50 μL N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) containing 1% trimethylchlorosilane (TMCS) as a catalyst. The mixture was thoroughly vortexed and heated at 70 °C for 30 min, conditions that have been demonstrated to achieve maximum derivatization efficiency for both monoterpenes and sesquiterpenes.

The inclusion of TMCS as a catalyst serves multiple functions in the silylation reaction. It acts as a proton acceptor and helps remove water formed during the reaction, thereby shifting the equilibrium toward complete derivatization. Following the reaction, samples were cooled to room temperature and filtered through 0.22 μm PTFE syringe filters to remove any particulates. The derivatized samples were analyzed within 12 h of preparation, as TMS derivatives maintain optimal stability under these storage conditions [28]. This comprehensive derivatization approach ensures reliable and reproducible analysis of the diverse range of bioactive compounds present in C. asiatica extracts.

2.5. Analysis of Plant Secondary Metabolites

The phytochemical compositions in plant extract were analyzed using an Agilent 6890N Series GC system coupled with a 5973 mass selective detector and auto sampler (Central Laboratory (Thailand) Co., Ltd., Bangkok, Thailand). Separation was performed on a DB-5MS capillary column (30.0 m × 0.25 mm i.d., 0.25 µm film thickness) containing 5% phenyl and 95% dimethyl-polysiloxane fused-silica. High-purity helium was used as the carrier gas at a constant flow rate of 1 mL/min.

The GC oven temperature program was initiated at 70 °C and then increased to 160 °C at a rate of 2 °C/min, followed by a further increase to 220 °C at 2 °C/min with a final hold time of 10 min, resulting in a total run time of 85 min. The injection was performed in split mode with an injection volume of 2 µL. The injector temperature was maintained at 230 °C.

Mass spectrometry was performed using an Agilent 6890N GC coupled with a 5973 MSD (Agilent Technologies, Santa Clara, CA, USA) with electron ionization (EI) mode at 70 eV. Mass spectra were recorded over the range of m/z 35–550 amu at a scan rate of 2.76 scans/s. The interface and ion source temperatures were maintained at 280 °C and 230 °C, respectively. The analysis was conducted at Central Laboratory (Thailand) Co., Ltd., Bangkok, Thailand.

For quantitative analysis, a mixture of n-alkane standards (C8–C40) was used as the internal reference for retention index calculation. The standard solution was prepared at a concentration of 50 μg/mL in n-hexane. Additionally, toluene-d8 and naphthalene-d8 were used as internal standards at a concentration of 10 μg/mL. The internal standards were added to each sample prior to GC-MS analysis.

Method detection limits (MDL) and limits of quantification (LOQ) were determined according to US EPA guidelines. The limit of detection (LOD) was calculated as 3 times the standard deviation of 7 replicate analyses of a low-level standard, while LOQ was set at 10 times the standard deviation. The signal-to-noise ratio (S/N) threshold was set at ≥3 for LOD and ≥10 for LOQ. The calculated LOD and LOQ values ranged from 0.05 to 0.15 μg/mL and 0.15 to 0.45 μg/mL, respectively, depending on the compound.

Compound identification was performed by comparing mass spectra with library data considering only matches with quality percentages above 90%. Chemical composition was based on retention times and retention indices calculated relative to n-alkanes.

2.6. Statistical Analysis

Statistical analyses utilized various methods to assess differences among CO₂ levels and plant extract chemical constituents. A one-way Analysis of Variance (ANOVA) was performed at the 95% confidence level, followed by Duncan’s Multiple Range Test for mean separations. This approach was selected based on established recommendations for analyzing multi-component chemical data.

Additionally, Tukey’s Honestly Significant Difference (HSD) test was employed for multiple comparisons when evaluating data presented in

Table 1. Statistical analysis of phytochemical composition in C. asiatica extract under different CO₂ concentrations.

Chemical Components	1000 ppm	800 ppm	600 ppm	Ambient	%CV
α-cubebene	0.83 ± 0.12 a	ND	0.17± 0.05 c	0.45 ± 0.08 b	8.65
α-copaene	10.60 ± 1.23 a	1.20 ± 0.15 d	2.61 ± 0.31 c	3.19 ± 0.42 b	12.34
epi-bicyclosesquiphellandrene	3.93 ± 0.58 a	0.33 ± 0.04 c	0.84 ± 0.11 b	1.02 ± 0.15 b	65.36
trans-caryophyllene	8.97 ± 0.89 a	6.34 ± 0.76 b	6.45 ± 0.68 b	1.75 ± 0.25 c	10.82
α-humulene	7.26 ± 0.85 a	4.16 ± 0.52 b	4.65 ± 0.48 b	1.55 ± 0.21 c	11.45
1H-cycloprop[e]azulene	5.07 ± 0.63 a	1.10 ± 0.18 c	1.15 ± 0.20 c	1.46 ± 0.23 b	9.78
germacrene D	4.72 ± 0.55 b	8.12 ± 0.92 a	4.35 ± 0.51 b	ND	13.21
caryophyllene oxide	7.01 ± 0.82 b	ND	1.27 ± 0.19 c	11.27 ± 1.31 a	14.56
neophytadiene	12.35 ± 1.45 b	12.90 ± 1.52 b	9.85 ± 1.12 c	16.84 ± 1.89 a	11.93

Note: Values are mean ± SE (n = 3). Different superscripts within rows indicate significant differences at p < 0.05 (Tukey’s HSD test). ^a–d Different superscripts within rows indicate significant differences at p < 0.05 (Tukey’s HSD test). ND = not detected. %Match = similarity match to NIST mass spectral library (only compounds with %Match > 90% are included). %CV = coefficient of variation.

, providing a more conservative assessment of significant differences between treatment means. Regression analysis was used to ascertain association trends, employing both linear and polynomial models where suitable.

For multivariate analysis, data preprocessing included standardization using Z-score transformation (Z = (X − μ)/σ) to ensure consistent scaling across variables. Principal Component Analysis (PCA) was employed to explore relationships among chemical constituents under different CO₂ conditions, with components selected using Kaiser’s criterion. Prior to analysis, sampling adequacy was confirmed through Kaiser–Meyer–Olkin and Bartlett’s tests.

3. Results

The results of this study examining the effects of various CO₂ concentrations (1000, 800, 600 ppm, and ambient conditions) on the chemical composition of Centella asiatica extracts are presented in three main sections.

3.1. Chemical Characterization and Analysis of Plant Extract Components in Centella asiatica

The chemical characterization of C. asiatica extracts through GC-MS analysis revealed an array of bioactive compounds responsive to varying atmospheric CO₂ levels. The chromatographic profiles demonstrated distinct variations in metabolite composition and relative abundances across different CO₂ treatments. Compounds were identified based on their mass spectral patterns and retention times, with identification confidence set at ≥90% match to standard libraries. This comprehensive analysis focuses on key metabolites, including terpenes, phenolic compounds, and other secondary metabolites, that play crucial roles in the plant’s biochemical response to environmental conditions.

As shown in Figure 2, the GC-MS chromatographic profiles demonstrate distinct variations in compound distribution across four CO₂ conditions (1000, 800, 600 ppm, and ambient). The chromatograms reveal clear differences in both peak patterns and intensities, with the majority of well-resolved peaks appearing in the retention time range of 18–40 min, corresponding to the elution of sesquiterpene and related compounds with high library match quality (≥90%).

Visual comparison of the chromatographic profiles reveals CO₂-dependent patterns in metabolite accumulation. The 1000 ppm CO₂ treatment (CO₂ 1000) displays the highest overall peak intensities, particularly in the 18–35 min region, where sesquiterpene compounds elute. The 800 ppm treatment (CO₂ 800) shows a distinctive prominent peak at approximately 24.45 min, representing enhanced germacrene D production, while maintaining moderate intensities for other compounds. The 600 ppm treatment (CO₂ 600) exhibits an intermediate profile with balanced peak distribution throughout the chromatogram.

In contrast, the ambient CO₂ condition (CO₂ Ambient) presents a markedly different profile characterized by lower overall peak intensities in the sesquiterpene region (18–35 min) but notably enhanced peaks in the later elution time, particularly the prominent peak at 58.44 min corresponding to neophytadiene. This shift in metabolite profile under ambient conditions demonstrates the profound influence of CO₂ availability on secondary metabolite biosynthesis pathways.

The quantitative analysis was based on peak %Area values, representing relative abundances within each chromatogram. The abundance scale (y-axis) comparison across all four conditions clearly shows that CO₂ enrichment enhances overall metabolite production, with peak intensities generally increasing from ambient to 1000 ppm conditions for most sesquiterpene compounds. However, certain compounds like neophytadiene show an inverse relationship, achieving maximum abundance under ambient conditions.

These chromatographic profiles correspond directly to the quantitative data presented in Table 1, where compounds identified with ≥90% match quality are reported with their respective %Area values. The visual differences observed in peak patterns and intensities across CO₂ treatments support the statistical analysis, confirming that atmospheric CO₂ concentration serves as a key regulatory factor in determining both the qualitative composition and quantitative distribution of bioactive compounds in C. asiatica extracts.

Chemical analysis at 1000 ppm CO₂ revealed the following compounds in descending order of concentration (Figure 3): neophytadiene (12.35%), α-copaene (10.60%), trans-caryophyllene (8.97%), α-humulene (7.26%), caryophyllene oxide (7.01%), 1H-cycloprop[e]azulene (5.07%), germacrene D (4.72%), epi-bicyclosesquiphellandrene (3.93%), and α-cubebene (0.83%). All compounds showed ≥90% match to standard libraries. Notably, all sesquiterpene compounds reached their peak concentrations at this CO₂ level, while germacrene D was not detected under ambient conditions.

Chemical analysis at 800 ppm CO₂ revealed distinct compound profiles. Neophytadiene showed the highest concentration (12.90%), followed by germacrene D (8.12%), trans-caryophyllene (6.34%), α-humulene (4.16%), α-copaene (1.20%), 1H-cycloprop[e]azulene (1.10%), and epi-bicyclosesquiphellandrene (0.33%). Notably, germacrene D was detected only at elevated CO₂ levels (800 and 1000 ppm), while α-cubebene was not detected at this concentration.

Compared to 1000 ppm CO₂, most sesquiterpene compounds showed substantial reductions, including α-copaene (88.7% decrease), epi-bicyclosesquiphellandrene (91.6% decrease), 1H-cycloprop[e]azulene (78.3% decrease), trans-caryophyllene (29.3% decrease), and α-humulene (42.7% decrease). In contrast, germacrene D reached its peak concentration at 800 ppm (8.12%), representing a 72% increase from its 1000 ppm level (4.72%), while neophytadiene remained relatively stable across both conditions.

Chemical analysis at 600 ppm CO₂ revealed intermediate metabolite concentrations between ambient and elevated CO₂ conditions. Compounds were detected in descending order, including neophytadiene (9.85%), trans-caryophyllene (6.45%), α-humulene (4.65%), germacrene D (4.35%), α-copaene (2.61%), caryophyllene oxide (1.27%), 1H-cycloprop[e]azulene (1.15%), epi-bicyclosesquiphellandrene (0.84%), and α-cubebene (0.17%).

Compared to 1000 ppm CO₂, all sesquiterpene compounds showed reduced concentrations, with α-copaene and epi-bicyclosesquiphellandrene exhibiting the greatest decreases (75.4% and 78.6%, respectively). However, trans-caryophyllene and α-humulene demonstrated relative stability, retaining 71.9% and 64.0% of their 1000 ppm levels, respectively.

Chemical analysis under ambient CO₂ conditions revealed a distinct metabolite profile (Figure 4). Neophytadiene showed the highest concentration (16.84%), followed by caryophyllene oxide (11.27%), both reaching their maximum levels under ambient conditions. Other compounds were detected at lower concentrations, including α-copaene (3.19%), trans-caryophyllene (1.75%), α-humulene (1.55%), 1H-cycloprop[e]azulene (1.46%), epi-bicyclosesquiphellandrene (1.02%), and α-cubebene (0.45%). Notably, germacrene D was not detected under ambient conditions.

Compared to 1000 ppm CO₂, neophytadiene increased by 36.4%, while caryophyllene oxide showed a 60.8% increase. In contrast, all sesquiterpene hydrocarbons showed marked decreases, including trans-caryophyllene (80.5% reduction), α-humulene (78.7% reduction), α-copaene (69.9% reduction), and 1H-cycloprop[e]azulene (71.2% reduction).

3.2. Statistical Analysis of Phytochemical Composition Under Different CO₂ Conditions

Statistical analysis of C. asiatica metabolite profiles under different CO₂ treatments (ambient, 600, 800, and 1000 ppm) revealed three distinct biosynthetic response patterns. Using ANOVA and Tukey’s HSD tests, we identified compound-specific responses where certain sesquiterpenes were maximized at elevated CO₂, others peaked at intermediate levels, and some metabolites favored ambient conditions. These differential responses provide insights for targeted cultivation strategies to enhance specific bioactive compound production.

GC-MS analysis of ethyl acetate extracts from C. asiatica revealed diverse compound classes, including sesquiterpenes, phenolics, and other secondary metabolites. Compound identification was based on mass spectral patterns, retention times, and retention indices, with acceptance criteria of ≥90% match quality to standard libraries.

One-way ANOVA revealed significant differences in chemical composition among CO₂ treatments for most compounds analyzed (Table 1). The sesquiterpene compounds showed varying degrees of response to CO₂ enrichment, with α-copaene exhibiting the strongest treatment effect (p < 0.001), followed by 1H-cycloprop[e]azulene and caryophyllene oxide. Notably, neophytadiene was the only compound showing no significant response to CO₂ treatments (p > 0.05), suggesting that its biosynthesis is regulated independently of carbon dioxide availability. This finding contrasts with the sesquiterpene compounds, which demonstrated clear responsiveness to elevated CO₂ conditions, indicating compound-specific metabolic regulation in C. asiatica.

Tukey’s HSD post hoc analysis revealed three distinct response patterns.

(1)

Maximum at 1000 ppm CO₂: α-copaene, trans-caryophyllene, α-humulene, 1H-cycloprop[e]azulene, and epi-bicyclosesquiphellandrene reached peak concentrations at 1000 ppm.

(2)

Maximum at 800 ppm CO₂: Germacrene D showed the highest concentration at 800 ppm and was not detected under ambient conditions.

(3)

Maximum under ambient conditions: Caryophyllene oxide and neophytadiene showed the highest concentrations under ambient conditions.

Notably, caryophyllene oxide was not detected at 800 ppm. Differences in degrees of freedom (df = 2,6 vs. 3,8) reflected varying numbers of detectable measurements across treatments. For example, α-cubebene was not detected at 800 ppm, while germacrene D was absent under ambient conditions. Coefficient of variation values ranged from 8 to 15% for most compounds, except for epi-bicyclosesquiphellandrene (65.36%).

3.3. Multivariate Analysis of Metabolite Profiles Under Different CO₂ Conditions

Multivariate analysis was conducted using correlation heatmap analysis and Principal Component Analysis (PCA) to examine metabolic responses to varying CO₂ conditions. The heatmap revealed correlations between compounds sharing biosynthetic pathways, while PCA identified primary factors driving metabolic variation across CO₂ treatments, providing insights into coordinated biochemical responses to carbon enrichment.

3.3.1. Correlation Analysis of Chemical Components

Correlation analysis was performed to examine relationships between chemical components under different CO₂ conditions. Pearson correlation coefficients were calculated based on relative abundances (%Area) across all treatments and visualized as a heatmap to identify compounds with similar response patterns and potential shared biosynthetic pathways.

The correlation analysis of nine high-confidence compounds (%Match > 90%) revealed distinct biosynthetic groupings under different CO₂ conditions (Figure 5). Strong positive correlations (r > 0.9, p < 0.001) were observed among structurally related sesquiterpenes, (1) α-copaene, epi-bicyclosesquiphellandrene, and 1H-cycloprop[e]azulene (r = 0.99–1.00) and (2) trans-caryophyllene and α-humulene (r = 0.98), reflecting coordinated regulation in the farnesyl pyrophosphate synthesis pathway.

Significant negative correlations were found between caryophyllene oxide and its precursors trans-caryophyllene (r = −0.57, p < 0.01) and α-humulene (r = −0.47, p < 0.01), indicating metabolic flux shifts toward oxidation products under elevated CO₂. Neophytadiene showed high positive correlation with caryophyllene oxide (r = 0.97, p < 0.001) but negative correlations with sesquiterpene precursors, suggesting similar CO₂ responses despite different biosynthetic origins. Germacrene D exhibited unique behavior with selective correlations, potentially due to CO₂-specific enzyme regulation, as it was undetectable under ambient conditions.

3.3.2. Principal Component Analysis of Metabolite Profiles

This report examines chemical constituents in plant extract using Principal Component Analysis (PCA) under different CO₂ conditions. The analysis consists of two main sections, the Principal Component Analysis of chemical components and the Score Plot analysis of chemical component distribution.

Principal Component Analysis (PCA) revealed distinct chemical composition patterns under varying CO₂ concentrations. Data suitability was validated using the Kaiser–Meyer–Olkin test (KMO = 0.82) [29] and Bartlett’s test of sphericity (p < 0.001) [30]. Four principal components with eigenvalues > 1 [31] were identified, with PC1 and PC2 explaining 84.38% of the total variance (Figure 6), exceeding the recommended 80% threshold [32]. PC1 accounted for 60.25% of the variance (eigenvalue = 4.82), while PC2 explained 24.13% (eigenvalue = 1.93) [33].

Loading analysis using a |0.7| threshold [34] showed PC1 positively correlated with sesquiterpenes, including α-copaene (0.92), α-cubebene (0.89), and trans-caryophyllene (0.88), which increased under elevated CO₂. Caryophyllene oxide (−0.72) and neophytadiene (−0.65) exhibited negative PC1 loadings, indicating inverse CO₂ relationships [35]. PC2 was characterized by high positive loading for germacrene D (0.85), reflecting its unique response pattern. Bootstrap validation (n = 1000) [36] confirmed component stability with minimal eigenvalue confidence interval overlap [37].

The loading plot (Figure 7) visualizes compound relationships in PC1-PC2 space, capturing 84.38% of the total variance. Vector length indicates contribution magnitude, while direction shows relationships with principal components. Three distinct response groups were identified: (1) compounds optimized at 1000 ppm CO₂ (blue vectors), including α-copaene, α-cubebene, and trans-caryophyllene, showing strong positive PC1 loadings (0.83–0.92) and coordinated biosynthetic regulation; (2) germacrene D (orange vector) with unique PC2 loading (0.85), reflecting peak accumulation at 800 ppm CO₂ and independent regulatory mechanisms; and (3) compounds favored under ambient conditions (red vectors), including caryophyllene oxide and neophytadiene, exhibiting negative loadings on both components (−0.55 to −0.72) and indicating inverse relationships with CO₂ concentration. The vector orientations demonstrate how CO₂ levels affect both individual compounds and overall metabolic coordination.

Principal Component Analysis revealed clear separation of samples by CO₂ treatment (Figure 8), with PC1 explaining 60.25% and PC2 explaining 24.13% of total variance (cumulative 84.38%). Sample distribution showed distinct clustering, as 1000 ppm CO₂ samples (positive PC1) were characterized by high levels of α-copaene (10.60%), trans-caryophyllene (8.97%), α-humulene (7.26%), 1H-cycloprop[e]azulene (5.07%), epi-bicyclosesquiphellandrene (3.93%), and α-cubebene (0.83%). The 800 ppm CO₂ samples (positive PC2, negative PC1) contained maximum germacrene D (8.12%) and high neophytadiene (12.90%), with α-cubebene and caryophyllene oxide not detected. Samples at 600 ppm CO₂ (quadrants 3–4) showed intermediate levels of trans-caryophyllene (6.45%), α-humulene (4.65%), germacrene D (4.35%), and α-copaene (2.61%). Ambient samples (negative PC1 and PC2) were dominated by neophytadiene (16.84%) and caryophyllene oxide (11.27%).

4. Discussion

Our comprehensive metabolomic approach revealed how carbon dioxide enrichment simultaneously affects multiple biosynthetic pathways in Centella asiatica. Analysis under varying carbon dioxide conditions showed significant changes in key secondary metabolites, particularly at 1000 ppm, where α-cubebene and trans-caryophyllene accumulation was significantly higher than in other conditions (p < 0.001). These findings align with previous research [38] demonstrating that elevated carbon dioxide levels enhance photosynthesis and increase biomass accumulation in plants. Similarly, studies on Coriandrum species [39] have shown enhanced secondary metabolite production under enriched atmospheric conditions.

However, response patterns varied among different compounds. Germacrene D exhibited a parabolic response pattern, reaching its maximum concentration at 800 ppm, while neophytadiene and caryophyllene oxide showed the highest concentrations under ambient conditions. These findings correspond with previous research [40,41] identifying species-specific responses to carbon dioxide enrichment in plants. Studies on various medicinal plants [42,43] have further demonstrated the complexity of these effects, which depend on environmental factors and plant species, potentially due to species-specific adaptation mechanisms. Research has shown [44] that variations in carbon dioxide levels differentially affect plant growth and development across species.

Principal Component Analysis (PCA) confirmed these differential response patterns, with the first two principal components explaining 84.38% of the variance. This aligns with previous research [45] showing a 33% increase in photosynthesis under elevated carbon dioxide conditions. However, studies [46,47] have cautioned that excessive carbon enrichment might adversely affect metabolism and nutrient absorption, necessitating the optimization of environmental conditions for target compounds. These findings correspond with observed species-specific physiological adaptations [48] under controlled environments ranging from 400 to 1000 μmol⋅mol⁻¹.

Carbon dioxide control is crucial for enhancing plant growth efficiency and compound synthesis. Research has shown [49] that increased carbon dioxide levels significantly enhanced plant productivity, with optimal efficiency at approximately 500 μmol·mol⁻¹. This corresponds with findings [50] that elevated carbon dioxide at 1200 ± 50 μL/L significantly increased the net photosynthetic rate, intercellular gas concentration, and dry matter accumulation. Furthermore, studies have demonstrated [51] that optimizing environmental carbon levels could significantly improve net photosynthetic rates. These findings support our experimental results showing increased key compound accumulation in C. asiatica under controlled atmospheric conditions.

In conclusion, this study demonstrates the potential of carbon dioxide control for enhancing key compound production in Centella asiatica, particularly sesquiterpenes, while emphasizing the need for compound-specific environmental optimization. Our analysis of the complex mixture of secondary metabolites provides insights into the overall metabolic response of C. asiatica to carbon enrichment, although future studies might focus on specific compound classes using targeted extraction methods. This aligns with research [52] recommending the incorporation of light and other environmental factors. Future research should analyze the molecular-level control mechanisms of these compounds’ synthesis under various atmospheric conditions and examine interactions with other environmental factors to determine optimal conditions for each key compound’s production, following established genomic research approaches [53].

In Centella asiatica, known for its medicinal properties and specific metabolic profile, changes in carbon dioxide levels can affect the concentration of key metabolites, such as asiaticoside and madecassoside, which are influenced by cultivation conditions [54]. Enhanced carbon dioxide leads to increased photosynthetic rates and nitrogen absorption, promoting growth without significant photosynthetic adaptation [55].

Elevated carbon dioxide generally affects plant metabolism by modifying photosynthesis, nutrient absorption, and metabolic pathways. It can lead to reduced photorespiration, affecting nitrogen uptake, as observed in studies of other plants, where C/N ratios changed due to altered carbon and nitrogen absorption dynamics [56]. Furthermore, increased carbon levels can influence the expression of genes related to sugar and starch metabolism, as observed in oriental melon seedlings, where increased root-zone carbon altered sugar and starch contents and metabolic activities [57]. Similar effects on nutrient absorption and metabolism have been demonstrated in common beans, where elevated carbon dioxide modified iron metabolism and reduced tissue nutrient accumulation [58].

While direct studies of C. asiatica under elevated carbon dioxide conditions are limited, these insights suggest that similar metabolic adjustments may occur, potentially affecting growth, nutrient absorption, and secondary metabolite production. Understanding the specific response of C. asiatica to increased carbon dioxide requires targeted studies focusing on its unique metabolic pathways and potential impacts on medicinal properties.

Sesquiterpene biosynthesis is a complex process regulated by various environmental and genetic factors, particularly carbon dioxide levels. Studies in Atractylodes lancea have demonstrated significant enhancement of sesquiterpenoid synthesis under mild shade conditions, affecting photosynthetic efficiency and intercellular carbon concentration [59]. This process is supported by the mevalonate (MVA) pathway, which is the primary pathway for sesquiterpene biosynthesis; it is influenced by carbon availability through its effect on acetyl-CoA production [60].

The regulation of sesquiterpene biosynthesis involves gene expression and enzymatic control mechanisms. Sesquiterpene synthase gene expression can be modulated by environmental factors, including carbon dioxide levels, as observed in Aquilaria sinensis [61]. In the broader context of plant adaptation and stress response, environmental carbon influences the expression of key enzymes and transcription factors associated with the MVA pathway [62].

The reduction of caryophyllene oxide under enriched atmospheric conditions involves multiple interconnected physiological and biochemical factors. Enhanced carbon dioxide increases photosynthesis and carbon fixation, leading to altered metabolic profiles [63]. These conditions influence plant metabolic profiles, resulting in increased carbohydrate and starch content while reducing nitrogen content in plant tissues [64]. Additionally, reduced oxidative stress under high carbon conditions affects the production of secondary metabolites, including terpenes [65].

Neophytadiene, a volatile organic compound found in plants, shows decreased levels under elevated carbon dioxide conditions due to multiple physiological and biochemical changes. Long-term exposure to high-carbon environments leads to reduced proteins associated with energy production pathways, such as the Calvin cycle, resulting in decreased photosynthesis and respiration in phytoplankton [66]. In terrestrial plants, increased carbon dioxide triggers photosynthetic acclimation, characterized by reduced leaf photosynthetic capacity influenced by nitrogen use efficiency [67]. These metabolic changes affect electron transport rates and photosystem II efficiency, influencing plant metabolic pathways and volatile compound production [68,69].

The regulation of sesquiterpene biosynthesis by carbon dioxide is thus a multifaceted process involving photosynthetic efficiency, precursor availability, and gene expression, highlighting the complex interactions between environmental conditions and plant metabolic pathways.

Studies have shown that essential oil composition under elevated carbon conditions is not static. In aromatic alpine plants like Angelica glauca and Nardostachys jatamansi, increased carbon dioxide levels alter secondary metabolite composition, including monoterpenes and sesquiterpenes [70]. Seasonal variations affect these changes, as demonstrated in Piper lhotzkyanum, where essential oil composition exhibits temporal and seasonal fluctuations [71].

The plant’s developmental stage also influences responses to environmental carbon. Glycine max seedlings show increased photosynthesis and biomass under elevated carbon conditions [72], while reproductive-stage plants like strawberries and tomatoes exhibit enhanced growth and development [73]. However, long-term exposure may reduce photosynthetic capacity through the downregulation of stomatal and mesophyll conductance [74].

The integration of carbon dioxide enrichment with SFE technology could create sustainable and efficient systems for large-scale essential oil production, although economic and technical barriers need to be addressed [75].

Environmental carbon management combined with ecological agricultural practices, as seen in Chinese medicine sectors, emphasizes carbon source reduction and sink efficiency, promoting low-carbon, high-efficiency farming models [76]. This approach supports carbon neutrality goals while improving overall herb quality and reducing environmental impact. Furthermore, supercritical fluid extraction aligns with increasing demand for sustainable and health-promoting products, as demonstrated in Cannabis sativa applications [77].

Controlled environment systems, such as vertical farms, enable the precise management of environmental factors, including carbon dioxide, to optimize bioactive phytochemical concentration and yield in medicinal plants [78].

A key limitation of our current study was the absence of phenotypic and growth-related parameters, which would have provided a more comprehensive understanding of how CO₂ enrichment affects not only the chemical composition but also the morphological and physiological aspects of C. asiatica. Future research should incorporate these measurements to establish clearer connections between metabolic changes and plant development under varying CO₂ conditions.

5. Conclusions

This study demonstrated the significant influence of CO₂ concentrations on secondary metabolite composition in Centella asiatica, revealing distinct compound-specific responses that have important implications for commercial cultivation. Under controlled conditions maintained by our custom-built environmental system (CV < 6%), sesquiterpene hydrocarbons, including α-copaene, trans-caryophyllene, and α-humulene, reached maximum levels at 1000 ppm CO₂, while germacrene D peaked at 800 ppm (8.12 ± 0.92%). In contrast, neophytadiene (16.84 ± 1.89%) and caryophyllene oxide (11.27 ± 1.31%) showed the highest concentrations under ambient conditions. Principal Component Analysis confirmed these treatment-specific responses, with PC1 and PC2 explaining 84.38% of total variance and revealing clear clustering patterns among different CO₂ treatments.

These findings have direct practical applications for optimizing C. asiatica cultivation strategies based on target metabolite profiles. For pharmaceutical applications requiring sesquiterpene-rich extracts, cultivation at 1000 ppm CO₂ would be optimal, while the production of neophytadiene and caryophyllene oxide can be achieved more cost-effectively under ambient conditions. Intermediate CO₂ enrichment at 800 ppm offers the best conditions for germacrene D production. The economic viability of CO₂ enrichment strategies depends on multiple factors, including infrastructure costs, energy requirements, market value differentials between specific metabolites, and production scale. While small-scale, high-value pharmaceutical production may justify the investment in CO₂ enrichment systems, large-scale cultivation for general extracts requires careful cost–benefit analysis.

The differential metabolic responses observed in this study suggest underlying regulatory mechanisms that warrant further investigation through targeted molecular approaches. Future research should focus on understanding the long-term stability of metabolite profiles under different CO₂ regimes, developing cost-optimization strategies for commercial implementation, investigating potential interactions between CO₂ and nutrient availability, and establishing rapid screening methods for quality control. These advances will enable more precise manipulation of biosynthetic pathways for enhanced bioactive compound production in C. asiatica and related medicinal plants, ultimately supporting the development of sustainable and economically viable cultivation practices for this important medicinal species.