Chemometric Analysis of GC-MS Chemical Profiles and Biological Activities of Three Citrus Essential Oils in Indonesia

1. Introduction

Citrus species, part of the Rutaceae family, are globally renowned for their popular fruits and diverse health benefits. Among them, sweet oranges (Citrus sinensis) play a significant role, constituting approximately 70% of the world's citrus production and consumption. Alongside sweet oranges, other widely cultivated and consumed Citrus species include tangerines or mandarins (Citrus reticulata), grapefruits (Citrus vitis), limes (Citrus aurantifolia), and lemons (Citrus limon) [1].

Indonesia has a substantial Citrus production of approximately 2.6 million tons per year, encompassing 255 different varieties, including sweet (C. sinensis), mandarin (C. reticulata), sour (C. aurantium), pomelo oranges (C. maxima), tangerines (C. nobilis), lemons, limes, etc [2,3]. Citrus, beyond its application as a condiment, is utilized in sweet delicacies in European nations, enhancing dishes such as pan-seared chicken with orange-brandy sauce or pork tenderloin with blood oranges [4]. Notably, C. limon and C. hystrix are extensively used as spices in Asian countries due to their distinctive scents and natural oils [5,6].

Citrus essential oils (EOs), obtained through hydro distillation of citrus peels, are rich sources of bioactive compounds with applications in the pharmaceutical and food industries [7]. Comprising a complex mixture of aldehydes, esters, alcohols, ketones, acids, monoterpenes, and sesquiterpenes [8], these EOs exhibit diverse biological activities, including antibacterial, antiviral, fungicidal, and antioxidant properties [9,10]. Beyond their antimicrobial effects, Citrus EOs also serve a crucial role as antioxidative agents, protecting organisms and tissues from damage caused by reactive oxygen species (ROS) [11].

Previous studies have unveiled the phytochemical makeup of essential oils (EOs) derived from Citrus species. Among these components, the pivotal active compound found in Citrus EOs is D-limonene. However, its concentration significantly fluctuates across Citrus species, spanning from 34.2% to 81.9% of the total compounds. Lemon essential oil exhibits stronger antioxidant activity, as assessed by ABTS and DPPH free radicals, compared to other Citrus EOs [12]. The major components of this EO are Limonene (67.1%), α-terpinene (8.0%), and α-pinene (11.0%) [13]. In another study, C. limon EO also demonstrated significant inhibition in the growth of Solanum lycopersicum and Lepidium sativum germinating seeds at a level of 100 µg/mL in comparison with C. myrtifolia and C. bergamia EOs [14]. Additionally, this EO was reported to exhibit moderate toxicity in animal gavage with 500 and 100 mg/kg doses at the sub-chronic stage [15].

Nevertheless, the precise phytochemical constituents responsible for the biological properties of Citrus EOs, especially across diverse Citrus species in Indonesia, remain unknown. This study aims to determine these associations by examining the chemical composition of EOs and their corresponding antioxidant, allelopathic, and cytotoxic characteristics, utilizing principal component analysis (PCA) and hierarchical cluster analysis (HCA). The significance of this study extends beyond uncovering the therapeutic potential inherent in Citrus EOs; it also delves into exploring their promising applications across agricultural, medicinal, and interdisciplinary domains.

2. Materials and Methods

2.1. Plant Materials and EO Extraction

The fruits of C. sinensis, C. limon, and C. hystrix were purchased commercially (Jogjakarta market). A total of 5 kg of each Citrus species was peeled manually and the peels were air dried for 12 hours before being extracted by a Clevenger hydro distillation system. Lactuca sativa and Raphanus sativus seeds were obtained commercially (PT. Panah Merah, Purwakarta, Indonesia). These seeds were tested for germination and more than 90% were alive before being used in a plant inhibitory assay.

2.2. DPPH Radical Scavenging Activity

The method described by Andriana et al. [16] was employed to assess the DPPH free radical scavenging activity of Citrus EOs. The absorbances of samples were measured at 517 nm and expressed as radical scavenging activity percentage.

2.3. Plant Inhibitory Potentials

Plant inhibitory assay was conducted using moist filter paper placed in a 12 well-plate (22.1 mm in diameter and 35 mm in depth) followed the method reported by Andriana et al. [17]. The seeds of the indicator plants were lettuce (L. sativa) and radish (R. sativus). Plant inhibitory potentials on L. sativa and R. sativus were expressed in inhibitory percentage of germination, shoot, and root size over control.

2.4. Evaluation of Cytotoxicity by the Brine Shrimp Lethality Assay

Cytotoxic activity of Citrus EOs was assessed using the brine shrimp (Artemia salina) lethality bioassay, with 5 mg of each extract dissolved in DMSO and serial dilutions in simulated seawater. After 24 hours, surviving brine shrimp nauplii were counted, and mortality was determined based on the absence of regulated forward movement within 30 seconds [18].

2.5. Total Phenolic Contents

Total phenolic contents in Citrus essential oils were determined using the Folin Ciocalteu (FC) reagent as reported previously [16] and expressed as milligrams of gallic acid equivalent (GAE) per gram of the sample.

2.6. Total Flavonoid Contents

Total flavonoid contents were evaluated based on the method reported previously and expressed as quantified and reported in milligrams of quercetin (QE) per gram of the sample [20].

2.7. Identification of Functional Groups by Fourier-Transform Infrared Spectroscopy (FTIR)

An FTIR analysis was performed to find the chemical functional groups in EO samples following the procedure described previously by Indrianingsih et al. [21]. The spectra were measured by attenuated total reflection (ATR) method and displayed as the percentage of transmittance.

2.8. Identification of chemical Constituents by GC-MS

The phytochemical constituents of Citrus EOs were identified by a GC-MS system (Agilent 7890B / MSD 5977 A, Agilent Technology, Inc., Santa Clara, California, United States) following the previous method [22]. Data peak processing was managed the Agilent Chem Station software, incorporating the NIST mass spectral library (Agilent Technology, CA, USA).

2.9. Statistical Analysis

The data analysis was conducted using MetaboAnalyst 5.0 software (https://metaboanalyst.ca/) with the one-way analysis of variance (ANOVA) method and Tukey's test with a 95% confidence interval (p<0.05) to determine the significance level among samples. A multivariate analysis was employed to reduce dimensional parameters, utilizing a correlation matrix consisting of 9 samples and their replicates as well as 103 variables (GC/MS peak area %) and antioxidant activities. Hierarchical cluster analysis was applied to illustrate clusters and interrelationships between samples, forming the basis for the hierarchical clustering algorithms [23].

3. Result

3.1. Assessment of DPPH Scavenging Activity, Total Phenolic and Flavonoid Contents

The antioxidant activities, total flavonoid, and phenolic contents of essential oils (EO) from three different types of citruses are shown in Table 1. Citrus sinensis EO demonstrated the strongest antioxidant activity compared to the other samples. Conversely, the total flavonoid contents of C. limon EO exhibited the highest value among all samples. There is a relationship between the total phenolic contents and the antioxidant activities of the samples.

3.2. Brine Shrimp Lethality Assay

Table 2 shows the brine shrimp lethality test of EO samples. The C. hystrix EO demonstrated the highest death rate, with an LC₅₀ value of 3188.23 ± 71.89 ppm. On the contrary, C. sinensis and C. limon essential oils exhibit a requirement exceeding 6000 ppm to induce mortality in brine shrimp nauplii.

3.3. Plant Inhibitory Activity

Table 3 illustrates the inhibitory activity of Citrus EOs on the germination and growth of R. sativus and L. sativa. The C. limon EO exhibited a significant inhibitory effect on R. sativus germination and shoot growth. At a concentration of 10,000 ppm, the observed inhibition rates on R. sativus germination and shoot length were the highest, measuring at 53.33 ± 11.55% and 97.41 ± 2.80%, respectively. On the other hand, C. hystrix EO demonstrated a strong inhibitory activity on the root length of R. sativus (95.73 ± 3.65%). The C. sinensis EO completely inhibited the germination and growth of L. sativa at a concentration of 10,000 ppm followed by the C. limon and C. hystrix EOs, respectively.

3.4. Identification of the Functional Group of Citrus EO

Figure 1 shows the FTIR spectra of essential oils from three species of Citrus. According to this figure, there are transmittances in wave numbers 3445, 2902, 2692, 2293, 1713, 1421, and 770 cm^-1. The wavenumber range of 3650-3250 cm^-1 indicated the presence of hydrogen bond (OH) presence in EO samples. C. hystrix EO showed a higher intensity in hydrogen bonds. Furthermore, all samples have a very low transmittance in wavenumber of 2902 cm^-1, indicating the presence of aliphatic compounds in all samples. A slight transmittance intensity was also shown at a wavenumber of 2293 cm^-1, indicating the presence of triple bond carbon because it was followed with spectra between 1600-1300 cm^-1. Moreover, wavenumber 1713 cm^-1 indicated the double bond carbon (C=C), and all samples seem to possess a high-intensity transmittance at 1421 cm^-1 that indicates the presence of Vinyl C-H in-plane bend. The final transmittance detected at 770 cm^-1 showed the content of the phenyl group in the EO samples [24].

3.5. Identification of Phytochemical Constituent of Essential oils of Citrus by GC-MS

Supplementary Material 1 shows the phytochemical constituents of Citrus EOs, while Figure 2 illustrates their GC-MS chromatograms. A total of 103 compounds from various chemical classes were identified in EO samples by a GC-MS system. Of which, D-limonene (27.86%), L-α-Phinene (16.64%), cis-(-)-1,2-Epoxy-p-menth-8-ene (5.24%), and Carvone (4.65 %) were accounted as major components of C. sinensis EO. On the other hand, p-Cymene (5.96%), Ethyl 2-(5-methyl-5-vinyltetrahydrofuran-2-yl) propan-2-yl carbonate (7.02%), Isopulegol (15.95%), trans-Carveol (6.79), and β-Pulegone (6.91) were the dominant compounds containing in C. hystrix EO. While for C. limon EO, L-α-Pinene (27.43%), D-limonene (36.41%), and p-Menthatriene (9.26%) were significant contents. These various chemical components might affect the biological activity of Citrus EO. However, which components contribute to their antioxidant, plant inhibitory, and cytotoxic properties require more investigation.

3.6. Chemometric Analysis based on GC-MS Chemical Profiles Concerning Antioxidant Activity

Figure 3 shows PCA score plots of GC-MS chemical profiles and antioxidant activities as the target cell of classification. PCA successfully reduced variables measured from all samples into two principal components, PC1 of 58.8 % and PC2 of 34.7 %, which explained about 93.1% of the variation in the dataset. Notably, C. limon and C. sinensis EOs exhibited similar antioxidant activity, so they are placed in the same negative ordinate (quadrants II and III).

Figure 4 displays the PCA loading plot based on the chemical profiles and antioxidant activities of EO samples determined by the GC-MS system. The Citrus EO samples were clearly dispersed along the PCs in the PCA plot. D-limonene and α-terpineol were among the most discriminating chemicals on negative PC1, while for negative PC2, they were caryophyllene and (+)-3-carene. Conversely, β-pinene and (-)-spathulenol exhibited high selectivity for positive PC1, however trans-p-mentha-1(7),8-dien-2-ol and trans-verbenol affected the positive PC2.

Figure 5 illustrates the HCA dendrogram, which reflects the relationship between EOs' chemical components and their antioxidant activities. This dendrogram categorized the EOs from Citrus species into two primary clusters. Cluster II grouped EOs from C. sinensis and C. limon, indicating their closeness in phytochemical components and antioxidant activities. In contrast, C. hystrix EO was situated in Cluster I, suggesting distinct properties compared to the other samples.

A heatmap analysis was conducted to assess the relationship between the chemical compounds of EO samples and their biological properties. This visualization, presented in Figure 6, showcased the top 20 compounds present in the EO samples derived from the percentage of peak area in GC-MS analysis, utilizing color intensity to represent their abundance.

In the assessment of antioxidant activity, the EO derived from C. limon exhibited notably high potential as an antioxidative agent, as evidenced by the percentage of DPPH scavenging activity. This observation indicated a dominant presence of antioxidant compounds in C. limon EO compared to the other EOs studied. Among the 20 identified compounds, D-limonene, linalool, α-terpineol, and β-myrcene were most abundant in both C. limon and C. sinensis EOs. These compounds likely contribute significantly to the antioxidant activity observed in these EOs. Nonetheless, further investigations are essential to precisely determine the specific roles of these compounds.

4. Discussion

In the current study, we evaluated the antioxidant, plant inhibitory, and cytotoxic properties of three distinct Citrus essential oils (EOs) originating from Indonesia as well as their phytochemical components. Utilizing chemometric techniques, Hierarchical Cluster Analysis (HCA), and Principal Component Analysis (PCA), we discriminated between EO samples to elucidate their proximity concerning antioxidant activity concerning their phytochemical compositions.

In terms of antioxidant properties, the EO from C. limon exhibited the highest potency at 66.11% DPPH scavenging activity, surpassing the activity levels of C. sinensis and C. hytrix EOs, which recorded values of 36.50% and 4.08%, respectively. This result was similarly reflected in the total flavonoid contents of the EO samples, with C. limon EO showing the highest concentration at 30.56 mg QE/g extract. These observations might be attributed to the notably elevated concentration of antioxidative agents in C. limon EO such as D-limonene compared to the other samples. This result aligned with prior studies that identified D-limonene as the predominant constituent in C. limon peel [25; 26] and leaf EO [27;28]. Notably, D-Limonene stands out as the primary volatile compound in lemon essential oil, at levels usually ranging between 70 and 48% [30]. Subsequent investigations have elucidated the diverse antioxidant properties of D-limonene through in vitro assays, including DPPH, ABTS, FRAP, iron chelating, hydroxyl radical scavenging, and superoxide radical scavenging assays, showcasing its efficacy in reducing reactive oxygen species (ROS) through varied mechanisms. Additionally, in vivo assessments have revealed that D-limonene enhances antioxidant levels and augments the protein expression of inducible cyclooxygenase-2 (COX-2) and in nitric oxide synthase (iNOS) UC rats [31].

In terms of plant growth inhibition, the C. hystrix EO demonstrated the strongest inhibitory activity. Prior research has shown that a methanol extract derived from C. hystrix, at a concentration of 10 mg/ml, effectively inhibited the germination of lettuce seeds and impeded the growth of their roots. This finding suggests the potential use of the extract as a bioherbicide in the future for weed management [32]. In this study, the main phytochemical compounds C. hystrix EO were Ethyl 2-(5-methyl-5-vinyltetrahydrofuran-2-yl) propan-2-yl carbonate (7.02%) and Isopulegol (15.95%). The scientific literature documents the phytotoxic activity of EOs, which manifest in varying degrees of inhibition on seed germination and radical elongation. This activity appears to be caused by monoterpenes, specifically oxygenated compounds, especially ketones, alcohols, aldehydes, and phenols [33].

Furthermore, in the field of cytotoxicity assays, which evaluates the capacity of cytotoxic substances to induce cellular harm or cell mortality, the essential oil derived from C. hystrix exhibited a greater mortality rate in comparison to other citrus oils. This higher cytotoxicity aligns with prior studies that showed C. hystrix EO had cytotoxic activity on Artemia salina and demonstrated a greater cytotoxicity effect on different cell lines including human cervix carcinoma (HeLa), murine melanoma (B16F10), and human lung fibroblast (MRC-5) [34]. Moreover, the current investigation demonstrates that the essential oil of C. hystrix possessed a notably elevated level of Isopulegol (15.95%). Isopulegol is a monoterpene found in different plant species that have been scientifically proven to possess pharmacological properties [35].

In the realm of biological activity, citrus EOs have demonstrated inhibitory effects against several microorganisms. For instance, the EO from C. limon has been shown to inhibit bacteria including Lactobacillus plantarum, Lactobacillus mesenteroides, and Escherichia coli [36]. Similarly, C. sinensis EO exhibits inhibitory activity against Gram-positive bacteria like Staphylococcus aureus, Lactobacillus monocytogenes, and Enterococcus faecium along with some Gram-negative bacteria including Salmonella enteritidis and Pseudomonas aeruginosa [37]. Furthermore, C. hystrix EO, has demonstrated inhibitory effect against various bacteria such as Staphylococcus aureus, Enterococcus faecalis, Escherichia coli, Staphylococcus epidermidis, and Proteus vulgaris [38]. The observed inhibitory effects of Citrus EOs against these bacteria are attributed to specific volatile components such as D-limonene and isopulegol, which are believed to contribute significantly to their antimicrobial properties [39,40].

The present study found the isopulegol and ethyl 2-(5-methyl-5-vinyltetrahydrofuran-2-yl) propan-2-yl carbonate as the major component present in C. hystrix EO. In contrast, a previous study reported different major compounds in the Citrus peel EO are sabinene, β-pinene, citronellal, limonene, terpinen-4-ol, and α-pinene [41]. Limonene was identified as the predominant constituent in nearly all Citrus peel EO samples, except C. hystrix and C. micrantha EOs, which were dominated by β-pinene. The most abundant compounds were monoterpenes, followed by sesquiterpenes and ester. According to another study, the kaffir lime peel EO has three primary chemical components: D-limonene (17.10%), 3-carene (13.77%), and γ-terpinene (12.56%). All three chemicals belong to the monoterpene hydrocarbon group and are characterized by the presence of double bonds, which are known as alkenes [42].

The variations in the primary chemical constituents of essential oils between the current findings and previous studies can be attributed to disparities in the geographical regions where the plants are cultivated, resulting in distinct chemical compositions. The chemical composition of essential oils within a plant is determined by various factors including the plant species, climate, geographical location, season, soil composition, extraction technique, and the specific plant portion used for oil extraction. These variations highlight the importance of considering multiple factors when studying and comparing essential oil compositions from different sources [43].

To understand the relation of antioxidant and phytochemical components of Citrus EOs as well as closeness among samples, principal component analysis (PCA) and hierarchy cluster analysis were employed. PCA and HCA analyses successfully discriminated EO samples into two groups, namely EO with high and low antioxidant activity. EOs of C. limon and C. sinensis were clustered in the same place showing the closeness of both samples. However, C. hystrix EO was put in a different cluster meaning the different properties of this sample to C. limon and C. sinensis EOs. PCA and HCA were widely used to discriminate many samples based on their similarity, for example, to cluster EO of Juniperus rigida [44] and Foeniculum vulgare [45]. In line with the previous study, the present study highlights the power of chemometric analysis to classify and cluster as well as to determine the relationship among observed variables.

5. Conclusions

In the present study, EOs extracted from the peels of three distinct Citrus species from Indonesia, namely C. sinensis, C. limon, and C. hystrix, underwent evaluation for their antioxidant potential, plant inhibitory effects, and lethality properties against brine shrimp. We employed Hierarchical Cluster Analysis (HCA) and Particle Component Analysis (PCA) to gauge the proximity of these samples concerning their antioxidant activity. Among the EOs tested, C. limon EO exhibited the most robust antioxidant activity against the DPPH free radical, while C. sinensis EO demonstrated the highest inhibitory effect on the growth of L. sativa, a vegetable plant. GC-MS analysis unveiled D-limonene as the predominant component in C. limon and C. sinensis EOs, constituting 36.41% and 27.86%, respectively. Meanwhile, C. hystrix EO featured Isopulegol as the major volatile compound, accounting for 15.95% of its composition. PCA and HCA analyses effectively classified the Citrus EO samples into two distinct categories. Specifically, C. limon and C. sinensis were grouped together as samples with high antioxidant activity, whereas C. hystrix EO exhibited lower activity. The clustering in HCA highlighted the proximity between C. limon and C. sinensis EOs. This study underscores the potent antioxidant, plant inhibitory, and brine shrimp cytotoxic effects exhibited by Citrus species EOs, suggesting their potential as sources for natural antioxidatives, herbicidal agents, and pharmaceutical substances. However, further elucidation is necessary to ascertain the specific volatile components responsible for these observed biological activities.