Supercritical CO 2 extraction of Phoradendron leucarpum (American mistletoe): optimizing oleanolic acid and polyphenol yields, monitoring ROS-specific neutralization

Phoradendron leucarpum (Ph. leucarpum) is a hemiparasitic plant native to Mexico that has been used in folk medicine for years to treat cancer, diabetes, and cardiovascular disease in different extraction forms. In this study, the effect of different parameters, including temperature, pressure, and co-solvent, on extract yield obtained by the supercritical CO 2 technique is studied, using 15 conditions defined through a Box-Behnken design. The level of the total polyphenols was studied using the Folin-Ciocalteu assay. High-performance liquid chromatography was performed to identify and quantify oleanolic acid (OA) yielded per condition. Ultimately, a novel technique is employed to evaluate the profile of each extract against physiologically relevant ROS, including singlet oxygen (1 O 2), hydroxyl radical (OH •), hydrogen peroxide (H 2 O 2), and superoxide anion (O 2 • −). Our results reveal optimal extraction conditions for enhanced OA yield and synergistic antioxidant responses of an extract with respect to a standard molecule. The response surface methodology analysis also demonstrated the significant effects of pressure, temperature, and co-solvent on polyphenol levels. Additionally, this work highlights an important disparity with traditional methods (e.g., ABTS and DPPH assays) for antioxidant capacity determination and raises important questions about the typical examination of ROS neutralization in the search for novel antioxidant compounds.

The bioactive compounds extracted from natural sources have gained considerable attention today due to their perceived safety, sustainability, environmental impact, and generally fewer side effects compared to synthetic drugs. Computational evidence suggests that disease genes act as central hubs in biological networks, while the targets of natural products tend to be more peripheral (Dančík et al., 2010). This distinction helps to explain why natural products often exhibit lower toxicity, making them promising candidates for drug discovery.

Among all natural sources, mistletoe has been used for centuries in herbal folk medicine to treat diseases such as hypertension, diabetes, epilepsy, and infertility, in the form of teas, powdered plant material, cold-water macerations, and topical poultices prepared from leaves and young stems. Mopai et al. (2025); Deeni et al. (2013); Szurpnicka et al. (2020). Additionally recent studies have proven some of the potential therapeutic properties of this historical family of plants in cardiovascular disease, cancer and other types of inflammatory-based disorders (Yu et al., 2016; Paller et al., 2023; Olas, 2024; Tunctan et al., 2017; Szurpnicka et al., 2020).

The phytochemical analysis has revealed that, depending on the host, climate, and geographical condition, mistletoe is a notable source of bioactive compounds such as lectins, viscotoxins, phenolic acids, flavonoids, alkaloids, and terpenoids, some of which possess well-documented antioxidant and anti-inflammatory properties (Soursouri et al., 2023; Yousefvand et al., 2022; Ishiwu et al., 2013; Kleszken et al., 2022). However, the majority of the studies are conducted on European and Asian mistletoe; hence, the potential therapeutic property of American mistletoe stayed unexplored. Phoradendron leucarpum (Ph. leucarpum), the most abundant mistletoe species in Mexico and the southern states of the USA, is a hemiparasitic plant dominated by various trees and shrubs, including oak and walnut (Kleszken et al., 2022; Nicoletti, 2023). Due to its hemiparasitic nature, it does not come into contact with the soil and directly feeds on the raw sap of its host. Consequently, its phytochemical composition varies depending on the host and climatic conditions. Given its abundance in these regions and its long-standing use in folk medicine, studying Ph. leucarpum was both scientifically and culturally relevant.

Among the diverse bioactive compounds produced by this mistletoe family, triterpenoids stand out as one of the most prominent metabolites. In particular, oleanolic acid (OA, 3 β -hydroxyolean-12-en-28-oic acid) has been identified as a major triterpenoid in related species of the Phoradendron genus, making it a logical target for extraction optimization in this study. Its pharmacological relevance, combined with its presence in mistletoe species, positions OA as a critical compound to evaluate in the context of bioactive compound extraction Pollier and Goossens (2012); Martín et al. (2014). Additionally, OA has been used to treat a variety of tumor cell lines, including MCF-7 and MCF-7/ADR human breast cancer cells, 1321N1 astrocytoma cell line, hepatocellular carcinoma, HCT-116 colorectal cancer cells, and others (Shanmugam et al., 2014). Liu et al. suggested OA as an inducer of apoptosis in MCs, an expression inhibitor of NF- κ B p65 and NLRP3, and malignant transformation of GES-1 cells through the PI3 K/AKT pathway (Liu et al., 2025).

In addition to triterpenes, polyphenols constitute another abundant and relevant group of target metabolites frequently reported in the mistletoe family. These compounds are distinguished by their multiple phenolic structural units (aromatic ring with hydroxyl groups) and exhibit a broad spectrum of biological activities, including antioxidant, antibacterial, anti-inflammatory, and antitumor effects (Zhang X. et al., 2023; Dilkalal et al., 2024; Yan et al., 2021). For instance, it has been demonstrated that polyphenols could suppress metastasis and angiogenesis in TNBC, leading to inhibition of tumor growth (Li et al., 2024). Moreover, due to their remarkable antioxidant activity, they could help prevent the onset and progression of neurodegenerative diseases (Zhang Y. et al., 2023).

On the other hand, achieving an optimal extract yield, sustainability, and selectivity for the targeted composition are the most important factors that could be controlled by choosing the proper extraction method for bioactive compounds Azmir et al., 2013). Over the years, depending on the nature of the plant and the expected composition yield, various extraction strategies, including maceration, pressing, microwave-assisted, ultrasound-assisted, and supercritical carbon dioxide extraction (SC-CO 2), have been used (Geow et al., 2021; Azmir et al., 2013). SC-CO 2 is a green method with properties between gas and liquid, leading to increased reactions that rarely occur in conventional systems (Martins et al., 2023). Due to the low critical temperature of CO 2 (31 °C), thermosensitive biological compounds could be safely extracted from the main matrix. In addition, by controlling temperature and pressure, the selectivity of the extract composition could be finely tuned (Herrero et al., 2010; da Silva et al., 2016). Moreover, by reaching 200 bar pressure, the density of supercritical fluid CO 2 becomes close to that of hexane, reaching to dissolve the lipophilic compounds (e.g., triterpenes and several polyphenols) without or with a lower amount of the co-solvent (Ampese et al., 2025). This method is particularly suitable for the extraction of triterpenoids such as oleanolic acid (OA), which are lipophilic in nature and require a precise balance of temperature and pressure for optimal recovery.

Although natural compounds commonly targeted in extraction processes might exert their effects at the cellular level through multiple molecular pathways, one of their most notable qualities is their antioxidant capacity. This property plays a crucial role in mitigating biological conditions associated with oxidative stress, a phenomenon characterized by an imbalance in the endogenous antioxidant system Mittler (2016). Chronic oxidative stress is known to trigger inflammation, which in turn contributes to the development of a wide range of pathologies, including cardiovascular diseases, cancer, diabetes, and neurodegenerative disorders (Halliwell, 2006; Pérez-Lebeña and Plou, 2021; de Almeida et al., 2022; Pizzino et al., 2017; Simpson and Oliver, 2020). Therefore, it is essential to conduct a thorough evaluation of such property to better assess its impact under more physiologically relevant conditions.

Currently, a diverse assortment of both in vitro and in vivo analytical methods is used to determine the antioxidant capacity of bioactive compounds, such as ABTS and DPPH assays, which are based on the scavenging of [2, 2 ′ -Azinobis-(3\ngdelimcl)-ethylbenzothiazoline-6-sulfonic acid] (A B T S • +) and the [2,2-di(4-tert-octylphenyl)-1-picrylhydrazyl] (D P P H •) radicals, respectively (Rumpf et al., 2023; Sun et al., 2018). Although these spectrophotometric assays profess reproducibility, speed, and relative sensitivity, the use of synthetic radicals is far from evaluating the real capacity of physiologically relevant ROS and their representation of diverse oxidative stress mechanisms (e.g., 1 O 2, OH •, H 2 O 2, and O 2 • −) (Magder, 2006; Abdal Dayem et al., 2017). Given that physiological ROS participate heterogeneously in various cellular processes and that specific ROS species are linked to particular diseases (Juan et al., 2021), singlet oxygen 1 O 2 is a key mediator in photo-oxidative stress and inflammatory processes, while the hydroxyl radical (OH •) is the most reactive and damaging ROS, capable of directly oxidizing critical biomolecules. Hydrogen peroxide (H 2 O 2) and superoxide anion (O 2 • −) also participate in redox processes and can contribute to the formation of more reactive species. This selection aligns with findings by Buenfil-Chi et al. (2025), which highlighted the importance of these ROS in providing a comprehensive assessment of antioxidant capacity in a physiologically relevant context and which was previously evaluated using a panel of antioxidant standards to establish their specific ROS-neutralization profiles. Additionally, these species address the limitations of traditional antioxidant assays, such as ABTS and DPPH, which do not differentiate between ROS-specific neutralization capacities. By targeting these four ROS, this study aims to offer a more robust and physiologically meaningful evaluation of the antioxidant potential of Ph. leucarpum extracts.

Therefore, in this research, by using the SC-CO 2 extraction method, we identified the optimal conditions of temperature, pressure, and co-solvent to effectively target OA and polyphenols as valuable bioactive compounds. Additionally, we employed the novel ROS-specific neutralization methodology, which was previously evaluated using a panel of antioxidant standards to establish their specific ROS-neutralization profiles Buenfil-Chi et al. (2025). This approach was applied to compare the antioxidant capacity of the bioactive compound extract with that of traditional methods, such as ABTS and DPPH assays. Our findings reveal a significant discrepancy in antioxidant capacity between ROS-specific neutralization and conventional methods, highlighting the importance of using physiologically relevant ROS as a more realistic scenario and demonstrating how slight changes in extract composition could synergistically modulate this neutralization capacity for each ROS species.

Materials

Ph. leucarpum was collected from a local ranch in Nuevo León, Mexico. A voucher specimen of this plant has been deposited in the herbarium of the Universidad Autónoma de Nuevo León, Mexico, under the voucher number 25614. The collection of the plant is compiled with relevant institutional, national, and international guidelines and legislation. Different parts of the plant were separated, and the materials were dried for 15 days in a well-ventilated room, placed in cardboard boxes, and protected from light and humidity. Subsequently, the dried stems were selected and cut into particles of 1–2 mm in size. Standards and reagents were purchased from Sigma-Aldrich.

Supercritical CO 2 extraction (SC-CO 2)

The extraction of air-dried stems of Ph. leucarpum was performed using the SC-CO 2 extraction technique. To identify the optimal treatments for maximizing the antioxidant properties of the final composition, 15 different treatments were designed by Box-Behnken using Minitab Statistical Software (Srivastava et al., 2015). Temperature (40, 55, 70 ° C), pressure (100, 250, 400 bar), and co-solvent flow rate (2, 4, 6 g/min) were considered as variables that could affect the phytochemical yield (Table 1). 20 g of dried Ph. leucarpum was added to an extraction vessel of 100 mL. The procedure starts by pumping the ethanol as co-solvent and CO 2 in a statistical mixture, thereafter heated to the target temperature in a heat exchanger, and pressurized. The Tuned flow was introduced to the extraction vessel and the extract was recovered after completing the extraction time (125 min) through the collector vessel’s valve (Thar-Waters 1000 F, Thar Technologies Inc., Pittsburgh, PA, United States). Throughout the process, the CO 2 flow rate was maintained at 40 mg/mL.

Concentration and dehydration

The obtained extracts by different treatments were centrifuged at 4,000 rpm, filtered through 0.22 μ m Corning syringe filters, and concentrated using nitrogen flow to prevent oxidation. This was followed by 3 days of lyophilization to remove any remaining water content from the extracts. All the extracts were weighed and stored at −20 °C until subsequent characterization.

Identification and quantification of OA

OA was identified using high-performance liquid chromatography (HPLC), equipped with a diode array detector (DAD) and an evaporative light scattering detector system (ELSD) (Agilent, 1200 series). The yield of OA of all samples was quantified using an external standard curve of OA, ranging from 0 to 200 ppm. In order to inject into the HPLC system, the extracted samples were dissolved in HPLC-grade methanol to a final concentration of 300 ppm. The separation was performed in a Zorbax SB-Aq reverse phase column (4.6 mm internal diameter × 150 mm length, 3.5 μ m particle size) with a flow rate of 1 mL/min, an injection volume of 10 μ L, and the column temperature maintained at 30 °C. A) HPLC-grade water with 0.1% formic acid, and B) HPLC-grade acetonitrile with 0.1% formic acid were used as stationary phases in the system. The solvent gradient was started by acetonitrile 70%; it reached to 100% acetonitrile at 10 min and was maintained for six more minutes. It was reduced to 70% after that and maintained this concentration for the next 4 min. Chromatograms were integrated using ChemStation HP-Agilent software for LC.

Total phenolic content

To evaluate the total phenolic content (TPC) of the extract, the Folin-Ciocalteu colorimetric method was applied, according to Domingues et al. (2012). 20 μ L of the diluted sample were oxidized under 100 μ L of Folin-Ciocalteu reagent in distilled water 10% (V/V); the reaction was neutralized with 80 μ L of sodium carbonate 7.5% (W/V). Absorbance was measured at 765 nm using a microplate reader (SynergyTM HT Multi-Detection, BioTek, Inc., Winooski, Vt., United States) following 1.5 h of incubation at 25 °C in the absence of light. Gallic acid was chosen as the standard; TPC was given as mg of gallic acid equivalents (GAE (mg)/1 g of dried extract).

ABTS and DPPH assay

The antioxidant activity of the samples was assayed according to Messaoud et al. (2024) using ABTS • + radical scavenging potential. 20 μ l of dissolved samples in DMSO-distilled water 1:1 was added to 180 μ l of ABTS • + solution (ABTS 7 mM + ammonium persulfate 2.45 mM, adjusted to 0.7 ± 0.02 units of absorbance at 734 nm). It was homogenized by pipetting, and the absorbance was recorded using a microplate reader (SynergyTM HT Multi-Detection, BioTek, Inc., Winooski, Vt., U.S.A.) at 734 nm.

In order to perform the DPPH assay, 20 μ l of dissolved extract in DMSO-distilled water 1:1 was added to 180 μ l of DPPH • stable solution (150 μ M DPPH in 80% methanol). Every 20 min until 120 min, the absorbance was read (Valdez-Morales et al., 2021).

During the procedure, the plate was protected from light. The antioxidant potential of the samples was expressed in Trolox equivalent (TE (mg)/1 g of dried extract).

ROS-specific neutralization capacity

In order to evaluate the ROS neutralization capacity of the extract against the four most important biological ROS (1 O 2, OH •, H 2 O 2, and O 2 • −), a novel technique based on photodynamic ROS generation and spectrofluorometric detection was applied according to Buenfil-Chi et al. (2025). In this technique, various ROS were generated by irradiating the specific types of photosensitizers using an optoelectronic setup developed in our group. In the following, the sensor, which acts as the probe molecule, and the extract, as the antioxidant, compete to neutralize the specific species. The whole process of ROS neutralization was studied indirectly by the fluorometric signal emission of the sensor.

The singlet oxygen sensor green (SOSG), 3 ′ -(p -hydroxyphenyl), fluorescein (HPF), Amplex UltraRed reagent (AR), and MitoSOX Red mitochondrial superoxide were used as the indicators of the presence of 1 O 2, OH •, H 2 O 2, and O 2 • − in the environment accordingly. All the mentioned fluorescent probes were provided by Invitrogen and the working solution was prepared based on the manufacturer’s data sheet. Rose bengal photosensitizer was used to generate 1 O 2, OH •, H 2 O 2 under irradiation at 532 nm. The methylene blue photosensitizer was applied to generate O 2 • − using 660 nm. The preliminary extract stock was prepared in DMSO-distilled water 1:1 to make sure both polar and nonpolar compounds are dissolved homogeneously, while the Standard OA was dissolved in pure DMSO. The samples were simply prepared in polystyrene cuvettes. The concentration of components was coming in the following. The extract concentration was 0.0136 mg/mL, the probe was 0.24 μ M, and the photosensitizers RB and MB were 0.67 μ M and 8.43 μ M, respectively. All the samples were diluted with 0.1 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) up to 2 mL. The same composition excluding the extract was considered as the control. The pH of the samples was adjusted to 7.4, equal to the physiological pH. In order to generate the ROS, each sample was irradiated for 3 min at room temperature. The distinct ROS neutralization potential of the extract against each species was obtained using (Equation 1), where ΔCtrl represents the change in the control intensity before and after irradiation, whereas ΔSample represents the change in the sample intensity before and after irradiation (Equations 2, 3).

Statistical analysis

Statistical analysis was performed using Minitab Statistical Software and MATLAB R2023b. To evaluate the effect of extraction treatment on extract yield, OA, and phenolic content, response surface methodology (RSM) was applied, and the data were fitted to a quadratic polynomial model to evaluate the relationship between the parameters and the response. All the assays were performed in triplicate, and the results were analyzed by ANOVA (analysis of variance) where P < 0.05 was considered significant, whereas effects 0.05 ≤ P < 0.10 were considered slight effects. To compare the differences between the treatments themselves, Tukey’s honestly significant difference with a significance level of α = 0.05 was performed. Principal component analysis (PCA) was also performed on the correlation matrix to explore relationships among he extract content and antioxidant activities, with the biplot constructed using the first two principal components (PC1 and PC2) and the variance explained reported.

Evaluation of the SC-CO 2 extraction process variables and extract yield

The SC-CO 2 extraction yield ranged between 0.1 and 3.08 g per 20 g of the dried plant. Treatment 12 (55 ° C, 100 bar, 2 g/min) with the lowest co-solvent flow rate and pressure, combined with a moderate temperature, yielded the lowest amount of extract, while treatment 8 (40 °C, 100 bar, 4 g/min) with the lowest temperature and pressure and an average co-solvent rate demonstrated the highest extract yield. This trend was aligned with Fahrudin et al. (2025), who reported that the extract yield escalates by increasing the co-solvent rate, up to an optimal point, and over that, the yield will decrease gradually. ANOVA analysis results confirmed that among all the parameters, the effect of co-solvent flow rate and the quadratic effect of temperature showed a slight effect on extraction yield (P < 0.1). It could be due to the change of density of the fluid, leading to an altered extract yield. While other linear, quadratic, or interaction parameters did not show a significant effect on the extract yield (Equation 4) (Table 2).

The treatment variables could strongly affect the extract yield, as well as the chemical nature of the targeted compounds. The solvent density could be modified by tuning the temperature or pressure. In the critical region, the density of the solvent could be easily determined by applying a slight change to the engaged factors. Hence, the response of the extraction yield and also the extract’s content are deeply dependent on the hired variables (Name et al., 2022).

Among all the treatments, treatment 3, with 1.45 g of yield (55 ° C, 400 bar, 2 g/min), and treatment 9, with 1.10 g of yield (55 ° C, 100 bar, 6 g/min), were identified as influential points characterized by large residuals. Notably, although the flow rate in treatment nine was three times higher compared to treatment 3 at the same temperature, treatment 3 exhibited a higher yield. It could be attributed to the high pressure in that treatment, which could lead to an increase in the fluid density, leading to more diffusion into the matrix and dissolving the solutes (Fraguela-Meissimilly et al., 2023). Moreover, it has been reported that, at the higher co-solvent rate, the exposure time of critical fluid with the matrix could be decreased; subsequently, the fluid will leave the extraction vessel unsaturated, leading to a lower amount of extract yield (treatment 9) (Moreira et al., 2015). Hence, such deviation from the model prediction highlights further potential areas for investigation.

Identification and quantification of OA

The presence of OA in the extract was identified at 210 nm (Liu et al., 2016) with a distinct peak appearing with the retention time of 2.6 min, as well as that of the OA standard. The quantification of the compound in each sample was performed by incorporating the area under the curve based on the curve of the external standard (y = 5.1068x + 17.138) with R 2 = 0.99 (Table 3).

The OA yield of each sample varied from 2 to 117 mg/1 g of dried extract. Sample 10, with 12% of OA yield, demonstrated the best treatment to obtain the purest subfraction of this compound (40 °C, 250 bar, 2 g/min). However, the large pentacyclic triterpene hydrocarbon core of OA makes the molecule strongly nonpolar, but the presence of a hydroxyl group on C-3 and a carboxyl group on C-28 gives the molecule a partial polarity in the deprotonated phase. Hence, using ethanol in low level (2 g/min) as a co-solvent, by modifying SC-CO 2 fluid, could fill the polarity gap between the OA and CO 2 (Xu et al., 2015; Liu et al., 2015.).

On the other hand, pressure and temperature are critical parameters, as they could be tuned to adjust the density of the SC-CO 2 fluid, thereby enabling targeting a specific component. At higher pressures, the density of the fluid increases, leading to increasing its strength. Likewise, as the temperature increases, the density of the fluid could drop. Moreover, the results of the model obtained by ANOVA (Equation 5; Table 2) demonstrated that the interaction of temperature and co-solvent could significantly affect the yield of OA (P < 0.05). It could be attributed to the density of the fluid inside the vessel as well as the polarity of SC-CO 2.

Hence, it would be easier to understand why in treatments 2 and 6, no OA was detected. In treatment 2, a CO 2 flow rate of 2 g/min and a pressure of 250 bar at 70 °C results in a significant decrease in fluid density, preventing it from reaching the optimal threshold. Conversely, in treatment 6, increasing the co-solvent flow rate to 6 g/min appears to affect the polarity of the fluid, rendering it unable to dissolve OA effectively.

Interestingly, treatment 9, conducted at a temperature of 55 °C and a pressure of 100 bar, appeared not to achieve the optimized density necessary to yield OA. Despite this, it produced the second-highest amount of OA among all treatments (71.45 mg per 1 g of dried extract), just after treatment 10. Although it was generally expected that increasing the co-solvent flow rate would decrease the yield of nonpolar compounds, in this case, reaching a flow rate of 6 g/min seems to compensate for the lack of flow density and results in an increased yield of OA.

The average OA yield of three central points was 64.46 ± 2.22 mg/1 g of dried extract (Samples 13, 14, and 15) (Table 3). This result not only was aligned with the importance of the density and polarity of the fluid into the vessel, but also the confidence of the results and their reproducibility was guaranteed due to the low variability in their mathematical amount.

Total phenolic content

The results of the Folin-Ciocalteu assay revealed that the highest phenolic content was attributed to the central points (55 °C, 250 bar, 4 g/min), with an average of 196.89 ± 3.22 GAE (mg)/1 g of dried extract (Figure 1). To evaluate the effect of three parameters on phenolic content yield, including pressure, temperature, and co-solvent flow rate, the RSM was performed and a quadratic equation was obtained (Equation 6). The coefficient of prediction was 0.99, indicating that the model efficiently captures the majority of the variability in the response. Among the examined factors, temperature showed a highly significant effect on the TPC (P < 0.05). According to Kasapoğlu et al. (2023), raising the vessel temperature from 40 to 60 ° C leads to an enhancement in the solubility of polyphenols. Moreover, the individual, quadratic and interaction effects of co-solvent on TPC were observed, which could be attributed to the changes in fluid’s polarity. At moderate co-solvent flow rate, the polarity of the fluid becomes suitable to yield more polyphenols. On the other hand, the interaction of co-solvent with temperature and pressure by altering the viscosity of the fluid could significantly affect the TPC. These results are consistent with previous findings (Allay et al., 2025; Wang et al., 2024). Pressure was another important factor, which, by affecting the viscosity of the fluid, showed a slight effect individually (P < 0.1), while its quadratic effect and interaction with co-solvent or temperature significantly affect TPC (P < 0.05) (Figure 2).

Antioxidant capacity determinations

To explore one of the potential biological characteristics of the extracts obtained under the different supercritical extraction conditions, the antioxidant potential of the bioactive plethora across the 15 treatments was evaluated using two of the most conservative assays (ABTS and DPPH) with a novel ROS-specific neutralization capacity method. According to the ABTS results in Figure 3A, only samples 5, 11, and the three central points (13, 14, and 15) demonstrated the highest radical scavenging activity (> 190 TE (mg)/1g of dried extract). In contrast, treatments 10 and 12 displayed lower activity, with the standard OA showing the weakest response. Analogously, the highest activity in the DPPH assay was attributed to sample 5 and the central points, whereas treatment 12 and the standard OA exhibited the lowest antioxidant response (Figure 3B). Although treatments 5, 11, 13, 14 and 15 showed average dominant ABTS responses compared to DPPH, a common general tendency in their antioxidant patterns was observed. This suggests that the ability to neutralize these two particular radicals arises from synergistic effects of multiple phytochemicals at equal masses, compared to the standard OA (Mokhtar et al., 2016; Zheng et al., 2021).

These results are not directly aligned with the TPC of the samples; therefore, lower DPPH values compared to ABTS suggest the differences in the chemical nature of the radicals and their neutralization pathways. Indeed, A B T S • + was found to react with both hydrophilic and lipophilic compounds due to its solubility in both aqueous and organic media, while D P P H • was more potent at reacting with hydrophobic compounds in organic solvents (Sadeer et al., 2020). However, despite the high hydrophobicity of OA, its neutralization capacity was negligible.

The profiles of ROS-specific neutralization capacities are shown in Figure 4 for each extract treatment. From these results, it was observed that the highest 1 O 2 neutralization was attributed to treatment 7 with 40%, while there was not a significant difference with treatments 1, 8, and standard OA (Figure 4A). Although treatment 7 contains a moderate level of OA and polyphenols, as shown in Table 3, it shows a higher neutralization level for this species. This particular result suggests that not only potent individual antioxidants are needed to neutralize such highly reactive species, but also that specific mixtures of compounds may contribute to potential synergistic effects. As demonstrated in the previous study, where different antioxidant standards were evaluated using this methodology (Buenfil-Chi et al., 2025), various individual polyphenols and triterpenes exhibited intrinsically low to medium responses for 1 O 2, which could explain the similar potency observed for this ROS type in all treatments (Figure 4A). Conversely, the differences in OH • neutralization between different treatments were not significant, and all were shown to have high neutralization potential (70–95 %), with the exception of the standard OA (Figure 4B). Regarding H 2 O 2, higher neutralization responses, they were observed for treatments 4, 5, 9, and 11 (Figure 4C). However, there was no correlation with the high content of either OA or phenolics. It was interesting to note that, although treatments 7 and 8 appear to be the most effective against 1 O 2, they presented the weakest neutralization capacity against H 2 O 2. On the other hand, treatment 8 demonstrated the highest level of antioxidant capacity against O 2 • − (Figure 4D). While significant differences were observed among certain treatments, they exhibited a moderate neutralization capacity, ranging from approximately 35%–65%, relative to 1 O 2 and OH •. It was important to mention that compounds obtained from all treatments were dissolved in DMSO due to their lipophilic nature, using DMSO as a co-solvent to create a water-soluble dispersion. However, if more polar or hydrophilic molecules are present, they independently dissolve in water. This implies that the diverse bioactive content coexists in the dispersion, which may enable interactions among the compounds and potentially contribute to the distinct ROS profiles observed.

To further elucidate the relationships between OA and polyphenol content with the observed antioxidant profiles, the results of PCA are expressed as follows: PC1 and PC2 explained 33.2% and 25.3% of the total variance, respectively, and revealed distinct correlation patterns (Figure 5). Interestingly, the level of OA showed a closer association with the neutralization of the OH •. This could be attributed to the inherent metal-chelating property of OA, which may bind transition metal ions like iron and copper, thereby potentially inhibiting Fenton-type reactions that generate this highly reactive species (Günther et al., 2026). This finding suggests that the contribution of OA to the overall antioxidant capacity, while not the dominant factor, appears mechanistically linked to specific ROS pathways. Conversely, a stronger and positive correlation was observed between the TPC and the scavenging of A B T S • + and D P P H • as well as the neutralization of H 2 O 2. This aligns with the role of phenolic compounds as electron and hydrogen atom donors, mechanisms central to these particular assays (Leopoldini et al., 2011). However, it is worth noting that the relationship between phenolics and H 2 O 2 is complex, as some studies indicate that phenolic compounds can both generate and scavenge this molecule, making the net effect dependent on their specific composition. Most notably, the PCA also highlighted an inverse relationship. TPC appeared negatively correlated with the neutralization of the O 2 • −, as suggested by their opposite directions along PC2. This could imply that the specific phenolic profile, rather than the total phenolic content alone, plays a decisive role in neutralizing this particular ROS. Given the structural diversity among phenolic compounds, it is possible that some may be more effective against O 2 • − than others, or that certain interactions within the complex extract mixture modulate this activity. Accordingly, an study on hop extracts found that phenolics like catechin and quercetin were the main contributors to DPPH scavenging, their activity against O 2 • − and H 2 O 2 differed significantly, underscoring the compound-specific nature of these activities (Gorjanović et al., 2013). Therefore, the observed negative correlation suggests that other unmeasured variables, such as the specific types of phenolic compounds present (e.g., flavonoids vs. phenolic acids) and their synergistic or antagonistic interactions with triterpenes like OA, may be significant contributors to the neutralization of O 2 • −. This reinforces the idea that a holistic antioxidant effect is not solely dependent on total phenolic content but is the result of a specific and synergistic interplay among the various bioactive compounds within the extract.

When comparing synthetic radical scavenging against physiological ROS neutralization, it was noticeable that treatments 10 and 12, which had low activity in DPPH and ABTS assays (Figure 3), also revealed low effectiveness in neutralizing 1 O 2 (Figure 4A). Nevertheless, these specific treatments are able to neutralize the remaining ROS, ranging from moderate (H 2 O 2 and O 2 • −) to relatively high (OH •) levels. These results indicate that the antioxidant responses obtained from ABTS and DPPH assays do not necessarily reflect the neutralization capacity of the extracts toward all individual physiological ROS.

Certainly, a prolonged exposure to specific ROS has been associated with particular diseases, such as glaucoma 1 O 2 (Murotomi et al., 2023), atherosclerosis (OH •) Lipinski (2011), cancer (H 2 O 2) Sies and Jones (2020), and cardiomyopathy (O 2 • −) Pitkanen and Robinson (1996). Although the present work was not intended to identify the compounds responsible for these responses, these functional assays provide an initial characterization of the extracts obtained under different extraction conditions and may guide future phytochemical investigations. Furthermore, since different ROS species are neutralized via various pathways, these findings highlight the need for a broad-spectrum antioxidant. This could be achieved through either a purified biomolecule or a combination of bioactive compounds that provides significant neutralization responses across different ROS types.

In the present study, Ph. leucarpum, a parasitic plant native to Mexico and commonly used in folk medicine, was investigated. The extraction process was carried out using the SC-CO 2 technique. The effects of temperature, pressure, and co-solvent on extract yield, as well as on the OA and polyphenol content, were evaluated under 15 experimental conditions determined by a Box-Behnken design. The OA, as a representative of the triterpene group, was identified and quantified using HPLC-DAD-ELSD. Treatment 10, with 2 g/min co-solvent, 250 bar, and 40 ° C produced the highest yield of OA. The TPC of the samples was also examined using the Folin-Ciocalteu assay. The best treatments for achieving the highest phenolic yield were primarily at the central point, with 55 ° C, 250 bar, and 4 g/mL co-solvent. Furthermore, the response surface methodology analysis of the TPC results revealed that temperature and co-solvent, had a significant impact, while pressure showed a slight individual effect, with its quadratic and interaction effects contributing significantly to the yield of this class of compounds.

The antioxidant capacity of different treatments was determined and compared using conventional synthetic-based radical assays, including ABTS and DPPH, along with a novel technique for ROS neutralization of physiological relevance (1 O 2, OH •, H 2 O 2, and O 2 • −). These results not only highlight the richness of OA and polyphenols in Ph. Leucarpum, but also reveal significant discrepancies where selective treatments with high neutralization capacity for synthetic radicals do not necessarily correspond to high neutralization of physiological ROS, and vice versa. The PCA analysis further revealed that TPC correlates positively with A B T S • +, D P P H •, and H 2 O 2, while OA shows a closer association with OH • scavenging, likely through metal-chelating mechanisms. Interestingly, TPC appeared negatively correlated with O 2 • − neutralization, suggesting that specific phenolic profiles rather than total phenolic content may be more relevant for neutralizing this particular ROS. Moreover, the observed antioxidant profiles cannot be directly attributed to the evaluated OA and polyphenol content, and are likely influenced by the overall phytochemical composition of the extracts and potential synergistic interactions among their constituents. Hence, by leveraging the potential of SC-CO 2 green technology, this work provides a foundation for future studies aimed at identifying the specific compounds responsible for the observed antioxidant profiles. Additionally, further research is needed to explore the precise phytochemical profiles of the extracts under each treatment, aiming to unravel the potential synergistic effect among the compounds contributing to the observed antioxidant profiles. These insights may support future studies toward the development of innovative, nature-based products for the cosmetics, food, and pharmaceutical industries, leveraging the rich herbal resources of the region.

Moreover, treatment 8 (40 ° C, 100 bar, and 4 g/min of co-solvent) was selected as the optimal condition for future studies. This treatment provided the highest extract yield and satisfactory level of bioactive compounds such as OA and polyphenols; it also showed a high antioxidant capacity against the most relevant biological ROS in the advanced ROS-specific neutralization method, which was developed in our group. Additionally, it operates at the lowest temperature and pressure among all the conditions that were tested, and it uses a moderate amount of co-solvent. Together, these factors help lower energy consumption, making this option more practical and cost-effective for potential industrial-scale applications.

Equations

where