Optimization of Subcritical Water Extraction Process for Polyphenols from Cinchona calisaya and Their Activity Analysis

TL;DR

This study explores a new method to extract polyphenols from Cinchona calisaya using subcritical water, showing better results than traditional methods and potential antimicrobial properties.

Contribution

The first study to use subcritical water extraction for polyphenols from Cinchona calisaya, optimizing key parameters and analyzing antimicrobial activity.

Findings

Optimal extraction conditions achieved higher total phenolic content in bark and heartwood compared to traditional hot water extraction.

Heartwood contained a higher content of phenolic compounds than bark, as identified by UHPLC-Q-Orbitrap HRMS.

Extracts showed strong inhibition of Staphylococcus aureus and weak inhibition of Escherichia coli at high concentrations.

Abstract

This study expands research on Cinchona calisaya, which has traditionally focused on alkaloids, to address the insufficient comprehensive utilization of its resources. It is the first to explore the feasibility of extracting phenolic compounds from Cinchona calisaya using subcritical water extraction (SCWE) technology. Combining single-factor experiments with Box–Behnken design-response surface methodology (BBD-RSM), the study optimized three key process parameters, namely extraction temperature, extraction time, and liquid-to-solid ratio, with total phenolic content (TPC) as the response variable. Optimal extraction conditions were 165 °C, 20 min, and a liquid-to-solid ratio of 70:1 mL/g. Under these conditions, the TPC values were 98.41 ± 1.06 mg/g for bark and 37.96 ± 1.18 mg/g for heartwood, significantly higher than those obtained by traditional hot water extraction (THWE). Ultra-high performance liquid chromatography-Q-orbitrap high-resolution mass spectrometry (UHPLC-Q-Orbitrap HRMS) identified 1198 and 1156 metabolites in the bark and heartwood, respectively, with a higher content of phenolic compounds in the heartwood. The extracts showed strong inhibitory activity against Staphylococcus aureus, weak inhibitory activity against Escherichia coli at high concentrations, and no inhibitory effect on Candida albicans. This study provides a theoretical basis and technical support for the comprehensive utilization of Cinchona calisaya resources and the green development of natural antioxidants and antimicrobial agents.

Figures

Full text

1. Introduction

Cinchona calisaya, commonly known as the quinine tree, is an evergreen medicinal plant belonging to the genus Cinchona (family Rubiaceae). Native to the mid- to high-altitude regions of the Andes Mountains in South America, it has been introduced and cultivated in South Asia, Southeast Asia, and Yunnan Province of China, and is recognized as one of the most classic and valuable medicinal plants worldwide [1]. Its bark, roots, branches, and stems are rich in nearly 30 quinine-type alkaloids, including quinine and quinidine, which serve as its core medicinal components [1]. The discovery of quinine as a monomer from Cinchona calisaya bark by Caventou and Pelletier in 1817 accelerated the global development of alkaloid preparation technologies for this species, leading to the synthesis of a series of quinine salts [2]. To date, research on Cinchona calisaya has remained centered on the extraction and application of these quinine-type alkaloids. This alkaloid-centric utilization pattern has resulted in a critical issue. Plant parts such as branches and heartwood, accounting for over 60% of the total plant biomass, are discarded as waste, as their alkaloid content is less than 0.1% [3,4]. This not only leads to a markedly insufficient comprehensive utilization rate of Cinchona calisaya resources but also represents a missed opportunity for valorizing underutilized biomass. As important plant secondary metabolites, phenolics have long attracted significant attention in the pharmaceutical, food, and cosmetics industries owing to their diverse biological activities, including antioxidant, anti-inflammatory, and antimicrobial properties [5]. However, there are only very few relevant reports on the systematic identification and extraction of phenolic compounds from Cinchona calisaya, and these non-alkaloid active components have not been effectively exploited, further exacerbating the underutilization of this resource. As the main underutilized parts of Cinchona calisaya, branches and heartwood are potential sources of phenolic antioxidants. Extracting phenolic compounds from these parts can significantly improve the comprehensive utilization rate of Cinchona calisaya resources and promote the sustainable development of the medicinal plant industry. To the best of our knowledge, only a limited number of previous studies have focused on extracting phenolic compounds from Cinchona calisaya [6]. This constitutes a distinct research gap, as the underutilization of Cinchona calisaya biomass could be mitigated by targeting its phenolic compounds, while the feasibility of efficient extraction and the biological potential of these compounds remain largely unexplored.

Organic solvent extraction is the most commonly used method for extracting phenolic compounds from plants [7]. However, solvents such as methanol, ethanol, n-hexane, and petroleum ether are volatile, flammable, toxic, and costly [8]. For products intended for functional food or pharmaceutical applications, stringent and energy-intensive solvent removal protocols are indispensable, limiting the safety and sustainability of this approach [9]. Thus, there is an urgent demand for novel extraction technologies characterized by low processing costs, mild operating conditions, short treatment times, and environmentally benign solvents. Subcritical water extraction (SCWE) is a promising green extraction technology using subcritical water as the solvent, maintained in a liquid phase by pressure exceeding its vapor pressure at a given temperature [10]. Compared with conventional organic solvent extraction, SCWE offers key advantages, as it requires minimal or no organic solvents to reduce environmental and safety risks and has a significantly shorter processing time of 1–50 min than traditional maceration [11]. Critically, the physicochemical properties of subcritical water, including decreased dielectric constant, viscosity, and surface tension, and increased diffusion coefficient, can be precisely regulated by adjusting temperature and pressure [12]. For example, the dielectric constant (ε) of water decreases from 80 at 25 °C to 25 at 250 °C, falling within the range of methanol (ε = 33) and ethanol (ε = 24) [13,14]. This property enhances the solubility of moderately polar compounds such as phenolics, making SCWE particularly suitable for extracting phenolic compounds from plant matrices by improving mass transfer and solubility while avoiding solvent-related drawbacks.

Given the research gap in the extraction of phenolic compounds from Cinchona calisaya and the advantages of SCWE, this study integrates SCWE with response surface methodology (RSM) to optimize the extraction conditions for Cinchona calisaya bark and heartwood, identify the optimal process, and clarify how extraction parameters affect the yield of phenolic compounds as well as their antioxidant and antimicrobial activities. The purpose of this study is therefore threefold. First, it aims to optimize SCWE conditions for extracting phenolic antioxidants from Cinchona calisaya bark and heartwood using RSM. Second, it seeks to compare extraction efficiency and matrix characteristics between the bark and heartwood matrices. Third, it intends to evaluate the antioxidant and antimicrobial activities of the phenolic extracts and clarify the effects of extraction parameters on these activities. By addressing these objectives, this study seeks to provide a strategy for improving the comprehensive utilization and sustainability of Cinchona calisaya resources while laying the groundwork for the development of natural antioxidants and antimicrobial agents from this underutilized biomass.

2. Results

2.1. Single-Factor Analysis

Single-factor experiments were conducted to investigate the effects of extraction temperature (Figure 1A), extraction time (Figure 1B), and liquid-to-solid ratio (Figure 1C) on the total phenolic content (TPC) in the bark and heartwood of Cinchona calisaya. With respect to extraction temperature (Figure 1A), TPC increased steadily from 105 °C to 165 °C, which corresponded to the maximum yield. A subsequent decrease in TPC was observed when the temperature was further increased to 180 °C. Regarding extraction time (Figure 1B), TPC exhibited a statistically significant increase (p-value < 0.05) when the extraction time was extended from 10 to 20 min. In contrast, further prolongation of the extraction time only resulted in a non-significant decline in TPC (p-value > 0.05). Among the three factors, the liquid-to-solid ratio (Figure 1C) exerted the most pronounced statistically significant effect on TPC (p-value < 0.05). Specifically, TPC increased progressively as the liquid-to-solid ratio increased from 30:1 to 70:1, but it declined when the ratio was further raised to 80:1. Collectively, the single-factor experiments showed the optimal extraction conditions as 165 °C, 20 min, and a liquid-to-solid ratio of 70:1 mL/g.

2.2. Response Surface Analysis

Based on the single-factor experimental results, extraction temperature (A), extraction time (B), and liquid-to-solid ratio (C) were selected as independent variables for optimization. With the TPC in Cinchona calisaya bark extract as the response index, the Box–Behnken design (BBD) was adopted using Design-Expert 13.0 software for experimental design, aiming to determine the optimal process parameters for phenolic compound extraction from Cinchona calisaya bark. The design and results of the response surface experiments are presented in Table 1. Through quadratic multiple regression fitting, the multiple regression equation for TPC was obtained:

$$[eqn]$$

The analysis of variance (ANOVA) results for this model are presented in Table 2. The model exhibited an F-value of 66.29 and a p-value of less than 0.0001, indicating that the model was extremely significant. In contrast, the lack of fit term of the model had a p-value of 0.8656 (>0.05), which was not significant. This demonstrated that the model had a good fit within the regression region and could effectively describe the experimental results.

Analysis of the linear terms showed that factors B (extraction time) and A (extraction temperature) had an extremely significant effect on TPC (p-value < 0.01), while factor C (liquid-to-solid ratio) had the smallest yet significant effect (p-value< 0.05). Extraction time was the decisive factor affecting TPC, and the order of effect of the linear terms on TPC was extraction time (B) > extraction temperature (A) > liquid-to-solid ratio (C). Analysis of the interaction terms revealed that the AC interaction was extremely significant (p-value < 0.01), indicating an extremely significant synergistic or interfering effect between these two factors. In contrast, the p-values of the AB and BC interactions were greater than 0.10, suggesting no significant synergistic or interfering effects between these factor pairs. The order of effect of the interaction terms on TPC was AC > BC > AB. Analysis of the quadratic terms indicated that A^2^, B^2^, and C^2^ all exerted an extremely significant effect on TPC, with the order of effect being A^2^ > B^2^ > C^2^. Error statistical analysis of the regression equation showed that the model had an R^2^ of 0.9884 and an adjusted R^2^ of 0.9735, meaning that 97.35% of the variation in the response value could be explained by the model.

Consequently, the regression model demonstrated a strong correlation with the actual experimental data and was able to effectively predict the experimental outcomes, suggesting that the application of this model is practical for forecasting the optimal preparation method for phenolic compounds extracted by subcritical water. As a result, the developed model is appropriate for investigating and predicting response values under different SCWE conditions.

Additionally, three-dimensional (3D) response surface plots were employed to demonstrate the interactions among the three variables and to assess the anticipated impact of each variable on the maximum levels of phenolic compounds (Figure 2). The elliptical contour lines signify a notable interaction between pairs of factors, with the center of the smallest ellipse indicating the peak or lowest point on the response surface [15]. These response surface plots illustrate both the relationships and the extent of interaction among different factors. Contour plots generated through response surface analysis provide a clearer and more intuitive visualization of the relationships between pairs of factors. In particular, elliptical contours denote a highly significant interaction between two variables, whereas circular contours suggest an absence of a significant interaction between them [16].

In Figure 2A, the effects of extraction temperature (A) and time (B) on TPC were examined at a constant liquid-to-solid ratio. The nearly circular contour lines in the 2D plot and the relatively flat curvature of the 3D response surface indicate a negligible interactive effect between these two factors. This result shows that temperature and time act independently, with no synergistic or antagonistic effect on phenolic compound extraction. Figure 2B illustrates the effects of extraction temperature (A) and liquid-to-solid ratio (C) on TPC at a fixed extraction time. The distinctly elliptical contour lines and the steep, sloped curvature of the 3D response surface reveal a strong and significant interactive effect between these variables. The elliptical shape of the contour lines demonstrates that the combined variation in temperature and liquid-to-solid ratio exerts a greater influence on phenolic compound extraction efficiency than either factor acting alone. In Figure 2C, the effects of liquid-to-solid ratio (C) and extraction time (B) on TPC were assessed at a constant extraction temperature. The near-circular contour lines and the gentle curvature of the 3D surface confirm a minimal interactive effect between these two factors. This outcome is similar to the weak interaction observed between temperature and time.

Collectively, the BBD-RSM experiments showed the optimal theoretical extraction conditions are 163.2 °C, 22.5 min, and a liquid-to-solid ratio of 68.6:1 mL/g. These parameters predict a maximum TPC of 98.94 mg/g. For practical operational feasibility, the optimal parameters were slightly adjusted to 165 °C, 20 min, and a liquid-to-solid ratio of 70:1 mL/g. Three replicate validation experiments were performed under these modified conditions to verify the regression model. The average TPC obtained was 98.41 ± 1.06 mg/g. This value is highly consistent with the theoretical maximum. This consistency validates the stability, reliability and predictive capacity of the established model. It also confirms the robustness of the identified optimal SCWE conditions for efficient phenolic compound extraction.

Compared with the traditional hot water extraction (THWE), the phenolic compounds extracted from Cinchona calisaya bark were 70.35 ± 0.52 mg/g, while that under the optimal SCWE conditions reached 98.41 ± 1.06 mg/g, showing a 40% increase in extraction efficiency with an extremely significant difference (p-value < 0.0001, Table 3). For the heartwood, the TPC by THWE was 18.27 ± 1.27 mg/g, and this value was greatly enhanced to 37.96 ± 1.18 mg/g under the optimal SCWE conditions, corresponding to a 108% increase.

2.3. Composition Analysis

Metabolite composition analysis was performed using ultra-high performance liquid chromatography-Q-orbitrap high-resolution mass spectrometry (UHPLC-Q-Orbitrap HRMS). After processing with Progenesis QI, a total of 1198 metabolites were identified in the Cinchona calisaya bark extract, with 556 metabolites detected in ESI^+^ mode and 642 in ESI^−^ mode (Table 4). Further retrieval from the Human Metabolome Database (HMDB, https://hmdb.ca, accessed on 25 August 2025) was performed, and unrecognized metabolites were excluded, resulting in a total of 922 valid metabolites. Among these, 372 metabolites were identified in ESI^+^ mode and 550 in ESI^−^ mode. For the Cinchona calisaya heartwood extract, 1156 metabolites were initially identified via Progenesis QI, including 569 in ESI^+^ mode and 587 in ESI^−^ mode. Subsequent retrieval from HMDB and exclusion of unrecognized metabolites yielded 888 valid metabolites, with 379 detected in ESI^+^ mode and 509 in ESI^−^ mode.

Furthermore, superclass classification analysis was conducted on the identified metabolites (Figure 3). The results showed that the chemical components of both bark and heartwood of Cinchona calisaya encompassed ten major categories, including lipids, phenylpropanoids, organic oxygen compounds, organic acids, alkaloids, and benzene compounds. Notably, significant differences were observed in the content and abundance of lipids, organic oxygen compounds, organic acids, and alkaloids between these two tissue types. Further statistical analysis based on HMDB subclasses indicated that the chemical profiles of both tissues were dominated by isoprenoid alcohols (166 in bark and 161 in heartwood), with phenol-related components occupying an important position. A total of 104 flavonoids and 104 organic oxygen compounds were detected in the bark, while 97 flavonoids and 108 organic oxygen compounds were found in the heartwood. Additionally, substantial amounts of fatty acyls, carboxylic acids and their derivatives, steroids, and benzene and its substituted derivatives were identified. Importantly, both tissues contained numerous low-abundance subclasses, reflecting the high chemical diversity of Cinchona calisaya. Analysis of secondary metabolites in the bark and heartwood of Cinchona calisaya revealed that quinine was the secondary metabolite with the highest relative content in both tissues, while phenolic compounds accounted for a significant proportion in both. As shown in Table S1, 246 phenolic compounds were detected in the bark, accounting for approximately 20.68% of the total metabolites, while 260 phenolic compounds were identified in the heartwood, with a relative content of about 33.8%. Although the relative proportion of phenolic compounds was higher in the heartwood, the absolute content of phenolic compounds in the bark remained significantly higher than that in the heartwood. Tables S2 and S3 indicated that phenolic compounds constituted a significant proportion of the top 20 most abundant secondary metabolites in the bark, further confirming that phenolics serve as a key active substance basis in this plant. Thus, abundant phenolics and other bioactive metabolites in Cinchona calisaya bark and heartwood underpin its core bioactivities, including antioxidant activity, and facilitate the advanced development and industrialization of high-value bioactive components.

2.4. In Vitro Activity Assays

2.4.1. Antioxidant Activity Assays

The extracts from the bark and heartwood of Cinchona calisaya are rich in phenolic compounds, including polyphenols and flavonoids. To further analyze their antioxidant activity, this study diluted the bark and heartwood extracts according to their TPC for activity determination. The extraction efficiencies of the THWE and the optimized SCWE were compared, with ascorbic acid used as a positive control to clarify differences in antioxidant capacity (Figure 4). The results showed that extracts from both processes exhibited effective free radical scavenging activity, with the SCWE-derived extracts showing more prominent antioxidant capacity. Specifically, the DPPH free radical scavenging IC_50_ values of SCWE-obtained bark and heartwood extracts were 0.1945 ± 0.0039 mg/mL and 0.1984 ± 0.0139 mg/mL (Figure 4A), respectively, while their ABTS^+^ free radical scavenging IC_50_ values were 0.5663 ± 0.0012 mg/mL and 0.3567 ± 0.0081 mg/mL (Figure 4B), respectively. Combined with the ferric reducing power assay results of Cinchona calisaya extracts (Figure 4C), these findings confirmed that the antioxidant activity of extracts obtained by SCWE was superior to that by THWE.

2.4.2. Antimicrobial Activity Assays

The bark and heartwood extracts of Cinchona calisaya exhibited significant antimicrobial activity, which may be mainly attributed to their abundant polyphenols, flavonoids, and relatively high alkaloid content (with quinine as a representative). In this study, quinine was selected as the positive control, as it has been well-documented to possess potent antimicrobial activity. As shown in Table 5, the antimicrobial activity of Cinchona calisaya bark extract was generally superior to that of the heartwood extract. At an extract concentration of 100 mg/mL, a certain inhibitory effect on Staphylococcus aureus was observed. When the concentration increased to 200 mg/mL, the extract exhibited weak inhibitory activity against Escherichia coli but almost no inhibitory activity against Candida albicans. Overall, the antimicrobial efficacy followed the sequence Staphylococcus aureus (Gram-positive bacteria) > Escherichia coli (Gram-negative bacteria) > Candida albicans (fungus), a pattern closely related to the permeability characteristics of the outer barrier structures of these three microorganisms and the accessibility of their target sites to antimicrobial substances.

To ensure the reliability of the antimicrobial activity results, a 5 mg/mL quinine ethanol solution was used as the positive control, with reference to the standards of the National Committee for Clinical Laboratory Standards (NCCLS). The results showed that the negative control group had no antimicrobial effect, while the inhibition zone diameter of the positive control against Staphylococcus aureus was >19 mm, indicating high sensitivity (Figure 5). Repeated experiments yielded stable and reliable results.

From a microbial structure perspective, Staphylococcus aureus, as a Gram-positive bacterium, has only a loose peptidoglycan layer (20–80 nm) in its outer cell wall, lacking an outer membrane. The above-mentioned antimicrobial pattern thus reflects a negative correlation between the complexity of the cellular barrier and antimicrobial efficacy. Specifically, Gram-positive bacteria are easier to inhibit than Gram-negative bacteria, and bacteria are easier to inhibit than fungi. Although factors such as the type of antimicrobial substances, action environment (pH, temperature), and drug resistance may cause fluctuations in specific antimicrobial activity values, this basic pattern remains unchanged. Such understanding can further provide theoretical support for the development of targeted antimicrobial agents, the analysis of mechanisms of action, and the selection of application scenarios.

3. Discussion

This study addressed the insufficient comprehensive utilization of Cinchona calisaya resources, traditionally focused solely on alkaloid extraction, by exploring SCWE for phenolic compounds. The results support green extraction of Cinchona calisaya bioactives, deepen understanding of non-alkaloid composition and activities, and advance high-value utilization of this medicinal plant.

The single-factor experimental results revealed that extraction temperature, extraction time, and liquid-to-solid ratio all induced a unimodal response in the TPC of the extracts, with the TPC rising to an optimal value before declining thereafter. This trend was attributed to the dynamic balance between the enhanced solubility of moderately polar phenolic compounds in subcritical water and the thermal degradation and oxidation of heat-sensitive phenolic compounds at elevated temperatures [17], and the improved solubility was associated with the reduced dielectric constant and polarity of subcritical water caused by increasing temperature [18]. Although the three factors exhibited an identical trend in their effects on TPC, their statistical significance and underlying action mechanisms differed significantly. Specifically, solvent penetration into the plant matrix required a specific time threshold to achieve effective solubilization of phenolic compounds [19]. When extraction time exceeded 20 min, the TPC showed no substantial increase and even risked slight thermal degradation of phenolic compounds. This finding further verified the aforementioned balance relationship. Analysis via the BBD-RSM identified extraction time as the decisive factor governing TPC, and the interaction between extraction temperature and liquid-to-solid ratio exerted a significant effect on TPC (p < 0.05). Thus, the synergistic regulation of these two parameters can optimize the extraction process and improve the efficiency of phenolic compound extraction in practical applications. Furthermore, the TPC of extracts obtained by SCWE was significantly higher than that by THWE, which is consistent with the findings of previous studies [20]. Additionally, this technique eliminates the need for toxic and volatile organic solvents, thereby reducing potential risks to product safety and the environment and demonstrating broad application prospects in the food and pharmaceutical industries.

Metabolomic analysis demonstrated that phenolic compounds serve as the core functional components in the bark (1198 metabolites) and heartwood (1156 metabolites) of Cinchona calisaya. Specifically, the bark contains 246 phenolic compounds, accounting for 20.68% of the relative content, whereas the heartwood comprises 260 phenolic compounds with a relative content of 33.8%, reflecting a higher relative proportion in the heartwood alongside a higher absolute content in the bark. This tissue-specific difference is hypothesized to be associated with the active accumulation of phenolic compounds in the bark under adverse stress conditions, which is consistent with the conclusion from existing literature that phenolic compounds, as key plant secondary metabolites, are implicated in environmental adaptation and stress resistance [21]. As a defensive barrier, the bark contains 12 unique secondary metabolites among the top 20 most abundant compounds. Among these metabolites, phenolic compounds, including protocatechuic acid and cinnamtannin A2, not only exhibit antimicrobial and antiviral activities but also absorb ultraviolet radiation to mitigate UV-induced damage [22,23]. In contrast, as a structural support organ, the heartwood contains 12 unique secondary metabolites among the top 20 most abundant compounds, with the phenolic compounds among them exhibiting functions that align with the demands for structural stability and long-term defense. Notably, 3,4-dicaffeoylquinic acid, a phenolic compound with high abundance in the heartwood, scavenges oxidants, retards lignin degradation and improves wood durability [24,25]. Rubiadin, by contrast, enhances preservative effects and preserves heartwood integrity [26]. Both compounds thus represent valuable candidates for development from the heartwood. Based on the present research findings and the requirements of resource development, Cinchona calisaya bark with high absolute phenolic content is ideal for pharmaceutical development, while heartwood with high relative phenolic proportion and stability fits wood protection.

In terms of in vitro biological activities, SCWE extracts exhibited superior antioxidant activity to THWE extracts, with lower IC_50_ values for DPPH and ABTS^+^ free radical scavenging. This is closely related to the higher TPC in SCWE extracts, as phenolic compounds with phenolic hydroxyl groups are well-recognized key antioxidants that scavenge free radicals by hydrogen atom donation [27]. For antimicrobial activity, the extracts had significant inhibition against Staphylococcus aureus (Gram-positive bacteria), weak activity against Escherichia coli (Gram-negative bacteria), and no effect on Candida albicans (fungi), following the inhibition difficulty order of Gram-positive bacteria < Gram-negative bacteria < fungi. This pattern stems primarily from structural differences in microbial cell barriers. Gram-positive bacteria lack an outer membrane, facilitating the penetration of antimicrobial components. Gram-negative bacteria possess a lipid outer membrane that blocks polar substances. Fungi have a chitin-glucan cell wall and an ergosterol-containing cell membrane, which further reduces the permeability of antimicrobial agents. Quinine was selected as the positive control to verify the reliability of the antimicrobial activity results [28]. Quinine was used as a positive control to verify result reliability. Notably, bark extracts generally exhibited superior antimicrobial efficacy compared to heartwood extracts, consistent with the higher absolute TPC and quinine in the bark. Specifically, Staphylococcus aureus has a loose peptidoglycan layer (20–80 nm) without an outer membrane, allowing most antimicrobial components to easily penetrate, act on the cell membrane, or interfere with intracellular metabolism, thereby exerting a significant inhibitory effect [29,30]. Escherichia coli has a thin peptidoglycan layer (2–3 nm) and a lipid outer membrane that selectively blocks polar antimicrobial compounds, with only hydrophobic components or those assisted by outer membrane disruptors able to penetrate effectively, resulting in moderate activity [31,32]. As a fungus, Candida albicans has a chitin-glucan cell wall and ergosterol-containing cell membrane, which hinder the action of most bacteria-targeting antimicrobial substances. Additionally, its tendency to form biofilm barriers further decreases agent permeability, leading to almost no inhibitory effect [33,34,35].

Despite the notable findings, this study has certain limitations. Metabolite identification relied solely on database matching, and structural confirmation and purification of key active components such as 3,4-dicaffeoylquinic acid remain unperformed, restricting in-depth exploration into their specific mechanisms of action. In vitro activity assays were conducted under controlled laboratory conditions, so the actual efficacy and safety of the extracts in in vivo settings or practical applications, including food preservation and pharmaceutical development, require further verification. Additionally, the study focused exclusively on phenolic compounds and quinine, leaving the biological activities of other metabolites, including lipids and organic oxygen compounds, unexplored, which may overlook other potential active components. Future research should focus on several key priorities. First, it is essential to purify and structurally confirm key active components so as to clarify their individual antioxidant and antimicrobial mechanisms and corresponding structure-activity relationships. Second, in vivo experiments should be conducted to evaluate the safety and efficacy of SCWE extracts while exploring their potential applications in functional foods, natural antioxidants and antimicrobial agents. Third, the scope of metabolite analysis needs to be expanded to cover understudied components, thus uncovering their biological activities and potential synergistic effects with phenolic compounds. Fourth, the SCWE process should be optimized for industrial-scale application to promote the large-scale production and comprehensive utilization of Cinchona calisaya resources.

4. Materials and Methods

4.1. Plant Material

Cinchona calisaya was provided by Yunnan Yuanfang Bioproducts Co., Ltd. It was identified as Cinchona calisaya (Rubiaceae, Cinchona) by Dr. Yang Wenguang from the Kunming Institute of Botany, Chinese Academy of Sciences, through molecular identification of its dried fruit branch leaves and corresponding dried vegetative leaves. Voucher specimens are deposited at the Yuanfang Biotechnology R&D Center of Yunnan Yuanfang Bioproducts Co., Ltd. (Qujing, China, No. 20251209). The bark of Cinchona calisaya was stripped, and the remaining heartwood (after bark removal) was cut into small sections of appropriate length. Both the bark and heartwood sections were then dried to a constant weight at 102 ± 3 °C using a DHG-9140A electric blast drying oven (YIHENG Scientific Instrument Co., Ltd., Shanghai, China). Afterwards, the dried bark and heartwood were separately ground and sieved to obtain 60–80 mesh Cinchona calisaya bark powder and heartwood powder, which were stored at 4 °C for subsequent use.

4.2. Extraction

First, a single-factor analysis was conducted to investigate the effects of different extraction temperatures (105, 120, 135, 150, 165 and 180 °C), different extraction times (10, 20, 30, 40, 50 and 60 min) and different liquid-to-solid ratios (30:1, 40:1, 50:1, 60:1, 70:1 and 80:1 mL/g) on the TPC in the extracts from Cinchona calisaya bark and heartwood, respectively. Then, accurately weigh 1 g (to an accuracy of 0.0001 g) of Cinchona calisaya bark powder and heartwood powder and add it into a high-pressure reactor (Yanzheng Instruments Co., Ltd., Shanghai, China. Model YZRR-100). Add ultrapure water according to the preset liquid-to-solid ratio, then introduce nitrogen gas to displace the air inside the reactor. Perform extraction at a specified temperature for a set period, during which the pressure inside the reactor is maintained at approximately 3.5 MPa. After the extraction is completed, cool down the reactor, relieve the pressure, and filter the mixture to collect the extract. Further concentrate the extract using a rotary evaporator and then make it up to a constant volume of 50 mL. Finally, the Folin–Ciocalteu standard curve method [36] was adopted to determine the TPC.

4.3. RSM Optimization

Based on the results of the single-factor experiments, the optimization was conducted with the TPC in the extract set as the response value (Y), and the extraction temperature (A), extraction time (B), and liquid-to-solid ratio (C) serving as the independent variables, respectively. The design of factors and levels for the response surface test is shown in Table 6.

4.4. Composition Analysis

The chemical composition of the extract was analyzed using an ultra-high performance liquid chromatography (UPLC, Vanquish, Thermo, Waltham, MA, USA) coupled with a high-resolution mass spectrometer (Q Exactive, Thermo, Waltham, MA, USA) (i.e., ultra-high performance liquid chromatography-Q Exactive hybrid quadrupole-orbitrap high-resolution accurate mass spectrometry, UHPLC-Q-Orbitrap HRMS). The specific pretreatment steps for the extract were as follows: Accurately pipette 10 mL of the extract and centrifuge it at 12,000 rpm for 10 min at 4 °C. Collect the supernatant and place it at −20 °C for 1 h to precipitate proteins. Then centrifuge the mixture again at 12,000 rpm for 10 min at 4 °C. Take the resulting supernatant for vacuum drying, and add 200 μL of 30% acetonitrile solution to reconstitute it. After vortexing, centrifuge the solution at 14,000 rpm for 15 min at 4 °C, and the final supernatant was subjected to chemical composition analysis. The chromatographic column used for UHPLC-Q-Orbitrap HRMS was Waters HSS T3 (100 × 2.1 mm, 1.8 μm). The column temperature was set at 40 °C, the injection volume was 2 μL, and the flow rate was 0.3 mL/min. Gradient elution was adopted, where mobile phase A was a 0.1% formic acid-water solution and mobile phase B was an acetonitrile solution (containing 0.1% formic acid). The elution gradient was as follows: 0–1 min with an A/B ratio of 100:0 (v/v); 2–13 min with an A/B ratio of 5:95 (v/v); and 13.1–17 min with an A/B ratio of 100:0 (v/v). For the high-resolution mass spectrometry, an electrospray ionization (ESI) source was employed. The parameters were set as follows: sheath gas pressure at 40 psi, auxiliary gas pressure at 10 psi, ion spray voltage at +3000 V/−2800 V, source temperature at 350 °C, and ion transfer tube temperature at 320 °C. The scanning mode was Full-scan-dd MS^2^, with both positive ion (ESI^+^) and negative ion (ESI^−^) modes. The mass range for the full scan was 70–1050 Da with a resolution of 70,000, while the mass range for the MS 2 scan was 200–2000 Da with a resolution of 17,500.

The raw data were subjected to preprocessing procedures, including baseline filtering, peak identification, peak matching, peak alignment, and retention time calibration using Progenesis QI 2.0 software (Waters Corporation, Milford, MT, USA), resulting in a data matrix containing time, mass-to-charge ratio, and peak intensity.

4.5. Antioxidant Activity Assay

To investigate the antioxidant activities of extracts obtained from Cinchona bark and heartwood using different extraction techniques, the optimized extracts from hot water extraction and SCWE were selected as the study subjects. The 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity [37], 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS^+^) radical scavenging activity [38], and ferric reducing antioxidant power [39] were then determined sequentially, with slight modifications based on previously reported methods. IC_50_ values were calculated using GraphPad Prism 9.0. It refers to the concentration of extracts needed to reduce the rate of a specific reaction to 50% of its maximum in an experimental system, serving as a key indicator for evaluating the antioxidant activity of the extracts.

4.6. Antimicrobial Activity Assay

The in vitro antimicrobial activity against Staphylococcus aureus (CMCC26003), Escherichia coli (CMCC26003) and Candida albicans (CMCC98001) was determined via a slightly modified Oxford cup method, with reference to the reported protocol [40,41]. All three strains were standard reference strains purchased from the Shanghai Microorganism Culture Collection Co., Ltd. (Shanghai, China, SMCC). Staphylococcus aureus and Escherichia coli were cultured in Nutrient Broth medium, whereas Candida albicans was cultured in Sabouraud Dextrose Agar medium. Both media were procured from Guangdong Huankai Microbial Science and Technology Co., Ltd. (Guangdong, China), and agar powder was purchased from Beijing Aoboxing Biotechnology Co., Ltd. (Beijing, China).

First, 10 mL of culture medium was poured into each Petri dish as the bottom layer. After the bottom layer cooled and solidified, Oxford cups were placed vertically on its surface, followed by the addition of another 10 mL of molten culture medium. Once the second layer of agar cooled and solidified, the Oxford cups were carefully removed. Prior to the experiment, all experimental tools inside the laminar flow hood were disinfected by ultraviolet sterilization, which ensured the entire antimicrobial and antifungal assay was conducted under aseptic conditions. This measure effectively prevented contamination of the experimental system by miscellaneous bacteria and thus guaranteed the reliability of the assay results. Specifically, the substance cooled in this step was the second layer of agar.

Bacteria cultured for three generations were scraped off with an inoculating loop and suspended in sterile distilled water. The concentration of the bacterial suspension was adjusted to 1 × 10^7^ CFU/mL using sterile distilled water via the McFarland turbidity method and then set aside for use. A 25 μL aliquot of the bacterial suspension was added to each culture medium and evenly spread across the medium surface with a sterile cotton swab. Subsequently, 100 μL of extract at different concentrations was added to each well formed by removing the Oxford cups. Notably, quinine is slightly soluble in water but highly soluble in ethanol; thus, an aqueous quinine solution could not be used as the positive control, and a 5 mg/mL quinine ethanol solution was used as the positive control, and sterile distilled water was therefore selected as an additional negative control. All experiments were performed in triplicate for each group.

The Petri dishes were incubated at 37 °C for 24 h in a bacterial incubator for bacterial strains and at 27 °C for 72 h in a fungal incubator for Candida albicans, after which the formation of inhibition zones was observed. The diameter of each inhibition zone (defined as the longest distance between two opposite edges) was measured using a vernier caliper with an accuracy of 0.02 mm. For non-circular inhibition zones (e.g., those with broken edges), diameters were measured in multiple directions and the average value was calculated.

4.7. Data Processing

Experimental data are presented as the mean of at least three replicate experiments and analyzed using Design Expert 13 (Stat-Ease Inc., Minneapolis, MN, USA). All figures were generated using Origin 2024b (Origin Lab Co., Northampton, MA, USA). Statistical significance was defined as a p-value less than 0.05. All results are presented as mean ± standard deviation or 95% confidence intervals.

5. Conclusions

This study breaks through the traditional limitation of Cinchona calisaya that only focuses on alkaloid utilization, systematically verifies for the first time the high-efficiency extraction capacity of SCWE technology for phenolic compounds in its bark and heartwood, and identifies the differences in composition and biological activities between the two. The green extraction process combined with clear antioxidant and antimicrobial activity data provides a theoretical basis and technical support for the comprehensive development of Cinchona calisaya resources (reducing waste such as heartwood) as well as the research and development of natural antioxidants and antimicrobial agents.