Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2025 Dec 18;11(12):18627–18641. doi: 10.1021/acsomega.5c07869

The Landscape of Anti-HBV Activity in Essential Oils: A Machine Learning-Based Virtual Screening Framework and Anti-HBV Activity of Acorus tatarinowii Schott

Guotai Yang †, Jinjun Li ‡, Tengyun Shu †, Wenjing Pan †, Qiong Wang †, Jiarao Wang †, Xinhan Pang ‡, Shaoxing Dai ‡,*, Min Xu †,§,*
PMCID: PMC13044680  PMID: 41939362

Abstract

Innovative and curative therapies for HBV infection are urgently needed. Essential oils (EOs) offer potential as natural therapeutic alternatives for modulating inflammation and antiviral activity. Extensive screening of anti-HBV EOs promises to yield drug candidates that suppress HBV immune evasion via the modulation of inflammatory pathways, thereby providing new attempts at HBV clearance. This study aimed to investigate the landscape of anti-HBV EOs and develop a virtual screening model to predict the efficacy of EOs against HBV. 292 plant samples were collected and analyzed in terms of extracting their EOs, determining their compositions via GC-MS, and testing their anti-HBV activity. Of the 292 EOs, five EOs exhibited significant anti-HBV activity, including those from Acorus tatarinowii Schott (ATEOs), Caryopteris forrestii, and Vitex agnus-castus. To develop a machine learning model for predicting anti-HBV EOs, we randomly selected 274 EOs to develop a virtual screening and prediction model using Random Forest (RF) and Support Vector Machine (SVM) algorithms. The remaining 18 EOs were used to validate the precision and accuracy of the model. Our machine learning models achieved high precision in predicting anti-HBsAg activity and cytotoxicity. ATEOs emerged as the most potent candidates, with α- and β-asarones identified as the primary active components responsible for the anti-HBV activities. ATEO exerted multiple functions in regulating the MAPK signaling pathway, suggesting multiple mechanisms in modulating inflammatory responses to alter HBV replication. These findings suggested that EOs promise to offer insights into potential aromatherapy applications against HBV. ATEOs exerted anti-HBV activity primarily by modulating the MAPK signaling pathway, making them potent candidates for anti-HBV aromatherapy targeting inflammatory pathway regulation to block persistent HBV infection.

graphic file with name ao5c07869_0010.jpg

graphic file with name ao5c07869_0008.jpg

1. Introduction

Hepatitis B virus (HBV) infection poses a significant risk for progression to chronic hepatitis, liver cirrhosis, liver failure, and hepatocellular carcinoma (HCC). Current standard treatments primarily consist of pegylated interferon alfa and nucleos­(t)­ide analogues. However, these therapy approaches cannot eliminate covalently closed circular DNA (cccDNA) or integrated HBV DNA. , Furthermore, the rate of HBsAg seroclearance remains suboptimal. As a result, there is an urgent need for the development of innovative curative therapies for HBV infection. Numerous efforts have been made on therapies targeting the life cycle of HBV; , however, a functional cure for HBV remains difficult to achieve. Key breakthroughs may lie in controlling the immune manipulation of HBV. HBV suppresses the immune response to achieve long-term infection, and one of the most critical mechanisms is that HBV infection causes abnormalities in inflammation-related pathway responses and ultimately alters the immune microenvironment of the liver. − For instance, HBV infection activates IL-35 promoter transcription via the JNK/c-Jun signaling pathway, resulting in an immunosuppression. Another study revealed that HBx promotes HBV replication by degrading UBXN7, which elevates the IKK-β level and activates the NF-κB signaling pathway. Therefore, drug discovery focusing on the modulation of inflammatory pathways to inhibit HBV immune evasion holds significant promise for the development of novel therapeutic agents against HBV. Significant progress has been made in this area, e.g., a biphenyl amide p38 MAPK inhibitor NJK14047 significantly reduces HBV pgRNA and cccDNA. However, further studies are still required to fully explore its potential. In this context, natural products, with their rich diversity of bioactive molecules, have emerged as a particularly promising source for identifying potential candidates.

Essential oils (EOs), significant components of aromatic plants, possess diverse pharmacological properties, such as antioxidant, anti-inflammatory, neuroprotective, and antimicrobial activities. − Among these, inflammation modulation and antivirus stand as the two primary ones. , Plenty of EOs have been demonstrated to modulate inflammatory pathways and inhibit the release of mediators by inflammatory cells. , Previous studies have also revealed EOs’ efficacy against a variety of viruses, including influenza virus, herpes simplex virus (HSV), dengue virus (DENV), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), HBV, etc. − Nevertheless, there is currently a lack of sufficient evidence linking the inflammatory pathway modulation and antiviral activity of EOs. Considering that controlling the strong immune evasion of HBV is an important and feasible option to cure it, screening immune-modulating EOs to confer anti-HBV activity holds significant research value.

Continuing our research on the antiviral properties of EOs and their mechanism of action, − we conducted a comprehensive study on a large number of EOs to identify new HBV inhibitors. First, we meticulously collected 292 plant samples from diverse global locations in over 30 countries, representing 32 families and 98 genera. Subsequently, we extracted the EOs from these samples and analyzed the chemical composition of the EOs using gas chromatography–mass spectrometry (GC-MS). In parallel, we randomly selected 274 EOs to test their anti-HBV activity for the first batch. This evaluation focused on their ability to inhibit HBsAg and HBeAg secretion and cytotoxicity. Utilizing these data, we developed a virtual screening and prediction model for potential anti-HBV EOs. The accuracy and practical applicability of the model were then validated using an additional 18 EOs. The application of machine learning in virtual screening has significantly accelerated drug discovery, and our model represents a critical advancement in the identification of HBV inhibitors. Through this systematic determination and prediction of anti-HBV activities of 292 EOs, we have provided the landscape of the anti-HBV activities in a large number of EOs with a comprehensive overview of their potential in combating HBV.

Among the hundreds of EOs, we first found that EOs extracted from Acorus tatarinowii Schott (ATEOs) exhibited the strongest inhibition of HBsAg secretion, marking them as potent anti-HBV candidates. A. tatarinowii has a well-established history of medicinal application, particularly in the treatment of neurological and psychiatric diseases with its major constituents, α- and β-asarones, such as stroke, dementia, depression, seizure, Alzheimer’s disease, and various mental disorders. − Additionally, the ethanol extract of A. tatarinowii has demonstrated antifungal properties, potentially contributing to strategies aimed at combating antibiotic resistance. The mechanisms by which ATEOs exert their biological activities are highly relevant to their inflammatory pathway regulations. For instance, ATEOs alleviate neuroinflammation by inhibiting the NF-κB signaling pathway and thus NLRP3 inflammasome activation. ATEOs’ major constituent, β-asarone, suppresses TNF-α expression through DNA methylation and c-Jun-mediated transcription modulation. Also, another study found that β-asarone exhibits anti-inflammatory effects by inhibiting NF-κB signaling and the JNK pathway in LPS activated BV-2 microglial cells. Despite these diverse therapeutic applications, it is noteworthy that the antiviral potential of A. tatarinowii remains largely unexplored. The mechanisms and specific targets of any potential antiviral activity remain to be elucidated, representing a novel area for research. Therefore, we investigated the primary anti-HBV components of ATEOs, and we analyzed the anti-HBV pathway via transcriptomic, network pharmacology analyses, and Western blotting assays.

Consequently, given the extensive antiviral and immunoregulatory potential of EOs, this study aimed to comprehensively analyze the anti-HBV activity across a broad spectrum of EOs, develop a robust virtual screening and prediction model to identify potential HBV inhibitors, and further explore the active components and underlying mechanisms of ATEOs, which exhibited strong anti-HBV activity.

2. Materials and Methods

2.1. Samples and Reagents

In this study, 292 plant samples were collected and used to extract the EOs. The MTT cell viability and cytotoxicity assay kit was purchased from Dalian Meilunbio Co., Ltd. (Dalian, China). The HBsAg and HBeAg ELISA detection kits were purchased from Shanghai Kehua Bioengineering Co., Ltd. (Shanghai, China). The RNAsimple Total RNA kit, TIANamp Virus DNA kit, and FastKing gDNA Dispelling RT SuperMix were purchased from Tiangen Biotech Co., Ltd. (Beijing, China). The ChamQ Universal SYBR qPCR Master Mix Q711D was obtained from Vazyme Biotech (Nanjing, China). p38 MAPK antibody (A0227), Phospho-p38 MAPK-T180/Y182 antibody (AP0526), NF-κB p65/RelA antibody (A2547), Phospho-NF-κB p65/RelA-S536 antibody (AP0475), JNK1/2/3 antibody (A4867), Phospho-JNK1-T183/Y185+JK2-T183/Y185 + JNK3-T221 antibody (AP1337), c-Jun antibody (A16905), Phospho-c-Jun-363 antibody (AP0048), ERK1/ERK2 antibody (A10613), Phospho-ERK1-T202+ERK2-T185 antibody (AP0485), c-Fos antibody (A0236), and Phospho-c-Fos-T232 antibody (AP0038) were purchased from Abclonal Biotechnology Co., Ltd. (Wuhan, China).

2.2. Preparation and Gas Chromatography–Mass Spectrometry (GC–MS) Analysis of Essential Oils (EOs)

Initially, 292 plant samples were collected, and their EOs were obtained by steam-distillation, pressing, and solvent extraction. Detailed information on all the aforementioned samples is shown in Supporting Information 2. Subsequently, GC–MS (Agilent US15173042) was applied to analyze the chemical ingredients of the EOs according to our previously reported method. Chromatographic separation was accomplished on a HP-5MS column (130 × 0.25 mm i.d.; film thickness: 0.25 mm; Agilent, USA). Helium was used as the carrier gas at a flow rate of 1.5 mL/min. The ATEOs were diluted with n-hexane (analytically pure) to 1 mg/mL, and 0.2 μL of sample was injected for analysis. The temperature was set as follows: 50 °C for 2 min, increased to 220 °C at a rate of 10 °C/min and held for 10 min, finally increased to 250 °C and held for 1 min. The injector temperature was set at 250 °C. The mass-selective detector was operated in an electron-impact ionization (EI) source (70 eV) and was used in full scan mode with a mass scan range from 50 m/z 50 to 550 m/z. The ingredients of these EOs were identified by NIST 14 (National Institute of Standards and Technology, Gaithersburg, MD, USA).

2.3. Cell Culture

Human hepatocellular carcinoma cells, HepG2.2.15 and HepAD38, were obtained from American Type Culture Collection (ATCC, Rockville, USA). The cells were cultured in MEM, supplemented with 10% FBS and 380 μg/mL G418, and incubated at 37 °C supplemented with 5% CO2. For HepAD38 cell culture, 400 ng/mL tetracycline was added into MEM.

2.4. Determination of Cell Viability and Inhibitory Rates of HBsAg and HBeAg

Cell viability after treating with EOs was measured with an MTT assay according to the manufacturer’s protocol (Dalian Meilunbio Co., Ltd.). Briefly, the HepG2.2.15 and HepAD38 cells (1.5 × 104 cells/well, 400 μL of medium/well) were cultured for 24 h and incubated for 72 h with the addition of EOs to a final concentration at 0.2 mg/mL. DMSO was used as the solvent for the EOs. PAC5 (30 μM, a derivative of sesquiterpenoid, can significantly eliminate the HBV cccDNA level and reduce HBV-related antigen expression) was also included as reference drug. Followed by adding 20 μL of MTT solution (5 mg/mL) to each well and incubating at 37 °C with 5% CO2 for 4 h, the culture medium was removed and 200 μL of DMSO (dimethyl sulfoxide) was added. The absorbance was recorded at 490 nm. The inhibitory rates of HBsAg and HBeAg were measured by HBsAg and HBeAg ELISA detection kits (Shanghai Kehua Bioengineering Co., Ltd.). Briefly, after treating the cells with EOs for 72 h, the culture supernatants were collected and the levels of HBsAg and HBeAg were detected according to the manufacturer’s protocol. To determine the IC50 and CC50 of EOs, cells were treated with various final concentrations (0.0125, 0.025, 0.05, 0.1, 0.2, and 0.4 mg/mL) of EOs for 72 h.

2.5. Construction and Processing of Essential Oil Molecular Features

Among the total 292 EOs obtained, we randomly selected 274 EOs and processed their component data from GC–MS analysis to create feature vectors of length 1184 in our training data set (Supporting Information 2). The remaining 18 EOs were used for the subsequent validation of model performance. Each vector position represents a specific compound’s proportion relative to the total compounds in the essential oil, with absent compounds assigned a value of zero. To assess the significance of each feature, we utilized the “feature_importances” function of the Random Forest (RF) algorithm. Various thresholds were applied to construct models to determine the optimal feature selection threshold. Furthermore, we attempted to transform the feature vector values into a binary form (0 for absence and 1 for the presence of a compound). This transformation improved the performance of the classification and regression models; therefore, we applied this transformation in our models.

2.6. Development of Classification Models

We developed classification models to predict anti-HBsAg activity and cytotoxicity using RF and Support Vector Machine (SVM) algorithms, both implemented through Scikit-learn. For the optimization of hyperparameters in the RF model, we adjusted “{n_estimators}”, and for the SVM model, we adjusted “{C}”; other parameters were set to their default values. Due to the imbalance in the data set, the F1 score was chosen as the evaluation metric for hyperparameter tuning. A grid search with 5-fold cross-validation was performed to identify the optimal parameters. The final prediction models were trained using the entire training data set with the determined optimal parameters. Due to the imbalance of the data set, we attempted to use SMOTE oversampling. However, although this method improved the model’s recall, it also reduced the precision (Figure S2A), and thus, we ultimately did not adopt oversampling.

2.7. Development of Regression Models

To predict anti-HBsAg activity and cytotoxicity, we also constructed regression models using RF and Support Vector Regression (SVR) methods within Scikit-learn. We adjusted “{n_estimators}” for the optimization of the hyperparameters of the RF model and “{C, kernel, gamma}” for the optimization of the hyperparameters of the SVR model. Other parameters were set to their default values. R 2 and MSE were used as evaluation metrics for hyperparameter tuning. A grid search with a 10-fold cross-validation was conducted to find the optimal settings. The data set was randomly split into training (90%) and test sets (10%) using the “train_test_split” function. After 100 iterations of random splits, the model that performed best on the test set was chosen as the final prediction model.

2.8. t-Distributed Stochastic Neighbor Embeding (t-SNE) Analysis

For visualization, we applied t-SNE using the “TSNE” function from “sklearn.manifold” on the EO feature vectors. Jaccard distance was used as the metric, which reflects the compositional differences between EOs: the more dissimilar the composition, the greater the distance between them on the plot. The perplexity parameter was set to 5 to accommodate the relatively small data set and achieve optimal visualization.

2.9. Determination of Inhibition of pg-RNA and Mature DNA

The pg-RNA was analyzed by RT-qPCR. Briefly, the HepAD38 cells were treated with ATEOs, α-, and β-asarones (0.2 mg/mL), for 6 days (treatments were changed once on day 3). Entecavir (ETV, 30 μM, a nucleos­(t)­ide analogue) can inhibit HBV replication and therefore was included as reference drug. The RNA was extracted using RNAsimple Total RNA kit (Tiangen Biotech Co., Ltd., Beijing, China) according to the manufacturer’s protocol. RT-qPCR was performed using FastKing gDNA Dispelling RT SuperMix (Tiangen Biotech Co., Ltd., Beijing, China) with the forward primer 5′-TGTTGTTAGACGACGAGGCA-3′ and the reverse primer 5′-TTCCCCACCTTATGAGTCCA-3′.

The levels of mature DNA encapsulated in Dane particles and secreted out of cells were quantified by the qPCR assay. Briefly, the HepAD38 cells were treated with ATEOs, α-, and β-asarones (0.4 mg/mL), for 6 days (treatments were changed once on day 3). The supernatant was collected after 6 d of drug treatment, and the cellular HBV DNA in the supernatant was extracted using TIANamp Virus DNA kit (Tiangen Biotech Co., Ltd., Beijing, China). The qPCR experiment was performed using ChamQ Universal SYBR qPCR Master Mix Q711D (Vazyme Biotech Co., Ltd., Nanjing, China) according to the manufacturer’s protocol. The primer sequences were as follows: forward 5′-GCTGGATGTGTCTGCGGC-3′, and reverse 5′-GAGGACAAACGGGCAACATAC-3′.

2.10. Network Pharmacology Analysis

Hepatitis B virus-related target genes were collected from Genecards (https://www.genecards.org/), and genes with relevance scores of ≥10 were chosen for analysis. Potential target proteins of the ingredients of ATEO zy-10 were collected by using the SEA Search server (https://sea.bkslab.org/). The Venn analysis tool was applied to obtain the common target genes between the predicted target of ATEO zy-10 and the HBV-related targets. The protein–protein interaction (PPI) was analyzed by the STRING database (https://cn.string-db.org/cgi/input.pl), with “Homo sapiens” defined as the current setting and a confidence level of 0.15. The relationships between active ingredients of ATEO zy-10, the related target gene, and HBV were visualized using Cytoscape 3.9.1. In the graphical network, each constituent and gene were described by the node, and edges encoded the interactions. The Database for Annotation, Visualization and Integrated Discovery (DAVID, https://david.ncifcrf.gov/home.jsp) and bioinformatics platform (http://www.bioinformatics.com.cn/) were applied to Gene Ontology (GO) enrichment analysis (including biological process, BP; molecular function, MF; and cellular component, CC) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.

2.11. Transcriptome Analysis

The HepG2.2.15 cells were cultured and treated with ATEOs and α- and β-asarones as described above. The total RNA was extracted by RNAsimple Total RNA Kit (Tiangen Biotech Co., Ltd., Beijing, China). Whole transcriptome sequencing was performed by Beijing Annoroad Gene Technology Co., Ltd. Data were analyzed with software and tools including fastQC, Hisat2, and featuresCounts.

2.12. Western Blotting Analysis

The HepG2.2.15 cells (1 × 105 cells/well, 2 mL of medium/well) were cultured overnight and incubated for 6 days with the addition of ATEO at a final concentration of 0.1 mg/mL. The cell lysates were collected, and the total cell proteins were extracted by treating with RIPA buffer. The protein samples were separated by SDS-PAGE gel electrophoresis and transferred to PVDF membranes. After blocking with 5% skim milk for 1.5 h, the membranes were incubated with primary antibodies overnight at 4 °C. Followed by incubating secondary antibodies for 1 h at room temperature, the membranes were exposed to enhanced the chemiluminescence substrate detection solution and then imaged (Gel view, GV6000). ImageJ software was applied to quantify the band density.

2.13. Statistical Analysis

Data were statistically analyzed by GraphPad Prism 10.1 software. Results are expressed as mean ± SD based on at least three independent experiments (n ≥ 3). Error bars represent the standard deviation of the mean value. Significant differences between the means of control and treated cells with p-value <0.05 were considered statistically significant. Part of the results were visualized using Weishengxin and CNSknowall platform.

3. Results

3.1. Diverse Sources and Composition of Essential Oils (EOs) from 292 Plant Samples

We conducted a comprehensive collection of 292 plant samples, extracted their EOs, and analyzed the compositions of EOs using gas chromatography–mass spectrometry (GC–MS). As shown in Figure A, the samples originated from over 30 countries, with about one-third sourced from more than ten regions in China. Significant contributions also came from India, Australia, Indonesia, and Italy. The plant samples covered a wide taxonomic diversity, spanning 32 families and 98 genera (Figure B). EOs were extracted from different plant parts with fruits being the predominant source (19.18%), followed by leaves, resins, peels, flowers, and roots (Figure C). The GC–MS analysis revealed that the number of constituents in the EOs varied considerably, ranging from 2 to 95, with a majority of EOs containing between 40 and 50 constituents (Figure D). In total, we identified 1184 individual components, primarily including 225 monoterpenes, 312 sesquiterpenes, and 179 aromatic compounds. These diverse sources and compositions of EOs provide a solid foundation for aromatherapy applications.

1.

1

Open in a new tab

Landscape of the anti-HBV activity of EOs. (A) The geographical distribution of EOs. (B) The botanical classification of EOs by family and genus. (C) The statistics on the sources of plant parts for EOs. (D) The composition quantities of the components of EOs. (E) The anti-HBsAg activity of EOs. (F) The anti-HBeAg activity of EOs. (E) The cytotoxicity of EOs.

3.2. Exploring the Anti-HBV Potential in a Large Number of EOs

To evaluate the anti-HBV potential within a diverse range of EOs, we conducted a systematic evaluation of 274 EOs for their ability to inhibit the secretion of HBV markers, HBsAg and HBeAg. Concurrently, we assessed the cytotoxicity of these EOs to ensure their safety for potential therapeutic use. As shown in Figure E–G and detailed in Supporting Information 2, 49 of the tested EOs demonstrated significant HBsAg-inhibition activity, achieving inhibition rates greater than 40% at a concentration of 0.2 mg/mL. Several EOs exhibited promising anti-HBV potential, particularly those derived from Acorus tatarinowii, Caryopteris forrestii, and Vitex agnus-castus. These findings not only mapped the anti-HBV activities across a broad spectrum of EOs but also highlighted specific EOs that may be valuable candidates for further exploration and development as therapeutic agents for chronic HBV infections. Our results indicate the potential of these EOs as natural alternatives in the fight against HBV.

3.3. Effective Machine Learning Models for Identifying EOs with Anti-HBV Potential

Using the data gathered from the experiments described above, we developed a virtual screening and prediction model to identify potential HBV inhibitors for aromatherapy (Figure A). This model integrates the compositional data obtained from GC–MS analysis with the results of HBV activity assays, specifically focusing on anti-HBsAg, anti-HBeAg, and cytotoxicity. Among the total 292 EOs obtained, we randomly selected 274 EOs and identified 1184 compounds using the proportion of each compound in the EOs as a feature vector. The labels of anti-HBsAg, anti-HBeAg, and cytotoxicity rates of EOs were used to assemble our training data set. We then attempted to establish classification and regression machine learning models using RF and SVM. Considering the small and high imbalance of the data set (49 EOs with more than 40% anti-HBsAg activity, 9 EOs with more than 40% anti-HBeAg activity, and 200 EOs with less than 30% cytotoxicity), it became challenging to simultaneously use high anti-HBV activity and low cytotoxicity as criteria for active compounds in the classification model. Therefore, after exploring various approaches such as activity threshold adjustments, feature filtering, and model parameter tuning, we opted to build separate models to predict anti-HBV activity and cytotoxicity (Supporting Information 3).

2.

2

Open in a new tab

Construction of virtual screening and prediction model of potential HBV inhibitors. (A) Schematic diagram of essential oil feature generation and model construction. (B–C) Performance of the classification models for predicting anti-HBsAg activity and cell toxicity. The values of the evaluation metrics are the averages from 20 iterations of holdout cross-validation. (D–E) Performance of the regression models for predicting anti-HBsAg activity and cell toxicity. The evaluation metrics are R 2 and MSE for the test data set (blue dots shown in the figures). (F) t-SNE plot of the training data set and the data set of EOs to be predicted. EOs with good anti-HBV activity in the training set are shown with red dots; EOs to be predicted are shown with yellow dots. (G) The distribution of 87 compound features filtered from data set comprised 1184 compounds in active and other EOs.

The machine learning model for anti-HBsAg activity (using 40% inhibition rate as the threshold value) and cytotoxicity (using 30% as the threshold value) showed relatively good prediction results (Figure B–E). However, due to the limited number of EOs with significant anti-HBeAg activity (inhibition rates over 40%), the corresponding model for anti-HBeAg activity was less effective (Figure S1A and B). Considering that HBsAg levels are highly correlated with HBV replication and are the key indication of HBV infection, we focused solely on the anti-HBsAg data in conjunction with cytotoxicity data to construct the virtual screening and prediction model for potential HBV inhibitors.

For the classification models, we compared the performance of RF and SVM. After determining the optimal parameters (Figure S3A and D), we quantified the overall model performance by averaging various classification metrics, such as accuracy, AUC, precision, recall, and F1 score, over 20 iterations of holdout cross-validation (Figure B, C and Figure S2B, C). The SVM model outperformed the RF model overall, with thresholds of 40% anti-HBsAg activity and 30% cytotoxicity, yielding the best performance. Despite the limitations inherent to our small data set, both prediction models achieved a precision exceeding 0.72, demonstrating their effectiveness.

For the regression models, we also compared the performance of RF and SVR. After identifying the optimal parameters (Figure S3B, C, E, and F), the final models were trained on 90% of the training data set, and their performance was quantified by the R2 and MSE values on the remaining 10% test data set (Figure D, E and Figure S2D). The regression prediction models for both anti-HBsAg activity and cytotoxicity performed better with SVR. Specifically, the R 2 and MSE for the anti-HBsAg activity regression prediction model were 0.674 and 0.054 (Figure D), while those for the cytotoxicity regression prediction model were 0.730 and 0.020 (Figure E), indicating a certain predictive capability for both models.

3.4. Mapping Diversity of EOs’ Composition and Validation of the Prediction Model Using New EOs

To investigate the similarity and spatial distribution of EOs based on their compound composition, we conducted a t-distributed stochastic neighborhood embedding (t-SNE) dimensionality reduction analysis on a data set of 292 EOs characterized by GC–MS data. These data include 274 EOs used to construct our machine learning model training data set and 18 EOs lacking anti-HBV activity experimental data (Supporting Information 2). As in the construction of the machine learning model, we used the compound components of the EOs as feature vectors for t-SNE analysis. In our analysis, five EOs exhibited more than 40% anti-HBsAg activity and less than 30% cytotoxicity, which were labeled as “active”. The remaining 269 EOs were labeled as “inactive”, while the 18 EOs awaiting activity predictions were designated as “new_data” (Figure F). The t-SNE results revealed that all three categories of EOs were dispersed throughout the analysis space, indicating substantial variation and diversity in the proportions of the compound components. However, only a small number of EOs displayed significant compositional differences from the others. Importantly, all components of the 18 EOs to be predicted were included in the 1184 molecular features of the training data set, suggesting that our models can effectively generate meaningful predictions for these samples. Nevertheless, due to the complexity of composition, although the map leveraged composition-only features to enable anti-HBV activity screening in EOs, it offered limited mechanistic interpretability.

Subsequently, we applied both classification and regression models to predict the anti-HBsAg activity and cytotoxicity of the 18 EOs. As shown in Table , the majority of predictions from the regression model were in agreement with those from the classification model. Combining the results from both classification and regression models, predictions were considered correct if at least one prediction result from either model agreed with the experimental results. The accuracy in predicting anti-HBsAg activity and cytotoxicity reached 83.3% and 66.7%, respectively. These external data validation results further confirm the robustness and validity of our predictive model.

1. Prediction of the Anti-HBV Activity of 18 New EOs .

Bench_code anti-HBsAg classification anti-HBsAg regression (%) experimental verification (%) prediction correctness cytotoxicity classification cytotoxicity regression (%) experimental verification (%) prediction correctness
jx-139(1) 1 44.9 70.4 yes 1 31.6 75.2 yes
jx-6 1 29.9 86.4 yes 0 15.3 23.5 yes
sx-52 0 –3.3 –1.8 yes 0 5.9 13.5 yes
sx-53 0 –10.0 –3.6 yes 0 16.9 8.2 yes
sx-60 0 –8.6 –4.5 yes 0 20.5 28.3 yes
zy-pl-108 0 –3.3 4.6 yes 0 17.8 7.7 yes
zy-pl-109 0 17.4 9.9 yes 0 15.8 11.3 yes
zy-pl-110 0 37.2 2.4 yes 1 38.6 19.1 no
zy-pl-130 0 26.8 94.2 no 1 48.7 90.6 yes
zy-pl-18 1 43.1 72.9 yes 0 4.9 38.6 no
zy-pl-20 0 7.0 –16.0 yes 0 16.4 57.7 no
zy-pl-50 0 32.7 36.7 yes 0 30.9 4.7 yes
zy-pl-51 0 32.7 20.5 yes 0 30.9 0.5 yes
zy-pl-52 0 32.7 33.4 yes 0 30.9 8.3 yes
zy-pl-53 1 11.8 59.0 yes 0 29.3 17.8 yes
zy-pl-54 0 –27.4 70.0 no 0 –3.2 40.4 no
zy-pl-55 0 2.3 79.6 no 0 4.9 34.8 no
zy-pl-6 0 –46.3 1.8 yes 0 –7.7 33.6 no
Open in a new tab
a

It is marked as “yes” in the “prediction correctness” column when at least one of the prediction results from either the classification or regression model was consistent with the experimental trend; otherwise, it is marked as “no”.

3.5. Uncovering Promising Anti-HBV EOs and Their Key Bioactive Components

In our analysis of the HBV inhibitory activities of the EOs utilized for modeling and prediction, we established criteria for highly active anti-HBV EOs. These criteria required an HBsAg inhibitory rate exceeding 40%, a cell toxicity below 30%, and an HBsAg inhibitory rate to cell toxicity ratio greater than 2. Following this assessment, we identified five EOs that exhibited high anti-HBV activity. Among these, two derived from Acorus tatarinowii, one derived from Vitex agnus-castus, one derived from Caryopteris forrestii, and one derived from Artemisia absinthium. The IC50 and CC50 values for these EOs were determined and are presented in Table S1, where the selectivity index (SI), calculated as CC5 0 against HBsAg divided by IC5 0, exceeded 2. Among the 292 EOs examined, the essential oil from Acorus tatarinowii (ATEOs) consistently demonstrated potent anti-HBV activity, establishing it as one of the most promising candidates for HBV inhibition.

In the 1184 molecular features of our virtual screening and prediction model, we identified 87 compound features that exhibited a strong correlation with the anti-HBV activity. Among these, six components were consistent with the potential active ingredients found in Artemisia argyi essential oil, which has documented anti-HBV activity (CAS number: 007212–44–4, 000099–85–4, 005989–27–5, 000127–91–3, 000470–82–6, 006753–98–6). Additionally, terpinen-4-ol (CAS number: 000562–74–3) was noted for its ability to impede the normal viral membrane fusion process. Meanwhile, most of the remaining identified compounds have not previously been reported with anti-HBV activity. The distribution of these features among highly active EOs and other EOs is illustrated in Figure G and detailed in Table S2. Our analysis revealed that α- and β-asarones were the predominant components in several highly active EOs, particularly in ATEOs, while being largely absent in other EOs. These findings suggested that these compounds may represent the most promising individual components for anti-HBV activity, and ATEOs have significant potential as HBV inhibitors. The key components of ATEOs, α- and β-asarones, appear to be critical in conferring anti-HBV activity.

3.6. Potent Anti-HBV Activity of EOs from Acorus tatarinowii (ATEOs)

We conducted a comprehensive evaluation of the anti-HBV activity of ATEOs that showed promising results during the initial screening. Our focus was on three specific ATEO samples: zy-10, jx-6, and zy-pl-18. As illustrated in Figure and Figure S4, these ATEOs exhibited significant inhibition of HBsAg in both HepG2.2.15 and HepAD38 cell lines with IC50 values below 0.11 mg/mL. Furthermore, these ATEOs also effectively suppressed HBeAg levels in HepG2.2.15 cells, achieving IC50 values under 0.17 mg/mL. The CC50 values of ATEOs in HepG2.2.15 and HepAD38 cell lines were higher than 0.22 and 0.13 mg/mL, respectively. This indicates that cell viability remained intact at concentrations of the IC50 values. Our investigation further revealed that ATEOs inhibited the production of mature HBV DNA and pg-RNA at subcytotoxic concentrations (Figure S5), indicating that ATEO treatment effectively impeded HBV replication and maturation at concentrations with a small impact on cell viability. Collectively, these findings demonstrate that ATEOs consistently inhibit the replication of HBV without compromising cell viability.

3.

3

Open in a new tab

ATEOs showed anti-HBV activity in terms of HBsAg and HBeAg inhibition with low cytotoxicity in HepG2.2.15 cells: (A) ATEO zy-10, (B) ATEO jx-6, and (C) ATEO zy-pl-18.

3.7. Identifying α- and β-Asarones as Key Inhibitors of HBV in ATEOs

As presented in Figures S6–S7 and Tables S3–S5, the GC–MS analysis identified that the chemical composition of ATEOs was predominantly composed of aromatics. Specifically, the aromatic content reached 87.59% in zy-10, 77.88% in jx-6, and 99.99% in zy-pl-18. Additionally, jx-6 was found to contain 13.67% terpenoids. Among these, α- and β-asarones served as the primary components of ATEOs, which aligns with previous reports. , Subsequently, we evaluated the anti-HBV activity of α- and β-asarones. As shown in Figure and Figure S8, both compounds effectively inhibited HBsAg in HepG2.2.15 and HepAD38 cells with IC50 values below 0.10 mg/mL, as well as HBeAg in HepG2.2.15 with IC50 values below 0.33 mg/mL. The cell viability (CC50 values higher than 0.29 mg/mL in HepG2.2.15 and 0.13 mg/mL in HepAD38, respectively) remained unaffected at a concentration of IC50 values of HBsAg. Furthermore, α- and β-asarones demonstrated inhibitory effects on mature HBV DNA and pg-RNA (Figure S9). These outcomes align with the results observed for ATEOs, suggesting that α- and β-asarones are the primary active components responsible for the anti-HBV activity of ATEOs.

4.

4

Open in a new tab

α- and β-asarones were identified as the main anti-HBV components of ATEOs with the consistent anti-HBV activity of ATEOs. (A–B) The molecular structures of α- and β-asarones. (C–D) α- and β-asarones showed anti-HBV activity in terms of HBsAg and HBeAg inhibition with low cytotoxicity in HepG2.2.15 cells.

3.8. Network Pharmacology Insights into the Potential Anti-HBV Mechanisms of ATEOs

To further explore the potential anti-HBV mechanisms of ATEOs, we conducted network pharmacology analysis. Owing to its superior SI and representative chemical profile, zy-10 was chosen from the three ATEOs for mechanistic studies. As shown in Figure , we identified 57 potential targets using the SEA target prediction server by analyzing six major components of ATEO zy-10. Additionally, we identified 2603 HBV-related disease genes using GeneCard. Our analysis revealed an overlap of nine potential genes between the predicated targets and disease genes (Figure A and C). The protein–protein interaction (PPI) network of these nine target genes indicated that JUN, FOS, and NFKB1 exhibited a higher degree of connectivity (Figure B). Moreover, we performed Gene Ontology (GO) enrichment analysis, which covered categories including Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). The BP enrichment results highlighted processes associated with transcription regulation, inflammatory responses, immune responses, and receptor-coupled signaling pathways, particularly regulation of the extracellular signal-regulated protein kinase (ERK) cascade (Figure D). Furthermore, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis using the David bioinformatics resource system. The top 30 singling pathways identified encompassed various categories, including viral diseases (e.g., Hepatitis B, Kaposi sarcoma-associated herpesvirus, and Human immunodeficiency virus 1), parasitic diseases (e.g., Leishmaniasis and Chagas disease), bacterial diseases (e.g., Yersinia infections), immune-related pathways (e.g., B cell receptor signaling pathway, Toll-like receptor signaling pathway, and T cell differentiation), cancer-related pathways (e.g., PD-1 checkpoint pathway in cancer and chemical carcinogenesis), etc. (Figure E). The results suggested that ATEOs may exert their antiviral effects through the modulation of regulating transcription and the stimulation of inflammatory and immune responses.

5.

5

Open in a new tab

Network pharmacology study showed that ATEO zy-10 may exert antiviral effects by regulating transcription and stimulating inflammatory and immune responses via the MAPK signaling pathway (comprised JUN, FOS, and NFKB1). (A) Venn diagram of compound targets of ATEO zy-10 and HBV-related targets. (B) The protein–protein interaction (PPI) network of 9 related targets. The bigger the degree of connection, the higher the degree of connection. (C) Compounds-target-disease network diagram. The green represents ATEO zy-10, the purple represents compounds, the blue represents disease, and the yellow represents targets. The right nodes represent 9 overlapping targets of ATEO zy-10 and HBV-related targets. (D) Gene Ontology (GO) enrichment analysis, including biological process, molecular function, and cellular component. (E) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis carried out by the DAVID database and visualized by the bioinformatics platform.

3.9. Transcriptomic Insights into ATEO’s Anti-HBV Activity via the MAPK Pathway

To further elucidate the mechanism of ATEOs counteracting HBV infection, we conducted a comprehensive transcriptomic analysis of ATEO (zy-10), α-, and β-asarones. Principal component analysis (PCA) revealed a clear separation between the ATEO-treated and DMSO-treated control samples (Figure A). Volcano plots demonstrated that a 6-day treatment with ATEO resulted in transcriptional alterations in 3650 genes, with 1804 genes showing upregulation and 1846 genes exhibiting downregulation (Figure B and C). We visualized the differentially expressed genes (DEGs) from all samples by using a clustered heat map (Figure D).

6.

6

Open in a new tab

Transcriptome analysis suggested that ATEO zy-10 might exert anti-HBV activity by modulating infection- and immune-related pathways. (A) Principal component analysis of subjects in ATEO zy-10 and DMSO. (B) The volcano plot of DEGs between ATEO zy-10 vs DMSO. (C) The bar graph of DEGs between ATEO zy-10 vs DMSO. (D) Hierarchical clustering heat map analysis of changing genes among treatment groups. (E) KEGG enrichment analysis of the DEGs.

Subsequent KEGG enrichment analysis of the DEGs (Figure E) indicated that ATEO treatment resulted in the regulation of infection-related pathways (e.g., Human T-cell leukemia virus 1, Human papillomavirus, Shigellosis, and Salmonella infections), immune-related pathways (e.g., autophagy, endocytosis, and MAPK signaling pathway), metabolic pathway, and neuro-related pathways (e.g., neurotrophin pathway, neurodegeneration pathway, and Alzheimer's disease). The infection- and immune-related pathways showed strong concordance with our network pharmacology analysis findings.

Of particular interest was the identification of the MAPK signaling pathway as a potential key mediator of ATEO for anti-HBV activity. This pathway, which encompasses the c-Jun/JNK, c-Fos/ERK, and NF-κB pathways (Figure S10), emerged as a central player in the mechanism of action. The prominence of the MAPK signaling pathway in our findings is especially noteworthy, as it aligns seamlessly with the results obtained from our network pharmacology analysis. This provides robust corroboration of our initial hypotheses. To further substantiate these findings, we conducted comprehensive transcriptome analyses focusing on α- and β-asarones, two primary components of ATEOs. The results of these analyses were particularly illuminating, as they provided additional evidence of the involvement of these compounds in the regulation of the MAPK signaling pathway (Figures S11 and S12).

3.10. Inhibition of the MAPK Pathway by the ATEO

To validate the involvement of the MAPK signaling pathway in the anti-HBV activity of ATEOs, we conducted Western blotting experiments on HepG2.2.15 cells treated with ATEO. The results demonstrate that ATEO (zy-10) treatment significantly impacts key proteins in the MAPK signaling pathway. Specifically, ATEO reduced the expression of c-Jun, JNK, p38 MAPK, and NF-κB, as shown in Western blotting and quantified by bar charts (Figure A and B). Furthermore, ATEO treatment resulted in a significant reduction in the phosphorylation levels of c-Jun, p38 MAPK, and NF-κB compared to the control group (p < 0.05) (Figure C). Interestingly, ATEO treatment led to an enhanced phosphorylation of c-Fos/ERK, despite a decrease in total ERK expression. This differential regulation of ERK signaling may indicate that ATEO modulates the MAPK pathway via a more complex mechanism.

7.

7

Open in a new tab

ATEO zy-10 altered the expression and phosphorylation of the MAPK signaling pathway. (A) The Western blotting analysis of c-Jun, JNK, c-Fos, ERK, p38 MAPK, NF-κB, and p53 in HepG2.2.15 treated with ATEO zy-10. (B) The analysis of protein levels after ATEO zy-10 treatment. (C) The analysis of phosphorylation levels after ATEO zy-10 treatment. The data are presented as mean ± SD from three replicate experiments. A p-value of less than 0.05 was considered statistically significant between the treatment and control.

In the context of persistent HBV infection, the HBx protein plays a crucial role in the HBV lifecycle by suppressing the inflammatory response. This is achieved by upregulating IL-35 expression via the activation of the c-Jun/JNK signaling pathway. HBx promotes ubiquitination degradation of UBXN7, subsequently activating NF-κB, leading to increased autophagy and enhanced HBV replication. Furthermore, HBsAg has been shown to enhance p38 MAPK phosphorylation and increase the expression of IL-12 p40, IL-12 p70, and IL-10 through the p38 MAPK and NF-κB pathways, thereby promoting HBV replication. AP-1, a heterodimer formed through either the homodimerization of Jun proteins or the heterodimerization of Jun and Fos proteins, augments the HBV core promoter activity. The transcription activity of HBV was reduced after the inhibition of c-Jun and c-Fos. Consequently, ATEO treatment led to decreased levels of c-Jun/JNK and NF-κB protein, coupled with the inhibition of c-Jun/JNK, p38 MAPK, and NF-κB activation. These changes may suppress the expression of relevant inflammatory factors, reduce autophagy, and alter HBV genome transcription, collectively inhibiting HBV replication. A previous study reported that protocatechuic acid inhibited HBV replication by activating the ERK1/2 pathway and subsequently inhibiting hepatocyte nuclear factors (HNF4α and HNF1α) in HepG2.2.15 cells, which may align with the mechanism of action of ATEO.

In addition, ATEO treatment did not affect p53 expression, suggesting that its modulation of the MAPK signaling pathway does not induce apoptosis or adversely affect cell viability. In conclusion, ATEO may act on the MAPK signaling pathway to exert complex regulatory functions and ultimately exhibit anti-HBV activity.

4. Discussion

Numerous studies have demonstrated that HBV infection evades innate immunity, making it difficult to achieve a functional cure. Dysregulation of inflammatory pathways is one of the primary mechanisms, particularly the extensively reported regulation of the MAPK pathway. , EOs, key active components of ethnic medicines, have demonstrated significant anti-inflammatory and antiviral activities. EOs hold promise for inhibiting HBV immune evasion by modulating inflammatory pathways, presenting promising prospects for the development of novel therapeutic agents against HBV. In this study, we collected 292 plant samples from diverse global regions, extracted their EOs, and evaluated their anti-HBV properties. The landscape of the anti-HBV activities in a large number of EOs was provided, offering insights into potential aromatherapy applications against HBV.

Based on GC–MS and anti-HBV activity data, we developed a virtual screening and prediction model to identify potential HBV inhibitors, focusing on HBsAg inhibition as a key indicator of HBV cure. Our model, incorporating both classification and regression components, demonstrated high accuracy in predicting anti-HBV activity, providing potent technical support for the extensive discovery of EOs with anti-HBV potential. Although our model is currently limited by the size of the data set, we expect its predictive power to improve as more data on EOs are incorporated.

Furthermore, our analysis of the anti-HBV activity landscape of EOs revealed that ATEOs exhibit potent efficacy against HBV. Further investigation confirmed that ATEOs effectively inhibit HBsAg and HBV mature DNA in HepG2.2.15 and HepAD38 cell lines. Our virtual screening and prediction model suggested α- and β-asarones, the primary components of ATEOs, as potential anti-HBV active agents. In vitro experiments also confirmed this result. Although α- and β-asarones are known for their pharmacological properties, including antioxidant, antiapoptotic, anti-inflammatory, and neuroprotective effects, their antiviral activity has not been previously reported. Our findings also highlighted that ATEOs exert their anti-HBV effects by modulating the MAPK signaling pathway. Network pharmacology analysis identified nine potential ATEO targets involved in transcription regulation, inflammatory responses, immune responses, and receptor-coupled signaling pathways. The key targets, JUN, FOS, and NFKB1, are integral to the MAPK pathway and likely contribute to antiviral activity by regulating transcription and stimulating immune and inflammatory responses. These results were further supported by transcriptomic analysis, which indicated that ATEO treatment regulated several metabolic and immune-related pathways with a significant correlation to the MAPK signaling pathway. We validated the role of the MAPK signaling pathway in the anti-HBV activity of ATEOs through Western blotting experiments.

Our findings indicate that ATEOs inhibit the expression and phosphorylation of c-Jun/JNK, p38 MAPK, and NF-κB, which may result from ATEOs counteracting the effects of HBV-induced related targets, subsequently modulating relevant inflammatory factors, autophagy pathways, and HBV genome transcription. Additionally, ATEO treatment resulted in the activation of the c-Fos/ERK pathway, potentially inhibiting the expression of HNF4α and HNF1α, ultimately suppressing HBV replication. These results suggest that ATEO potentially acts on multiple targets for the complex regulation of the MAPK signaling pathway. Despite the current lack of evidence of direct targets, our results depict a mechanism by which ATEO inhibits HBV transcription and translation through the MAPK pathway. Previous studies have reported that α- and β-asarones inhibited the activation of NF-κB and c-Jun/JNK signaling pathways. , Several targets on the MAPK signaling pathway, including p38 MAPK, NF-κB, JNK, and c-Jun, have been associated with HBV infection. ,,,, Integratively, this finding strongly suggests that α- and β-asarones are likely to be the primary active ingredients in ATEOs that are responsible for their anti-HBV.

In considering the potential limitations of this study, several factors merit discussion. First, continued addition of EO data on GC–MS and anti-HBV activity is imperative for constructing the landscape of anti-HBV activity in EOs. Additionally, aromatherapy based on ATEOs needs further investigation, e.g., mode of administration, in vivo studies, etc. Moreover, more direct evidence of ATEOs’ role in the MAPK signaling pathway needs to be further explored. Despite these limitations, we conduct an in-depth and whole-process investigation into the anti-HBV activity of various EOs, and we provide new candidates for anti-HBV on the basis of inflammation modulation.

5. Conclusions

In conclusion, this study provides a comprehensive analysis of the anti-HBV activity of a large number of EOs and introduces a highly effective virtual screening and prediction model for identifying potential HBV inhibitors. The model exhibits robust predictive capabilities and broad coverage with the potential for further optimization through the incorporation of additional EOs. Importantly, we find that ATEOs effectively suppress HBsAg production and inhibit HBV DNA maturation while exhibiting low cytotoxicity. These effects are mediated by the modulation of the MAPK signaling pathway, which influences HBV genome transcription and contributes to modulating HBV’s immune regulation. α- and β-asarones were identified as the primary anti-HBV active components of the ATEOs. The comprehensive evaluation and predictive analysis of anti-HBV activities across numerous EOs provide valuable insights into the potential aromatherapy applications in HBV treatment. The findings of this research establish ATEOs as promising candidates for the inhibition of HBV with inflammation regulation properties. Further studies should be conducted to explore the mechanisms of action of ATEOs in order to support the development of more effective anti-HBV aromatherapies.

Supplementary Material

ao5c07869_si_001.pdf (2.8MB, pdf)

Acknowledgments

The authors are grateful to the members of the Analytical Group in State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany (KIB) and Analysis and Test Center, School of Life Science and Technology, Kunming University of Science and Technology for measurements of all spectra. The authors thank Kenny Xu (Dulwich International High School Programme Suzhou) for contribution to cell culture and partial activity assays. This work was supported by the National Natural Science Foundation of China (32470716), Yunnan Fundamental Research Projects (202301AS070048, 202302AG050003, and 202105AD160008), “Xingdian Talent Support Program” of Yunnan Province (XDYC–CYCX-2023-0005), and Yunnan Provincial Central Guidance for Local Science and Technology Development Fund Project (202507AB040007).

The data used and analyzed during the current study is provided in the manuscript and the Supporting Information. The anti-HBV activity data and the code of machine learning-based virtual screening framework are also openly accessible via the Essential Oils Database webservice (http://essentialoils.miningbiology.com/DownLoad.php).

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c07869.

  • Supporting Information 1: Construction and optimization of the virtual screening model; Anti-HBV activity validation of ATEOs in HepAD38 cells; Validation of inhibitory effects on HBV DNA and pg-RNA; GC–MS analysis for ATEOs; Transcriptome analysis of α- and β-asarone; Supporting Information 2: 1184 molecular features of the virtual screening model; Basic information and anti-HBV activity of 274 EOs; Supporting Information 3: Example code for the virtual screening model (PDF)

#.

G.Y., J.L., and T.S. contributed equally to this work.

G.Y.: Methodology, Data curation, Formal analysis, Investigation, Writing – original draft, Visualization. J.L.: Methodology, Validation, Visualization. T.S.: Methodology, Data curation, Validation, Visualization, Revision. W.P.: Investigation, Data curation. Q.W.: Investigation, Formal analysis. J.W.: Investigation, Data curation. X.P.: Data curation, Visualization. S.D.: Resources, Supervision, Conceptualization, Writing – review and editing, Visualization, Data curation. M.X.: Resources, Supervision, Conceptualization, Project administration, Writing – review and editing, and gave final approval of the version to be submitted. All authors gave feedback on the final manuscript and gave final approval of the version to be submitted.

The authors declare no competing financial interest.

References

  1. Yang X., Li H., Sun H., Fan H., Hu Y., Liu M., Li X., Tang H., Ou J. H. J.. Hepatitis B virus-encoded microRNA controls viral replication. J. Virol. 2017;91(10):e01919. doi: 10.1128/JVI.01919-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Martinez M. G., Boyd A., Combe E., Testoni B., Zoulim F.. Covalently closed circular DNA: The ultimate therapeutic target for curing HBV infections. J. Hepatol. 2021;75(3):706–717. doi: 10.1016/j.jhep.2021.05.013. [DOI] [PubMed] [Google Scholar]
  3. Wong G. L. H., Gane E., Lok A. S. F.. How to achieve functional cure of HBV: Stopping NUCs, adding interferon or new drug development? J. Hepatol. 2022;76(6):1249–1262. doi: 10.1016/j.jhep.2021.11.024. [DOI] [PubMed] [Google Scholar]
  4. Yardeni D., Ghany M. G.. Review article: hepatitis B-current and emerging therapies. Aliment Pharmacol Ther. 2022;55(7):805–819. doi: 10.1111/apt.16828. [DOI] [PubMed] [Google Scholar]
  5. Jeng W. J., Papatheodoridis G. V., Lok A. S. F.. Hepatitis B. Lancet. 2023;401(10381):1039–1052. doi: 10.1016/S0140-6736(22)01468-4. [DOI] [PubMed] [Google Scholar]
  6. Gane E. J.. Future anti-HBV strategies. Liver Int. 2017;37(Suppl 1):40–44. doi: 10.1111/liv.13304. [DOI] [PubMed] [Google Scholar]
  7. Verrier E. R., Colpitts C. C., Sureau C., Baumert T. F.. Hepatitis B virus receptors and molecular drug targets. Hepatol Int. 2016;10(4):567–573. doi: 10.1007/s12072-016-9718-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Kuipery A., Gehring A. J., Isogawa M.. Mechanisms of HBV immune evasion. Antiviral Res. 2020;179:104816. doi: 10.1016/j.antiviral.2020.104816. [DOI] [PubMed] [Google Scholar]
  9. Op den Brouw M. L., Binda R. S., van Roosmalen M. H., Protzer U., Janssen H. L., van der Molen R. G., Woltman A. M.. Hepatitis B virus surface antigen impairs myeloid dendritic cell function: a possible immune escape mechanism of hepatitis B virus. Immunology. 2009;126(2):280–289. doi: 10.1111/j.1365-2567.2008.02896.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Li Y., Wu C., Lee J., Ning Q., Lim J., Eoh H., Wang S., Hurrell B. P., Akbari O., Ou J. J.. Hepatitis B virus e antigen induces atypical metabolism and differentially regulates programmed cell deaths of macrophages. PLoS Pathog. 2024;20(3):e1012079. doi: 10.1371/journal.ppat.1012079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Wang L., Zeng X., Wang Z., Fang L., Liu J.. Recent advances in understanding T cell activation and exhaustion during HBV infection. Virol Sin. 2023;38(6):851–859. doi: 10.1016/j.virs.2023.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Li X., Zhu Q., Ye B., Zhu C., Dong Y., Ni Q.. JNK/c-Jun pathway activation is essential for HBx-induced IL-35 elevation to promote persistent HBV infection. J. Clin Lab Anal. 2023;37(5):e24860. doi: 10.1002/jcla.24860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Yuan S., Xu J., Wang M., Huang J., Ma S., Liu Y., Ke Y., Zeng X., Wu K., Wang J., Tian X., Zheng D., Yousaf T., Naz W., Sun J., Chen L., Guo D., Guo M., Sun G.. HBV x protein induces degradation of UBXN7, a novel negative regulator of NF-κB signaling, to promote HBV replication. Cell Mol. Gastroenterol Hepatol. 2023;15(1):179–195. doi: 10.1016/j.jcmgh.2022.09.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kim S. Y., Kim H., Kim S. W., Lee N. R., Yi C. M., Heo J., Kim B. J., Kim N. J., Inn K. S.. An effective antiviral approach targeting Hepatitis B virus with NJK14047, a novel and selective biphenyl amide p38 mitogen-activated protein kinase inhibitor. Antimicrob. Agents Chemother. 2017;61(8):e00214. doi: 10.1128/AAC.00214-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Bezerra F. W. F., Salazar M. d. L. A. R., Freitas L. C., de Oliveira M. S., dos Santos I. R. C., Dias M. N. C., Gomes-Leal W., Andrade E. H. d. A., Ferreira G. C., Carvalho R. N. d.. Chemical composition, antioxidant activity, anti-inflammatory and neuroprotective effect of Croton matourensis Aubl. leaves extracts obtained by supercritical CO2 . J. Supercrit. Fluids. 2020;165:104992. doi: 10.1016/j.supflu.2020.104992. [DOI] [Google Scholar]
  16. Ferreira O. O., Franco C. J. P., Varela E. L. P., Silva S. G., Cascaes M. M., Percário S., de Oliveira M. S., Andrade E. H. A.. Chemical composition and antioxidant activity of essential oils from leaves of two specimens of Eugenia florida DC. Molecules. 2021;26(19):5848. doi: 10.3390/molecules26195848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Franco C. d. J. P., Ferreira O. O., Cruz J. N., Varela E. L. P., de Moraes Â. A. B., Nascimento L. D. d., Cascaes M. M., Souza Filho A. P. d. S., Lima R. R., Percário S., Oliveira M. S. d., Andrade E. H. d. A.. Phytochemical profile and herbicidal (phytotoxic), antioxidants potential of essential oils from Calycolpus goetheanus (Myrtaceae) specimens, and in silico study. Molecules. 2022;27(15):4678. doi: 10.3390/molecules27154678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Mezzomo L., Butzge J., Fiorio M., Bastiani C., Mezzari A., Fuentefria A., Apel M. A.. Origanum essential oil and antifungal activity: a systematic review. Chem. Biodivers. 2025;22(6):e202402296. doi: 10.1002/cbdv.202402296. [DOI] [PubMed] [Google Scholar]
  19. de Sousa D. P., Damasceno R. O. S., Amorati R., Elshabrawy H. A., de Castro R. D., Bezerra D. P., Nunes V. R. V., Gomes R. C., Lima T. C.. Essential oils: chemistry and pharmacological activities. Biomolecules. 2023;13(7):1144. doi: 10.3390/biom13071144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pires Costa E., Maciel Dos Santos M., de Paula R. A., da Silva D. A., Lopes R. P., Teixeira R. R., Gonçalves R. V.. Antioxidant and anti-inflammatory activity of eugenol, bis-eugenol, and clove essential oil: an in vitro study. ACS. Omega. 2025;10(28):31033–31045. doi: 10.1021/acsomega.5c04146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Shen C. Y., Jiang J. G., Zhu W., Ou-Yang Q.. Anti-inflammatory effect of essential oil from Citrus aurantium L. var. amara Engl. J. Agric. Food Chem. 2017;65(39):8586–8594. doi: 10.1021/acs.jafc.7b02586. [DOI] [PubMed] [Google Scholar]
  22. Lee S. C., Hsu J. S., Li C. C., Chen K. M., Liu C. T.. Protective effect of leaf essential oil from Cinnamomum osmophloeum Kanehira on endotoxin-induced intestinal injury in mice associated with suppressed local expression of molecules in the signaling pathways of TLR4 and NLRP3. PLoS One. 2015;10(3):e0120700. doi: 10.1371/journal.pone.0120700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Oriola A. O., Oyedeji A. O.. Essential oils and their compounds as potential anti-influenza agents. Molecules. 2022;27(22):7797. doi: 10.3390/molecules27227797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Wani A. R., Yadav K., Khursheed A., Rather M. A.. An updated and comprehensive review of the antiviral potential of essential oils and their chemical constituents with special focus on their mechanism of action against various influenza and coronaviruses. Microb Pathog. 2021;152:104620. doi: 10.1016/j.micpath.2020.104620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leka K., Hamann C., Desdemoustier P., Frédérich M., Garigliany M. M., Ledoux A.. In vitro antiviral activity against SARS-CoV-2 of common herbal medicinal extracts and their bioactive compounds. Phytother Res. 2022;36(8):3013–3015. doi: 10.1002/ptr.7463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Zhang Z., Fu X., Wang Y., Wang J., Feng S., Zhao Z., Zheng L., Zhang J., Zhang X., Peng Y.. In vivo anti-hepatitis B activity of Artemisia argyi essential oil-loaded nanostructured lipid carriers. Study of its mechanism of action by network pharmacology and molecular docking. Phytomedicine. 2023;116:154848. doi: 10.1016/j.phymed.2023.154848. [DOI] [PubMed] [Google Scholar]
  27. Yao Z. W., Liu H., Zhou R., Feng M. Y., Wang F., Qin X. J., Chen X. X., Zheng C. B., Luo R. H., Yang L. M., Cen S., Xiong S. D., Liu H. Y., Zheng Y. T.. Non-volatile acylphloroglucinol components from Eucalyptus robusta inhibit Zika virus by impairing RdRp activity of NS5. Bioorg Chem. 2021;116:105303. doi: 10.1016/j.bioorg.2021.105303. [DOI] [PubMed] [Google Scholar]
  28. Chen S., Xiang J., Han J., Yang-Feng, Li H., Chen H., Xu M.. Essential oils from spices inhibit cholinesterase activity and improve behavioral disorder in AlCl(3) induced dementia. Chem. Biodivers. 2022;19(1):e202100443. doi: 10.1002/cbdv.202100443. [DOI] [PubMed] [Google Scholar]
  29. Chen S. X., Yang K., Xiang J. Y., Raymond Kwaku O., Han J. X., Zhu X. A., Huang Y. T., Liu L. J., Shen S. B., Li H. Z., Li X. C., Feng Y., Xu M.. Comparison of chemical compositions of the pepper EOs from different cultivars and their AChE inhibitory activity. Nat. Prod. Commun. 2020;15(11):1–6. doi: 10.1177/1934578X20971469. [DOI] [Google Scholar]
  30. Wang Y. F., Zheng Y., Cha Y. Y., Feng Y., Dai S. X., Zhao S., Chen H., Xu M.. Essential oil of lemon myrtle (Backhousia citriodora) induces S-phase cell cycle arrest and apoptosis in HepG2 cells. J. Ethnopharmacol. 2023;312:116493. doi: 10.1016/j.jep.2023.116493. [DOI] [PubMed] [Google Scholar]
  31. Wang Y. F., Zheng Y., Feng Y., Chen H., Dai S. X., Wang Y., Xu M.. Comparative Analysis of Active Ingredients and Potential Bioactivities of Essential Oils from Artemisia argyi and A. verlotorum . Molecules. 2023;28(9):3927. doi: 10.3390/molecules28093927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liang J., Zheng Y., Tong X., Yang N., Dai S.. In silico identification of anti-SARS-CoV-2 medicinal plants using cheminformatics and machine learning. Molecules. 2023;28(1):208. doi: 10.3390/molecules28010208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wang M., Tang H. P., Wang S., Hu W. J., Li J. Y., Yu A. Q., Bai Q. X., Yang B. Y., Kuang H. X.. Acorus tatarinowii Schott: a review of its botany, traditional uses, phytochemistry, and pharmacology. Molecules. 2023;28(11):4525. doi: 10.3390/molecules28114525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Wang N., Wang H., Pan Q., Kang J., Liang Z., Zhang R., Moreno S.. The combination of β-asarone and icariin inhibits Amyloid-β and reverses cognitive deficits by promoting mitophagy in models of Alzheimer’s disease. Oxid Med. Cell Longev. 2021;2021:7158444. doi: 10.1155/2021/7158444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Zhao X., Liang L., Xu R., Cheng P., Jia P., Bai Y., Zhang Y., Zhao X., Zheng X., Xiao C.. Revealing the antiepileptic effect of α-asaronol on pentylenetetrazole-induced seizure rats using NMR-based metabolomics. ACS Omega. 2022;7(7):6322–6334. doi: 10.1021/acsomega.1c06922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Wang Z., Qiu Y., Hou C., Wang D., Sun F., Li X., Wang F., Yi H., Mu H., Duan J.. Synthesis of hyaluronan-amikacin conjugate and its bactericidal activity against intracellular bacteria in vitro and in vivo . Carbohydr. Polym. 2018;181:132–140. doi: 10.1016/j.carbpol.2017.10.061. [DOI] [PubMed] [Google Scholar]
  37. Xu Z., Zhou X., Hong X., Wang S., Wei J., Huang J., Ji L., Yang Y., Efferth T., Hong C., Li C.. Essential oil of Acorus tatarinowii Schott inhibits neuroinflammation by suppressing NLRP3 inflammasome activation in 3 × Tg-AD transgenic mice. Phytomedicine. 2023;112:154695. doi: 10.1016/j.phymed.2023.154695. [DOI] [PubMed] [Google Scholar]
  38. Yi M., Wang D., Chen Y., Xu X., Dai X.. β-Asarone suppresses TNF-α expression through DNA methylation and c-Jun-mediated transcription modulation in scratch-injured neuronal cells. J. Biochem Mol. Toxicol. 2021;35(7):e22798. doi: 10.1002/jbt.22798. [DOI] [PubMed] [Google Scholar]
  39. Lim H. W., Kumar H., Kim B. W., More S. V., Kim I. W., Park J. I., Park S. Y., Kim S. K., Choi D. K.. β-Asarone (cis-2,4,5-trimethoxy-1-allyl phenyl), attenuates pro-inflammatory mediators by inhibiting NF-κB signaling and the JNK pathway in LPS activated BV-2 microglia cells. Food Chem. Toxicol. 2014;72:265–272. doi: 10.1016/j.fct.2014.07.018. [DOI] [PubMed] [Google Scholar]
  40. Zuo D., Chen Y., Cai J. p., Yuan H. Y., Wu J. Q., Yin Y., Xie J. W., Lin J. M., Luo J., Feng Y., Ge L. J., Zhou J., Quinn R. J., Zhao S. J., Tong X., Jin D. Y., Yuan S., Dai S. X., Xu M.. A hnRNPA2B1 agonist effectively inhibits HBV and SARS-CoV-2 omicron in vivo . Protein Cell. 2022;14(1):37–50. doi: 10.1093/procel/pwac027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sulkowski M. S., Agarwal K., Ma X., Nguyen T. T., Schiff E. R., Hann H. L., Dieterich D. T., Nahass R. G., Park J. S., Chan S., Han S. B., Gane E. J., Bennett M., Alves K., Evanchik M., Yan R., Huang Q., Lopatin U., Colonno R., Ma J., Knox S. J., Stamm L. M., Bonacini M., Jacobson I. M., Ayoub W. S., Weilert F., Ravendhran N., Ramji A., Kwo P. Y., Elkhashab M., Hassanein T., Bae H. S., Lalezari J. P., Fung S. K., Yuen M. F.. Safety and efficacy of vebicorvir administered with entecavir in treatment-naïve patients with chronic hepatitis B virus infection. J. Hepatol. 2022;77(5):1265–1275. doi: 10.1016/j.jhep.2022.05.027. [DOI] [PubMed] [Google Scholar]
  42. Wang L., Zhu Q., Zhang J. D., Zhang Y., Ni X., Xiang K., Jiang J., Li B., Yu Y., Hu H., Zhang M., Wu W., Zeng J., Yan Z., Dai J., Sun K., Zhang X., Chen D., Feng S., Sach-Peltason L., Young J. A. T., Gao L.. Discovery of a first-in-class orally available HBV cccDNA inhibitor. J. Hepatol. 2023;78(4):742–753. doi: 10.1016/j.jhep.2022.12.014. [DOI] [PubMed] [Google Scholar]
  43. Li X., Duan S., Chu C., Xu J., Zeng G., Lam A. K., Zhou J., Yin Y., Fang D., Reynolds M. J., Gu H., Jiang L.. Melaleuca alternifolia concentrate inhibits in vitro entry of influenza virus into host cells. Molecules. 2013;18(8):9550–9566. doi: 10.3390/molecules18089550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Kim C. J., Kwak T. Y., Bae M. H., Shin H. K., Choi B. T.. Therapeutic potential of active components from Acorus gramineus and Acorus tatarinowii in neurological disorders and their application in Korean medicine. J. Pharmacopuncture. 2022;25(4):326–343. doi: 10.3831/KPI.2022.25.4.326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Zhu H., Ali I., Hussain H., Hussain M., Wang X. B., Song X., Luo G., Zhang Z., Wang Z., Wang D.. Extraction and purification of cis/trans asarone from Acorus tatarinowii Schott: Accelerated solvent extraction and silver ion coordination high-speed counter-current chromatography. J. Chromatogr A. 2021;1643:462080. doi: 10.1016/j.chroma.2021.462080. [DOI] [PubMed] [Google Scholar]
  46. Jan R. H., Lin Y. L., Chen C. J., Lin T. Y., Hsu Y. C., Chen L. K., Chiang B. L.. Hepatitis B virus surface antigen can activate human monocyte-derived dendritic cells by nuclear factor kappa B and p38 mitogen-activated protein kinase mediated signaling. Microbiol. Immunol. 2012;56(10):719–727. doi: 10.1111/j.1348-0421.2012.00496.x. [DOI] [PubMed] [Google Scholar]
  47. Ren J. H., Tao Y., Zhang Z. Z., Chen W. X., Cai X. F., Chen K., Ko B. C., Song C. L., Ran L. K., Li W. Y., Huang A. L., Chen J.. Sirtuin 1 regulates hepatitis B virus transcription and replication by targeting transcription factor AP-1. J. Virol. 2014;88(5):2442–2451. doi: 10.1128/JVI.02861-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Dai X. Q., Cai W. T., Wu X., Chen Y., Han F. M.. Protocatechuic acid inhibits hepatitis B virus replication by activating ERK1/2 pathway and down-regulating HNF4α and HNF1α in vitro . Life Sci. 2017;180:68–74. doi: 10.1016/j.lfs.2017.05.015. [DOI] [PubMed] [Google Scholar]
  49. Wang S., Chen Z., Hu C., Qian F., Cheng Y., Wu M., Shi B., Chen J., Hu Y., Yuan Z.. Hepatitis B virus surface antigen selectively inhibits TLR2 ligand-induced IL-12 production in monocytes/macrophages by interfering with JNK activation. J. Immunol. 2013;190(10):5142–51. doi: 10.4049/jimmunol.1201625. [DOI] [PubMed] [Google Scholar]
  50. Balakrishnan R., Cho D. Y., Kim I. S., Seol S. H., Choi D. K.. Molecular mechanisms and therapeutic potential of α- and β-asarone in the treatment of neurological disorders. Antioxidants (Basel) 2022;11(2):281. doi: 10.3390/antiox11020281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Dai S., Chen X., Yu Y., Zang G., Tang Z.. Immunization with lentiviral vector-modified dendritic cells encoding ubiquitinated hepatitis B core antigen promotes Th1 differentiation and antiviral immunity by enhancing p38 MAPK and JNK activation in HBV transgenic mice. Mol. Med. Rep. 2018;18(5):4691–4699. doi: 10.3892/mmr.2018.9487. [DOI] [PubMed] [Google Scholar]
  52. Yang Q., Zhang Q., Zhang X., You L., Wang W., Liu W., Han Y., Ma C., Xu W., Chen J., Yang H., Wan P., Zhou Y., Liu Y., Wu K., Yang Z., Wu J., Ou J. H. J.. HoxA10 Facilitates SHP-1-catalyzed dephosphorylation of p38 MAPK/STAT3 to repress Hepatitis B virus replication by a feedback regulatory mechanism. J. Virol. 2019;93(7):e01607. doi: 10.1128/JVI.01607-18. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao5c07869_si_001.pdf (2.8MB, pdf)

Data Availability Statement

The data used and analyzed during the current study is provided in the manuscript and the Supporting Information. The anti-HBV activity data and the code of machine learning-based virtual screening framework are also openly accessible via the Essential Oils Database webservice (http://essentialoils.miningbiology.com/DownLoad.php).

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES