Artemisia argyi extract subfraction exerts an antifungal effect against dermatophytes by disrupting mitochondrial morphology and function

Dermatophytosis is a common fungal skin disease caused by dermatophytes, which can be contracted on the head, hands, and feet ^[1]. This disease is widespread throughout the world and more prevalent in tropical and subtropical regions, and the incidence is quite high in China. Large numbers of people are negatively affected by the clinical manifestations, which are mainly erythaematous, blistering, pustular lesions with pronounced itching and varying degrees of symptoms depending on the site of onset ^[2]. Among the fungal species isolated from skin infections, the anthropophilic dermatophyte Trichophyton rubrum is the most common and accounts for 69.5% ^[3], and others include Trichophyton mentagrophytes and Microsporum gypseum ^[4]. The commonly used clinical topical drugs for the treatment of dermatophytosis are mainly imidazole and allylamine antifungals ^[5]. Despite their significant efficacy, they are accompanied by a high number of side effects and adverse reactions and present carcinogenicity, teratogenicity, and acute toxicity ^[6]. In addition, the abuse of broad-spectrum highly effective antibacterial drugs and glucocorticoids has led to increased resistance of dermatophytes, which are prone to recurrence and not easily curable ^[7]. Hence, investigating for efficient, low-toxicity alternatives with potent broad-spectrum antifungal activity are urgently needed to control dermatophytes.

China is rich in medicinal and edible plant resources, and increasing evidence suggests that numerous secondary metabolites in plants exhibit excellent antifungal activity, such as ellagic acid, licochalcone A, magnoflorine and loureirin A ^[8-11]. Meanwhile, plant resources have obvious advantages owing to their fewer side effects, diversified formulations, and less susceptibility to drug resistance ^[12]. Therefore, we are committed to excavating the functional plants with high antibacterial activity and clarifying their active components and anti-dermatophytic mechanism.

Artemisia argyi Levl.et Vant. belongs to Compositae and is widely distributed in Aisa, especially in China. A. argyi are not only traditional medicinal plants, but also abundant in nutrients and functional constituents as dietary additives. For example, ancient Chinese traditionally picked the buds and leaves of A. argyi around the Tomb-Sweeping Day and consumed as infusion, green dumpling and other forms of food supplements ^[13]. More importantly, A. argyi exhibits multiple biological activities and is widely used. The 2020 edition of the Chinese Pharmacopoeia indicated that the external use of A. argyi can remove dampness and relieve itching. However, A. argyi has been recorded as early as in the Compendium of Materia Medica as a good therapy for ringworm more than 400 years ago, and people with beriberi (foot infection with dermatophytes) can clip A. argyi in socks for treatment. Until today, A. argyi is still applied to prevent and cure skin diseases according to traditional Chinese folk usage, such as bathing and soaking feet with A. argyi. These records indicate that A. argyi water extract (AAWE) may have significant anti-dermatophytic activity, but the scientific evidence has been lacking. Modern pharmacological studies show that A. argyi is mainly abundant in volatile oils, flavonoids, phenolic acids, terpenes and other chemical components and presents significant antibacterial and antiviral effects ^[14]. At present, almost all studies have focused on the antifungal activity of A. argyi volatile oil ^[15,16]. However, few reports have reveled the antifungal activity of A. argyi water extract while the essential oil is very low and almost completely volatilized due to the high temperature during processing and preparation. And there has been a lack of effective drugs against the major clinical pathogenic fungi, including T. rubrum, T. mentagrophytes, and M. gypseum. In addition, the safety of A. argyi water extract has been proved by previous reports ^[17]. Therefore, it is urgent to confirm the inhibitory activity of AAWE against dermatophytes for promoting its scientific application.

In this work, we first evaluated the antifungal activity against dermatophytes of AAWE and its subfractions (AAWE1-AAWE6) using three main dermatophytes (T. rubrum, T. mentagrophytes, and M. gypseum). Then, ultra-performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-QTOF-MS) and high-performance liquid chromatography with a diode array detector (HPLC-DAD) assays were performed to characterize the chemical composition of the outstanding active subfraction (AAWE6) from AAWE. Furthermore, the anti-dermatophytic mechanism of AAWE6 was elucidated by physiological and biochemical detection, transcriptome analysis and key genes validation. Finally, the results were fully confirmed by molecular docking technology, enzymatic key targets verification and metabolite content measurements. Due to the innovative and systematic nature of our research, we obtained a patent licence (ZL202111161301.9) in 2022. In brief, our findings provide new insights into the anti-dermatophytic effects of A. argyi and the potential development of A. argyi as a therapeutic agent against fatal pathogens of human infections.

A. argyi was provided by Qichun County, Hubei Province, China. The specimens of these materials (S200711) were deposited in the Traditional Chinese Medicine Resource Center of Hubei University of Chinese Medicine and authenticated by Professor Dahui Liu (one of the authors of this study).

T. rubrum (No. 340195), T. mentagrophytes (No. 340405), and M. gypseum (No. 340196) were purchased from BeNa Culture Collection (BNCC, China). Terbinafine was supplied by Shanghaiyuanye Bio-Technology Co., Ltd. (Shanghai, China). Column chromatography was performed on a D101 macroporous resin (Donghong Chemical Co., China). Eupatilin, jaceosidin and casticin were obtained from Chengdu Herbpurity Co., Ltd. (Chengdu, China). MS-grade acetonitrile, methanol, and formic acid were purchased from Merck (Darmstadt, Germany). Other chemicals or solvents were of analytical grade and used without further purification.

Plant materials (6.0 kg), A. argyi water crude extract (AAWE, 714.2 g) and its six subfractions, aqueous elution portion (AAWE1, 262.9 g), 10% ethanol elution portion (AAWE2, 29.4 g), 20% ethanol elution portion (AAWE3, 45.6 g), 40% ethanol elution portion (AAWE4, 65.2 g), 60% ethanol elution portion (AAWE5, 13.3 g) and 95% ethanol elution portion (AAWE6, 3.9 g), were reported in our previous studies ^[18]. Briefly, the air-dried A. argyi (6.0 kg) was ultrasonically extracted with pure water at room temperature 2 times and concentrated by a rotary evaporator (50 °C) to afford a crude extract (714.2 g). Then the crude extracts (450.0 g) were added to a D101 macroporous resin chromatography and eluted with EtOH/H₂O (0:100, 10:90, 20:80, 40:60, 60:40, 95:5 v/v) to yield six subfractions (AAWE1-AAWE6).

T. rubrum, T. mentagrophytes and M. gypseum were cultured on potato dextrose agar (PDA) medium for 7-10 days, and hyphae were transferred into liquid Sabouraud medium for 7 days at 28 °C to prepare suspension solutions (1-5 × 10⁵ CFU·mL⁻¹).

The antifungal activity of AAWE and its six subfractions against dermatophytes was determined by the growth rate method according to a previous study ^[19]. AAWE was dissolved in pure water and then added to PDA to prepare drug-containing plates with concentrations of 0, 1.25, 2.5, 5 and 10 mg·mL⁻¹. Under aseptic conditions, three fungal cultures with similar growth rates were selected, and the 5 mm fungus cake was cut and placed in the centre of blank control medium (CK), drug-containing medium and positive drug medium (terbinafine, 5 μg·mL⁻¹). Each treatment was cultured in a biochemical incubator at 28 °C, and the hyphal growth diameter was measured when the CK group occupied the whole plate. Similarly, AAWE1-AAWE6 was dissolved in ethanol to prepare a drug-containing medium with a final concentration of 2.5 mg·mL⁻¹ for screening the active subfraction. The ethanol content was the same in the blank and drug-containing groups. In addition, the antifungal effects of the highest active subfraction (AAWE6) were evaluated with concentration gradients of 0, 39.06, 78.125, 156.25, 312.5, 625, and 1250 μg·mL⁻¹. Finally, the inhibition rate (IR) was calculated by the following formula: IR (%) = 100 × (Dc−Ds)/(Dc−D0), where Dc, Ds and D0 represent the diameters of the blank control group, extract-treated group and fungus cake, respectively.

The minimum inhibitory concentrations (MICs) of AAWE6 on three dermatophytes were assayed by the broth microdilution method based on the Clinical and Laboratory Standards Institute (CLSI) M38-A2 ^[20]. In a 96-well plate, 100 μL of the prepared spore suspension (1-5 × 10⁵ CFU·mL⁻¹) was added to each well, which contained 100 μL of AAWE6 (concentrations ranging from 0 to 1250 μg·mL⁻¹) in RPMI 1640 (Solarbio, China) medium, with terbinafine (5 μg·mL⁻¹) as a positive control. The negative control group was treated with RPMI 1640 medium. Subsequently, the 96-well plate was placed at 28 °C for 96 h. The MIC was defined as the lowest concentration capable of visually inhibiting 100% of fungal growth by two people. To obtain the minimal fungicidal concentration (MFC), 100 µL of the culture medium was taken from wells showing no visible growth and spread on PDA plates, which were then incubated at 28 °C for 7 d. The MFC was determined as the minimum drug concentration of no more than 5 colonies on the plate or no live fungal growth. All assays were carried out in three independent experiments performed in triplicate.

UPLC-QTOF-MS analysis was performed to identify the chemical components of the active subfraction AAWE6 using a Waters Acquity I-Class ultra-performance liquid chromatograph combined with a Xevo G2-XS quadrupole time-of-flight mass spectrometer (Waters, Milford, Massachusetts, USA). The chromatographic column, chromatography and mass spectrometry conditions used for analysis refer to our previous work ^[21].

Quantitative analysis of the main chemical components from AAWE6, including eupatilin, jaceosidin and casticin, was performed by HPLC-DAD (Shimadzu, Japan). The analytical column was an Agilent ZORBAX Eclipse XDB-C18 column (250 mm × 4.6 mm, 5 μm). An ordinary solvent system of A) 0.1% (v/v) formic acid in water and B) acetonitrile was used. The gradient elution program was 0-12 min, 10-20% B; 12-25 min, 20-25% B; 25-50 min, 25-65% B, 50-55 min, 65-10% B; and 55-60 min, 10% B. The detection wavelength, flow rate, column temperature and injection volume were 330 nm, 0.8 mL·min⁻¹, 30 °C and 10 μL, respectively.

The ultrastructural changes in T. rubrum and M. gypseum treated with AAWE6 were observed using transmission electron microscopy (TEM). The above spore suspension was added to liquid Sabouraud medium and incubated in a 28 °C constant temperature shaker for 7 d. Then, the shaken spore fluid was treated with AAWE6 for 24 h, centrifuged at 12000 rpm for 10 min at 4 °C, and washed three times with PBS buffer (pH 7.4) to collect the deposits. After the general procedure of dehydration, infiltration, embedding and cutting as described previously ^[22], fresh hyphal samples treated with AAWE6 (MIC, 2MIC) and untreated (control) samples were selected and prepared for observation by TEM (Talos L120C, Thermo Fisher, USA). In addition, the grayscale of the internal structure of mitochondria was analyzed and relatively quantified by ImageJ to show the damage degree of AAWE6 to mitochondria.

We weighed 0.1 g of fungal deposits treated as described above from each group, added 1 mL of extracting solution (PBS, pH 7.4), ground the deposits on ice, and centrifuged them at 12000 rpm for 10 min to obtain the supernatant for further analysis. The activity of squalene epoxidase, β-(1, 3)-D-glucan synthase and chitin synthase was determined according to a quantitative detection kit (Shanghai Fusheng Industrial Co., Ltd.). The enzyme activity in the samples was calculated by the concentration of the standard substance and the OD value at 450 nm.

Three replicates of both the control group and AAWE6-treated group used for the above analyses were also used for the transcriptome analysis. Total RNA extraction, quantification, qualification, cDNA library construction (three for the control group, and three for the AAWE6-treated group), and transcriptome sequencing were completed by Allwegene Tech Co., Ltd. (Beijing, China). The identification of the differentially expressed genes (DEGs) between groups was carried out according to a previous report ^[23]. Quantitative real-time PCR (RT-qPCR) was conducted using Real Universal Colour PreMix (SYBR Green) as the manufacturer’s instructions (Tiangen, China). The expression level of genes was normalized with β-tubulin ^[9], and the primer sequences used in this study are listed in Table S1.

Interactions between the chemical components in AAWE6 and key targets were studied by molecular docking. The 2D structure of the chemical composition in AAWE6 was downloaded using the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) and converted to PDB format with Openbabel. Concurrently, all crystal structures of the key targets with distinctive ligands and comparatively higher resolution were collected from the RCSB protein data bank (PDB, http://www.pdb.org/). After that, molecular docking between the compound and core target was analysed using AutoDock Vina software, and the binding energy was calculated. Finally, the docking results of a high degree of combination were displayed visually by PyMOL software.

The sample solution for measurement was prepared using the same method applied for the determination of squalene epoxidase activity as described above. Briefly, 0.1 g samples of the control group and AAWE6-treated group (MIC, 2MIC) were ground into homogenate after adding 1 mL of extracting solution. The activities of succinate dehydrogenase (SDH) and malate dehydrogenase (MDH), and the contents of pyruvic acid (PA) and citric acid (CA) in the supernatant were determined by an assay kit (Nanjing Jiancheng Bioengineering Institute, China). Similarly, the effects of different treatment time of AAWE6 (12 h, 24 h, 48 h and 72 h) on these indexes were measured.

A 0.1 g mycelium sample was made into a 10% homogenate using 9 times the volume of boiling water. The samples were boiled for 10 min in a boiling water bath, and the supernatant was collected after centrifugation to determine the ATP content. The samples of ATPase activity were extracted with 9 times the volume of PBS buffer (pH 7.4) and homogenized in an ice water bath. All the determination procedures were carried out in accordance with the kit instructions of Nanjing Jiancheng Bioengineering Institute (Nanjing, China).

Experimental data are presented as the mean ± standard deviation (SD). All figures were generated using GraphPad Prism 8. The difference between the control and the treatment groups was assessed by one-way analysis of variance (ANOVA), and a p-value less than 0.05 indicated a significant difference. All experiments were performed at least in triplicate.

A. argyi has been used as an herb and a food in Asia for thousands of years that exhibits antifungal activity, but its effects against dermatophytes has not been reported. First, cultures of T. rubrum, M. gypseum and T. mentagrophytes were checked for evaluating its anti-dermatophytic effect. Different concentrations of AAWE were applied on drug-containing medium to culture three dermatophytes. The results showed that AAWE inhibited the growth of the three dermatophytes in a concentration-dependent manner (Fig. 1A). When the concentration of AAWE reached 5 mg·mL⁻¹, the inhibition rate on the three dermatophyte mycelia was more than 60%, and more than 90% at a concentration of 10 mg·mL⁻¹ (Fig. 1B-D). These results indicated that AAWE has the potential to be developed as a clinical antifungal agent.

Figure 1.  Antifungal activity of A. argyi water extract (AAWE) against T. rubrum, T. mentagrophytes and M. gypseum, and screening of its active subfractions. (A) Antifungal effect of AAWE on three dermatophytes. (B-D) Inhibition rate (IR) of different concentrations of AAWE (1.25, 2.5, 5, 10 mg·mL⁻¹) on T. rubrum, T. mentagrophytes and M. gypseum, respectively. (E-G) The colony diameters of 6 subfractions isolated by D101 macroporous resin on T. rubrum, T. mentagrophytes and M. gypseum, respectively. * indicates p < 0.05 and ** indicates p < 0.01 relative to the control by ANOVA. The data are presented as the mean ± standard deviation (n = 3).

To determine the main antifungal active fraction of AAWE, the extract was further fractionated by macroporous resin D101 chromatography using EtOH/H₂O (0-95%, v/v) gradient elution to obtain subfractions AAWE1-AAWE6. The antifungal activity of six subfractions was evaluated using a colony diameter analysis at a concentration of 2.5 mg·mL⁻¹ (Fig. S1). As shown in Fig. 1E-G, the AAWE6 treatment had the smallest colony diameter (close to the original fungus cake diameter) for the three dermatophytes and was comparable to the effect of terbinafine treatment (P < 0.01). Thus, the most effective subfraction AAWE6 was screened out as the target fraction for further evaluation of the anti-dermatophytic effect and molecular mechanism.

To further validate that AAWE6 is an ideal subfraction, the anti-dermatophytic effect of AAWE was assessed using the above dermatophytes. Similarly, the inhibitory effect of AAWE6 on dermatophytes was enhanced with an increase in AAWE6 content (Fig. 2A). By contrast, AAWE6 showed a stronger inhibitory effect on T. rubrum and M. gypsum, and weaker effect on T. mentagrophytes. The inhibition rate of AAWE6 was over 70% on T. rubrum and M. gypsum at 625 μg·mL⁻¹ (Fig. 2B-D). More precisely, AAWE6 exhibited lower MIC and MFC values against T. rubrum (312.5 μg·mL⁻¹ for MIC and 2.5 mg·mL⁻¹ for MFC), M. gypseum (312.5 μg·mL⁻¹ for MIC and 2.5 mg·mL⁻¹ for MFC) and T. mentagrophytes (625 μg·mL⁻¹ for MIC and 5 mg·mL⁻¹ for MFC). Besides, compared with the MICs of plant extracts/compounds against dermatophytes reported in recent years, the advantage of AAWE6 is highlighted (Table 1). Therefore, the most effective subfraction AAWE6 might be a reliable source for mining natural antifungal compounds.

Figure 2.  Antifungal activity of the most effective active subfraction AAWE6 against T. rubrum, T. mentagrophytes and M. gypseum, and its chemical composition analysis using UPLC-QTOF-MS and HPLC-DAD. (A) Antifungal effect of AAWE6 on three dermatophytes. (B-D) IR of different concentrations of AAWE6 (39.06, 78.125, 156.25, 312.5, 625, 1250 μg·mL⁻¹) on T. rubrum, T. mentagrophytes and M. gypseum, respectively. (E) Base peak intensity (BPI) chromatogram in the negative ion mode of AAWE6 by UPLC-QTOF-MS. (F) Quantitative analysis of AAWE6 using HPLC-DAD, peak 4, 9, and 11 represent jaceosidin, eupatilin and casticin, respectively. The data are presented as the mean ± standard deviation (n = 3).

In the negative mode, the UPLC-QTOF-MS method was used to clarify the bioactive constitutions of subfraction AAWE6. The base peak intensity (BPI) chromatogram is shown in Fig. 2E. For eleven compounds, ion information was obtained, of which peaks 1-9 and 11 were flavonoids (Table 2) ^[27]. Flavonoids are mainly substituted by hydroxyl groups and oxymethyl groups with different numbers and positions, which mainly show the signal of [M-H]^- and a continuous CH₃· removal signal ^[28]. Collectively, these results showed that flavonoids were the main type of antifungal compounds in AAWE6.

Furthermore, the contents of three main flavonoids in AAWE6 were determined by HPLC-DAD at 330 nm: jaceosidin (peak 4), eupatilin (peak 9) and casticin (peak 11) (Fig. 2F). Their calibration curves were established as Y₁ = 52526X + 4218.6 (R² = 0.9993) for jaceosidin, Y₂ = 42143X + 31366 (R² = 0.9996) for eupatilin, along with Y₃ = 34480X + 20024 (R²=0.9993) for casticin. Obviously, eupatilin had the highest contents (131.16 ± 4.52 mg·g⁻¹) in AAWE6, followed by casticin (9.87 ± 0.35 mg·g⁻¹) and jaceosidin (4.17 ± 0.18 mg·g⁻¹) in succession. In addition, almost all of jaceosidin, eupatilin and casticin were distributed in AAWE6 but not in other subfractions (Fig. S2). This fully highlighted the importance of these flavonoids in antifungal activity.

The antifungal activity of AAWE6 has been confirmed, and its mechanism was subsequently explored to provide more scientific support for development and application. TEM images revealed clear morphological alterations and structural damage of T. rubrum cells induced by AAWE6 treatment, which were mainly manifested by disintegration of the cell wall and membrane, and disruption of the internal structure of mitochondria. Specifically, the control group was structurally intact and the internal mitochondria were clearly visible (Fig. 3A). In contrast, the AAWE6-treated group exhibited some cytoplasm shrunken away from the cytoderm, loose internal mitochondrial structure and invisible cristae (Fig. 3B and C). In addition, glycogen granules were significantly reduced and autophagosomes were increased after AAWE6 treatment. More importantly, the microstructure of M. gypsum showed similar damage under the action of AAWE6 (Fig. S3). These results demonstrated that AAWE6 could break through the protective barrier of the fungus and disrupt its vital sites of energy metabolism.

Figure 3.  Disruption of T. rubrum cell wall integrity, cell membrane function and ultrastructure treated with AAWE6 (MIC, 2MIC). (A-C) The ultramicroscopic structures under transmission electron microscopy (TEM) in the control group, MIC group, and 2MIC group, respectively. CW, cell wall; CM, cell membrane; MI, mitochondria; GG, glycogen granule; ATG, autophagosome. (D) The grayscale of the internal structure of mitochondria. (E-G) The inhibitory effects of AAWE6 on β-(1, 3)-D-glucan synthase, chitin synthase and squalene epoxidase activities, respectively. * indicates p < 0.05, ** indicates p < 0.01 and *** indicates p < 0.001 relative to the control by ANOVA. The data are presented as the mean ± standard deviation (n = 3).

According to numerous studies, β-(1, 3)-D-glucan synthase and chitin synthase are the key enzymes for fungal cell wall integrity, catalyzing the production of β-1,3-glucan and chitin ^[10,29]. Thus, their activities were utilized to evaluate the integrity of the cell wall in T. rubrum treated with AAWE6. The activity of both key enzymes was significantly reduced in the presence of AAWE6 at MIC and 2MIC concentrations (P < 0.05 or P < 0.01) (Fig. 3D and E). Moreover, the effect of AAWE6 on the cell membrane function of T. rubrum was demonstrated by measuring squalene epoxidase activity. The results in Fig. 3F show that the activity of squalene epoxidase decreased with increasing AAWE6 concentration. Taken together, AAWE6 exerted antifungal activity by disrupting the integrity of the cell wall and the function of the cell membrane of T. rubrum.

Because transcriptomics can comprehensively and objectively show gene changes after drug treatment, it is often used to explore the mechanism of drug action. To investigate the molecular mechanism of the most effective subfraction (AAWE6) against dermatophytes, transcriptome analysis was performed on T. rubrum samples from the control and AAWE6-treated (MIC) groups. With |log₂(fold change)| ≥1 and a Q-value (the corrected P value) less than 0.05 as the screening conditions, a total of 2018 DEGs were upregulated and 1918 DEGs were downregulated (Fig. 4A). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the above genes showed that the upregulated DEGs were mainly enriched in the peroxisome, ABC transporters and glutathione metabolism pathways (Fig. S4). This result indicated that the organism was in a possible oxidative stress state after exposure to AAWE6. Downregulated DEGs were significantly enriched in the citrate cycle (TCA cycle) pathway, as well as the pathways involved in fatty acid biosynthesis and metabolism, glycolysis/gluconeogenesis and amino sugar and nucleotide sugar metabolism (Fig. 4B).

Figure 4.  Transcriptome analysis of T. rubrum between the control group and AAWE6-treated group (MIC). (A) 2018 DEGs were upregulated and 1918 DEGs were downregulated in the volcano plot. (B) KEGG enrichment analysis of downregulated DEGs with TCA cycle as the most significant pathway. (C) Overall distribution of the 16 downregulated DEGs in the TCA cycle pathway.

As shown in Fig. 4C, there were 16 DEGs involved in the TCA cycle. All of these genes were downregulated and were distributed in almost all major processes in the pathway. It is worth pointing out that the synthesis of several important metabolites was significantly inhibited, including phosphoenolpyruvic acid, acetyl-CoA, α-ketoglutaric acid, and oxaloacetate. They are not only important parts of the TCA cycle, but also crucial bridges connecting carbohydrate, lipid and amino acid metabolism. Overall, the results indicated that AAWE6 might play an anti-dermatophytic effect by inhibiting the TCA cycle of the respiratory metabolism and then affecting the metabolism of nutrients in the whole body.

Through the above transcriptome results, an expression heatmap of 16 DEGs related to the TCA cycle pathway in each sequencing sample was generated, and it is shown in Fig. 5A. Eight DEGs were selected for RT-qPCR to validate the reliability of the transcriptome, including TERG_08639 (encoding phosphoenolpyruvate carboxykinase, PEPCK), TERG_06560 (encoding pyruvate dehydrogenase, PDH), TERG_04125 (encoding citrate synthase, CS), TERG_07814 (encoding isocitrate dehydrogenase, IDH), TERG_02628 (encoding 2-oxoglutarate dehydrogenase, OGDC), TERG_04056 (encoding succinyl-CoA synthetase, SCS), TERG_00102 (encoding fumarate hydratase, FH) and TERG_03762 (encoding malate dehydrogenase, MDH). The results were consistent with transcriptome (Fig. 5B-I), indicating that AAWE6 significantly inhibited the TCA cycle of T. rubrum.

Figure 5.  Comparative analysis of transcriptome results and quantitative real-time PCR (RT-qPCR) data. (A) The expression heatmap of 16 DEGs related to TCA cycle pathway in each sequencing sample. (B-I) Comparative analysis of the expression and transcriptome sequencing results of eight genes, and they encode phosphoenolpyruvate carboxykinase, (PEPCK), pyruvate dehydrogenase (PDH), citrate synthase (CS), isocitrate dehydrogenase (IDH), 2-oxoglutarate dehydrogenase (OGDC), succinyl-CoA synthetase (SCS), fumarate hydratase (FH) and malate dehydrogenase (MDH), respectively. * indicates p < 0.05, ** indicates p < 0.01 and *** indicates p < 0.001 relative to the control by ANOVA. The data are presented as the mean ± standard deviation (n = 3).

In the above study, we concluded that PDH, CS, IDH, SDH and MDH, which are involved in the TCA cycle, may be important targets of AAWE6 antifungal activity. To identify the key targets and potential antifungal components in AAWE6, the binding ability of 11 compounds in AAWE6 to 5 proteins was studied by molecular docking. By calculating the binding energy, we found that the binding energies of 11 compounds to MDH and SDH were lower, among which jaceosidin and eupatilin were particularly prominent (Fig. 6A). This demonstrated that MDH and SDH may be the key antifungal targets, and jaceosidin and eupatilin were the key antifungal components. Therefore, we further explored the preferred binding sites of jaceosidin and eupatilin to two key targets.

Figure 6.  Molecular docking to explore potential antifungal components in AAWE6. (A) Binding energy heatmap for molecular docking of 11 compounds from AAWE6 and 5 important targets in TCA cycle. (B and C) The preferred docking position and interactions of jaceosidin and eupatilin with MDH, respectively. (D and E) The preferred docking position and interactions of jaceosidin and eupatilin with SDH, respectively.

Since jaceosidin and eupatilin differ only one CH₃· group in structure, they were highly similar to the binding sites of the same target. As shown in Fig. 6B, the two methoxy groups and the hydroxyl groups at positions 4' and 7 of jaceosidin could generate hydrogen bonds with ARG80, GLY12, HIS187 and VAL13 in MDH. In addition to two methoxy groups, the hydroxyl group at positions 5 and 7 and the carbonyl group at position 4 of eupatilin formed six hydrogen bonds with MDH in ARG80, GLY12, HIS187 and ILE116 (Fig. 6C). In the process of interaction between two compounds and SDH, they all produced five hydrogen bonds through three oxygen atoms and one atom (Fig. 6D and E). The difference was that the hydrogen atom of jaceosidin was on 5-OH while that of eupatilin was on 7-OH. They shared four of the five binding sites with the SDH protein. By the above results, MDH and SDH may be the key targets for AAWE6 to exert antifungal effects, and jaceosidin and eupatilin have excellent direct binding ability with them.

SDH and MDH are two important oxidoreductases in the TCA cycle ^[30] that also represent potential key targets in this study. Therefore, we examined the effect of different concentrations of AAWE6 on SDH and MDH activity of T. rubrum. After AAWE6 treatment, SDH activity increased by 2.1 times at the MIC concentration and decreased by 45.6% at the 2MIC concentration (Fig. 7A). The MDH activity after MIC and 2MIC treatment was 92.7% and 10.2% of that of the untreated group, respectively (Fig. 7B). Moreover, AAWE6 inhibited the contents of PA and CA in T. rubrum, which are two key organic acids in aerobic respiration, in a concentration-dependent manner (Fig. 7C and D). In addition, MDH activity and PA content in M. gypsum were also significantly inhibited by AAWE6 treatment (Fig. S3). These results fully suggested that AAWE6 seriously disrupted the process of TCA cycle of dermatophytes. Still further, in order to find the key regulatory protein targets, we examined the effect of AAWE6 on SDH, MDH, PA and CA of T. rubrum in a temporal manner (Fig. S5). As a result, MDH activity existed the most significant time dependence.

Figure 7.  Determination of several enzyme activities and metabolite contents of T. rubrum closely related to the TCA cycle after AAWE6 treatment. (A)The activity of succinate dehydrogenase (SDH). (B) The activity of malate dehydrogenase (MDH). (C) The content of pyruvic acid (PA). (D) The content of citric acid (CA). (E) The content of ATP. (F) The activity of ATPase. * indicates p < 0.05, ** indicates p < 0.01 and *** indicates p < 0.001 relative to the control by ANOVA. The data are presented as the mean ± standard deviation (n = 3).

ATP synthesis is an equally important part of mitochondrial biochemical reactions, and its content directly affects the normal energy metabolism of fungi ^[31]. Fig. 7E and F show the effects of AAWE treatment on the ATP content and ATPase activity of T. rubrum, respectively. Compared with the control group, the ATP content of the AAWE6-treated groups decreased with increasing exposure concentration by more than 70%. ATPase activities in T. rubrum treated with the MIC and 2MIC of AAWE6 decreased by 32.8% and 39.1%, respectively. In short, AAWE6 hindered the energy generation and consumption of mitochondria of T. rubrum.

As a common medicinal and edible plant used to prevent and treat skin diseases, A. argyi has been reported to have antifungal activity owing to the essential oil, but the anti-dermatophytes effect of its water extract (classic application) has not been confirmed. In this study, we investigated the antifungal effect of A. argyi water extract (AAWE) on three dermatophytes in this work and confirmed that AAWE6 separated by D101 macroporous resin was the most efficient subfraction (Fig. 8). Considering the efficacy and clinical importance, T. rubrum was mainly used for further mechanistic studies. Subsequently, physiological, biochemical and transcriptome analysis revealed the subfraction AAWE6 from A. argyi exerted antifungal effects against T. rubrum by disrupting mitochondrial morphology and function. Moreover, qualitative and quantitative analysis of the chemical composition from AAWE6 and molecular docking with key targets elucidated the potential antifungal components. Briefly, this study presents the first report on the anti-dermatophytic effect of A. argyi extract, which ensures that we are granted patents in China (ZL202111161301.9). Importantly, it provides a scientific support for the development of clinical antifungal agents.

Figure 8.  Schematic diagram of screening the most effective subfraction (AAWE6) against dermatophytes and exploring its molecular mechanisms of A. argyi.

Study reported that the water extract of A. argyi can inhibit the growth of common microbial strains, such as Staphylococcus aureus, Escherichia coli and Bacillus subtilis ^[14]. However, there is no scientific report on the prevention and treatment of dermatophyte infections with A. argyi extract. In addition, folk medicine often uses A. argyi to soak feet or fumigate to disinfect and relieve itching, but also without scientific evidence. In this study, the MIC of the most effective subfraction AAWE6 (MIC 312.5 μg·mL⁻¹ for T. rubrum) was significantly lower than that of traditional Chinese medicine extracts and monomer components in our study that have been widely reported (Table 1), such as matrine (MIC 780 μg·mL⁻¹) from Sophora flavescens ^[4] and oxyresveratrol (500 μg·mL⁻¹) from Morus alba ^[26]. On the basis of its excellent efficacy (Fig. 1 and 2), our systematic research will promote its further application to a great extent. In addition, it also offers a new idea for A. argyi as a health-promoting food to prevent human pathogens.

Elucidating the molecular mechanism of drugs is an important guarantee for drug application. The fungal cell wall and cell membrane are important protective barriers for the organism against external osmotic shocks ^[32]. However, the reduction of β-(1, 3)-D-glucan synthase, chitin synthase and squalene epoxidase activities in the presence of AAWE6 indicated that the cell structure of T. rubrum was attacked, which was consistent with the damage to the cell wall and cell membrane structure observed by TEM (Fig. 3). Although AAWE6 exhibited obvious inhibitory effect, it was not prominent, suggesting that it differed from the mechanism of traditional antifungal drugs such as terbinafine and ketoconazole ^[33]. Importantly, the cell wall and cell membrane damaged by the extract of AAWE6 were enough to promote the drug to swarm in and exert its effect. The disintegration of the internal structure of mitochondria was the strongest evidence. Moreover, autophagy forms a bilayer membrane structure in the cytoplasm, wraps the damaged material, and transports it to lysosomes for degradation ^[34]. The increase in autophagosomes also reflected the damaged state of cells. In addition, the decrease in glycogen granules indicated that glucose metabolism in T. rubrum cells was inhibited. Therefore, we believe that AAWE6 might be based on the “small holes” cutted in the cell wall and cell membrane, then affecting the structure and function of the granary (mitochondria), and eventually leading to cell collapse (Fig. 8).

Transcriptome sequencing technology provides comprehensive and rapid access to sequence and expression information for almost all transcripts in cells or tissues with a given state ^[35]. Thus, this technology will help fully elucidate the antifungal mechanism of AAWE6 at the molecular level. The TCA cycle was the most significant pathway for the enrichment of differentially expressed genes, and all the enriched DEGs were downregulated (Fig. 4). The TCA cycle is a general metabolic pathway in aerobic organisms and mainly distributed in mitochondria ^[36]. The results were consistent with our guess that AAWE6 destroyed not only the mitochondrial structure of T. rubrum but also its function. Importantly, the TCA cycle is also the hub that connects the metabolism of carbohydrates, lipids, and amino acids ^[37]. The KEGG enriched pathways contained pyruvate metabolism, fatty acid metabolism and glycolysis/gluconeogenesis, which were less significantly enriched than the TCA cycle. Combined with the decrease in glycogen in TEM images, these results suggested that the antifungal mechanism of AAWE6 was centred on inhibiting the TCA cycle and indirectly affecting nutrient metabolism.

Furthermore, we conducted a series of experiments to validate the mitochondrial function of T. rubrum after treated with AAWE6. RT-qPCR confirmed the downregulation of genes encoding multiple enzymes in the TCA cycle at the genetic level, in agreement with transcriptomic data (Fig. 5). Moreover, molecular docking techniques were used to predict the binding ability of 11 compounds identified from AAWE6 to the dominant targets (PDH, CS, IDH, SDH and MDH) associated with the TCA cycle (Fig. 6). Surprisingly, SDH and MDH were the targets that bound more tightly to all compounds. MDH plays a key metabolic role in the pathway of aerobic energy production and can catalyse the oxidative dehydrogenation of malate to oxaloacetate ^[38]. In the present study, the inhibition of AAWE6 on MDH in T. rubrum was not only significantly concentration-dependent (Fig. 7B), but also excellent time-dependent (Fig. S5). Referring to the results of gene expression and molecular docking, MDH may be the key target for AAWE6 to exert its antifungal effect. In addition, the binding energy of eupatilin and jaceosidin (< −8.0 kcal/mol) to these targets was significantly lower than that of the other components, and they generated multiple hydrogen (≥ 5) bonds with the preferred binding sites of SDH and MDH. Numerous studies have shown that flavonoids, particularly eupatilin and jaceosidin, are quality markers of A. argyi ^[39,40]. Currently, reports on the activity of eupatilin and jaceosidin generally focus on their anti-inflammatory, antioxidant, and antitumor activities ^[41]. Therefore, our study will provide insights for their application in the treatment of dermatophyte infections.

PA is the final product of the glycolytic pathway, which enables the interconversion of sugars, fats and amino acids in vivo through the acetyl-CoA and TCA cycles ^[42]. While the condensation of acetyl-CoA with oxaloacetate to form CA is thought to be the starting step of the TCA cycle ^[43]. In this study, the contents of PA and CA also decreased as the increase of AAWE6 concentration and time (Fig. 7C and D). These results fully confirmed that AAWE6 significantly inhibited the TCA cycle pathway of T. rubrum. In addition to the TCA cycle, the ATP synthesis machinery is also an indispensable biochemical reaction for mitochondrial energy metabolism ^[44]. Previous studies have shown that mitochondrial cristae structure can directly affect the function of the mitochondrial respiratory chain complex and inhibit the synthesis of ATP ^[45]. Along with the damage to the mitochondrial crista structure caused by AAWE6, this subfraction also had an inhibitory effect on ATP content and ATPase activity based on the disruption of energy metabolism (Fig. 7E and F). As shown above, all evidence demonstrated that subfraction AAWE6 from A. argyi extract exerted antifungal effects against dermatophytes by disrupting mitochondrial morphology and function.

In summary, in this work, we demonstrate that AAWE6 (separated by D101 macroporous resin), a subfraction from the water extract of A. argyi, is effective at inhibiting common clinical dermatophytes. The chemical composition analysis and molecular docking predictions indicated that its antifungal effect was closely attributable to eupatilin and jaceosidin contained in AAWE6. The potential antifungal mechanism of AAWE6 is summarized in three stages: 1) cutting holes in the wall (breaking through the cell membrane and cell wall); 2) attacking the granary (destroying the structure and function of mitochondria); and 3) cell death. Briefly, our study can provide novel insights into the application of A. argyi and its subfraction AAWE6 as anti-dermatophytic medicine therapies.