Malawi University of Science and Technology, Thyolo, Malawi
Fungal infections pose a significant global health burden, with drug resistance severely limiting treatment options. While Artemisia annua L., has been widely studied for its antifungal properties in various regions, there is however limited research on the antifungal potential of A. annua L., cultivated in Malawi. Given that environmental factors influence the plant’s phytochemical biosynthesis, composition and yield, its therapeutic activity may differ from those reported elsewhere. This study aimed to evaluate the antifungal potential of A. annua L. grown in Malawi through in-vitro assays and explore the molecular interactions and pharmacokinetic profiles of literature-reported bioactive compounds from A. annua using in-silico approaches. Antifungal efficacy of the plant extracts was evaluated through in-vitro susceptibility testing using the disk diffusion method. Qualitative phytochemical analysis was employed to identify the presence of key bioactive groups. In parallel, molecular docking studies were carried out in PyRx to examine interactions between literature-sourced compounds and selected fungal targets (CYP51 and AFSQS). Additionally, the pharmacokinetic properties, toxicity, and drug-likeness of five top-performing compounds were predicted using ADMET lab 2.0. Aqueous extracts of A. annua L. demonstrated stronger antifungal activity than ethanolic extracts against Candida spp. and Cryptococcus neoformans in disk diffusion assays. Phytochemical tests confirmed the presence of key bioactive classes. Literature-sourced flavonoids showed strong binding affinities to fungal targets CYP51 and AFSQS, indicating promising molecular interactions relevant to antifungal therapy. ADMET profiling further supported the suitability of these compounds, highlighting good drug-likeness, minimal predicted toxicity, and acceptable pharmacokinetic features, despite limited absorption potential. All five compounds were predicted to localize to mitochondria, suggesting a possible intracellular mode of antifungal action. Together, these findings support the therapeutic relevance of A. annua derived flavonoids as promising leads for antifungal drug development.
1.1 Background
Fungal infections represent a pressing yet neglected threat to public health [1]. They can cause severe disease and death and impose a substantial economic burden on healthcare systems [2]. These infections are responsible for more than 1.5 million deaths globally per year, primarily in those with compromised immune function [3]. They are also associated with increased illness rates and substantial healthcare costs, resulting in $6.7 billion in hospitalization costs in the United States in 2018 [3]. Data generated by the Global Action Funds for Fungal Infections (GAFFI), suggest an estimated 47.6 million Africans suffer from fungal diseases, of which 1.7 million suffer annually from a serious fungal infection [4]. On the other hand, Kalua et al., (2018) estimated that 1,338,523 people in Malawi (7.54%) are affected by serious fungal infections. The HIV epidemic in Malawi has highlighted Cryptococcus and Candida infections as important opportunistic fungal infections [5].
Cryptococcosis, caused by Cryptococcus neoformans, is an opportunistic fungal infection predominantly affecting immunocompromised individuals, especially those with HIV/AIDS [6]. The magnitude of Cryptococcosis among HIV patients varies from 1–10% in Western countries as opposed to almost a one third of HIV-infected individuals in sub-Saharan Africa where it is associated with high mortality[7]. The infection varies from asymptomatic colonization to severe disseminated disease, often targeting the CNS. It primarily enters via the respiratory tract but is rarely part of the normal human flora [8]. C. neoformans is frequently isolated in patients with HIV/AIDS, those undergoing immunosuppressive treatment, and individuals with specific cancers or lymphoproliferative conditions [9][10]. On the other hand, Candidemia which is caused by Candida species is a leading healthcare-associated bloodstream infection, ranking fourth in ICU prevalence based on patient profiles and risk factors[11,12] Neonates and immunocompromised individuals are most affected, leading to prolonged hospitalization and increased healthcare costs [13,14]. Mortality rates may exceed 70%, particularly with drug-resistant strains [13]. Despite its severity, candidemia remains an opportunistic infection requiring urgent medical attention [11].
Despite the growing threat of fungal infections, current antifungal drugs face limitations. The development of resistance to antifungals across diverse pathogenic fungal genera is an increasingly common and devastating phenomenon [1]. Currently available antifungals include only five classes, and their utility and efficacy in antifungal treatment are limited by one or more of innate or acquired resistance in some fungi, poor penetration into “sequestered” sites, and serious adverse effects/toxicities which require frequent patient reassessment and monitoring [15]. As a result of these challenges, treatment options for invasive fungal infections are limited, and patients at highest risk often have multiple comorbidities, including immunosuppression, which may limit the effectiveness of therapy even in the absence of drug resistance [16]. In addition, most fungal drugs are limited by drug–drug interactions that prevent their prolonged use or dosage escalation [16]. Despite serious fungal infections being reported in all major hospitals in Malawi, major broad-spectrum antifungals are still not included on the essential drug list in Malawi, and are only available as specially imported drugs for specific diseases [5].
Medicinal plants, such as Artemisia annua L., have been used for various treatment purposes in different regions of the world for thousands of years [17]. The plant A. annua L., has shown promise as a source of antimalarial and antimicrobial compounds [18]. The species has become a subject of particular interest due to the 2015 Nobel Prize awarded for detecting the sesquiterpene lactone artemisinin in it and proving its antimalarial activities [19]. Recent studies have shown antifungal activity of Artemisia annua L., against fungal pathogens [20,21] Artemisia annua L., contains many different classes of compounds and they include; specific sesquiterpene lactones [22], essential oil with mono and sesquiterpenes [23], flavonoids, coumarins, phenolic acids, tannins, saponins, polyalkenes, phytosterols, fatty acids, and proteins [23], including enzyme proteins[24]. Each class contains several bioactive compounds that contribute to the therapeutic effects of the plant [25].
While A. annua L. cultivated in various regions has demonstrated antifungal properties, its chemical profile and biological activity can vary depending on environmental factors such as soil type, climate, and altitude[26,27,28]. These geographical influences may lead to differences in the concentration, yield or presence of key phytochemicals, potentially affecting therapeutic efficacy [26]. Despite its cultivation and traditional use in Malawi, there is a lack of studies examining the antifungal potential of A. annua L. growing in Malawi. Furthermore, limited in-silico analyses have been conducted to explore how A. annua’s bioactive compounds interact with fungal target proteins, particularly those relevant to pathogenic strains common in the Malawian setting. It is therefore in line with these aforementioned gaps that this study employ in-vitro experiments to examine antifungal potential of A. annua L., growing in Malawi and in-silico experiments to identify potential antifungal agents within bioactive compounds from A. annua L., reported in literature.
MATERIALS AND METHODS
The objective of this research was to evaluate antifungal potential of A. annua L., cultivated in Malawi through in-vitro experiments, and explore molecular interactions of literature-reported bioactive compounds from A. annua with selected fungal protein targets using in-silico approaches. In this chapter, the comprehensive methodology adopted for this study is presented. The research design, research approach, study setting, experimental procedure, data and collection techniques, data analysis, ethical consideration and study limitations are explained.
Setting
The study was conducted across four distinct settings. The initial phase involved plant sample collection at ANAMED Malawi in Ntcheu, Kamuzu University of Health Sciences (KUHeS) Pharmacy Laboratory where plant extracts were prepared, Malawi Liverpool Wellcome Trust microbiology laboratory where fungal isolates were collected. The subsequent laboratory testing and experimentation phase, which included both in-vitro and in-silico methodologies, was conducted at Malawi University of Science and Technology (MUST) in Thyolo district, located in the southern region of Malawi. All laboratory procedures were carried out at the university’s microbiology and chemistry laboratories, ensuring appropriate environmental conditions for analysis.
Plant collection, identification and extract preparation
Artemisia annua L., plant leaves were collected at ANAMED Malawi herbal gardens. The plant materials were identified, authenticated, and voucher specimens were deposited in the national herbarium under accession number 89608. Fresh leaves of A. annua L. were thoroughly washed with sterile distilled water to remove surface contaminants, air dried for one week and ground into powder using sterile manual grinder. This was stored in air-tight glass containers protected from light and heat at room temperature for one week. [29]. Two solvents; water and ethanol were used as extraction solvents. 50g of A. annua powder were weighed using analytical beam balance and were macerated in 500ml of water and ethanol (in separate conical flasks). These were shaken periodically after every 2 hours. After 24 hours the slurry of Artemisia annua in ethanol and water were filtered with Whatman No.1 filter paper with 11?μm pore size and concentrated under reduced pressure in a rotavapor at 50°C to recover the ethanol and water [30,31]. Using equation 1, percentage yield for the plant extracts were calculated.
Percentage yield=weight of dry extractinitial weight of A3 leaves×100
………… (Equation 1)
Calculated yield percentage for the plant extracts are presented in the Table 1.
Table 1: Percentage yield of Aqueous and ethanolic plant extracts.
|
Solvent |
Weight of dry extract (g) |
Initial weight of plant material (g) |
Percentage Yield (%) |
|
Water (500ml) |
7.98 |
50 |
15.96 |
|
Ethanol (500ml) |
3.1 |
50 |
7.02 |
The aqueous and ethanolic dry extracts were stored in small, airtight bottles and refrigerated at approximately 4?°C until required for testing [30].
Collection and preparation of Fungal Isolates
Four petri dishes of Candida species isolates and four for Cryptococcus neoformans were collected from the MLW Laboratory. Change of custody forms were signed as can be seen in Appendix 2 and 3. All the organisms were treated as potentially infectious and were handled with utmost care. These were transferred in a microbe transfer box and transferred using a private car to the medical microbiology lab at Malawi University of Science and Technology (MUST) in Thyolo where isolates were inoculated into new agar plates and incubated at the optimum temperature (30oC) for 24hrs.
Preparation and Standardization of Fungal Cultures for Antifungal Assays
For the preparation and standardization of fungal cultures for antifungal assays, fungal cells were carefully harvested from Candida species and Cryptococcus neoformans isolates in Petri dishes using a sterile loop. The collected cells were transferred into sterile tubes containing normal saline solution and vortexed briefly to create uniform suspensions [32]. The fungal suspensions were then standardized to a concentration of 1 × 10? CFU/ml [33] by measuring the cell density, and adjustments were made with sterile solution where required. These standardized suspensions were subsequently used directly in antifungal assays (disk diffusion method) for evaluating antifungal activities.
Preparation of Extract Concentrations
Aqueous extracts were dissolved in sterile distilled water [34] while ethanolic plant extracts were dissolved in Dimethylsulfoxide (DMSO) prior to diluting them into stock concentrations [30]. Sterile distilled water was used as the negative control for the experiment but DMSO control was not included as the final DMSO concentration used for dissolution was below 1% a level generally regarded as non-toxic and non-antifungal in biological assays[31]. The formula; C1V1=C2V2 was used to prepare four concentrations of each plant extracts. The concentrations were; 100mg/ml, 50mg/ml , 25mg/ml and 10mg/ml . The same concentration formula was used to prepare 0. 03mg/ml of Ketaconazole drug (control)[32]. Sterilization of materials was done by autoclaving.
Preparation of diffusion disks
The diffusion disks were made by cutting out circles, with a diameter of 6mm, from filter papers using a paper borer. The disks were put in a beaker closed with aluminium foil and autoclaved at 121?°C for 15 minutes to ensure sterility. The disks were dipped in different concentrations of aqueous, ethanolic extracts and controls. The disks were left to air dry before use.
Antifungal activity
The antifungal activities of the aqueous and ethanolic plant extracts of Artemisia annua L. was determined against Candida species and Cryptococcus neoformans. Fungal strains were tested on Sabouraud dextrose agar (SDA)[32]. Sterilized paper disks were loaded with different amount of A. annua aqueous and ethanolic extracts (100, 50, 25, and 10mg/ml
) and applied on the surface of agar plates. Ketaconazole of 0.03mg/ml concentration and sterile distilled water were used as positive control and negative control respectively. Ketoconazole was chosen due to its widespread use as a positive control in antifungal studies and its well-established broad-spectrum activity against against both Candida species and Cryptococcus neoformans [35,32,36,37]. All the plates were incubated in an upright position at 27°C for 24 h. The experiment was done in triplicates. The inhibition zones were measured on the underside of the plates using a ruler after 48hrs [34]. The Minimum Inhibition Concentration (MIC) was defined as the lowest drug concentration, resulting in a clear zone of growth inhibition around the disk after conventional incubation period [38].
Phytochemical Screening of Artemisia annua L.
The extracts were subjected to qualitative phytochemical screening of bioactive compounds using standard chemical tests. The aqueous and ethanolic extracts of the leaves were screened for the presence of phenols, saponins, glycosides, terpenoids, alkaloids, flavonoids, steroids, and tannins. The qualitative results are expressed as (+) for positive and (−) for negative results. The screening process was conducted following established methodologies described by [39,40,41].
In-silico Experiments
Preparation of ligands
Bioactive components that are found in A. annua L. have been investigated in the literature [17]. 60 compounds known for their antimicrobial properties were sourced from PubChem (https://pubchem.ncbi.nlm.nih.gov/) [42], a comprehensive chemical database. The compounds were downloaded in the 3D SDF (Structure Data File) format and converted to PDBQT format with Open Babel [43].
Preparation of proteins
Crystal protein structures were obtained from the Protein Databank (http://www.rcsb.org/pdb). Lanosterol 14-alpha demethylase (CYP51) (PDB code: 7RYX, resolution: 2.10 Å) [44], fungal squalene synthase (SQS) (PDB code: 7wgi, resolution: 2.50 Å) [45] were selected. The CYP51 and AFSQS proteins were downloaded in PDB format. All water molecules, non-interacting ions and inhibitors were removed. Polar hydrogen atoms were added. Discovery Studio (Biovia Inc.) software was used for protein preparation [17].
Molecular docking
Molecular docking study was conducted using PrYx software. The grid boxes were carefully defined to encompass the active site of the target proteins. For CYP51 the center of the grid was set at coordinates X = 10.5864, Y = 11.9377, and Z = 17.8629, while the dimensions of the grid were specified as X = 124.2885, Y = 91.8645, and Z = 95.1212. For AFSQS the center of the grid was set at coordinates X = 23.93472, Y = 48.0920, and Z = 218.12692, while the dimensions of the grid were specified as X = 72.35446, Y = 59.12754, and Z = 68.45785. This ensured the grid was appropriately sized to accommodate potential binding interactions. A total of 60 compounds were analysed at once using PrYx software, with ketoconazole and Terbinafine serving as the control for CYP51 and AFSQS respectively. Binding affinities were analyzed, ranging from -4.9 to -10.4 Kcal/mol. Compounds with higher binding affinities were shortlisted for further analysis, given their potential to exhibit strong interactions with the target. Ketoconazole and terbinafine served as the control compound to validate the docking results and ensure reliability. Ketoconazole targets CYP51 and terbinafine targets AFSQS both crucial enzymes in fungal ergosterol biosynthesis. Two target proteins from fungal pathogens are shown in Figure 1. These were retrieved from the protein databank (PDB) (http://www.rcsb.org/pdb).
Figure 1: Protein structures obtained from the Protein Data bank (a.CYP51, b. AFSQS)
ADMET Profiling of top performing compounds
ADMET profiling of the top-performing phytochemicals was conducted using ADMETlab 2.1, an online predictive platform for assessing drug-likeness and pharmacokinetic properties. The analysis included evaluation of absorption, distribution, metabolism, excretion, and toxicity parameters to determine the compounds' suitability as potential antifungal agents. Canonical SMILES of each compound were obtained in Pubchem database and input into the software.
Data Analysis
Data obtained from In-vitro experiments were analysed using SPSS (IBM SPSS Statistics) in which ANOVA test was employed to examine the effects of two independent variables on the antifungal activity of Artemisia annua L. extracts. Post hoc (Turkey HSD) was conducted to compare the means of antifungal activities between different solvent-concentration combinations to determine significant differences. The molecular docking results were systematically analyzed using visualisation tools like Discovery studios (Biovia.inc) and Pymol (3.1.3) to assess the binding affinities and interaction profiles of the selected bioactive compounds with the target proteins (CYP51 and AFSQS). ADMET profiling results were analyzed based on key descriptors such as human intestinal absorption, blood–brain barrier permeability, cytochrome P450 interactions, and toxicity risks using ADMETlab 2.0 online software.
RESULTS
Results for In-vitro Experiment
Antifungal efficacy of aqueous and ethanolic extracts of Artemisia annua L., against Candida spp. and Cryptococcus neoformans was observed by measuring the zones of inhibiton. The zones of inhibition observed as shown in the Figure 2 indicate the varying degrees of antifungal activity exhibited by each extract.
Figure 2: Zones of inhibition for ethanolic plant extracts (A), aqeuous extracts (B) against C.neoformans and ethanolic plant extracts (C) and aqeuous extracts (D) against Candida species.
The antifungal activity of aqueous and ethanolic extracts of Artemisia annua L. against Candida spp. and Cryptococcus neoformans is highlighted in Table 2. Statistical analysis (ANOVA followed by post hoc Tukey's test) revealed that the aqueous extract exhibited significantly greater antifungal activity than the ethanolic extract across all tested concentrations (p?C. neoformans (19.5?±?0.3?mm), followed by Candida spp. (17.8?±?0.4?mm). Ethanolic extracts demonstrated moderate inhibition, with zones ranging from 12.3?±?0.5?mm to 15.6?±?0.2?mm, depending on concentration. Ketoconazole (0.03?mg/mL), used as the positive control, showed the highest overall activity (19.3?±?0.2?mm for Candida spp. and 18.5?±?0.4?mm for C. neoformans), which was significantly greater than both extracts (p?
Table 2: Descriptive statistics showing the mean zones of inhibition of Artemisia annua L. plant extracts against Candida spp and Cryptococcus neoformans.
|
Treatment |
Pathogen |
Concentration |
Mean IZ (mm) |
SD |
|
Ethanolic Extract |
Candida spp. |
100?mg/mL |
15.67 |
0.52 |
|
50?mg/mL |
13.67 |
0.52 |
||
|
25?mg/mL |
12.50 |
0.55 |
||
|
10?mg/mL |
11.67 |
0.52 |
||
|
C. neoformans |
100?mg/mL |
15.33 |
0.52 |
|
|
50?mg/mL |
13.67 |
0.52 |
||
|
25?mg/mL |
13.33 |
0.52 |
||
|
10?mg/mL |
12.33 |
0.52 |
||
|
Aqueous Extract |
Candida spp. |
100?mg/mL |
17.83 |
0.75 |
|
50?mg/mL |
15.83 |
0.75 |
||
|
25?mg/mL |
14.33 |
0.52 |
||
|
10?mg/mL |
13.33 |
0.52 |
||
|
C. neoformans |
100?mg/mL |
19.50 |
0.55 |
|
|
50?mg/mL |
16.67 |
0.52 |
||
|
25?mg/mL |
14.67 |
0.52 |
||
|
10?mg/mL |
12.67 |
0.52 |
||
|
Ketoconazole |
Candida spp. |
0.03?mg/mL |
19.33 |
0.52 |
|
C. neoformans |
0.03?mg/mL |
18.50 |
0.55 |
4.2 Phytochemical screening
For phytochemical screening tests were done to check for the presence of phenolics, saponins, glycosides, terpenoids, alkaloids, flavonoids and tannins in aqueous and ethanolic extracts of A.annua L., Table 3 presents the qualitative screening results for these phytochemical groups. The presence of bioactive compounds is indicated by a (+) sign, showing their detection in the respective extracts and (-) sign shows absence. These results reveals the presence of several bioactive compounds in both aqueous and ethanolic extracts, including phenolics, flavonoids, alkaloids, terpenoids, tannins, and glycosides all of which are known to exhibit antimicrobial properties. Notably, saponins were detected only in the aqueous extract, which may contribute to its comparatively stronger antifungal activity, as saponins are known to disrupt fungal cell membranes. These findings suggest that the antifungal efficacy observed in the bioassays is likely due to the synergistic action of multiple phytoconstituents. Moreover, the variation in compound profiles between the two solvents highlights the influence of solvent polarity on phytochemical solubility and extraction efficiency.
Table 3: Qualitative analysis of phytochemical groups in aqueous and ethanolic extracts of Artemisia annua L.
|
Phytochemical Group |
Aqueous Extract |
Ethanolic Extract |
|
Phenolics |
+ |
+ |
|
Saponins |
+ |
– |
|
Glycosides |
+ |
+ |
|
Terpenoids |
+ |
+ |
|
Alkaloids |
+ |
+ |
|
Flavonoids |
+ |
+ |
|
Tannins |
+ |
+ |
In-silico Experiments Results
Molecular Docking Results
Molecular docking results for selected compounds are outlined In Table 4. Eight compounds selected from the 60 compounds contains higher docking scores and offers good interaction with target proteins. Ketaconazole and terbinafine were used as controls for CYP51 and AFSQS proteins respectively. The selected compounds are represented in Table 4 showing their docking scores in Kcal/mol, key amino acids and key interactions.
Table 4: Selected compounds with docking scores, key amino acid residues and key interactions.
|
Compound |
Target Protein |
Docking Score (kcal/mol) |
Key Amino Acids |
Key Interactions |
|
Apigenin |
CYP51 |
–9.4 |
GLN316, ASP233, TYR229, ILE205 |
Hydrogen Bond, Pi-Anion, Pi-Pi Stacked |
|
AFSQS |
–8.5 |
PHE55, TYR74, LEU192, VAL188 |
Pi-Pi Stacked, Hydrogen Bond, Pi-Alkyl |
|
|
Diosmetin |
CYP51 |
–8.9 |
GLN316, ASP233, TYR229, ILE205 |
Hydrogen Bond, Pi-Anion, Pi-Pi Stacked |
|
AFSQS |
–9.0 |
ARG53, TYR74, PHE55, LEU220 |
Hydrogen Bond, Pi-Pi Stacked, Pi-Alkyl |
|
|
Chryseriol |
CYP51 |
–9.0 |
MET100, PHE241, ARG98, SER382 |
Pi-Pi Stacked, Hydrogen Bond, Alkyl |
|
AFSQS |
–8.5 |
ARG53, PHE55, TYR74, LEU220 |
Hydrogen Bond, Pi-Pi Stacked, Pi-Alkyl |
|
|
Genkwanin |
CYP51 |
–9.0 |
PHE241, SER508, LEU95 |
Pi-Pi Stacked, Hydrogen Bond, Alkyl |
|
AFSQS |
–8.6 |
TYR74, PHE55, LEU192, VAL188 |
Pi-Pi Stacked, Hydrogen Bond, Pi-Alkyl |
|
|
Acacetin |
CYP51 |
–9.0 |
ASP233, TYR229, MET313, ILE205 |
Pi-Anion, Pi-Pi Stacked, Hydrogen Bond |
|
AFSQS |
–8.5 |
TYR74, PHE55, LEU192, VAL188 |
Hydrogen Bond, Pi-Pi Stacked, Pi-Alkyl |
|
|
Kaempferol |
CYP51 |
–8.8 |
ASP233, ILE205, MET313, TYR229 |
Pi-Anion, Pi-Pi Stacked, Hydrogen Bond |
|
AFSQS |
–8.2 |
PHE297, CYS298, LEU220, VAL188 |
Pi-Pi Stacked, Hydrogen Bond, Pi-Alkyl |
|
|
Luteolin |
CYP51 |
–8.8 |
PHE241, LEU95, VAL242, TYR229 |
Pi-Pi Stacked, Hydrogen Bond, Alkyl |
|
AFSQS |
–8.3 |
TYR74, PHE55, LEU220, VAL188 |
Pi-Pi Stacked, Hydrogen Bond, Pi-Alkyl |
|
|
Tamarixetin |
CYP51 |
–8.8 |
ASP233, TYR229, MET313, ILE205 |
Pi-Anion, Pi-Pi Stacked, Hydrogen Bond |
|
AFSQS |
–8.0 |
PHE297, ASP184, LEU220, VAL188 |
Hydrogen Bond, Pi-Anion, Pi-Alkyl |
|
|
Terbinafine |
AFSQS |
–8.2 |
TYR74, PHE55, LEU192, VAL188 |
Pi-Pi Stacked, Hydrogen Bond, Pi-Alkyl |
|
Ketoconazole |
CYP51 |
–10.2 |
THR507, PHE241, MET509, TYR126 |
Halogen Bond, Pi-Pi Stacked, Alkyl |
ADMET Profiles of selected compounds from A.annua L.,
To evaluate the pharmacokinetic behavior and safety profile of the top-performing compounds ADMET evaluation of the five prioritized compounds diosmetin, apigenin, chryseriol, acacetin, and luteolin was done. Table 5 is showing a balanced profile of drug-likeness, pharmacokinetic potential, and safety parameters. All compounds satisfied Lipinski’s Rule, indicating favorable oral drug-like characteristics, and displayed high oral bioavailability despite poor intestinal absorption predictions. Acacetin exhibited the highest drug-likeness (QED?=?0.756) and all the compounds were predicted to localize to mitochondria, potentially influencing their intracellular activity. Most compounds demonstrated low predicted hepatotoxicity risks, which is important because it suggests a favorable safety profile and reduces the likelihood of liver-related toxicity during systemic use. Consistent trends across all five compounds included poor blood–brain barrier penetration, high plasma protein binding, moderate clearance rates, and short half-lives (≤?0.90 h), suggesting rapid systemic elimination. Toxicity alerts such as AMES, DILI, and skin sensitization were uniformly moderate, while hERG inhibition risk remained minimal.
Table 5 : ADMET Profiles of Selected Compounds from A. annua L.,
|
Property |
Diosmetin |
Apigenin |
Chryseriol |
Acacetin |
Luteolin |
|
Molecular Weight (MW) |
300.26 |
270.05 |
300.06 |
284.07 |
286.050 |
|
logP / logS |
3.19 / –3.91 |
3.31 / –3.84 |
3.202 / –3.71 |
3.65 / –3.70 |
3.40 / –3.85 |
|
Drug-likeness (QED) |
0.672 (Good) |
0.632 (Moderate) |
0.672 (Fair) |
0.756 (Excellent) |
0.601 (Moderate) |
|
Hepatotoxicity (H-HT) |
0.1 (Safe) |
0.1 (Safe) |
0.06 (Safe) |
0.1 (Safe) |
0.1 (Safe) |
|
Lipinski Rule |
Accepted |
Accepted |
Accepted |
Accepted |
Accepted |
|
PAINS / ALARM NMR |
No alert / 3 alerts |
No alert / 2 alerts |
1 alert / 3 alerts |
No alert / 2 alerts |
1 alert / 3 alerts |
|
HIA / F30% / Caco-2 |
Poor / High / –4.92 |
Poor / High / –4.85 |
Poor / High / –4.91 |
Poor / High / –4.83 |
Poor / High / –5.11 |
|
P-gp Substrate / Inhibitor |
Substrate / Non-inhibitor |
Substrate / Non-inhibitor |
Substrate / Non-inhibitor |
Substrate / Non-inhibitor |
Substrate / Non-inhibitor |
|
PPB / Fu / VD (L/kg) |
96.04% / 6.73% / 0.657 |
97.25% / 5.47% / 0.510 |
95.93% / 7.13 / 0.67 |
97.22% / 4.92% / 0.625 |
95.60% / 5.01% / 0.612 |
|
BBB Penetration |
Moderate |
Moderate |
Moderate |
Moderate |
Moderate |
|
CYP3A4 (Inhibitor / Substrate) |
Moderate / Non-substrate |
Moderate / Non-substrate |
Moderate / Non-substrate |
Moderate / Non-substrate |
Moderate / Non-substrate |
|
Clearance / Half-life |
7.07 / 0.855h |
7.02 / 0.86h |
7.075 / 0.87 h |
5.90 / 0.696 h |
6.75 / 0.798 h |
|
AMES / DILI / hERG |
Borderline / High / Low |
Borderline / High / None |
Borderline / High / Low |
Borderline / High / Low |
Borderline / High / Low |
|
Skin Sens. / Carcinogenicity |
High risk / Low risk |
High risk / Low risk |
High risk / Low risk |
High risk / Low risk |
High risk / Low risk |
|
Biodegradability / Fish Tox. |
Not readily / High risk |
Not readily / High risk |
Not readily / High risk |
Not readily / High risk |
Not readily / High risk |
|
Subcellular Localization |
Mitochondria |
Mitochondria |
Mitochondria |
Mitochondria |
Mitochondria |
DISCUSION
The rise of drug-resistant fungal pathogens is a growing health challenge that calls for new and effective treatments [46]. Current antifungal agents, such as azoles and polyenes, often face issues like toxicity, side effects, and the development of resistant strains [47]. These problems highlight the need for alternative solutions, including biologically active compounds found in medicinal plants[48]. These plant-derived phytochemicals offer potential antifungal and antibacterial activities and could play a key role in overcoming antimicrobial resistance [49]. The current study explored potential antifungal agents from Artemisia annua L. that would work against some biological targets in selected fungal pathogens using in-vitro and in-silico approaches.
5.2 Antifungal Activities of Artemisia annua L., Plant Extracts
Artemisia annua L. has demonstrated antifungal activity against Candida spp. and C. neoformans, indicating its potential as a source of promising antifungal agents. The descriptive statistics in Table 2 highlight the zone of inhibition for the ethanoic and aqueous extracts from Artemisia annua L., across varying concentrations (100mg/ml, 50mg/ml, 25mg/ml, and 10mg/ml) against Candida spp. and Cryptococcus neoformans, with ketoconazole as a control. Both aqueous and ethanolic extracts showed antifungal activity against Candida spp and cryptococcus neoformans. These results are supported by several studies that have shown that artemisia annua’s aqueous and ethanolic extracts have antifungal activity[50,51,47,52]. However, a study conducted in Pakistan by Ikram et al., (2015) reported no antifungal activity of Artemisia annua L. on Candida albicans [53]. This suggest that the antifungal activity of Artemisia annua L. may be influenced by some factors. For example, Nageeb et al., (2014) reported that Artemisia annua’s chemical composition is influenced by the habitat in which it grows consequently affecting its medicinal efficacy.
It has been reported that the choice of extraction solvent significantly impacts the extraction efficiency and the bioactivity of the isolated compounds[54]. This study reports that aqueous extracts consistently showed higher mean inhibition zones compared to the ethanoic extracts, particularly at higher concentrations. For example, the aqueous extract demonstrated higher antifungal activity, with mean inhibition zones of 17.83 mm and 19.5 mm for Candida spp. and C. neoformans, respectively, at 100 mg/ml. The multiple comparisons provide pairwise analyses of the treatments, offering detailed insights into their relative efficacy. The aqueous extract consistently showed significantly greater antifungal activity than the ethanoic extract (mean difference = 2.08, p < 0.05). The enhanced antifungal activity of the aqueous extract may be linked to its ability to extract polar compounds like saponins identified exclusively in this extract which are known for their strong bioactivity [55]. These findings align with Oniha et al., (2021) who observed that aqueous extracts often retain a higher concentration of water-soluble bioactive compounds, enhancing their antifungal potency.
Azole antifungal agents are essential in treating fungal infections. Ketoconazole, used as the standard, consistently showed the highest inhibition zones, averaging 19.33 mm and 18.5 mm for Candida spp. and C. neoformans, respectively and significantly outperformed both extracts. Aqueous extracts were less effective than ketoconazole (mean difference = -3.31, p < 0.05) and the mean difference between ketoconazole and the ethanoic extract was 5.40 (p < 0.05), emphasizing the superior efficacy of synthetic antifungal agents. Its effectiveness underscores its established role as a benchmark in antifungal studies [57,58]. However, the promising performance of the aqueous extract, particularly at higher concentrations, suggests its potential as a natural alternative, especially in cases where synthetic drug resistance is a concern[49,56] and in regions with limited access to synthetic drugs. These findings are consistent with studies indicating the potency of aqueous extracts for antifungal activity[47,57,59] These results also highlight the necessity of further exploring the phytochemical composition of the extracts to identify specific compounds responsible for their antifungal activity. Isolating and optimization of these compounds can enhance the efficacy of plant-derived treatments.
The observed standard deviations across the treatments indicate low variability, reflecting the reproducibility and reliability of the experimental procedures. This consistency aligns with the principles of reproducibility in scientific research, which emphasize the importance of obtaining consistent results under similar conditions [60]. Furthermore, the dose-dependent nature of antifungal efficacy demonstrated in this study highlights the potential of plant-derived extracts as effective therapeutic agents. These findings are consistent with broader research into the antimicrobial potential of plant extracts, which have shown promising results in combating fungal pathogens [61] [62]. The integration of such extracts into therapeutic applications offers a valuable avenue for addressing antimicrobial resistance [63].
Artemisia annua L., has been reported to contain a diversity of bioactive compounds [64,65]. In this study the qualitative phytochemical analysis of aqueous and ethanolic extracts of Artemisia annua reveals a rich diversity of bioactive compounds. Both extracts contained a broad spectrum of phytochemical groups, including phenolics, glycosides, terpenoids, alkaloids, flavonoids, and tannins, all of which are known for their significant therapeutic properties, such as antioxidant, anti-inflammatory, and antimicrobial activities[64,66,67]. Notably, saponins were detected only in the aqueous extract, reflecting the differential solubility of compounds depending on the solvent used. This may suggest that aqueous extraction is more effective for isolating saponins, which are widely recognized for their antifungal properties [68,69,70]. However, a study conducted in Pakistan by Ikram et al., (2015) reported the presence of saponins in ethanolic extracts. This can be supported by a study by Nageeb et al., (2014) who reported that Artemisia annua’s chemical composition is influenced by the habitat. The consistent presence of phenolics, glycosides, terpenoids, alkaloids, and flavonoids in both extracts highlights the pharmacological potential of Artemisia annua, particularly for its antifungal efficacy, which aligns with previous studies emphasizing their roles in microbial inhibition [71,72]. The presence of tannins further enhances the medicinal profile of these extracts, contributing to their observed bioactivity against fungal pathogens. Overall, these findings underline the importance of solvent selection in maximizing the recovery of bioactive compounds, particularly for applications in antifungal drug discovery and development.
The exploration of antifungal agents derived from Artemisia annua L., represents a growing area of interest in natural product-based drug discovery. This study focused on computational screening of 60 phytochemical compounds, ultimately selecting potential antifungal agents based on molecular docking results against two fungal target proteins: CYP51 and AFSQS. The use of established antifungal drugs like ketoconazole (CYP51) and terbinafine (AFSQS), as positive controls provided a comparative framework to evaluate the efficacy of these selected compounds.
The molecular docking results revealed that apigenin,diosmetin,chryseriol, genkwanin, acacetin, kempferol, luteolin, tamarixetin as potential antifungal agents with with diosmetin, apigenin, chryseriol, acacetin and luteolin exhibiting excellent binding affinity, with docking scores surpassing terbinafine, a control drug, suggesting their potential as effective antifungal agents. The success of these selected compounds can be attributed to their strong ligand-protein interactions, which are crucial for effective enzyme inhibition in fungal pathogens. Previous studies have demonstrated that flavonoid-based compounds exhibit potent antifungal properties, showing promising potential for antifungal therapy and prevention by significantly reducing fungal virulence and downregulating antifungal resistance-associated genes [73,74]. This activity is crucial as it not only weakens fungal pathogenicity but also mitigates drug resistance, thereby enhancing the efficacy of existing antifungal treatments and providing a foundation for the development of novel, plant-derived antifungal therapeutics.
Apigenin showed higher docking score against all the target proteins, with predominant hydrogen bond, Pi-Anion and Pi-Pi stacked interactions which facilitates stability within the proteins binding pockets [75]. Some studies have reported antifungal activity of apigenin and its mechanism of action. For example, it is reported that apigenin downregulates genes linked to drug resistance and ergosterol biosynthesis. It also significantly reduces biofilm formation and hyphal transition, both key virulence factors in Candida infections [73]. Futhermore, apigenin disrupts fungal cell membrane integrity, induces mitochondrial dysfunction, increases reactive oxygen species (ROS), and triggers apoptosis in Candida albicans through mitochondrial calcium signaling [76,77,78]. These results align with invitro findings that apigenin inhibits the growth of various Candida species, with minimum inhibitory concentrations (MIC) ranging from 0.10 to 0.15 mg/mL and minimum fungicidal concentrations (MFC) from 0.15 to 0.30 mg/mL [76]. These findings highlight apigenin as a promising antifungal candidate, capable of targeting multiple resistance pathways and mitigating fungal pathogenicity.
Similary, diosmetin showed high docking scores (-8.9Kcal/mole, -9.0Kcal/mol) across all the target proteins, with predominant Pi-Pi stacking, hydrogen bonding, and Pi-Anion interactions facilitating enhanced stability within the protein binding pockets. Zhao et al. (2023) provide further biological relevance for diosmetin’s antimicrobial effects, demonstrating that the flavonoid effectively suppresses microbial growth by attenuating inflammation and ferroptosis, a form of iron-dependent cell death associated with oxidative stress and metabolic disruption [79]. This suggests that diosmetin’s efficacy against fungal targets extends beyond direct enzyme inhibition, as it may also contribute to fungal cell damage by modulating oxidative stress pathways. The attenuation of inflammation-related signaling further supports its therapeutic role, as fungal infections often induce strong inflammatory responses that compromise host immune defenses.
Furthermore, chryseriol, and acacetin similarly demonstrated significant interactions with conserved amino acid residues like TYR 229, PHE 241, ASP 233, reinforcing their potential as CYP51 inhibitors. CYP51 (14-alpha demethylase) is essential for ergosterol synthesis in fungi, a key component of the fungal cell membrane. Inhibiting CYP51 disrupts membrane integrity and leads to fungal cell death [74]. Computational and molecular docking studies have identified specific flavonoid-like compounds with strong binding affinity to CYP51, outperforming standard antifungal drugs in docking scores [74,80]. Most tested flavonoids exhibit low cytotoxicity to human cells, supporting their potential for safe therapeutic use [73]. Antifungal activity of Chryseriol has also been reported and just like other flavones, chryseriol may act by modulating key metabolic enzymes and disrupting fungal cell processes [81]. Whilst acacetin’s diverse pharmacological profile includes anti-inflammatory, anticancer, and antimicrobial effects has been reported suggesting its potential as an antifungal agent, though more targeted studies are needed [81]. Acacetin on the other hand is noted for its broad antimicrobial activity, including inhibition of microbial growth, but direct evidence of its antifungal potency is limited. Its diverse pharmacological profile includes anti-inflammatory, anticancer, and antimicrobial effects, suggesting potential as an antifungal agent, though more targeted studies are needed[82].
Apart from CYP51 inhibition, selected compounds also exhibited strong interactions with AFSQS, a key target involved in fungal metabolism. AFSQS plays a crucial role in squalene epoxidation, a key step in fungal lipid metabolism, making it a validated drug target for antifungal therapy [45]. Several flavonoids in this study like Genkwanin, Tamarixetin, and Luteolin demonstrated Pi-Pi stacking and hydrogen bonding interactions, stabilizing their binding within the AFSQS pocket. This mode of binding has previously been noted in similar molecular docking studies, where flavonoid compounds exhibited high-affinity binding to fungal squalene epoxidase enzymes, disrupting lipid biosynthesis and leading to fungal cell death [83,84]. In a similar study molecular docking revealed strong binding affinities for apigenin driven by hydrophobic and electrostatic interactions with critical residues in the squalene epoxidase active site [85]. Although the antifungal activity of Genkwanin and Tamarixetin has not been extensively studied, their antioxidant and anti-inflammatory properties offer strong support for their potential role in fungal inhibition. By reducing oxidative stress and modulating inflammatory responses, these compounds may help weaken fungal survival mechanisms, enhance host defense, and disrupt key virulence factors, reinforcing their viability as antifungal agents. On the otherhand Luteolin inhibited growth of Candida albicans with the minimal inhibitory concentration of 37.5 µg/mL and showed ability to act as biofilm formation inhibitors [86].
A notable observation in this study was the consistent superiority of flavonoid-based compounds over steroidal phytochemicals such as Stigmasterol and Beta-Sitosterol. Although Stigmasterol with docking score; -10.3Kcal/mole and -10.4Kcal/mole for CYP51 and AFSQS repectively demonstrated strong binding affinity across CYP51 and AFSQS , its interactions were predominantly Pi-Alkyl and Van der Waals forces, which may contribute to lipid perturbation rather than direct enzyme inhibition [87]. In contrast, flavonoid compounds such as Diosmetin and Apigenin exhibited broader structural adaptability, enabling multi-target interactions across CYP51 and AFSQS. This versatility is a crucial advantage in drug development, as compounds that engage multiple fungal targets could mitigate drug resistance mechanisms often observed in pathogenic fungi [88]
The comparison with two control drugs ketoconazole and terbinafine, further strengthened the findings. Ketoconazole, a well-established azole-based CYP51 inhibitor, exhibited a docking score of -10.2, slightly outperforming Diosmetin but with fewer interactions with critical amino acid residues. Terbinafine, an AFSQS inhibitor, recorded a docking score of -8.2, placing it below several selected flavonoids, including apigenin (-8.5) and chryseriol (-8.5), which engaged key residues more effectively. This finding aligns with studies reporting flavonoid compounds as potential alternatives to terbinafine, given their broader bioavailability and minimal toxicity . For example, apigenin, showed a potential as an antifungal agent for treating dermatophytosis in mice, with complete recovery comparable to standard drug terbinafine in 12 days[84].
The ADMET profiling of five selected phytocompounds from Artemisia annua L. Diosmetin, apigenin, chryseriol, acacetin, and luteolin revealed a consistent pattern of drug likeness and moderate pharmacokinetic and safety properties that support their candidacy as antifungal leads. All five compounds complied with Lipinski’s Rule of Five, indicating favorable oral bioavailability profiles, at least from a molecular property standpoint [89]. Acacetin exhibited the highest drug likeness (QED?=?0.756), implying strong structural suitability for drug development, while Luteolin showed the lowest (QED?=?0.601), yet still within acceptable range for natural product scaffolds. LogP values ranged from 3.19 to 3.65 across the compounds, suggesting adequate lipophilicity for membrane permeability, while logS values between –3.70 and –3.91 indicated low to moderate aqueous solubility typical for polyphenolic molecules. Importantly, hepatotoxicity predictions flagged Chryseriol with a probability of 0.06, suggesting a potential safety, whereas the remaining compounds scored 0.1, indicating low predicted hepatotoxicity and hence a safer profile for systemic application [90]. This distinction could guide structure activity relationship (SAR) refinement to improve tolerability.
Although all compounds demonstrated high predicted oral bioavailability (F30%), their HIA scores were poor, and Caco-2 permeability values were below threshold, pointing to possible absorption limitations. However, they were uniformly non-inhibitory to P-glycoprotein, reducing concerns for efflux mediated resistance. Plasma protein binding (PPB) percentages ranged from 95.60% to 97.25%, suggesting high systemic retention, while volume of distribution (VD) and fraction unbound (Fu) values were within typical bounds for moderate tissue penetration. Blood–brain barrier (BBB) penetration predictions were moderate for all compounds, which may be acceptable or even advantageous depending on intended tissue targets. All five showed non-substrate behavior for CYP3A4, paired with weak to moderate inhibition indicating manageable risks for drug–drug interactions. Toxicity markers like hERG inhibition were absent or low across the board. Skin sensitization and fish toxicity were consistently high-risk, necessitating environmental and dermal safety evaluation in future formulations.
Interestingly, all compounds were predicted to localize to mitochondria, which may suggest intracellular antifungal mechanisms involving oxidative stress or mitochondrial disruption [91]. Given the natural origin and multi-target potential of flavonoids, their profiles are consistent with previously studied scaffolds in antifungal drug development [92]. The ADMET evaluation indicates that A. annua-derived compounds, particularly acacetin chryseriol and diosmetin, possess promising drug-like properties, modest toxicity risk, and acceptable pharmacokinetics for further optimization. These findings complement the molecular docking and phytochemical screening results, reinforcing their potential as lead antifungal candidates.
CONCLUSION
This study evaluated the antifungal potential of Artemisia annua L. cultivated in Malawi and explored the pharmacological relevance of literature-reported compounds through in-silico techniques. The in-vitro assays revealed that aqueous extracts exerted stronger inhibitory effects against Candida spp. and Cryptococcus neoformans than ethanolic extracts, supporting the presence of potent, water-soluble bioactive compounds within the locally cultivated plant. Phytochemical screening confirmed the presence of multiple secondary metabolite classes which are known to exhibit broad-spectrum antifungal activity. These findings highlight not only the therapeutic value of A. annua in the context of fungal infections but also the importance of evaluating medicinal plants within their local agroecological environments, given how environmental factors influence phytochemical composition. This aspect is particularly relevant for resource-limited settings, where indigenous medicinal plants could serve as accessible and cost-effective alternatives to synthetic antifungal agents.
Complementing the in-vitro findings, the in-silico analysis of literature-reported compounds provided deeper insight into the pharmacological potential of specific flavonoids. Molecular docking revealed strong interactions between key fungal targets (CYP51 and AFSQS) and compounds such as diosmetin, apigenin, acacetin, and chryseriol, with binding affinities comparable to or exceeding those of reference drugs. The ADMET profiling of these compounds further validated their drug-likeness, low toxicity profiles, and favorable pharmacokinetic characteristics despite absorption limitations reinforcing their candidacy for further development. Notably, the predicted mitochondrial localization of all five compounds suggests a possible intracellular antifungal mechanism that warrants further exploration. Taken together, this work underscores the untapped therapeutic potential of A. annua L. cultivated in Malawi and positions these literature-derived flavonoids as promising leads for antifungal drug development. The integrative research strategy adopted here provides a model for future studies aiming to bridge local biodiversity with rational drug discovery.
RECOMMENDATION
To advance these findings, it is recommended that future studies focus on isolating and characterizing specific bioactive compounds from A. annua L. grown in Malawi, followed by targeted antifungal testing. Experimental validation of molecular docking predictions using enzyme inhibition assays against CYP51 and AFSQS is essential to confirm the functional activity of literature-identified compounds. In addition, in-vivo studies should be pursued to assess bioavailability, toxicity, and pharmacokinetic behavior. Structural optimization through molecular dynamics simulations and QSAR modeling could enhance potency and drug-likeness. Exploring synergistic effects with existing antifungals may offer strategies to combat resistance and reduce toxicity. Finally, mechanistic studies investigating gene regulation and virulence suppression by selected flavonoids would help elucidate their therapeutic potential at the molecular level. Such integrative efforts will be instrumental in transforming phytochemicals from A. annua into viable antifungal therapeutics.
CONFLICT OF INTEREST: The authors declare that they have no competing interest.
FUNDING: Funding for this research was provided by the NORHED II Grant scholarship.
AUTHOR DECLARATIONS
Ethics approval
This study did not involve experiments on human or animal subjects, and therefore did not require written consent for its execution. However, the research protocol was approved by the Malawi University of Science and Technology Ethics Committee (MUSTREC). Additionally, a formal request for the collection fungal pathogens (Candida spp., and Cryptococcus neoformans) isolates was sent to Malawi Liverpool Wellcome Trust Laboratory manager who facilitated the collection process.
Consent to participate
Not applicable. This study did not involve human participants, human data, or animal subjects.
Consent for publication
Not applicable. The manuscript does not contain any individual person’s data in any form (including images or videos).
Availability of data and materials
All data generated or analyzed during this study are included in this published article. Additional datasets are available from the corresponding author upon reasonable request.
REFERENCES
Madalitso Muhakeya, Wilson Msaku, Willard Mbewe, Andrew Mtewa, Exploring Potential Antifungal Agents from Artemisia annua L., Using In-vitro and In-silico approaches., Int. J. of Pharm. Sci., 2025, Vol 3, Issue 10, 2913-2935. https://doi.org/10.5281/zenodo.17458582
10.5281/zenodo.17458582