Skip to Content
BiomoleculesBiomolecules
  • Article
  • Open Access

26 January 2026

14 Pages

New Pyridinone Alkaloid and Polyketide from the Cordyceps-Colonizing Fungus Pseudogymnoascus roseus

,
,
,
and
1
Jiangsu Key Laboratory for Biofunctional Molecules, College of Life Science and Chemistry, Jiangsu Second Normal University, Nanjing 210013, China
2
State Key Laboratory of Microbial Diversity and Innovative Utilization, Institute of Microbiology, Chinese Academy of Sciences, University of Chinese Academy of Sciences, Beijing 100101, China
3
Shandong Academy of Pharmaceutical Sciences, Jinan 250101, China
*
Author to whom correspondence should be addressed.

Abstract

One new pyridinone alkaloid pseudogymnone A (1) and one new tricyclic polyketide penijanthinone C (2), together with six known compounds, harzianic acid (3), 3-methyl-2-(2-nonenyl)-4(1H)-quinolinone (4), emodic acid (5), alaternin (6), violaceol-I (7), and violaceol-II (8), were obtained from the Cordyceps-colonizing fungus Pseudogymnoascus roseus. The structures and absolute configurations of the isolated compounds were elucidated through a combination of NMR and MS spectroscopic analyses, ECD calculations, and X-ray crystallography. Compound 3 exhibited obvious cytotoxicity against A549 (IC50 = 4.2 µM) and MGC (IC50 = 3.8 µM) cell lines. Integrated network pharmacology and molecular docking analyses indicated that compound 3 exerts potential anti-gastric-cancer effects by modulating multiple cancer-related signaling pathways, with EGFR identified as a potential target of compound 3.

1. Introduction

Natural fungal products are pharmacologically promising and structurally diverse, serving as crucial sources for drug discovery [1]. In 2023 and 2024 alone, more than 1400 new natural fungal products were reported, including polyketides, terpenoids, alkaloids, peptides, and steroids [2,3]. These metabolites display a broad spectrum of bioactivities including anti-inflammatory, antifungal, antibacterial, cytotoxic, and enzyme inhibitory effects, with some outperforming clinical drugs [2,3]. Fungi thriving in specialized and competitive environments tend to biosynthesize diverse structural secondary metabolites [4], which are not only crucial for their survival and growth but also exhibit various biological activities with potential pharmaceutical applications [5]. Recent studies on secondary metabolites produced by Cordyceps-colonizing fungi, inhabiting the fruiting bodies of Cordyceps sinensis, have revealed novel structural compounds with promising bioactivities. Examples include a novel gentianellane-type sesterterpenoid versicolorin A, which showed significant β-glucuronidase inhibitory activity and was isolated from Aspergillus versicolour ZJUTE2 [6]; the antifungal 4H-chromen-4-one derivatives coniochaetones E–I, obtained from Fimetariella sp. [7]; the cytotoxic p-terphenyls gliocladinin A and its β-d-glucoside gliocladinin B, isolated from Gliocladium sp. strain XZC04-CC-302 [8]; and the cytotixic cycloaspeptides F and G, derived from Isaria farinose [9]. Therefore, Cordyceps-colonizing fungi are prolific sources of bioactive natural fungal products with promising potential for future pharmacological development.
Our research aims to discover bioactive metabolites produced by fungi residing in underexplored ecological niches. We obtained the fungal strain Pseudogymnoascus roseus by isolating it from the sclerotia of Cordyceps sinensis (Berk.) Sacc., which were collected from the cold, alpine environment of Kangding, Sichuan Province, China. A subsequent chemical analysis of the crude extract of P. roseus produced one new pyridinone alkaloid pseudogymnone A (1) and one new tricyclic polyketide penijanthinone C (2) together with six known compounds: harzianic acid (3) [10], 3-methyl-2-(2-nonenyl)-4(1H)-quinolinone (4) [11], emodic acid (5) [12], alaternin (6) [13], violaceol-I (7) [14], and violaceol-II (8) [14] (Figure 1). Cytotoxicity assays revealed that compound 3 exhibited significant inhibitory effects against gastric cancer cell lines. A network pharmacology analysis was applied to predict the possible mechanism of compound 3, which was further validated by molecular docking. The comprehensive details of the isolation, structural characterization, biological evaluation, network pharmacology analysis, and molecular docking of these compounds are described herein.
Figure 1. Chemical structures of compounds 1–8.

2. Materials and Methods

2.1. Molecular Identification

To identify the isolate, we combined a morphological examination with a molecular analysis of the rDNA ITS region (GenBank Accession No. DQ117451). This fungus was deposited at the Institute of Microbiology, Chinese Academy of Sciences, Beijing, under the accession number S161.

2.2. General Experimental Procedure

The experimental approaches employed in this study were adapted from previously reported protocols [15]. Comprehensive descriptions of the methods are included in the Supplementary Method S1.

2.3. Fungal Materials, Cultivation, Fermentation, and Isolation

The fungus P. roseus was isolated from the sclerotia of C. sinensis (Berk.) Sacc. collected from Kangding, Sichuan, China. After 10 days of incubation on PDA at 25 °C, mycelial agar plugs (0.5 cm3 each) were used to inoculate a liquid seed medium (1% malt extract, 0.4% glucose, and 0.4% yeast extract, pH 6.5) in an Erlenmeyer flask. Shake cultivation (170 rpm, 5 d, 25 °C) provided the seed culture. For large-scale fermentation, autoclaved rice in Fernbach flasks was inoculated with a spore suspension (1 × 106 spores/mL) and incubated statically at 25 °C for 40 days.
The fermentation product was extracted multiple times with ethyl acetate (4 × 500 mL). The combined organic layers were concentrated under reduced pressure to afford a crude extract (4.2 g). This crude material was subjected to vacuum column chromatography over silica gel, eluted with a gradient of petroleum ether–ethyl acetate. The 30% ethyl acetate fraction (286 mg) was collected and further purified by Sephadex LH-20 gel column chromatography (Cytiva, Uppsala, Sweden; eluent: 1:1 CH2Cl2–MeOH). The combined relevant fractions were then refined by reversed phase HPLC (column: Agilent Zorbax SB-C18, 9.4 × 250 mm, 5 µm; elution program: 20% MeOH–H2O isocratic for 2 min, followed by a gradient from 20% to 90% MeOH–H2O over 50 min; flow rate: 2 mL/min, Agilent Technologies, Santa Clara, CA, USA) to yield compound 1 (4.7 mg, tR 48.9 min) and compound 2 (7.2 mg, tR 39.5 min).The 60% ethyl acetate fraction (123 mg) was purified by RP-HPLC (20% MeOH–H2O for 5 min, followed by a gradient from 20% to 80% MeOH–H2O over 30 min), affording compound 3 (6.9 mg, tR 27.6 min). Similarly, the 70% ethyl acetate fraction (261 mg) was separated by RP-HPLC (25% MeOH–H2O for 5 min, followed by a gradient from 25% to 80% MeOH–H2O over 30 min), yielding compounds 4 (6.2 mg, tR 23.2 min), 5 (7.4 mg, tR 17.6 min), 6 (4.1 mg tR 15.4 min), 7 (5.1 mg, tR 26.3 min), and 8 (3.7 mg, tR 25.4 min).
Pseudogymnone A (1): yellow oil; [ α ] D 25 = +252 (c 0.3, CH3OH); UV (MeOH) λmax (log ε) 207.1 (2.13), 241.0 (1.92), 296.0 (0.88) nm; IR (neat) νmax 3192, 2962, 2930, 1641, 1556, 1455, 1225 cm−1. 1H, 13C NMR and HMBC data see Table 1 and Supplementary Figures S1–S3; HR-ESIMS m/z 406.1989 [M + Na]+ (calcd for C23H29NO4Na, 406.1994).
Table 1. NMR data of pseudogymnone A (1) in CDCl3.
Penijanthinone C (2): yellow powder; [ α ] D 25 = +71 (c 0.1, CH3OH); UV (MeOH) λmax (log ε) 218.0 (2.46), 229.0 (2.58), 288 (1.76) nm; IR (neat) νmax 3192, 2962, 2930, 1641, 1556, 1455, 1225 cm−1. 1H, 13C NMR, and HMBC data see Table 2 and Supplementary Figures S7, S8 and S11; HR-ESIMS m/z 385.1650 [M − H]− (calcd for C22H25O6, 385.1651).
Table 2. NMR data of penijanthinone C (2) in CDCl3.

2.4. X-Ray Crystal Structure Analysis

A single crystal of compound 2 was obtained by employing the vapor-phase diffusion technique utilizing acetone−H2O (10:1), and data were collected using a Rigaku RAPID IP diffractometer (Rigaku, Tokyo, Japan) with graphite-monochromated Mo Kα radiation, λ = 0.71073 Å at 173(2) K. Crystal data: C22H26O6, 2(CH4O), M = 450.51, space group orthorhombic, P2(1)2(1)2(1); unit cell dimensions a = 7.2615(15) Å, b = 10.299(2) Å, c = 31.986(6) Å, V = 2392.2(8) Å3, Z = 4, D(calcd) = 1.251 mg/mm3, μ = 0.093 mm−1, F(000) = 968. Crystallographic data for compound 2 have been deposited with the Cambridge Crystallographic Data Centre (deposition number CCDC 2518640). The crystal structure was solved and refined using SHELXL-97 [16] with anisotropic non-H atoms and riding H atoms. After SADABS absorption correction, 5487 unique reflections produced a final R1 = 0.0456 and wR2 = 0.1085 [I > 2σ(I)].

2.5. ECD Calculation Methods

The absolute configurations of compounds 1 and 2 were assigned by comparing their experimental ECD spectra with the Boltzmann-weighted TD-DFT simulations. Full details regarding the computational procedures are provided in the Supplementary Method S2.

2.6. MTS Assay

Cell proliferation was evaluated with the MTS assay, following established protocols, and IC50 values were determined via nonlinear regression (SPSS Statistics 19.0) [17,18]. The human cancer cell lines used in this assay have been described in our previous work [19]. Further experimental details are available in the Supplementary Method S3.

2.7. Network Pharmacology

The SMILES of compound 3 was submitted to SwissTargetPrediction (http://www.swisstargetprediction.ch/ (accessed on 11 November 2025)), PharmMapper (http://lilab-ecust.cn/pharmmapper/ (accessed on 11 November 2025)), and TargetNet (http://targetnet.scbdd.com/ (accessed on 11 November 2025)) to predict its potential targets. Gastric cancer-related targets were collected from the GeneCards (https://www.genecards.org/ (accessed on 11 November 2025)) and OMIM (https://www.omim.org/ (accessed on 11 November 2025)) databases.
A Venn diagram of common targets between compound 3 and gastric cancer was constructed using Venny 2.1.0. Subsequently, the overlapping targets were processed through the STRING (https://string-db.org/ (accessed on 20 November 2025)) database to construct a protein–protein interaction (PPI) network. PPIs with a medium confidence (score > 0.4) were included. Cytoscape 3.9.1 was used to construct the key target network and to systematically analyze network parameters. To elucidate the potential mechanism of compound 3 against gastric cancer, we performed a functional enrichment analysis using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) (https://davidbioinformatics.nih.gov/ortholog.jsp (accessed on 2 December 2025)).

2.8. Molecular Docking

The crystal structure of the human EGFR ligand-binding domain (PDB ID: 9GFE) was retrieved from the Protein Data Bank. The 3D structure of harzianic acid (PubChem CID: 139585999) was obtained from PubChem and energy minimized under the MMFF94 force field. Molecular docking simulations were performed with AutoDock Vina 1.2.3 The receptor was prepared with PyMOL 2.5.5 by removing crystallographic water molecules, ions, and any co-bound ligands. A docking grid that enclosed the entire protein was defined using AutoDock Vina 1.2.3. Both the receptor and ligand were converted to the required PDBQT format with ADFRsuite 1.03. The exhaustiveness parameter for the global search was set to 32; all other parameters were kept at their default values. The top-ranked pose reported by Vina was taken as the binding mode and subsequently visualized with PyMOL 2.5.5.

2.9. Use of AI-Assisted Writing Tools

During the preparation of this manuscript, Kimi was employed as an auxiliary tool for grammar checking and sentence structure refinement to enhance the readability of the text. It is crucial to emphasize that all critical academic arguments, data interpretations, and conclusions originated solely from and are the full responsibility of the authors. The tools were strictly confined to optimizing the clarity of the language.

3. Results

3.1. Structure Elucidation

The molecular formula of pseudogymnone A (1) was assigned as C23H29NO4Na based on its HRESIMS (m/z 406.1989 [M + Na]+; Δ –0.5 mmu). The analysis of its 1H and 13C NMR data (Table 1) and HSQC spectrum (Figure S4) showed the presence of one carbonyl amide carbon (δC 157.4), twelve aromatic or olefinic carbons (including eight sp2 methines and four sp2 quaternary carbon), four sp3 methines (two oxygenated), three methylenes, three methyl groups, and one hydroxyl proton (δH 9.63). The NMR signals of the aromatic region of compound 1 matched well with those of the 1,4-dihydroxy-5-phenyl-2-pyridinone moiety, which is a key skeleton of the anticancer agent TMC-69 [14]. HMBC correlations from H-6 to C-2, C-4, C-5, and C-1′, and from H-2′ and H-6′ to C-5, together with 1H-1H COSY correlations (Figure 2 and Figure S5), as well as IR absorptions (3192, 1641, and 1556 cm−1), further confirmed the presence of 1,4-dihydroxy-5-phenyl-2-pyridinone moiety. 1H-1H COSY correlations (Figure 2) define the isolated spin system of the side chain from the C-7 to C-17 subunit. In addition, HMBC correlations from H-7 (δH 4.85) to C-11 (δC 79.8) and from H-11 (δH 4.00) to C-7 (δC 80.9) indicated that C-7 and C-11 were connected via the same oxygen atom to form the tetrahydropyran ring. Key HMBC correlations from H-7 to C-4, C-3, and C-2 identified the tetrahydropyran ring at the C-3 position of the pyridinone moiety, completing the planar structure of compound 1, as described in Figure 1.
Figure 2. Key 1H-1H COSY and HMBC correlations of compounds 1 and 2.
The relative configuration of pseudogymnone A (1) was established based on NOESY correlations and the 1H NMR coupling constants. The trans-diaxial orientation of H-8 and H-7 was determined by the large 3J7,8 value of 8.5 Hz. The C-12/C-13 olefinic bond was determined as E-geometry based on the large coupling constant (14 Hz) of H-12 and H-13. Additionally, NOESY correlations (Supplementary Figure S6) observed for H-11 with H-7 and H3-17 revealed that H-11, H-7, and H3-17 are on the same face of the tetrahydropyran ring. However, the configuration of C-14 remained unassigned owing to the lack of relevant NOESY cross-peaks.
ECD calculations at the CAM-B3LYP/6-311G (d,p) level were performed for four stereoisomers: (7S,8S,11R,14R)-1a, (7S,8S,11R,14S)-1b, (7R,8R,11S,14S)-1c, and (7R,8R,11S,14R)-1d. The subsequent comparison revealed that the experimental ECD curve of compound 1 (Figure 3 and Table S1) aligned conclusively with the Boltzmann-weighted theoretical spectrum of (7S,8S,11R,14R)-1a, thereby establishing its absolute configuration (Figure 3).
Figure 3. Experimental ECD of compound 1 in MeOH and the calculated ECD spectra of 1a–1d.
Penijanthinone C (2) was obtained as a pale-yellow powder. Its molecular formula was assigned as C22H26O6 (corresponding to 10 degrees of unsaturation) based on HRESIMS (m/z 385.1650 [M − H]−; Δ –0.1 mmu). The 1D NMR data of compound 2 (Table 2) and HSQC spectrum (Figure S9) displayed signals for two exchangeable protons (δH 5.80 and 3.43), five methyl groups, three methylenes, one sp3 methine, eight olefinic or aromatic carbons (two protonated), two sp3 quaternary carbons that are oxygenated (δC 76.2 and 83.4, respectively), one carbonyl carbon (δC 206.7), one α,β-unsaturated carbonyl carbon (δC 191.9), and one unsaturated ester carbonyl carbon (δC 166.1). These data were consistent with all NMR resonances and implied that compound 2 contains three rings. The comparison of the NMR data for compound 2 with those of penijanthinone A suggested a similar tricyclic polyketide core [20]. HMBC correlations (Figure 2 and Figure S11) from H2-1 to C-5, C-13, and C-14; from H-4 to C-5; and from H-7 to C-5, C-6, C-8, and C-13 as well as 1H-1H COSY correlations (Figure 2 and Figure S10) of H2-1/H2-2/H2-3/H-4/H3-15 established a tetrahydronaphthalenol substructure (rings A and B), with the C-15 methyl group attached to the C-4 position. Further HMBC cross-peaks from H3-16 to C-8, C-9, and the carbonyl carbon C-10 (δC 206.7); from H3-17 to C-11, C-10, and C-12 (δC 191.9); and from H-7 to C-9 and C-12 indicated the presence of a cyclohexane-1,3-dione moiety (ring C) fused to the tetrahydronaphthalenol substructure at C-8 and C-13, thus completing the tetrahydrophenanthrene-2,4(1H,3H)-dione tricyclic core skeleton. Other HMBC correlations from H-3′ to the C-1′ (δC 166.1) ester carbonyl carbon; C-5′, C-4′, and C-2′ from H3-4′ to C-3′ and C-2′; and from H3-5 to C-3′, C-2′, and C-1′ established a 2-methylbut-2-enoic acid moiety located at the C-11 position of tetrahydrophenanthrene-2,4(1H,3H)-dione to form 2-methylbut-2-enoate. This was further confirmed based on the chemical shift in the oxygenated sp3 quaternary carbon C-11 (δC 83.4) and the molecular weight of compound 2.
Ultimately, the structure of compound 2 was unequivocally confirmed by the single-crystal X-ray crystallography (ORTEP plot), which also enabled the determination of its relative configuration, as shown in Figure 4.
Figure 4. X-ray crystallography ORTEP plot of compound 2.
The absolute configuration of compound 2 was determined by comparing its ECD spectrum with those calculated via the TDDFT method at the CAM-B3LYP/6-311G(d,p) level for two possible stereoisomers (4R,9S,11S)-2a and (4S,9R,11R)-2b (Figure 5 and Table S2). The close match with the calculated spectrum of compound 2a confirmed the absolute configuration of 4R,9S,11S for compound 2.
Figure 5. Experimental ECD of compound 2 in MeOH and calculated ECD spectra for 2a and 2b.
The known compounds 3–8, isolated from the crude extract, were identified as harzianic acid (3) [10], 3-methyl-2-(2-nonenyl)-4(1H)-quinolinone (4) [11], emodic acid (5) [12], alaternin (6) [13], violaceol-I (7) [14], and violaceol-II (8) [14], respectively, based on comparative NMR and MS analyses with published data.

3.2. Cytotoxicity

Compounds 1–3 were tested for cytotoxicity against a panel of human tumor cell lines. Compound 3 exhibited significant activity, with IC50 values of 3.8 µM (against MGC cells) and 4.2 µM (against A549 cells). In contrast, the positive control paclitaxel exhibited potent activity against the tested cells, showing low IC50 values (e.g., 1.6 × 10−5 µM for Hep3B-2 cells) (Table 3).
Table 3. Cytotoxicity of compounds 1–3.

3.3. Network Pharmacology Prediction for Compound 3

The intersection of compound 3 and gastric cancer was visualized by a Venn diagram, identifying 266 intersection targets (Figure 6B). Subsequently, a “compound–gastric cancer target” network (Figure 6C) was established by Cytoscape 3.9.1.
Figure 6. Identification and analysis of intersection targets shared by harzianic acid (3) and gastric cancer. (A) Chemical structure of harzianic acid (3). (B) Venn diagram showing the overlapping targets harzianic acid (3) and gastric cancer. (C) Compound–disease–target network of harzianic acid (3) and gastric cancer. (D,E) Go functional enrichment analysis and KEGG pathway analysis of harzianic acid (3) and gastric cancer. (F) The PPI network of harzianic acid (3) and gastric cancer.
To delineate the mechanism, GO functional annotation and KEGG pathway enrichment analyses were performed on the shared targets. The GO enrichment analysis (Figure 6D) revealed that these targets were primarily involved in receptor-mediated signaling transduction, peptidyl-tyrosine phosphorylation, and the regulation of cell proliferation and apoptosis. Furthermore, the molecular functions of these targets were varied, ranging from nuclear receptor and protein tyrosine kinase activity to serine-type endopeptidase and histone kinase activity. These molecular functions are closely associated with the processes of abnormal proliferation, apoptosis evasion, and invasion–metastasis in gastric cancer cells [21,22]. Importantly, the KEGG pathway enrichment analysis (Figure 6E) indicates that the major enriched pathways included the cancer pathway MAPK and PI3K-Akt signaling pathways.
These intersection targets were imported into STRING to construct a PPI network, which was then topologically analyzed through two rounds of CytoNCA screening using median value cutoffs (Figure 6F). This analysis identified 13 core targets, including EGFR, HSP90AA1, AKT1, and others. They are involved in several classical signaling pathways, including PI3K-Akt and MAPK signaling pathways, as well as pathways in cancer [23], which aligned with the findings from the KEGG pathway enrichment analysis. As a receptor tyrosine kinase, the key gene EGFR regulated multiple signaling pathways associated with cell proliferation, apoptosis, invasion, and metastasis [24]. Collectively, these synergistic multi-target interactions may cooperatively modulate the gastric cancer microenvironment and contribute to the reversal of the malignant phenotype.

3.4. Molecular Docking Analysis of Compound 3

The interaction analysis between the EGFR and compound 3 indicates that the protein and ligand form a stable complex predominantly driven by electrostatic interactions, together with hydrophobic interactions and hydrogen bonds (Figure 7). Notably, a salt bridge interaction is formed between the ligand and the EGFR: the side chain ammonium group of Lys745 serves as a positive charge center and establishes a salt bridge with the carboxyl group of compound 3 [25]. With respect to hydrophobic interactions, multiple hydrophobic residues form a compact hydrophobic network with the ligand [26]. Specifically, Leu718 establishes dual hydrophobic contacts with the carbon skeleton of the ligand, and Lys728, Leu792, and Leu844 also participate in hydrophobic interactions, while the main chain amino group of Cys797 interacts with the oxygen atom of the ligand to form a hydrogen bond. Although the number of hydrogen bonds is relatively limited, this interaction acts synergistically with the salt bridge and hydrophobic interactions to stabilize the complex. As a result, the ligand achieves stable and specific binding by forming a salt bridge with Lys745 within the protein binding pocket, engaging in dense hydrophobic interactions with multiple leucine residues, and being further stabilized by a hydrogen bond involving Cys797. This binding mode, characterized by electrostatic interactions as the dominant driving force in concert with multiple non-covalent interactions, provides an important structural basis for elucidating the molecular recognition mechanism between the ligand and protein. A negative binding affinity indicates the feasibility of binding, and, generally, a lower value corresponds to stronger binding [27]. In this complex, the docking results yielded a binding affinity of −7.79 kcal/mol for the EGFR and harzianic acid (3), indicating a favorable binding interaction between the protein and the ligand.
Figure 7. The molecular docking of harzianic acid (3) complexed with the EGFR. In the figure, the small molecule is shown as green sticks, and the protein is represented by a light-cyan cartoon. Hydrogen bond, hydrophobic, and salt bridge interactions are shown as yellow, gray, and blue dashed lines, respectively.

4. Discussion

Pseudogymnone A (1) is a new pyridinone alkaloid that shares the same core skeleton as the known compound TMC-69 [28] but differs in its substituents. Specifically, compound 1 features a 3-methyl-1-pentenyl group at C-11, whereas TMC-69 lacks substitution at C-11 and instead possesses a longer 6-methyl-1,3,5-octatrienyl polyene chain at the C-10 position, while penijanthinone C (2) is a new tricyclic polyketide, bearing a similar structural skeleton to penijanthinones A and B [20]. The main difference is that compound 2 is substituted by a side chain at the C-11 position, whereas penijanthinone A carries a longer side chain at the C-3 position. Furthermore, compound 2 lacks two hydroxyl groups located at C-2 and C-4 of penijanthinone A.
In addition, bioactivity assays showed that harzianic acid (3) exhibited selective cytotoxicity against both MGC (IC50 = 3.8 µM) and A549 (IC50 = 4.2 µM) cell lines. Harzianic acid (3) was first described in 1994 as an antifungal agent [10]. Compound 3 inhibited the growth of Pythium irregulare and Sclerotinia sclerotiorum [29], eradicated methicillin-resistant Staphylococcus pseudintermedius at 16 mg/L [30], and acted as a Gram-positive-selective membrane disruptor of Bacillus subtilis [31]. Furthermore, compound 3 has also been identified as a novel natural product inhibitor of fungal acetohydroxyacid synthase (AHAS), retaining efficacy against herbicide-resistant enzyme variants. Its selectivity, arising from a unique binding mode involving the C3′ side chain that is incompatible with the plant AHAS active site, coincides with its reported plant growth-promoting properties [32]. To date, only one brief report has noted that compound 3 exhibits weak anti-liver-cancer activity with an IC50 of 58.92 ± 1.81 μM against SMMC-7721 cells.
The molecular docking analysis predicted that compound 3 binds to the EGFR with a binding affinity of −7.79 kcal mol−1, with the carboxyl group of compound 3 forming a salt bridge to Lys745; hydrophobic interactions with Leu718, Leu792, and Leu844; and a hydrogen bond with Cys797, consistent with the network pharmacology prediction that the EGFR is a potential key target of this compound. This finding aligns with the ongoing evolution of EGFR-directed anticancer agents, which have progressed from conventional monoclonal antibodies [33] to innovative fusion proteins [34] and naturally derived small molecules [35]. However, the present work is constrained by the low fermentation yield of harzianic acid (3), which limited further bioactivity evaluations beyond cytotoxicity assays. Moreover, its proposed interaction with the EGFR is based solely on network pharmacology and molecular docking predictions.

5. Conclusions

In summary, two new compounds, pseudogymnone A (1, a pyridinone alkaloid) and penijanthinone C (2, a tricyclic polyketide), along with six known compounds were obtained from the fungus P. roseus. The biological evaluation revealed that harzianic acid (3) exhibits potent and selective cytotoxicity against gastric cancer (MGC) and lung adenocarcinoma (A549) cell lines. Integrated network pharmacology predictions and molecular docking analyses implied that the EGFR is a potential target of compound 3. This study expands the chemical space of Cordyceps-colonizing fungal metabolites and identifies compound 3 as a potential EGFR-interacting scaffold that merits further experimental validation as part of future anti-gastric-cancer drug discovery efforts.

Supplementary Materials

The following Supplementary Materials can be downloaded at https://www.mdpi.com/article/10.3390/biom16020187/s1: Method S1. General experimental procedure; Method S2. ECD calculation methods; Method S3. MTS assay; Figure S1. 1H NMR spectrum of pseudogymnone A (1; 500 MHz, CDCl3); Figure S2. 13C NMR spectrum of pseudogymnone A (1; 125 MHz, CDCl3); Figure S3. HMBC spectrum of pseudogymnone A (1; 500 MHz, CDCl3); Figure S4. HSQC spectrum of pseudogymnone A (1; 500 MHz, CDCl3); Figure S5. 1H-1H COSY spectrum of pseudogymnone A (1; 400 MHz, CDCl3); Figure S6. NOESY spectrum of pseudogymnone A (1; 500 MHz, CDCl3); Figure S7. 1H NMR spectrum of penijanthinone C (2; 600 MHz, CDCl3). Figure S8. 13C NMR spectrum of penijanthinone C (2; 125 MHz, CDCl3); Figure S9. HSQC spectrum of penijanthinone C (2; 600 MHz, CDCl3); Figure S10. 1H-1H COSY spectrum of penijanthinone C (2; 600 MHz, CDCl3); Figure S11. HMBC spectrum of penijanthinone C (2; 600 MHz, CDCl3); Table S1. ECD conformers of pseudogymnone A (1); and Table S2. ECD conformers of penijanthinone C (2).

Author Contributions

Conceptualization, L.L.; methodology, J.L. and L.L.; resources, J.L. and L.L.; data curation, J.L., L.L., Y.G. and J.W.; writing—original draft preparation, J.L., L.L., Y.G. and J.W.; writing—review and editing, J.L., Y.G. and L.L.; visualization, Y.G., J.W., L.L. and F.W.; funding acquisition, J.L., F.W. and L.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Major Project of Natural Science Research in Jiangsu Higher Education Institutions (24KJA350001), the National Natural Science Foundation of China (32022002), and the APC was funded by the 2023 Talent Development Special Funds Support Project of Jinan City (202333007).

Institutional Review Board Statement

Not applicable.

Data Availability Statement

All data generated or analyzed in this study are available within the manuscript and are available from the corresponding author upon request.

Conflicts of Interest

The authors declare no conflict of interest.

Abbreviations

NMRNuclear Magnetic Resonance
MSMass Spectrometry
ECDElectronic Circular Dichroism
IRInfrared
ESIMSElectrospray Ionization Mass Spectrometry
HRESIMSHigh-Resolution Electrospray Ionization Mass Spectrometry
CCColumn Chromatography
RP HPLCReversed Phase High-Performance Liquid Chromatography
MeOHMethanol
CH2Cl2Dichloromethane
EtOAcEthyl Acetate
PDAPotato Dextrose Agar
PPIsProtein–Protein Interactions
GOGene Ontology
KEGGKyoto Encyclopedia of Genes and Genomes
DAVIDDatabase for Annotation, Visualization, and Integrated Discovery
PDBProtein Data Bank
TDDFTTime-Dependent Density Functional Theory
COSYCorrelation Spectroscopy
HMBCHeteronuclear Multiple-Bond Correlation
HSQCHeteronuclear Single Quantum Coherence
NOESYNuclear Overhauser Effect Spectroscopy
DMSODimethyl Sulfoxide
EGFREpidermal Growth Factor Receptor

References

  1. Kała, K. Advances in fungal natural products: Insights into bioactivity and therapeutic potential. Pharmaceuticals 2025, 18, 1313. [Google Scholar] [CrossRef] [Scilit]
  2. Shi, Y.; Ji, M.; Dong, J.; Shi, D.; Wang, Y.; Liu, L.; Feng, S.; Liu, L. New bioactive secondary metabolites from fungi: 2023. Mycology 2024, 15, 283–321. [Google Scholar] [CrossRef] [Scilit]
  3. Bao, M.; Shi, Y.; Gong, X.; Guo, Y.; Wang, J.; Chen, X.; Liu, L. New bioactive secondary metabolites from fungi: 2024. Mycology 2025, 16, 961–987. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Zakariyah, R.F.; Ajijolakewu, K.A.; Ayodele, A.J.; Folami-A, B.I.; Samuel, E.P.; Otuoze, S.O.; Abdulrauf, L.B.; Ahmed, R.N. Progress in endophytic fungi secondary metabolites: Biosynthetic gene cluster reactivation and advances in metabolomics. Bull. Natl. Res. Cent. 2024, 48, 44–62. [Google Scholar] [CrossRef] [Scilit]
  5. Giurini, E.F.; Godla, A.; Gupta, K.H. Redefining bioactive small molecules from microbial metabolites as revolutionary anticancer agents. Cancer Gene Ther. 2024, 31, 187–206. [Google Scholar] [CrossRef] [Scilit]
  6. Ying, Y.; Qiu, Y.; Tang, J.; Lin, Y.; Xu, S.; Ni, J.; Zhang, Y.; Tong, C.; Mao, W. A new sesterterpenoid with β-glucuronidase inhibitory activity from the Cordyceps-colonising fungus Aspergillus versicolour ZJUTE2. Nat. Prod. Res. 2025, 1–8. [Google Scholar] [CrossRef] [Scilit]
  7. Deng, L.; Niu, S.; Liu, X.; Che, Y.; Li, E. Coniochaetones E–I, new 4H-chromen-4-one derivatives from the Cordyceps-colonizing fungus Fimetariella sp. Fitoterapia 2013, 89, 8–14. [Google Scholar] [CrossRef] [Scilit]
  8. Guo, H.; Hu, H.; Liu, S.; Liu, X.; Zhou, Y.; Che, Y. Bioactive p-terphenyl derivatives from a Cordyceps-colonizing isolate of Gliocladium sp. J. Nat. Prod. 2007, 70, 1519–1521. [Google Scholar] [CrossRef] [Scilit]
  9. Zhang, Y.G.; Liu, S.C.; Liu, H.W.; Liu, X.Z.; Che, Y.S. Cycloaspeptides F and G, cyclic pentapeptides from a Cordyceps-colonizing isolate of Isaria farinosa. J. Nat. Prod. 2009, 72, 1364–1367. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  10. Sawa, R.; Mori, Y.; Linuma, H.; Naganawa, H.; Hamada, M.; Yoshida, S.; Furutani, H.; Kajimura, Y.; Fuwa, T.; Takeuchi, T. Harzianic acid, a new antimicrobial antibiotic from a fungus. J. Antibiot. 1994, 47, 731–732. [Google Scholar] [CrossRef] [Scilit]
  11. Moon, S.-S.; Kang, P.M.; Park, K.S.; Kim, C.H. Plant growth promoting and fungicidal 4-quinolinones from Pseudomonas cepacia. Phytochemistry 1996, 42, 365–368. [Google Scholar] [CrossRef] [Scilit]
  12. Alvi, K.A.; Nair, B.; Gallo, C.; Baker, D. Screening of microbial extracts for tyrosine kinase inhibitors. J. Antibiot. 1997, 50, 264–266. [Google Scholar] [CrossRef] [Scilit]
  13. Choi, J.S.; Lee, H.J.; Kang, S.S. Alatemin, cassiaside and rubrofusarin gentiobioside, radical scavenging principles from the seeds of cassia tora on 1, 1-diphenyl-2-picrylhydrazyl (DPPH) radical. Arch. Pharmacal Res. 1994, 17, 462–466. [Google Scholar] [CrossRef] [Scilit]
  14. Hirano, N.; Kohno, J.; Tsunoda, S.; Nishio, M.; Kishi, N.; Okuda, T.; Kawano, K.; Komatsubara, S.; Nakanishi, N. TMC-69, a new antitumor antibiotic with Cdc25A inhibitory activity, produced by Chrysosporium sp. TC1068: Taxonomy, fermentation and biological activitie. J. Antibiot. 2001, 54, 421–427. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. Lin, J.; Huo, R.-Y.; Hou, L.; Jiang, S.; Wang, S.-L.; Deng, Y.-L.; Liu, L. New polyketides from the basidiomycetous fungus Pholiota sp. J. Asian Nat. Prod. Res. 2023, 25, 674–682. [Google Scholar] [CrossRef] [Scilit]
  16. Sheldrick, G.; Sheldrick, G.M.; Sheldrick, G.M. SHELXL-97, A Program for Crystal Structure Refinement; University of Göttingen: Göttingen, Germany, 1997. [Google Scholar]
  17. Shi, Y.; Sun, X.Q.; Zhang, J.X.; Zhang, R.H.; Hong, K.; Xue, Y.X.; Qiu, H.; Liu, L. New cytotoxic γ-lactam alkaloids from the mangrove-derived fungus Talaromyces hainanensis sp. nov. guided by molecular networking strategy. J. Agric. Food Chem. 2024, 72, 17431–17443. [Google Scholar] [CrossRef] [Scilit]
  18. Huo, R.; Dong, J.; Liu, G.; Shi, Y.; Liu, L. Cytotoxic anthrone–cyclopentenone heterodimers from the fungus Penicillium sp. guided by molecular networking. Chin. J. Nat. Med. 2025, 23, 1259–1267. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Liu, L.; Chen, X.; Li, D.; Zhang, Y.; Li, L.; Guo, L.; Cao, Y.; Che, Y. Bisabolane sesquiterpenoids from the plant endophytic fungus Paraconiothyrium brasiliense. J. Nat. Prod. 2015, 78, 746–753. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  20. Chen, M.; Zheng, Y.Y.; Chen, Z.Q.; Shen, N.X.; Shen, L.; Zhang, F.M.; Zhou, X.J.; Wang, C.Y. NaBr-induced production of brominated azaphilones and related tricyclic polyketides by the marine-derived fungus Penicillium janthinellum HK1-6. J. Nat. Prod. 2019, 82, 368–374. [Google Scholar] [CrossRef] [Scilit]
  21. Li, H.; Guan, B.; Liu, S.; Liu, H.; Song, L.; Zhang, G.; Zhao, R.; Zhou, C.; Gao, P. PTPN14 promotes gastric cancer progression by PI3KA/AKT/mTOR pathway. Cell Death Dis. 2023, 14, 188. [Google Scholar] [CrossRef] [Scilit]
  22. Zhao, T.; Chen, Z.; Liu, W.; Ju, H.; Li, F. Identification of hub genes associated with gastric cancer via bioinformatics analysis and validation studies. Int. J. Gen. Med. 2023, 16, 4835–4848. [Google Scholar] [CrossRef] [Scilit]
  23. Gu, J.Y.; Li, F.J.; Hou, C.Z.; Zhang, Y.; Bai, Z.X.; Zhang, Q. Mechanism of icariin for the treatment of osteoarthritis based on network pharmacology and molecular docking method. Am. J. Transl. Res. 2023, 15, 5071–5084. [Google Scholar] [PubMed]
  24. Tomuleasa, C.; Tigu, A.B.; Munteanu, R.; Moldovan, C.S.; Kegyes, D.; Onaciu, A.; Gulei, D.; Ghiaur, G.; Einsele, H.; Croce, C.M. Therapeutic advances of targeting receptor tyrosine kinases in cancer. Signal Transduct. Target. Ther. 2024, 9, 201. [Google Scholar] [CrossRef] [Scilit]
  25. Rezaei, M.; Sakong, S.; Groß, A. Molecular modeling of water-in-salt electrolytes: A comprehensive analysis of polarization effects and force field parameters in molecular dynamics simulations. J. Chem. Theory Comput. 2023, 19, 5712–5730. [Google Scholar] [CrossRef] [Scilit]
  26. Bertalan, É.; Konno, M.; del Carmen Marín, M.; Bagherzadeh, R.; Nagata, T.; Brown, L.; Inoue, K.; Bondar, A.-N. Hydrogen-bonding and hydrophobic interaction networks as structural determinants of microbial rhodopsin function. J. Phys. Chem. B 2024, 128, 7407–7426. [Google Scholar] [CrossRef] [Scilit]
  27. Shami, A.; Alharbi, N.K.; Al-Saeed, F.A.; Alsaegh, A.A.; Al Syaad, K.M.; Abd El-Rahim, I.H.A.; Mostafa, Y.S.; Ahmed, A.E. In silico subtractive proteomics and molecular docking approaches for the identification of novel inhibitors against Streptococcus pneumoniae strain D39. Life 2023, 13, 1128. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  28. Kohno, J.; Hirano, N.; Sugawara, K.; Nishio, M.; Hashiyama, T.; Nakanishi, N.; Komatsubara, S. Structure of TMC-69, a new antitumor antibiotic from Chrysosporium sp. TC 1068. Tetrahedron 2001, 57, 1731–1735. [Google Scholar] [CrossRef] [Scilit]
  29. Vinale, F.; Flematti, G.; Sivasithamparam, K.; Lorito, M.; Marra, R.; Skelton, B.W.; Ghisalberti, E.L. Harzianic acid, an antifungal and plant growth promoting metabolite from Trichoderma harzianum. J. Nat. Prod. 2009, 72, 2032–2035. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  30. De Filippis, A.; Nocera, F.P.; Tafuri, S.; Ciani, F.; Staropoli, A.; Comite, E.; Bottiglieri, A.; Gioia, L.; Lorito, M.; Woo, S.L.; et al. Antimicrobial activity of harzianic acid against Staphylococcus pseudintermedius. Nat. Prod. Res. 2021, 35, 5440–5445. [Google Scholar] [CrossRef] [Scilit]
  31. Ouyang, X.; Hoeksma, J.; Beenker, W.A.G.; van der Beek, S.; den Hertog, J. Harzianic acid exerts antimicrobial activity against Gram-positive bacteria and targets the cell membrane. Front. Microbiol. 2024, 15, 1332774. [Google Scholar] [CrossRef] [Scilit]
  32. Xie, L.; Zang, X.; Cheng, W.; Zhang, Z.; Zhou, J.; Chen, M.; Tang, Y. Harzianic acid from Trichoderma afroharzianum is a natural product inhibitor of acetohydroxyacid synthase. J. Am. Chem. Soc. 2021, 143, 9575–9584. [Google Scholar] [CrossRef] [Scilit]
  33. Ruan, D.-Y.; Wu, H.-X.; Xu, Y.; Munster, P.N.; Deng, Y.; Richardson, G.; Yan, D.; Lee, M.-A.; Lee, K.-W.; Pan, H.; et al. Garsorasib, a KRAS G12C inhibitor, with or without cetuximab, an EGFR antibody, in colorectal cancer cohorts of a phase II trial in advanced solid tumors with KRAS G12C mutation. Signal Transduct. Target. Ther. 2025, 10, 189. [Google Scholar] [CrossRef] [Scilit]
  34. Hallaji, M.; Allahyari, M.; Teimoori-Toolabi, L.; Yasami-Khiabani, S.; Golkar, M.; Fard-Esfahani, P. Targeted cancer treatment using a novel EGFR-specific Fc-fusion peptide based on GE11 peptide. Sci. Rep. 2025, 15, 5107. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Dipa, C.D.; Hossain, S.; Chy, M.M.K.; Rahman, M.S.F.; Kayes, T.; Easha, A.M.; Moin, A.T.; Vakare, S.N.; Patil, R.B.; Sarkar, B.K. In silico exploration of anticancer plant phytochemicals for EGFR-targeted lung cancer therapy. Sci. Rep. 2025, 15, 27809. [Google Scholar] [CrossRef] [Scilit] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.