Introduction

Fungi own a developed secondary metabolism that is attractive for the potentiality to produce abundant secondary metabolites possessing meaningful structural features, including polyketides, terpenoids, alkaloids, steroids, and miscellaneous types (Keller 2019; Liu et al. 2017). These metabolites not only possess diversified structural skeletons but also exhibit a very wide range of bioactivities, many of which have been developed into life-saving medicines and agrochemicals (Helaly et al. 2018). Studies of fungal secondary metabolites have yielded a large number of important pharmaceuticals and agrochemicals, such as penicillins (nonribosomal peptide, isolated from Penicillium species), cephalosporin C (nonribosomal peptide, isolated from Acremonium chrysogenum), griseofulvin (halogenated polyketide, isolated from P. griseofulvum, P. aethiopicum, P. coprophilum, and other Penicillium species), lovastatin (polyketide, isolated from Aspergillus terreus and Monascus purpureus), and fumagillin (meroterpenoid, isolated from Aspergillus fumigatus), which have afforded significant society-changing benefits (Fox, 2008). Moreover, natural products remain one of the most significant therapeutic agents and lead compounds in medicine, especially crucial in the development of effective therapies for cancer, malaria, microbial infections, neurological/cardiovascular diseases, and immunopathy (Newman and Cragg 2016). It is estimated that a significant proportion of approved natural-sourced therapeutics are actually derived from microorganisms, particularly fungi (Newman and Cragg 2020). Nowadays, the upward trend of chemical research into the secondary metabolites of fungi continues undiminished, which has yielded the discovery of a mass of intriguing compounds (Gao and Zhang 2023; Xu et al. 2021).

Endophytic fungi living harmoniously in host plants are considered a treasure house for the discovery of novel secondary metabolites with high medical potentiality. Recently, endophytic fungi have drawn tremendous attention from pharmaceutical researchers. Over the last few years, endophytic fungi have proven to be an abundant source of natural bioactive metabolites with important application values in the pharmaceutical industry (Gupta et al. 2020). Up to now, Taxol, a miscellaneous diterpenoid alkaloid characterized from the endophytic fungus Metarhizium anisopliae living in the bark of Taxus tree, is one of the most potential anticancer agents (Zhang et al. 2009). An endophytic fungus Neurospora sp., isolated from the seed of Nothapodytes foetida, was found to be a potential source of anticancer drug lead compound camptothecin (a pentacyclic quinoline alkaloid), when cultured in Sabouraud liquid culture media (Puri et al. 2005). As a result, endophytic fungi have been considered novel sources of anticancer lead molecules (Chandra 2012). It has reported that more than 100 various fungal species have been found to synthesize over 200 meaningful anticancer compounds, which demonstrate antiproliferative/cytotoxic activities against more than 60 different human malignant tumor cells (Hridoy et al. 2022). The structures of these anticancer fungal secondary metabolites include alkaloids and nitrogen-containing heterocycles, benzo[j]fluoranthenes, chromones, coumarins, depsidones, depsipeptides, ergochromes, esters, lactones, lignans, peptides, polyketides, quinones, spirobisnaphthalenes, terpenes (diterpenes, sesquiterpenes, triterpenes), and xanthones (Hridoy et al. 2022). It should be noted that most of these active compounds have been studied at the molecular level to elucidate the corresponding mode of action. Due to the limited yields and isolation difficulties, these secondary metabolites may not be available to perform in vivo investigations in animal models.

Prenylated indole alkaloids are a class of secondary metabolites isolated from marine and terrestrial derived fungi (Klas et al. 2018). The main types include brevianamide, marcfortine, paraherquamide, penicimutamide, sclerotiamide, asperparaline, avrainvillamide, stephacidin, malbrancheamide, notoamide, versicolamide, chrysogenamide, and the recently discovered taichunamide (Klas et al. 2018). Chemically, prenylated indole alkaloids contain aromatic and isoprene groups, as well as a characteristic bicyclo[2.2.2]diazaoctane ring structure unit. Biogenetically, prenylated indole alkaloids are composed of proline and isopentenyl tryptophan. These metabolites possess a complex ring system, and their stability, novelty, and unique structure are of great research value in biosynthesis, organic synthesis, and biological evolution (Greshock et al. 2008; Finefield et al. 2012; Miller and Williams 2009). In our continued interests to explore metabolites from endophytic fungi (Li et al. 2021; Su et al. 2022; Xu et al. 2020; Yuan et al. 2020), the endophytic fungus Aspergillus japonicus TE-739D was isolated from fresh leaves of Nicotiana tabacum and selected for further chemical investigations. Chemically guided isolation of the fungal strain TE-739D yielded eight structurally diverse prenylated indole alkaloids, including an undescribed compound, namely aspertaichamide B (ATB, compound 1), together with seven previously discovered derivatives (compounds 2 − 8) (Fig. 1). Based on HRESIMS, NMR, and ECD calculations, their structures were finally determined. The new compound ATB was found to display considerable anti-tumor activity against the gastric cancer MFC cell line both in vitro and in vivo. In this study, the separation, structural determination, and anti-tumor activity of the new compound 1 are described.

Fig. 1
figure 1

Chemical structures of 1–8

Materials and methods

General experimental procedures

The optical rotation was acquired by a Jasco P-1020 digital polarimeter and the UV data was gained by a Shimadzu UV-2700 spectrophotometer. NMR data were gained by an Agilent DD2 NMR spectrometer at 500 (1H) and 125 (13C) MHz. HRESIMS experiments were carried out on a Waters ACQUITY UPLC I-Class-Vion IMS Q-TOF mass spectrometer. Column chromatography (CC) was carried out by silica gel, RP-18, and Sephadex LH-20.

Fungal material

The endophytic fungus A. japonicus TE-739D was isolated from fresh leaves of Nicotiana tabacum that was obtained from Enshi Autonomous Prefecture, Hubei Province, China. Fungal identification was achieved by DNA amplification and the internal transcribed spacer (ITS) sequencing (GenBank accession number: PP126510). This strain has been deposited both in the Tobacco Research Institute of Chinese Academy of Agricultural Sciences and in the China General Microbiological Culture Collection Center under a patent depository number 40901.

Fermentation, extraction, and isolation

This fungus was cultured in PDA medium at 27 °C for 4 days to get the seed cultures. The agar plugs, which are cut into appropriate pieces, were introduced into a 1-L Erlenmeyer flask containing with 300 mL liquid Sabouraud medium (component: 4% glucose, 1% peptone, and 300 mL distilled water). In order to obtain trace chemical compositions, 200 flasks were prepared under static conditions for 30 days; afterwards, the whole fermentation liquor were combined and extracted with sufficient EtOAc. The solvent was removed in vacuo to yield a total of 30.6 g residue, which was chromatographed on silica gel VLC with an increasing petroleum ether (PE)–EtOAc (30:1, 10:1, 5:1, 2:1, and 1:1, v/v) and CH2Cl2–MeOH (20:1, 10:1, and 1:1, v/v) mixtures to afford eight fractions (A–H).

Firstly, fraction D (2.1 g), eluted by PE–EtOAc 2:1, was separated by CC (CH2Cl2–MeOH, from 40:1 to 20:1, v/v) to obtain two major subfractions D1 (0.3 g) and D2 (1.6 g) by TLC analysis. D1 was separated by prep.-TLC (CH2Cl2–MeOH, 20:1, v/v) to give compound 6 (30.5 mg), while D2 was isolated by CC (CH2Cl2–acetone, 15:1, v/v) to give compounds 4 (20.6 mg) and 5 (17.8 mg). Then, fraction E (2.6 g), eluted by PE–EtOAc 1:1, was refractionated via RP-18 with an increasing gradient of MeOH–H2O (10–100%) to give five subfractions E1 − E5. E3 (0.9 g) was applied to Sephadex LH-20 to give 1 (200 mg). E4 (0.5 g) was separated by prep.-TLC (CH2Cl2–MeOH = 20:1) to give 3 (11.6 mg). E5 (0.4 g) was subjected to prep.-TLC (CH2Cl2–MeOH, 20:1) and Sephadex LH-20 to generate 2 (10.9 mg). Finally, fraction F (1.8 g), eluted by CH2Cl2–MeOH 20:1, was separated by a combination of CC (CH2Cl2–MeOH = 25:1) and Sephadex LH-20 to give 7 (9.8 mg) and 8 (12.0 mg), respectively.

Aspertaichamide B (1): colorless oil; [α] + 26.0 (c = 0.13, MeOH); UV (MeOH) λmax (log ε) 230 (4.32), 260 (4.34), 388 (3.68) nm; ( −)-HRESIMS m/z 506.2289 [M − H] (calcd for C28H32N3O6, 506.2291); and 1H and 13C NMR data (CD3OD) are shown in Table 1.

Table 1 NMR spectroscopic data for compound 1 (1H at 500 MHz and 13C at 125 MHz, measured in CD3OD)

Computational section

Detailed quantum chemical calculations were provided in Supplementary materials.

Anti-tumor assay

Chemicals and reagents

2',7'-Dichlorodihydrofluorescein diacetate (DCFH-DA), Sigma-Aldrich (USA). RPMI-1640 medium, FBS, and trypsin–EDTA, HyClone (USA). Annexin V-FITC/PI apoptosis detection kit, KeyGEN (China). CCK8 and PEG400, MedChemExpress (China).

Cell culture and cell viability assay

The gastric cancer MFC cells were preserved in an incubator at 37 °C with a humidity of 5% CO2. In a 96-well plate, MFC cells were inoculated with an initial 3 × 104 cells. After 12 h, the medium was removed and RPMI-1640 with 10% FBS and ATB (0.2, 0.5, 1, 2, 4, 8, 16, and 32 µmol/L) was introduced and the culture was preserved for 24 − 48 h. Then, 10 µL of CCK-8 solution was introduced. Absorbance was measured at 450 nm.

Colony formation test

Approximately 1000 cells were incubated in 6-well microplates. Following the treatment of 24 h, the medium without the drug was introduced for 10 days. In order to facilitate visualization, the resulting colonies were immobilized by methanol and stained by using crystal violet solution.

Cell wound healing assay

MFC cells were inoculated in a six-well plate for 24 h to form adherent monolayers. A scratch was made on the cell surface with the tip of the pipette. Then 0, 2, and 4 µM ATB were introduced for 24 h.

Determination of cell death

In summary, a 6-well plate was inoculated with 1 × 106 cells per well. After overnight adhesion, cells were removed twice and exchanged with 2 and 4 µM ATB for 24 h. Cells were collected. After multiple washes, cells were re-suspended.

Animal experiments

BALB/c nude mice were obtained from Shanghai SLAC Laboratory. For the subcutaneous tumor model, 100-µL tumor cell suspension with 5 × 105 MFC cells was subcutaneously inoculated for each mouse. The tumor growth was observed for every 3 days. Tumor volume (mm3) = 0.5 × length × width2. The mice were randomly given ATB (dissolved in PEG400, saline, and ethanol) at 10 mg/kg and 15 mg/kg, intraperitoneally injected once a day to the endpoint.

Results

Isolation and identification of the producing strain

The producing strain A. japonicus TE-739D was isolated from fresh leaves of cultivated tobacco (Nicotiana tabacum L.), which was collected from Enshi Autonomous Prefecture, Hubei Province, China, in August 2018. The fresh leaves of tobacco were suffered with strict surface sterilization treatment (75% ethanol for 1 min, 2.5% sodium hypochlorite for 30 s, and 75% ethanol for 1 min) (Zhang et al. 2018). Therefore, the obtained fungal strain A. japonicus TE-739D was considered an endophyte residing in tobacco. This fungus was identified by morphological characters cultured on different culture media (Fig. 2A) and the sequencing of the ITS of the rRNA locus. The ITS sequence, which displayed 99% identical to that of A. japonicus (GenBank accession, KY199566), has been submitted to GenBank with the accession number of PP126510. To determine the evolutionary position of the fungus TE-739D, a phylogenetic analysis was performed based on the ITS sequence with those from other Aspergillus species. The results showed that the strain TE-739D was clustered with species in genus Aspergillus with a high confidence (Fig. 2B).

Fig. 2
figure 2

A Morphology of A. japonicus TE-739D on different culture media (PDA, potato dextrose agar; MEA, malt extract ager; SDA, Sabouraud dextrose agar) and B neighbor-joining tree based on ITS nucleotide sequences

Structure elucidation

The producing strain A. japonicus TE-739D was statically cultured in liquid Sabouraud medium. The whole fermentation liquor was extracted with EtOAc to yield EtOAc extracts, which were then chromatographed on successive silica gel column to yield compounds 1 − 8.

Aspertaichamide B (ATB, compound 1) was acquired as colorless oil (MeOH). The molecular formula was C28H33N3O6 as observed in ( −)-HRESIMS (m/z 506.2289 [M − H], calcd for C28H32N3O6, 506.2291) (Fig. S1). The 1H NMR data (Table 1) showed four olefinic/aromatic signals at δH 7.39 (d, H-4), 6.41 (d, H-5), 7.44 (d, H-25), and 5.84 (d, H-26), one oxymethine signal at δH 6.31 (s, H-10), nine protons ranging from δH 1.88 to 3.48, two methoxyl groups at δH 3.65 (s, H3-30) and 3.13 (s, H3-31), and four methyls [δH 0.63 (s, H3-23), 1.18 (s, H3-24), 1.54 (s, H3-28), and 1.46 (s, H3-29)] (Fig. S2). The 13C NMR together with DEPT spectra (Fig. S3) indicated the existent of 28 carbons, namely, six methyls (including two methoxyl), four methylenes, two saturated methines, four olefinic/aromatic methines, and 12 nonprotonated carbons (including three carbonyls and three oxygenated/nitrogenated sp2 carbons) (Table 1). The abovementioned functional groups showed similarity to those of previously reported prenylated indole alkaloids (Li 2010; Whyte et al. 1996). The diagnostic carbonyl signals at δC 168.9 (C-12) and 173.4 (C-18) as well as nitrogenated sp2 nonprotonated carbons at δC 60.8 (C-11) and 67.2 (C-17) suggested a bicyclo[2.2.2]diazaoctane framework (Kato et al. 2007), which was proved by the 1H − 1H COSY correlations of H2-20/H-21 and H2-14/H2-15/H2-16 (Fig. S4) and by the HMBCs of H-21 with C-12, H2-20 with C-16 and C-18, and H2-14 with C-17 (Figs. 3 and S6). Further NMR spectra of ATB suggested that ATB was identical to that of aspertaichamide A (Chen et al. 2024), except for an additional methoxyl group observed in ATB (Fig. S5). The key HMBC correlation from H3-31 to C-2 positioned the methoxyl at C-2 (Fig. 3). The gross structure of ATB was therefore elucidated.

Fig. 3
figure 3

COSY and HMBC correlations for 1

Since ATB and aspertaichamide A may be derived from the same biosynthetic pathway, the stereochemistry of ATB was tentatively determined to be the same as aspertaichamide A (Chen et al. 2024). The observed NOE interactions between H3-30 and H3-31 suggested that both methoxyls were placed on the same face of the cyclohexane (Fig. S7). However, key NOE correlations were missing, which resulted in the relative configurations unsolved. We proposed four relative configurations, named as 1a, 1b, 1c, and 1d, corresponding to (2S*,10S*,11R*,17S*,21S*)-1, (2S*,10S*,11R*,17S*,21R*)-1, (2R*,10R*,11R*,17S*,21R*)-1, and (2R*,10R*,11R*,17S*,21S*)-1, respectively. These configurations were computed by NMR calculations at the mPW1PW91-SCRF/6–31 + G(d,p) level. The calculation results suggested that configuration 1d exhibited lower MAE (mean absolute error) and CMAE (corrected mean absolute error) values than the other configurations (Table 2). Additionally, DP4 + analysis afforded a 100% probability, and linear regression analysis further revealed the closest agreement between the calculated and experimental values (Table 2). Based on these results, we finally determined 1 as 2R*,10R*,11R*,17S*,21S*. Subsequently, we established the absolute configurations of 1 to be 2R,10R,11R,17S,21S through TDDFT ECD calculations at the PBE1PBE/6-311G(d) level (Fig. 4). Molecular orbital (MO) analysis revealed that the negative Cotton effects at ~ 375 nm were corresponded from HOMO to LUMO, while at ~ 240 nm corresponded from HOMO-5 to LUMO (Fig. 5). Additionally, the positive Cotton effects at ~ 340 nm were associated with HOMO-3 to LUMO, and these at ~ 260 nm were associated with HOMO-1 to LUMO + 1. It should be noticed that the stereostructure of ATB were different from aspertaichamide only at C-2 and C-10, which was very interesting, and deserved to be further studied for the related enzymes in the biosynthesis pathway.

Table 2 Comparison of the key parameters of four possible relative configurations in NMR computations
Fig. 4
figure 4

Experimental and calculated ECD spectra of 1

Fig. 5
figure 5

Key MOs in the important transitions regarding ECD spectra

In addition, the known compounds 2 − 8 were characterized as notoamide B (2) (Kato et al. 2007), sclerotiamide (3) (Whyte et al. 1996), ( −)-versicolamide B (4) (Tsukamoto et al. 2009), stephacidin A (5) (Greshock et al. 2008), notoamide C (6) (Kato et al. 2007), waikialoid A (7) (Wang et al. 2012), and waikialoid B (8) (Wang et al. 2012), respectively, by comparison of their spectroscopic data with literatures.

Anti-tumor activity

ATB inhibited gastric cancer MFC cells proliferation and migration

To evaluate the in vitro anti-tumor efficacy of the isolated compounds 1 − 8, we conducted a series of cytotoxicity assays on gastric cancer MFC cells. The results revealed that ATB (1) exerted a pronounced inhibitory effect on MFC cell viability, as illustrated in Fig. 6A. Specifically, ATB reduced cell viability in a time-dependent manner, with IC50 values determined to be 15.6 µM at 24 h and 7.2 µM at 48 h (doxorubicin as positive control with an IC50 value of 108.1 nM at 48 h). This indicates that prolonged exposure to ATB enhanced its cytotoxic efficacy against MFC cells. Additionally, ATB significantly suppressed the clonogenic potential of MFC cells in a dose-dependent manner, as depicted in Fig. 6B. At concentrations of 2 µM and 4 µM, ATB inhibited colony formation by 57.4% and 62.1%, respectively, compared to the control group (P < 0.01). This data suggested that ATB not only reduced cell survival but also diminished the ability of MFC cells to proliferate over time. Furthermore, migration assays (wound healing) demonstrated that ATB effectively inhibited the migratory capacity of MFC cells. Specifically, at 2 µM and 4 µM concentrations, ATB reduced cell migration by 49.9% and 27.4%, respectively, when compared to the control (P < 0.05) (Fig. 6C). These results collectively underscore the potential of ATB as a potent inhibitor of both cell proliferation and migration in gastric cancer.

Fig. 6
figure 6

ATB inhibited the MFC cells proliferation and migration. A MFC cells were conducted with the indicated concentrations of ATB for indicated times. B ATB inhibited colony formation in MFC cells. C The wound healing assay in MFC cells with ATB for 24 h. *P < 0.05; **P < 0.01

ATB promoted apoptosis of gastric cancer MFC cells

To further investigate the mechanism underlying the anti-cancer effects of ATB, we examined its influence on apoptosis in MFC cells. The treatment with ATB resulted in a significant increase in apoptotic cell populations, as shown by flow cytometry analysis. In the group treated with 2 µM ATB, there was a marked elevation in both early (6.86% vs. 2.32%) and late (6.07% vs. 0.18%) apoptotic cells, compared to the untreated control group (P < 0.01). Similar trends were observed in the 4 µM ATB treatment group, which exhibited early apoptosis rates of 8.97% and late apoptosis rates of 7.98% (P < 0.01), as illustrated in Fig. 7A and B. These findings indicate that ATB not only inhibits cell growth but also actively promotes programmed cell death in gastric cancer cells.

Fig. 7
figure 7

ATB induced apoptosis in MFC cells. A MFC cells with ATB for 24 h and the detection of apoptosis. B The percentages of apoptotic cells. *P < 0.05; **P < 0.01

ATB promoted induction of ROS (reactive oxygen species) in gastric cancer MFC cells

To elucidate the potential mechanisms driving ATB-induced apoptosis, we investigated the effect of ATB on the production of reactive oxygen species (ROS) in MFC cells. Using the DCFH-DA assay, it was observed that ATB treatment significantly elevated ROS levels within the cells. This was evident from the increased fluorescence intensity of DCF, indicating enhanced ROS generation (Fig. 8A). Specifically, treatment with 2 µM and 4 µM ATB elevated ROS levels to 42.12% and 43.75%, respectively, compared to 18.58% in the control group (P < 0.05; Fig. 8B and C). These results suggest that ATB may induce oxidative stress as a key mechanism contributing to its pro-apoptotic and anti-tumor effects in gastric cancer cells.

Fig. 8
figure 8

ATB promoted induction of ROS in MFC cells. A Fluorescent images taken by a fluorescence microscope with DCFH-DA staining in MCF cells at indicated concentrations of ATB. B DCF fluorescence distribution of cells with ATB. C ROS levels at indicated concentrations of ATB. *P < 0.05; **P < 0.01

ATB inhibited tumor growth

To translate the in vitro findings into an in vivo context, we assessed the anti-tumor efficacy of ATB using a subcutaneous tumor model in nude mice, established with MFC cells. The in vivo experiments revealed that ATB treatment significantly inhibited tumor growth, as indicated by reduced tumor volumes and weights (Fig. 9A). Specifically, mice treated with 10 mg/kg and 15 mg/kg doses of ATB exhibited a 37.5% and 44.3% reduction in tumor volume, respectively, compared to the control group (Fig. 9B). Correspondingly, the average tumor weight in these groups was reduced by 37.5% and 43.7%, respectively (P < 0.05; Fig. 9C). Importantly, ATB treatment did not result in significant changes in the body weight of the treated mice, suggesting that the compound was well tolerated at these doses (Fig. 9D). These in vivo results further corroborate the potential of ATB as an effective anti-cancer agent against gastric cancer, providing a strong foundation for further investigation and potential clinical development.

Fig. 9
figure 9

ATB inhibited tumor growth. A ATB inhibited tumor growth in the MFC-derived subcutaneous mouse model. B The tumor volume of mice. C The tumor mass in ATB-treated mice. D The body weights of MFC-derived subcutaneous mouse model with ATB. *P < 0.05; **P < 0.01

Discussion

Prenylated indole alkaloids, featuring an uncommon bicyclo[2.2.2]diazaoctane framework biosynthetically originated from tryptophan and isoprene residues, are a group of structurally intriguing specialized metabolites produced by genera Aspergillus sp. and Penicillium sp. (Ding et al. 2010; Greshock et al. 2008; Liu et al. 2018). These metabolites possess unique ring systems and stereochemical characteristics and have become targets for synthesis and biosynthesis (Finefield et al. 2012; Miller and Williams 2009). Up to now, more than 40 kinds of prenylated indole alkaloids, including notoamides, paraherquamides, brevianamides, stephacidins, taichunamides, and asperparalines, have been isolated and reported (Klas et al. 2018). Besides, these alkaloids have also shown multitudinous biological properties including antifeedant, antiscolic, bacteriostatic, calmodulin inhibitory, and cytotoxic activities (Klas et al. 2018). Prenylated indole alkaloids derived from L-tryptophan and L-proline are renowned for their significant cytotoxic properties. For instance, stephacidin B demonstrated potent cytotoxicity against various human tumor cells, exhibiting IC50 values ranging from 0.06 to 0.46 µM (Qian-Cutrone et al. 2002). Sclerotiamide C showed strong cytotoxicity against HeLa, A549, HepG2, and SMMC7721 cells with an IC50 around 1.6 µM (Guo et al. 2022). Additionally, fumitremorgin C acted as a specific inhibitor of the breast cancer resistance protein and had the potential to reverse the resistance of certain tumor cell lines (Li 2010). Stephaochratidin A showed potent effects on human melanoma cell A375, human renal cell carcinoma cell 786-O, and human nonsmall cell lung cancer cell H1299 (Zou et al. 2024). Recently, Chen et al. reported a new prenylated indole alkaloid, namely, aspertaichamide A, from A. taichungensis 299 (Chen et al. 2024). Aspertaichamide A was found to possess strong cytotoxic activity and induce apoptotic death of the AGS cells. In this study, the endophytic fungus A. japonicus TE-739D was isolated from fresh leaves of Nicotiana tabacum and selected for further chemical investigations. As compared with other species belonging to the Aspergillus genus, the secondary metabolites of A. japonicus have not been extensively investigated and only led to the characterization of limited metabolites (Hayashi et al. 2000; Wang et al. 2022). As a result, eight structurally diverse prenylated indole alkaloids including an undescribed compound aspertaichamide B (ATB) were isolated and identified from this fungal strain. With the aid of spectroscopic methods, including HRESIMS, NMR, NMR calculations with DP4 + probability analysis, and ECD calculations, the structure of ATB was elucidated. It should be noted that the structure of ATB is very similar with the previously reported compound aspertaichamide A (Chen et al. 2024). The 3-pyrrolidone dimethylbenzopyran moiety in ATB and aspertaichamide A may be essential for their cytotoxic activities. Moreover, since aspertaichamide A possessed strong cytotoxic activity against the AGS cells, it is interesting to assess the cytotoxic activity of ATB against AGS cells in further studies.

In vitro cytotoxic effects against the gastric cancer MFC cell line revealed that ATB demonstrated considerable activity. Additionally, ATB dose-dependently inhibited the colony formation of MFC cells and significantly reduced their migration ability. Further studies showed that ATB promoted apoptosis in gastric cancer cells, including early and late apoptosis. ATB also increased the production of ROS in gastric cancer cells. In vivo studies suggested that ATB inhibited the growth of MFC cell-derived subcutaneous tumors, without significantly affecting the average mice body weight. Excessive ROS triggers mitochondrial dysfunction, increasing the permeability of the outer mitochondrial membrane and leading to the release of cytochrome C into the cytoplasm, which activates caspase-9 and initiates the apoptotic process (Hou et al. 2020). ROS also enhances the expression of death receptors on the cell surface, such as Fas and TNF receptor. Caspase-8 is activated when these receptors bind to their respective ligands, initiating apoptotic signaling (Roberge et al. 2014). Excess ROS can also induce endoplasmic reticulum (ER) stress, leading to the unfolded protein response (UPR). If the UPR fails to restore normal ER function, caspase-12 and caspase-4 may be activated, leading to apoptosis (Zappavigna et al. 2019). Additionally, oxidative stress can directly damage DNA, causing single-stranded or double-stranded breaks. DNA damage sensors, such as p53, are activated, promoting the expression of pro-apoptotic genes like Bax while inhibiting anti-apoptotic genes like Bcl-2, ultimately driving the cell toward apoptosis (Srinivas et al. 2019). NRF-2 and KEAP-1 play crucial roles in the cellular response to oxidative stress. KEAP-1 regulates the stability and activity of NRF-2, ensuring that the antioxidant defense mechanisms are only activated when necessary. NRF-2, in turn, helps protect cells from oxidative damage by regulating the expression of antioxidant and detoxification genes (Bellezza et al. 2018). However, when NRF-2 upregulation is combined with other oncogenic lesions, it provides a survival advantage for cancer cells (DeNicola et al. 2011).

Overall, this study demonstrated that ATB inhibited cell proliferation, promoted apoptosis, and increased ROS production in gastric cancer cells, and exhibited inhibitory effects on tumor growth in an in vivo model, indicating the potential of the fungal metabolite ATB as a promising anti-tumor lead compound. In future experiments, our group will continue focusing on the mechanism by which ATB induces apoptosis in gastric cancer cells by promoting oxidative stress. By analyzing the expression levels of key genes in the KEAP-1/NRF-2/HO-1 pathway, we will assess whether ATB has a regulatory effect on these genes. Additionally, molecular docking and surface plasmon resonance techniques will be used to analyze the binding interactions between ATB and key proteins, further elucidating the mechanism of ATB’s anti-gastric cancer activity.