Discovery of highly potent marine-derived compound Penicolinate H reveals SREBP-1 mediated lipogenesis as a druggable vulnerability in gastric cancer

Abstract

Background

Gastric cancer (GC) is a highly aggressive and fatal disease, with limited treatment options. Altered cellular lipid metabolism is a hallmark of cancer that contributes to GC progression. However, the determinants of lipid metabolism in GC and new agents that target lipid metabolic pathways are poorly defined. The aim of this study was to identify potentially effective lead anti-GC compounds by developing and exploring natural marine products. Furthermore, we sought to uncover viable therapeutic targets for GC through the underlying mechanisms of action of the compounds as an anticancer agent.

Methods

The chemical structures of the metabolites produced by the crinoid-derived fungus Penicillium brocae SYSU-CJ17 were elucidated using advanced spectroscopic techniques. These metabolites were screened for their growth-inhibitory effects on GC cell lines. Among them, Penicolinate H (Pen-H) demonstrated the most significant anti-cancer activity. High-throughput RNA sequencing of Pen-H-treated GC samples revealed differentially expressed genes, and transcriptomic data integrated with bioinformatics analyses highlighted the potential pathways and target genes through which Pen-H might exert its anti-cancer effects. Further investigations, including rescue experiments, endogenous affinity pull-down assays, cellular thermal shift assays (CETSA), surface plasmon resonance (SPR) assays, molecular docking, and in vitro analyses, confirmed the interaction between Pen-H and SREBP-1. The feasibility of SREBP-1 as a therapeutic target for GC is supported by single-cell transcriptome analysis, bioinformatics evaluation of GC patient data, and in vitro studies. Additionally, the chemosensitization effect of Pen-H was confirmed by in vivo and in vitro experiments.

Results

Our findings reveal a novel marine-derived compound, Pen-H, which inhibits GC growth and metastasis both in vitro and in vivo by suppressing SREBP-1 mediated lipogenesis. Bioinformatic analysis indicated that SREBP-1 was highly expressed in GC tissues, and high SREBP-1 transcript levels were negatively correlated with prognosis in GC patients. SREBP-1 depletion significantly inhibits the proliferation, migration, and invasion of GC cells. Mechanistic studies have revealed that targeting SREBP-1 by Pen-H significantly reduces de novo fatty acid synthesis and that the anti-GC efficacy of Pen-H is SREBP-1 dependent. Moreover, combination treatment with Pen-H and 5-fluorouracil (5-Fu) resulted in enhanced inhibition of cell proliferation and tumor growth compared to monotherapy.

Conclusion

Taken together, these findings highlight that SREBP-1 is an effective therapeutic target in GC and that Pen-H is a promising SREBP-1 inhibitor and a candidate for GC treatment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-07323-3.

Introduction

Gastric cancer (GC) is one of the most aggressive cancer types. According to the 2024 Cancer Statistics Report, GC ranks fifth in terms of incidence and mortality among all tumors [1]. Although some patients can benefit from surgery and chemotherapy, the overall survival rate remains low, especially in patients with advanced GC [2]. Hence, new therapeutic strategies and approaches are urgently needed to treat GC. Trastuzumab in combination with chemotherapy increases the median overall survival of patients with HER2-positive advanced GC. Nevertheless, the clinical application of this targeted therapy is considerably constrained by the relatively low prevalence of HER2-positive cases, which account for only approximately 10–20% of all GC patients [3]. This limitation underscores the need for further research to expand the therapeutic landscape to a broader patient population. Dysregulated cellular lipid metabolism is now widely recognized as a hallmark of cancer [4]. Various cancer types exhibit enhanced lipid uptake, storage, and lipogenesis, which collectively contribute to accelerated tumor growth and progression [5]. However, the complexity and diversity of lipid metabolic pathways, along with the multitude of molecular targets involved, present significant challenges in identifying core therapeutic targets. This remains a critical obstacle to the development of targeted strategies for cancer treatment based on the modulation of lipid metabolism.

Biomarker-based precision medicine is significantly constrained by the limited sensitivity and specificity of biomarkers in predicting drug responsiveness or resistance. Furthermore, despite extensive biomarker profiling, effective therapeutic targets are lacking in most patients. As an alternative to biomarker-driven drug prediction, functional drug screening has emerged as a promising strategy for advancing precision oncology [6]. The marine environment has an immense biodiversity and is a unique resource for drug discovery. Approximately 70% of FDA-approved anticancer drugs are derived from natural sources. Among these, natural marine products have attracted significant interest from researchers due to their diverse molecular structures [7, 8]. Based on a functional drug screening strategy targeting anticancer natural products, we identified a novel and potent anticancer lead compound, Penicolinate H (Pen-H), derived from the crinoid-associated fungus Penicillium brocae SYSU-CJ17. Pen-H selectively inhibited the growth and survival of GC cells and demonstrated significant efficacy in suppressing tumor growth and metastasis in preclinical models. Preliminary mechanistic investigations suggested that the anticancer effects of Pen-H may be mediated through the regulation of SREBP-dependent lipid metabolic reprogramming.

Sterol regulatory element-binding proteins (SREBPs), including SREBP-1 and SREBP-2, are transcription factors that play crucial roles in the maintenance of lipid homeostasis by regulating the expression of enzymes involved in the synthesis of endogenous cholesterol, fatty acids, triacylglycerols, and phospholipids [9]. SREBP-1 is involved in cholesterol and fatty acid metabolism [10], and SREBP-2 is exclusively involved in cholesterol metabolism [11]. Acetyl-CoA carboxylase (ACC) and fatty acid synthase (FASN) are the key enzymes in fatty acid synthesis, whereas 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) and squalene monooxygenase (SQLE) are the rate-limiting enzymes in cholesterol synthesis [12, 13]. Emerging evidence suggests that SREBPs and their target genes are activated and upregulated in several cancers [10, 14, 15], indicating that SREBPs are promising targets for anticancer therapies. However, the functional roles of SREBPs in GC remain unclear. Given the complexity of SREBP-regulated metabolic pathways and their numerous downstream targets, the specific isoform responsible for mediating the core metabolic pathways relevant to GC treatment is yet to be identified. Further in-depth investigations are crucial to clarify the therapeutic potential of SREBP isoforms as molecular targets and facilitate the development of targeted therapies for GC.

In this study, we found that SREBP-1 was highly expressed in GC tissues and was significantly associated with poor prognosis in patients with GC. Targeting SREBP-1 effectively suppressed GC proliferation and metastasis. Furthermore, we demonstrated that Pen-H, a novel SREBP-1 inhibitor, exhibited robust anti-tumor activity against GC, both in vitro and in vivo by inhibiting SREBP-1-mediated lipogenesis. Additionally, our findings revealed that Pen-H enhanced the sensitivity of GC cells to the first-line chemotherapeutic agent 5-fluorouracil. Collectively, our results establish SREBP-1 as a promising therapeutic target for GC and highlight Pen-H as a potent SREBP inhibitor with significant potential for GC treatment.

Materials and methods

Compounds and antibodies

Penicolinate H (purity > 98%) was produced by Penicillium brocae SYSU-CJ17, which was obtained from a crinoid collected from Xisha. Penicillium brocae SYSU-CJ17 was fermented at room temperature for 30 days using a rice medium containing 3% sea salt; then it was separated and purified using silica gel chromatography, gel column chromatography, and high-performance liquid chromatography (HPLC). Fifteen secondary metabolites, containing three new structural compounds, were identified. The chemical structures of the compounds were deduced using spectroscopic techniques (including high-resolution electrospray ionization mass spectrometry and 1D and 2D nuclear magnetic resonance spectroscopy) and electronic circular dichroism calculations (Fig. S1, S2, S3, Table S1). 5-Fu (purity > 98%) was supplied by GlpBio (Guangzhou, China). Antibodies against the following proteins were utilized, with their sources and dilution ratios specified as follows: GAPDH (1:1,000, Cell Signaling Technology, NO.2118S), β-actin (1:1,000, Cell Signaling Technology, NO.4970S), PARP-1 (1:1,000, Cell Signaling Technology, NO.9542S), c-caspase-7 (1:1,000, Cell Signaling Technology, NO.9491S), N-Cadherin (1:1,000, Cell Signaling Technology, NO.13116S), Snail (1:1,000, Cell Signaling Technology, NO.3879S), SREBP-1 (1:1,000, Proteintech Group; NO.14088–1-AP), SREBP-2 (1:1,000, Proteintech Group; NO.28212–1-AP), FASN (1:1,000, Proteintech Group; NO.10624–2-AP), ACC (1:1,000, Proteintech Group; NO.21923–1-AP), HMGCR (1:1,000, Proteintech Group; NO.13533–1-AP), SQLE (1:1,000, Proteintech Group; NO.12544–1-AP).

Cell culture

GES-1, AGS, MKN-45, HGC-27, SNU-1, and KATO III cells were obtained from the American Type Culture Collection (ATCC), and the human GC cell line MGC-803 was obtained from China Academia Sinica (Shanghai, PR China). All cells were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin and maintained at 37 °C in an atmosphere containing 5% CO₂.

Patients and specimens

All normal and GC tissues were provided by the Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, Guangdong Province, China. Specimens were obtained with informed consent and confirmed by pathological experts at the hospital. All patients were diagnosed with advanced gastric adenocarcinoma. The Clinical Research Ethics Committee of Sun Yat-sen University approved all clinical experiments (approval no. SYSKY-2023–763-01, granted on 2023–07-21).

Animal experiments

The Committee for Ethics of Animal Experimentation approved the animal experiments, which followed the guidelines for animal experiments of the Sun Yat-sen University (approval no. SYSU-IACUC-2023-000429, granted on 2023–03-14). Male BALB/c nude mice (four weeks old, 16–18 g) were obtained from the Experimental Animal Center of Sun Yat-sen University. To establish GC xenograft tumors, a suspension of 2 × 10⁶ MKN45 cells in 100 μL of PBS and Matrigel mixture (1:1) was subcutaneously injected into both flanks of the mice. When the gross tumor volume reached 50 mm³, the mice were randomly divided into specific groups. They were treated intraperitoneally (i.p.) with 100 μL of either vehicle (DMSO), Pen-H (5 mg/kg or 10 mg/kg; i.p.; once daily), or Pen-H (5 mg/kg; i.p.; once daily) in combination with 5-FU (10 mg/kg; i.p.; twice weekly). The tumor size and body weight were monitored twice per week. On the final day of the experiment, the mice were euthanized and their tumors were collected for further analysis. For orthotopic mouse xenografts, four-week-old male BALB/c nude mice were incised, 1 × 10⁶ MKN45 cells stably expressing luciferase were injected into the subserosal layer of the stomach with a needle, and the mice were randomized into two groups. They were treated intraperitoneally (i.p.) with 100 μL of either vehicle (DMSO) or Pen-H (5 mg/kg; i.p.; once daily). Tumors that had grown in the stomach wall and metastasized to the liver or other organs were monitored using bioluminescence imaging. The body weights of the mice were also monitored. At the end of the studies, blood samples were collected from mice via orbital venipuncture under isoflurane anesthesia. Approximately 1.0–1.5 ml of blood was obtained from each animal and transferred into pre-labeled 2 ml microtubes containing 10% EDTA as an anticoagulant. Samples were immediately processed for hematological and biochemical analyses. Mice were sacrificed and tumors were dissected and weighed. We used a power analysis based on preliminary data to ensure adequate statistical power (typically 80% or higher) and minimize type II errors. Calculations were performed using G*Power, with an assumed effect size derived from prior studies and a significance level of 0.05.

Cell viability assay

A total of 1000 cells in 100 μL of culture medium were seeded into 96-well plates. After 24 hours, compounds were serially diluted to various concentrations and added in 50 μL of culture medium per well, achieving final concentrations of 0–10 μM. After 96 h of incubation, the CCK-8 kit was added. The cells were then incubated at 37 °C for 150 min in a 5% CO₂ environment, and absorbance was measured at 450 nm. Data were expressed as percentages, with the control group serving as a 100% reference. IC50 values were determined using GraphPad Prism 7.0 software.

Colony formation assay

A total of 1000 cells in 2 mL of culture medium were seeded into 6-well plates. After 24 h, the compounds were added at the indicated concentrations. The culture medium was periodically changed, and the compounds were added to the specified concentrations. The culture was terminated when visible colonies formed. After discarding the supernatant and washing with phosphate buffered saline (PBS), the cells were fixed with 4% paraformaldehyde for 15 min. The fixative was then removed and crystal violet staining was performed for 30 min. The staining solution was washed with running water, and the dish was air-dried.

Western blot analysis

The cell culture dishes were washed on ice with ice-cold PBS and treated with ice-cold lysis buffer for approximately 60 min. The cells were scraped off, transferred to microcentrifuge tubes, and centrifuged at 20,000 g for 15 min. The protein concentration in each cell lysate was determined using the BCA protein assay. Subsequently, 30 μg of protein from each lysate was loaded onto a polyacrylamide gel, separated by SDS-PAGE for 100 min, and transferred to PVDF membranes for 150 min. The membranes were incubated in blocking solution for 60 min at room temperature, rinsed in 1X TBST for 5 min with constant rocking, and incubated with primary antibody solution overnight at 4 °C with gentle rocking. After three rinses with 1×TBST, the membranes were incubated with the appropriately diluted secondary antibody for 60 min at room temperature. Visualization was performed using a ChemiDoc Molecular Imager System.

Patient-derived organoid (PDO) culture

Tumor samples were obtained from a patient diagnosed with gastric adenocarcinoma. The patient characteristics were as follows: male, 42 years old, Asian, primary gastric cancer tumor, classified as AJCC stage IB with a histological grade of 3, derived from a surgical resection specimen. Organoids were cultured from freshly dissected tumors of PDX xenografts when the tumors reached approximately 500 mm³. The tumors were finely minced and placed in a 50 mL conical tube containing a digestion mix of serum-free DMEM/F-12 medium and 1 mg/mL collagenase IV (Sigma), then incubated for 1 h at 37 °C. The isolated organoids were mixed with 50 μL of Matrigel and seeded into 96-well plates. The culture medium comprised nicotinamide (5 mM), N-acetylcysteine (1.25 mM), SB202190 (500 nM), EGF (100 ng/mL), neuregulin 1 (5 nM), Hepes (10 mM), FGF 10 (20 ng/mL), A83-01 (500 nM), Y-27632 (5 mM), B27 supplement (1×), FGF 7 (5 ng/mL), penicillin/streptomycin/glutamine (100 mg/mL), R-Spondin 3 (250 ng/mL), and phenol red-free DMEM/F-12 with primocin (50 mg/mL). Supplemented culture medium (1 mL) was added to each well and the organoids were maintained in a humidified atmosphere at 37 °C with 5% CO₂. After one week, the PDOs were treated with DMSO or Pen-H for an additional three days. Representative images were captured using a fluorescence microscope, and cell viability was assessed using Cell Titer-Glo.

Migration and invasion assay

The cells were evenly cultured in 6-well plates until they reached 90% confluence. A pipette tip was used to create a straight scratch (wound) on the cell monolayer, and the plate was photographed under a microscope. Debris was removed, fresh complete growth medium was added to each well, and Pen-H was added at the indicated concentrations. After an additional 24 h, the plate was photographed again to observe changes in scratch healing. The percentage wound closure was calculated by comparing the wound area at each time point with that at baseline.

Cell invasion assays were performed using 24-well transwell plates. Before adding the cells, serum-free medium was placed in Matrigel-coated polycarbonate membrane transwell chambers and incubated for 1 h. Cells were harvested from the culture using trypsin-ethylenediaminetetraacetic acid and resuspended in serum-free medium. A suspension of 2 × 10⁴ cells in 300 μL of serum-free medium was placed in the upper chamber of each transwell insert, and 500 μL of complete growth medium (with serum) to the lower well of the 24-well plate. The cells that invaded the lower surface of the membrane were fixed with 4% paraformaldehyde and stained with crystal violet. Three random fields were imaged under a microscope and the cells were counted for statistical analysis.

Caspase-3/7 activity assay

A total of 5 × 10⁴ cells in 2 mL of culture medium were seeded into 6-well plates and treated with a specific concentration of Pen-H. After 72 h, the cells were scraped off, transferred to microcentrifuge tubes, and centrifuged at 5,000 rpm for 5 min. The pelleted cells were resuspended in 180 μL of cold PBS. Then, 25 μL of the cell suspension and 25 μL of caspase-3/7 working solution were added to each well of a 96-well white plate, followed by gentle rocking. The fluorescence was measured using a luminescence detector. The protein concentration in each cell lysate was determined using the BCA protein assay. The final caspase-3/7 activity was calculated as the ratio of caspase-3/7 fluorescence to protein absorbance.

Total RNA-sequence

MKN-45 and HGC-27 cells were treated with vehicle (DMSO) or Pen-H (5 μM) for 48 h prior to RNA extraction. Transcriptomic sequencing was performed by BGI Tech (China) using a MGISEQ2000 SE50 machine to validate the sequence libraries. Gene Set Enrichment Analysis (GSEA v.3.0) was used to rank genes based on shrunken limma log2 fold changes. Genes with expression changes of ≥ 1.5-fold (increase or decrease) were clustered using the k-means clustering algorithm in Cluster 63. Enrichment analysis of GO and Reactome pathways for these genes was conducted with GSEA v.4.1.0, considering p adjust (FDR) < 0.05 as statistically significant. Clusters were displayed using the R statistical package (http://www.rproject.org/).

Quantitative real-time PCR

Total cellular RNA was extracted using the TRIzol reagent, followed by chloroform, isopropanol, and ethanol. RNA was reverse-transcribed into cDNA. The cDNA was combined with primers and SYBR Green (11201ES08*, YEASEN, Shanghai, China) according to the manufacturer’s instructions and detected using a BIO-RAD CFX96 (Bio-Rad). After the fluorescence values were collected, melting curve analysis was performed. The primer sequences used for the qRT-PCR assays are listed in Table S2.

Small interfering RNA (siRNA) transfection

For siRNA knockdown, cells were transfected with either control or SREBF1 siRNA using DharmaFECT and Opti-MEM following the manufacturer’s instructions. The control siRNAs were non-specific oligonucleotides without complementarity to any human gene. Additionally, two different siRNA sequences specifically targeting SREBF1 were used to rule out sequence-specific effects, because similar results were obtained for both sequences. The siRNA sequences are listed in Table S3. All the siRNAs were synthesized by Sangon Biotech (Shanghai, China).

Cellular thermal shift assay (CETSA)

HGC-27 cells were cultured in a 15-cm dish until they reached the desired confluence. The cells were then treated with either vehicle or 5 μM Pen-H. After a 3-hour incubation at 37 °C, the cells were collected and resuspended in 1 mL of PBS containing lysis buffer along with the vehicle or Pen-H to maintain the initial concentration. Equal amounts of cell lysate were aliquoted into multiple microcentrifuge tubes and heated to a set temperature for 3 min using a thermal cycler. The suspension was subjected to three freeze-thaw cycles using liquid nitrogen to lyse the cells. The tubes were centrifuged at 20,000 g for 20 min at 4 °C to separate soluble proteins from aggregates. The protein concentration in each supernatant was measured using a protein assay. Proteins were detected by western blot analysis, and their intensities were quantified using ImageJ software.

Molecular docking

Molecular docking was performed using AutoDock Vina software. The ChemDraw software was used to determine the structure of the compound. MM2 calculations were conducted to minimize conformational energy. The crystal structures of SREBP1 were retrieved from the UniProt database. PYMOL 2.3.4 was used to perform dehydration and ligand removal on the receptor protein, and AutoDockTools was applied to make modifications, such as hydrogenation and charge balancing. Docking studies were performed using AutoDock Vina, and PYMOL was used to visualize the docking results.

Surface plasmon resonance (SPR) analysis

The binding affinity of Pen-H for SREBP-1 was evaluated using a Biacore 8 K instrument (GE Healthcare, Uppsala, Sweden). Recombinant SREBP-1 was immobilized on a CM5 sensor chip following standard amine-coupling instructions. PBS (G0002, pH 7.2–7.4; Servicebio, Wuhan, China) with 5% DMSO was used as the running buffer. Various concentrations of Pen-H were injected as analytes. The binding assay was conducted at 25 °C with a flow rate of 30 μL/min using PBS buffer. The affinity constants were determined using a 1:1 Langmuir binding model and BIA evaluation software.

Measurement of cellular neutral lipids

A total of 10⁶ cells diluted in 2 mL of medium were grown in 6-well plates and treated with indicated concentrations of Pen-H, followed by incubation at 37 °C. After 4 d, the cells were rinsed with PBS and fixed with 4% warm paraformaldehyde for 15 min. To observe intracellular lipid accumulation, the cells on 6-well glass slides were incubated with the Lipid Droplets Red Fluorescence Assay Kit with Nile Red (C2051S, Beyotime) for 20 min at 37 °C. Fluorescently labeled lipid droplets were observed under a confocal microscope (Nikon, Tokyo, Japan).

Determination of cellular free fatty acids

Cellular free fatty acid levels were measured using the Amplex Red Free Fatty Acid Assay Kit (S0215S, Beyotime). Cells were homogenized in 200 μL of BeyoLysis™ Buffer A for Metabolic Assay, the homogenate was then centrifuged at approximately 12,000×g for 5 min at 4 °C, and the supernatant was collected for subsequent analysis. Then, 50 μL free fatty acid assay reaction buffer was added into each well and incubate at 37 °C for 30 min, and absorbance was measured at 570 nm.

Immunohistochemistry (IHC)

Tumors from a mouse model of subcutaneous GC were fixed in formalin and embedded in paraffin. Tissue blocks were dewaxed, rehydrated, and treated to block endogenous peroxidase activity. Five-micron-thick tumor sections were incubated with specific antibodies overnight at 4 °C. After washing with PBS, the slides were incubated with secondary antibody for 30 min. The sections were then stained with DAB and hematoxylin (both from Servicebio), dehydrated, and sealed with neutral resin. Images were captured using a Nikon microscope.

Bioinformatics analysis

Single-cell RNA-seq data (GSE183904) of GC were downloaded from the Gene Expression Omnibus (GEO) database. SREBF1 mRNA levels in GC samples were obtained from the TCGA database. Kaplan-Meier survival data for GC (GSE26253) were downloaded from the GEO database. Drug sensitivity and expression correlation data were downloaded from the Genomics of Drug Sensitivity in Cancer (GDSC) and Cancer Therapeutics Response Portal (CTRP) databases. Bioinformatics analysis was conducted using R (v.3.4) in the edgeR package. Volcano plot analysis was conducted using Sangerbox 3.0,57, an online bioinformatics analysis platform.

Statistics

All experiments were performed in at least three biological replicates (independent experiments) and, where applicable, technical replicates (triplicate wells for cell-based assays). The experimental data were analyzed using GraphPad Prism 7. In vitro results are shown as the mean ± SD from three independent experiments, whereas in vivo results are presented as the mean ± SEM. Statistical significance was assessed using a two-tailed Student’s t-test. Statistical significance was set at p < 0.05.

Results

Identifying bioactive penicolinates from the crinoid-derived fungus penicillium brocae SYSU-CJ17

In our continued search for bioactive natural products from crinoid-derived fungi, we explored the chemical composition of the metabolites produced by the crinoid-derived fungus Penicillium brocae SYSU-CJ17. This investigation led to the discovery of three new structural compounds (1, 3, and 4) and 12 known compounds (Fig. 1A). The chemical structures of the compounds were deduced using spectroscopic techniques (including high-resolution electrospray ionization mass spectrometry and 1D and 2D nuclear magnetic resonance spectroscopy) (Figs. S1, S2, S3, Table S1). Compounds 1 − 15 were tested for their growth-inhibitory effects in a panel of GC cell lines. Among them, compound 3 (Pen-H) showed excellent activity (Fig. S4A-C). Therefore, Pen-H was selected as the lead compound for further investigation of its function in GC. Pen-H was isolated as yellow powder, the molecular formula of which C₂₆H₃₆N₂O₆ was deduced from its HRESIMS spectrum at m/z 473.2650 [M + H] ⁺ (calcd for C₂₆H₃₆N₂O₆, 473.2646). Comprehensive comparison of the 1D NMR data of Pen-H with those of the known compound penicolinate A (Table S1) suggested a close structural relationship between the two compounds. A notable difference was the replacement of the methoxy group signal with three sp³ non-protonated carbon signals (68.15, 71.1, and 63.88 ppm). In addition, ¹H-¹H correlation spectroscopy (COSY) correlations between H-18, H-19, and H-20 were observed. Hence, these results indicate the presence of a glycerol fragment. Finally, the HMBC correlations from δ_H 4.33(2 H, dd, J = 11.3 Hz) to δ_C 165.96 (C-7’, C) revealed the planar structure of Pen-H. Owing to the existence of a long flexible chain, it is difficult to verify an accurate configuration using ECD experiments. According to the research conducted by Talontsi [16], the chirality of compounds with a single glycerol fragment can be determined using optical rotation. The optical rotation value of Pen-H is  = −4.5 (c 0.001, MeOH), indicating that the configuration of Pen-H is 19S (Fig. 1B).

Fig. 1.

Structurally diverse penicolinates from crinoid epipsymbiosis fungus penicillium exert anti-GC effect. (A) The structures of bioactive penicolinates from crinoid epipsymbiosis fungus penicillium specie penicilliumbrocae SYSU-CJ17. (B) Chemical structure of pen-H. (C) Cell viability curve for pen-H in a panel of GC cell lines and nonmalignant GES-1 cells in indicated concentrations for 96 h. Cell viability was measured with CCK-8 reagents. The data were shown as percentage and vehicle-treated cells set at 100. (D) Total number of viable cells was counted after treatment of indicated GC cell lines with vehicle or pen-H for 96 h. (E) Left: effect of pen-H on the colony formation of GC cells. Colonies were stained with crystal violet. Right: analysis of the results by using image J software. (F) Influence of pen-H on expression levels of apoptosis-related proteins in indicated GC cell lines. Immunoblotting analysis was carried out in cells treated with vehicle or indicated concentrations of pen-H for 48 h. (G) Pen-H inhibited patient-derived xenograft (PDX)-derived organoid growth. PDX-derived organoids were treated with vehicle (DMSO) or indicated concentrations of pen-H. Four days later, representative images were captured by a fluorescence microscope (top three rows) or a standard light microscope (bottom row). Scale bar represents 200 µM. (H) GC cells were treated with vehicle (DMSO) or indicated concentrations of pen-H for 24 h. Wound healing assays were performed to detect cell migration. (I) The invasion ability of GC cells treated with vehicle (DMSO) or indicated concentrations of pen-H were evaluated by transwell assay. (J) Immunoblotting showing the expression of EMT related proteins in indicated GC cells after treatment with vehicle (DMSO) or indicated concentrations of pen-H for 48 h. Data were shown as the mean ± SD. Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3

Pen-H inhibits GC cells proliferation, migration and invasion

Pen-H is a novel picolinic acid derivative. The structural features of penicolinates endow them with good biological activities, such as antimalarial, antituberculosis, antibacterial, and cytotoxic activities [17]. In our study, we aimed to investigate inhibitory effects of Pen-H on a panel of GC lines, including both tumor and normal cells. The results showed that Pen-H displayed strong viability inhibition on GC cell lines in a concentration-dependent manner, while having minimal effect on the gastric mucosa cell line GES-1 (Fig. 1C, Table S4). Additionally, Pen-H inhibited proliferative ability and clone formation capacity in GC cells (Fig. 1D, E, Fig. S5B, C) and induced the expression of apoptosis-related proteins, such as cleaved PARP-1 and cleaved caspase-7, in GC cell lines (Fig. 1F, Fig. S5A, D, H). Further examination using 3D organoids demonstrated that Pen-H effectively inhibits the growth and survival of GC patient-derived tumor organoids (PDO) (Fig. 1G).

Given that metastasis is an indicator of poor prognosis in advanced GC, we examined the effect of Pen-H on GC metastasis. Wound healing and transwell assays showed that Pen-H significantly inhibited GC cell migration and invasion (Fig. 1H, I; Fig. S5E, F). Additionally, western blot analysis revealed that Pen-H decreased the expression of EMT-related proteins such as N-cadherin and Snail in GC cells (Fig. 1J, Fig. S5G). Collectively, these results suggest that Pen-H inhibits the proliferation, migration, and invasion of GC cells.

Pen-H triggers apoptosis in GC through blockade of the SREBP metabolic pathway

To identify the core transcriptional network regulated by Pen-H in GC cells, we performed RNA-seq analysis of MKN-45 and HGC-27 cells treated with Pen-H. A Venn diagram illustrating the common differentially expressed genes (DEGs) revealed that 112 genes were upregulated and 268 genes were downregulated by up to 1.5-fold (Fig. 2A). Gene Ontology (GO) enrichment analysis of these altered transcripts revealed that the apoptotic pathway was upregulated and the lipid metabolic process was downregulated after Pen-H treatment (Fig. 2B). Analysis of transcripts commonly downregulated by Pen-H using the Reactome database revealed that genes involved in lipid metabolism regulated by SREBF were among the most highly enriched (Fig. 2C). Genes involved in lipid metabolism were strongly suppressed by Pen-H treatment, whereas those related to apoptosis were upregulated (Fig. 2D). Further analysis using GSEA indicated that the lipid metabolic processes were the most downregulated pathways affected by Pen-H (Fig. 2E). Furthermore, compared with GES-1 cells, the activated form of SREBPs was significantly increased in MKN-45 and HGC-27 cell lines (Fig. 2F).

Fig. 2.

Pen-H triggers apoptosis in GC cells through blockade of the SREBP metabolic pathway. (A) Venn diagram of the number of genes with expression significantly (1.5-fold) upregulated or downregulated, which is detected by RNA-seq of MKN-45 and HGC-27 cells treated for 48 h with 5 μM pen-H. (B) GSEA of top enriched gene sets in MKN-45 and HGC-27 cells treated with pen-H. The upregulated and downregulated gene sets from GO platforms were output by GSEA. (C) According to the RNA-seq sequencing results of MKN-45 and HGC-27 cells cultured for 48 h under 5 μM pen-H conditions, the significantly downregulated signaling pathways were analyzed by Reactome databases and displayed in the form of bubble map. (D) Heatmap and hierarchical clustering displaying the fold changes of gene expression detected by RNA-seq in MKN-45 and HGC-27 cells treated with pen-H (5 μM) for 48 h, compared to vehicle (DMSO). Correctly display genes with log2 > 0.585 expression changes under at least one condition. Genes are displayed in rows, and the standardized count of each sample is displayed in columns. Red indicates upregulated, and blue indicates down-regulated expression level. Lipid metabolic process and apoptotic process signature genes that were altered in expression are displayed. (E) GSEA of the lipid metabolism signatures in HGC-27 cells treated with pen-H (5 μM). (F) Western blot for SREBP-1 and SREBP-2 expression in human GC cells and gastric normal cells. (G, H) Volcano plot reflecting lipid metabolic gene expression alterations in MKN-45 and HGC-27 cells after pen-H (5 μM) treatment. (I, J) qRT-PCR analysis of the indicated genes in MKN-45 and HGC-27 cells treated with DMSO or with pen-H (5 μM) for 48 h. (K) Immunoblotting of SREBP-1 and important enzymes (ACC1 and FASN) involved in fatty-acid synthesis pathway in GC cells treated with indicated concentrations of pen-H for 48 h. (L) Immunoblotting of SREBP-2 and rate-limiting enzymes (HMGCR and SQLE) involved in cholesterol-biosynthesis pathway in GC cells treated with indicated concentrations of pen-H for 48 h. Data were shown as the mean ± SD. Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3

Considering that lipid metabolism is crucial for oncogenesis and progression of GC, we speculated that Pen-H might attenuate GC development by inhibiting lipid metabolism. The volcano plot reflected a decrease in lipid metabolic gene expression following Pen-H treatment (Fig. 2G, H). To validate the RNA-Seq results, we performed qRT-PCR and western blotting on Pen-H-treated MKN-45 and HGC-27 cells. The results showed a marked decrease in the expression of genes involved in fatty acid synthesis and cholesterol metabolism, including the rate-limiting enzymes FASN, ACC, HMGCR, and SQLE (Fig. 2I–L). Collectively, these results indicate that Pen-H triggers apoptosis in GC by blocking the SREBP pathway and regulating the expression of genes involved in lipid metabolism.

Inhibitory effect of pen-H on GC cells is fatty acids-dependent

Considering the inhibitory effect of Pen-H on the SREBP pathway, which is involved in lipid metabolism, we first examined its effects on the cellular lipid pool to determine whether Pen-H affects lipid metabolism in GC cells. Impaired intracellular lipid accumulation was confirmed by confocal microscopy, showing that Pen-H treatment led to a significant reduction in the number of lipid droplets in both MKN-45 and HGC-27 cells, as determined by Nile red staining (Fig. 3A, B). These results indicated that Pen-H may exert an inhibitory effect on GC cells by suppressing SREBP-mediated lipid metabolism. As lipid metabolism includes fatty acid and cholesterol metabolism, we investigated the role of fatty acids and cholesterol in the inhibitory effect of Pen-H on GC cells.

Fig. 3.

Lipid metabolic remodeling by pen-H in GC cells is mediated via SREBP-1 induction. (A, B) GC cells were treated with indicated concentrations of pen-H for 24 h, followed by labeling of lipid droplets with Nile red. Representative images of red fluorescence for pen-H-treated and control cells were shown. Scale bar, 20 μm. (C-F) Exogenous fatty-acid supply rather than cholesterol rescued GC cell death caused by treatment with pen-H for 96 h. n = 3. (G) Exogenous fatty-acid supply rather than cholesterol rescued GC cell clonogenicity inhibition caused by treatment with pen-H (left). GC cell clonogenicity were quantified by using image J software (right). (H) GC cells were treated with indicated concentrations of pen-H for 24 h. Cellular free fatty acid levels were measured using amplex Red free fatty acid assay kit. (I) The degree of siRNA knockdown efficiency for SREBF1 is quantified by western blot. (J) HGC27 cells were transfected with siSREBF-1 or control siRNA for 48 h and treated with vehicle or pen-H for another 24 h. Cells were harvested for determining cell growth by counting viable cells. (K) The degree of siRNA knockdown efficiency for SREBF2 is quantified by western blot. (L) HGC27 cells were transfected with siSREBF-2 or control siRNA for 48 h and treated with vehicle or pen-H for another 24 h. Cells were harvested for determining cell growth by counting viable cells. (M) Left: melt curves of SREBP-1 in HGC27 cells after DMSO or pen-H (5 μM) treatment by CETSA. Right: SREBP-1 protein levels were quantified by using image J software. (N) Surface plasmon resonance (SPR) analysis of binding affinity of pen-H to SREBP-1. The K_d value was calculated based on the fitted curves. (O) The predicted 2D and 3D binding mode between pen-H with SREBP-1 according to docking simulation using AutoDock Vina. Data were shown as the mean ± SD. Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3

GC cells were treated with Pen-H in serum-free media with or without supplementation of palmitic acid or cholesterol. Quantitative lipid analysis showed that exogenous lipid supplement (fatty acid or cholesterol) effectively increased intracellular given lipid levels in Pen-H-treated cells, confirming intact uptake pathways (Fig. 3C, E). Fatty acid, but not cholesterol, supplementation significantly rescued Pen-H-induced inhibition of cell viability and colony formation (Fig. 3D, F, G). Meanwhile, Pen-H could inhibit the level of cellular free fatty acid in a dose dependent manner (Fig. 3H). These data indicate that cholesterol is not a limiting factor for GC cell growth under these experimental conditions. Instead, Pen-H primarily exerts its anti-tumor effects by disrupting fatty acid metabolism.

Identification of pen-H as a potent SREBP-1 inhibitor

Transcriptome analysis revealed that genes involved in lipid metabolism regulated by SREBF were among the most highly enriched genes, and further research indicated that the inhibition of Pen-H in GC cells was fatty acid-dependent. Given that SREBP-1 preferentially activates genes involved in fatty acid synthesis and Pen-H suppresses SREBP-1-mediated lipogenesis, SREBP-1 may be a potential target of Pen-H. To investigate whether Pen-H reduced cell viability by inhibiting SREBP-1 activity, we used SREBF1 siRNA to specifically silence SREBP-1 expression. The results demonstrated that the inhibitory effect of Pen-H on cell survival was markedly attenuated in SREBF1 siRNA-treated cells compared with that in siControl-treated cells (Fig. 3I, J). However, when the same knockdown strategy was applied to SREBF2 (Fig. 3K, L), the inhibitory effect of Pen-H on GC cells remained unaffected. These findings indicate that the anti-cancer activity of Pen-H is specifically dependent on SREBP-1 but not on SREBP-2, further supporting the role of SREBP-1 as a critical mediator of the effects of Pen-H.

To confirm SREBP-1 target engagement by Pen-H in cells, we performed a CETSA assay. The results indicated that SREBP-1 protein denatured at 41 and 43 °C, but its stability was significantly enhanced in the presence of Pen-H (Fig. 3M). The binding affinity of Pen-H to SREBP-1 was assessed using an SPR assay, which confirmed a high affinity between Pen-H and SREBP-1 (Fig. 3N). To further investigate the interaction between Pen-H and SREBP-1, molecular docking simulations were performed using AutoDock Vina software (version 1.1.2). Docking analysis was performed between the receptor protein and the ligand, and the results were analyzed using Discovery Studio. The binding analysis revealed that Pen-H forms hydrogen bonds with the residue TRP784 and exhibits hydrophobic interactions with residues ALA791, LEU790, PRO853, TYR855, and PRO783, with a Glide SP score of −5.4 kcal/mol (Fig. 3O). The hydrogen bond distances were also calculated (Table S5).

Taken together, these results imply that dysregulated SREBP-1 mediated fatty acid metabolism is responsible, at least in part, for the induction of cell death in GC.

SREBP-1 is overexpressed in GC tumors and predicts poor prognosis

Our findings demonstrated that Pen-H exerts antitumor effects by modulating SREBP-1-mediated fatty acid metabolism, indicating that SREBP-1 may serve as a promising therapeutic target for GC. To evaluate the potential of SREBP-1 as viable therapeutic target, further investigation of its functional role in GC progression is warranted.

To investigate the potential functions of SREBP-1 (encoded by SREBF-1) in GC progression, we queried the SREBF-1 mediated de novo fatty acid synthesis pathway gene expression datasets from human benign and primary GC samples. Analysis of TCGA and GTEx databases revealed that de novo fatty acid synthesis pathway was significantly elevated in GC tissues compared to normal gastric tissues (Fig. 4A), whereas de novo cholesterol synthesis pathway was significantly decreased (Fig. S6A). To better define the expression characteristics of SREBP-1 in patients with GC, we performed a single-cell transcriptome analysis (GSE183904). Because GC originate from gastric epithelial cells, we classified all single cells into distinct clusters, including epithelial cells, based on their genomic characteristics (Fig. 4B, Fig. S6B). Among the various cell types, SREBP-1 was highly expressed in epithelial cells (Fig. 4C, D). We found that SREBP-1 was markedly upregulated in tumors compared to normal tissues, which was primarily attributed to epithelial cells (Fig. 4E–G). In addition, SREBF-1 transcript levels were significantly higher in gastric tumor tissues than in normal tissues in TCGA database (Fig. 4H). This finding was supported by western blot analysis of eight pairs of GC specimens and their corresponding non-tumor specimens, which showed higher SREBP-1 expression in tumor tissues (Fig. 4I, J). Moreover, Kaplan–Meier survival analysis of published tumor datasets (GSE26253, GSE84426) showed that higher tumor SREBF-1 mRNA levels were significantly associated with poor survival in patients with GC (Fig. 4K, Fig. S6C). Taken together, these results suggest that SREBP-1 is overexpressed in GC and is associated with tumor progression.

Fig. 4.

SREBP-1 is highly expressed in GC tissues and predicts poor prognosis. (A) GSVA scores of SREBF-1 mediated fatty-acid biosynthesis pathway genes compared with GC tissues and non-diseased gastric tissue. (B–G) Single-cell sequencing analyses of gastric cancer ( > 200,000 cells) comprising 48 samples from 31 patients across clinical stages and histologic subtypes (GSE183904). The single cells were classified into distinct clusters according to their genomic characteristics (B). The mRNA levels of SREBF1 in various cell types in all tissue samples (C, D). The mRNA levels of SREBF1 in various cell types in normal and tumor samples (E). The expression level of SREBF1 in epithelial cells in tumors was higher than that in normal tissues (F, G). (H) Gene expression of SREBF1 in human GC tissue and matched normal tissue was analyzed in the TCGA database using log2 (TPM +1) for log scale. (I) Western blot analysis showing the expression of SREBP-1 in GC tissue and adjacent normal tissue. (J) Gray values of the blots after normalization to GAPDH were statistically analyzed. (K) Kaplan-meier survival plots of SREBF1. GC patients were grouped by the trichotomization of the SREBF1 expression (Q1 vs. Q4, lower quartile vs. upper quartile, GSE26253). All data were shown as the mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001; n = 3, Student’s t test

SREBP-1 inhibition attenuates the proliferation, migration, and invasion of GC cells by suppressing lipogenesis

To further validate the effect of SREBP-1 on the survival of GC cells, we used siRNA to silence SREBF-1 expression in the GC cell lines, MKN45 and HGC27. We investigated the effects of SREBP-1 knockdown on GC cell proliferation, migration, and invasion to confirm SREBP-1 is a potential therapeutic target for GC. Silencing SREBF-1 with specific siRNAs significantly inhibited cell proliferation (Fig. 5A) and colony formation (Fig. 5B). However, SREBF-2 knockdown did not inhibit the growth of GC cells (Fig. S7A). SREBF-1 silencing also increased the expression of apoptosis-related proteins such as cleaved PARP-1 and cleaved caspase-7 (Fig. 5C). Given that metastasis is the leading cause of death in patients with GC, we assessed the role of SREBP-1 in modulating GC metastasis. Wound healing and transwell assays indicated that siSREBF-1 inhibited GC cell migration and invasion (Fig. 5D, E; Fig. S7B, C). Moreover, SREBP-1 knockdown decreased the expression of EMT-associated proteins, such as N-cadherin and Snail (Fig. 5F). Additionally, siSREBF-1 reduced lipid droplet formation (Fig. 5G), fatty acid synthesis (Fig. 5H), and expression of key enzymes involved in fatty acid synthesis in GC cells (Fig. 5I).

Fig. 5.

SREBP-1 inhibition attenuates the proliferation, migration, and invasion of GC cells by suppressing lipogenesis. (A) GC cells were transfected with two different SREBF1 siRnas against SREBF1. Three and five days later, viable cell numbers were counted. (B) GC cell lines were infected as in (A), fourteen days later, colonies were counted. Left: effect of siSREBF1 on the colony formation of GC cells. Colonies were stained with crystal violet. Right: analysis of the results by using image J software. (C) Immunobloting analysis of apoptosis-related protein in GC cells transfected with siSREBF1 or control siRNA and incubated for 48 h. Representative blot, n = 3. (D) Wound healing assays assay of GC cells after SREBP-1 knockdown to investigate the effects of cell migration. (E) Transwell assay of GC cells after SREBP-1 knockdown to investigate the effects of cell invasion. (F) Immunoblotting analysis of EMT related proteins in indicated GC cells after transfected with siSREBF1 or control siRNA and incubated for 48 h. Representative blot, n = 3. (G) GC cells were transfected with siSREBF1 or control siRNA and incubated for 48 h, followed by labeling of lipid droplets with Nile red. Representative images of red fluorescence for siSREBF1-treated and control cells were shown. Scale bar, 20 μm. (H) GC cells were transfected with siSREBF1 or control siRNA and incubated for 48 h, cellular free fatty acid levels were measured using amplex Red free fatty acid assay kit. (I) Immunobloting analysis of important proteinss (ACC1 and FASN) involved in fatty-acid synthesis pathway in GC cells transfected with siSREBF1 or control siRNA and incubated for 48 h. Representative blot, n = 3. (J) Chemical structure of Fatostatin. (K) Cell viability curves of MKN-45 and HGC-27. CCK-8 analysis was performed after treatment of MKN-45 and HGC-27 cells with different concentrations of Fatostatin for 96 h. (L) Total number of viable cells was counted after treatment of MKN-45 and HGC-27 cells with vehicle or Fatostatin for 96 h. (M) Colony formation was counted after treatment of MKN-45 and HGC-27 cells with vehicle or different concentrations of Fatostatin. (N) Western blotting of apoptosis-related proteins after treatment of MKN-45 and HGC-27 cells with vehicle or different concentrations of Fatostatin for 48 h. Data were shown as the mean ± SD. Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3

Fatostatin, a specific inhibitor of SREBPs that controls the expression of multiple key enzymes in the lipid and cholesterol synthesis pathways (Fig. 5J), inhibits SREBP maturation by blocking the translocation of SCAP from the endoplasmic reticulum to the Golgi apparatus [18]. In parallel, we evaluated the effects of the SREBP inhibitor Fatostatin on GC cell lines. Consistent with the outcomes observed following SREBP-1 knockdown, fatostatin significantly inhibited cell viability in a concentration-dependent manner (Fig. 5K). Furthermore, Fatostatin suppressed cell proliferation (Fig. 5L) and colony formation capacity (Fig. 5M), while upregulating the expression of apoptosis-related proteins, including cleaved PARP-1 and cleaved caspase-7 (Fig. 5N), in GC cell lines.

Collectively, these findings suggest that SREBP-1 inhibition suppresses GC cell survival and metastasis through fatty acid metabolic reprogramming and that SREBP-1 is a promising target for GC treatment.

Pen-H substantially suppresses GC tumor growth and metastasis in vivo

Having confirmed that Pen-H inhibited GC cell proliferation, migration, and invasion by suppressing SREBP-1 mediated lipogenesis in vitro, we assessed its effects on GC tumor growth in vivo. We used the MKN-45 subcutaneous xenograft model in BALB/c-nu/nu male mice that were intraperitoneally administered either vehicle or Pen-H (Fig. 6A). The tumor volume and body weight were monitored every other day. Fifteen days after Pen-H administration, the tumors were excised and weighed. The results demonstrated that Pen-H significantly suppressed tumor growth in a dose-dependent manner compared to that in the vehicle group (Fig. 6B–D). Additionally, Pen-H administration showed no obvious toxic side effects in mice, as indicated by a stable body weight and no significant pathological changes in the lungs, heart, liver, spleen, or kidney (Fig. S8A-C). Hematological and biochemical evaluations indicated that Pen-H had no significant effect on the well-being of the mice, considering blood, liver, and renal evaluations (Table S6, S7). H&E staining revealed a distinct enlargement of the intercellular spaces and tumor cell necrosis, whereas immunohistochemical Ki67 staining showed that Pen-H strongly inhibited cancer cell proliferation (Fig. 6E). Western blot analysis of xenograft tumors indicated that Pen-H significantly decreased the expression of SREBP-1, ACC1, and FASN compared to that in the control group (Fig. 6F). Nile red staining of tumor tissues confirmed that Pen-H inhibition of SREBP-1 resulted in impaired cellular fatty acid pools (Fig. 6G).

Fig. 6.

Pen-H substantially suppresses GC tumor growth and metastasis in vivo. (A) Schematics illustrating MKN45 xenograft tumor establishment and treatment. Male BALB/c nude mice were implanted with MKN45 cells and intraperitoneally administered with 5 mg/kg or 10 mg/kg of pen-H once daily. (B) Subcutaneous tumor volume was measured every three days. (C) Tumors were harvested and taken pictures by using a digital camera. (D) Tumor weight was measured at the last day. (E) H&E staining and immunohistochemical analysis of tumor tissues. Tumor tissues were fixed, dehydrated, and embedded into paraffin blocks. Tissue sections were cut from paraffin blocks, stained with hematoxylin and eosin (H&E), or immunolabeled to detect the expression of ki-67. Scale bar, 100 µm. (F) Immunoblotting analysis of SREBP1, ACC1 and FASN in tumor tissues. (G) Representative nile red staining of tumor tissue. (H) MKN45 cells stable expressing pLenti-firefly luciferase-EGFP were injected into the stomach wall of BALB/c nude mice, and randomly divided into indicated groups. GC progression were monitored by bioluminescence at 14th or 21st day. (I) Representative tumor images of orthotopic xenograft mice model. Circles indicate the primary gastric cancer lesions in situ, while arrows highlight metastatic lymph nodes. (J) Bioluminescent images of in vivo tumor growth status (left) and representative photos of tumors in stomach and liver sites on the last day (right). (K) H&E staining and immunohistochemical analysis of gastric carcinoma. Scale bars, 200 mm. Data were shown as mean tumor volume ± standard error of the mean (SEM). * p < 0.05, ** p < 0.01, *** p < 0.001

Metastasis is a major cause of GC-associated death, and the liver is the most common site of metastasis. To further evaluate the effect of Pen-H on gastric tumor progression and metastasis in vivo, we developed an orthotopic GC model by implanting MKN45-luciferase cells into the gastric serosa of immunodeficient mice. After confirming successful model establishment, mice were randomly divided into vehicle- and Pen-H-treated groups, and tumor growth was monitored longitudinally using bioluminescence imaging. On Day 14, distant metastasis had not yet occurred in the control group; however, the orthotopic tumor fluorescence intensity was markedly reduced in the Pen-H-treated group, indicating that Pen-H effectively inhibited local tumor progression. By Day 21, vehicle-treated mice developed liver and peritoneal metastases, whereas tumor progression and dissemination were significantly suppressed in the Pen-H-treated group (Fig. 6H–J). Histological analysis (H&E staining) and Ki67 immunohistochemical staining of the primary tumor site further confirmed that Pen-H exerted a strong inhibitory effect on GC progression (Fig. 6K). Collectively, these findings demonstrate that Pen-H not only suppresses local tumor growth, but also prevents distant metastasis by targeting SREBP-1 in vivo without apparent toxicity.

Pen-H enhances 5-fluorouracil (5-fu) chemosensitivity in GC

Currently, the standard postoperative adjuvant therapy for advanced GC in clinical practice remains predominantly chemotherapy-based, with 5-fluorouracil (5-Fu) serving as the cornerstone of chemotherapy regimens [19]. According to clinical guidelines, most targeted therapies are recommended in combination with chemotherapeutic agents to optimize the therapeutic outcomes [20]. Accumulating evidence suggests that the dysregulation of lipid metabolism may contribute to the development of chemotherapy resistance in cancer. Consequently, targeting fatty acid metabolism has emerged as a promising therapeutic strategy to overcome drug resistance, attracting increasing attention in cancer research [21]. Given that Pen-H exerts its anticancer effects primarily by modulating SREBP1-mediated lipogenesis, we further investigated whether Pen-H, as a targeted inhibitor, could enhance the chemosensitivity of cancer cells to 5-Fu treatment.

Analysis of the Genomics of Drug Sensitivity in Cancer (GDSC) database demonstrated a negative correlation between the expression levels of key genes involved in the SREBF1-mediated fatty acid metabolism pathway and the chemosensitivity of gastric cancer cells to 5-Fu (Fig. 7A), implying that upregulation of these genes may promote chemotherapy resistance. To corroborate these findings, we conducted additional analyses using data from The Cancer Therapeutics Response Portal (CTRP) database, which revealed a negative association between the SREBF1-mediated fatty acid metabolism pathway and 5-Fu chemosensitivity (Fig. 7B). Collectively, these findings suggest that aberrant overexpression of SREBF1 in tumors may contribute to decreased chemosensitivity to 5-Fu.

Fig. 7.

Pen-H enhances 5-fu chemosensitivity in gastric cancer. (A) Correlation between GDSC anti-GC drug sensitivity and gene expression in the SREBP1-mediated fatty acid metabolism pathway. (B) Correlation between CTRP 5-fu sensitivity and gene expression in the SREBP1-mediated fatty acid metabolism pathway. (C) Total number of viable cells was counted after treatment of MKN-45 and HGC-27 cells with pen-H (1.25 μM), 5-fu (2.5 μM) or combination for 96 h. (D) Total number of colony formations was counted after treatment of MKN-45 and HGC-27 cells with pen-H (1.25 μM), 5-fu (2.5 μM) or combination for 10 days. (E) PDX-derived organoids were treated with pen-H (1.25 μM), 5-fu (2.5 μM) or combination. Four days later, representative images were captured by a fluorescence microscope (top three rows) or a standard light microscope (bottom row). Scale bar represents 200 µM. (F) Immunoblotting showing the expression of apoptotic markers in GC cells exposed to vehicle, pen-H (1.25 μM), 5-fu (2.5 μM) or combination for 48 h. (G) GC cells were treated with pen-H (1.25 μM), 5-fu (2.5 μM) or combination for 24 h. Wound healing assays were performed to detect cell migration. (H) The invasion ability of GC cells treated with with pen-H (1.25 μM), 5-fu (2.5 μM) or combination were evaluated by transwell assay. Data were shown as the mean ± SD. Student’s t test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3. (I-K) Effects of the indicated treatments on the growth of MKN45 cell-based xenografts. Male nude mice carrying gastric cancer xenografts were randomly divided into four groups and given pen-H (5 mg/kg; i.P.; once daily), 5-FU (10 mg/kg; i.P.; two times per week), or combination for 15 days. Representative tumor image and tumor weight at the end time point were captured. Changes in tumor volume (mean tumor volume ± SEM) in different groups (G), representative images of xenograft tumors harvested after treatment for 15 days (H), and mean tumor weight ± SEM (I) are shown, * p < 0.05, ** p < 0.01, *** p < 0.001

Therefore, we further investigated the effect of Pen-H on enhancing sensitivity to 5-Fu both in vitro and in vivo. The results demonstrated that the combination of Pen-H and 5-Fu significantly enhanced the inhibition of cell proliferation (Fig. 7C) and colony formation (Fig. 7D, Fig. S9A) compared with 5-Fu monotherapy. Further evaluation using 3D organoid models revealed that Pen-H markedly potentiated the inhibitory effect of 5-Fu on GC PDOs (Fig. 7E). Additionally, combination treatment upregulated the expression of apoptotic markers, including c-caspase-7 and c-PARP-1, in GC cells (Fig. 7F). Furthermore, the combined treatment exhibited a more pronounced suppression of GC cell invasion and migration than either treatment alone (Fig. 7G, H; Fig. S9B-D).

To evaluate the antitumor effects of Pen-H and 5-Fu in vivo, we established a BALB/c nude mouse xenograft model using MKN-45 gastric cancer cells. The combined treatment with Pen-H and 5-Fu demonstrated a significantly stronger inhibitory effect on tumor growth than either treatment alone (Fig. 7I–K). Furthermore, the combined administration did not induce significant toxic side effects in the major organs of the mice at effective doses (Fig. S9E-G). These results indicated that Pen-H enhanced the sensitivity of GC cells to 5-Fu, potentiating the inhibitory effects of 5-Fu on tumor growth both in vitro and in vivo. Collectively, these findings suggested that Pen-H synergistically enhanced the therapeutic efficacy of 5-Fu in GC.

Discussion

The dysregulation of lipid metabolism is one of the most prominent metabolic alterations in cancer [22]. In recent years, increasing evidence has shown that lipid metabolism supports tumorigenesis and disease progression by enhancing lipid synthesis, storage, and catabolism [4]. Lipid metabolism encompasses various complex metabolic pathways, the roles of which vary significantly across cancer types. The complexity and diversity of lipid metabolic pathways, along with the multitude of molecular targets involved, pose significant challenges in identifying core therapeutic targets [23]. In the era of precision medicine, it is crucial to accurately analyze the key lipid metabolic pathways specific to particular cancer types and employ functional drug screening strategies to identify effective druggable targets and lead compounds. This approach is of foremost importance for the development of targeted therapies. Our study analyzed transcriptome data and found that the expression of de novo fatty acid synthesis gene sequences was significantly higher in GC tissues than in normal gastric tissues, and that dysregulation of fatty acid metabolism contributed to the progression of GC. These findings prompted us to focus on the aberrant fatty acid metabolism in GC cells.

De novo fatty acid synthesis is regulated by two key enzymes, ACC1 and FASN [24], which are closely linked to the progression of GC [25, 26]. ACC1 and FASN expression is controlled by SREBP-1, a major transcription factor involved in lipid metabolism. In this study, Pen-H inhibited the expression of SREBP-1 and SREBP-2. However, only the addition of fatty acids significantly counteracted the inhibitory effects of Pen-H on GC cells, which led us to focus on SREBP-1-mediated lipogenesis in GC. Analysis of patient samples revealed that SREBF-1 is overexpressed in GC tumors, and its high expression is associated with poor patient survival. Silencing SREBP-1 significantly reduced GC cell survival and induced apoptosis, highlighting its essential role in GC cell viability. In previous investigations targeting SREBP for cancer therapy, insufficient attention has been paid to the interplay between different SREBP isoforms within lipid metabolic pathways and their distinct functional roles across various cancer types. Our study not only demonstrates that SREBP-1, rather than SREBP-2, is the key determinant in lipid metabolism remodeling, but also provides a unique opportunity for an effective therapeutic intervention for GC. Based on our findings, we propose that future research on SREBP-targeted antitumor therapies should emphasize the identification of the specific SREBP isoform that predominantly drives tumor progression in a particular cancer type. In addition, drug development efforts should prioritize the creation of isoform-specific inhibitors to enable precise and effective therapeutic interventions.

Although no cocrystal structure of an SREBP-inhibitor complex exists, the mechanisms of action of several compounds have been elucidated [27]. However, current SREBP inhibitors have limited efficacy and no small-molecule inhibitors for the treatment of malignant tumors have been approved for clinical trials. Therefore, it is crucial to identify and characterize novel and effective SREBP inhibitors for the treatment of malignant cancers. Fatostatin, an SREBP antagonist, is commonly used as a positive control. In this study, the use of fatostatin further validated SREBP1 as a promising drug target for gastric cancer treatment. In the present study, we identified a novel SREBP inhibitor, Pen-H, from a natural product library. SPR, CETSA, and molecular docking analyses demonstrated that Pen-H selectively binds to SREBP and effectively inhibits its activity. Pen-H significantly suppresses the expression of genes involved in lipid metabolism in GC cells. Using xenograft models derived from GC cell lines, we demonstrated that Pen-H markedly inhibited GC tumor growth in vivo. Through a comparative analysis of Pen-H and Fatostatin, we observed that Pen-H exhibited a lower IC50 value and demonstrated superior anticancer efficacy, highlighting its potential as an effective therapeutic agent. Additionally, Pen-H was non-toxic to major organs and did not affect the body weight in mouse models. These findings suggest that Pen-H is an excellent lead compound for exploring the pharmacological roles of SREBP in cancer.

Chemotherapeutic drugs effectively inhibit tumor cell proliferation and survival via various mechanisms. However, tumor cells often develop intrinsic and acquired chemoresistance under chemotherapeutic stress, limiting the effectiveness of chemotherapy and leading to treatment failure. Increasing evidence suggests that alterations in lipid metabolism are closely related to drug resistance in tumors [28–30]. Abnormal fatty acid metabolism is crucial for tumor metabolic adaptation and contributes to treatment resistance by reducing drug accumulation and affecting signaling pathways, cellular lipid saturation, ER stress, and ferroptosis [21]. To date, no studies have investigated whether targeting lipid metabolism can reverse chemotherapy resistance in GC. Although 5-Fu is the first choice of treatment for advanced GC, its efficacy is often limited by drug resistance. To further support our rationale, we analyzed the GDSC and CTRP databases and found that the expression of key genes in the SREBP1-mediated fatty acid metabolism pathway was negatively correlated with the sensitivity of GC cells to 5-Fu. Specifically, high SREBP1 expression was associated with reduced sensitivity to 5-Fu, suggesting that SREBP1-mediated fatty acid metabolism contributes to chemoresistance. Furthermore, we found that the combination of Pen-H and 5-Fu synergistically decreased the proliferation, migration, and invasion of GC cells. Pen-H appears to sensitize GC cells to the anti-cancer effects of 5-Fu, suggesting that SREBP inhibition may have broad clinical applications in GC. These findings indicate that targeting SREBP could be an effective strategy for alleviating chemotherapy resistance. The combined effect of Pen-H and chemotherapeutic drugs for treating chemotherapy-resistant GC warrants further in-depth studies.

Our findings not only provide Pen-H as a promising drug lead for GC but also demonstrate that SREBP-1 can be explored as an effective therapeutic target in this disease. As SREBP-1 mediated lipogenesis may play a significant regulatory role in other tumors, Pen-H may serve as an efficient tool for future SREBP-based drug discovery. This will help inspire researchers to explore all possible combinations of factors for novel drug design rather than limiting themselves to traditional drug characteristics.

Conclusions

In this study, we identified a highly potent SREBPs inhibitor, Pen-H, by screening natural compounds from marine fungi. Pen-H exhibited ultrapotent anti-GC activity, both in vitro and in vivo by suppressing fatty acid metabolism (Fig. 8). Based on a functional drug screening strategy aimed at identifying novel therapeutic targets, we further demonstrated that the overexpression of SREBP-1 is closely associated with GC progression. These findings revealed that SREBP-1 is a key regulator of lipid metabolism and a promising therapeutic target for GC. Correlation analysis of gene expression and drug sensitivity revealed that high expression of SREBP-1 was negatively associated with the chemosensitivity of gastric cancer cells to 5-Fu. Furthermore, Pen-H enhanced the chemotherapeutic efficacy of 5-Fu by inhibiting SREBP-1 activity.

Fig. 8.

A proposed model illustrating therapeutic targeting SREBP-1 with marine-derived compound penicolinate-H for gastric cancer via regulating fatty acid metabolism

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

GC

Gastric cancer

SREBP

Sterol regulatory element binding protein

Pen-H

Penicolinate H

5-Fu

5-fluorouracil

ACC

Acetyl-CoA carboxylase

FASN

Fatty acid synthase

HMGCR

3-hydroxy-3-methylglutaryl coenzyme A reductase

SQLE

Squalene monooxygenase

Author contributions

Jianjiao Chen: Writing – original draft, Investigation, Formal analysis. Haishan Cui: Writing – original draft, Formal analysis. Xiaolu Wang: Writing – original draft, Formal analysis. Yechun Zeng: Methodology, Conceptualization. Dongyue Pan: Conceptualization. Jun Chen: Resources. Zhenhua Zhang: Conceptualization. Yana An: Data curation. Zhanfeng Gu: Formal analysis. Guodi Cai: Investigation, Methodology. Hong Wang: Supervision. Mengya Yu: Resources. Bin Yang: Resources. Shengning Zhou: Resources. Jianan Tan: Resources. Lan Liu: Supervision. Junjian Wang: Supervision, Project administration, Funding acquisition. Jing Li: Supervision, Project administration, Funding acquisition. Fanghai Han: Supervision, Project administration, Funding acquisition. All data were generated in-house, and no paper mill was used. All authors agree to be accountable for all aspects of work ensuring integrity and accuracy. All authors approved the publication of this manuscript.

Funding

This research was supported by the National Natural Science Foundation of China (82273956, 82003253), Guangdong Basic and Applied Basic Research Foundation (2022B1515130008), Guangzhou Science and Technology Plan Project (202206080007), the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (No. 24xkjc016) and Talent introduction project of The Affiliated Guangdong Second Provincial General Hospital (YY2024-008, YY2024-0089).

Data availability

The data that support the findings of this study are available upon request from the corresponding author.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Commitee of Sun Yat-sen University. The Clinical Research Ethics Committee of Sun Yat-sen University approved all clinical experiments (approval nos. SYSKY-2023–763-01). The Committee for Ethics of Animal Experimentation approved the animal experiments, which followed the guidelines for animal experiments at Sun Yat-sen University (approval nos. SYSU-IACUC-2023-000429).

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.