Skip to main content
Download PDF

LSD1 is a targetable vulnerability in gastric cancer harboring TP53 frameshift mutations

Clinical Epigenetics volume 17, Article number: 26 (2025) Cite this article

Abstract

Background

TP53 mutations are linked to aggressive progression and chemoresistance in gastric cancer (GC). Frameshift mutation is the second most common mutation type of TP53. However, the consequences of this mutation type in GC were not well understood, and targeted therapies for cancer patients harboring frameshift mutations were also not established. Histone methylation significantly influences tumorigenesis in TP53-mutated cancers, and related inhibitors are emerging as specific therapeutic strategies.

Methods and results

By treating GC cell lines harboring various TP53 mutation types with a library of histone demethylase inhibitors, we identified that GSK690, a reversible inhibitor of lysine-specific demethylase 1 (LSD1), selectively inhibits GC cells harboring TP53 frameshift mutations without nuclear localization sequence (NLS) (termed TP53 Frameshift NLS), which accounts for 89% TP53 frameshift mutations in GC patients. GSK690 showed significant specific inhibition in vitro and in vivo against this subtype by inducing G1/S cell cycle arrest via the LSD1-CCNA2 axis. Importantly, dual-luciferase assays and ChIP-qPCR confirmed that the loss of transcriptional repression activities of p53 in drives LSD1 upregulation in TP53 Frameshift NLS cancer cells.

Conclusions

In summary, our results indicate that the nuclear localization deficiency of p53 accounts for increased expression of LSD1 in TP53 Frameshift NLS GCs. GSK690 inhibits cell cycle progression and tumor growth by suppressing aberrantly activated LSD1-CCNA2 signaling in this GC subtype, counteracting malignant proliferation and thereby providing a precise therapeutic strategy for GC patients with TP53 Frameshift NLS.

Introduction

Gastric cancer (GC) is a global health concern, with the fifth incidence and mortality rates worldwide [1]. TP53 mutations are one of the most common genetic alterations in GC [2]. Unfortunately, GCs harboring TP53 mutations often exhibit chemoresistance, resulting in limited benefits from traditional chemotherapy for these patients [3, 4]. Therefore, there is an urgent need to identify potential targets and develop effective therapies for TP53 mutant GCs.

Multiple mutation types in TP53, including missense mutations, frameshift mutations, deletion, and nonsense mutations, have been reported in human cancers, including GC. Among these, frameshift mutation is the second most common type after missense mutations [5]. However, frameshift mutations are complex, as different mutation sites can produce distinct p53 mutants. In recent years, research on TP53 hotspot mutations has made significant progress, with several drugs entering clinical trial phases [6]. In contrast, advancements in targeting GCs with TP53 frameshift mutations have been limited due to the inherent complexity of the protein structure [7,8,9]. When frameshift mutation occurs before the nuclear localization signal of p53, the resultant p53 mutants fail to translocate to the nucleus and then exert its transcription factor activity [10]. We refer to this mutation type as TP53 Frameshift NLS. Mechanisms by which TP53 Frameshift NLS cancers maintain their malignant proliferation and how to efficiently treat these patients remain to be elucidated.

Studies have indicated that cancers with TP53 hotspot mutations can exert a gain-of-function (GOF) effect through the nuclear accumulation of mutant p53 protein, which promotes oncogenesis and progression [11,12,13]. In contrast, TP53 frameshift mutations typically result in the complete loss of p53 protein or a nuclear localization deficiency [14], suggesting a greater tendency toward loss of function rather than gain of function, similar to mechanisms that maintain malignant proliferation in various cancers [15]. However, does the loss of p53 function solely affect traditional downstream targets of the p53 signaling pathway? Notably, tumors harboring TP53 mutations often exhibit alterations in histone methylation levels [16]. Furthermore, our previous research also demonstrated that the histone demethylase KDM4C can mediate the senescence defense in gastric cancer harboring TP53 mutations [17]. Furthermore, the sensitivity to the inhibitors of histone methylation modulators, such as DZNep, varies among cancer patients with different TP53 mutation types [17, 18]. Therefore, we hypothesize that the histone methylation process may play a role in sustaining malignant proliferation in GCs following the loss of p53 function. To test this hypothesis, we treated GC cells with a library of histone demethylase inhibitors and found that GSK690, an inhibitor of LSD1, could efficiently kill TP53 Frameshift NLS but show negligible or weak activities against other GC types. Mechanistically, the loss of p53 transcriptional repression in TP53 Frameshift NLS GCs led to the upregulation of LSD1, which promotes cell cycle progression via the CCNA2/CDK2, thereby maintaining malignant proliferation of cancer cells. Our findings provide a novel therapeutic strategy for TP53 Frameshift NLS cancers and elucidated the underlined mechanism.

Materials and methods

Cell lines

Four GC cell lines (AGS, NCI-N87, MKN45, and HGC-27) were purchased from ATCC. The human GC cell line SNU-1 was obtained from Boster Biologics (Wuhan, China), and the human esophageal adenocarcinoma cell line SK-GT-4 was purchased from Meisen CTCC (CTCC-007-0167, China). AGS and SNU-1 are gastric adenocarcinoma cell lines with wild-type (WT) TP53. NCI-N87 is a gastric adenocarcinoma cell line harboring R248Q TP53 mutation. HGC-27 is a gastric adenocarcinoma cell line harboring P153fs TP53 mutation. SK-GT4 is an esophageal adenocarcinoma cell line harboring Q100Ter TP53 mutation (lacking NLS). All cell lines were cultured in DMEM or RPMI 1640(SNU-1, SK-GT-4) supplemented with 10% FBS and maintained at 37 °C with 5% CO2.

Immunoblotting assay

Cell and tissue samples were lysed using RIPA lysis buffer supplemented with 1% Protease Inhibitor Cocktail (CWBIO, CW2200S) and 1 mM PMSF (Beyotime, ST506). The samples were then centrifuged for 30 min at 14,000 rpm after ultrasonication. The bicinchoninic acid kit (Yeasen, #20201ES86) was used to measure protein concentration. Total proteins were separated by a 10% precast protein plus gel (Yeasen, #36252-ES10) and transferred to PVDF membranes (Millipore, IPVH00010). The PVDF membranes were blocked with 5% milk for 1 h and incubated with primary antibodies at 4 °C overnight, followed by incubation with secondary antibodies (Jackson, #115-035-003 and #111–035-003, 1:10,000) for 2 h. Finally, the bands were visualized using an ECL Kit (Vazyme, #E412-02). The following antibodies were used for Western blotting: anti-LSD1 (CST, #4218, 1:1000), anti-p53 (ABclonal, A19585, 1:1000) and anti-Cyclin A2 (ABclonal, A19036, 1:1000).

Cell proliferation and colony formation assays

Cells were treated separately and plated into 96-well or 6-well plates at a density of 2000 cells per well and treated with GSK690 (MCE, HY-117226A) at different concentrations. For cell proliferation assays, cells in the 96-well plates were incubated in medium containing 10% Cell Counting Kit 8 (Beyotime, CCK-8) at 37 °C for 2 h. Afterward, OD values at 450 nm were measured using a microplate reader. For colony formation assays, cells in the 6-well plates were fixed with 4% paraformaldehyde fix solution (Sangon, E672002) after being treated with GSK690 for 10–15 days; GSK690 was changed every 2 days during treatment. The colonies were stained with 0.1% crystal violet (Beyotime, C0121) and photographed using a scanner. The colonies were counted using ImageJ. The above experiments were repeated at least 3 times.

Histone demethylation drug screening

A total of 3000 cells were plated into 96-well plates and then sequentially treated for 4 days with one of the inhibitors from the epigenetics compound library (MCE, #HY-L005, 5 μM). Afterward, cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay. The drugs used in this study are listed in Supplementary Table S1.

Half maximal inhibitory concentration (IC50) assay

Cells were plated in 96-well plates at a density of 3000 cells per well and treated with various concentrations of GSK690 or GSK2879552. After 4 days, cell proliferation was assayed using the CCK-8 method, and the relative inhibition rate and IC50 values were calculated using GraphPad Prism 8.0.

Quantitative reverse transcription PCR (qRT-PCR)

Total RNA was extracted from tumor cells using the RNA Isolater Total RNA Extraction Reagent (Vazyme, R401-01) and then reverse transcribed into cDNA using the HiFiScript cDNA Synthesis Kit (CWBIO, CW2569M). Gene expression levels were measured by qRT-PCR using the 2 × SuperFast Universal SYBR Master Mix (CWBIO, CW3888M). The relative levels of gene expression were normalized to the expression levels of β-actin and calculated using the 2−△△Ct method. The relevant primer sequences are listed in Supplementary Table S2.

Cell cycle assay

Cells treated with GSK690 or DMSO were collected and fixed overnight in 70% ethanol at − 20 °C. The cells were stained using a cell cycle and apoptosis analysis kit (Beyotime, C1052) and analyzed using a NovoCyte flow cytometer (Agilen).

5-Ethynyl-2′-deoxyuridine (EdU) assay

The EdU assay was performed using the BeyoClick™ EdU Cell Proliferation Kit with Alexa Fluor 594 (Beyotime, C0078L) according to the manufacturer’s instructions. An Olympus fluorescence microscope was used to capture fluorescent images.

In vivo anti-cancer effect of GSK690

The 5-week-old male nude mice were purchased from SPF (Beijing, China) and maintained under specific pathogen-free conditions. A total of 5 × 106 HGC-27 cells were resuspended in a 1:1 mixture of Matrigel Matrix (Yeasen, #40183ES08) and PBS, which were injected dorsally into the right subcutaneous side of the nude mice. When tumors became visible, the mice were randomly assigned to the control, GSK690 5 mg/kg, or GSK690 10 mg/kg treatment groups and were administered GSK690 or DMSO every other day for 21 days. Every 3 days, the size of the tumors and the body weight of the mice were measured using calipers and electronic scales, respectively. After 21 days, the mice were sacrificed, and the tumors were removed, photographed, and weighed. Tumor volume was calculated using the formula: volume = length × width2 × 0.5.

Immunohistochemistry (IHC)

Xenograft tumors were fixed in 10% neutral formalin fix solution (Sangon, E672001) for 24 h after resection and then embedded in paraffin. Fixed tumors and tissues were processed into 4 mm paraffin sections, which were dried at 72 °C, dewaxed, dehydrated, antigen-repaired, sealed, and incubated overnight with primary antibodies at 4 °C. Secondary antibody incubation, DAB staining, and other related procedures were performed on the 2nd day. Images were captured using an Olympus microscope. The primary antibodies used in the assay included anti-LSD1 (CST, #4218, 1:200) and anti-Cyclin A2 (ABclonal, A19036, 1:200).

Chromatin immunoprecipitation quantitative PCR (ChIP-qPCR)

Initially, cells were fixed with formaldehyde at a final concentration of 1% for 10 min. The cell lysates were sheared by sonication to generate 400–800 bp fragments. ChIP assays were performed using anti-p53 (ABclonal, A19585), anti-H3K9me2 (Abcam, ab1220) and IgG (Beyotime, A0208) via a ChIP Assay Kit (Beyotime, P2078). The immunoprecipitation products were purified and quantified by qPCR. The relevant primer sequences are listed in Supplementary Table S2.

Dual-luciferase reporter assays

The wild-type or mutant promoter regions of LSD1 were cloned into the reporter gene vector PGL3-basic, respectively. The resultant reporter plasmids were co-transfected with the p53 overexpression plasmid into cells, and 48-h post-transfection, the samples were subjected to dual-luciferase reporter assays using the dual-luciferase reporter assay kit (Vazyme, DL101-01) according to the manufacturer’s instructions. The relevant primer sequences are listed in Supplementary Table S2.

Study approval

The animal experiments were approved by the Institutional Animal Care and Use Committee of the affiliated hospital of Jiangnan University (approval number: JN.No2024093-0b0301207 [522]).

Statistical analyses

GraphPad Prism 8.0 was used for all statistical analyses. An unpaired Student’s t-test was used to analyze differences between 2 groups. One-way analysis of variance (ANOVA) was used to analyze differences among multiple groups. P < 0.05 indicated a significant difference.

Result

The deficiency in p53 nuclear translocation due to TP53 frameshift mutations is common in GC and correlates with poor prognosis

Analyses of the TCGA GC dataset revealed that TP53 was one of the most frequently mutated genes in GC, particularly prevalent in Asian GC populations (Fig. 1A, B). Among the various types of TP53 mutations, frameshift mutations ranked second only to hotspot mutations, establishing them as the second most common mutation type of TP53 in GC (Fig. 1C). The complexity associated with frameshift mutations leads to differing mechanisms of action, as distinct mutation sites generate various mutant p53 proteins. Notably, frameshift mutations occurring prior to the NLS render the altered p53 protein incapable of translocating into the nucleus, thereby depriving its transcriptional activity; this subtype has been designated as TP53 Frameshift NLS GC (Fig. 1D). Further analyses of TCGA multiple GC datasets since 2010 revealed that the incidence of nuclear translocation defects linked to frameshift mutations was as high as 89% GC harboring TP53 frameshift mutations (Fig. 1E). Overall, approximately 17.3% of GC patients exhibited nuclear translocation deficiencies resulting from TP53 frameshift mutations (Fig. 1F). Unfortunately, the data indicated that these patients often had poorer prognoses than other GC populations (Fig. 1G). These findings demonstrate that nuclear translocation deficiencies caused by TP53 frameshift mutations are common in GC and are associated with unfavorable prognoses, underscoring the urgent need to identify potential therapeutic targets for this subtype.

Fig. 1
figure 1

TP53 Frameshift NLS is common in GC and links to poor prognosis. A Frequently mutated genes in GCs. The data were sourced from the TCGA GC database. B The probability of TP53 mutations occurring in different human groups was assessed. C The classification and incidence rates of different mutation types in TP53. D The diagram of TP53’s structural domains and mutations was based on the analysis of GC datasets in TCGA since 2010. E The proportion of NLS deletions in frameshift mutations of TP53. F The percentages of TP53 frameshift mutations and other mutation types in the TCGA GC dataset. G Kaplan–Meier survival analysis evaluated the overall survival (OS) of GC patients with different TP53 mutation types

Full size image

Screening a library of histone demethylase inhibitors to identify potential drugs to specifically kill TP53 Frameshift NLS GC

We assessed the expression of the p53 protein in six cancer cell lines harboring various TP53 mutations or WT used in our study. Due to failure to acquire additional TP53 frameshift NLS GC cell lines aside from HGC-27, we selected the esophageal adenocarcinoma cell line SK-GT-4, which harbors Q100Ter TP53 mutation and lacks NLS, to validate related results in TP53 frameshift NLS GC cells (HGC-27). Our results indicated that the expression levels of p53 in TP53 Frameshift NLS GCs were significantly lower than those in GC patients harboring TP53 WT and point mutation types (Fig. 2A). This raised the question of how these p53-deficient GCs maintained their rapid proliferation phenotype. Previous studies reported that the levels of histone methylation varied to different extents in GC populations with TP53 frameshift mutations [16]. Based on this, we designed a screening experiment using a library of histone demethylase inhibitors to treat three GC cells with different mutation types (AGS, NCI-N87, and HGC-27), aiming to identify drugs that specifically kill TP53 Frameshift NLS GC (Fig. 2B). The drug screening results demonstrated that GSK690, a reversible inhibitor of LSD1, effectively inhibited the proliferative activity of HGC-27, a TP53 Frameshift NLS GC cell line. Interestingly, GSK690 showed minimal proliferation-inhibitory activities toward GC cells with WT TP53 (AGS) or point mutation TP53 (NCI-N87) (Fig. 2C). Molecular docking analyses using the previously reported structure of LSD1 (PDB ID: 9WFG) [19] and confirmed the association between LSD1 and GSK690 (Fig. 2D). To further validate whether GSK690 specifically targeted TP53 Frameshift NLS GC, we measured the IC50 values of GSK690 in six cancer cells with different TP53 mutation types. The results revealed that the IC50 value of GSK690 in TP53 Frameshift NLS GC was significantly lower than that in WT and hotspot mutation subtypes (Fig. 2E, F). Interestingly, another LSD1 inhibitor, GSK2879552, also exhibited high selective inhibitory activity against TP53 Frameshift NLS cancer cells (Fig. S1A, B). These findings suggest that LSD1 inhibitors, particularly GSK690, might serve as a specific therapeutic agent for TP53 Frameshift NLS GC.

Fig. 2
figure 2

GSK690 specifically inhibits GC cells harboring TP53 Frameshift NLS. A Immunoblotting of baseline expression levels of p53 in GC cells with different mutation types was performed. B Schematic diagram of screening epigenetic drugs for the specific treatment of TP53 GC was created. C Sequential treatment of GC cells with TP53 wild-type, TP53 hotspot mutations, and TP53 Frameshift NLS was conducted using individual histone demethylase inhibitor compounds (5 μM) from a library, followed by CCK-8 assay to estimate inhibition rates and generate a heatmap. D Predicted binding mode of GSK690 in the LSD1 active site (9FWG). Visualized using AutoDock Vina and Pymol. E, F GC cells were treated with GSK690 at different concentrations, and the inhibition rates were calculated using the CCK-8 assay. The IC50 values of GSK690 in various GC cell lines were derived from the fitted curves, and a heatmap was generated. The relevant IC50 data for the heatmap could be found in Fig. S2A

Full size image

GSK690 inhibits TP53 Frameshift NLS GC in vitro

To further validate the specific inhibitory effect of GSK690 on TP53 Frameshift NLS GC, we treated GC cells with different types of TP53 mutations using various concentrations of GSK690. The CCK-8 and colony formation assays indicated that GSK690 exerted a strong inhibitory effect on TP53 Frameshift NLS cancer (HGC-27 and SK-GT-4) in a dose-dependent manner but showed weak proliferation-inhibitory effect in GC cells with WT (AGS) and point mutation (NCI-N87) TP53 (Fig. 3A, S2B, C). This specific inhibitory effect of GSK690 was further validated in the two TP53 Frameshift NLS cancer cells using both colony formation assays and EdU cell proliferation assays (Fig. 3B–E). Taken together, these data suggest that GSK690 could efficiently inhibit the malignant proliferation of TP53 Frameshift NLS GC in a dose-dependent manner in vitro.

Fig. 3
figure 3

GSK690 inhibits TP53 Frameshift NLS GCs in vitro. A–C AGS, NCI-N87, HGC-27, and SK-GT-4 cells were treated with different concentrations of GSK690 (5, 10 μM) or DMSO, and were then subjected to cell proliferation assays (A), colony formation assays (B), and EdU assays (C). D, E Statistical data from the colony formation (B) assays and EdU assays (C) were presented. *P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001. P values were determined by a two-tailed unpaired t-test. Data were presented as mean ± SEM of three independent experiments

Full size image

GSK690 inhibits TP53 Frameshift NLS GC in vivo

To further evaluated the growth-inhibitory effect of GSK690 on TP53 Frameshift NLS GC, we constructed a xenograft mouse model using HGC-27 cells and administered GSK690 at different concentrations (0, 5 and 10 mg/kg) via intraperitoneal injection (Fig. 4A). Consistent with the in vitro findings, GSK690 significantly inhibited the in vivo growth of TP53 Frameshift NLS tumors in a dose-dependent manner (Fig. 4B–D). Furthermore, as the concentration of GSK690 increased, the expression level of the tumor proliferation marker Ki-67 correspondingly decreased (Fig. 4E). Notably, we did not observe significant changes in the body weight of the nude mice during the GSK690 treatment (Fig. 4F). Additionally, in subsequent histological examinations of the heart, liver, spleen, lungs, and kidneys, we did not observe obvious histological changes in these organs (Fig. 4G), indicating that GSK690 did not exhibit obvious in vivo toxicity. Together, these data demonstrate that GSK690 could inhibit the malignant proliferation of TP53 Frameshift NLS GC in a dose-dependent manner in vivo.

Fig. 4
figure 4

GSK690 inhibits TP53 Frameshift NLS GCs in vivo. A Schematic diagram of drug administration in a xenograft mouse model. B Representative images of xenograft tumors with the indicated treatments at the end point (n = 7). C Scatter bar chart of xenograft tumors with the indicated treatments at the end point (n = 7). D Tumor growth curves for each group during the treatment period (n = 7). E Representative Ki-67 immunohistochemical staining in tumor tissues derived from the indicated groups. Scale bar, 50 μm. F The body weight of mice in each group during the treatment period (n = 7). G Representative images of hematoxylin and eosin (HE) staining of heart, liver, spleen, lung, and kidney tissues from designated groups. Scale bar, 100 μm

Full size image

GSK690 induces G1/S phase cell cycle arrest in TP53 Frameshift NLS GC by inhibiting the LSD1-CCNA2 axis

As a specific reversible inhibitor of LSD1, GSK690 has shown significant proliferation-inhibitory effects on TP53 Frameshift NLS GC, suggesting that LSD1 may play a critical role in maintaining the malignant proliferation of this GC subtype. To investigate potential mechanism of LSD1 in TP53 Frameshift NLS GC, we categorized TP53 Frameshift NLS GC patients in the TCGA GC cohort into high and low expression groups of KDM1A(LSD1 coding gene) (Fig. 5A). Differentially expressed genes between the two groups predominantly enriched in the “Cell Cycle” pathway (Fig. 5B), suggesting that LSD1 may promote malignant proliferation of TP53 Frameshift NLS GC by affecting the cell cycle. Flow cytometry analyses revealed that GSK690 dose-dependently induced G1-S phase arrest in TP53 Frameshift NLS GC, and similar results were obtained in KDM1A knockdown experiments using KDM1A-specific siRNAs (Fig. 5C, D). To identify key factors mediating LSD1 inhibition-induced G1-S phase arrest, we analyzed these differentially expressed genes and found that CCNA2, a crucial cell cycle regulator, is the most significantly upregulated gene in the KDM1A high expression group (Fig. 5A). Previous research has shown that LSD1 can upregulate the expression of CCNA2 by modulating the process of histone methylation [20], and increased CCNA2 promotes the cell cycle progression by forming a complex with CDK2 [21]. Therefore, we hypothesized that CCNA2 is a key downstream molecule regulated by LSD1 in cell cycle control. To validate this hypothesis, we treated the GC cell lines AGS, NCI-N87, and HGC-27 with various concentrations of GSK690. The results indicated that CCNA2 expression remained unaffected in AGS and NCI-N87 cells but was inhibited in HGC-27 cells by GSK690 or KDM1A knockdown (Fig. 5E). In the HGC-27 xenograft tumors, both Western blotting and IHC results further indicated that GSK690 suppress CCNA2 expression (Fig. 5F, S3A). A previous study has indicated that LSD1 regulates CCNA2 expression by decreasing H3K9me2 levels in the CCNA2 promoter region [20]. Consistent with these findings, our ChIP analysis revealed that obvious H3K9me2 enrichment was observed in the CCNA2 promoter in NCI-N87 and AGS cells but not in HGC-27 cells. More importantly, GSK690 could induce H3K9me2 enrichment in the CCNA2 promoter in HGC-27 cells (Fig. 5G). Collectively, these data suggest that CCNA2 serves as a critical downstream target of LSD1 to regulate the cell cycle progression in TP53 Frameshift NLS GC cells.

Fig. 5
figure 5

LSD1 promotes cell cycle progression through CCNA2. A Volcano maps for differentially expressed genes. Red dots, significantly upregulated genes. Blue dots, significantly downregulated genes. Gray dots, nondifferentially expressed genes. B KEGG pathway enrichment analyses of genes exhibiting significant differences between high and low expression levels of KDM1A in the TCGA GC cohort. C, D Apoptosis assays were performed using flow cytometry in HGC-27 and SK-GT-4 cells treated with GSK690 C or transfected with si-KDM1A D. E The effects of GSK690 treatment (5 and 10 μM) or KDM1A knockdown on the expression of CCNA2 in GC cell lines. F Immunohistochemical staining results of CCNA2 in HGC-27 cell-derived xenografted tumors. Scale bar, 50 μm. G ChIP analyses using IgG and H3K9me2 antibodies to assess the abundance of H3K9me2 in the CCNA2 promoter in cancer cells

Full size image

LSD1 overexpression resulting from p53 transcriptional deficiency leads to accelerated proliferation in cancer cells harboring TP53 Frameshift NLS

Based on the aforementioned results, we speculated that GSK690 specifically inhibits TP53 Frameshift NLS tumors due to their significantly higher LSD1 expression than other cancer cells with hotspot mutation or WT TP53. RNA sequencing results from the TCGA GC cohort revealed that KDM1A and CCNA2 expression was significantly increased in TP53 Frameshift NLS GCs compared with other GC tumors (Fig. S3B). Moreover, the expression of KDM1A is strongly correlated with that of CCNA2 (Fig. S3C). Further validation through RT-qPCR and Western blotting confirmed the upregulated KDM1A/LSD1 expression in TP53 Frameshift NLS cancer cells compared with other GC cell lines (Fig. 6A), suggesting that the aberrantly high expression of LSD1 was responsible for maintaining the malignant proliferation of TP53 Frameshift NLS GC. To verify this speculation, we knocked down KDM1A in two TP53 Frameshift NLS cancer cell lines, HGC-27 and SK-GT-4, and then measured their proliferation activities (Fig. 5E, 6B). Consistent with the results of GSK690 treatment experiments, KDM1A knockdown significantly inhibited malignant proliferation of these two TP53 Frameshift NLS cancer cells (Fig. 6C, D). We speculated that the loss of p53-mediated transcriptional suppression due to the nuclear distribution deficiency of p53 in TP53 Frameshift NLS cancer cells contributes to the increased expression of KDM1A/LSD1 in these cells. Western blotting results showed that p53 knockdown upregulated LSD1 and CCNA2 expression in TP53 WT AGS cells, while p53 overexpression had the opposite effect in HGC-27 cells (TP53 Frameshift NLS) (Fig. 6E, S3D). Interestingly, we observed a negative relative trend between LSD1 and p53 expression across various cancer cell lines (Fig. 2A, 6B). To determine whether p53 transcriptionally suppresses KDM1A, we predicted a potential binding region for p53 in the KDM1A promoter using Jasper (Fig. 6F). Dual-luciferase and ChIP-qPCR experiments conducted in AGS cells with WT TP53 confirmed that p53 binds to this region and inhibits KDM1A transcription (Fig. 6G, H). Taken together, these data indicate that p53 can transcriptionally inhibit KDM1A, and nuclear localization deficiency of p53 causes aberrant high expression of LSD1 in TP53 Frameshift NLS GC cells, which is crucial for maintaining their malignant proliferation.

Fig. 6
figure 6

P53-mediated transcriptional inhibition of KDM1A expression suppresses malignant proliferation in GC cells harboring TP53 Frameshift NLS. A The relative mRNA and protein expression of KDM1A/LSD1 was detected in cancer cell lines harboring various TP53 mutation types using qRT-PCR and Western blotting, respectively. B The knockdown efficiency of KDM1A in SK-GT-4 cells was verified using qRT-PCR and western blotting. C, D Effects of KDM1A knockdown on the cell proliferation (C) and colony formation (D) of TP53 Frameshift NLS cancer cells. E The effects of TP53 knockdown or overexpression on the LSD1 expression in TP53 WT(AGS) or TP53 Frameshift NLS (HGC-27) GC cells, respectively. F The predicted binding sites of p53 in the KDM1A promoter via Jasper. G Relative intensity of dual-luciferase reporter activity in AGS cells co-transfected with KDM1A reporter gene vectors (or their mutants) and TP53 overexpression plasmids. H ChIP products obtained using IgG and p53 antibodies were analyzed for the abundance of predicted binding sites in the KDM1A promoter using real-time quantitative PCR. I Graphical Abstract for this study

Full size image

Discussion

GC is the fifth most common cancer worldwide [1]. In recent years, significant progress has been made in the precision treatment of GC. However, it is undeniable that traditional chemotherapy remains the primary clinical approach [3, 4]. GC is a highly heterogeneous disease, and approximately 50% of GC patients harbor TP53 mutations [2]. Unfortunately, this GC subpopulation often displays a phenotype of chemoresistance [4]. TP53 frameshift mutations represent the second most common mutation type in TP53 [5]. Regrettably, research on TP53 frameshift mutations is limited, and no effective treatment strategies have been established [7,8,9]. In this study, we define a GC subtype, TP53 Frameshift NLS, which lacks of nuclear p53 and encompasses 89% of frameshift mutations of TP53. We identified GSK690 as a specific epigenetic modulator targeting this GC subtype. Moreover, we showed that GSK690 induces cell cycle arrest in TP53 Frameshift NLS GC by inhibiting the LSD1-CCNA2 axis. In addition, we revealed that p53 transcriptionally inhibit KDM1A, and p53 nuclear deficiency contributes to aberrant overexpression of KDM1A/LSD1, which explains why TP53 Frameshift NLS GCs are especially sensitive to GSK690.

Targeting enzymatic machinery of chromatin regulation appears to be a promising anti-cancer strategy[22]. Inhibitors of DNA methyltransferases and histone deacetylases (HDACs) have shown partial efficacy in certain cancer types. Aberrant histone methylation reprogramming has been observed in p53 mutant cancers [16], suggesting potential application of histone demethylase inhibitors in kill cancers carrying TP53 mutations. Through screening a library of histone demethylase inhibitors, we revealed that GC cells with different TP53 status showed different sensitivities to different inhibitors. Notably, we found GSK690, a reversible inhibitor of LSD1, specifically inhibits TP53 Frameshift NLS cancers, including GC. LSD1 is the first histone demethylase and a FAD-dependent amine oxidase specific to H3K9me1/2 and H3K4me1/2. LSD1 has been demonstrated to promote tumorigenesis in various cancers, including GC [23,24,25,26]. In addition, multiple LSD1 inhibitors have been developed to suppress cancer growth [27,28,29]. However, the exact mechanism mediating LSD1 overexpression in GC and which GC patients are highly sensitive to LSD1 inhibitors are largely unclear.

Notably, we observed obviously differing sensitivities to GSK690 in GCs with different TP53 mutation statuses, determined by their varying levels of LSD1 expression. Further exploration revealed that p53 can transcriptionally repress the expression of LSD1. Interestingly, previous studies have suggested the regulatory effect of p53 on KDM1A. For example, Zhou et al. observed that increased KDM1A expression in TP53 mutant breast cancer compared with TP53 WT tumors. Moreover, they further demonstrated that pharmacological activation of p53 inhibits LSD1 and DNMT1 expression and then restore the expression of ERV (endogenous retroviruses), triggering viral mimicry response and promoting anti-cancer immunity [30]. However, the exact mechanism that p53 regulates KDM1A/LSD1 expression is unclear.

As a transcription factor, p53 exerts complicated functions mainly by transcriptionally regulating more than 300 target genes, which is often context-dependent. In this study, we reveal that p53 transcriptionally inhibits KDM1A. Consequently, LSD1 expression is increased in TP53 Frameshift NLS cancer cells due to the nuclear localization deficiency of p53 and then the loss of p53-mediated transcription suppression on KDM1A, promoting cell cycle progression and tumor proliferation. Interestingly, LSD1 can interact with p53 and inhibit p53 functions by regulating its methylation status at K370 [31]. In addition, Scoumanne et al. reported LSD1 knockdown induced cell cycle arrest and growth inhibition in breast cancer cells in both a p53-dependent and -independent manners [32]. These data suggest an interactive feedback regulation between p53 and LSD1. However, whether this regulatory loop is universal or context-dependent remains to be elucidated.

GSK690 is a specific reversible inhibitor of LSD1 and serves as a versatile starting point for the development of reversible LSD1 inhibitors [33, 34]. Unfortunately, its antitumor activity has only been characterized as one of the combinatorial agents in rhabdomyosarcoma cells [35], while its effects in other cancers remain completely unknown. KDM1A has been identified as an oncogene in various cancers [23,24,25], suggesting that GSK690, as a reversible inhibitor, may represent a potential therapeutic strategy for certain malignancies with high LSD1 expression. Numerous LSD1 inhibitors have demonstrated the ability to suppress gastric cancer proliferation or modify the GC immune microenvironment [28, 29, 36]. In contrast, GSK690 exhibits differential sensitivity against GC cells with various TP53 mutation statuses, thereby enabling the identification of a more precisely defined patient population. In this study, we, for the first time, identify that p53 transcriptionally suppress KDM1A expression. Furthermore, we discover that GSK690 exerts its antitumor effects by inhibiting the LSD1/CCNA2 axis to induce cell cycle arrest in TP53 Frameshift NLS cancer cells. In addition, no obvious adverse effects were observed in the xenograft mouse model systemically administrated GSK690, thereby identifying a potential therapeutic strategy for this subtype of GC. Importantly, GSK2879552, the other irreversible LSD1 inhibitor, also showed the same specific inhibitory effect as GSK690 in TP53 Frameshift NLS cancer cells, further confirming the key role of LSD1 and the therapeutic potential of LSD1 inhibitors in this subtype. Further preclinical studies and clinical trials are desired to promote the application of LSD1 inhibitors in treating TP53 Frameshift NLS cancer.

There are several limitations in this study. Firstly, it primarily focuses on gastric cancer and esophageal adenocarcinoma, which share similar histological characteristics. Therefore, further research is needed to determine whether these conclusions are applicable to other types of cancer. Additionally, while the nuclear import defects and transcriptional inactivation resulting from the disruption of the p53 nuclear localization deficiency are anticipated, the inability to detect the p53 mutant protein resulting from TP53 frameshift mutations using conventional antibodies limits our capacity to characterize this subtype using visual techniques such as immunofluorescence.

Conclusion

In summary, this study defines a novel subtype of gastric cancer and identifies GSK690 as a potential therapeutic strategy for TP53 Frameshift NLS GC. Mechanistically, GSK690 inhibits the LSD1-CCNA2 axis to kill this subtype, LSD1 upregulation is due to the transcriptional deficiency of p53.

Data availability

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

Abbreviations

qRT-PCR :

Quantitative reverse transcription polymerase chain reaction

CHIP:

Chromatin immunoprecipitation

CCK-8:

Cell counting Kit-8

DAB:

Diaminobenzidine

DMEM :

Dulbecco’s modified Eagle’s medium

DMSO:

Dimethyl sulfoxide

ECL:

Electrochemiluminescence

EdU:

5-Ethynyl-2′-deoxyuridine

GC:

Gastric cancer

IHC:

Immunohistochemistry

NLS:

Nuclear localization sequence

OS:

Overall survival

PBS:

Fetal bovine serum

PMSF:

Phenylmethanesulfonyl fluoride

PVDF:

Polyvinylidene fluoride

TCGA:

The cancer genome atlas

TP53 Frameshift NLS :

TP53 frameshift mutations without nuclear localization sequence

WT:

Wild type

References

  1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74:229–63.

    Article  PubMed  Google Scholar 

  2. Yeoh KG, Tan P. Mapping the genomic diaspora of gastric cancer. Nat Rev Cancer. 2022;22:71–84.

    Article  CAS  PubMed  Google Scholar 

  3. Kim MP, Lozano G. Mutant p53 partners in crime. Cell Death Differ. 2018;25:161–8.

    Article  CAS  PubMed  Google Scholar 

  4. Iacopetta B. TP53 mutation in colorectal cancer. Hum Mutat. 2003;21:271–6.

    Article  CAS  PubMed  Google Scholar 

  5. van Beek EJAH, Hernandez JM, Goldman DA, Davis JL, McLoughlin KC, Ripley RT, et al. Correction to: rates of TP53 mutation are significantly elevated in African American patients with gastric cancer. Ann Surg Oncol. 2020;27:963.

    Article  PubMed  Google Scholar 

  6. Ashling W. Rezatapopt revolutionizes research directions in TP53-mutated solid tumors. Supplements and Featured Publications. 2024;1.

  7. Hassin O, Oren M. Drugging p53 in cancer: one protein, many targets. Nat Rev Drug Discov. 2023;22:127–44.

    Article  CAS  PubMed  Google Scholar 

  8. Zandi R, Selivanova G, Christensen CL, Gerds TA, Willumsen BM, Poulsen HS. PRIMA-1Met/APR-246 induces apoptosis and tumor growth delay in small cell lung cancer expressing mutant p53. Clin Cancer Res. 2011;17:2830–41.

    Article  CAS  PubMed  Google Scholar 

  9. Sallman DA, DeZern AE, Garcia-Manero G, Steensma DP, Roboz GJ, Sekeres MA, et al. Eprenetapopt (APR-246) and azacitidine in TP53-mutant myelodysplastic syndromes. J Clin Oncol. 2021;39:1584–94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Marchenko ND, Hanel W, Li D, Becker K, Reich N, Moll UM. Stress-mediated nuclear stabilization of p53 is regulated by ubiquitination and importin-alpha3 binding. Cell Death Differ. 2010;17:255–67.

    Article  CAS  PubMed  Google Scholar 

  11. Lang GA, Iwakuma T, Suh Y-A, Liu G, Rao VA, Parant JM, et al. Gain of function of a p53 hot spot mutation in a mouse model of Li-Fraumeni syndrome. Cell. 2004;119:861–72.

    Article  CAS  PubMed  Google Scholar 

  12. Olive KP, Tuveson DA, Ruhe ZC, Yin B, Willis NA, Bronson RT, et al. Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell. 2004;119:847–60.

    Article  CAS  PubMed  Google Scholar 

  13. Brosh R, Rotter V. When mutants gain new powers: news from the mutant p53 field. Nat Rev Cancer. 2009;9:701–13.

    Article  CAS  PubMed  Google Scholar 

  14. Sethi N, Kikuchi O, McFarland J, Zhang Y, Chung M, Kafker N, et al. Mutant p53 induces a hypoxia transcriptional program in gastric and esophageal adenocarcinoma. JCI Insight. 2019;4.

  15. Wang Z, Burigotto M, Ghetti S, Vaillant F, Tan T, Capaldo BD, et al. Loss-of-function but not gain-of-function properties of mutant TP53 are critical for the proliferation, survival, and metastasis of a broad range of cancer cells. Cancer Discov. 2024;14:362–79.

    Article  PubMed  Google Scholar 

  16. Rahnamoun H, Hong J, Sun Z, Lee J, Lu H, Lauberth SM. Mutant p53 regulates enhancer-associated H3K4 monomethylation through interactions with the methyltransferase MLL4. J Biol Chem. 2018;293:13234–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Wang K, Gong Z, Chen Y, Zhang M, Wang S, Yao S, et al. KDM4C-mediated senescence defense is a targetable vulnerability in gastric cancer harboring TP53 mutations. Clin Epigenetics. 2023;15:163.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Cheng LL, Itahana Y, Lei ZD, Chia N-Y, Wu Y, Yu Y, et al. TP53 genomic status regulates sensitivity of gastric cancer cells to the histone methylation inhibitor 3-deazaneplanocin a (DZNep). Clin Cancer Res. 2012;18:4201–12.

    Article  CAS  PubMed  Google Scholar 

  19. Jasmine S, Mandl A, Krueger TEG, Dalrymple SL, Antony L, Dias J, et al. Characterization of structural, biochemical, pharmacokinetic, and pharmacodynamic properties of the LSD1 inhibitor bomedemstat in preclinical models. Prostate. 2024;84:909–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Lim S, Janzer A, Becker A, Zimmer A, Schüle R, Buettner R, Kirfel J. Lysine-specific demethylase 1 (LSD1) is highly expressed in ER-negative breast cancers and a biomarker predicting aggressive biology. Carcinogenesis. 2010;31:512–20.

    Article  CAS  PubMed  Google Scholar 

  21. Chen Q, Shen P, Ge W-L, Yang T-Y, Wang W-J, Meng L-D, et al. Roundabout homolog 1 inhibits proliferation via the YY1-ROBO1-CCNA2-CDK2 axis in human pancreatic cancer. Oncogene. 2021;40:2772–84.

    Article  CAS  PubMed  Google Scholar 

  22. Jambhekar A, Anastas JN, Shi Y. Histone lysine demethylase inhibitors. Cold Spring Harb Perspect Med. 2017;7:a026484.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Gong Z, Li A, Ding J, Li Q, Zhang L, Li Y, et al. OTUD7B deubiquitinates LSD1 to govern its binding partner specificity, homeostasis, and breast cancer metastasis. Adv Sci (Weinh). 2021;8: e2004504.

    Article  PubMed  Google Scholar 

  24. Gao S, Chen S, Han D, Wang Z, Li M, Han W, et al. Chromatin binding of FOXA1 is promoted by LSD1-mediated demethylation in prostate cancer. Nat Genet. 2020;52:1011–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Jiang Q, Stachelscheid J, Bloehdorn J, Pacholewska A, Aszyk C, Grotenhuijs F, et al. Oncogenic role and target properties of the lysine-specific demethylase KDM1A in chronic lymphocytic leukemia. Blood. 2023;142:44–61.

    CAS  PubMed  Google Scholar 

  26. Liu Y-W, Xia R, Lu K, Xie M, Yang F, Sun M, et al. LincRNAFEZF1-AS1 represses p21 expression to promote gastric cancer proliferation through LSD1-Mediated H3K4me2 demethylation. Mol Cancer. 2017;16:39.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Zheng YC, Duan YC, Ma JL, Xu RM, Zi X, Lv WL, et al. Triazole-dithiocarbamate based selective lysine specific demethylase 1 (LSD1) inactivators inhibit gastric cancer cell growth, invasion, and migration. J Med Chem. 2013;56:8543–60.

    Article  CAS  PubMed  Google Scholar 

  28. Ma QS, Zhang YF, Li CY, Zhang WX, Yuan L, Niu JB, et al. Discovery of novel tranylcypromine-based derivatives as LSD1 inhibitors for gastric cancer treatment. Eur J Med Chem. 2023;251: 115228.

    Article  CAS  PubMed  Google Scholar 

  29. Liu HM, Xiong XP, Wu JW, Chen HX, Zhou Y, Ji SK, et al. Discovery of acridine-based LSD1 inhibitors as immune activators targeting LSD1 in gastric cancer. Eur J Med Chem. 2023;251: 115255.

    Article  CAS  PubMed  Google Scholar 

  30. Zhou X, Singh M, Sanz Santos G, Guerlavais V, Carvajal LA, Aivado M, et al. Pharmacologic activation of p53 triggers viral mimicry response thereby abolishing tumor immune evasion and promoting antitumor immunity. Cancer Discov. 2021;11:3090–105.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Huang J, Sengupta R, Espejo AB, Lee MG, Dorsey JA, Richter M, et al. p53 is regulated by the lysine demethylase LSD1. Nature. 2007;449:105–8.

    Article  CAS  PubMed  Google Scholar 

  32. Scoumanne A, Chen X. The lysine-specific demethylase 1 is required for cell proliferation in both p53-dependent and -independent manners. J Biol Chem. 2007;282:15471–5.

    Article  CAS  PubMed  Google Scholar 

  33. Mould DP, Alli C, Bremberg U, Cartic S, Jordan AM, Geitmann M, et al. Development of (4-Cyanophenyl)glycine derivatives as reversible inhibitors of lysine specific demethylase 1. J Med Chem. 2017;60:7984–99.

    Article  CAS  PubMed  Google Scholar 

  34. Mould DP, Bremberg U, Jordan AM, Geitmann M, McGonagle AE, Somervaille TCP, et al. Development and evaluation of 4-(pyrrolidin-3-yl)benzonitrile derivatives as inhibitors of lysine specific demethylase 1. Bioorg Med Chem Lett. 2017;27:4755–9.

    Article  CAS  PubMed  Google Scholar 

  35. Haydn T, Metzger E, Schuele R, Fulda S. Concomitant epigenetic targeting of LSD1 and HDAC synergistically induces mitochondrial apoptosis in rhabdomyosarcoma cells. Cell Death Dis. 2017;8: e2879.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Dai X-J, Liu Y, Wang N, Chen H-X, Wu J-W, Xiong X-P, et al. Novel acridine-based LSD1 inhibitors enhance immune response in gastric cancer. Eur J Med Chem. 2023;259: 115684.

    Article  CAS  PubMed  Google Scholar 

Download references

Funding

Wuxi Taihu Lake Talent Plan for Leading Talents in Medical and Health Profession, and Wuxi Medical Key Discipline (ZDXK2021002).

Author information

Authors and Affiliations

  1. Department of Gastrointestinal Surgery, Affiliated Hospital of Jiangnan University, Wuxi, 214062, Jiangsu, China

    Suzeng Wang, Kaiqing Wang, Hao Cheng & Bojian Fei

  2. Wuxi Cancer Institute, Affiliated Hospital of Jiangnan University, Wuxi, 214062, Jiangsu, China

    Suzeng Wang, Chunyu Yang, Junhui Tang, Kaiqing Wang, Hao Cheng, Surui Yao & Zhaohui Huang

  3. Laboratory of Cancer Epigenetics, Wuxi School of Medicine, Jiangnan University, Wuxi, 214122, Jiangsu, China

    Suzeng Wang, Chunyu Yang, Junhui Tang, Kaiqing Wang, Hao Cheng, Surui Yao, Zhaohui Huang & Bojian Fei

Authors
  1. Suzeng Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Chunyu Yang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Junhui Tang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Kaiqing Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Hao Cheng
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Surui Yao
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Zhaohui Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Bojian Fei
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

This project was conceived and supervised by B.F. and Z.H. The experimental work was primarily completed by S.W. and C.Y. Bioinformatics data analysis was performed by J.T. The animal experiments included participation from H.C. The experimental materials and guidance were mainly provided by S.Y. The manuscript was prepared by Z.H. and S.W. All authors reviewed the manuscript.

Corresponding authors

Correspondence to Zhaohui Huang or Bojian Fei.

Ethics declarations

Ethics approval and consent to participate

This study was performed in accordance with the Declaration of Helsinki. The animal experiments were approved by Institutional Animal Care and Use Committee of the affiliated hospital of Jiangnan University (approval number: JN.No20240930b0301207-[522]).

Consent for publication

Not applicable.

Competing interest

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, S., Yang, C., Tang, J. et al. LSD1 is a targetable vulnerability in gastric cancer harboring TP53 frameshift mutations. Clin Epigenet 17, 26 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13148-025-01829-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13148-025-01829-9

Keywords

Clinical Epigenetics

ISSN: 1868-7083

Contact us