Icariin-curcumol inhibits histone H3K18 lactylation and FOXM1 expression to enhance the sensitivity of prostate cancer cells to docetaxel

Abstract

Background

Histone lactylation has emerged as an epigenetic driver of tumor chemoresistance. Our prior work identified the phytochemical combination icariin-curcumol (Ica-Cur) as a potential therapeutic agent against docetaxel (DTX)-resistant prostate cancer (PCa). This study aimed to investigate the mechanistic link between histone lactylation and DTX resistance in PCa, and evaluates Ica-Cur’s regulatory role in this process.

Methods

DTX-resistant LNCaP/R cells were generated from parental LNCaP PCa cells. Xenograft models were established in BALB/c nude mice using both cell lines. Interventions included pharmacological modulation of glycolysis (sodium lactate [Nala], a glycolysis activator and 2-deoxy-D-glucose [2-DG], a glycolysis inhibitor) and genetic silencing of forkhead box M1 (FOXM1) via lentiviral constructs (sh-FOXM1). The enrichment of histone H3K18 lactylation (H3K18la) at the FOXM1 promoter was validated. Tumor growth, lactate levels, lactate dehydrogenase (LDH) activity, proliferation, and apoptosis were systematically analyzed.

Results

Resistant LNCaP/R models exhibited significant upregulation of H3K18la and FOXM1 compared to controls. Nala increased lactate production, enhanced H3K18la deposition, and stimulated proliferation while suppressing apoptosis. Conversely, 2-DG reduced H3K18la deposition and inhibited proliferation. FOXM1 expression was directly regulated by H3K18la, with sh-FOXM1 reducing LDH activity, inhibiting proliferation, and inducing apoptosis. Ica-Cur restored DTX sensitivity by suppressing H3K18la and FOXM1 expression.

Conclusion

These findings identify H3K18la-mediated FOXM1 activation as a novel mechanism underlying DTX resistance in PCa. Ica-Cur may represent a promising therapeutic agent by targeting lactylation-dependent epigenetic regulation and FOXM1-driven transcriptional activity, supporting its clinical potential for overcoming chemoresistance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12935-025-03927-3.

Introduction

Prostate cancer (PCa) represents the most prevalent malignancy of the male reproductive system and ranks as the most frequently diagnosed cancer among men worldwide [1]. Globally, it represents the sixth leading cause of cancer-related mortality in males, posing a significant public health burden [2]. While localized PCa is amenable to curative interventions such as surgical resection, radiotherapy and hormone therapy, advanced cases frequently evolve into castration-resistant prostate cancer (CRPC) following androgen deprivation therapy (ADT) and remains incurable [3]. Although docetaxel (DTX) serves as a cornerstone chemotherapeutic agent for PCa, its clinical utility is frequently compromised by the inevitable development of resistance, which drives disease progression toward more aggressive phenotypes [4]. Consequently, novel strategies to counteract DTX resistance are urgently needed to achieve durable therapeutic responses and improve survival outcomes.

Accumulating evidence underscores the critical role of epigenetic alterations in PCa pathogenesis [5]. Among these, histone lysine lactylation (Kla) has recently emerged as a novel post-translational modification. This process involves the enzymatic addition of lactoyl groups—derived from lactate, a metabolic byproduct of the Warburg effect— to lysine residues on histones [6]. Notably, aberrant histone lactylation has been implicated in malignant transformation and chemoresistance across multiple cancer types [7]. Lactate-mediated histone lactylation can enhance cell susceptibility to malignancy by regulating target gene activity [8]. For instance, lactate-induced histone H3K18 lactylation (H3K18la) promotes USP39 expression and deubiquitination activity, exacerbating endometrial cancer progression [9]. Histone H3K9 lactylation (H3K9la) enhances glioblastoma resistance to temozolomide via transcriptional upregulation of LUC7L2 [10]. In PCa, elevated lactate production and dysregulated histone lactylation have been documented [11]. However, the specific contribution of this epigenetic mechanism to DTX resistance remains uncharacterized, warranting systematic investigation.

Plant-derived natural products have garnered significant attention as potential anti-cancer agents, particularly for PCa, owing to their favorable safety profiles and low systemic toxicity [12]. Icariin (Ica), the principal bioactive flavonoid extracted from Epimedium brevicornu (Yinyanghuo), exhibits diverse pharmacological properties against diseases, including tumorigenesis [13], male reproductive disorders [14], and bone and joint diseases [15]. Curcumol (Cur), a sesquiterpenoid isolated from Curcuma zedoaria (Ezhu), demonstrates broad biological activities encompassing anti-inflammatory, anti-cancer, and neuroprotective functions [16]. In previous studies, our team reported that the Ica-Cur combination exerts additive therapeutic effects against PCa [17]. Further mechanistic investigations revealed that this synergy enhances DTX sensitivity by affecting the Warburg effect [18]. Building upon these findings, we hypothesize that Ica-Cur may modulate histone lactylation dynamics to attenuate target gene activation and counteract DTX resistance in PCa.

This study aimed to evaluate the therapeutic potential of Ica-Cur against DTX-resistant PCa using both in vitro and in vivo models. Our results demonstrate that the anti-resistance efficacy of Ica-Cur is mechanistically linked to H3K18la modification. Specifically, Ica-Cur suppresses forkhead box M1 (FOXM1) expression by inhibiting H3K18la enrichment at the FOXM1 promoter, thereby restoring DTX sensitivity in resistant PCa cells.

Methods

Cell culture

Human prostate cancer LNCaP cells (AW-CCH042, Abiowell, Changsha, China) were cultured in RPMI-1640 medium containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. DTX-resistant sublines (LNCaP/R) were generated through progressive exposure of parental LNCaP cells to escalating docetaxel (DTX) concentrations (0.1, 0.2, 0.5, 1, 5, 10 nM) [19], with resistance confirmed by stable proliferation in 10 nM DTX. All cells were grown in an incubator at 37℃ with 5% CO₂.

For DTX response analysis, parental LNCaP cells were treated with 10 nM DTX for 6, 12, and 24 h. To assess histone lactylation dynamics, both LNCaP and LNCaP/R cells were exposed to 10 nM DTX, 25 mM sodium lactate (Nala; HY-134545, MCE, Monmouth Junction, NJ, USA) [20], a glycolysis activator, and 20 mM 2-deoxy-D-glucose (2-DG; HY-13966, MCE) [21], a glycolysis inhibitor, for 24 h. FOXM1 functional studies utilized LNCaP/R cells transfected with FOXM1-targeting lentiviral constructs (sh-FOXM1#1–3; HonorGene, Changsha, China) using Lipofectamine 2000 (11668019, Invitrogen, CA, USA) for 48 h, followed by24 h DTX (10 nM) treatment. Target sequences are as follows: sh-FOXM#1: GAGAGTGAAAACGCAGATTCATA; sh-FOXM#2: CGCAGATTCATAATGAAAACTAG; sh-FOXM#3: GTGTTTAAGCAGCAGAAACGACC; and sh-NC (negative control): GTTAGCCGTAAGTTAAGGCCAAT. To clarify the role of Ica-Cur, LNCaP/R cells were co-treated with 10 nM DTX [19], 35 µg/mL Ica, and 25 µg/mL Cur [18] for 24 h under standardized culture conditions (37℃, 5% CO₂).

Animal models

All animal procedures received ethical approval from the Medical Ethics Committee of Hunan University of Medicine (No. 202409083). Male BALB/c nude mice (6 weeks old) were purchased from Hunan SJA Laboratory Animal Co., Ltd. and maintained under specific pathogen-free (SPF) conditions with free access to food and water. Following a 7-day acclimatization period, interventions were initiated.

LNCaP and LNCaP/R cells (2 × 10⁸ cells/mouse) were subcutaneously inoculated into the right axilla of mice, as previously described [19]. On day 5 post-inoculation, mice received daily intraperitoneal administration of 0.2 g/kg Nala [22], or oral gavage of 80 mg/kg Ica [23] and 60 mg/kg Cur [24] for consecutive 30 days. The administered doses of Ica and Cur were determined based on previously established protocols demonstrating robust pharmacological activity following 30-day treatment regimens [23, 24]. All experimental groups exhibited 100% survival with no observed neurotoxic manifestations (e.g., tremors, abnormal locomotion). Tumor volumes were recorded using vernier calipers every 3–4 days and calculated as 0.5 × length × width². On day 35, mice were euthanized by intraperitoneal injection of 150 mg/kg sodium pentobarbital. Tumors were isolated and weighed.

Immunofluorescence (IF)

Paraffin-embedded tumor sections and cell-adherent slides were subjected to IF analysis. Tissue sections underwent antigen retrieval in Tris-EDTA buffer via microwave heating, followed by gradual cooling to ambient temperature. Cultured cells were fixed in 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. After blocking with 5% bovine serum albumin (BSA), samples were incubated overnight at 4℃ with anti-Pan-Kla (1:50, PTM-1401RM, PTO BIO, Hangzhou, China), anti-LDHA (1:50, 21799-1-AP, Proteintech, Rosemont, IL, USA), and anti-H3K18la (1:50, PTM-1406RM, PTO BIO) antibodies. The next day, samples were incubated with goat anti-rabbit IgG H&L antibody (AWS0005a, Abiowell) at room temperature, with nuclear counterstaining via DAPI. Fluorescent images were acquired using a Motic BA210E microscope (Motic, Xiamen, China).

Western blot

PCa cells and tissues were lysed in radioimmunoprecipitation (RIPA) buffer (AWB0136, Abiowell). Lysates were clarified by centrifugation, and protein concentrations were quantified via bicinchoninic acid (BCA) assay (AWB0104, Abiowell). Denatured proteins (95℃, 10 min) were resolved on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrotransferred to nitrocellulose membranes. Membranes were blocked with 5% skim milk in TBST and incubated overnight at 4℃ with primary antibodies. After TBST washes, membranes were incubated with secondary antibodies for 1 h at room temperature. Chemiluminescent signals were developed using enhanced chemiluminescence (ECL) substrate (AWB0005, Abiowell) and imaged on a ChemiScope6100 system (CLiNX, Shanghai, China). Band intensities were quantified using ImageJ software (NIH, Bethesda, MD, USA). The antibodies used are presented in Table 1.

Table 1.

Antibodies used for Western blot analysis

Antibodies	Cat. no. and company	Source	Dilution ratio
LDHA	21799-1-AP, Proteintech	Rabbit	1:5000
Pan-Kla	PTM-1401RM, PTO BIO	Rabbit	1:1000
H3K18la	PTM-1427RM, PTO BIO	Rabbit	1:1000
H3K23la	PTM-1413RM, PTO BIO	Rabbit	1:2000
H3K9la	PTM-1419RM, PTO BIO	Rabbit	1:2000
H4K5la	PTM-1407RM, PTO BIO	Rabbit	1:1000
H4K8la	PTM-1415RM, PTO BIO	Rabbit	1:1000
H4K12la	ab177793, Abcam	Rabbit	1:10000
Histone H4	PTM-1009, PTO BIO	Rabbit	1:1000
Histone H3	ab1791, Abcam	Rabbit	1:1000
FOXM1	13147-1-AP, Proteintech	Rabbit	1:2000
β-actin	AWA80002, Abiowell	Rabbit	1:5000

Cell counting kit-8 (CCK8) assay

Cells (5 × 10³/well) were seeded in 96-well plates. CCK8 reagent (AWC0114a, Abiowell; 10 µL/well) was added, followed by 4 h incubation. Absorbance (OD value) at 450 nm was measured using a MB-530 microplate reader (HEALES, Shenzhen, China). Triplicate technical replicates were performed for each condition.

EdU (5-Ethynyl-2’-deoxyuridine) staining

Proliferation was assessed using the Cell-Light EdU Apollo567 Kit (C10310, RiboBio, Guangzhou, China). Cells were pulsed with 50 µM EdU overnight, fixed in 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. Apollo staining (30 min) and Hoechst 33,342 nuclear counterstaining (30 min) were performed sequentially. Images were captured using a Motic BA210E fluorescence microscope.

Lactate and LDH quantification

Lactate and lactate dehydrogenase (LDH) levels were measured using commercial kits (A019-2-1 and A020-2, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following manufacturer protocols. Reactions were incubated at 37℃ in a water bath, and absorbance was recorded using a HEALES MB-530 spectrophotometer.

Flow cytometry

Apoptosis was analyzed via Annexin V-APC/PI dual staining (KGA1030, KeyGen BioTECH, Nanjing, China). Cells were resuspended in binding buffer and stained with Annexin V-allophycocyanin (APC) and propidium iodide (PI) in the dark. Fluorescence was quantified using a Beckman Coulter CytoFLEX flow cytometer (Beckman Coulter, Fullerton, CA, USA).

Quantitative real-time PCR (qRT-PCR)

Total RNA was extracted with Trizol (15596026, Thermo Scientific, Portsmouth, NH, USA), quantified via Micro Drop ultra-micro spectrophotometer (BIO-DL, Shanghai, China), and reverse-transcribed using HiFiScript cDNA Synthesis Kit (CW2569M, CWBIO, Taizhou, China). qRT-PCR was performed on a QuantStudio1 System (Thermo Scientific) with UltraSYBR Mixture (CW2601M, CWBIO). Relative mRNA expression was calculated using the 2^−ΔΔCt method, normalized to β-actin. Primer sequences are provided in Table 2.

Table 2.

Primer sequences for qRT-PCR

Gene	Forward (5’−3’)	Reverse (5’−3’)
Human FOXM1	GTTCTGATGGACTGGGCTCC	ACCTTAACCTGTCGCTGCTC
Human β-actin	ACCCTGAAGTACCCCATCGAG	AGCACAGCCTGGATAGCAAC
Mouse FOXM1	ACCAGTCAGGAAGCATCACC	TGCTCAGGACAACCTGAAAGG
Mouse β-actin	ACATCCGTAAAGACCTCTATGCC	TACTCCTGCTTGCTGATCCAC

Chromatin immunoprecipitation (ChIP)-qPCR

Chromatin from 3 × 10⁶ cells was cross-linked with 1.1% formaldehyde, quenched with excess glycine, and sonicated to fragments. DNA fragments were verified by 1.5% agarose gel. Immunoprecipitation used ChIP-grade anti-H3K18la antibody (PTM-1427RM, PTO BIO) and a commercial ChIP kit (ab500, Abcam, Cambridge, UK). Purified DNA was analyzed by qPCR for FOXM1 promoter enrichment.

Dual-luciferase (LUC) reporter assay

The FOXM1 promoter was cloned into psiCHECK-2 (HonorGene). LNCaP/R cells were transfected with the reporter plasmid and treated with or without Nala or 2-DG for 24 h. Firefly and Renilla luciferase activities were measured using the Dual-LUC kit (E1910, Promega, Madison, WI, USA). Promoter activity was expressed as Firefly/Renilla ratio.

Statistical analysis

Data were presented as mean ± standard deviation (SD) and were statistically analyzed using GraphPad Prism 9.0. The unpaired t-test was used for comparisons between two groups. One- or two-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was used for comparisons between multiple groups. P < 0.05 was considered significantly different.

Results

DTX-resistant PCa cells exhibit elevated histone lactylation levels

To investigate the mechanisms underlying DTX resistance in PCa, we established LNCaP and DTX-resistant LNCaP/R xenograft models. Tumor tissues from LNCaP/R mice demonstrated significantly higher levels of pan-lysine lactylation (pan-Kla) and lactate dehydrogenase A (LDHA) compared to those from LNCaP mice (Fig. 1A and B), suggesting a strong association between DTX resistance and enhanced histone lactylation in vivo.

Fig. 1.

DTX-resistant PCa cell lines exhibit elevated lactylation levels. A IF staining of tumor sections from LNCaP and LNCaP/R xenograft mice using pan-Kla and LDHA antibodies. Scale bar = 25 μm. B Western blot analysis of pan-Kla and LDHA in tumor tissues. ***P < 0.001 vs. LNCaP-Tumor. LNCaP and LNCaP/R cells were treated with 25 mM Nala for 24 h. C Western blot analysis of pan-Kla levels in LNCaP and LNCaP/R cells. D IF staining of pan-Kla and LDHA in LNCaP and LNCaP/R cells. Scale bar = 25 μm. E Lactate and lactate dehydrogenase (LDH) levels in LNCaP and LNCaP/R cells. F CCK8 assay evaluating cell viability. G EdU staining to assess proliferation. Scale bar = 50 μm. H Apoptosis analysis by flow cytometry. ***P < 0.001 vs. LNCaP + DTX. ###P < 0.001 vs. LNCaP/R + DTX

To elucidate the mechanisms linking DTX therapy to histone lactylation, LNCaP cells were treated with DTX (10 nM, 24 h). In vitro analyses demonstrated that DTX treatment robustly suppressed LNCaP cell proliferation, reduced pan-Kla and LDHA expression, and decreased LDH activity and lactate accumulation (Figure S1A-S1D). These data collectively indicate that DTX treatment attenuates histone lactylation in PCa cells, potentially contributing to its anti-tumor efficacy.

To explore the functional role of histone lactylation in DTX resistance, we supplemented DTX-treated LNCaP and LNCaP/R cells with Nala, a lactylation inducer. DTX-treated LNCaP/R cells exhibited baseline elevations in pan-Kla, LDHA, lactate, and LDH levels compared to DTX-treated LNCaP cells (Fig. 1C-E). Nala treatment further amplified these parameters in both cell lines (Fig. 1C-E), concomitant with enhanced proliferation (Fig. 1F and G) and reduced apoptosis (Fig. 1H). Strikingly, LNCaP/R cells displayed intrinsic resistance to apoptosis, which was exacerbated by Nala-mediated lactylation (Fig. 1F-H). These results demonstrate that hyper-lactylation drives proliferative advantage and apoptotic resistance in DTX-resistant PCa cells.

Targeting lactylation restores DTX sensitivity in resistant PCa cells

To interrogate the therapeutic potential of lactylation inhibition, we treated DTX-treated LNCaP and LNCaP/R cells with 2-DG, a glycolysis inhibitor. 2-DG administration significantly suppressed pan-Kla and LDHA expression in both DTX-treated cell lines (Fig. 2A and B), paralleled by reduced lactate and LDH levels (Fig. 2C). Functionally, 2-DG attenuated proliferation (Fig. 2D and E) and induced apoptosis (Fig. 2F) in DTX-exposed LNCaP/R cells, effectively reversing their aggressive phenotype to levels comparable to parental LNCaP cells (Fig. 2). These findings establish that glycolytic inhibition disrupts lactylation-dependent survival mechanisms, thereby resensitizing resistant cells to DTX.

Fig. 2.

Lactylation inhibition reduces DTX resistance in PCa cells. LNCaP and LNCaP/R cells were treated with 20 mM 2-DG for 24 h. A Western blot analysis of pan-Kla and LDHA. B IF staining of pan-Kla and LDHA. Scale bar = 25 μm. C Lactate and LDH level measurements. D CCK8 assay for viability assessment. E EdU staining to assess proliferation. Scale bar = 50 μm. F Apoptosis analysis by flow cytometry. *P < 0.05; **P < 0.01; ***P < 0.001 vs. LNCaP + DTX. #P < 0.05; ##P < 0.01; ###P < 0.001 vs. LNCaP/R + DTX

H3K18la hyperactivation is associated with DTX resistance in PCa

Comparative analysis of tissues revealed elevated lactate and LDH levels in tumors relative to para-carcinoma tissues, with maximal accumulation observed in DTX-resistant LNCaP/R tumors (Figure S2A and B). Consistent with this, serum lactate and LDH levels were significantly higher in LNCaP/R-bearing mice compared to LNCaP controls (Figure S2C and D), establishing a systemic correlation between DTX resistance and hyper-lactate metabolism in PCa. To identify lactylation sites mechanistically linked to resistance, we profiled histone lactylation marks (H3K9la, H3K18la, H3K23la, H4K5la, H4K8la, H4K12la) across tumors and matched para-carcinoma tissues. Among these, H3K18la exhibited the most pronounced upregulation in tumors (Fig. 3A), a finding corroborated by IF staining showing elevated H3K18la intensity in tumor tissues versus matched para-carcinoma regions (Fig. 3B and C). Strikingly, LNCaP/R tumors displayed the highest H3K18la levels, whereas serum H3K18la remained unchanged across models (Figure S2E), suggesting tissue-specific lactylation dysregulation.

Fig. 3.

H3K18la levels are upregulated in PCa and modulated by DTX. A Western blot analysis of site-specific histone lactylation in tumor and para-carcinoma tissues from LNCaP and LNCaP/R xenograft mice. B-C. IF staining of H3K18la in tumor tissues and para-carcinoma tissues. Scale bar = 25 μm. **P < 0.01; ***P < 0.001 vs. LNCaP-Tumor. ###P < 0.001 vs. LNCaP/R-Tumor. D Western blot analysis of site-specific histone lactylation in LNCaP and LNCaP/R cells. E IF staining of H3K18la in LNCaP and LNCaP/R cells. Scale bar = 25 μm. ***P < 0.001 vs. LNCaP

In vitro validation using LNCaP and LNCaP/R cells confirmed that H3K18la was the most differentially regulated lactylation site between sensitive and resistant cells (Fig. 3D). IF quantification further demonstrated an increase in H3K18la levels in LNCaP/R cells compared to parental LNCaP (Fig. 3E). Notably, DTX treatment (10 nM) induced time-dependent suppression of H3K18la in LNCaP cells, with maximal reduction observed at 24 h (Figure S2F). These data collectively implicate H3K18la hyperactivation as a hallmark of DTX resistance and a potential target for therapeutic modulation.

Histone lactylation activates FOXM1 transcription in PCa

FOXM1 dysregulation is closely associated with DTX resistance and Warburg effect in PCa [25, 26]. To delineate the epigenetic regulation of FOXM1, ChIP-qPCR analysis revealed that DTX treatment (10 nM, 24 h) significantly reduces the enrichment of H3K18la at the FOXM1 promoter region in both LNCaP and LNCaP/R cell lines (Fig. 4A). This observation suggests that DTX-mediated suppression of FOXM1 gene expression is mechanistically linked to the attenuation of H3K18la levels.

Fig. 4.

Histone lactylation regulates FOXM1 expression in LNCaP/R cells. A ChIP-qPCR analysis of H3K18la enrichment in the FOXM1 promoter region after DTX treatment. ***P < 0.001 vs. LNCaP. ###P < 0.001 vs. LNCaP/R. B ChIP-qPCR analysis of H3K18la enrichment in the FOXM1 promoter region after Nala or 2-DG treatment. C Dual-LUC analysis of FOXM1 promoter activity after Nala or 2-DG treatment. D-E. qRT-PCR D and western blot E of FOXM1 mRNA expression and protein levels in LNCaP/R cells after Nala or 2-DG treatment. *P < 0.05; **P < 0.01; ***P < 0.001 vs. LNCaP/R

To further validate the lactylation-dependent regulation of FOXM1, we modulated lactylation levels using 2-DG and Nala. In LNCaP/R cells, H3K18la enrichment at the FOXM1 promoter decreased upon 2-DG treatment but increased with Nala supplementation (Fig. 4B). Consistent with these findings, Dual-LUC assays confirmed that 2-DG attenuated FOXM1 promoter activity, whereas Nala enhanced it (Fig. 4C). Correspondingly, qRT-PCR and Western blot analyses showed that 2-DG downregulated both FOXM1 mRNA and protein levels, while Nala upregulated FOXM1 expression in LNCaP/R cells (Fig. 4D). Collectively, these data demonstrate that histone lactylation directly activates FOXM1 transcription in DTX-resistant PCa cells.

FOXM1 drives proliferation and LDH release in DTX-resistant PCa cells

Given the lactylation-dependent regulation of FOXM1, we investigated its functional role in sustaining DTX resistance. Transfection of FOXM1-specific shRNAs (sh-FOXM1#1–3) into LNCaP/R cells achieved efficient FOXM1 knockdown, with sh-FOXM1#3 exhibiting the strongest suppression (Fig. 5A). Notably, FOXM1 silencing did not alter H3K18la levels (Fig. 5B), confirming that lactylation acts upstream of FOXM1 regulation. Functional assays revealed that FOXM1 knockdown significantly reduced LDH activity (Fig. 5C), suppressed cell proliferation (Fig. 5D and E) and increased apoptosis rates (Fig. 5F) in LNCaP/R cells. These findings establish FOXM1 as a critical mediator of proliferative survival and LDH release in DTX-resistant PCa.

Fig. 5.

FOXM1 modulates proliferation and lactylation in LNCaP/R cells. A qRT-PCR and western blot of FOXM1 silencing efficiency in LNCaP/R cells. B Western blot analysis of H3K18la levels in LNCaP/R cells. C LDH level quantificationin LNCaP/R cells. D CCK8 assay for viability. E EdU staining to assess proliferation. Scale bar = 50 μm. F Apoptosis analysis by flow cytometry. *P < 0.05; **P < 0.01; ***P < 0.001 vs. LNCaP/R + DTX + sh-NC

Ica-Cur inhibits histone lactylation and FOXM1 expression

Through in vivo and in vitro analysis, we evaluated the therapeutic potential of Ica-Cur in modulating histone lactylation and overcoming DTX resistance. In DTX-resistant LNCaP/R cells, co-treatment with DTX (10 nM), Ica (35 µg/mL), and Cur (25 µg/mL) for 24 h significantly inhibited proliferation (Figures S3A-S3B), paralleled by a marked reduction in LDH activity (Figure S3C). Critically, Ica-Cur downregulated both pan-Kla and H3K18la levels (Figure S3D), accompanied by suppression of FOXM1 mRNA and protein expression (Figure S3E). These results establish that Ica-Cur disrupts the lactylation-FOXM1 axis, thereby attenuating proliferative capacity in DTX-resistant PCa cells.

To validate these findings in vivo, LNCaP/R xenograft models were treated with Ica-Cur or/and Nala. Tumors exhibiting hyper-lactylation (induced by Nala) demonstrated accelerated growth kinetics, which was robustly inhibited by Ica-Cur treatment (Fig. 6A-6C). No significant intergroup differences in body weight were detected throughout the study (Fig. 6D). IF and Western blot analyses revealed that Nala elevated pan-Kla, H3K18la, and LDHA levels in tumors, consistent with enhanced lactylation-driven metabolic reprogramming. Strikingly, Ica-Cur not only reduced baseline lactylation markers but also reversed Nala-induced hyper-lactylation (Fig. 6E and F). Concomitantly, Ica-Cur monotherapy suppressed FOXM1 expression in tumors, while combinatorial treatment with Nala and Ica-Cur abolished the pro-tumorigenic effects of Nala on FOXM1 upregulation (Fig. 6G). These data demonstrate Ica-Cur reverses lactylation-mediated DTX resistance in vivo.

Fig. 6.

Ica-Cur and histone lactylation regulate DTX sensitivity in vivo. A Xenograft tumors in BALB/c nude mice 35 days after subcutaneous injection of LNCaP/R cells. B Tumor volume calculated as 0.5 × length × width². C Tumor weight at day 35. D Body weight monitoring. E IF staining of pan-Kla, H3K18la, and LDHA in tumor sections. Scale bar = 25 μm. F Western blot analysis of pan-Kla in tumor tissues. G qRT-PCR and western blot of FOXM1 mRNA and protein levels in tumors. *P < 0.05; **P < 0.01; ***P < 0.001 vs. LNCaP/R. &&&P < 0.001 vs. Nala. #P < 0.05; ##P < 0.01; ###P < 0.001 vs. Ica-Cur

Discussion

Plant-derived natural products have emerged as promising candidates for developing novel therapeutic agents in cancer prevention and treatment [27]. These phytochemicals act through multiple interconnected pathways to inhibit carcinogen activation, suppress the growth and metastasis of cancer cells, and counteract drug resistance [28]. For example, the flavonoid luteolin has been demonstrated to inhibit androgen receptor activity, eliminate PCa stem cells, suppress tumor growth, and induce apoptosis [29]. Similarly, the sesquiterpene lactone artesunate enhances DTX sensitivity in PCa cells and promotes ferroptosis when administered with DTX [30]. Building on our previous findings regarding Ica-Cur’s ability to mitigate DTX resistance, this study investigates its underlying mechanism of action. Our results reveal a strong correlation between elevated H3K18la levels and DTX resistance in PCa. Ica-Cur treatment decreased H3K18la enrichment, induced apoptosis, and inhibited proliferation through FOXM1 downregulation, ultimately enhancing DTX sensitivity in PCa models.

The Warburg effect, a pathological hallmark of cancer metabolism, drives metabolic reprogramming to sustain the bioenergetic and biosynthetic demands of rapidly proliferating tumor cells [31]. Tumor microenvironment-derived lactate (imported via monocarboxylate transporters, MCTs) or endogenously generated lactate via the Warburg effect (mediated by LDH) can modulate chromatin-mediated gene expression through histone lactylation [32]. Elevated intratumoral lactate and LDH levels correlate strongly with Gleason grade, cancer aggressiveness, and poor prognosis in PCa [33–35]. Lactate accumulation promotes tumor progression through multifaceted mechanisms, including epithelial-mesenchymal transition (EMT), metastasis, immune escape, and chemoresistance [36]. Lactate functions as a pivotal oncometabolite within tumor microenvironment, orchestrating lipid metabolic rewiring and potentiating invasive phenotypes [37]. Emerging evidence implicates histone lactylation, a lactate-dependent post-translational modification, as a novel epigenetic modulator of oncogenic signaling and therapeutic resistance [38, 39]. Lactate depletion induces histone lactylation deficiency, which subsequently enhances macrophage-mediated phagocytosis of PCa cells [40]. Lactylation-associated differential genes have been proposed as robust prognostic biomarkers for PCa [41, 42]. Prior studies have also documented upregulated histone lactylation and H3K18la in human PCa cells in response to lactate accumulation [11, 43]. In this study, our data demonstrated that DTX-resistant PCa models exhibit heightened histone lactylation and H3K18la levels. Modulation of histone lactylation directly influenced cell proliferation and apoptotic susceptibility in PCa models. Experimental induction of lactylation via Nala significantly promoted cellular proliferation and suppressed apoptotic rates in both LNCaP and LNCaP/R cell lines. Conversely, inhibition of lactylation via 2-DG attenuated proliferative capacity and increased apoptosis in these models. Critically, lactylation induction via Nala exacerbated LNCaP/R xenograft tumor growth in vivo. These suggest a potential mechanistic link between histone lactylation and DTX chemoresistance. Targeting histone lactylation may offer a novel strategy to improve DTX’s clinical efficacy in PCa.

Emerging evidence indicates that natural compounds play crucial roles in regulating histone lactylation during disease pathogenesis. For instance, 20(S)-ginsenoside Rh2 targets METTL3 to reduce lactylation levels and overcome resistance to all-trans retinoic acid in acute myeloid leukemia cells [44]. Andrographolide inhibits p300 to decrease lactate production and H3Kla modifications, thereby mitigating calcific aortic valve disease [45]. Fargesin suppresses non-small cell lung cancer progression by binding to PKM2, which attenuates aerobic glycolysis and reduces H3Kla modifications [46]. Evodiamine inhibits HIF-1α lactylation to disrupt angiogenesis and ferroptosis, thereby restraining PCa progression [47]. In our study, Ica-Cur treatment not only suppressed tumor growth but also reduced histone lactylation, counteracting the effects of Nala. These findings suggest that Ica-Cur enhances PCa sensitivity to DTX by modulating histone lactylation dynamics.

This study identified elevated H3K18la levels in DTX-resistant PCa cells. Prior research has implicated H3K18la in cancer progression and metastasis. For example, H3K18la enrichment at gene promoters activates transcription of transmembrane nucleoporins, impairing CD8 + T cell function and promoting PD-L1-mediated immune escape in non-small cell lung cancer [48]. Similarly, H3K18la modifications drive transcriptional activation of immunoglobulin superfamily members, facilitating epithelial-mesenchymal transition and metastasis in gastric cancer [49]. In breast cancer, LDHA-mediated H3K18la modification establishes a feedforward loop that accelerates tumorigenesis [50]. Our work revealed FOXM1 as a key downstream target of H3K18la modification in LNCaP/R cells. H3K18la specifically activated FOXM1 transcription in LNCaP/R cells, with lactylation enhancing promoter activity and H3K18la enrichment at the FOXM1 promoter. Conversely, reducing histone lactylation suppressed FOXM1 transcription, decreased promoter activity, and diminished H3K18la enrichment. These results confirmed that H3K18la directly targets FOXM1 to promote its transcriptional activation. Furthermore, in vivo experiments demonstrated that Ica-Cur reduces H3K18la levels and inhibits FOXM1 activity.

However, this study also has some limitations. Comprehensive systemic toxicological evaluations of Ica and Cur—including pharmacokinetic half-life monitoring, hematological parameters, serum biochemical profiling, and histopathological analyses—will be prioritized in subsequent investigations. The clinical relevance of histone lactylation levels to PCa prognosis and patient survival remains unexamined, and clinical samples were not collected to validate findings from cell and animal models. Additionally, specific deletion of the H3K18la site was not performed to investigate its functional consequences. Although the regulatory relationship between H3K18la and FOXM1 was established, further studies are required to determine whether H3K18la enhances PCa DTX sensitivity via FOXM1. The role of lactylation-mediated post-translational modifications in PCa initiation, metastasis, and chemotherapy resistance remains largely unexplored, warranting extensive investigation.

In conclusion, this study advances our understanding of Ica-Cur’s mechanism in enhancing PCa DTX sensitivity. Elevated histone lactylation levels were strongly linked to DTX resistance in PCa, while Ica-Cur treatment reduced H3K18la modifications and suppressed FOXM1 activity, thereby overcoming DTX resistance through epigenetic modulation. These findings position Ica-Cur as a promising therapeutic candidate for addressing chemotherapy resistance in PCa.

Author contributions

Data curation, Formal analysis, and Visualization: WX, YL, TT, XY, XH, LLi, CZ, YC, LLiu, MF, and HD; Methodology: YL; Investigation and Supervision: QH; Conceptualization and Writing-original draft: WX; Resources and Writing-review & editing: WS. All authors agree to be accountable for the content of the work.

Funding 

This study was supported by the Project of Traditional Chinese Medicine Administration of Hunan Province (No. B2023034), Doctoral Scientific Research Starting Foundation of Hunan University of Medicine (No. 202409), National Natural Science Foundation of China (No. 82405421), Hunan Provincial Hygiene and Health Commission Health Research Project (No. W20243165), Hunan provincial innovation and entrepreneurship training program for college students (No. S202412214034), Key Project of Hunan Provincial Department of Education (No. 24A0765), and Hunan Provincial Furong Plan Young Talents for Science and Technology Innovation (No. 2025JJ70415).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Medical Ethics Committee of Hunan University of Medicine (No. 202409083).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.