Ginsenoside Rh2 Suppresses the Fanconi Anemia Pathway by Inhibiting NF-κB-Mediated FANCL Transcription in Bladder Cancer

Abstract

Objectives

Cisplatin, a frontline chemotherapeutic agent for bladder cancer (BC), induces DNA interstrand crosslinks that are primarily repaired through the Fanconi anemia (FA) pathway. Hyperactivation of this repair mechanism contributes to cisplatin resistance, underscoring the need for FA-targeted sensitizers. This study investigated the effect of ginsenoside Rh2 on FA signaling and cisplatin sensitivity in bladder cancer cells.

Methods

Bladder cancer cell lines (T24, 5637, and RT4) were treated with cisplatin, with or without Rh2 pretreatment. FANCI/FANCD2 (ID2) complex monoubiquitination, FANCD2 foci formation, and interactions with downstream repair proteins (FANCP, FANCQ, PCNA) were examined. FANCL expression was analyzed at the transcriptional level, and rescue experiments were performed by FANCL overexpression. NF-κB signaling involvement was assessed using pharmacological agonists. A T24 xenograft model was used to validate in vivo efficacy.

Results

Cisplatin induced ID2 complex monoubiquitination, confirming FA pathway activation. Rh2 pretreatment abolished this modification and reduced FANCD2 foci formation, leading to persistent interstrand crosslinks without affecting intrastrand repair. Rh2 disrupted FANCD2- FANCP/FANCQ/PCNA interactions and selectively suppressed FANCL transcription. Overexpression of FANCL restored ID2 monoubiquitination despite Rh2 exposure. NF-κB agonists reversed Rh2-induced FANCL downregulation and FA inhibition. In vivo, Rh2 combined with cisplatin significantly reduced tumor growth in T24 xenografts, whereas NF-κB stimulation counteracted this effect.

Conclusion

Ginsenoside Rh2 suppresses NF-κB signaling to transcriptionally downregulate FANCL, thereby impairing FA pathway-mediated DNA repair and enhancing cisplatin cytotoxicity in bladder cancer. These findings highlight Rh2 as a potential combinatorial agent to overcome platinum resistance.

Graphical Abstract

Introduction

Bladder cancer (BC) ranks among the most prevalent malignancies of the urinary tract and is notorious for its high recurrence rate and resistance to treatment. ¹ Although surgical resection, chemotherapy, radiotherapy, and immunotherapy have advanced in recent years, outcomes for patients with advanced or muscle-invasive disease remain unsatisfactory. ² Platinum-based regimens-most notably cisplatin-continue to represent the standard first-line option; however, a substantial proportion of patients either fail to respond initially or develop resistance over the course of therapy.^3,4 This highlights an urgent need to elucidate the molecular basis of therapeutic resistance and to devise strategies that can improve treatment efficacy.

The Fanconi anemia (FA) pathway is a key DNA damage response system that safeguards genomic integrity by repairing interstrand crosslinks, such as those induced by platinum compounds.^5,6 Upon DNA damage, pivotal FA proteins including ID2 complex are monoubiquitinated, triggering downstream homologous recombination repair. In bladder cancer, heightened FA pathway activity has been linked to increased DNA repair capacity and reduced sensitivity to DNA-damaging therapies, ⁷ implicating this pathway as a promising target for overcoming chemoresistance.

Ginsenoside Rh2 (Rh2), a bioactive component of ginseng, has attracted increasing attention for its potential to enhance the efficacy of conventional cancer therapies. ⁸ Beyond its intrinsic antitumor activities-such as inhibiting proliferation, inducing apoptosis and autophagy, and suppressing metastasis-Rh2 has been shown to sensitize tumor cells to both radiotherapy and chemotherapy. ⁹ Preclinical studies demonstrate that Rh2 augments the cytotoxic effects of ionizing radiation, ¹⁰ in part by suppressing pro-survival signaling pathways such as NF-κB ¹¹ and Wnt/β-catenin, ¹² thereby amplifying radiation-induced DNA damage and apoptotic signaling. In chemotherapy, Rh2 has been reported to potentiate the effects of agents including cisplatin, doxorubicin, paclitaxel, and oxaliplatin, often reversing intrinsic or acquired drug resistance.^13-16 Mechanistically, Rh2 can inhibit key resistance-associated pathways such as PI3K/Akt/mTOR and NF-κB, down-regulate anti-apoptotic proteins, and modulate drug efflux pumps, leading to enhanced drug-induced cell death. Importantly, Rh2 influences the DNA damage response by increasing γH2AX accumulation, activating ATM/ATR signaling, and potentially attenuating homologous recombination efficiency, thereby reducing DNA repair capacity and heightening sensitivity to DNA-damaging treatments. ¹⁷ These findings support the rationale for developing Rh2 as an adjuvant to improve therapeutic outcomes in cancer.

In this study, we have identified a previously unrecognized mechanism by which Rh2 enhances cisplatin sensitivity of BC to DNA-damaging treatments. Specifically, we show that Rh2 down-regulates the transcriptional expression of FANCL, the E3 ubiquitin ligase within the Fanconi anemia (FA) core complex, thereby impairing monoubiquitination of the FANCD2-FANCI (ID2) complex and attenuating FA pathway activity. Given that the FA pathway is a critical determinant of interstrand crosslink repair and chemoresistance, this finding expands the current understanding of Rh2’s molecular targets and positions FANCL suppression as a novel strategy to disrupt DNA repair and overcome therapy resistance.

Materials and Methods

Study Duration

All experiments started from January 2022 to December 2024.

Cell Culture

The human bladder cancer cell lines T24 (ATCC HTB-4), 5637 (ATCC HTB-9), and RT4 (ATCC HTB-2) were obtained from the American Type Culture Collection (ATCC, USA). T24 and 5637 cells were maintained in RPMI-1640 medium (Gibco, Thermo Fisher Scientific, USA) supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco) and 1% penicillin–streptomycin (Gibco). RT4 cells were cultured in McCoy’s 5A medium (Gibco) supplemented with 10% FBS and 1% penicillin–streptomycin. All cell lines were incubated at 37°C in a humidified atmosphere containing 5% CO₂ and were authenticated by short tandem repeat (STR) profiling within the last 3 years. Mycoplasma contamination was routinely tested and confirmed to be negative prior to use.

For drug treatments, cisplatin (Sigma-Aldrich, USA, Cat. P4394) was used at 10 μM for 12 h. Rh2 (Sigma-Aldrich, USA, Cat. G0550) was applied as a 24 h pretreatment at 20 μM prior to subsequent assays. The USP1 inhibitor ML323 (MedChemExpress, China, Cat. HY-12782) was used at 5 μM for 4 h. Actinomycin D (Sigma-Aldrich, Cat. A1410) was applied at 5 μg/mL for the indicated times to block transcription. Cycloheximide (CHX; Sigma-Aldrich, Cat. C7698) was used at 50 μg/mL for protein stability assays. MG132 (MedChemExpress, Cat. HY-13259) was used at 10 μM for 6 h to inhibit proteasomal degradation. SMAC mimetic AZD5582 (Sigma-Aldrich, Cat. SML2900) was used at 1 μM for 4 h to stabilize NF-κB-inducing kinase, GSK3 inhibitor CHIR99021 (Sigma-Aldrich, Cat. SML1046) was used at 5 μM for 12 h to stabilize β-catenin. SC79 (Sigma-Aldrich, Cat. 123871) was used at 2 μg/mL for 6 h to phosphorylate Akt at Thr308. Anisomycin (Sigma-Aldrich, Cat. A9789) was used at 1 µg/mL for 12 h to trigger the ribotoxic-stress-JNK axis. Colivelin (MedChemExpress, Cat. HY-P1061) was used at 1 µM for 6 h to activate STAT3. For USP1 knockdown, siRNA duplexes targeting human USP1 ¹⁸ (5’ -UCG​GCA​AUA​CUU​GCU​AUC​UUA -3′, GenePharma, China) were transfected at a final concentration of 50 nM using Lipofectamine RNAiMAX (Invitrogen, Thermo Fisher Scientific, Cat. 13778150) according to the manufacturer’s instructions. Cells were harvested at the indicated times after treatment for protein or RNA analyses.

Western Blot

Cell lysates were prepared in RIPA buffer and protein concentrations measured using the BCA assay (Thermo Fisher Scientific, Cat. 23227). Equal amounts (40 µg) of protein were separated by SDS-PAGE, transferred to PVDF membranes, and blocked with 5% non-fat milk in TBST. Membranes were incubated with the following primary antibodies at 1:2000 dilution overnight at 4°C: FANCI (Thermo Fisher Scientific, Cat. A300-212 A), FANCD2 (Thermo Fisher Scientific, Cat. MA1-16570), FANCP/SLX4 (Thermo Fisher Scientific, Cat. PA5-45039), FANCQ/XPF (Thermo Fisher Scientific, Cat. PA5-117118), PCNA (Thermo Fisher Scientific, Cat. 13-3900), FANCA (Thermo Fisher Scientific, Cat. 14657), FANCC (Thermo Fisher Scientific, Cat. H00002176-B01P), FANCM (Thermo Fisher Scientific, Cat. PA5-68191), FANCF (Abcam, USA, Cat. ab105266), FAAP24 (Abcam, Cat. ab122800), FAAP100 (Abcam, Cat. ab224178), FANCL (Abcam, Cat. ab272618), FANCB (Abcam, Cat. ab84536), USP1 (Cell Signaling Technology, USA, Cat. 4933), UAF1/WDR48 (Abcam, Cat. ab97343), GAPDH (Cell Signaling Technology, Cat. 2118), p65 (Cell Signaling Technology, Cat. 8242), p-p65 Ser536 (Cell Signaling Technology, Cat. 3033), β-catenin (Cell Signaling Technology, Cat. 8480), p-β-catenin Ser37 (Cell Signaling Technology, Cat. 9561), Akt (Cell Signaling Technology, Cat. 9272), p-Akt Thr308 (Cell Signaling Technology, Cat. 9275), JNK (Cell Signaling Technology, Cat. 9252), p-JNK Thr183/Tyr185 (Cell Signaling Technology, Cat. 4671), STAT3 (Cell Signaling Technology, Cat. 4904), and p-STAT3 Tyr705 (Cell Signaling Technology, Cat. 9145). After washing, membranes were incubated with HRP-conjugated secondary antibodies (anti-rabbit IgG or anti-mouse IgG, depending on host species; Cell Signaling Technology, Cat. 7074 for rabbit, Cat. 7076 for mouse) at 1:5000 dilution for 1 h at room temperature. Bands were visualized using enhanced chemiluminescence (Promega, USA, Cat. W1001) detection and quantified by densitometry.

FANCD2 Foci Immunofluorescence Assay

Cells were seeded on sterile glass coverslips and subjected to cisplatin treatments. Where indicated, ice-cold cytoskeleton buffer (10 mM PIPES [pH 6.8], 100 mM NaCl, 300 mM sucrose, 3 mM MgCl₂, 1 mM EGTA, 0.5% Triton X-100, 1 mM DTT, 1× protease/phosphatase inhibitors) was used to remove soluble proteins by 2 min, then 4% paraformaldehyde (PFA) in PBS was used for immediate fixation in for 10 min at room temperature. After three PBS washes, cells were permeabilized with 0.5% Triton X-100 in PBS for 5 min and blocked in 5% BSA in PBS for 1 h. Coverslips were incubated overnight at 4°C with anti-FANCD2 primary antibody (1:1000 in blocking buffer). After PBS washes, cells were incubated with Alexa Fluor-conjugated secondary antibody (goat anti-mouse IgG, Invitrogen, Thermo Fisher Scientific, Cat. A-11005, 1:1,000, 1 h, room temperature), counterstained with DAPI (Sigma-Aldrich, Cat. D9542), and mounted using ProLong Gold Antifade Mountant (Thermo Fisher Scientific, Cat. P36930). Images were acquired on a laser-scanning confocal microscope (60×/63× oil objective) using matched exposure settings across conditions; z-stacks (0.4 µm steps) were collected and maximum-intensity projections generated in ImageJ. Quantification was performed blinded on ≥100-200 nuclei per condition from at least three independent experiments; cells with ≥5 FANCD2 foci/nucleus were scored as foci-positive, and both percentage of foci-positive cells and mean foci per nucleus were reported. Statistical analysis was carried out using two-tailed t-tests or one-way ANOVA with appropriate post hoc tests, as specified in the figure legends.

Long-Fragment Quantitative PCR (qPCR Blocking) Assay

The formation and repair kinetics of DNA interstrand crosslinks were quantified using this method. Briefly, genomic DNA was extracted from bladder cancer cells at the 0, 6, and 24 h time points after treatment and quantified spectrophotometrically. Genomic DNA from treated and untreated (NC) bladder cancer cells was purified and amplified with SYBR Green using the following human HBB locus primers: long amplicon (∼10.1 kb), Primer F: 5-GCT​GAG​TTC​TCT​GGC​TGT​GTT​C-3; R: 5′- CCA​GGA​GAA​GTC​AGG​GTA​GGA​A -3’; short amplicon (∼122 bp), Primer F: 5′- TGC​ACG​TGG​ATC​TGT​CCG​AA -3’; R: 5′- GCA​CCT​GAC​TCT​CTC​CAC​CA -3’. Lesion frequency was derived from Ct values as follows: ΔCt_long = Ct(treated)_long - Ct(NC)_long and ΔCt_short = Ct(treated)_short - Ct(NC)short; relative amplification for each amplicon RA = 2^−ΔCt; normalized relative amplification RA_norm = RA_long/RA_short. Assuming a Poisson distribution of blocking lesions, the interstrand crosslinks frequency per 10 kb was calculated as λ(10 kb) = −ln(RA_norm) × (10,000/L), where L is the long-amplicon length (bp). To confirm assay specificity for interstrand crosslinks, parallel DNA samples were subjected to de-crosslinking by incubation at 65°C for 2 h in 50 mM Tris-HCl (pH 8.0) prior to amplification, which restored long-amplicon amplification to NC levels.

Dot Blot

Genomic DNA of BC cells treated with cisplatin and Rh2 at indicated time was isolated from cells using a phenol-chloroform based kit, RNase A-treated, and quantified by UV spectrophotometry. For each sample, 0.5 µg DNA was brought to equal volume and denatured in 0.4 M NaOH, 10 mM EDTA for 10 min in room temperature, then chilled on ice. 2 µL denatured DNA per spot was applied to a positively charged nitrocellulose membrane using a dot-blot manifold or by manual spotting. Membranes were neutralized in 2× SSC, air-dried, and cross-linked (UV 150 mJ/cm² for 30 min). After blocking in 5% non-fat milk/TBST for 1 h at room temperature, membranes were incubated overnight at 4°C with an anti-cisplatin DNA adducts antibody, clone ICR4 (Merck, USA, Cat. MABE416, 1:1000 in 5% milk/TBST). Following washing, membranes were incubated with HRP-conjugated secondary antibody at 1:5000 dilution for 1 h at room temperature, and developed by chemiluminescence. To control for loaded DNA, the same membrane was briefly stained with 0.02% methylene blue in 0.3 M sodium acetate (pH 5.2), imaged, and used to normalize signal intensity. Dots were quantified by ImageJ, and cisplatin-DNA adduct signal was reported as enhanced chemiluminescence intensity normalized to methylene blue total-DNA signal and expressed relative to the control sample.

Immunoprecipitation

1 × 10⁷ BC cells were rinsed in ice-cold PBS and lysed for 30 min on ice in IP lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.5% Triton X-100, 1 mM EDTA, 1 × protease/phosphatase inhibitors). Lysates were cleared by centrifugation (14 000 × g, 15 min, 4°C), protein concentration was measured by BCA assay, and equal amounts of protein (1 mg) were precleared with Protein A/G magnetic beads (Thermo Fisher Scientific, Cat. 88803) at 4°C for 30 min. For IP, supernatants were incubated overnight at 4°C with 1 µg anti-FANCD2 antibody with gentle rotation, followed by addition of Protein A/G beads for 2 h at 4°C. Beads were washed 5 times with lysis buffer (first two washes at 150 mM NaCl, then optionally one high-salt wash at 300 mM NaCl to increase stringency), and bound proteins were eluted by boiling in 2× Laemmli sample buffer (65.8 mM Tris-HCl, [pH 6.8], 2.1% SDS, 26.3% (w/v) glycerol, 0.01% bromophenol blue) for 5 min. Eluates and matched inputs (1% of total) were resolved by SDS-PAGE and transferred to PVDF. Western blots were probed with primary antibodies against FANCP/SLX4, FANCQ/XPF, and PCNA to detect co-immunoprecipitated proteins; membranes were also reprobed for FANCD2 to confirm IP efficiency. Isotype IgG control IPs and no-antibody bead controls were included to assess nonspecific binding. Where indicated, lysates were treated with 50 µg/mL ethidium bromideduring incubation to test whether interactions were DNA-bridged. Signals were quantified by densitometry and co-IP levels were normalized to the amount of immunoprecipitated FANCD2 and to input controls.

Polysome Profiling

Cells were pretreated with cycloheximide (CHX, 100 µg/mL, 10 min, 37°C) to freeze translating ribosomes, rinsed twice with ice-cold PBS + CHX (100 µg/mL), and lysed on ice for 10 min in polysome lysis buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM MgCl₂, 1% Triton X-100, 0.5% sodium deoxycholate, 1 mM DTT, 100 µg/mL CHX, RNase inhibitor). Lysates were cleared (14 000 × g, 10 min, 4°C) and equal A₂₆₀ units were layered onto prechilled 10-50% (w/v) sucrose gradients prepared in gradient buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM MgCl₂, 1 mM DTT, 100 µg/mL CHX). Gradients were centrifuged in a Beckman SW41Ti rotor (Beckman, USA) at 210 000 × g for 2 h at 4°C. 1 mL/fraction of gradients were fractionatedwith continuous A₂₅₄ monitoring to identify monosome (80S) and polysome peaks; in selected controls, 20 mM EDTA was added to collapse polysomes and verify peak assignments. RNA from each fraction was extracted by acid phenol-chloroform, spiked, and reverse-transcribed. FANCL mRNA were quantified by RT-qPCR. For each sample, the proportion of FANCL mRNA in monosome vs polysome fractions was calculated as the summed qPCR signal (2^–ΔCt or absolute copy number) across fractions corresponding to the 80S peak (“monosome”) and ≥2-ribosome peaks (“polysome”), expressed as a percentage of total FANCL mRNA across the gradient. Data were derived from ≥3 independent experiments, with distribution profiles plotted as fraction index vs normalized FANCL abundance, and shifts from polysome to monosome were interpreted as reduced translation engagement, whereas unchanged distribution indicated preserved initiation/elongation efficiency.

Regular Quantitative PCR (qPCR)

Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Cat. No. 15596018) according to the manufacturer’s protocol. cDNA synthesis was performed using the PrimeScript RT reagent kit (Takara, Japan, Cat. No. RR037 A), and was conducted with TB Green Premix Ex Taq II (Takara, Cat. No. RR820 A) on a QuantStudio 6 Flex Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific, Cat. 4485691). The mRNA levels of the target gene were normalized to GAPDH. Primer sequence were listed below:

FANCL, F: 5′- GGA​GTG​CAA​CAG​CAC​GCA​GAA​T -3’; R: 5′- CTG​CTC​AGC​TTA​ATT​CCC​AGG​G -3’; GAPDH, F: 5′- GGT​GTG​AAC​CAT​GAG​AAG​TAT​GA -3’; R: 5′- GAG​TCC​TTC​CAC​GAT​ACC​AAA​G -3’.

Xenograft Mouse Model and in Vivo Drug Treatment

Twenty BALB/c nude mice (4-6 weeks old, 16-20 g) were maintained under pathogen-free conditions with ad libitum access to food and water. Human bladder cancer T24 cells (5 × 10⁶ in 100 µL PBS mixed 1:1 with Matrigel) were subcutaneously injected into the right flank of each mouse. When tumors reached approximately 100 mm³, mice were randomly assigned into four groups (n = 5 per group): (1) control group, receiving intraperitoneal injection of vehicle; (2) cisplatin group, receiving cisplatin at 5 mg/kg intraperitoneally once every 5 days; and (3) cisplatin + Rh2 group, receiving Rh2 at 20 mg/kg by oral gavage daily in combination with cisplatin at 5 mg/kg intraperitoneally once every 5 days; (4) cisplatin + Rh2 + NF-κB agonist, receiving Rh2 at 20 mg/kg by oral gavage daily in combination with cisplatin at 5 mg/kg intraperitoneally once every 5 days in as well as 0.1 mg/kg AZD5582 intravenously once a week. Tumor volume was calculated as V = (length × width ² )/2. After the treatment period for 4 weeks, mice were euthanized, and tumors were excised, weighed, and processed for subsequent biochemical and histological analyses. All animal experiments were approved by the institutional animal care and use committee (IACUC) of Soochow University (KY2023-039-01) and conducted in accordance with relevant guidelines. All animal procedures complied with institutional guidelines and were conducted in accordance with the ARRIVE 2.0 guidelines.

Immunohistochemistry (IHC)

Xenograft tumors (established with Matrigel; Corning, Cat. 356231) were excised, fixed in 10% neutral-buffered formalin for 48 h, processed, and paraffin-embedded. Sections (4 µm) were deparaffinized in xylene and rehydrated through graded ethanol to water. Endogenous peroxidase was quenched with 3% H₂O₂ in PBS for 10 min. Heat-induced epitope retrieval was performed in citrate buffer (10 mM sodium citrate, pH 6.0) at 95°C for 15 min, followed by cooling to room temperature. After blocking with 5% normal goat serum for 30 min, sections were incubated overnight at 4°C with primary antibodies diluted in antibody diluent: FANCL (1:200) and phospho-NF-κB p65 (Ser536) (1:200). Slides were washed in PBS/T and incubated with an HRP-conjugated anti-rabbit detection system for 30 min in room temperature. Signal was developed with DAB substrate (Dako, USA Cat. K3468) and nuclei were counterstained with hematoxylin, followed by dehydration, clearing, and mounting. FANCL and p-p65 staining was quantified by H-score.

Statistical Analysis

All data are presented as mean ± SEM unless otherwise indicated. Two-group comparisons used two-tailed unpaired Student’s t-tests. Multi-group comparisons used One-way ANOVA. All statistical analyses were conducted using SPSS 20.0, and a two-tailed P-value less than .05 was considered as statistical significance. For in vitro experiments, a sample size of n = 3 independent biological replicates per condition was chosen based on widely accepted practice in molecular and cellular biology, ensuring reproducibility while limiting technical variability. For in vivo xenograft experiments, n = 5 mice per group was selected according to previous studies using similar bladder cancer xenograft models and is sufficient to detect treatment-related differences in tumor volume while minimizing animal use. All group sizes were determined before experimentation and no animals or data points were excluded.

Results

Rh2 Disrupts the Cisplatin-Induced FA Pathway Activation in BC Cells

Three bladder cancer T24 (high grade invasive urothelial carcinoma, Grade III, c.378 C>G, p.Tyr126*, poorly differentiated), 5637 (invasive urothelial carcinoma, Grade II, p53 c.839 G>C, p.Arg280Thr, moderately differentiated) and RT4 (non-invasive papillary urothelial carcinoma, Grade I, wild type p53, well differentiated) ¹⁹ cell lines were exposed to cisplatin. Although these cell lines represent different origins, p53 genotypes and different differentiation grades as known to all, monoubiquitinated ID2 complex was observed upon cisplatin stimulation in all three cell lines (Figure 1A-C), indicating that the activation mechanism of the FA signaling pathway was conserved in the bladder cancer cell line. However, pretreatment with Rh2 followed by cisplatin exposure abolished the monoubiquitination of ID2 complex (Figure 1A-C). FANCD2 foci by immunofluorescence were substantially diminished by Rh2 pretreatment (Figure 1D-F). The levels of DNA interstrand crosslinks were quantified by qPCR blocking assay, and the intrastrand crosslinks were quantified by dot blot assay over time in three bladder cancer cells. Notably, although cisplatin treatment induced both interstrand and intrastrand crosslinks, the pretreatment of Rh2 selectively increased the persistence of interstrand crosslinks, as evidenced by a slower decline during the repair phase compared with cisplatin alone (Figure 1G-I). In contrast, the kinetics of intrastrand crosslink removal were comparable between with and without Rh2 pretreatment (Figure 1J-L), indicating that Rh2 specifically impaired FA-mediated interstrand crosslinks repair without affecting the clearance of intrastrand crosslinks repair. Now, we determined that Rh2 was capable of inhibiting FA pathway activation in bladder cancer cells in vitro.

Figure 1.

Rh2 impairs FA pathway activation and selec tively delays interstrand crosslink repair in bladder cancer cells. (A-C) Western blot assay show the monoubiquitination (top band highlighted by red arrows) of FANCI and FANCD2 in T24 (A), 5637 (B), and RT4 (C) bladder cancer cell lines stimulated by cisplatin and Rh2. (E-F) Immunofluorescence show the FANCD2 nuclear foci in T24 (D), 5637 (E), and RT4 (F) bladder cancer cell lines stimulated by cisplatin and Rh2, and the percentage of cells with foci was compared among “NC”, “Cisplatin” and “Cisplatin/Rh2” at below. “*” indicates significant difference (P < .05) compared with “NC”, and “#” indicates significant difference (P < .05) compared with “Cis” as determined by One-way ANOVA based on three independent biological replicates (the same with below). (G-I) Quantification of interstrand crosslink repair kinetics in bladder cancer cells by qPCR-blocking following “Cisplatin” and “Cisplatin/Rh2”. Cells were harvested at 0, 6, and 24 h after drug exposure, and genomic DNA was subjected to amplification of a ∼10 kb target (HBB locus) normalized to a nested ∼120 bp amplicon from the same region. Interstrand crosslink frequency (lesions/10 kb) was calculated based on the normalized relative amplification using a Poisson distribution model. Dashed lines indicate parallel samples subjected to de-crosslinking treatment (65°C for 2 h) prior to qPCR to confirm assay specificity, which restored long-amplicon amplification to near baseline. “Cisplatin/Rh2” selectively delayed interstrand crosslink removal compared with “Cisplatin”, whereas de-crosslinking abolished the apparent lesions in both groups. “*” indicates significant difference (P < .05) compared “Cis” as determined by One-way ANOVA based on three independent biological replicates. (J-L) Dot blot assay reveal that Rh2 do not affect the removal kinetics of cisplatin-induced intrastrand crosslinks in T24 (J), 5637 (K), and RT4 (L) bladder cancer cell lines. NC: untreated BC cells; Cis: 10 μM cisplatin treatment for 12 h; Cis/Rh2: 20 μM Rh2 pretreatment for 24 h prior to cisplatin

The interaction between FANCD2 and FANCP (SLX4), FANCQ (ERCC1) as well as PCNA by immunoprecipitation was also diminished in bladder cancer T24, 5637 and RT4 cells (Figure 2A-C), implying that Rh2 disrupted the subsequent function of nucleotide excision repair (NER) or homologous recombination (HR) triggered by activated FA pathway.

Figure 2.

Rh2 impairs the recruitment process of downstream DNA repair factors mediated by FANCD2. (A-C) Western blot assay show the interaction between FANCD2 and FANCP/FANCQ/PCNA in T24 (A), 5637 (B), and RT4 (C) bladder cancer cell lines stimulated by cisplatin and Rh2. Weakened interaction between FANCD2 and FANCP/FANCQ/PCNA in “Cis/Rh2” compared to “Cis” highlighted by red arrows

Rh2 Reduces FANCL Transcription by MAPK/ERK Inhibition

To determine at which stage Rh2 impeded monoubiquitination of ID2 complex, we examined representative proteins from different functional modules of the FA pathway, including FANCA, FANCC, FANCF (assembly), FANCM, FAAP24 (damage recognition), FANCL, FANCB, FAAP100 (catalytic module) and USP1, UAF1 (deubiquitination). Among these, only FANCL protein levels were markedly reduced upon Rh2 treatment in cisplatin-stimulated cells (Figure 3A). Overexpression of FANCL restored ID2 monoubiquitination in the presence of Rh2, whereas siRNA-mediated USP1 knockdown or USP1 inhibition by the ML323 did not prevent the Rh2-induced reduction of ID2 monoubiquitination (Figure 3B-D). qPCR analysis further revealed that Rh2 significantly suppressed FANCL mRNA level, while Actinomycin D chase assays showed no difference in mRNA decay rates between with and without Rh2 treatment (Figure 3E), indicating that Rh2 indeed reduced FANCL mRNA level but not at mRNA stability. Polysome profiling demonstrated no change in the distribution of FANCL mRNA between monosome and polysome fractions (Figure 3F), suggesting that Rh2 neither altered translation initiation or elongation efficiency of FANCL. Moreover, treatment with cycloheximide (CHX) or the proteasome inhibitor MG132 did not affect FANCL protein stability (speed of FANCL degradation) in Rh2-treated cells (Figure 3G), thereby excluding the probable protein degradation of FANCL. All evidence above supported the conclusion that Rh2 suppressed FANCL expression primarily at the transcriptional level.

Figure 3.

Transcription of FANCL is diminished by Rh2. (A) Western blot assay show the expression of FANCA, FANCC, FANCF, FANCM, FAAP24, FANCL, FANCB, FAAP100, USP1 and UAF1 in bladder cancer cell lines stimulated by cisplatin and Rh2. FANCL expression is reduced by Rh2. (B-D) Western blot assay show the mono-ubiquitination of ID2 complex in T24 (B), 5637 (C), and RT4 (D) bladder cancer cell lines additionally treated with FANCL ectopic plasmid, USP1 siRNA and ML323 (for USP1 activity inhibition). FANCL over-expression is determined to rescue ID2 complex activity. (E) The relative mRNA levels of FANCL are detected by qPCR in T24, 5637, and RT4 bladder cancer cell lines additionally treated with Actinomycin D for 2, 4 and 6 h. A linear regression model is fitted with expression level of FANCL as the time, group, and their interaction term (time × group) as independent variables. The interaction term is used to test whether the temporal trends (slopes) differed significantly between the two groups. One-way ANOVA is performed on the fitted model to obtain F-statistics and P-values for each effect. A non-significant interaction term indicated no difference in the rate of change between groups. (F) The relative mRNA levels of FANCL in monosome and polysome fractions are determined by qPCR. Ratio of two fractions are compared among “NC”, “Cisplatin” and “Cisplatin/Rh2”. (G) Western blot assay show the expression of FANCL in T24, 5637, and RT4 bladder cancer cell lines additionally treated with cycloheximide for 2, 4, 8 h and MG132. One-way ANOVA is performed to show that either cycloheximide or MG132 does not affect the degradation speed of FANCL compared between “Cisplatin” and “Cisplatin/Rh2”. “*” indicates significant difference (P < .05) compared with “NC”, and “#” indicates significant difference (P < .05) compared with “Cisplatin”. LOE: FANCL over-expression; UKD: USP1 knockdown; ActD_number: Actinomycin D as a transcription inhibitor for indicated hour treatment; CHX: cycloheximide as a translation inhibitor; MG132: proteasome inhibitor

As previously described, Rh2 has been shown to inhibit multiple classical signaling pathways, including NF-κB, ²⁰ Wnt/β-catenin, ²¹ Axl, ²² JNK/AP-1, ²³ and IL-6/JAK2/STAT3, ²⁴ across various disease models. To identify which pathway contributed to Rh2-induced suppression of FANCL, BC cells were co-treated with Rh2 and pathway-specific agonists to test rescue of FANCL expression. Notably, the activities of Axl and JNK/AP-1 were not suppressed by Rh2 in our bladder cancer cell system (Figure 4A-C). Although pathway activation were all successfully confirmed (The PI3K/Akt pathway is a principal downstream effector of Axl signaling, so using a PI3K/Akt agonist can bypass the direct Axl autophosphorylation inhibition by Rh2) (Figure 4A-C), both mRNA and protein of FANCL were only elevated by NF-κB agonist in the presence of Rh2 (Figure 4A-F).

Figure 4.

Rh2 blocks NF-κB signaling pathway in bladder cancer. (A-C) Western blot assay show the expression of FANCL as well as the phosphorylation of p65 (Ser536), β-catenin (Ser37), Akt (Thr308), JNK (Thr183/Tyr185), STAT3 (Tyr705) to determine the activity of NF-κB, Wnt/β-catenin, PI3K/Akt, JNK/AP-1 and JAK2/STAT3 pathways in T24 (A), 5637 (B), and RT4 (C) bladder cancer cell lines additionally treated with CHIR99021 (NF-κB-inducing kinase, GSK3 inhibitor), SC79 (Akt activator), anisomycin (JNK activator), colivelin (STAT3 activator). (D-F) The relative mRNA levels of FANCL are detected by qPCR in T24 (D), 5637 (E), and RT4 (F) bladder cancer cell lines additionally treated with agonists of the pathway mentioned above. “*” indicates significant difference (P < .05) compared with “Cisplatin”, and “#” indicates significant difference (P < .05) compared with “Cisplatin/Rh2”

Rh2 Potentiates the Cytotoxicity of Cisplatin Against Bladder Cancer Cells in Vivo

Given the pivotal role of the FA signaling pathway in repairing DNA damage induced by platinum-based and other anticancer agents, Rh2 may serve as a potent enhancer of the therapeutic efficacy of radiotherapy and chemotherapy. In a xenograft mouse model established by subcutaneous implantation of T24 cells, intraperitoneal administration of cisplatin combined with Rh2 resulted in a marked reduction in tumor volume compared with treatments lacking Rh2, whereas NF-κB agonist reversed this effect of Rh2, restoring tumor size. (Figure 5A). Mechanistically, Rh2 impaired the monoubiquitination of ID2 complex and consistently downregulated FANCL expression (Figure 5B). In parallel, NF-κB pathway activity was attenuated, as evidenced by decreased phosphorylated p65 levels (Figure 5B). Immunohistochemical analysis further confirmed the diminished expression of FANCL (Figure 5C) and p-p65 (Figure 5D) in Rh2-treated tumors.

Figure 5.

Enlarged effect of cisplatin by Rh2 on bladder cancer in vivo. (A) Rh2 enhances cisplatin efficacy in a T24 xenograft model (n = 5), significantly reducing tumor volume. (B) Western blot assay show the mono-ubiquitination of FANCD2, FANCI, the phosphorylation of p65 as well as the expression of FANCL in xenografted tumor. Rh2 inhibits ID2 complex monoubiquitination, suppresses FANCL expression, and attenuated NF-κB signaling, as shown by decreased p-p65 levels. (C, D) Immunohistochemistry show the positive staining of FANCL (C) and p-p65 (D) in xenografted tumors. Scale bar, 0.02 mm. “*” indicates significant difference (P < .05) compared with “NC”, and “#” indicates significant difference (P < .05) compared with “Cisplatin”, and “^” indicates significant difference (P < .05) compared with “Cisplatin/Rh2”

In summary, we demonstrated that Rh2 could suppress NF-κB pathway, thereby reducing FANCL expression, inhibiting FA pathway activation, and ultimately enhancing the cytotoxic effect of cisplatin against bladder cancer.

Discussion

Platinum-based chemotherapeutic agents, particularly cisplatin, remain the backbone of systemic therapy for muscle-invasive and metastatic bladder cancer, often in combination regimens. These agents exert their cytotoxic effects primarily through the formation of DNA adducts, including intrastrand and interstrand crosslinks, which disrupt DNA replication and transcription. The main cytotoxic effect of cisplatin arises from intrastrand crosslinks, but the interstrand crosslinks it generates are particularly lethal, as they completely block strand separation. ²⁵ Consequently, cisplatin is often used as an ICL-inducing model drug in studies of the FA pathway and related mechanisms, although it is not a “pure” ICL inducer such as mitomycin C or psoralen.^26,27 Despite initial sensitivity, many patients experience relapse due to acquired resistance, which severely limits the long-term effectiveness of treatment. Resistance to platinum compounds can arise from multiple factors, including increased drug efflux, inactivation by thiol-containing molecules such as glutathione, and-critically-enhanced DNA damage repair capacity. ²⁸ Overcoming such resistance is a major clinical challenge in urothelial carcinoma therapy and necessitates new strategies to selectively weaken the tumor’s DNA repair capabilities while minimizing toxicity to normal tissues.

Among the mechanisms enabling tumor cells to withstand platinum-induced genotoxic stress, DNA damage repair pathways play a central role. Enhanced repair efficiency allows cells to tolerate otherwise lethal DNA lesions, restoring replication fork progression and genomic stability. ²⁹ Platinum-induced DNA lesions are recognized and processed through multiple pathways, including NER for intrastrand adducts, HR for double-strand break intermediates, and the FA pathway for interstrand crosslinks.^30,31 Mounting evidence indicates that aberrant activation or overexpression of key DNA repair factors confers a survival advantage to tumor cells, contributing directly to chemoresistance. In bladder cancer, high expression of DNA repair genes, particularly within the FA and HR pathways, has been associated with poor clinical response to cisplatin-based chemotherapy.^32,33 The FA signaling pathway is especially critical in resolving interstrand crosslink lesions, which are among the most cytotoxic forms of DNA damage caused by platinum drugs. Upon sensing an interstrand crosslink, the FA core complex-comprising multiple FANC proteins-activates its E3 ubiquitin ligase subunit FANCL, which monoubiquitinates the ID2 complex. This modification facilitates the recruitment of structure-specific nucleases to incise the damaged DNA, followed by lesion bypass through translesion synthesis and repair completion via HR. Without proper activation of the FA pathway, cells accumulate unresolved interstrand crosslinks, leading to replication fork collapse, double-strand breaks, and apoptosis. Therefore, inhibition of FA pathway activity represents a promising strategy to sensitize tumor cells to interstrand crosslink-inducing agents like cisplatin and mitomycin C.

Within the FA pathway, FANCL occupies a central position as the catalytic E3 ubiquitin ligase responsible for the monoubiquitination of ID2 complex.^34,35 This post-translational modification is the irreversible commitment step in FA pathway activation, enabling the transition from lesion recognition to nucleolytic processing. Loss-of-function mutations in FANCL result in defective interstrand crosslink repair, hypersensitivity to crosslinking agents, and increased chromosomal instability. ³⁶ While much is known about FANCL’s biochemical role within the core complex, comparatively little is understood about its upstream regulation. Understanding how FANCL expression and activity are modulated under physiological and stress conditions could reveal therapeutic entry points to modulate FA pathway output. To date, there have been no definitive reports specifically identifying transcriptional regulators or signaling pathways that directly control FANCL expression. While certain transcription factors, such as E2F family members and NF-Y, are known to broadly regulate DNA repair gene expression, ³⁷ no study has conclusively mapped FANCL’s promoter occupancy or defined enhancer elements governing its transcription. Likewise, post-transcriptional regulation of FANCL, such as by microRNAs or RNA-binding proteins, remains unexplored. This gap in knowledge presents both a challenge and an opportunity: elucidating FANCL’s regulatory inputs could allow for selective modulation of FA pathway activity in tumors without broadly impairing other DNA repair processes.

Rh2, an active saponin from Panax ginseng, has emerged as a multi-target antitumor agent with pleiotropic effects on signaling networks. In various malignancies, Rh2 has been shown to suppress pro-survival and pro-proliferative pathways including NF-κB, Wnt/β-catenin, PI3K/Akt, JNK/AP-1, Axl, and IL-6/JAK2/STAT3 as mentioned in the previous text. These effects converge on reduced expression of anti-apoptotic proteins, inhibition of epithelial–mesenchymal transition, suppression of angiogenesis, and sensitization to chemotherapy. Notably, NF-κB inhibition is a recurrent finding across tumor models and other diseases,^38-42 suggesting that Rh2 may act on upstream kinases such as IKK, interfere with p65 nuclear translocation, or alter co-activator recruitment to NF-κB target promoters. Several studies have also linked Rh2 to modulation of DNA damage repair. Rh2 treatment has been reported to impair HR efficiency, suppress expression of NER components, and increase sensitivity to genotoxic drugs and ionizing radiation.^8,9 The underlying mechanisms are incompletely understood, but likely involve interference with damage response signaling cascades, transcriptional down-regulation of repair genes, and possible disruption of chromatin remodeling at damage sites. These findings position Rh2 as a promising candidate for combination therapy strategies aimed at exploiting synthetic lethality in repair-deficient tumor cells.

In the present study, we have revealed that Rh2 suppresses NF-κB activity, leading to decreased FANCL expression and compromised monoubiquitination of ID2 complex, thereby impairing FA pathway activation. Functionally, this attenuation of interstrand crosslink repair sensitized bladder cancer cells to cisplatin both in vitro and in a T24 xenograft mouse model, resulting in significantly reduced tumor growth. Importantly, co-administration of an NF-κB agonist reversed the Rh2-induced suppression of FANCL and FA pathway activity, restoring tumor size toward control levels. This establishes a novel mechanistic link between NF-κB signaling and FANCL regulation, and identifies NF-κB as a potential upstream determinant of FA pathway capacity in bladder cancer.

Nevertheless, several limitations of our study warrant discussion. First, the molecular mechanism by which Rh2 inhibits NF-κB remains undefined. Further work is required to determine whether Rh2 directly targets IKK complex activity, disrupts p65 phosphorylation or nuclear translocation, or modulates transcriptional co-activators. Second, while the FA pathway primarily mediates repair of interstrand crosslink lesions, platinum agents also induce other types of DNA damage, such as intrastrand crosslinks and monoadducts, which are repaired by NER and other mechanisms. Rh2’s capacity to enhance cisplatin efficacy may therefore involve additional effects beyond FA pathway inhibition, including modulation of parallel repair systems or apoptotic signaling.

Finally, the Rh2-NF-κB-FANCL axis identified here may represent a conserved regulatory mechanism across multiple tumor types, given the widespread role of NF-κB in cancer biology and the ubiquitous involvement of the FA pathway in genome maintenance. Future studies should evaluate whether this pathway operates similarly in other malignancies, and whether Rh2 can be generalized as a sensitizing agent for crosslinking chemotherapy and radiotherapy. Detailed dissection of Rh2’s molecular targets, coupled with pharmacodynamic studies in clinically relevant models, will be essential to advance its translation into therapeutic regimens. In clinical settings, elevated FANCL or NF-κB activity could serve as predictive biomarkers for platinum resistance, helping to stratify patients who are less likely to respond to conventional regimens. By pharmacologically targeting this pathway, either through Rh2 or more selective derivatives, it may be possible to re-sensitize resistant tumors while sparing normal tissues. Rh2’s multitarget profile and natural origin also suggest the feasibility of its incorporation into combination regimens with existing DNA-damaging agents, potentially allowing for dose reduction of cytotoxic drugs and mitigating systemic toxicity. Moreover, since the FA pathway and NF-κB signaling are conserved across multiple malignancies, the Rh2-based strategy may have broader applicability beyond bladder cancer, particularly in platinum-refractory ovarian, lung, or head-and-neck cancers. Moving forward, pharmacokinetic, toxicity, and synergistic efficacy studies in clinically relevant models will be critical to bridge these findings toward translational and clinical application.

Conclusion

Ginsenoside Rh2 suppresses NF-κB signaling to transcriptionally downregulate FANCL, thereby impairing FA pathway-mediated DNA repair and enhancing cisplatin cytotoxicity in bladder cancer. These findings highlight Rh2 as a potential combinatorial agent to overcome platinum resistance.

Supplemental Material

Supplemental Material - Ginsenoside Rh2 Suppresses the Fanconi Anemia Pathway by Inhibiting NF-κB-Mediated FANCL Transcription in Bladder Cancer

Supplemental Material for Ginsenoside Rh2 Suppresses the Fanconi Anemia Pathway by Inhibiting NF-κB-Mediated FANCL Transcription in Bladder Cancer by Chen Li, Guanglin Lv, Zhuang Xiong, and Zhenfan Wang in Dose-Response