CF10 Displays Improved Synergy with Oxaliplatin in -Null and Wild-Type CRC Cells from Increased Top1cc and Replication Stress.

1. Introduction
Colorectal cancer (CRC) is the 3rd leading cause of cancer-related deaths in the U.S., and incidence is increasing, particularly in younger individuals [1]. The primary first-line therapy for metastatic CRC (mCRC) is the 5-fluorouracil (5-FU)-based regimen FOLFOX (Folinic acid (leucovorin), 5-FU, Oxaliplatin) [2]. FOLFOX is also a preferred adjuvant chemotherapy regimen for CRC patients with stage III and high-risk stage II disease [3]. While FOLFOX is generally more effective than 5-FU/LV, oxaliplatin (OXA) adds potential risk for serious neuropathy [4]. This emphasizes the importance of understanding the molecular basis of its therapeutic benefit in the context of 5-FU-based chemotherapy to select patients who will benefit despite the increased risk of toxicity. Further, mortality for mCRC patients and recurrence among patients with stage III and high-risk stage II treated with FOLFOX remains unacceptably high, and new therapies are needed [5,6].
A limitation to the therapeutic benefit of FOLFOX relative to 5-FU/LV has been reported in CRC lacking wild-type (WT) p53. TP53 mutations or deletions are found in ~60% of all CRC tumors [7]. Although TP53 status is not currently used to stratify CRC patients, the FOLFOX component drugs are known to exhibit TP53-dependent activity. The cytotoxicity of 5-FU results from both RNA- and DNA-directed processes, and both processes display TP53 dependence [7,8,9]. The cytotoxicity of OXA also displays TP53 dependence [10]. Several clinical studies have shown that the therapeutic benefit of FOLFOX vs. 5-FU/LV is TP53-dependent, and patients treated with FOLFOX who do not express WT TP53 are at increased risk for poor outcomes and decreased overall survival [11,12,13].
To overcome the limitations of 5-FU-based therapies, our lab is developing a 2nd-generation FP polymer, CF10, as an improved treatment for mCRC. The core of CF10 consists of 10 FdUMP units linked via a phosphodiester DNA backbone. Cytarabine (AraC) at the 3′-terminus and polyethylene glycol (PEG6) at the 5′-terminus improve potency and reduce exonucleolytic degradation. CF10 is highly potent against CRC cells regardless of MSI/MSS status, KRAS mutation status, or other clinically used factors used to stratify CRC patients for treatment. CF10 is highly effective in TP53-null cancer cell lines and is being investigated as an improved therapy for TP53-null AML, which displays poor outcomes due to chemoresistance [14]. The improved potency of CF10 per molar equivalent of 5-FU is due to both its more efficient intracellular conversion to FdUMP and its unexpected activity as a DNA topoisomerase 1 (Top1) poison [15], making it a unique TS/Top1 dual inhibitor. CF10 also causes more DNA double-strand breaks (DSBs) than 5-FU, due to greater Top1 poisoning, leading to increased replication stress [16,17]. Our laboratory has demonstrated that CF10 exhibits improved antitumor activity compared with 5-FU in multiple in vivo CRC models [18,19].
In this study, we examined whether the combination of CF10 with OXA (COXA) is likely to be a safe and effective therapeutic strategy for treating advanced CRC. While fundamental to 5-FU activity, leucovorin was not included in these studies because standard culture media contain high folate levels, obviating any benefit from leucovorin co-treatment. However, they do not fully mimic the low folate environment in human serum [20,21,22]. We expect that COXA-based therapeutic regimens can improve outcomes for mCRC patients and that limitations in the efficacy of 5-FU + OXA (FOXA)-based therapies associated with TP53 status will be reduced or eliminated with COXA-based therapies, adding to the previously established therapeutic benefit of CF10 relative to 5-FU.

2. Materials and Methods

2.1. Cell Lines and Reagents
HCT116 WT (RRID: CVCL_0291) and HCT116 TP53−/− (RRID: CVCL_HD97) cell lines were obtained from GRCF Bio repository and Cell Center from Johns Hopkins School of Medicine (Baltimore, MD, USA), and LS174T (RRID: CVCL_1384) and Caco-2 (RRID: CVCL_0025) cells were from ATCC (Manassas, VA, USA). Cellular authentication and STR profiling were performed by ATCC and GRCF Bio repository (Supplementary Table S1). Cells were cultured in recommended media and regularly tested for Mycoplasma contamination. CF10 was obtained from STPharma and dissolved in sterile saline (Baxter International, Deerfield, IL, USA) (Cat. No. 2F7124). CF10 concentrations were determined from A260 UV absorbance using a spectrophotometer and using the extinction coefficient for ssDNA. Arabinosyl cytidine (AraC) was purchased from Thermo Fisher Scientific (Waltham, MA, USA, Cat. No. 449560010). Clinical-grade 5-fluorouracil (5-FU) (NDC 63323-117-00) was purchased from the WFUSM clinical pharmacy as a 50 mg/10 mL vial (Fresenius Kabi USA, LLC, Wilson, NC, USA). All drugs were filtered using a 0.22-μm syringe filter. Cells were maintained at 37 °C with 5% CO2. All culture media were supplemented with 10% FBS (Gibco) (Grand Island, NY, USA), 20% FBS for Caco-2 cells, 1% penicillin–streptomycin (Sigma) (St. Louis, MO, USA), and 1% L-Glutamine (Sigma).

2.2. Cell Viability and Synergy Analyses
All cells were treated with varying concentrations of CF10, 5-fluorouracil (5-FU), and Oxaliplatin (OXA) for 72 h. Cells were exposed to varying concentrations of each drug: 1 μM to 0.0001 μM for CF10; 10 μM to 0.001 μM for 5-FU and OXA. Cell viability was determined using Aqueous One CellTiter-Glo (Promega, Cat. No. G3581, RRID: SCR_006724) (Madison, WI, USA) according to the manufacturer’s instructions. To determine drug interactions with COMBENEFIT, cells were treated simultaneously with both agents (e.g., OXA + CF10, OXA + 5-FU) in an 8 × 9 concentration matrix spanning biologically relevant doses, as indicated in Figure 1. The effect of drug combinations on cell viability was also determined using CellTiter-Glo. Optical absorbance values were measured using a SpectraMAx iD3 spectrophotometer plate reader (Molecular Devices, San Jose, CA, USA, RRID: SCR_023920). Synergy and EC50 values for combinations were evaluated using the highest single agent (HSA) analysis model in the COMBENEFIT program.

2.3. Modified Clonogenic Assay/Colony Forming Assay
For colony-forming assays, approximately 500 cells were seeded into each well of a 24-well plate and allowed to attach for 24 h. Following attachment, cells were treated with single or combination drug treatments including CF10, 5-FU, and OXA, with drug combinations administered at ratios of 1:10 for CF10 + OXA and 1:1 for 5-FU + OXA. Ratios were selected based on previous studies demonstrating that CF10 cytotoxicity can be observed at drug concentrations 10×, 100×, and even 1000× lower than those of 5-FU. Meanwhile, OXA doses were selected based on literature reporting in similar cell lines. Cells were exposed to varying concentrations of each drug, ranging from 0.1 μM to 0.00123 μM for CF10 and 1 μM to 0.0123 μM for 5-FU and OXA, for 72 h. After drug treatment, cells were washed with PBS, fresh medium was added, and plates were incubated for 4 days to allow for colony formation. CellTiter-Glo (Promega, Cat no. G3581) was added to the plates, and the absorbance was read at 562 nm on a SpectraMax iD3 spectrophotometer plate reader (Molecular Devices, RRID: SCR_023920) (San Jose, CA, USA). Isobologram plots were generated to visually represent the interaction between the drugs. The Chou–Talalay equation and CompuSyn Software V1.0 (ComboSyn, Paramus, NJ, USA, RRID: SCR_022931) were used to determine the Combination Index (CI) and Dose Reduction Index (DRI) The CI was used to determine the type of drug interaction: CI < 1 indicates synergism, CI = 1 indicates an additive effect, and CI > 1 indicates antagonism. The equation for CI is given below:
The DRI for CF10 was calculated using the formula below, measuring how many fold the dose of CF10 is reduced in combination with OXA, compared to single drug treatment:

2.4. Statistical Analysis
Statistical analysis was performed using GraphPad Prism 7 software (RRID: SCR_002798). Data were expressed as mean and standard Error Mean (SEM) from at least three independent experiments. An ordinary one-way ANOVA was performed, followed by Sidak’s multiple comparisons test, with a single pooled variance.

2.5. Determination of DNA Damage
To assess the extent of DNA damage, cells were stained for γH2AX, a marker of DNA double-strand breaks, using the primary anti-phospho-γH2AX (Ser139) Rabbit Ab (Cell Signaling, Beverly, MA, USA) (Cat. No. 2577S, RRID: AB_2118009) and the secondary anti-Rabbit IgG Fab2 Ab conjugated with Alexa-488 (Cell Signaling, Cat. No. 4412S, RRID: AB_AB_1904025). For all washing steps, 1× PBS without Ca2+ or Mg2+ was used to protect the DNA from any DNase-mediated degradation.

2.6. Detection of Apoptosis and Cell Cycle Using Annexin V/PI Assay
HCT116 cells were treated for 48 h with COXA and FOXA at the indicated doses, or with the corresponding single doses of 5-FU and CF10. Cells were stained using the Alexa Fluor 488 Annexin V/Dead Cell Apoptosis Kit (Invitrogen, Cat. No. V13241), following the manufacturer’s protocol. Similarly, cell cycle analysis was performed by treating cells for 24 and 48 h, fixing them in 70% ethanol overnight, washing with PBS, and using a propidium iodide flow cytometry kit Abcam (Waltham, MA, USA) (ab139418). Flow cytometry was performed using BD FACS Canto II Analyzer (BD Biosciences, Franklin Lakes, NJ, USA), RRID: SCR_019627). Data were processed using FCS Express 6 Flow Cytometry Software (De Novo Software, Pasadena, CA, USA RRID: SCR_016431) or https://floreada.io, (date accessed 20 January 2025, web-based software) (RRID: SCR_025286). Gating strategies for Annexin V/PI staining involved isolating single, healthy, and apoptotic cells based on size and complexity (FSC/SSC) and then applying a quadrant gate on the Annexin V (x-axis) vs. PI (y-axis) plot to separate live (Ann-V−/PI−), early apoptotic (Ann-V+/PI−), late apoptotic (Ann-V+/PI+), and necrotic cells (Ann-V−/PI+). Accurate gating involved excluding doublets (FSC-A vs. FSC-H) and ensuring apoptotic cells with reduced size (lower FSC) are included.

2.7. Immunofluorescence
Human CRC cells were grown in Cell View chambered glass-bottom culture dishes (Greiner Bio-One, Monroe, NC, USA Item No. 627870) or in Fisher cover glass in a 12-well plate, seeded at a density of 12,000 cells per chamber. After a 24-h attachment period, cells were treated with 10 μM 5-fluorouracil (5-FU), 0.1 μM CF10, 10 μM oxaliplatin (OXA), or combinations of CF10 + OXA (COXA, 1:10 ratio) and 5-FU + OXA (FOXA, 1:1 ratio) for 48 h. Following treatment, cells were fixed with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) for 15 min. After fixation, cells were permeabilized with 0.1% Triton-X in PBS for 15 min to allow antibody access to intracellular targets. To render DNA-protein crosslinks more accessible to an antibody, the cells were incubated with 1% SDS at room temperature for 5 min, then washed three to five times with wash buffer (1% BSA in PBST). Cells were then blocked for 1 h at room temperature in 1% BSA in PBST. After blocking, cells were incubated overnight at 4 °C with the primary antibody at a 1:1000 dilution in 1% BSA/PBS. The following day, cells were washed three times with PBST to remove unbound primary antibody, then incubated with a secondary antibody conjugated to a fluorophore at the appropriate dilution for 1 h at room temperature in the dark. After the final wash with PBS to remove excess secondary antibody, cells were stained using SlowFade Gold antifade reagent with DAPI (Invitrogen, Carlsbad, CA, USA Cat. No. S36939) and imaged using an Olympus FV4000 confocal microscope, Tokyo, Japan, (RRID: SCR_017015).

2.8. Detection of Topoisomerase 1 Cleavage Complex (Top1cc) Formation
HCT116 cells were stained with anti-Top1cc Mouse (Millipore Sigma, Danvers, MA, USA, Cat. No. MABE1084, RRID: AB_2756354) and anti-phospho-γH2AX (Ser139) Rabbit Ab (Cell Signaling, Cat. No. 2577S, RRID: AB_2118009) primary antibodies. For secondary antibodies, anti-rabbit IgG Alexa 488 (Cell Signaling, Cat. No. 4412S, RRID: AB_10694746) and anti-mouse IgG Alexa 657 (Cell Signaling Technology, Cat# 4410, RRID: AB_1904023) were used. Coverslips were mounted on glass slides or directly in Cell View chambered glass-bottom culture dishes using SlowFade Gold antifade reagent with DAPI (Invitrogen, Cat. No. S36939). Slides were kept in the dark for 15 min at room temperature, and then confocal microscopy was performed using an Olympus FV4000 confocal microscope (Tokyo, Japan).

2.9. Immunoblotting
Human colorectal cancer cell lines were cultured in McCoy’s 5A (modified) medium (Paisley, Scotland, UK, Cat. No. 16600082) supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin at 37 °C in a 5% CO2 atmosphere. Cells were seeded in 6-well plates at approximately 300,000 cells per well and allowed to adhere overnight before being treated with the indicated drug concentrations or control treatments for the specified durations. Drug concentrations used for Western blots were as follows: CF10 at 0.01 µM, OXA at 2.5 µM, and 5-FU at 1 µM. Higher doses are needed in IF to visualize and quantify cellular localization, whereas lower doses in WB allow for the accurate detection of protein changes without signal saturation. Following treatment, cells were washed twice with cold PBS and lysed directly in the 6-well plates using RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitor cocktails. Cell lysates were collected by scraping, transferred to microcentrifuge tubes, incubated on ice for 30 min with intermittent vortexing, homogenized, and centrifuged at 10,000 rpm for 30 min at 4 °C to remove insoluble debris. The supernatant containing total protein was collected and quantified using the BCA Protein Assay Kit (Thermo Fisher, Cat. No. 23225) according to the manufacturer’s instructions. Equal amounts of protein (20–40 µg) were mixed with 2× Laemmli sample buffer containing β-mercaptoethanol, boiled at 95 °C for 10 min, separated by SDS-PAGE on a 10% polyacrylamide gel, and transferred to nitrocellulose membranes using a semi-dry BioRad Trans Blot Turbo system (Hercules, CA, USA, RRID: SCR_023156). Nitrocellulose membranes were blocked with 5% non-fat dry milk in TBS-T for 1 h at room temperature, then incubated with primary antibodies against p53 (Cell Signaling, Cat. No. 2524, RRID: AB_331743), p21 (Cell Signaling, Cat. No. 2947, RRID: AB_823586), β-actin (Cell Signaling, Cat. No. 4970, RRID: AB_2223172), Phospho-Chk1 (Ser345) (133D3) (Cell Signaling Technology, Danvers, MA, USA, Cat. No. 2348, RRID: AB_331212), Phospho-Chk2 (Thr68) (Cell Signaling Technology Cat. No. 2197, RRID: AB_2080501), Chk1 (Cell Signaling Technology Cat# 37010, RRID: AB_3662851), Chk2 (Cell Signaling Technology Cat# 6334, RRID: AB_11178526), and Thymidylate Synthase (D5B3) (Cell Signaling Technology Cat# 9045, RRID: AB_2797693) diluted in TBS-T overnight at 4 °C. Following primary antibody incubation, membranes were washed three times with TBS-T and incubated with HRP-conjugated secondary (Cell Signaling Technology, Cat# 7074, RRID: AB_2099233) antibodies for 1 h at room temperature. Membranes were developed using WesternBright ECL substrate (Thermo Fisher, Cat. No. K-12045-D20) and imaged on ChemiDoc (Bio-Rad, RRID: SCR_019037). Band intensity quantification was performed using ImageJ 1.8.0 (RRID: SCR_003070). Protein band intensities were normalized to the loading control (β-actin) and expressed as fold change relative to the control treatments, with statistical analysis performed using GraphPad Prism.

2.10. Investigation of DNA Damage Using Neutral Comet Assay
To quantify the extent of DNA double-strand breaks, a neutral comet assay was performed using 0.1% (v/v) H2O2 treatment for 1 h as a positive control. HCT116 cells were treated with synergistic combinations along with the single doses of 5-FU and CF10. After 48 h of treatment, cells were harvested to prepare a single-cell suspension. A cell suspension at 106 cells/mL was diluted in molten 0.5% low-melting-point agarose (LMPA) prepared in PBS at a 1:10 (v/v) ratio. 100 μL of the cell suspension in LMPA was plated on glass slides coated with 1% normal melting point agarose (NMPA). Cell lysis was performed using lysis buffer (pH 10.0; 2.5 M NaCl, 10 mM Tris-HCl, 100 mM EDTA, and 1.2% Triton X-100) overnight at 4 °C. Slides were neutralized in 1× TBE buffer for 30 min at room temperature. Electrophoresis was performed using cold 1× TBE at 1 V/cm for 1 h. DNA precipitation was performed by submerging the slides in a solution of 7.5 M NH4OAc (13.4% v/v) and absolute EtOH (86.6% v/v) for 30 min at room temperature Slides were again submerged in 70% ethanol for 30 min at room temperature, then air-dried overnight. DNA was stained with Vista Green dye solution (1:10,000 dilution in deionized water; Cell BioLab, San Diego, CA, USA, Cat. No. 235003). Slides were visualized using a FITC filter on a Keyence All-In-One Fluorescence Microscope (BZ-X700, Itasca, IL, USA, RRID: SCR_016979), with a 10× magnification. Data were analyzed with ImageJ 1.8.0 (RRID: SCR_003070) software using the OpenComet v1.3 plugin.

3. Results

3.1. Increased Synergy of CF10 + OXA in TP53-WT and TP53-Null CRC Cells
We evaluated synergy for CF10 + OXA (COXA) vs. 5-FU + OXA (FOXA) using the highest single agent (HSA) synergy model with COMBENEFIT (Figure 1A) and using a modified colony formation assay (Supplementary Figure S1A–D) [23] using the Combination Index algorithm of CompuSyn (Figure 1B) [24]. Synergy for COXA was identified in TP53-WT HCT116, TP53-null HCT116, TP53-WT LS174T, and TP53-null Caco-2 cells over a range of drug interaction space. Synergy for COXA was greater in TP53-null HCT116 and Caco-2 cells (Figure 1A), although the potency of the individual drugs was greater in the TP53-WT HCT116 cells (Figure 1C). For FOXA, strong synergy was observed only in TP53-null HCT116, LS174T, and Caco-2 cells at higher non-clinically relevant doses, with antagonism for a few 5-FU + OXA combinations observed in the TP53-WT HCT116 cells (Figure 1A). Similarly, COMPUSYN analysis of the COXA combination displayed lower combination index values, indicating increased combinatorial potency compared to FOXA (Figure 1B), and greater synergy based on polyonograms (Supplementary Figure S1C,D). The 50 percent effect combination index scores calculated by CompuSyn were CI = 0.002 for COXA and CI = 0.98 for FOXA in TP53-null HCT116 cells, CI = 0.12 for COXA and CI = 0.59 for FOXA effect in TP53-WT HCT116 cells, CI = 0.04 for COXA and CI = 1.3 for FOXA in LS174T cells, and CI = 0.08 for COXA and CI = 0.93 for FOXA in Caco-2 cells (Figure 1B). Collectively, these results reveal that CF10 and OXA are synergistic across multiple CRC cell lines with differing p53 tumor suppressor status.

3.2. TP53-Dependent Cell Cycle Arrest Limits 5-FU + OXA Synergy
TP53 modulates the cellular response to DNA-damaging anticancer drugs, in part, by activating cell-cycle checkpoints to delay cell-cycle progression and allow DNA repair before the cells enter S-phase (G1/S-checkpoint) and mitosis (G2/M-checkpoint) [25,26]. TP53 also upregulates Bax and other proteins to activate the mitochondrial apoptotic pathway in cells that experience prolonged and extensive DNA damage [27]. We investigated the molecular mechanisms underlying the differential synergistic effects with FOXA relative to COXA in TP53-WT HCT116 and TP53-null HCT116 cells by examining cell cycle perturbations, the induction of apoptosis, and the activation of tumor suppressor p53 and downstream effector p21, in cells treated with the single drugs (CF10, 5-FU, OXA) and the FOXA and COXA (Figure 2A–D). Drug concentrations were selected based on IC50 values of single drugs.
We first quantified levels of the tumor suppressor p53 and downstream cell cycle effector p21 in TP53-WT and TP53-null HCT116 isogenic mutants, LS174T, and Caco-2 cells (Figure 2A,B; Supplementary Figures S2 and S3). COXA-treatment resulted in a ~20-fold increase in both p53 and p21 levels compared to PBS, and these levels were significantly higher than those of the component drugs (CF10 and OXA) and the FOXA combination. In contrast, FOXA-treated cells did not display significantly increased p53 or p21 relative to single-agent OXA (Figure 2A,B). Consistent with previous studies demonstrating 5-FU activates the G1/S-checkpoint in cancer cells in a TP53-dependent manner, HCT116 TP53-WT cells showed increased G1-phase cells with 5-FU treatment relative to TP53-null cells at 24 h (57% vs. 33%) (Figure 2C,D; Supplementary Figure S4), and similar trends were observed for FOXA but not for CF10 or COXA. Thus, although CF10/COXA more strongly activates p53/p21 than 5-FU/FOXA, p53-dependent cell-cycle perturbations are not as evident as for 5-FU/FOXA. These results are consistent with findings that increased FOXA synergy in TP53-null cells depends on decreased G1 cell cycle arrest and increased S and G2/M cell cycle arrest. Increased G1 arrest in TP53-WT cells presumably allows for DNA repair and limits lethal DNA damage [28]. Similar effects were observed at 48 h (Supplementary Figure S4).
Since sustained p53 levels can activate apoptosis, we then evaluated the sub-G0 cell population by flow cytometry for TP53-WT and TP53-null cells (Supplementary Figure S4C). All treatments increased sub-G0 cell populations in WT cells relative to TP53-null cells, but the effects were greater with CF10/COXA than with 5-FU/FOXA. Treatment with CF10, 5-FU, and OXA decreased levels of Sub-G0 cells in TP53-WT vs. TP53-null HCT116 cells, which were approximately 40% vs. 30% for CF10, 32% vs. 15% for 5-FU, and 41% vs. 35% for OXA at 48 h (Supplementary Figure S4). Moreover, this population showed an overall increase in combination with single-agent OXA, with similar trends of reduced sub-G0 apoptosis in TP53-WT vs. TP53-null cells. This yielded approximately 50% vs. 35% for COXA and 40% vs. 15% for FOXA (Supplementary Figure S4C).
To further examine treatment effects on apoptosis, we analyzed the percentage of cells undergoing early and late apoptosis using a Live/Dead assay. Compared to untreated controls, COXA and CF10 induced the highest percentage of cells with both double-positive Annexin V and PI staining (late apoptosis) and AnnexinV+/PI− cells (early apotosis) (Figure 3A) while FOXA, 5-FU (10 µM), and OXA (10 µM) were less effective at inducing both early apoptosis (Figure 3B) and late apoptosis (Figure 3C). Significant differences in late apoptosis for HCT 116 TP53-WT vs. TP53-null cells were 21.75% and 8.37% (COXA), and 21.75% and 10.31% (CF10), with 5-FU displaying 10% and 6%, OXA displaying 9% and 3%, and FOXA at 12% and 6% (Figure 3B,C). The decrease in late apoptotic/DEAD cells and the increase in early apoptotic cells in the TP53-null mutant cell line, compared to TP53-WT cells at 48 h, prompted us to investigate whether apoptosis was slower in TP53-null cells. To address this, we treated WT HCT116 and TP53-null cells at 48 and 72 h, and Western blot analysis showed an increase in caspase-3 cleavage in HCT116 WT cells for OXA and COXA at 48 h, and for CF10, 5-FU, and FOXA at 72 h. Conversely, CF10, 5-FU, COXA, and FOXA treatments in p53-null cells showed the cleavage of caspase 3 at 72 h but not at 48 h via Western blot (Supplementary Figure S5). Our Western blot and Annexin V results suggest that COXA induces more and earlier caspase-3 cleavage compared to FOXA in WT HCT116 cells and that TP53-null cells undergo a slower apoptotic process.

3.3. OXA Enhances CF10-Induced Top1cc in CRC Cells
DNA topoisomerase 1 (Top1) is the sole target of camptothecin analogs [29], such as Irinotecan, which is widely used for CRC treatment. The formation of Top1 cleavage complexes (Top1cc) is also a pivotal component of the mechanism contributing to the increased cytotoxicity of CF10 compared to 5-FU [30]. Top1cc is converted to DNA DSBs through collision with the replication or transcription machinery, leading to stalled replication forks and increased replication stress [31]. To explore the importance of Top1cc for COXA cytotoxicity, we treated cells with either a single agent or drug combinations for 48 h. We assessed Top1cc formation by immunofluorescence, with DAPI staining to delineate cell nuclei and gH2AX to quantify DNA double-strand breaks (Figure 4A, Supplementary Figure S6). COXA resulted in increased Top1cc in WT and TP53-null HCT116 isogenic mutants, LS174T, and Caco-2 CRC cells compared to FOXA, as determined by CellProfiler quantification of Top1cc mean intensity per nucleus (Figure 4B). OXA co-treatment was associated with increased DSBs with both 5-FU and CF10. However, the sub-nuclear localization of DNA DSBs differed for the FOXA and COXA combinations. Specifically, COXA DSBs were primarily co-localized with Top1cc, whereas with FOXA, there was reduced co-localization with Top1cc and increased localization proximal to the nuclear envelope, particularly in TP53-WT cells. These results pinpoint an important pathway leading to lethal DNA DSBs caused by Top1cc formation, which is significantly increased with COXA compared to FOXA.

3.4. OXA Enhances CF10-Induced DNA Damage Response
The component nucleotides of CF10 (FdU, AraC), by design, are incorporated into DNA, leading to Top1cc and resulting in DNA DSBs and enhanced replication stress [30]. OXA causes intra- and interstrand DNA crosslinks, leading to DNA DSBs [32]. To further investigate enhanced DNA DSB formation with OXA co-treatment with 5-FU and CF10, we performed a Comet assay under neutral conditions (Figure 5A) and Western blots for TS and biomarkers of the DNA damage response at 48 h (pChk1/Chk1; pChk2/Chk2) (Figure 5B–D). Drug concentrations used for Western blots were as follows: CF10 at 0.01 µM, OXA at 2.5 µM, and 5-FU at 1 µM. These were chosen because higher prior doses were required in immunofluorescent assays to visualize and quantify cellular localization. In contrast, lower doses in WB enabled accurate detection of protein changes without signal saturation. Comet tails were visualized by fluorescence microscopy and quantified by Comet Analyzer for all treatments (5-FU, OXA, COXA, FOXA) relative to PBS control (Figure 5A, Supplementary Figure S7). Increased DNA tail percentages were observed in TP53-null (35%) cells compared with WT cells (25%), consistent with increased cell cycle arrest. Meanwhile, DNA tail percentages in LS174T (45%) and Caco-2 (35%) reflected relatively high and lower 5-FU drug sensitivities for FOXA treatment. Similarly, DNA tail percentages for COXA were 45% in TP53-WT vs. 42% in TP53-null cells, indicating no cell cycle arrest differences between the two (Figure 5A). Meanwhile, increased CF10 drug-sensitive LS174T and Caco-2 cells displayed DNA tail percentages of 50% and 40%, respectively. To further determine if our combinations induced greater activation of biomarkers of DNA DSBs, we performed Western blots for TS (Figure 5B; Supplementary Figure S8), Chk1/pChk1 (Figure 5C; Supplementary Figure S9), and Chk2/pChk2 (Figure 5D; Supplementary Figure S10). The literature shows that TS levels are repressed in OXA-treated TP53-WT CRC cells through the p53-miR34a axis [12]. However, we are the first to show that TS is also downregulated in TP53-null cells with OXA treatment (Figure 5B), suggesting a p53-independent mechanism. Consistent with previous studies, CF10 and 5-FU both increased TS levels, consistent with treatment-induced formation of the FdUMP/TS/reduced folate ternary complex that releases translational repression of TYMS (Thymidylate Synthase) mRNA. This effect was alleviated by OXA co-treatment with both 5-FU and CF10, particularly in TP53-WT cells. Similar to its effects on TS levels, OXA also reduced total CHK1 both as a single agent and in the FOXA and COXA combinations, and these effects were greater in TP53-WT cells than in TP53-null cells. Increased pChk1 was detected with CF10 relative to 5-FU, and with COXA relative to FOXA, with the latter effect being much greater in TP53-null cells and in WT cells, which showed OXA-induced reduction in pChk1. In contrast to its effects on TS and Chk1 levels, OXA treatment did not decrease total Chk2 levels in either WT or TP53-null cells, and, similar to pChk1 results, OXA selectively increased pChk2 in TP53-null cells but did not enhance pChk2 levels with 5-FU or CF10 in TP53-WT co-treated cells. Similar results were seen for LS174T and Caco-2 cells (Supplementary Figures S8–S10).

4. Discussion
5-FU remains an important component of therapeutic regimens for CRC, notably as an integral part of FOLFOX, a preferred frontline treatment for mCRC and also used to treat stage III and high-risk stage II CRC patients. Our lab previously demonstrated improved efficacy of CF10 relative to 5-FU across multiple in vitro and in vivo studies [19,20,31,33], consistent with CF10 being a more effective and less toxic alternative to 5-FU, suitable for clinical development. We recently reported improved antitumor activity of CF10/LV relative to 5-FU/LV in a syngeneic, orthotopic model of CRC liver metastasis, consistent with CF10 development proceeding similarly to 5-FU with respect to leucovorin co-treatment. The present study, for the first time, investigates the potential of CF10 to be combined effectively with OXA in therapeutic regimens analogous to FOLFOX. Overall, our studies confirm the potency advantage of single-agent CF10 relative to 5-FU and further establish that CF10 synergizes with OXA, with greater synergy observed for CF10 + OXA (COXA) than for 5-FU + OXA (FOXA) in CRC cells (Figure 1).
The TP53 tumor suppressor is frequently mutated or deleted in CRC [34] and in many other malignancies. TP53 status is an important determinant of cellular response to 5-FU [31], OXA [12], and other anticancer drugs. However, to the best of our knowledge, the effects of TP53 status on 5-FU + OXA synergy were not previously reported. We evaluated 5-FU + OXA synergy in an isogenic pair of CRC cell lines (HCT116 TP53+/+, TP53−/−), TP53-WT LS174T, and TP53-null Caco-2 cells, using both the highest single agent method in COMBENEFIT (Figure 1A) and the Combination Index method implemented in CompuSyn (Figure 1B). Synergy for 5-FU + OXA was observed only for a limited region of drug interaction space and only in TP53-null cells. In contrast, synergy for CF10 + OXA was observed in both WT and TP53-null CRC cells, and the magnitude of synergy was greater than that for 5-FU + OXA and spanned a broader range of drug interaction space. These findings have potential implications for the clinical development of CF10 and indicate that CF10 development could benefit from a synergistic combination with OXA, especially in patients with TP53-null tumors, which may be resistant to conventional treatments such as FOLFOX [15]. Our studies represent the first attempt to exploit the increased DNA-directed activity of the polymeric FP CF10 compared with 5-FU to develop a novel, highly effective combination treatment for CRC. Likewise, the potential clinical benefits of COXA versus FOXA in TP53-deficient patients may increase responses currently unattainable, given the strong TP53 dependence of 5-FU and the lack of therapeutic benefit in patient survival with the addition of OXA. Further, our potential to use lower doses of OXA and CF10 in patients could significantly mitigate OXA-induced neurotoxicity and the unwanted metabolites observed with 5-FU co-treatment.
Our findings that synergy was not detected for 5-FU + OXA in TP53+/+ cells, with evidence of antagonism in some regions of the drug interaction space (Figure 1A), prompted us to investigate the cell-cycle dependence of drug combination treatments. 5-FU stabilized p53 and induced p53-dependent G1 arrest at 24 h, consistent with previous studies (Figure 2), and similar effects were also observed for the FOXA combination at 24 h. Theoretically, G1/S arrest could prevent TP53-WT cells from entering S-phase and undergoing replication-mediated DNA damage, which could explain the limited synergistic interactions with OXA. In contrast, although CF10 and COXA strongly stabilized p53 and induced p21 (Figure 2A,B), no significant p53-dependent cell-cycle effects were detected with these treatments, and they induced predominantly S-phase arrest regardless of TP53 status (Figure 2C,D). Our cell cycle results are consistent with other literature findings regarding 5-FU effects on p53-null CRC organoids [33]. Since p53 is also important for activating apoptosis, we evaluated whether there were p53-dependent differences in apoptotic fractions using a Live/Dead assay. CF10 and COXA were much more effective than 5-FU and FOXA at inducing apoptosis based on AnnexinV expression, regardless of p53 expression. However, a greater percentage of WT-TP53 cells were also permeable to PI, suggesting that CF10-induced apoptosis is more rapid in WT-TP53 cells (Figure 3). These findings were confirmed by Western blot for cleaved caspase 3, which was detected at earlier timepoints in CF10- and COXA-treated TP53-WT cells than in TP53-null cells (Supplementary Figure S5). Overall, these findings support CF10 as a significantly better inducer of apoptosis in CRC cells, accounting for the increased sensitivity observed in dose–response curves compared to 5-FU, especially in combination with OXA.
Mechanistically, CF10 and COXA’s potency advantage relative to 5-FU and FOXA in CRC cells is associated with increased Top1cc formation (Figure 4A,B) that coincides with increased DNA DSBs (Figure 4A,C and Figure 5A), consistent with increased replication stress [35]. The causes of replication stress are multifaceted and include deoxynucleotide depletion and imbalance that result from TYMS repression and TS inhibition. In previous studies, we demonstrated that CF10’s increased potency relative to 5-FU is associated with greater TS inhibition, facilitating Top1cc formation by misincorporating FdU into DNA under thymidine-depleted conditions, thereby inhibiting the religation step of Top1 catalysis. In the current study, we show that COXA induced significantly greater Top1cc formation compared to FOXA in both TP53-WT and p53-null cells (Figure 4A,B, Supplementary Figure S6). The addition of OXA to CF10 reduced TS levels in both TP53-WT and p53-null cells, as shown by Western blot (Figure 5B), creating a thymine-deficient environment that better encourages the formation of Top1ccs. The mechanism by which OXA downregulates TS, although controlled by the p53-miR24a axis to our knowledge, appears to be active in TP53-null HCT116 and Caco-2 cells, suggesting an alternative mechanism that controls this downregulation. To the best of our knowledge, we are the first to report it. In addition to decreasing TS levels, OXA decreased total CHK1 (Figure 5C), which is strongly activated by CF10 and COXA, potentially contributing to decreased activation of the intra-S-phase checkpoint and to cells with extensive DNA damage advancing into mitosis. Our results, suggesting that OXA induces downregulation of total CHK1, p-Chk1, and p-Chk2 but not total CHK2, have been previously reported [36]. However, to our knowledge, this has not been reported for p53-deficient cell lines. Interestingly, single-agent OXA also induced Top1cc, indicating that at least two different types of therapy-induced Top1cc are present in CRC cells treated with COXA. Overall, Top1cc and DNA DSBs are slightly decreased in TP53-null cells relative to wild-type (Figure 4A,B), consistent with these cells showing slightly reduced sensitivity to CF10 [33] and COXA (Figure 1C). However, TP53 dependence, as it pertains to Top1cc formation and DNA DSBs, is reduced relative to 5-FU and FOXA.
Overall, this study addresses the need for improved treatment options for advanced and metastatic CRC, which are strongly associated with p53 mutation or null status. Although a significant limitation of this study is the lack of an in vivo model demonstrating the improved antitumor activity of COXA compared to FOXA, follow-up studies will include evaluation of COXA in a more clinically relevant setting, such as a syngeneic orthotopic CRC liver metastasis model [19]. The present mechanistic studies will inform future efforts to further uncover the enhanced benefits of CF10 as a novel FP for the treatment of CRC.

5. Conclusions
Overall, our findings demonstrate that synergy with CF10 + OXA (COXA) is stronger than for 5-FU + OXA (FOXA) and is evident in TP53-WT and TP53-null CRC cells, while 5-FU + OXA synergy is restricted to TP53-null cells. The mechanistic basis for CF10 + OXA synergy involves enhanced Top1-mediated DNA damage, in part through OXA-mediated downregulation of TS and thymineless conditions, presumably promoting increased FdUTP incorporation into DNA and leading to enhanced Top1cc. The increased synergy for COXA in conjunction with our previous studies that demonstrated that CF10 + LV is more potent than 5-FU + LV to CRC cells and is highly effective in mouse models of CRC including liver-metastatic disease with less systemic toxicity demonstrate that CF10 clinical development could proceed using similar combinatorial strategies that have enabled 5-FU-based chemotherapy to become a preferred treatment option for treatment of locally advanced and mCRC.