The DDW (deuterium %=25 ppm) used in the present study was provided by HYD LLC. Regular ddH₂O or DDW was used to dissolve RPMI-1640 medium powder (Gibco; Thermo Fisher Scientific, Inc.). A total of two CRC cell lines, HT-29 (cat. no. STCC10801G) and DLD-1 (cat. no. STCC10810G), were purchased from Wuhan Servicebio Technology Co., Ltd. and authenticated by STR profiling. Cells were cultured in conditioned RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin solution (Sigma-Aldrich; Merck KGaA) at 37°C in a humidified atmosphere containing 5% CO₂. The control group contained cells that were cultured in medium prepared with regular ddH₂O, whereas the cells in the DDW group were cultured in medium prepared with DDW.

The EdU assay was performed using a commercial kit (cat. no. C10310-1; Guangzhou RiboBio Co., Ltd.). HT-29 and DLD-1 cells were seeded into 6-well plates at a density of 1×10⁵ cells per well. After culturing in serum-free medium for 72 h, cells were incubated with 0.1% EdU solution for 3 h at 37°C, then fixed with 4% paraformaldehyde (PFA) for 30 min at room temperature. Cells were then incubated with appollo staining complex for 30 min at room temperature to react with EdU, followed by incubation with Hoechst 33342 staining reagent for 30 min at room temperature to label the cell nuclei. Fluorescent staining was recorded using the Operetta CLS high-content analysis system (PerkinElmer, Inc.). A total of five visual fields were randomly selected and the number of EdU positive cells was assessed using ImageJ software (version 1.53 K; National Institutes of Health). The following formula was used to determine the percentage of EdU positive cells: EdU positive cells (%)=(EdU positive cells/total cells) ×100%.

HT-29 and DLD-1 cells were seeded into 6-well plates at a density of 1×103 cells per well and cultured with conditioned medium. The medium was changed every 3 days. After 10 days, the cells were fixed with 4% PFA for 30 min at room temperature and then stained with 0.1% crystal violet for 20 min at room temperature. The colonies consisting of at least 40 cells were counted under an inverted light microscope (ECLIPSE Ts2; Nikon Corporation) using ImageJ software.

DMEM powder and F-12 medium powder (both from Gibco; Thermo Fisher Scientific, Inc.) were dissolved with regular ddH2O or DDW. Both media were mixed at a ratio of 1:1 to prepare the DMEM/F-12 medium. HT-29 and DLD-1 cells were seeded into ultra-low attachment 24-well plates (Corning, Inc.) at a density of 1×10³ cells per well and cultured in DMEM/F12 medium supplemented with 1% B27, 20 ng/ml human FGF and 20 ng/ml human EGF (Invitrogen™; Thermo Fisher Scientific, Inc). Fresh medium was added every 3 days. After 2 weeks, cell spheres were counted under an inverted light microscope (ECLIPSE Ts2; Nikon Corporation), and the sphere formation efficiency (SFE) was calculated using the following formula: SFE (%)=number of spheres/1,000 ×100%.

HT-29 and DLD-1 cells were cultured in 6-well plates until they formed confluent monolayers. A gap was created across the middle of the well using a 200 µl sterile tip. The cells were then cultured in serum-free medium for an additional 48 h. An inverted light microscope (ECLIPSE Ts2; Nikon Corporation) was used to capture images both before and after the 48-h incubation period to evaluate the migration rate. Experiments were repeated in quintuplicate. A visual field was randomly selected, and the same field was imaged for every time point. The area of blank space was measured using ImageJ software. The following formula was used to assess the wound healed ratio: Wound healed ratio (%)=[1-(area^{48 h}/area^{0 h})] ×100%.

HT-29 and DLD-1 cells were suspended in serum-free medium. The inner membranes of Transwell inserts (cat. no. 3464; Corning, Inc.) were coated with 10% Matrigel (cat. no. 354234; Corning, Inc.), followed by the seeding of 2×10⁴ cells with 200 µl serum-free DMEM into each upper chamber. A total of 750 µl DMEM supplemented with 10% FBS was added to each lower chamber of the 24-well plate. After incubating for 24 h at 37°C, the cells were fixed with 4% PFA for 30 min at room temperature and then stained with 0.1% crystal violet for 20 min at room temperature. Cells that successfully invaded the bottom side of membrane were visualized using an inverted light microscope (ECLIPSE Ts2; Nikon Corporation).

Intracellular ROS levels were assessed using the 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) method. HT-29 and DLD-1 cells were incubated with the DCFH-DA probes (Sigma-Aldrich; Merck KGaA) for 30 min at 37°C and then treated with H2O2 (50 µM) for 30 min at 37°C. Cells were then fixed with 4% PFA for 30 min at room temperature. The fluorescence intensity of DCF was recorded using the Operetta CLS High-Content Analysis System (PerkinElmer, Inc.).

Total RNA was isolated using TRIzol™ reagent (Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer's instructions. RT-qPCR was performed using the SuperScript™ III Platinum™ One-Step qRT-PCR Kit (cat. no. 11732020; Invitrogen; Thermo Fisher Scientific, Inc.). The sequences of the primers used for qPCR were as follows: FoxM1 forward, 5′-ACGTCCCCAAGCCAGGCTC-3′ and reverse, 5′-CTACTGTAGCTCAGGAATAA-3′; cyclin D1 (CCND1) forward, 5′-CGCTTCCTGTCGCTGGAGCC-3′ and reverse, 5′-CTTCTCGGCCGTCAGGGGGA-3′; matrix metalloproteinase 9 (MMP9) forward, 5′-TTCTGCCCGGACCAAGGATA-3′ and reverse, 5′-ACATAGGGTACATGAGCGCC-3′; and β-actin (reference gene) forward, 5′-CATGTACGTTGCTATCCAGGC-3′ and reverse, 5′-CTCCTTAATGTCACGCACGAT-3′. The thermocycling conditions were as follows: 50°C for 15 min and 95°C for 2 min, and then 40 cycles of 95°C for 15 sec and 60°C for 30 sec. The relative gene expression was calculated using the 2^−∆∆Cq method (18).

FoxM1-overexpressing (OE-FoxM1) and negative control (OE-NC) lentiviral particles (1×10⁸ TU/ml) were purchased from Shanghai GeneChem Co., Ltd., both originating from the same plasmid backbone (Ubi-MCS-3FLAG-SV40-puromycin). HT-29 and DLD-1 cells were seeded into 6-well plates at a density of 1×10⁵ cells per well and incubated overnight. HT-29 and DLD-1 cells were transduced with the specific lentivirus at a multiplicity of infection of 10. After a 24-h incubation, fresh medium was changed to terminate the transduction. Stable infected cells were selected by puromycin (5 µg/ml) incubation for 48 h, and the surviving cells were continuously maintained in fresh DMEM with 10% FBS and 1.25 µg/ml puromycin for passaging. FoxM1 expression in the recombinant cells was validated by RT-qPCR and western blotting before subsequent experiments.

Total proteins from HT-29 and DLD-1 cells were extracted using RIPA lysis buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) and quantified using the BCA Protein Assay Kit (cat. no. P0012; Beyotime Institute of Biotechnology). Proteins (40 µg per lane) were separated by 4–20% SDS-PAGE (cat. no. P0468S; Beyotime Institute of Biotechnology) and subsequently transferred onto PVDF membranes (MilliporeSigma). The membranes were blocked with 5% fat-free milk (Bio-Rad Laboratories, Inc.) for 1 h at room temperature and incubated with primary antibodies (1:1,000 dilution by PBS) against β-actin (cat. no. 66009-1-Ig), octamer-binding transcription factor-4 (OCT-4; cat. no. 11263-1-AP), Nanog (cat. no. 14295-1-AP), MMP-9 (cat. no. 10375-2-AP), CCND1 (cat. no. 26939-1-AP) and FoxM1 (cat. no. 13147-1-AP) (all Proteintech Group, Inc.) at 4°C overnight. Protein signals were labeled with HRP-conjugated secondary antibodies (1:5,000 dilution by PBS; cat. nos. SA00001-1 or SA00001-2; Proteintech Group, Inc.) for 1 h at room temperature and visualized using an ECL kit (cat. no. WBKLS0500; MilliporeSigma). The densitometry was measured using ImageJ software.

Continuous data are expressed as mean ± standard deviation. Differences between two groups were analyzed using the unpaired Student's t-test or Mann-Whitney U test as appropriate. One-way ANOVA was used to compare the mean differences among three or more groups, and the least significant difference test was used as a post-hoc test. The data analysis for this paper was generated using GraphPad Prism 10 software (Dotmatics). P<0.05 was considered to indicate a statistically significant difference.

To assess the biological functions of DDW in CRC malignancy, HT-29 and DLD-1 cells were cultured in conditioned medium prepared with regular ddH₂O or DDW. It was demonstrated that, compared with the control (regular ddH₂O) treatment, DDW treatment reduced the proportion of EdU-positive cells (Fig. 1A) and the number of colonies formed (Fig. 1B). These results indicated that DDW could inhibit the proliferation of CRC cells in vitro. Notably, the results also revealed that DDW treatment impaired the sphere formation efficiency of HT-29 and DLD-1 cells (Fig. 1C). This finding implied that DDW might interfere with the stem cell-like phenotype of CRC cells. Therefore, the expression levels of two typical stemness markers were analyzed using western blotting. The findings demonstrated that, compared with the control group, the protein levels of Nanog and OCT-4 were significantly downregulated in the DDW treatment group (Fig. 1D). These results indicated that DDW treatment impaired the self-renewal ability of HT-29 and DLD-1 cells.

Subsequently, wound healing and Transwell assays were performed to evaluate the effect of DDW on cancer cell mobility. Compared with the control treatment, the migrated distances of HT-29 and DLD-1 cells were significantly reduced when treated with DDW (Fig. 2A and C). Additionally, there was a notable reduction in the number of invaded cells for both the HT-29 and DLD-1 cells treated by DDW in the Transwell assays (Fig. 2B and D). These findings suggested that DDW treatment is associated with a significant reduction in CRC cell migration and invasion.

To determine whether the ROS/FoxM1 axis was involved in the aforementioned findings, ROS production in response to DDW treatment was evaluated in the CRC cells. The findings revealed that, compared with the control treatment, DDW treatment significantly reduced the ROS levels in the HT-29 and DLD-1 cell lines (Fig. 3A). Subsequently, RT-qPCR and western blotting were used to evaluate FoxM1 expression. Compared with the control treatment, DDW treatment significantly reduced the mRNA and protein levels of FoxM1 (Fig. 3B and C). Moreover, the expression levels of CCND1 and MMP9, two typical downstream targets of FoxM1 (19,20), were lower in HT-29 and DLD-1 cells treated with DDW, compared with the control treatment (Fig. 3B and C). Furthermore, H₂O₂ was utilized to induce intracellular ROS production. Compared with the cells treated by DDW alone, additional treatment with H₂O₂ not only blocked the antioxidant effect of DDW (Fig. 3A) but also rescued the expression of FoxM1 (Fig. 3D). These results indicated that DDW treatment downregulated FoxM1 expression possibly through inhibiting ROS production.

To assess whether the anticancer effects of DDW were dependent on the inhibition of FoxM1, FoxM1 was overexpressed in HT-29 and DLD-1 cells using lentiviral transduction. The mRNA and protein expression levels were significantly upregulated in OE-FoxM1 cells compared with the OE-NC cells (Fig. 4). Furthermore, compared with the OE-NC lentiviral transduction, the inhibitory effects of DDW on the proliferation (Fig. 5A and B), stemness (Fig. 5C and D), migration (Fig. 6A) and invasion (Fig. 6B) of HT-29 and DLD-1 cells were significantly abrogated by OE-FoxM1 lentiviral transduction. In addition, overexpression of FoxM1 also rescued the expression levels of CCND1 and MMP-9 in CRC cells treated with DDW, as demonstrated using RT-qPCR (Fig. 7A and C) and western blotting (Fig. 7B and D). These results validated that DDW may exert its tumor suppressive effects through inhibiting FoxM1 expression.