The HCT116 human colon cancer cell line was purchased from the National Collection of Authenticated Cell Cultures of China (Shanghai, China). HCT116 cells were cultured in McCoy’s 5A medium (Wisent, Nanjing, China) supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA), 100 μg/ml streptomycin, and 100 units/ml penicillin G in 5% CO₂ at 37°C and subcultured by trypsinization when confluent.

Inactivation of the GSTA4 gene was carried out using the lentiCRISPRv2 system (Addgene, MA, USA) (24). Guide RNA (gRNA) sequences were designed to target the human GSTA4 gene using CRISPRdirect (https://crispr.dbcls.jp) (25). Selected gRNAs ( Supplementary Table S1 ) were cloned into the pLentiCRISPRv2 vector and transformed into E. coli Stbl3 competent cells (TransGen Biotech, Beijing, China) as previously reported (24, 26). The positive transformants were screened by PCR and plasmids were extracted using a TaKaRa MiniBEST Plasmid Purification Kit (Takara, Dalian, China). To delete a large fragment in the GSTA4 gene, HCT116 cells were co-transfected with gRNA66 and gRNA214 ( Supplementary Table S1 and Figure 1 ) using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, USA) according to the instructions of the manufacturer. Transfected cells were screened by 4 µg/ml puromycin followed by a maintenance concentration of 1 µg/ml puromycin. Clones surviving at 1 µg/ml puromycin were sub-cultured in a 24-well plate. DNA was extracted for mutant screening. Deletion of the GSTA4 gene fragment was detected by PCR amplification using specific primers ( Supplementary Table S1 ) followed by Sanger sequencing.

Cell viability was determined using the CCK-8 assay kit (Beyotime Biotechnology, Shanghai, China). Cells were seeded in 96-well plates (5 × 10³ cells/well) and incubated overnight at 37°C. Following a 48-hour treatment with 5-FU or oxaliplatin, 10 μl of CCK-8 solution was added to each well and incubated at 37°C for 4 h. The absorbance was measured at 450 nm using a Varioskan LUX microplate reader (Thermo Fisher Scientific, MA, USA).

Cell proliferation was determined using the BeyoClick™ EdU-555 kit (Beyotime Biotechnology) according to the instructions of the manufacturer. Briefly, cells were seeded in 6-well plates at a density of 2.5 × 10⁵ cells/well and incubated at 37°C overnight. Cells were treated with 5-FU or oxaliplatin for 48 h and incubated with 10 μM 5-ethynyl-2′-deoxyuridine (EdU) at 37°C for 2 h. Cells were then harvested by trypsinization, fixed with 4% paraformaldehyde at room temperature (RT) for 15 min, and permeabilized with 0.3% Triton X-100 in PBS at RT for 10 min. Cells were washed, stained with 500 μl of Click Additive Solution at RT for 30 min, and sorted on a DxFLEX flow cytometer (Beckman Coulter, Suzhou, China). EdU-positive cells were analyzed using CytExpert software.

Cells were seeded into 6‐well plates at a density of 1,000 cells/well and cultured in 5% CO₂ at 37°C for 10 days. The clones were fixed with methanol and stained with 0.1% crystal violet for 30 min. The number of clones was counted.

Apoptosis was analyzed using an Annexin-V/PI Staining kit (Vazyme, Nanjing, China) according to the instructions of the manufacturer. In brief, cells were harvested by trypsinization, washed with PBS, and counted. Cells (1 × 10⁵) were resuspended in 100 µl binding buffer and incubated with 5 µl Annexin V-FITC and 5 µl propidium iodide (PI) at RT for 10 min. Following staining, 400 µl binding buffer was added and cells were sorted on a DxFLEX flow cytometer (Beckman Coulter). The percentage of total apoptotic cells was calculated.

ROS was detected using a ROS Assay Kit (Beyotime Biotechnology). Briefly, cells were treated with PBS, 5-FU, and oxaliplatin, respectively, for 48 h. Following treatment, 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA) was added to a final concentration of 10 μM and incubated at 37°C for 30 min. Cells were harvested, washed with PBS, and re-seeded into a 96-well plate at a density of 2 × 10⁴ cells/well. The fluorescence intensity was measured at 525 nm using an excitation wavelength of 488 nm at the Varioskan LUX microplate reader (Thermo Fisher Scientific). The intensity of fluorescence was normalized to the fluorescence intensity of the PBS control.

Western blotting was performed as described previously (6). Cells were lysed using lysis buffer and the concentration was analyzed by the BCA assay (Beyotime Biotechnology). Whole-cell lysates were separated by SDS-PAGE, transferred to a PVDF membrane (Millipore, Shanghai, China), and blocked with 5% non-fat dry milk in Tris-buffered saline with 0.1% Tween 20 (TBST). The primary antibodies used in the study included anti-AKT (1:1,000, Affinity Biosciences, Changzhou, China), anti-p-AKT (1:1,000, Cell Signaling Technology, Shanghai, China), anti-p38 MAP kinase (1:1,000, Affinity Biosciences), anti-phosphorylated p38 (1:1,000, Cell Signaling Technology), and anti-β-actin (1:2,000, Sangon, Shanghai, China). HRP-conjugated species-specific secondary antibodies were purchased from Beyotime Biotechnology. Signals were generated by enhanced chemiluminescence (Biosharp, Hefei, China) and captured by the Odyssey Fc system (LI-COR Biosciences, Lincoln, NE).

The animal study was approved by the Institutional Animal Care and Use Committees at Nantong University. Male, 6‐week‐old specific pathogen-free BALB/cJGpt-Foxn1^nu /Gpt nude mice were used for tumor xenograft. A total of 30 mice were randomly and equally divided into six groups, with 5 mice per group. HCT116 (3 groups) or GSTA4-deficient HCT116 (HCT116^ΔGSTA4) cells (3 groups) were subcutaneously injected into the flanks of the mice (5 × 10⁶ cells/side) for xenograft tumor growth. One week after xenografting, mice were intraperitoneally injected with 5-FU (30 mg/kg), oxaliplatin (3 mg/kg), and saline as control, once a week for 4 weeks. Mouse body weight was recorded and tumor size was measured weekly. The tumor size was calculated as described previously (27). Mice were euthanized after 4-week treatment and tumors were removed for measuring size, weighing, and histology.

Data are expressed as means ± SEM from at least three independent experiments. Student’s t-test or two-way ANOVA was used for comparisons between groups using the GraphPad Prism 8 software (GraphPad Software, CA, USA). P <0.05 was considered statistically significant.

To inactivate GSTA4 in HCT116 cells, we co-transfected HCT116 cells with pLentiCRISPR-gRNA66 and pLentiCRISPR-gRNA214 ( Supplementary Table S1 ), which resulted in a 1,135 bp homozygous deletion across exons 1 and 2 in the GSTA4 gene ( Figure 1A ). Deletion of this fragment was confirmed by PCR amplification and sequence analysis ( Figure 1B and Supplementary Figure S1 ). This GSTA4-deficient clone was named HCT116^ΔGSTA4 and was used for the subsequent studies. Because GSTA4 is a detoxifying enzyme for 4-HNE, inactivation of GSTA4 in HCT116^ΔGSTA4 was further confirmed by a decreased detoxifying capability for 4-HNE. The viability of HCT116^ΔGSTA4 cells significantly decreased compared to the parental HCT116 cells when exposed to various concentrations of 4-HNE for 48 h ( Figure 1C ), indicating a reduced 4-HNE-detoxifying ability of HCT116^ΔGSTA4 cells.

To investigate the effect of GSTA4 inactivation, we first determined the cell proliferation of HCT116^ΔGSTA4 cells and compared them to the parental HCT116 cells. Inactivation of GSTA4 significantly reduced cell proliferation after culturing for 24, 48, and 72 h compared with HCT116 cells ( Figure 2A ). Furthermore, clonogenic assays demonstrated that the size and number of clones formed by HCT116^ΔGSTA4 cells remarkably decreased compared with clones formed by HCT116 cells ( Figures 2B, C ). Additionally, decreased proliferation of HCT116^ΔGSTA4 cells was confirmed by the EdU incorporation assay ( Figures 2D, E ). The fluorescence-activated cell sorting (FACS) analysis showed that the proportion of EdU-positive cells significantly decreased for HCT116^ΔGSTA4 cells compared with HCT116 cells (32.6 ± 2.9% vs. 39.1 ± 4.6%, P <0.05). Finally, apoptosis was analyzed to determine whether the inactivation of GSTA4 induced apoptosis. FACS analysis showed that, when cells left untreated, there were no significant differences in the proportions of the early, late, and total apoptotic cells for HCT116^ΔGSTA4 cells compared to those of HCT116 cells ( Supplementary Figure S2 , Ctrl). These results suggest that GSTA4 contributes to cancer cell proliferation while having no effect on apoptosis.

GSTs, as phase II detoxifying enzymes, are associated with anti-cancer drug resistance (28). To investigate whether GSTA4 is involved in chemoresistance, we determined cell survival rates for HCT116 and HCT116^ΔGSTA4 cells after exposure to 5-FU and oxaliplatin, two first-line chemotherapeutic agents for colorectal cancer. As shown in Figure 3 , the survival rates significantly decreased for HCT116^ΔGSTA4 cells compared with HCT116 cells following treatment with 5-FU at the doses of 5, 15, and 30 μM for 48 h. Similarly, significantly decreased survival rates were also noted for HCT116^ΔGSTA4 cells exposed to oxaliplatin at the doses of 0.5 and 5 µM for 48 h. Notably, we were unable to find significantly increased proportions of early, late, and total apoptosis in HCT116 cells exposed to 5-FU (5 μM) and oxaliplatin (5 μM) compared with untreated controls ( Figure S2A–D ). Furthermore, inactivation of GSTA4 (HCT116^ΔGSTA4) did not increase apoptosis compared with the parental HCT116 cells when treated with 5-FU (5 μM) and oxaliplatin (5 μM), suggesting that loss of GSTA4 augmented chemotherapeutic reduction of cell viability without affecting apoptosis.

We next examined cell proliferation using the EdU incorporation assay following treatment with 5-FU and oxaliplatin. In comparison with untreated controls, the proportion of EdU-positive cells decreased for both HCT116 and HCT116^ΔGSTA4 cells after treatment with 5-FU and oxaliplatin. Treatment with 5-FU (5 µM) for 48 h resulted in a 56.28 ± 4.93% decreased proportion of EdU-positive cells (from 32.59 ± 1.56% to 13.24 ± 0.98%, P <0.01) for HCT116^ΔGSTA4 cells compared to a 19.22 ± 6.82% decrease (from 40.46 ± 2.40% to 30.53 ± 4.08%, P <0.05) for HCT116 cells ( Figures 4A–C ). Comparably, treatment with oxaliplatin (0.5 µM) for 48 h also resulted in a 56.77 ± 17.65% decreased proportion of EdU-positive cells (from 33.63 ± 4.04% to 10.01 ± 8.87%, P <0.01) for HCT116^ΔGSTA4 cells compared to a 11.77 ± 9.79% decrease (from 38.40 ± 5.39% to 18.86 ± 14.21%, P <0.05) for HCT116 cells ( Figures 4D–F ). These results indicate that inactivation of GSTA4 increases the sensitivity of colorectal cancer cells to chemotherapeutic agents via inhibition of proliferation.

Many chemotherapeutic agents act as DNA damage inducers, leading to cancer cell death. Inactivation of GSTs often induces ROS production that causes DNA damage (29). However, chemotherapeutic agents may also modulate GST expression and ROS production (5, 30). Therefore, we measured ROS production in HCT116^ΔGSTA4 cells compared with that in HCT116 cells and found that inactivation of GSTA4 increased ROS production in untreated HCT116^ΔGSTA4 cells in contrast to HCT116 cells. Furthermore, ROS production also increased significantly in HCT116^ΔGSTA4 cells treated with 5-FU and oxaliplatin compared with HCT116 cells treated with the same drugs ( Figure 5A ). To determine whether the inactivation of GSTA4 promotes DNA damage, we analyzed phosphorylated H2AX (γH2AX), a hallmark of double-stranded DNA breaks, by FACS analysis following treatment with 5-FU and oxaliplatin. No significant change was seen in the proportion of γH2AX-positive cells between untreated HCT116^ΔGSTA4 and HCT116 cells (0.7 ± 0.82% vs 0.84 ± 0.98%, P = 0.859), suggesting that loss of GSTA4 has no effect on DNA damage in unstressed cells. Following a 48-hour treatment with 5-FU, the proportion of γH2AX-positive cells increased to 12.73 ± 1.06% and 4.23 ± 0.28% for HCT116^ΔGSTA4 and HCT116 cells, respectively ( Figures 5B, C , P <0.01). In contrast, although the proportion of γH2AX-positive cells increased for both HCT116^ΔGSTA4 (1.67 ± 2.42%) and HCT116 cells (3.69 ± 6.00%) after treatment with oxaliplatin for 48 h compared with untreated controls, no significant difference in the proportion of γH2AX-positive cells was seen between HCT116^ΔGSTA4 and HCT116 cells ( Supplementary Figures 3A, B , P = 0.618), indicating that the inactivation of GSTA4 has limited effect on low-dose oxaliplatin-induced crosslinking of DNA.

To further investigate the mechanisms by which inactivation of GSTA4 inhibits proliferation, we examined proliferative signaling pathways including phosphoinositide 3 kinase (PI3K) and MAPK. As shown in Figure 6 , no remarkable change was noted in the expression of AKT (a downstream kinase in the PI3K signaling pathway) and p38 MAPK. However, phosphorylated AKT (p-AKT) and phosphorylated p38 (p-p38), the activated forms of AKT and p38, remarkably decreased in HCT116^ΔGSTA4 cells compared with HCT116 cells, indicating inhibition of AKT and p38 signaling pathways. Additionally, treatment of HCT116 cells with 5-FU and oxaliplatin slightly decreased p-AKT and p-p38 pathways in HCT116 cells while being prone to increasing p-AKT and p-p38 in HCT116^ΔGSTA4 cells. These results suggest that inactivation of GSTA4 reduces proliferation by blocking PI3K/AKT and p38 MAPK signaling pathways and that activation of AKT and p38 pathways in HCT116 CRC cells occurs in a GSTA4-dependent manner.

Finally, we investigated the effect of GSTA4-deficiency on tumorigenesis using a xenograft tumor model. We injected nude mice with 5 × 10⁶ cells per side of the flank and treated them with saline, 5-FU, and oxaliplatin, respectively. For saline-treated mice, the average tumor size significantly decreased for HCT116^ΔGSTA4-derived xenografts compared with HCT116-derived xenografts (P <0.001), indicating reduced tumor growth for GSTA4-deficient HCT116 cells ( Figure 7A ). For mice treated with 5-FU and oxaliplatin, the average tumor size also significantly decreased for xenografts derived from HCT116^ΔGSTA4 cells compared with HCT116 cells ( Figures 7B, C , P <0.05 for both 5-FU and oxaliplatin groups). Of note, both 5-FU and oxaliplatin reduced tumor size in HCT116-derived xenografts compared with saline control. However, no significant difference was found between 5-FU and oxaliplatin ( Figure 7D , P = 0.75). In contrast, for HCT116^ΔGSTA4-derived xenografts, 5-FU, but not oxaliplatin, remarkably reduced tumor size compared to saline control ( Figure 7E , P <0.01 and P = 0.22 for 5-FU and oxaliplatin, respectively). This is consistent with the findings that 5-FU, but not oxaliplatin, at the tested doses, induced more γH2AX foci in HCT116^ΔGSTA4 cells. These results indicate that GSTA4 promotes in vivo tumorigenesis and contributes to chemoresistance.