Abstract
Background/Aim: Ferroptosis is a nonapoptotic type of cell death that is dependent on iron and involves the accumulation of reactive oxygen species. Ferroptosis suppressor protein 1 (FSP1) and glutathione peroxidase 4 (GPX4) are ferroptosis regulators that inhibit ferroptosis through independent pathways. This study assessed the prognostic value of GPX4 and FSP1 expression in colorectal cancer (CRC). We also examined the effects of FSP1 and GPX4 inhibition on cell survival of CRC cells. Materials and Methods: This study included 206 surgical specimens from Stage II or III CRC patients. FSP1 and GPX4 expression was analyzed immunohistochemically, and the association of their expression levels with clinical outcome was evaluated. We also examined the effects of FSP1 and GPX4 inhibitors on the cell proliferative capacity of CRC cell lines. Results: Overall survival and recurrence-free survival were reduced in patients with high expression of FSP1 or GPX4, and those with both GPX4 and FSP1 expression showed worse prognosis. Positivity of both FSP1 and GPX4 was an independent poor prognostic factor for CRC patients. In CRC cells, the combination of GPX4 and FSP1 inhibitors led to more effective cell death than either inhibitor alone. Conclusion: High expression of both GPX4 and FSP1 is a significant poor prognostic factor for CRC. Simultaneous inhibition of GPX4 and FSP1 to induce ferroptosis may be a novel therapeutic strategy in CRC.
Colorectal cancer (CRC) is the second most common cause of cancer-related mortality and the third most frequent cancer worldwide (1). The incidence of CRC is rapidly increasing, with over 1.1 million deaths and over 2.2 million cases expected in 2030 (2). Screening programs and endoscopic treatment have enabled early detection, and advances in perioperative management in surgery and preoperative and postoperative chemotherapy and radiation therapy have improved the prognosis for CRC patients. However, approximately 15%-20% of patients with CRC experience distant metastases or local recurrence even after receiving adjuvant chemotherapy and extensive resection (3, 4). Therefore, better understanding of the mechanisms of carcinogenesis and progression of CRC is required to help develop novel therapeutic targets.
Ferroptosis is an iron-induced programmed cell death in which uncontrolled high production of reactive oxygen species (ROS) causes lipid peroxide accumulation and oxidative damage to cell membranes (5). Because ferroptosis differs from the cell death mechanisms of apoptosis and necrosis, ferroptosis can be induced to cause cell death even in apoptosis-resistant cancer cells (5). For this reason, novel chemotherapeutic strategies targeting ferroptosis are attracting attention (6).
Ferroptosis is negatively regulated by glutathione peroxidase 4 (GPX4), an antioxidant that uses reduced glutathione to directly reduce phospholipid hydroperoxides and protect cells from ferroptosis (7). Studies have shown that the poor prognosis of many carcinomas is correlated with high GPX4 expression (8-11). In 2019, studies showed that apoptosis-inducing factor mitochondria 2 (AIFM2) exhibited functions in suppressing ferroptosis; it was thus renamed as ferroptosis regulatory protein 1 (FSP1) (12-14). Distinct from GPX4, FSP1 reduces coenzyme Q to produce ubiquinol and prevents lipid peroxidation of cell membranes, thereby regulating ferroptosis independently of glutathione (13-15). Therefore, targeting both FSP1 and GPX4, which inhibit ferroptosis through different pathways, is expected to lead to additional therapeutic possibilities. Previous studies have shown that the expression of GPX4 in CRC is related to poor prognosis (16, 17). However, few studies have been published on the significance of FSP1 expression. Furthermore, the prognostic significance of the co-expression of GPX4 and FSP1 and their correlations with clinicopathological features in CRC patients is unclear. To this end, this study examined the levels of FSP1 and GPX4 expression in CRC patients and assessed the predictive value of their expression patterns for patient prognoses. Furthermore, we examined the effects of inhibition of ferroptosis in CRC cells by suppressing both FSP1 and GPX4.
Materials and Methods
Patients and study specimens. Tumor specimens were collected from 206 consecutive patients who received curative surgery for CRC at the Tottori University Hospital from January 2007 to December 2012. The patients were categorized as Stage II or III using the 8th Edition of the International Cancer Control-TNM classification. Patients who were lost to follow-up or who underwent emergency surgery were excluded. Patients were followed every three months until five years had passed since their last surgery or until death. This study was approved by the Tottori University Institutional Review Board (No. 23A111).
Immunohistochemical analysis. Immunohistochemistry was performed using a Histostainer-36A (Nichirei Biosciences, Tokyo, Japan). Paraffin embedded and formalin-fixed tissues were cut into 4-μm-thick slices. After deparaffinization, the samples were autoclaved in 10 mM citrate buffer (pH 6.0) at 121°C for 10 min. The sections were blocked with 3% H2O2 for 5 min, followed by incubation with primary antibodies for 60 min. The primary antibodies were antibodies against FSP1 (20886; ProteinTech Group, Chicago, IL, USA) and GPX4 (ab125066; Abcam, Cambridge, UK); both were used at a dilution of 1:100. The slides were then incubated with secondary antibody MAX-PO(MULTI) (Nichirei Biosciences) for 30 min, stained with DAB (Nichirei Biosciences), and counterstained with hematoxylin.
A blinded assessment was conducted on the expression of FSP1 and GPX4 in tissue samples. FSP1 expression was classified as follows: <40% stained cells, FSP1 negative; and ≥40% stained cells, FSP1 positive. GPX4 expression was classified as follows: <50% stained cells, GPX4 negative; and ≥50% stained cells, GPX4 positive. Two of the authors (C.Y. and Y.K.) and a certified pathologist (Y.U.) assessed immunolabeling; in every case, a consensus was obtained.
Cell culture and cell lines. Three human CRC cell lines DLD1, SW620, and HT29 were procured from the American Type Culture Collection (Rockville, MD, USA). Cells were cultured in Roswell Park Memorial Institute medium (RPMI 1640; FUJIFILM Wako Pure Chemical Co., Osaka, Japan) supplemented with 1% penicillin-streptomycin solution (FUJIFILM Wako Pure Chemical Co.) and 10% fetal bovine serum (Cosmo Bio Co., Ltd, Tokyo, Japan) under humidified air with 5% CO2 at 37°C. In some experiments, cells were treated with iFSP1, a FSP1 inhibitor (7162; R&D System, Minneapolis, MN, USA); (1S,3R)-RSL3, a GPX4 inhibitor (RSL3; 19288; Cayman Chemical, Ann Arbor, MI, USA); Liprostatin-1, a ferroptosis inhibitor (Lipro-1; 17730; Cayman Chemical); necrostatin-1, a necrosis inhibitor (Necro-1; n9037; Sigma-Aldrich, St Louis, MO, USA); and Z-VAD-FMK caspase inhibitor V1, an apoptosis inhibitor (Z-VAD; S8102; Selleckchem, Houston, TX, USA).
Western blotting analysis. DLD1, SW620, and HT29 cells were seeded at 2.0×105 cells in 6-cm dishes and cultured for 72 h. Cells were harvested in RIPA buffer (Nacalai Tesque, Kyoto, Japan) and centrifuged at 21,432 × g for 10 min at 4°C. The supernatants were collected, and protein concentrations were measured using the Bradford Protein Assay (Takara Bio Inc., Shiga, Japan). Protein samples were separated on 12% Mini-PROTEAN TGX Precast Gels (Bio-Rad, Hercules, CA, USA) and transferred to 0.2-μm polyvinylidene difluoride membranes (Bio-Rad). The membranes were blocked for 1 h at room temperature with 2% ECL Prime Blocking Agent (Cytiva, Tokyo, Japan). The membranes were incubated for 16 h at 4°C with the same primary antibodies used in immunohistochemical staining: FSP1 (1:1,000) and GPX4 (1:1,000). Primary antibody against β-actin (1:2,000) (sc-47778; Santa Cruz Biotechnology, Dallas, TX, USA) was used as a loading control. After washing the membrane, peroxidase-conjugated anti-mouse and anti-rabbit secondary antibodies (Cytiva) were reacted for 1 h at room temperature. The protein signals were detected using ECL Prime Western blotting detection reagent (Cytiva) and quantified using Image Quant 500 (Cytiva). ImageJ software was used to quantify the intensity of the protein bands.
Cell proliferation assay. DLD1, SW620, and HT29 cells were treated with FSP1 or GPX4 inhibitors alone or in combination. The cells were seeded in 96-well plates at 5.0×105 cells/well and incubated overnight. After 24 h, the cells were incubated with iFSP1 (6 μM), RSL3 (1 μM), iFSP1 with RSL3, or DMSO (0.1%) and incubated for two days. After 48 h, Cell Counting Kit-8 solution (CCK-8; DOJINDO, Kumamoto, Japan) was added to each well in accordance with the manufacturer’s protocol, and the cells were incubated for 1 h. Cell viability was evaluated by measuring absorbance using an Infinite F50 microplate reader (TECAN, Kawasaki, Japan) set at 450 nm/620 nm.
Cell death inhibition assay. DLD1, SW620, and HT29 cells were treated with inhibitors for ferroptosis regulators and various cell death pathways. The cells were seeded in 96-well plates at 5.0×105 cells/well and incubated overnight. After 24 h, the cells were incubated with iFSP1 (6 μM) and RSL3 (1 μM) or DMSO (0.1%). Lipro-1 (1 μM), Necro-1 (1 μM), and Z-VAD (10 μM) were added to cells. After 48 h of incubation, CCK-8 solution was added to each well. The cells were incubated for 1 h. An Infinite F50 microplate reader was used to measure absorbance.
Statistical analysis. Receiver operating characteristic curve analysis was used to determine the Youden index. The percentage of GPX4 and FSP1 positivity, determined using the Youden index, was used as the cutoff value. The Kaplan–Meier method was used to assess overall survival and recurrence-free survival, and differences were compared using the log-rank test. Univariate and multivariate analyses were performed using Cox proportional hazards models. For correlations between FSP1 and GPX4 expression and clinicopathologic features of patients, the chi-square test was used; for categorical variables, the Fisher exact test was performed. The means were compared using Student’s t-test. Excel 2021 (Microsoft) and SPSS version 25.0 (IBM, Armonk, NY, USA) were used for all statistical analyses A p-value <0.05 was regarded as statistically significant.
Results
FSP1 and GPX4 expression in CRC patients is negatively correlated with prognosis. Immunohistochemical analysis was performed on resected specimens from patients with CRC to assess the expression of FSP1 and GPX4. We categorized the specimens as either positive or negative on the basis of the proportion of positively stained tumor cells (Figure 1). Among the 206 patients, 150 patients (72.8%) were FSP1-positive, and 102 patients (49.5%) were GPX4-positive.
Immunohistochemical staining for FSP1 and GPX4 in tissue specimens of colorectal cancer patients. Representative samples from positive and negative groups are shown. Both FSP1 and GPX4 proteins are localized in the cytoplasmic region. Magnification: ×100, scale bar: 100 μm. A) FSP1-positive sample. B) FSP1-negative sample. C) GPX4-positive sample. D) GPX4-negative sample.
We next examined the relationship between FSP1 and GPX4 expression levels and clinicopathological factors. There were no correlations between clinicopathological factors and FSP1 or GPX4 expression (Table I). We then investigated the relationship between FSP1 and GPX4 expression and the prognosis of CRC patients. Patients in the FSP1-positive group had a significantly worse prognosis in both OS (p<0.001) and recurrence-free survival (RFS; p<0.001) compared with the FSP1-negative group (Figure 2A and B). Compared with patients in the GPX4-negative group, patients in the GPX4-positive group had significantly worse OS (p<0.001) and RFS (p<0.001) (Figure 2C and D).
Association between FSP1 and GPX4 expression and clinicopathologic factors in colorectal cancer patients.
Kaplan–Meier survival curves for overall survival and recurrence-free survival in colorectal cancer patients classified by FSP1 and GPX4 expression. A, B) Survival determined by FSP1 expression. C, D) Survival determined by GPX4 expression. E, F) Survival determined by expression of FSP1 and GPX4. *p<0.05; **p<0.01; ***p<0.001.
We then examined the prognoses of patients by grouping them into three categories by the expression of FSP1 and GPX4: 85 patients (41.3%) were positive for both FSP1 and GPX4, 39 patients (18.9%) were negative for both FSP1 and GPX4, and 82 patients (39.8%) were positive for either FSP1 or GPX4. The group positive for both GPX4 and FSP1 had a significantly lower RFS (p<0.001) and OS (p<0.001) than the other two groups (Figure 2E and F). The group that was negative for both FSP1 and GPX4 had the longest OS and RFS compared with the other groups.
In multivariate analysis, age ≥70 years, CEA ≥5 ng/m, and both FSP1- and GPX4-positivity (p<0.001) were found to be independent prognostic factors for OS (Table II). In the multivariate analysis of RFS, age ≥70 years, lymph vessel invasion, and both FSP1-and GPX4-psitivity (p<0.001) were independent prognostic factors (Table III). These results indicate that high expression of FSP1 or GPX4 is a substantially poor prognostic factor in CRC patients, while the combined high expression of both represents an even worse prognostic factor.
Univariate and multivariate analyses of clinicopathologic factors related to overall survival.
Univariate and multivariate analyses of clinicopathologic factors related to recurrence-free survival.
FSP1 and GPX4 inhibition causes growth inhibition in CRC cell lines. We next examined the expression of FSP1 and GPX4 in three human CRC cell lines (DLD1, SW620, and HT29) by western blotting. FSP1 expression was low in DLD1 and SW620 cells and almost undetectable in HT29 cells; GPX4 expression was moderate in DLD1 and SW620 cells and high in HT29 cells (Figure 3).
Western blotting analysis of FSP1 and GPX4 expression in colorectal cancer cell lines.
To evaluate the extent of ferroptosis caused by FSP1 and GPX4 inhibition in CRC cells, we used iFSP1, a FSP1 inhibitor, RSL3, a GPX4 inhibitor, and Lipro-1, a ferroptosis inhibitor, and the survival of the cells was evaluated. iFSP1 reduced cell viability in DLD1 and SW620 cells, both of which express FSP1 (Figure 4A and B). In HT29 cells, where FSP1 was nearly undetectable, there was no significant reduction in cell number by iFSP1 (Figure 4C). All three cell lines, which express GPX4, exhibited significant cell death in response to RSL3 (p<0.001, Figure 4A-C). HT29 cells, which showed high expression of GPX4, showed more cell death after RSL3 treatment than the other two cell lines (Figure 4C). Moreover, co-treatment of iFSP1 and RSL3 resulted in greater cell death across all cell lines compared to single treatments (Figure 4A-C).
Effect of FSP1 and RSL3 inhibition in colorectal cancer cell lines. iFSP1 and RSL3 were added alone or in combination, and cell viability was evaluated by using the CCK8 assay. The effect of Lipro-1 addition on cell death was examined. A) DLD1. B) SW620. C) HT29. iFSP1: FSP1 inhibitor; RSL3: GPX4 inhibitor; Lipro-1: liprotatin-1, ferroptosis inhibitor. *p<0.05; **p<0.01; ***p<0.001; NS: Not significant.
Cell death by FSP1 and GPX4 inhibitors is induced by ferroptosis. To determine whether the death induced by the inhibition of FSP1 and GPX4 was caused by ferroptosis, we used inhibitors against other forms of cell death. The addition of Lipro-1, the ferroptosis inhibitor, blocked the growth suppression of iFSP1 and RSL3 (Figure 4A-C). These results suggest that iFSP1- and RSL3-induced cell death was from ferroptosis.
However, treatment with Necro-1, a necrosis inhibitor, or Z-VAD, an apoptosis inhibitor, did not prevent cell death induced by iFSP1 and RSL3 (Figure 5A-C). These results indicate that ferroptosis rather than necrosis or apoptosis is the main cause of cell death induced by iFSP1 and RSL3.
iFSP1 and RSL3 cell death in colon cancer cell lines evaluated using WST assay. Lipro-1, Necro-1, and Z-VAD were added to the cell lines treated as indicated. A) DLD1. B) SW620. C) HT29. iFSP1: FSP1 inhibitor; RSL3: GPX4 inhibitor; Lipro-1: Liprotatin-1, ferroptosis inhibitor; Necro-1: Necrostatin-1, necrosis inhibitor; Z-VAD: Z-VAD-FMK Caspase Inhibitor VI, apoptosis inhibitor. *p<0.05; **p<0.01; ***p<0.001; NS: Not significant.
Discussion
This study investigated the levels of FSP1 and GPX4 expression, which are ferroptosis regulators, in CRC patients. Our results showed that high expression of FSP1 or GPX4 was a negative prognostic factor in CRC patients. Furthermore, patients with both FSP1 and GPX4 positivity showed the worst prognosis. To the best of our knowledge, this is the first study demonstrating the association of the combination of FSP1 and GPX4 with prognosis in patients with CRC. We further used inhibitors against FSP1 and GPX4 to induce cell death in CRC cell lines. The degree of cell death by ferroptosis correlated with the presence of FSP1 and GPX4 expression. Moreover, combined inhibition of FSP1 and GPX4 induced more cell death than inhibition of FSP1 or GPX4 alone. These results show that higher levels of both FSP1 and GPX4 expression are related to poorer prognosis in CRC patients; furthermore, regulation of both of these factors may induce more ferroptosis in cancer cells.
Apoptosis was the first recognized type of programmed cell death, followed by autophagy and necrosis. Elucidation of the mechanisms underlying the regulation of cell death, which is dysregulated in tumor cells, has been a research focus in identifying novel cancer treatments (18, 19). In 2012, researchers discovered ferroptosis, a very specific non-programmed form of cell death (5). Ferroptosis is an iron-dependent mechanism that leads to lipid hydroperoxide accumulation, in which the cell membrane breaks down and cell death is induced. GPX4 is a ferroptosis regulator that reduces lipid peroxidation and protects cells from oxidative damage by using glutathione as a reducing agent to reduce and eliminate toxic phospholipid hydroperoxides (20-22). GPX4 is a promising therapeutic target in a variety of carcinomas (23-27). Notably, there are some cancer cell lines in which inhibition of GPX4 does not induce ferroptosis, which suggested another regulatory mechanism for ferroptosis and led to the discovery of FSP1 (13, 14). FSP1, previously called AIFM2, reduces coenzyme Q in an NADH-dependent manner and inhibits lipid peroxidation; it was thus established as a ferroptosis regulator that is not dependent on GPX4 (13, 14). FSP1 is also being investigated for its therapeutic potential in multiple cancer types (28, 29).
Various studies have been conducted on pharmacological FSP1 and GPX4 inhibitors to explore their potential as therapeutic agents, and research has shown that the simultaneous suppression of FSP1 and GPX4 significantly and synergistically induces ferroptosis (12-14, 30, 31). Therefore, concurrent control of FSP1 and GPX4 is a promising therapeutic strategy for cancer drug treatment.
Gotorbe et al. stated that GPX4 inhibition in cell lines where FSP1 was knocked out significantly reduced cell viability in colon cancer cells in a synergistic manner (31). In our study, the combination treatment with iFSP1, the FSP1 inhibitor, and RSL3, the GPX4 inhibitor, resulted in more potent cell death in the three CRC cell lines, DLD1, SW620, and HT29 compared to each inhibitor alone (Figure 4A-C). In the HT29 cell line, while iFSP1 alone did not cause significant cell death, the combination with RSL3 resulted in more cell death compared to RSL3 alone (p<0.01, Figure 4C). The inhibition of cell viability caused by the FSP1 and GPX4 inhibitors correlated with the levels of FSP1 and GPX4 expression, while the combination therapy resulted in further reduction in cell viability, independent of FSP1 and GPX4 expression levels (Figure 4A-C). These findings imply that combined inhibition of FSP1 and GPX4 may be a promising pharmacological method for treating CRC by inducing ferroptosis.
Study limitations. First, the sample size for IHC was small. Second, while we investigated FSP1 and GPX4 inhibition using in vitro experiments, we did not evaluate the inhibitors in vivo. Additional validation through animal studies will be necessary.
Conclusion
High expression of FSP1 or GPX4 is a negative prognostic factor in CRC, and high expression of both FSP1 and GPX4 is an even stronger negative prognostic factor. In colon cancer cell lines, combined inhibition of FSP1 and GPX4 induces higher levels of ferroptosis compared to single-agent inhibition. Simultaneous inhibition of FSP1 and GPX4 is a promising new therapeutic strategy for CRC.
Acknowledgements
The Authors thank Gabrielle White Wolf, PhD, from Edanz (https://jp.edanz.com/ac), for editing a draft of this manuscript.
Footnotes
Authors’ Contributions
CY and YK designed the study, analyzed the data, and drafted the article. IR, YT, KK, and MY contributed to the acquisition of clinical data. YF provided final approval of the manuscript. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors have no conflicts of interest to disclose regarding this study.
Funding
None.
- Received September 11, 2024.
- Revision received September 20, 2024.
- Accepted September 23, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).











