Colorectal cancer (CRC), is one of the most frequently occurring digestive malignancies worldwide and accounts for approximately 9.4% and 10.1% of all cancers in man and woman (Boyle and Langman, 2000; Benson et al., 2018). Environmental and hereditary factors, such as diet, dysbacteriosis, and genetic variations, have been found to be the most significant risk factors for CRC morbidity (Boyle and Langman, 2000; Grady, 2003; Garrett, 2019; Vernia et al., 2021). The main treatment strategies for CRC include endoscopic therapy for early stage CRC and radical surgery with adjuvant systemic therapy for advanced CRC (Benson et al., 2014; Benson et al., 2018). Preventing of CRC etiology and optimizing systemic therapeutic strategies against the disease are both beneficial strategies for improving overall CRC patient survival.

Laboratory-based and clinical studies have found high dietary fat intake, as well as bile acid metabolism by gut microbiota, account for the increasing risk of CRC incidence (Boyle and Langman, 2000; Garrett, 2019; Ocvirk and O'Keefe, 2021). Previous study on correlations between CRC and cholesterol metabolism provided evidence of confirming serum cholesterol as carcinogenic factor on CRC progression, and an abnormal accumulation of formed secondary bile acids also presented promoting effects on CRC (Jacobs et al., 2012). Thus, drugs that target lipid metabolism may have potential benefits in CRC adjuvant therapy. Stain, an HMG-CoA reductase inhibitor which exerts significant effects in lowering cholesterin (TC) and low-density lipoprotein (LDL) levels, was found to exhibit protective effects against CRC by reducing the dilatation activity and enhancing the chemotherapeutic sensitivity of CRC cells (Bardou et al., 2010). In addition, studies have demonstrated the anti-tumor effects of fibrates, which function mainly in reducing LDL and triglyceride (TG) levels, against CRC and other digestive neoplasms (Chen et al., 2020; Kong et al., 2021; Li et al., 2021). Cholesterol accumulation, which exceeded the ability of liver conversion and intestinal absorption, was reported as high risk factor of CRC (Waluga et al., 2018; Gu et al., 2022). Squalene epoxidase (SQLE), a cholesterol synthesis promoting factor, has been proved to suppress the apoptosis of CRC cells and lead to poor disease outcome (Li et al., 2022). And the inhibition impacts of cholesterol on CRC cell apoptosis has also been reported via miR-33a-PIM3 pathway (Wang et al., 2019). As ezetimibe blocks lipid absorption in the small intesines, which is the main organ for bile acid re-absorption, it may also have potential anti-CRC effects, even though correlational approaches to prove this are lacking. Furthermore, the high toxicity of adjuvant chemotherapy may partially counteract its benefits in CRC patients, especially in older patients under combined chemotherapeutic regimens (Moth et al., 2016; Benson et al., 2018). Studies on novel candidates for adjuvant therapy against CRC, which focus on reversing damages counteracted benefits, are still being carried out. In this study, we preliminarily investigated the effects and possible mechanisms of ezetimibe against CRC.

CRC, which is the fourth most common type of cancer, causes approximately 394,000 deaths worldwide annually (Boyle and Langman, 2000). In the last decades, its incidence in younger individuals has significantly increased (Stoffel and Murphy, 2020). Epidemiological analyses on life style habits have shown that a westernized dietary pattern, characterized by high fat intake and low fiber ingestion, could be a contributing factor to CRC morbidity (Boyle and Langman, 2000; Stoffel and Murphy, 2020; Vernia et al., 2021). With the inadequacies of current therapeutic approaches, overall survival is still poor in advanced stage CRC patients (Boyle and Langman, 2000). Previous studies pointed out the antitumor potential of ezetimibe against several cancers involving urinary system, digestive system and genital system (Gu et al., 2022), with clinical trials not provided evidence for adverse impacts of ezetimibe on cancer risk (Peto et al., 2008; Kobbero Lauridsen et al., 2017). Based on the approach of reducing lipid absorption in the small intestines, we evaluated the potential effects of ezetimibe against CRC for the development of a novel adjuvant therapeutic strategy.

Cellular experiments showed that ezetimibe decrease proliferation, inhibit invasion, and stimulate apoptosis in CRC cell lines (Figures 1, 2). IC₅₀ was used to evaluate CRC cell viability in this study (Sebaugh, 2011). HCT116 cell viability was found to decrease by half when ezetimibe concentrations reached 73.71 μM and 56.85 μM at 24 and 48 h, respectively (Figure 1B). A similar trend was observed in the Caco2 cell line, with IC₅₀s of 79.27 μM and 44.33 μM at 24 and 48 h, respectively (Figure 1C). The clone-formation capacity of CRC cells, which represents the proliferative viability of the cells (Dobson et al., 2020), also decreased following treatment with ezetimibe (Figure 1D). CRC cell invasiveness was found to significantly decrease following ezetimibe treatment in a dose- and time-dependent, as determined through the wound healing assay (Figures 1E, F). In addition, CRC cell apoptosis increased following treatment with effective ezetimibe concentrations as determined through the fluorescence microscopic and flow cytometric assays (Figures 2D, E); these findings indicated the toxic effects of ezetimibe against CRC.

Energy metabolism in cancer cells is more complex than simple glycolysis, as initially reported, and involves the use of a wide variety of substrates (Ferro et al., 2020). To satisfy the energy demands of cancer cells, mitochondria have to be trafficked from host immune cells, highlighting the important role played by mitochondria as powerpacks in cancer cells (Ferro et al., 2020; Saha et al., 2022). The ATP Binding Cassette transporter (ABCC) subfamily of ABCA1 and ABCC3, as potential targets for ezetimibe (Figure 3), reported of intervening the drug resistance in cancers (Wang and Smith, 2014; Grube et al., 2018; Belisario et al., 2020; Ramírez-Cosmes et al., 2021; Gao et al., 2022), thus providing evidence for the possibility of using ezetimibe in CRC adjuvant therapy. We found ezetimibe-induced CRC cell death to be associated with mitochondrial dysfunction. The JC-1 fluorochrome, a cyanine dye widely used to distinguish energized mitochondria from de-energized ones (Perelman et al., 2012), was used to measure CRC △Ψm in this study. The microscopic and flow cytometric assays consistently showed an increase in the proportion of JC-1 monomers with respect to JC-1 aggregates in CRC cells with increase in ezetimibe concentration (Figures 2C–F), indicating a decrease in △Ψm in a dose-dependent manner. Excessive accumulation of intracellular ROS can induce DNA damage and finally, cell death (Filomeni et al., 2015). Following treatment with cell lethal concentrations of ezetimibe, there was a significant accumulation of intracellular ROS in both HCT116 and Caco2 cells (Figures 2B, G). The findings of these experiments revealed that mitochondrial damage occurred during ezetimibe-induced CRC cell death, indicating that an energy metabolism-related mechanism underlies the pharmacological action of ezetimibe against CRC.

The Comparative Toxicogenomics Database (Davis et al., 2019) and Drugbank online (Law et al., 2014) were used to determine the possible signaling targets of ezetimibe in CRC cells, and 29 ezetimibe-targeted genes were identified, mainly clustered into groups of glycometabolism and lipometabolism (Figures 3A, B; Table 2). The GEPIA-based analysis (Tang et al., 2017) showed that eight out of 29 genes i.e., ABCA1, ABCC3, HMGCR, LDLR, PCSK9, PGR, PLA2G7, and SCARB1, can significantly discriminate between CRC tumors and normal tissues (Figure 3C; Table 2). Among the eight targeted genes, HMGCR, LDLR, PCSK9, PLA2G7, and SCARB1 were overexpressed in CRC tumor tissues as compared to normal tissues (Figures 3F–K), while ABCA1, ABCC3, and PGR exhibited low expression levels in CRC tumor tissues (Figures 3D–I). The expression levels of PGR vary between different CRC stages and are significantly correlated with CRC patient overall survival (Figure 3I), as its expression levels have been shown to be negatively correlated with CRC prognosis (Zhang et al., 2021). Similarly, in previous studies, ABCC3 differential expression in different CRC stages (Figure 3E) was found not to be significantly correlated with CRC prognosis (Kim et al., 2020). UGT2B15 and UGT2B7, which are non-CRC tumor-associated genes that were identified in the normal expression group, were found to be correlated with overall survival in CRC patients (Table 2). In this study, using the GEPIA database, we analyzed the correlation between the 29 targeted genes and GLS, HIF1-α, mTOR, TNF, and p53, which are recognized as key molecules involved in cellular metabolism (Table 2). Of the eight CRC-related genes, TNF was found to be significantly correlated with two of them i.e., ABCA1 and PLA2G7 (Table 2). In addition, of these eight genes, GLS, a key enzyme involved in glutamine metabolism, which plays an important role in cancer metabolism (Mates et al., 2019; Masisi et al., 2020), was associated with ABCA1, PLA2G7, and SCARB1 (Table 2). p53, an important antitumor transcription factor in CRC cells (Liebl and Hofmann, 2021), was found to be associated with four of the CRC-related expression genes i.e., ABCA1, PCSK9, PGR, and SCARB1 (Table 2). A total of 6 and seven CRC-related expression genes were found to be significantly correlated with mTOR and HIF1-α, respectively, in the mTOR signaling pathway (Table 2), indicating that mTOR signaling is an important part of the mechanisms underlying the effects of ezetimibe against CRC. Low density lipoprotein receptor (LDLR), which is significantly correlated to mTOR according to the bioinformatic analysis (Table 2), mediates the hepatic endocytosis to consume Niemann-Pick C1-Like 1 (NPC1L1) mediated lipid absorption (Gu et al., 2022; Liao et al., 2022; Zhang et al., 2022). AS ezetimibe exerts its suppression on cholesterol absorption with blockage to NPC1L1, it may be potential regulator of mTOR signal for the lipid homeostasis in CRC progression. Proprotein convertase subtilisin/kexin type 9 (PCSK9) and HMG-CoA reductase (HMGCR), both significantly high expressed in CRC tumor as ezetimibe potential targets (Figure 3), also confirmed of tight correlation to mTOR (Table 2). Studies have also reported that the PCSK9 inhibitor and HMGCR suppressor statins possess anti-CRC effects (Navarese et al., 2015; Gu et al., 2022; Shailes et al., 2022), which further stand for the potential pivotal role of mTOR signal mediation by ezetimibe in CRC. The ezetimibe targeted phospholipase A2 group VII (PLA2G7) has also been found to have tight correlation with cancer cachexia (Morigny et al., 2021) and put forward as therapeutic target for prostate cancer (Vainio et al., 2011), supporting the possibility of ezetimibe used in clinical therapy in advanced cancer stages.

Western blotting and the quantitative real-time polymerase chain reaction (qRT-PCR) assay carried out in our study revealed that ezetimibe attenuated CRC invasiveness, and induced CRC apoptosis and autophagy, which were accompanied by a decrease in mTOR phosphorylation at the molecular level. The mTOR^pSer2448 protein, which is recognized as a key indicator of mTOR signaling activation, was found to be negatively correlated with disease prognosis in advanced stage CRC patients (Vainio et al., 2011; Muller et al., 2013; Navarese et al., 2015; Mates et al., 2019; Kim et al., 2020; Masisi et al., 2020; Zhang et al., 2020; Liebl and Hofmann, 2021; Morigny et al., 2021; Liao et al., 2022; Shailes et al., 2022; Zhang et al., 2022), its expression levels significantly decreased in both HCT116 and Caco2 cells following effective ezetimibe treatment (Figures 4A, B); however, mTOR transcription and expression levels were not significantly different between the ezetimibe-treated and control groups (Figures 4A–C). MMPs play crucial roles in tumorigenesis as they promote angiogenesis, invasiveness, and immune evasion in a wide variety of cancers (Vainio et al., 2011; Muller et al., 2013; Navarese et al., 2015; Gobin et al., 2019; Masisi et al., 2020; Zhang et al., 2020; Liebl and Hofmann, 2021; Lodge et al., 2021; Morigny et al., 2021; Liao et al., 2022; Shailes et al., 2022; Zhang et al., 2022). Ezetimibe induced MMP-2 and MMP-9 downregulation in HCT116 and Caco2 cells (Figures 4A, B); this finding, together with the decreased wounding healing capacity found in the scratch assay, demonstrated the inhibitory effects of ezetimibe against CRC invasiveness. Ezetimibe induced CRC cell apoptosis and autophagy at effective concentrations. In this study, we observed an increase and a decrease in both BAX and Bcl-2 mRNA and protein levels, respectively, in the CRC cell lines; this finding, together with the caspase cascade activation observed (Figures 4A–C), provided evidence for our point of the CRC cell apoptosis promotion effects of ezetimibe. The expression levels of Beclin-1, the pre-autophagosomal structure positive correlated protein (Tran et al., 2021), significantly increased under treatment with increasing ezetimibe concentrations (Figures 4A–C). In addition, the expression levels of the key autolysosome membrane formation protein, LC3 II, were also found to significantly increase in ezetimibe-treated HCT116 and Caco2 cells in a dose-dependent manner (Figures 4A–C).

The apoptosis and mitochondrial dysfunction phenotype observed in ezetimibe-treated CRC cells was partially reversed by the mTOR activator, MHY1485 (Figure 5), indicating that ezetimibe induces cell death and mitochondrial dysfunction in CRC cells through the mTOR signaling pathway. The levels of mTOR^pSer2448 proteins significantly decreased in the ezetimibe-treated group as compared to the negative control group; these levels were partially upregulated following the addition of MHY1485 to the CRC cells. The proportion of apoptotic HCT116 and Caco2 cells decreased in the ezetimibe + MHY1485-treated group as compared to the ezetimibe-treated group, and this was accompanied by homogeneous △Ψm recovery and a reduction in intracellular ROS levels.

The anti-CRC effects of ezetimibe, which involve the promotion of CRC cell death without damage to vital organs involved in drug metabolism, were confirmed using a xenograft tumor mouse model (Figure 6). Ezetimibe significantly decreased the volumes of xenograft tumors in nude model mice subjected to treatment as compared to control mice; however, there was no significant difference in mouse weight between the two groups (Figures 6A–C). No significant morphological changes were observed in the livers and kidneys of mice in the control and ezetimibe-treated groups as determined through the H&E staining analysis (Figure 6D); however, visible mitochondrial morphological damage, with cytophagosome development, was observed by TEM in the xenograft tumor tissues of ezetimibe-treated mice (Figure 6E). HCT116 xenograft mouse tumor immunohistochemical staining showed an increase in the expression levels of apoptosis/autophagy-related proteins and a decrease in the expression levels of invasiveness-related proteins, considering the downregulation of mTOR^pSer2448 expression in the ezetimibe-treated group (Figures 6F, G). In vivo findings support our hypothesis that ezetimibe induces CRC cell death via mTOR signaling-dependent mitochondrial dysfunction.

Despite the demonstrated potential anti-CRC effects of ezetimibe, our study still has some limitations. First, the in-depth mechanisms of ezetimibe-induced mitochondrial dysfunction, as well as correlations to the tumor microenvironment in CRC, still need to be investigated. More evidence needs to be gathered on the clinical application of ezetimibe in the treatment of CRC before it be taken from the bench to the bedside for CRC treatment.

We demonstrated in vitro and in vivo that ezetimibe elicits its anti-CRC effects by inhibiting cell proliferation and invasion, and promoting apoptosis and autophagy. In addition, ezetimibe-induced apoptosis and autophagy were found to be correlated with mTOR signaling-dependent mitochondrial dysfunction (Figure 7). The findings of this study highlight the potential value of ezetimibe for use in CRC adjuvant therapy and for optimized selections in individual therapeutic schedules in future.