ACSM5 Regulates Ferroptosis in Hepatocellular Carcinoma by Up‐Regulating POR and Modulating Lipid Metabolism

ABSTRACT

The medium‐chain fatty acyl‐CoA synthetase‐5 (ACSM5) plays a crucial role in the development of some cancers. However, its impact on liver cancer is still not clear. In this study, we found that the proliferation ability of LM3 and HepG2 cells was significantly inhibited after ACSM5 was overexpressed, and this change was blocked by the ferroptosis inhibitor deferoxamine. ACSM5 increased the levels of malondialdehyde (MDA) and lipid reactive oxygen species (ROS), reduced the level of glutathione (GSH), and thus triggered ferroptosis. Furthermore, ACSM5 promoted the upregulation of cytochrome P450 oxidoreductase (POR). Knocking down POR blocked the promoting effect of ACSM5 on ferroptosis in HCC. Moreover, ACSM5 promoted the generation of arachidonic acid and thus increased the sensitivity to ferroptosis. In summary, our findings indicate that ACSM5 induces ferroptosis in hepatocellular carcinoma (HCC) by upregulating POR. The metabolic transformation of linoleic acid to arachidonic acid was also promoted by ACSM5; therefore, sensitivity to ferroptosis was increased.

ACSM5 induced ferroptosis in HCC by up‐regulating cytochrome P450 oxidoreductase (POR). The metabolic transformation of linoleic acid to arachidonic acid was also promoted by ACSM5; thereby, the sensitivity to ferroptosis was elevated.

Abbreviations

ACSL4

long‐chain fatty acid‐CoA ligase 4

ACSM5

medium‐chain fatty acyl‐CoA synthetase‐5

ARA

arachidonic acid

COA

coenzyme A

GSH

glutathione

LPCAT3

lysophosphatidylcholine acyltransferase 3

LPOs

lipid peroxides

MDA

malondialdehyde

POR

cytochrome P450 oxidoreductase

PUFAs

polyunsaturated fatty acids

ROS

reactive oxygen species

RSD

relative standard deviation

VIP

variable importance in the projection

1. Introduction

Hepatocellular carcinoma, a type of tumor with a high degree of malignancy, has a relatively high incidence and mortality rate worldwide and constitutes a significant threat to public health [1]. The pathogenesis of hepatocellular carcinoma is complex [2]. Various liver‐infecting viruses, parasites, and toxins can damage the liver. Moreover, they activate intracellular pathways and kinases in hepatocytes, resulting in liver injury and the occurrence of liver cancer. Furthermore, the occurrence and development of hepatocellular carcinoma involve the regulation of multiple genes, multiple stages, and multiple steps [3]. Thus, clarifying the molecular mechanism underlying the occurrence of liver cancer is highly important for identifying effective molecular therapeutic targets.

Ferroptosis is a unique form of cell death discovered in recent years that is driven by iron‐dependent lipid peroxidation. Its main feature is the lethal accumulation of intracellular lipid peroxides, which damage cell membranes, proteins, and DNA, thereby leading to cell death [4]. Ferroptosis in cells involves three biological processes: lipid metabolism, the participation of ROS, and iron regulation. Any biological activity within cells that is involved in regulating the aforementioned processes can directly or indirectly modulate ferroptosis [5, 6]. Ferroptosis plays a crucial role in the progression and treatment of tumors. Various metabolic disorders, including lipid metabolism, an elevated demand for iron ions, and changes in the redox environment, typically occur in tumors [7, 8]. Numerous studies have verified that many tumor suppressors and tumor promoters are involved in regulating ferroptosis in cancer cells. For example, the P53 gene can promote ferroptosis in cells by downregulating the SCL7A11 subunit and thereby reducing the uptake of cysteine (Cys) by System xc‐ [9, 10]. The liver is one of the main sites for lipid synthesis and catabolism. Therefore, abnormal lipid metabolism, iron metabolism, and the accumulation of reactive oxygen species in liver cancer cells suggest that ferroptosis is closely related to the occurrence and development of liver cancer [11, 12, 13]. POR, which serves as the sole electron donor of the P450 enzyme family, possesses oxidoreductase activity [14]. Recent studies have revealed that polyunsaturated fatty acids (PUFAs) are the primary substrates that induce ferroptosis. ACSL4 catalyzes the combination of free arachidonic acid (ARA) or adrenic acid (Ada) with coenzyme A to form AA‐CoA or Ada‐CoA derivatives, respectively. LPCAT3 subsequently promotes their esterification reaction to generate membrane phosphatidylethanolamine, resulting in AA‐PE or Ada‐PE. Thereafter, POR and ALOX induce the peroxidation of polyunsaturated fatty acid‐containing phosphatidylethanolamine (PUFA‐PE) to generate peroxidized lipids, which serve as fuel for ferroptosis. POR is abundantly expressed in both normal liver tissues and tumor tissues. Additionally, the liver is a place where numerous redox reactions take place. POR plays crucial roles in the occurrence and development of liver cancer and ferroptosis [15].

ACSM5 is a member of the medium‐chain acyl‐CoA synthetase family (ACSMs). The ACSMs family is a group of enzymes that can primarily activate medium‐chain fatty acids into fatty acyl coenzyme A esters. This is the first step in activating fatty acids for fatty acid metabolism [16, 17]. The members of the ACSMs family are closely related to tumors. ACSM1 and ACSM3 in prostate cancer affect tumor occurrence and development by inhibiting ferroptosis in tumor cells [18, 19]. The low expression of ACSM5 is closely related to the occurrence and development of lung adenocarcinoma, papillary thyroid cancer, and prostate cancer as well as a poor prognosis [20, 21, 22]. As indicated in recent studies, the expression of ACSM5 is downregulated in the majority of tumors, and ACSM5 has a certain tumor‐suppressing function. Nevertheless, the specific tumor‐suppressing mechanism involved remains unclear. Similarly, the role of ACSM5 in liver cancer has not been confirmed. This article presents further research on the role of ACSM5 in liver cancer.

2. Materials and Methods

2.1. Cell Culture and Transfection

LO2, HepG2, HCCLM3, Hep3B, Huh‐7, and MHCC97H cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS and 100 U/mL penicillin and streptomycin. Lentivirus and plasmids purchased from General BIOL (Anhui China) were used for transfection according to the product instructions.

2.2. qRT–PCR

TRIzol reagent (Thermo, Ambion) was used to lyse the sample, total RNA was obtained, and a cDNA synthesis kit (Thermo, #K1622) was used to perform reverse transcription according to the product specifications. The primer sequences are presented in Table S1.

2.3. Western Blotting

Samples were added to a 10% gel for electrophoresis. After electrophoresis, the proteins were transferred to a membrane and hybridized with the appropriate antibody. Finally, the immunoreactivities were detected using a chemiluminescent ECL reagent. The antibodies used were as follows: ACSM5 (dilution 1:1500; cat. no. 67334‐1‐Ig, Proteintech), POR (dilution 1:1000; cat. no. ab180597, Abcam), and anti‐beta actin (dilution 1:1000; Abcam, ab8226).

2.4. Co‐IP

Untreated LM3 cells were lysed in cell lysis buffer containing protease and phosphatase inhibitors. The total cell protein extracts were collected. Then, we added a specific antibody (ACSM5, cat. no. 67334‐1‐Ig; Proteintech; POR, cat. no. ab180597; Abcam) that binds to the target protein and forms an immune complex. Protein A/G magnetic beads were added to bind with the specific antibody. The desired proteins were collected by centrifugation after the magnetic beads were eluted. The 55 kDa and 78 kDa protein bands were detected by western blot, and the bands were exposed after incubation with ACSM5, POR antibody, and heavy chain‐free light chain secondary antibody.

2.5. Immunohistochemistry

Immunohistochemistry (IHC) was performed as described previously. The tumor sections were deparaffinized, blocked, incubated with antibodies, and subsequently counterstained. The sections were examined under a light microscope to detect histological alterations.

2.6. Immunofluorescence Colocalization

Untreated LM3 and HepG2 cells were fixed with 4% paraformaldehyde solution and incubated with PBS containing 0.1%–0.3% Triton X‐100, 5%–10% goat serum, and 0.05%–0.1% DAPI for half an hour. After the cells were washed, antibodies labeled with two different fluorescent dyes were mixed in appropriate proportions and added to the samples. After incubation overnight, the unbound fluorescent antibodies were washed away, and the two antigens were observed under a fluorescence microscope.

2.7. Cell Viability and Proliferation

Cell counting kit‐8 (CCK8) assay, 5‐Ethynyl‐2′‐deoxyuridine (EdU) cell proliferation detection experiment, and plate colony formation assay were performed to detect cell viability and proliferation.

2.8. Pan‐Cancer, Kaplan–Meier, KEGG–GO and Pearson Analysis

The data for pan‐cancer analysis was sourced from the TIMER 2.0 database (http://timer.cistrome.org/). RNA sequencing data from patients with tumors and control samples were obtained from the publicly available Cancer Genome Atlas liver cancer (TCGA‐LIHC) and GEO databases. Subsequently, Kaplan–Meier analysis, differential expression analysis, Pearson correlation analysis, and KEGG–GO analysis were performed using the survival, limma, corrplot, and clusterProfiler packages, respectively, for the R Programming Language. In KEGG and GO analysis, patients were divided into ACSM5‐high and ACSM5‐low groups based on the cutoff value, which was obtained through “survival” R package. Patients whose values were below the cutoff value were classified into the ACSM5‐low group, whereas those whose values were above the cutoff value were classified into the ACSM5‐high group. In the TCGA cohort, there were 197 patients in the ACSM5‐high group and 173 patients in the ACSM5‐low group. In the GSE14520 cohort, the ACSM5‐high group included 68 patients, whereas the ACSM5‐low group included 148 patients. GSE84044 was used to analyze the association between the expression of ACSM5 and pathological grade based on the Scheuer system in patients with HBV‐related liver fibrosis.

2.9. Assessment of Fe²⁺, Lipid Peroxide, ROS, MDA and GSH Level

The Ferrous Iron Colorimetric Assay Kit (elabscience, #E‐ BC‐K773‐M) was used to detect the Fe²⁺ levels according to the product instructions. The level of lipid ROS in HCC cells was examined using a Cellular ROS Assay Kit (Abcam, ab 186,029) according to the manufacturer's protocols. A Lipid Peroxidation MDA Assay Kit (Beyotime, #S0131S) was used to detect the MDA levels according to the product instructions. A reduced glutathione (GSH) assay kit (Nanjing Jiancheng Bioengineering Institute, #A006‐2‐1) was also used.

2.10. Transmission Electron Microscopy

Transmission electron microscopy was performed at Nanchang University using a standard procedure.

2.11. Detection of Free Fatty Acid Levels

We detected the contents of 31 free fatty acids using gas chromatography (Agilent 7820, USA) and quadrupole mass spectrometry detection (Agilent 5977, USA) systems.

2.12. Construction of Lentivirus and Xenograft Mouse Models

ACSM5‐overexpressing lentiviruses were constructed from the lentiviral vector pLenti‐CMV‐GFP‐Puro, and the pLenti‐CMV‐GFP‐Puro empty vector was used as a control after the cell lines were transfected for subsequent experiments. The sequences are shown in Table S2. POR was knocked down by shRNA, and its sequence is shown in Table S3. The mouse strain used was a six‐week‐old male nude mouse. LM3 cells overexpressing ACSM5 were injected into the right armpits of nude mice to induce tumor formation in vivo. The injection volume was 0.1 mL, which contained 5 × 10 ^ 6⁶ LM3 cells. Animal procedures were approved by the Experimental Animal Center of Nanchang University and were in line with animal ethics guidelines.

2.13. Statistical Analysis

Statistical analyses were conducted using one‐way ANOVA and the Newman–Keuls multiple comparison test with GraphPad Prism 9 (GraphPad Software, San Diego, CA). The data are presented as the means ± SEMs from at least three independent experiments. To evaluate the contribution or importance of metabolites in lipid metabolism, the value of VIP (variable importance in the projection) was utilized. The higher the VIP value was, the greater the importance of this metabolite in groups. A p value of 0.05 or less was considered significant.

3. Results

3.1. ACSM5 Is Downregulated in Liver Cancer and Indicates a Poor Prognosis

The expression levels of ACSM5 in 33 cancer types in the TCGA cohort demonstrated that ACSM5 was downregulated in 16 cancer types, including liver cancer (Figure 1A). Similar to those in the TCGA database, liver cancer tissues in the GEO cohort presented lower expression of ACSM5. Moreover, low ACSM5 expression was associated with high stage and a lower survival rate in patients with liver cancer (Figure 1B,C). In addition, compared with that in adjacent normal tissues, ACSM5 was also expressed at low levels in HCC tumor tissues, which was consistent with the above results (Figure 1D–F). Interestingly, patients with HBV‐related liver fibrosis with high necroinflammatory activity (G2–4) and high fibrosis stage (S2–4) had lower levels of ACSM5, which suggested that ACSM5 may play an important role in the progression of liver fibrosis and the transformation from fibrosis to liver cancer (Figure S1).

FIGURE 1.

ACSM5 is poorly expressed in HCC and is related to poor prognosis. (A) The expression levels of ACSM5 in 33 cancer types in the TCGA cohort. (B, C) The expression of ACSM5 in HCC tissues and adjacent normal tissues and the overall survival in HCC of patients with HCC were analyzed using data from the TCGA and GEO databases. (D) The expression of ACSM5 in HCC tissues and adjacent normal tissues was determined by western blotting. (E) The expression of ACSM5 in HCC tissues and adjacent normal tissues was determined by RT–qPCR. (F) Representative immunohistochemistry images showing ACSM5 staining in normal and tumor tissues from patients with HCC. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

3.2. Overexpression of ACSM5 Inhibits the Proliferation of HCC Cells

An examination of ACSM5 expression across 5 liver cancer cell lines revealed that ACSM5 expression is lower in liver cancer cells, in which LM3 and HepG2 cells have the lowest expression (Figure 2A). Thus, we selected LM3 and HepG2 cells for further study. ACSM5 was stably overexpressed in LM3 and HepG2 cells, and the efficiency of its overexpression was subsequently tested (Figure 2B). The results of the CCK8, EDU, and colony formation assays revealed that ACSM5 decreased the viability and proliferative ability of LM3 and HepG2 cells (Figure 2C–E). ACSM5 was knocked down in 97H cells, and cell plate cloning and cell counting kit‐8 (CCK8) assays were then performed. The results showed that the knockdown of ACSM5 increased the proliferation of 97H cells (Figure S2).

FIGURE 2.

ACSM5 inhibits the proliferation of HCC cells. (A) ACSM5 expression in HCC cells was detected by RT–qPCR and western blotting. (B) LM3 and HepG2 cells were treated with a lentiviral vector, and the expression of ACSM5 was detected by RT–qPCR and western blotting. (C–E) Cell proliferation was tested by CCK8, EDU, and colony formation assays. ***p < 0.001.

3.3. Overexpression of ACSM5 Promotes Ferroptosis in HCC Cells

The viability of ACSM5‐overexpressing HCC cell lines treated with inhibitors of ferroptosis, autophagy, apoptosis, and necroptosis was determined by CCK8 assays. As shown in Figure 3A, there was no significant difference in cell viability between the control group and the groups treated with the apoptosis inhibitor ZVAD‐FMK, the autophagy inhibitor 3‐methyladenine, or the necroptosis inhibitor necrosulfonamide. Viability was significantly increased in cells treated with the ferroptosis inhibitor ferrostatin‐1 when ACSM5 was overexpressed. Electron microscopy images revealed some changes in mitochondrial morphology, including the reduction or disappearance of mitochondrial cristae and the rupture of the mitochondrial membrane (Figure 3B). To determine the typical changes in ferroptosis indices induced by ferroptosis inducers and explore whether ACSM5 has synergistic effects with ferroptosis inducers, we used the ferroptosis inducer erastin to induce ferroptosis. As shown in Figure 3C, the overexpression of ACSM5 in the LM3 and HepG2 cell lines led to increases in the MDA, lipid, ROS, and Fe²⁺ levels. Conversely, the level of GSH was reduced when ACSM5 was overexpressed. The effect became even more remarkable after treatment with erastin. These results demonstrated that ACSM5 is a positive regulator of ferroptosis in HCC cells.

FIGURE 3.

ACSM5 promotes ferroptosis in HCC cells. (A) LM3 and HepG2 cells were treated with DMSO, ferrostatin‐1 (2.0 μM), ZVAD‐FMK (10.0 μM), 3‐methyladenine (5 mM) or necrosulfonamide (0.5 μM) for 48 h, and cell death was evaluated in a CCK8 assay. (B) Electron microscopy image of the cells overexpressing ACSM5. The orange arrow indicates the reduction or disappearance of mitochondrial cristae, and the blue arrow indicates mitochondrial membrane rupture. (C) The levels of GSH, MDA, lipid ROS, and Fe²⁺ in HCC cells were compared between the DMSO and erastin (10.0 μM) groups. **p < 0.01, ***p < 0.001, ****p < 0.0001.

3.4. ACSM5 Is Associated With the POR

Based on the analysis of genes activated or suppressed by ACSM5 obtained from the TCGA and GEO databases, GO analysis revealed that ACSM5 was involved in the lipid metabolism pathway (Figure 4A). In addition, ACSM5 and its associated genes were enriched for certain functions, especially in the cytochrome P450 metabolic pathways (Figure 4B). POR, a cytochrome P450 oxidoreductase, is a crucial constituent of ferroptosis. The Pearson score was used to estimate the connection between ACSM5 and POR. As shown in Figure 4C, the expression of POR was positively correlated with ACSM5. Immunofluorescence staining revealed that ACSM5 and POR had highly overlapping regions, which confirmed the colocalization of ACSM5 and POR in the cytoplasm (Figure 4D). Coimmunoprecipitation (co‐IP) revealed that the ACSM5 protein interacted with the POR protein (Figure 4E). When ACSM5 was overexpressed, both the mRNA and protein expression levels of POR were upregulated (Figure 4F). A Cytochrome P450 Reductase (CPR) Activity Assay Kit was also used to detect POR enzyme activity. The results revealed that POR enzyme activity increased in LM3 and HepG2 cells when ACSM5 was overexpressed (Figure S3).

FIGURE 4.

ACSM5 is associated with the POR. (A) GO analysis of genes activated or suppressed by ACSM5 in the TCGA and GEO databases. (B) KEGG functional analysis of the differential ACSM5 expression groups in the GEO and TCGA databases. (C) Pearson scores were used to estimate the connection between ACSM5 and POR in the GEO and TCGA databases. (D) Colocalization of ACSM5 and POR in the cytoplasm. (E) Coimmunoprecipitation (co‐IP) results for untreated LM3 cells. (F) POR expression in ACSM5‐overexpressing HCC cell lines was measured by RT–qPCR and western blotting. ***p < 0.001.

3.5. ACSM5 Promotes Ferroptosis in HCC Cells by Regulating POR

POR was stably knocked down by shRNA, and its efficiency was tested (Figure 5A). We selected one of the three shRNAs that exhibited the highest knockdown efficiency of POR (Figure 5B). The results of the EdU, CCK8, and colony formation assays demonstrated that POR silencing antagonized the ACSM5 overexpression‐mediated inhibition of cell viability and proliferation (Figure 5C–E). Moreover, the increases in Fe²⁺, lipid ROS, and malondialdehyde (MDA) levels were attenuated by POR knockdown in HCC cells overexpressing ACSM5. Additionally, the reduction in glutathione (GSH) was also diminished following POR knockdown in ACSM5‐overexpressing HCC cells (Figure 5F).

FIGURE 5.

ACSM5 promotes ferroptosis in HCC by regulating POR. (A) The expression levels of ACSM5 in LM3 and HepG2 cells transfected with shNC or shACSM5 were measured by RT–qPCR and western blotting. (B) The expression levels of ACSM5 and POR were measured by western blotting. (C–E) Cell proliferation in the indicated cell lines was tested by EDU, CCK8, and colony formation assays. (F) The levels of GSH, MDA, lipid ROS, and Fe²⁺ in HCC cells were compared between the ACSM5‐overexpressing group and the ACSM5‐knockdown group. ***p < 0.001, ****p < 0.0001.

3.6. ACSM5 Enhances the Susceptibility of Liver Cancer Cells to Ferroptosis by Increasing the Level of Arachidonic Acid

As previously mentioned, GO analysis and GSEA indicated that ACSM5 was enriched mainly in metabolic pathways, particularly in linoleic acid metabolism. To further elucidate how ACSM5 regulates fatty acid metabolism, we subsequently employed gas chromatography–mass spectrometry to detect changes in the free fatty acid content of liver cancer cells overexpressing ACSM5. The fatty acids in the samples presented a low relative standard deviation (RSD), which confirmed the reliability of this experiment (Figure 6A). The main changes in fatty acid content are shown in Figure 6B, which revealed that after ACSM5 was overexpressed, the content of medium‐chain fatty acids in the cells was significantly reduced; the contents of some long‐chain fatty acids were decreased, among which the content of linoleic acid was significantly decreased, whereas the content of arachidonic acid was significantly increased. The variable importance in the projection (VIP) value of fatty acid metabolism suggested that overexpressing ACSM5 played a significant role in regulating the metabolism of linoleic acid and arachidonic acid (Figure 6C). To explore whether ARA can increase the sensitivity to ferroptosis, we used erastin to induce ferroptosis, added arachidonic acid (ARA) to liver cancer cells and then detected the intracellular contents of Fe²⁺, MDA, lipid, ROS, and GSH. The results revealed that treatment with erastin and arachidonic acid resulted in lower levels of GSH with higher levels of MDA, lipid ROS and Fe²⁺, which suggested that arachidonic acid could increase the ferroptosis sensitivity of liver cancer cells (Figure 6D).

FIGURE 6.

ACSM5 enhances the susceptibility of liver cancer cells to ferroptosis by increasing the level of arachidonic acid. (A) Relative standard deviation (RSD) of the fatty acid sample. (B) Comparison of the free fatty acid contents between the control group and the ACSM5‐overexpressing group. (C) The VIP value of fatty acid metabolism. (D) The levels of GSH, MDA, lipid ROS, and Fe²⁺ in HCC cells treated with arachidonic acid (10.0 μM) and erastin (10.0 μM). *p < 0.05, ***p < 0.001, ****p < 0.0001.

3.7. ACSM5 Suppresses the Growth of Cell Xenografts in Nude Mice

In addition to the crucial role of ACSM5 in vitro, we further explored the role of ACSM5 in tumorigenesis in vivo in nude mice. As shown in Figure 7A–C, ACSM5 overexpression inhibited tumor growth in nude mice, which presented a smaller tumor volume, a slower growth rate, and a lower tumor weight. POR was also upregulated in ACSM5‐overexpressing tumors in vivo (Figure 7D,E). When ACSM5 was overexpressed, the levels of ferroptosis‐related indicators, including MDA, lipid ROS, and Fe²⁺, were increased, whereas the opposite trend was observed for GSH in vivo (Figure 7F). These results are consistent with the above in vitro experimental results.

FIGURE 7.

ACSM5 overexpression reduces the tumor volume and upregulates the expression of POR and ferroptosis in tumors in vivo. (A) Representative images showing xenograft tumors on Day 24 after subcutaneous injection (n = 5). (B, C) Tumor sizes were measured with a ruler on Days 3, 7, 12, 18, 24, and 28, and tumor weights were determined on Day 24. (D) The expression of ACSM5 in mouse tumors was tested by RT–qPCR and western blotting. (E) Representative immunohistochemistry images showing ACSM5 and POR staining in normal and tumor tissues from mouse tumors. (F) The levels of GSH, MDA, lipid ROS and Fe²⁺ in mouse tumor tissue. ***p < 0.001, ****p < 0.0001.

4. Discussion

ACSM5, a molecule involved in lipid metabolism, is expressed at low levels in multiple types of tumors, indicating a poor prognosis for patients with various malignancies. Our research revealed that ACSM5 plays a similar role in liver cancer. We conducted in vivo and in vitro cell experiments and confirmed that ACSM5 can suppress the proliferation ability and viability of liver cancer cells. Moreover, ferroptosis inhibitors can attenuate this inhibitory effect. Our subsequent experiments demonstrated that ACSM5 induces ferroptosis in liver cancer cells by upregulating the expression of POR and increasing ferroptosis sensitivity by regulating fatty acid metabolism in HCC.

Fatty acid metabolism is intimately associated with the occurrence and development of multiple diseases, including tumors [23, 24, 25]. The ACSMs family, which serves as crucial molecules in regulating lipid metabolism, is involved in modulating the occurrence and development of various diseases as well as tumors. Scholars have discovered that mice with the ACSM3 gene knockout exhibit an accumulation of lauric acid, which in turn upregulates the HNF4A‐P38 MAPK pathway and triggers mitochondrial dysfunction and metabolic syndrome [26]. Additionally, silencing the ACSM1 and ACSM3 genes in prostate cancer results in increased toxic levels of medium‐chain fatty acids and promotes ferroptosis in prostate cancer cells [18]. Previous studies have revealed that ACSM5, a key molecule for the thickening of the ligamentum flavum, is highly susceptible to DNA promoter methylation regulated by DMNT1, leading to downregulated expression. This prompts lipid accumulation in the tissue and ultimately results in hypertrophy of the ligamentum flavum [27, 28, 29]. In liver cancer tissues, ACSM5 is also easily regulated by DMNT1 to promote DNA promoter methylation. The downregulation of ACSM5 leads to the accumulation of various medium‐chain fatty acids and promotes the proliferation of liver cancer cells, which is one of the factors associated with the poor prognosis of patients with liver cancer. However, the specific mechanism of this effect has not been elucidated [30]. Through in vivo and in vitro experiments, we confirmed the crucial role of ACSM5 in the proliferation and activity of liver cancer cells. Moreover, our data clearly revealed that the upregulation of ACSM5 triggers marked changes in various fatty acids. The levels of some medium‐chain fatty acids significantly decreased, whereas the contents of certain long‐chain fatty acids increased. Notably, the content of linoleic acid in liver cancer cells was significantly reduced, whereas the content of arachidonic acid was significantly increased.

Peroxisome proliferator‐activated receptors (PPARs), which are nuclear transcription factors, play important roles in the regulation of energy metabolism [31]. In recent years, we have reported that PPARs play important roles in the occurrence and development of many tumors. In liver cancer, PPARs can regulate the progression of liver cancer through glycolipid metabolism and inflammatory reactions [32]. Interestingly, arachidonic acid was found to be the natural ligand of PPARs and could regulate their expression [33, 34]. Wang et al. reported that arachidonic acid regulated the proliferation of lung cancer cells by modulating the PPARγ signaling pathway [35]. In sebaceous adenomas, arachidonic acid can regulate the PPARγ signaling pathway to modulate phospholipid biosynthesis [36]. These findings suggest that ARA can regulate the development of tumors through the PPAR signaling pathway. In our study, KEGG analysis revealed that the PPAR signaling pathway was significantly enriched in the ACSM5‐high group. We hypothesized that the increase in ARA induced by ACSM5 stimulated the PPAR signaling pathway, which could explain the enrichment of the PPAR signaling pathway in the ACSM5‐high population. ACSM5 may affect the expression of PPARs indirectly through the regulation of ARA, which needs further exploration. In addition, we found that fatty acid degradation was also enriched in the ACSM5‐high group. Similarly, the degradation of medium‐chain fatty acids occurred after ACSM5 overexpression, which was consistent with the above findings. These results suggested that the changes in the lipid profile were essentially consistent with the results of the functional analysis shown in Figure 4.

Disorders of lipid metabolism constitute one of the bases for the occurrence of ferroptosis. Lipid peroxides are the ultimate executors leading to ferroptosis [37]. The lipids involved in ferroptosis are generally polyunsaturated fatty acids (PUFAs). This is because they are more likely to undergo oxidation reactions than saturated and monounsaturated fatty acids. Initially, arachidonic acid and adrenic acid were regarded as the main PUFAs that participate in ferroptosis. Research and verification revealed that linoleic acid, linolenic acid, and other polyunsaturated fatty acids promote ferroptosis [38]. Free polyunsaturated fatty acids (PUFAs) cannot directly participate in inducing ferroptosis. Only PUFAs integrated into membrane lipids (PLs) and peroxidized by reactive oxygen species (ROS) to form lipid peroxides (LPOs) can exert ferroptotic effects. Long‐chain fatty acid‐CoA ligase 4 (ACSL4) can catalyze the binding of intracellular free PUFAs with coenzyme A (COA). The PUFA‐COA formed after binding is catalyzed by lysophosphatidylcholine acyltransferase 3 (LPCAT3) for esterification and integrated into PL to form substances highly prone to peroxidation, namely, PUFA‐PL [39, 40]. Once LPOs are generated, they can undergo a reduction reaction under the action of the GPX4 protein. Moreover, GPX4 can inhibit the activity of lipoxygenase, thereby suppressing the lipid peroxidation reaction. However, if the balance between the synthesis and clearance of LPOs is disrupted, resulting in a lethal accumulation of LPOs and undermining the stability of the membrane, ferroptosis is induced [41]. Previous studies have suggested that polyunsaturated fatty acids, such as linoleic acid, linolenic acid, arachidonic acid, and adrenic acid, can all participate in ferroptosis in organisms. An increase in the content of these polyunsaturated fatty acids can increase the ferroptosis sensitivity of cells [42]. However, the main substrates of ACSL4 are arachidonic acid and eicosapentaenoic acid. A study on intestinal‐type gastric cancer revealed that when the process of linoleic acid metabolism converting into arachidonic acid in intestinal‐type gastric cancer cells was blocked, the ferroptosis sensitivity of tumor cells was significantly reduced. After treatment with arachidonic acid, the ferroptosis sensitivity of intestinal‐type gastric cancer cells was restored. Arachidonic acid is an important direct substrate for ferroptosis [43].

The CYP450 enzyme system is a family of heme‐containing proteins with powerful oxidizing capabilities and belongs to the class of mono‐oxygenases. This system is highly abundant in the liver. Most members of this family are distributed mainly in the mitochondria, Golgi apparatus, and peroxisomes. The main function of these enzymes in the liver is to participate in the metabolism of endogenous substances and a variety of exogenous substances, including drugs and environmental compounds. The main process of their action is to convert lipid‐soluble substances into water‐soluble substances so that they can be excreted from the body, which can play a detoxification role in organisms [44, 45, 46]. POR serves as a crucial electron donor for the cytochrome P450 enzyme system [47]. It is also highly expressed in the liver and takes part in the metabolism of numerous endogenous and exogenous substances [48, 49]. Moreover, as a crucial molecule for ferroptosis, POR does not mediate ferroptosis via System Xc–GPX4. Rather, it promotes ferroptosis by facilitating lipid peroxidation [41, 50]. On the one hand, POR's participation in various oxidative metabolic reactions in the body generates a substantial amount of reactive oxygen species (ROS). On the other hand, after polyunsaturated fatty acids bind to membrane phospholipids to form PUFA‐PEs, POR facilitates the peroxidation reaction between PUFAs‐PEs and ROS, thereby providing fuel for ferroptosis. As previously mentioned, we analyzed the function of ACSM5, in which ACSM5 promoted various metabolic activities of the cytochrome P450 enzyme family. ACSM5 is interconnected with the metabolic pathways involved in POR by regulating these metabolic activities. Our further experiments confirmed that there is a strong correlation between ACSM5 and POR in liver cancer tissues and that the ACSM5 and POR proteins interact. Overexpressed ACSM5 increases the level of lipid peroxidation in liver cancer cells by upregulating the expression of POR, ultimately inducing ferroptosis in these cells. Nevertheless, our study revealed the role of ACSM5 in fatty acid metabolism, which was confined to in vitro experiments. More in‐depth exploration requires in vivo experiments to be carried out for verification.

In conclusion, for the first time, we revealed the mechanism by which ACSM5 inhibits the growth of liver cancer cells by upregulating POR to trigger ferroptosis in HCC. In addition, we investigated the function of ACSM5 in the context of lipid metabolism in liver cancer cells, in which ACSM5 promoted the metabolic transformation of linoleic acid to arachidonic acid, thereby increasing sensitivity to ferroptosis. These findings provide a valuable reference, enabling a deeper exploration of ferroptosis and paving the way for the discovery of novel therapeutic targets in the fight against liver cancer.

Author Contributions

Zhengqiang Wu: conceptualization, project administration, writing – original draft. Xiaofeng Xiong: data curation, formal analysis, methodology, writing – review and editing. Mingyi Dong: data curation, investigation, software. Linfei Luo: resources, validation. Zixiang Huang: resources, validation, visualization. Kedong Xu: methodology, validation. Lianwu Zhao: data curation, software. Fenfen Wang: investigation, supervision, visualization. Zhili Wen: conceptualization, project administration, writing – review and editing.

Ethics Statement

The animal studies were reviewed and approved by the Experimental Animal Welfare Ethics Committee of Nanchang Leyou Biotechnology Co. Ltd. Animal experiment approval number (IACUC Issue No): RYE2023121301. Registry and the Registration No. of the Study/Trial: Not applicable.

Consent

The study informed consent was obtained from all individual participants.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1. Table S1. The primer sequences of ACSM5, POR and GAPDH.

Table S2. The ACSM5 insert fragment sequence for overexpression.

Table S3. The sequences of the shRNA for POR.

Figure S1. ACSM5 expression in patients with HBV‐related liver fibrosis. (A) Patients with high necroinflammatory activity (G2–4) had lower levels of ACSM5. (B) Patients with high fibrosis stage (S2–4) had lower levels of ACSM5.

Figure S2. Knockout of ACSM5 improved the proliferation of 97H cells. (A, B) western blot and qRT–PCR were used to test the expression of ACSM5 in 97H cells after the knockout of ACSM5. (C) Cell plate cloning of 97H cells after the knockout of ACSM5. (D) Cell Counting Kit‐8 assay after the knockout of ACSM5.

Figure S3. Changes in POR enzyme activity after ACSM5 was overexpressed in LM3 and HepG2 cells.