Cancer is the leading cause of human mortality and a significant barrier to increasing life expectancy in countries worldwide (Bray F et al., 2021). Primary liver cancer is the sixth most commonly diagnosed cancer and the third leading cause of cancer death worldwide in 2020, with approximately 906,000 new cases and 830,000 deaths (Sung H et al., 2021). In most areas, morbidity and mortality rates are two to three times higher for men than women (Sung H et al., 2021). Primary liver cancer ranks fifth in global incidence and second in male mortality (Sung H et al., 2021). Hepatectomy or liver transplantation are the two main surgical procedures used to treat primary liver cancer (Maki H et al., 2022), but hepatic ischemia-reperfusion injury (HIRI) is inevitable with hepatectomy or liver transplantation. HIRI may lead to liver dysfunction or even failure, causing severe inflammatory and stress reactions in patients. It can seriously affect various organs and systems, affecting their prognosis and quality of life (Nastos C et al., 2014). There are many studies on the pathophysiological mechanisms of HIRI, but they are still inconclusive and controversial. The concept of ferroptosis was first introduced by Dr Brent R. Stockwell in 2012 (Dixon S J et al., 2012). It is an iron-dependent mode of cell death caused by lipid peroxidation and massive accumulation of reactive oxygen radicals. It is a novel form of programmed cell death distinct from apoptosis, cell necrosis, and cell autophagy. Ferroptosis is associated with various biological contexts, from development to ageing, immunity and cancer^[6]. Recent studies have shown that ferroptosis is closely related to the pathophysiology of many diseases, with ferroptosis being directly linked to ischemia-reperfusion injury (IRI) in many organs and, of course, HIRI being inextricably linked to ferroptosis (Li J. 2020; Stockwell B R. 2022). However, the relationship between HIRI and ferroptosis has been poorly studied, so we have used bioinformatics analysis to investigate the relationship between HIRI and ferroptosis (Figure 1) and provide a direction and foundation for future research.

HIRI is a complex pathophysiological process involving multiple factors acting together to exacerbate liver damage, dysfunction and increased structural damage when blood flow is restored after a period of insufficient or interrupted blood flow to the liver (Jaeschke H. 2003; Klune J R et al., 2010). Not only does it affect the physiological functions of the liver, leading to liver dysfunction or even failure in patients, but the severe stress and inflammatory response brought about by HIRI can cause acute dysfunction of other organs such as the heart, kidneys and lungs, and even acute multi-organ failure. HIRI is, therefore, not just a little complication but a severe complication with devastating systemic effects that severely affect patients’ prognosis and quality of survival (Nastos C et al., 2014). Whereas hepatectomy or liver transplantation is currently the predominant surgical procedure for treating primary liver cancer (Maki H et al., 2022), both inevitably result in HIRI. Therefore, the question of how to reduce or even avoid the harmful effects of HIRI has been a hot topic in clinical work and has not been effectively addressed.

The pathophysiological mechanisms of HIRI are currently known in terms of anaerobic metabolism, oxygen-free radical production, Ca²⁺ overload, Kupffer cell production, activation of the complement system, apoptosis and necrosis (Jaeschke H. 2003; Montalvo-Jave E E et al., 2008; Klune J R et al., 2010). However, many unknown pathophysiological mechanisms exist and have not been confirmed by research. The pathophysiological mechanisms of HIRI have therefore been controversial in clinical practice. Ferroptosis is a newly discovered mode of cell death and has been shown to play an essential role in HIRI (Li J. 2020; Stockwell B R. 2022). As a complement to the pathophysiological mechanisms of HIRI, there are fewer studies in this area, so we explored the relationship between the two in theory employing bioinformatics analysis.

Through our studies, we have found that differences in expression at the gene level occur in the liver before and after reperfusion after a period of ischaemia. We define p-value<0.05 and |logFC|≥1 as DEGs. These DEGs determine the differential expression of proteins. As seen from our plot of the volcano (Figure 2A), the vast majority of these DEGs are up-regulated genes. We, therefore, suggest that the differential proteins expressed by DEGs may be a cause of HIRI. Of these DEGs, 11 of these genes are FRGs (Figure 2B) (Table 1). Most of these FRGs are Driver genes (Zhou N et al., 2020), genes that drive ferroptosis, have a positive regulation of ferroptosis, and have a Mark gene (Zhou N et al., 2020) that does not regulate ferroptosis but which marks the occurrence of ferroptosis, reinforcing the existence of a strong link between ferroptosis and HIRI. In GO analysis of these 11 genes (Figure 3B) with KEGG analysis (Figure 4B), the necessary conditions were provided for us to validate ferroptosis with HIRI further. In the PPI network constructed for these 11 genes (Figure 5A), expression was strongly linked between genes, but the CREB5 gene was not linked to any other gene and could therefore be targeted for elimination in subsequent studies. We analysed the PPI network again using statistical methods (Figure 5B), narrowed it down and screened it again (Figure 5C), and screened for five Hub genes (Table 2). For the five Hub genes GO analysis (Figure 6B) with KEGG analysis in (Figure 7B), which is an essential basis for our following study. In a protein interaction prediction of Hub genes, there were 20 genes with which they interacted that were co-expressed in some way (Figure 8A), which provided a direction for our following study. We applied the study data from the dataset GSE15480 to provide a preliminary validation of our study. As shown in Figure 9B, the FRGs in this dataset overlap with the 11 genes screened in Table 1 and contained all of the Hub genes we screened. This further suggests that ferroptosis plays an essential role in HIRI with these five genes.

Activation transcription factor 3 (ATF3) is a basic leucine zipper (bZIP) DNA binding protein, a member of the ATF/CREB family of transcription factors, and the stress response induces widespread expression of ATF3 (Hai T et al., 1999). Furthermore, HIRI can lead to severe stress reactions in the organism. It has been shown that ATF3 promotes erastin-induced ferroptosis by suppressing system Xc^– (Wang L et al., 2020). ATF3 plays an essential role in ferroptosis (Dai C et al., 2020). In HIRI, ATF3 expression is significantly increased, significantly up-regulated and is a very important DEG (Zabala V et al., 2019; Zhong Y et al., 2019). In a comprehensive study of HIRI factor binding proteomics and transcriptomics (Huang S et al., 2019): it was shown that ATF3 is up-regulated by DEGs. This study reveals that ATF3 may be a core factor in HIRI and has potential clinical implications. In our study, ATF3 was the Hub gene, so we speculate that ferroptosis plays an essential role in HIRI in which ATF3 plays an important role.

Interleukin-6 (IL6) is a cytokine with pleiotropic activity that plays an essential role in the acute immune response of the body (Akira S et al., 1993). IL6 has been shown to be required for the induction of ATF3 (Hai T et al., 1999). In contrast, in bioinformatic analysis of ferroptosis in myocardial IRI (Wang Z et al., 2022), IL6 played an essential role in both. In our study, ATF3 and IL6 were screened as Hub genes in the ferroptosis bioinformatic analysis of HIRI; they were intricately linked in the PPI network and intergenic interaction prediction, and ATF3 and IL6 had the same maximum score when scored by the Cytoscape software CytoHubba functional MCC algorithm. We, therefore, hypothesise that ATF3 and IL6 have a central role in the induction of ferroptosis in HIRI. Also, the Hub genes screened in a raw letter analysis of IRI mechanisms in patients undergoing coronary artery bypass grafting (Shen Z et al., 2019) were identical to the Hub genes IL6, IL1B and PTGS2 predicted by our study, and PTGS2 is a Marker gene, and IL6 and IL1B are both Driver genes, thus reinforcing that ferroptosis has an essential role in HIRI.

How to mitigate or avoid the harm caused by HIRI to patients has always been a significant concern in clinical work and is currently a hot topic of research in clinical work. Ferroptosis, one of the pathophysiological mechanisms of HIRI, is beneficial in reducing HIRI by inhibiting the occurrence of ferroptosis. Numerous studies have shown (Kim K M et al., 2020; Yamada N et al., 2020; Liu H et al., 2021; Luo L et al., 2021) that some drugs, such as ferrostatin-1, alpha-tocopherol, deferoxamine, etc. or a low iron diet can reduce HIRI. Therefore, we have also done research in this area and predicted drug-gene interactions for the five Hub genes that were screened (Figure 8B). These predicted drugs could be a reference for future clinical work in targeting HIRI.

MicroRNA (miRNA) is a class of small, endogenous, non-coding RNA molecules of approximately 20–40 nucleotides in length (Saliminejad K et al., 2019). It has a compelling physiological role in regulating gene expression, controlling early development, cell proliferation, apoptosis, cell death, lipid metabolism and cell differentiation, directly affecting tissues, organs and even our entire system^[40]. miRNA dysregulation has been demonstrated as a causal factor in disease progression, and miRNAs can be used as biomarkers and therapeutic targets (Paul P et al., 2018). In clinical work, we can use miRNA to monitor organ status, diagnose diseases, treat diseases, etc. Numerous studies have shown that miRNA expression is altered in IRI in organs such as the heart, brain, kidney and intestine and that differential miRNA expression occurs in HIRI. miRNA mediates HIRI, while HIRI can be alleviated by regulating miRNA expression (Cao M et al., 2021; Ingram H et al., 2022; Ye L et al., 2022). We performed miRNA expression prediction for five Hub genes (Figure 8C), further revealing the relationship between ferroptosis and HIRI, which can be used to diagnose the occurrence of ferroptosis in HIRI and may provide a new approach to mitigate HIRI in clinical work. The miRNAs we predicted have the same results as those predicted in the raw letter analysis of IRI mechanisms in coronary artery bypass grafting patients (Shen Z et al., 2019) and different ones. Among them, we predicted has-miR-128-3p, and it has been shown that inhibition of miR-128-3p by Tongxinluo protects human cardiomyocytes from ischemia/reperfusion injury via upregulation of p70s6k1/p-p70s6k1 (Chen G et al., 2017). We predicted hsa-miR-24-3p, and it has been shown that rosuvastatin alleviates ischemia/reperfusion injury in cardiomyocytes by downregulating hsa-miR-24-3p to target up-regulated uncoupling protein 2 (Wei W et al., 2019). We, therefore, hypothesise that there is also some benefit in alleviating ferroptosis in HIRI by regulating miRNA levels.

Although we have done a great deal of work on the relationship between HIRI and ferroptosis, we acknowledge that our study has some limitations. We have only speculated on the link between HIRI and ferroptosis in theory and have used another HIRI-related dataset for preliminary validation of this. We have not further validated our conclusions in animal experiments or clinical trials. Also, our data selection is somewhat limited in that the datasets we selected were all about HIRI in liver transplantation and not about HIRI in liver resection. Therefore, these limitations will also be the focus of our future research efforts to investigate the link between HIRI and ferroptosis in depth and to design and implement rigorous animal or clinical trials to test the theoretical speculations we have obtained. We will apply the results of our research to clinical work to reduce the risks of HIRI, improve the prognosis of patients and enhance their quality of life.

Ferroptosis, a newly discovered mode of cell death, plays an essential role in HIRI. The 11 DEGs and 5 Hub genes of HIRI that we screened by the bioinformatic analysis were closely associated with ferroptosis. Further bioinformatic analysis methods of these genes are beneficial in revealing the potential biological pathways and mechanisms of ferroptosis in HIRI, providing a reference for further in-depth studies. The targeted drugs and miRNA predicted for the five Hubs provide some reference value for clinical work in assessing the diagnosis of ferroptosis in HIRI, the extent of its occurrence and the mitigation of HIRI by means of ferroptosis inhibition. The five Hub genes can be used as potential biomarkers and therapeutic targets for HIRI and also provide some theoretical basis and research directions to explore the relationship between HIRI and ferroptosis further.Oliveros, 2007, R Core Team. R, 2022.