Abstract

   Hepatic steatosis significantly elevates the vulnerability of the graft
   to ischemia-reperfusion (I/R) injury during liver transplantation (LT).
   We investigated the protective role of insulin-induced gene 2 (Insig2)
   in steatotic liver’s I/R injury and underlying mechanisms. Employing
   mouse model with Insig2 knock-out or hepatocyte-specific overexpression
   and high-fat diets to induce steatosis, we subjected these mice to
   hepatic I/R injury. The primary hepatocytes isolated from steatotic
   liver were used in in vitro hypoxia/reoxygenation (H/R) experiment. Our
   integrated in vivo and in vitro approach uncovered that Insig2
   deficiency exacerbated steatotic liver’s damage following hepatic I/R
   injury, whereas its overexpression offers protection. Mechanically,
   integrative analysis of transcriptome, proteome, and metabolome found
   that Insig2 deficiency disturbed lipid metabolism and oxidative stress
   homeostasis, particularly inhibiting GPX4 expression to induce
   ferroptosis. Furthermore, chemical inhibition of ferroptosis reversed
   the deleterious effect of Insig2 deficiency; whereas the protective
   influence of Insig2 overexpression was negated by the target inhibition
   of GPX4, leading to an exacerbation of hepatic I/R damage. These
   insights underscored the potential of the Insig2-GPX4 axis as a
   therapeutic target, presenting a novel avenue for enhancing the
   resilience of steatotic liver grafts against I/R injury.

   graphic file with name 41420_2025_2406_Figa_HTML.jpg

   Subject terms: Transcriptomics, Metabolic disorders

Introduction

   Liver transplantation (LT) continues to stand out as the most
   efficacious treatment for individuals afflicted with acute liver
   failure, end-stage liver disease, or advanced liver cancer [[48]1,
   [49]2]. Given the increasing gap between organ availability and demand,
   there is an inevitable trend in utilizing marginal liver donors, which
   encompass older donors, donors after cardiac death, and steatotic
   donors [[50]3]. Approaches to enhance the viability of these grafts may
   significantly expand the pool of donors eligible for LT. With the
   growing morbidity of non-alcoholic fatty liver disease (NAFLD)
   worldwide, steatotic liver has become increasingly common in donor
   organs [[51]4, [52]5]. The presence of steatosis renders the liver more
   sensitive and less tolerant to ischemia-reperfusion (I/R) injury, as
   lipid deposition in hepatocytes complicates the underlying
   pathophysiologic mechanism [[53]6, [54]7]. The utilization of steatotic
   graft was at a higher risk for allograft dysfunction and primary
   nonfunction [[55]8]. Current studies have shown that liver steatosis
   greater than 10% is associated with early allograft dysfunction,
   long-term graft loss, and tumor recurrence [[56]9]. Several
   interventions aimed at mitigating preservation injury in steatotic
   livers have been explored in prior studies [[57]10]. Nonetheless, novel
   graft preservation solutions and machine perfusion techniques have yet
   to realize their benefits in clinical practice. In-depth investigations
   into the mechanisms underlying I/R injury in the context of steatotic
   liver are crucial for identifying therapeutic strategies to ensure
   graft safety and further expand the donor pool.

   Ferroptosis was reported to contribute to liver injury and dysfunction
   in LT [[58]11]. Ferroptosis is characterized by excessive lipid
   peroxide buildup and disrupted cellular redox equilibrium, which often
   involves the impaired function of the lipid repair enzyme glutathione
   peroxidase 4 (GPX4), significant iron accumulation and the peroxidation
   of polyunsaturated fatty acids (PUFAs) [[59]12, [60]13]. Furthermore,
   ferroptosis in hepatocytes has been recognized the major cell death
   type involved in steatotic liver I/R injury [[61]14, [62]15].

   Insulin-induced gene 2 (Insig2) is an integral protein located in the
   endoplasmic reticulum and is modulated by insulin signals. Its primary
   function involves regulating the activity of sterol regulatory
   element-binding protein 1c (SREBP-1c), thereby participating in the
   synthesis of cholesterol and fatty acids [[63]16, [64]17].
   Additionally, Insig2 plays a crucial role as a regulator of energy
   production and cellular immune metabolism [[65]18–[66]20]. Insig2 was
   recently characterized to be involved in detection of abnormal lipids
   and reactive oxygen species (ROS), and in maintenance of intracellular
   homeostasis [[67]21, [68]22]. Furthermore, our previous research
   revealed Insig2 protected hepatic I/R injury via remolding glucose
   metabolism [[69]23]. Insig2 modulated the translocation of SCAP-SREBP1
   and the activation of SREBP1, which ultimately led to the increased
   transcription of stearoyl-CoA desaturase 1, thereby inhibiting
   ferroptosis [[70]24]. However, the function of Insig2 in steatosis
   graft injury remains to be elucidated.

   Considering the role of Insig2 in preserving intracellular immune and
   metabolic balance, we hypothesized that Insig2 could likewise exert a
   protective effect in steatosis hepatic I/R injury, and also aimed to
   investigate the mechanisms underlying this effect as well as the
   therapeutic possibilities for marginal donor recovery.

Results

Hepatic Insig2 expression is downregulated during I/R or H/R in steatotic
liver

   In the rat model of steatosis LT, we observed the transcriptional
   expression levels of Insig2 in the transplanted liver were
   significantly decreased during LT ([71]GSE193764, Fig. [72]1A). We
   aimed to validate this finding in mouse model of hepatic I/R injury.
   After 8 weeks of HFD diet feeding, mice exhibited elevated serum and
   hepatic TG levels compared to chow diet (Fig. [73]1B). The H&E staining
   revealed moderate hepatic steatosis, characterized by the presence of
   lipid vacuoles [[74]10, [75]25] (Fig. [76]1C). The HFD mice were
   subjected to 1.5 h of ischemia, followed by 6 h of reperfusion.
   Moreover, the steatosis primary hepatocytes were isolated from the HFD
   mice’s liver, and subjected to 6 h of hypoxia, followed by 6 h of
   oxygenation. The endogenous expression of Insig2 protein was evaluated
   by western blot. As anticipated, the results demonstrated that I/R
   injury downregulated Insig2 expression in steatosis hepatocytes (Fig.
   [77]1D, E).

Fig. 1. Expression of Insig2 is down-regulated after I/R injury in steatotic
liver.

   [78]Fig. 1
   [79]Open in a new tab

   A The change of Insig2 mRNA levels in mice orthotopic liver
   transplantation with steatotic graft from [80]GSE193764 Database. B, C
   The serum and liver tissue TG levels in mice fed HFD for 8 weeks or
   normal diets (n = 4/group). D Western blotting analysis of endogenous
   Insig2 expression in the steatotic liver subjected to sham operation or
   I/R injury (n = 3/group). E Western blotting analysis of Insig2 protein
   expression in primary hepatocytes with steatosis subjected to H/R
   stimulation (n = 2/group). *p < 0.05, **p < 0.01.

Insig2 knockout in mice with diet-induced steatotic liver aggravates liver
damage following I/R injury

   Firstly, Insig2 KO and WT mice were fed with HFD diet for 8 weeks, all
   mice developed moderate hepatic steatosis. We found that the hepatic TG
   accumulation evidenced by Oil Red O staining was not different by
   Insig2 deficiency (Fig. [81]2A). To document the role of Insig2 on
   steatosis hepatic I/R injury, Insig2 KO and WT mice with steatotic
   liver were subjected to I/R surgery. The serum ALT and AST levels were
   markedly elevated in Insig2 KO mice compared to WT controls (Fig.
   [82]2B). Liver necrosis area, indicated by H&E staining, was also
   significantly greater in Insig2 KO mice (Fig. [83]2C). Besides, Insig2
   deficiency was associated with increased levels of inflammatory
   cytokines (IL-6, IL-1β, TNF-α) in serum and heightened hepatic mRNA
   expression of proinflammatory cytokines and chemokines (IL-6, IL-1β,
   TNF-α, Ccl2, Cxcl10), as depicted in Fig. [84]2D, E. IF assays revealed
   heightened infiltration degree of F4/80 and MPO-positive cells in
   Insig2 KO liver (Fig. [85]2F). Furthermore, we quantified the
   expression of apoptosis-related proteins in Insig2 KO and WT liver
   subjected to I/R injury. Our results demonstrated that Insig2 KO
   increased pro-apoptotic protein Bax and decreased anti-apoptotic Bcl2
   in KO mice (Fig. [86]2G). TUNEL staining supported these findings that
   Insig2 KO exacerbated cell apoptosis in steatotic liver post I/R injury
   (Fig. [87]2H).

Fig. 2. Insig2 deficiency exacerbates hepatic I/R injury in steatotic liver.

   [88]Fig. 2
   [89]Open in a new tab

   A Oil Red O staining of liver sections in WT and Insig2-KO mice fed
   with HFD for 8 weeks. B. The serum ALT and AST levels in WT and
   Insig2-KO mice with steatotic liver following sham (n = 3/group) and
   I/R surgery (n = 8/group). C Representative H&E staining liver sections
   from liver tissues of WT and Insig2-KO mice with steatitic liver
   following sham operation and I/R injury (n = 4/group). D The serum
   levels of inflammatory cytokines (IL-6, IL-1β, TNF-α) in WT and
   Insig2-KO mice with steatotic livers following I/R (n = 6/group). E
   qPCR analysis of the mRNA levels of proinflammatory cytokines and
   chemokines (IL-6, IL-1β, TNF-α, Ccl2, and Cxcl10) in WT and Insig2-KO
   mice with steatotic livers following I/R (n = 4/group). F
   Representative F4/80 and MPO IF staining (red) and statistics in liver
   sections of WT and Insig2-KO mice following sham or I/R surgery
   (n = 4/group). G Western blotting analysis of endogenous
   apoptosis-related protein expression in the steatotic liver subjected
   to I/R injury (n = 4/group). H TUNEL staining (green) and statistics in
   liver sections of WT and Insig2 KO mice following sham or I/R surgery
   (n = 4/ group). *p < 0.05, **p < 0.01.

Hepatocyte-specific Insig2 overexpression alleviated liver steatosis damage
following I/R injury

   To further delineate the functional role of Insig2 in hepatic I/R
   injury with steatosis, hepatocyte-specific Insig2 overexpression
   (Insig2 OE) mice were generated via AAV8-Insig2 intravenous injection,
   with AAV8-control-infected mice as parallel controls. After 8 weeks of
   HFD diet feeding, Insig2 OE and control mice with steatotic liver were
   established and successfully confirmed (Fig. [90]3A). Prior to I/R,
   hepatic TG accumulation remained unchanged (Fig. [91]3B). At 6 h
   post-I/R, Insig2 OE mice exhibited significantly lower serum ALT and
   AST levels compared to controls (Fig. [92]3C), along with reduced
   hepatic necrosis (Fig. [93]3D). Furthermore, Insig2 OE mice displayed
   decreased levels of the secretion and mRNA expression of
   proinflammatory factors, reduced inflammatory cell infiltration in
   liver (Fig. [94]3E–G). Western blot and TUNEL staining revealed that
   Insig2 OE significantly mitigated apoptosis and modulated apoptotic
   protein expression (Fig. [95]3H, I). Collectively, these findings
   indicated that Insig2 overexpression ameliorated hepatic damage,
   inflammation, and apoptosis during I/R in steatotic livers.

Fig. 3. Hepatic Insig2 overexpression ameliorates hepatic I/R injury in
steatotic liver.

   [96]Fig. 3
   [97]Open in a new tab

   A Western blot analysis of AAV8-Insig2 and AAV8-NC transfection
   efficiency in steatotic liver. B Oil Red O staining of liver sections
   in AAV8-NC and AAV8-Insig2 mice fed with HFD for 8 weeks. C The serum
   ALT and AST levels in AAV8-NC and AAV8-Insig2 mice with steatotic liver
   following sham (n = 3/group) and I/R surgery (n = 8/group). D
   Representative H&E staining liver sections from liver tissues of
   AAV8-NC and AAV8-Insig2 mice with steatitic liver following sham
   operation and I/R injury (n = 4/group). E The serum levels of
   inflammatory cytokines (IL-6, IL-1β, TNF-α) in AAV8-NC and AAV8-Insig2
   mice with steatotic livers following I/R (n = 6/group). F qPCR analysis
   of the mRNA levels of proinflammatory cytokines and chemokines (IL-6,
   IL-1β, TNF-α, Ccl2, and Cxcl10) in AAV8-NC and AAV8-Insig2 mice with
   steatotic livers following I/R (n = 4/group). G Representative F4/80
   and MPO IF staining (red) and statistics in liver sections of AAV8-NC
   and AAV8-Insig2 mice following sham or I/R surgery (n = 4/group). H
   Western blotting analysis of endogenous apoptosis-related protein
   expression in the steatotic liver subjected to I/R injury
   (n = 4/group). I TUNEL staining (green) and statistics in liver
   sections of AAV8-NC and AAV8-Insig2 mice following sham or I/R surgery
   (n = 4/ group). *p < 0.05, **p < 0.01.

Insig2 mitigated inflammation and apoptosis in steatotic hepatocytes from H/R
stimulation

   Primary hepatocytes, derived from mice treated with AAV8 for either
   Insig2 knockdown or overexpression at a uniform age, were treated with
   OA:PA and then exposed to H/R stimulation (Fig. [98]4A–D). Consistent
   with in vivo findings, Insig2 knockdown hepatocytes under steatotic
   conditions exhibited markedly increased levels of inflammatory
   cytokines (IL-6, IL-1β, TNF-α) post-H/R (Fig. [99]4B). Additionally,
   pro-apoptotic proteins (Bax, Cleaved-caspase 3) were upregulated,
   whereas the anti-apoptotic protein (Bcl2) was downregulated in these
   knockdown cells (Fig. [100]4C). Conversely, hepatocytes with Insig2
   overexpression showed significantly diminished inflammation and
   apoptosis levels (Fig. [101]4E, F).

Fig. 4. Insig2 alleviates primary hepatocytes inflammation and apoptosis
during H/R.

   [102]Fig. 4
   [103]Open in a new tab

   A Insig2 protein expression in primary hepatocytes extracted from mice
   infected with shRNA or shInsig2 AAV8 followed by incubation of OA:PA. B
   mRNA levels of proinflammatory factors (IL-6, IL-1β, TNF-α) in
   Insig2-knockdown hepatocytes with steatosis after H/R stimulation
   (n = 3/group). C Expression of apoptosis-related proteins in
   Insig2-knockdown hepatocytes with steatosis after H/R stimulation
   (n = 3/group). D Insig2 protein expression in primary hepatocytes
   extracted from mice infected with control or Insig2 overexpressing AAV8
   followed by incubation of OA:PA. E mRNA levels of proinflammatory
   factors in Insig2-overexpression hepatocytes with steatosis after H/R
   stimulation (n = 3/group). F Expression of apoptosis-related proteins
   in Insig2-overexpression hepatocytes with steatosis after H/R
   stimulation (n = 3/group). *p < 0.05, **p < 0.01.

Multi-omics analysis indicates that Insig2 regulated lipid remodeling and
ferroptosis in steatosis I/R injury

   To explore the mechanisms underlying Insig2’s role in steatotic hepatic
   I/R injury, a multi-omics strategy was employed, analyzing
   transcriptomic, proteomic, and metabolomic data from steatotic liver
   tissues of Insig2 KO and WT mice post-I/R (Fig. [104]5A). The
   hierarchical clustering and volcano mapping delineated the profiles of
   DEGs and DEPs (Fig. S1). The KEGG enrichment analysis revealed
   significant enrichment of lipid metabolism and ferroptosis pathways in
   both transcriptomic and proteomic data (Fig. [105]5B, C). Additionally,
   the GO pathway analysis identified various differential pathways
   related to intermediary metabolism, oxidoreductase activity, and immune
   response (Fig. S2).

Fig. 5. Transcriptome, proteome, and metabolome integrative analysis reveals
disturbed pathways in Insig2-KO mice with steatotic liver subjected to I/R
injury.

   [106]Fig. 5
   [107]Open in a new tab

   A The experimental design of the muti-omics approach. B The bubble plot
   of KEGG pathway enrichment of differentially expressed genes. C The
   bubble plot of KEGG enrichment pathway of differential expressed
   proteins. D The heatmaps of dysregulated metabolites in metabolomic
   analysis. E The substrate levels of oxidized phospholipids that
   contained polyunsaturated fatty acids (PUFAs) chains (arachidonic acid
   (ArA) and adrenic acid (AdA)) in metabolomic analysis (n = 6/group). F
   The levels of oxidized lipids in metabolomic analysis (n = 6/group).
   *p < 0.05, **p < 0.01.

   The metabolome was particularly insightful in characteristic of
   ferroptosis, as it detected altered metabolites that were directly
   involved in iron metabolism and lipid peroxidation process. The
   metabolomic analysis uncovered pronounced disparities in lipid profiles
   between Insig2 KO and WT mice, with a notable increase in fatty acids
   (FAs), phosphatidylethanolamines (PEs) and phosphatidylcholines (PCs)
   in Insig2 KO mice with steatotic livers (Fig. [108]5D). Further
   evaluation of substrates of PUFAs peroxidation and products of
   oxidative lipids, revealed elevated levels of oxidized arachidonic acid
   (ArA) and adrenic acid (AdA) metabolites, notably 12-HETE and 15-HETE
   (Fig. [109]5E, F). Collectively, these findings indicated that Insig2
   deficiency disrupted lipid metabolism and enhanced ferroptosis,
   exacerbating hepatic I/R injury in the context of steatosis.

Insig2 protects against steatosis I/R injury partly by inhibiting ferroptosis

   To ascertain Insig2’s role in ferroptosis, we measured serum and
   hepatic iron levels, the NADPH/NADP^+ and GSH/GSSG ratios, MDA levels,
   and mRNA expressions of ferroptosis biomarkers (PTGS2, ALOX12) in the
   liver tissues. We found that the elevated iron concentrations in serum
   and liver tissues (Fig. [110]6A, B), increased PTGS2 and ALOX12
   expression (Fig. [111]6C), and higher hepatic MDA levels (Fig. [112]6D)
   were observed in Insig2 KO mice following steatosis I/R injury compared
   to WT mice. By contrast, the hepatic NADPH/NADP^+ and GSH/GSSG ratios
   were diminished, indicating an attenuated antioxidant capacity in
   Insig2 KO mice (Fig. [113]6E, F).

Fig. 6. Insig2 defiency facilitates I/R-induced ferroptosis in steatotic
liver.

   [114]Fig. 6
   [115]Open in a new tab

   A Relative PTGS2 and ALOX12 mRNA levels in liver (n = 6/group), (B, C).
   Iron contents in liver tissue and serum (n = 6/group), (D–F). Hepatic
   levels of NADPH/NADP^+, MDA, and GSH/GSSG (n = 6/group) were detected
   at sham and I/R groups from Insig2 KO and WT mice with steatotic liver.
   *p < 0.05, **p < 0.01.

   To further establish if Insig2 protected steatotic liver against I/R
   injury partly through the inhibition of ferroptosis, we administrated
   Fer-1 (a selective ferroptosis inhibitor) to Insig2 KO and WT mice
   before I/R surgery. As anticipated, Fer-1 preconditioning successfully
   rescued the exacerbating effects of Insig2 deficiency on I/R-induced
   liver damage (Fig. [116]7). Moreover, the pro-inflammatory and
   pro-apoptotic effects enhanced by the deficiency of Insig2 were
   significantly restored by Fer-1 (Fig. [117]7).

Fig. 7. Insig2 protects against I/R injury in steatotic liver by partly
regulating ferroptosis.

   [118]Fig. 7
   [119]Open in a new tab

   A Graphical representation of experimental design. Fer-1 was i.v.
   injected 1 h before I/R of Insig2 KO and control WT mice with steatotic
   liver to inhibit ferroptosis. B The serum ALT and AST levels in the
   indicated group (n = 5/group). C Representative H&E staining liver
   sections from indicated group (n = 4/group). D The serum levels of
   inflammatory cytokines (IL-6, IL-1β, TNF-α, n = 4/group). E qPCR
   analysis of the mRNA levels of proinflammatory cytokines (IL-6, IL-1β,
   TNF-α, n = 4/group) and ferroptosis indicator (PTGS2, ALOX12,
   n = 4/group). F Representative F4/80 and MPO IF staining and TUNEL
   staining in liver sections (n = 4/group). *p < 0.05, **p < 0.01.

Insig2 inhibits GPX4-dependent ferroptosis in steatosis I/R injury

   Further, we explored how Insig2 affected ferroptosis. Among all key
   genes and proteins in the ferroptosis pathways, we found that GPX4 was
   both down-regulated in Insig2 KO mice compared to WT mice (Fig. S3). We
   confirmed a significant reduction in hepatic GPX4 expression in
   Insig2-deficient livers, suggesting a mechanism by which Insig2
   modulates GPX4 level in steatotic liver (Fig. [120]8A, B). Furthermore,
   the AAV8-Insig2 viruses were injected into 8-week HFD-fed mice, then
   RSL3 (a competitively GPX4 binding inhibitor) was administrated to
   confirm the role of GPX4 in Insig2 OE-induced steatotic liver I/R
   damage (Fig. S4). After hepatic I/R injury, RSL3 significantly reversed
   the protective effect of Insig2 overexpressing (Fig. [121]8C, D). To
   illustrate, RSL3 administration activated the ferroptosis, upregulated
   the expression of inflammatory cytokines, and accelerated cell
   apoptosis (Fig. [122]8E–G). In summary, these results indicated that
   Insig2 could suppress ferroptosis in a GPX4-dependent manner to protect
   against steatotic liver injury under I/R.

Fig. 8. Insig2 regulates I/R injury by inhibiting GPX4-dependent ferroptosis
in steatotic liver.

   [123]Fig. 8
   [124]Open in a new tab

   A, B The mRNA and protein levels of GPX4 were evaluated in livers from
   WT and Insig2 KO mice with steatotic liver (n = 6/group for mRNA
   samples, n = 4/group for protein samples). C–G RSL3 was i.p. injected
   1 h before I/R of Insig2 OE and control mice with steatotic liver to
   inhibit GPX4. The serum ALT and AST (C, n = 5/group), necrosis areas in
   liver section (D, n = 4/group), serum inflammatory cytokines (E,
   n = 4/group), expression of proinflammatory cytokines and ferroptosis
   indicator (F, n = 4/group), Representative F4/80 and MPO IF staining
   and TUNEL staining in liver sections (G, n = 4/group). *p < 0.05,
   **p < 0.01.

Discussion

   In the face of the high mortality on the liver transplant waitlist and
   the scarcity of liver donors, donors with steatosis are frequently
   considered for use as marginal grafts. Growing evidence supported that
   mild-moderate macrosteatotic grafts could be successfully used for LT
   with appropriate donor-recipient matching [[125]26]. However, numerous
   studies have indicated that the presence of hepatic steatosis was prone
   to I/R injury, thereby raising the potential for poor post-transplant
   complications [[126]27, [127]28]. Factors such as mitochondrial
   dysfunction, metabolism disorder, lowered antioxidant capacity, and
   amplified inflammatory response were believed to contribute to the
   increased vulnerability of steatotic livers to I/R injury [[128]9,
   [129]29]. There is an urgent demand for treatments that can mitigate
   the harmful effects of I/R injury in steatotic livers. In our study,
   for the first time, we examined Insig2’s protective effects and
   underlying mechanisms against steatotic liver I/R injury, providing
   valuable insights for devising strategies to mitigate graft damage when
   utilizing steatotic donors.

   Insig2 is essential in the body’s reaction to insulin and a variety of
   biological signals, exhibiting widespread expression throughout
   numerous tissues, with a particularly heightened presence in
   hepatocytes of the liver [[130]30, [131]31]. Extensive research has
   highlighted Insig2’s significant involvement in maintaining glucose and
   lipid balance, primarily through the Insig-SREBP pathway
   [[132]32–[133]34]. Furthermore, several studies have underscored
   Insig-SREBP’s regulatory importance in cell death, redox equilibrium,
   and inflammatory response, which were closely related to I/R injury
   [[134]21, [135]22, [136]35]. In our study, the expression of Insig2 was
   found to be significantly downregulated in both rat liver
   transplantation and mouse model of I/R injury in steatotic liver,
   suggesting its potential role in regulating steatosis hepatic I/R
   injury. Subsequently, we utilized Insig2 knockout and
   hepatocyte-specific Insig2 overexpression mice to investigate its
   biological role in steatotic I/R injury. By administering HFD to mice
   for 8 weeks, we successfully established a mouse model of liver
   steatosis with a macrosteatosis (MaS) degree ranging from ~30 to 60%
   [[137]10, [138]25]. This model was utilized as a donor liver to
   simulate the clinical application of moderate MaS donor livers.
   Notably, we observed that neither Insig2 knockout nor overexpression
   altered the predominant determinant of hepatic steatosis during HFD
   feeding—TG. We speculated that the possible reason lies in the
   complementary roles of Insig1 and Insig2 in regulating lipids,
   particularly TG (Fig. [139]S5) [[140]36]. However, Insig2 did modulate
   the extent of liver damage, apoptosis, and inflammatory responses
   following I/R. This also indicated that lipid substances such as
   cholesterol and free fatty acids played a more significant role in the
   susceptibility to hepatic I/R injury [[141]37]. To determine its
   downstream mechanism, transcriptomic and proteomics analysis suggested
   that Insig2 KO led to impairment of lipid and energy metabolic
   pathways, including ferroptosis. The present investigation provided
   evidence that Insig2 KO expedited iron accumulation, increased
   oxidative stress (MDA), and weakened the in vivo antioxidant defense
   systems (NADPH, GSH). Crucially, pharmacological inhibition of
   ferroptosis significantly counteracted the exacerbated outcomes
   associated with Insig2 deficiency in steatotic hepatic I/R injury.
   Hence, our research findings indicated that Insig2 protected steatotic
   liver damage induced by I/R through the partly suppression of
   ferroptosis in hepatocytes.

   Ferroptosis, a form of cell death characterized by iron-mediated lipid
   peroxidation, assumes a pivotal role in the LT process and the
   advancement of hepatic I/R injury [[142]38–[143]40]. The condition of
   iron overload and lipid peroxidation stress are more commonly observed
   in steatosis hepatic I/R [[144]15]. Jiao’s research indicated that
   ferroptosis is the predominant form of cell death occurring in
   hepatocytes in steatotic liver I/R injury [[145]14]. Our previous study
   demonstrated that Insig2 could upregulate the pentose phosphate pathway
   (PPP), which is an important glucose metabolism pathway involved in the
   regulation of NADPH, GSH, and ROS [[146]23]. Recent reports also
   revealed that PPP imposed potential regulatory roles in ferroptosis
   [[147]41]. Given the interplay between lipid metabolism and iron
   homeostasis, Insig2 may indirectly influence the levels of PUFAs within
   the cell membrane. These PUFAs are the primary substrates for lipid
   peroxidation in the process of ferroptosis [[148]42, [149]43]. Based on
   metabolome sequencing, the difference in the hepatic content of
   metabolites was been fully characterized. We found that the level of
   oxidative lipids, oxidized AA metabolites were significantly increased
   in Insig2 KO liver tissues upon I/R injury, which was in accordance
   with the result of transcriptome and proteome.

   Additionally, we found that Insig2 and the ferroptosis inhibitor
   modulated the levels of cellular necrosis and apoptosis in liver. These
   mice exhibited similar trends in lipid peroxidation and ferroptotic
   activity, indicating a potential interplay among ferroptosis and
   different modes of cell death [[150]12]. While these forms of cell
   death are distinct in their mechanisms, they can interact in complex
   ways. Studies have shown that lipid peroxides could regulate NF-κB and
   Casepase signals to initiate apoptosis [[151]44]. To illustrate, under
   conditions of iron overload, cells may simultaneously undergo
   ferroptosis as well as apoptosis or necrosis. It is well-recognized
   that ROS generation and lipid peroxidation contribute to hepatic injury
   through the promotion of inflammation and cell death [[152]45,
   [153]46]. In this context, the inflammatory cytokines and infiltration
   of immune cells may serve as mediators of the interactions between
   ferroptosis, apoptosis, and necrosis pathways in steatotic liver
   subjected to I/R injury.

   To investigate the specific mechanisms by which Insig2 regulates
   ferroptosis in steatosis hepatic I/R, we have identified a
   molecule—GPX4, which exhibited significant changes at both the
   transcriptional and protein levels from the sequencing data.
   Ferroptosis is regulated by cyst(e)ine/GSH/GPX4 antioxidant pathway,
   and GPX4 plays a cornerstone role in protecting cells by reducing lipid
   hydroperoxides to their corresponding alcohols, thereby preventing the
   propagation of oxidative damage [[154]13, [155]38]. GPX4-mediated
   ferroptosis is closely related to the progression of free fatty
   acid-induced steatosis [[156]47]. Additionally, programmed cell death
   orchestrated by GPX4-dependent pathways is essential for the
   pathophysiology of I/R injury in the liver and other organs [[157]48,
   [158]49]. Our research first introduced Insig2 as a novel regulator of
   GPX4 in steatotic hepatic I/R injury. Insig2 could block SREBP
   processing to maintain lipid homeostasis, which could indirectly affect
   the level of PUFAs and other oxidative lipids. In vivo oxidative stress
   levels and SREBP factors could affect the expression of GPX4, thereby
   regulating the sensitivity of cells to ferroptosis [[159]50, [160]51].

   However, our research also had limitations that were important to
   highlight. Firstly, we did not elucidate the detailed molecular
   mechanism by which Insig2 influenced GPX4 expression. We proposed that
   Insig2/SREBP axis might be responsible to lipid contents remodeling.
   Secondly, although our preliminary research in steatotic graft was
   encouraging, further clinical investigations were essential to
   critically scrutinize the applicability of our findings.

   In conclusion, our research found that Insig2 in hepatocytes protected
   against I/R injury in steatotic liver by regulating GPX4-dependent
   ferroptosis. Therefore, targeting Insig2/GPX4 axis may be a promising
   approach of steatotic graft injury treatment.

Materials and methods

Animals and study approval

   Inbred wild-type (WT) male C57BL/6 mice (4 weeks old) were purchased
   from Hangzhou Medical College. The Insig2 knock-out (KO) mice of
   C57BL/6 were kindly provided by Prof. Di Wang, Zhejiang University
   School of Medicine. All animals were accommodated at 12-h light/dark
   cycle, with unrestricted access to both water and food. Steatotic liver
   in mice was induced by feeding with high-fat diet (HFD, 60 kcal% fat)
   for 8 weeks (Research Diet, D12492, USA). The animal protocols were
   observed in the guidelines of the National Institute of Health (NIH
   Publications NO. 88−23, revised in 1996). The experiments were approved
   by the Zhejiang Provincial People’s Hospital (Affiliated People’s
   Hospital), Hangzhou Medical College (Approval No. 20250110966929).

Hepatic I/R surgery

   The surgical technique was employed to replicate the graft damage
   observed in LT. Each experimental group included 4–8 mice. If a mouse
   died during the surgical procedure, its data were excluded from the
   analysis. Initially, the hepatic artery and portal vein branches
   supplying the right and triangular lobes were temporarily occluded
   using a precision microvascular clamp for a duration of 1.5 h.
   Subsequently, the clamp was removed to restore blood flow. The sham
   group performed the same surgery except for the hepatic ischemia. At
   6 h post-reperfusion, the animals were euthanized, and both their
   circulating blood and liver samples were collected for further
   analysis.

Liver biochemical measurement and triglyceride determination

   Serum alanine aminotransferase (ALT) and aspartate aminotransferase
   (AST) were measured by Chemray 800 chemistry analyzer (Shenzhen,
   China). Serum inflammatory cytokines (IL-6, IL-1β, and TNF-α) were
   quantified by enzyme-linked immunosorbent assay (ELISA) kits
   (Invitrogen, USA) according to manufacturer’s instructions. The
   triglycerides (TG) in serum and liver tissues were determined using a
   Triglyceride assay kit (NJJCBIO, Nanjing, China) according to the
   protocol.

Quantitative real‐time PCR and Western blot

   Total RNA was isolated from hepatic tissues and cells utilizing the
   Total RNA Isolation Kit (Vazyme, Nanjing, China), adhering to the
   protocol provided. One microgram of RNA was converted to complementary
   DNA (cDNA) using the First Strand cDNA Synthesis Kit (Vazyme, Nanjing,
   China). Subsequent quantitative real-time PCR (qPCR) analysis was
   conducted on the ABI QuantStudio 6 Flex platform (Thermo Fisher, USA),
   employing the SYBR Green PCR Master Mix (Vazyme, Nanjing, China). The
   mRNA levels were standardized against β-actin expression. Primer
   sequences for the target genes are detailed in Supplementary Table
   [161]S1.

   Protein extraction from liver samples was carried out using
   radioimmunoprecipitation assay lysis buffer (Fudebio, Hangzhou, China),
   with quantification performed via the BCA Protein Assay Kit (Fudebio,
   Hangzhou, China). The proteins were resolved by SDS-PAGE and then
   transferred onto polyvinylidene fluoride membranes (Millipore, USA).
   Membranes were exposed to primary antibodies at 4 °C for an overnight
   period, followed by incubation with secondary antibodies (Fudebio,
   Hangzhou, China) for 2 h at ambient temperature. Protein band intensity
   was assessed using an ECL reagent (Fudebio Hangzhou, China) and
   Fluorescence Chemiluminescence Image Analyzer (ProteinSimple, USA),
   with β-actin as a reference protein. A comprehensive list of antibodies
   utilized is presented in Supplementary Table [162]S2. The original
   Western blot bands are provided in [163]Supplementary Materials.

Histological analysis

   The steatotic liver tissues from mice fed with HFD were fixed in OCT
   compound. The degree of steatosis was determined by Oil-Red-O staining.
   Ischemic liver lobes were processed with 4% paraformaldehyde for
   fixation, followed by dehydration and paraffin embedding. Sections
   (4 μm) from the paraffin blocks were subjected to staining with
   hematoxylin and eosin (H&E), immunofluorescence (IF), and terminal
   deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL).
   Additional methodological details are outlined in [164]Supplementary
   Information. Histopathological imaging was conducted using a light
   microscope (OLYMPUS, Japan). A list of antibodies utilized for staining
   is provided in Supplementary Table [165]S2.

In vitro hypoxia/reoxygenation (H/R) model

   The primary hepatocytes were separated as previously described
   method—two-step collagenase perfusion [[166]52]. The cells were
   cultured at serum-free DMEM medium (37 °C, 5% CO[2]) overnight. The
   steatosis hepatocytes were incubated with oleic acid/palmitic acid
   (OA/PA, 500 μM/250 μM, Kunchuang Biotechnology, Xian, China) for 2
   days. For H/R model, the cells were subjected to hypoxia (94% N[2], 1%
   O[2], and 5% CO[2]) in sugar-free, serum-free DMEM medium for 6 h. Then
   the cells transferred to normal air conditions in normal DMEM medium
   for 6 h.

Small interference RNA (siRNA) transfection

   BNL CL.2 cells (Pricella, Wuhan, China) by STR profiling were seeded in
   six-well plates at 2 × 10^6 cells/well. After reaching ~70% confluence,
   the cells were transfected with the indicated siRNA (Tsingke, Beijing,
   China) using transfection reagent (Yeason, Shanghai, China) for 48 h.
   The sequences of siInsig2 duplexes were F: 5′- CGGUGUUCGUGGGUAUAAA-3′
   and R: 5′- UUUAUACCCACGAACACCG-3′, siInsig1 duplexes were F: 5′-
   CACUCAGUUUCUUGUGUAU-3′ and R: 5′- AUACACAAGAAACUGAGUG-3′.

Adeno‐associated‐virus 8 (AAV8) vectors construction

   AAV8 vectors for Insig2 knockdown and overexpression were constructed
   by the Vigene Biosciences Co., Ltd (Shandong, China). The control group
   received an AAV8-thyroxine-binding globulin (TBG) vector that was
   devoid of any genetic material. The mice were injected through the tail
   vein with a viral solution carrying 3 × 10^11 copies of the AAV8 vector
   genomes, all in a 100 μL volume.

Transcriptome, proteome, and metabolome sequencing and analysis

   Liver samples from Insig2 KO and controlled WT mice on an HFD feeding
   that underwent I/R injury were collected for multi-omics analysis.
   Metware Biotechnology Co., Ltd (Wuhan, China) executed multi-omics
   sequencing, including sample preparation, extraction, and detection.
   Transcriptomic and proteomic analyses identified differentially
   expressed genes/proteins (DEGs/DEPs) with p value < 0.05 and fold
   change > 1.3. Metabolites were differentially expressed based on
   Variable Importance in Projection (VIP) score > 1 and p value < 0.05.
   Pathway enrichment analysis using the Kyoto Encyclopedia of Genes and
   Genomes (KEGG) and Gene Ontology (GO) was performed for the DEG/DEPs,
   with pathways having p value < 0.05 deemed statistically significant.

Detection of iron concentration and oxidative stress indicators

   The iron concentration in serum and liver tissue was quantified using
   the Iron Assay Kit (Solarbio, Beijing, China). The Malondialdehyde
   (MDA), Glutathione/glutathione disulfide (GSH/GSSG), NADPH/NADP^+ were
   measured using Assay Kit (Beyotime, Shanghai, China) according to
   manufacturer’s instructions.

Application of ferroptosis inhibitor and agonist

   Ferrostatin-1 (Fer-1, Selleck, S7243, USA) and RSL3 (Selleck, S8155,
   USA) were dissolved in DMSO and subsequently diluted with a vehicle
   solution composed of PEG300, Tween 80, and normal saline. These
   compounds were administered intravenously and intraperitoneally at a
   dosage of 10 mg/kg, 1 h prior to the initiation of I/R surgery,
   respectively. For the mouse treatment protocol, animals were randomly
   assigned to either the control or the drug treatment group.

Statistical analysis

   All experiments were performed a minimum of three times and
   reproducible results were obtained. Data were analyzed and visualized
   utilizing GraphPad Prism (v9.0), with summary statistics depicted as
   mean ± standard deviation. For pairwise group comparisons, a two-tailed
   Student’s t-test was applied; while one-way ANOVA with a post hoc test
   was utilized for assessing differences across multiple groups. Images
   that typify the average results for each group were carefully selected
   for presentation. P value < 0.05 was considered statistically
   significant.

Supplementary information

   [167]Supplementary materials^ (14.6MB, docx)
   [168]Original western blots^ (15.2MB, pdf)

Acknowledgements