ABSTRACT
-
Background/Aims
Dysregulated cholesterol metabolism is a hallmark of hepatocellular carcinoma (HCC) that drives tumor initiation and progression. However, clinical targeting of cholesterol metabolism has yielded limited benefits due to stringent feedback in tumor cells. Identifying a central mediator capable of restoring cholesterol homeostasis within the cell’s intrinsically fine-tuned regulatory framework is urgently needed.
-
Methods
We integrated a proteomic dataset from patients with cholesterol-dysregulated HCC into a global cholesterol metabolic regulatory network to identify potential therapeutic targets for disrupted cholesterol homeostasis. The prognostic significance of the candidate targets was further validated in an independent cohort through immunohistochemistry. Functional and mechanistic studies were conducted in vitro using HCC cell lines and in vivo using mouse models. The pharmacological efficacy of the candidate agent was evaluated in both subcutaneous and orthotopic HCC mouse models.
-
Results
ER lipid raft-associated 1 (ERLIN1), a pivotal regulator of cholesterol metabolism reprogramming, was identified as an independent favorable prognostic indicator in HCC. ERLIN1 constrains HCC progression both in vitro and in vivo by stabilizing the INSIG1–SCAP–SREBP2 axis and maintaining the metabolic balance of intracellular cholesterol. Under hypoxia, impaired factor-inhibiting hypoxia-1-dependent hydroxylation of ASB11 at asparagine residues 90 and 92 enhances ASB11-mediated ERLIN1 degradation. Pharmacological targeting of this axis using zoledronic acid (ZoA) attenuated HCC progression by weakening the ASB11–ERLIN1 interaction and restoring cholesterol homeostasis.
-
Conclusions
ERLIN1 represents a druggable metabolic vulnerability in cholesterol-dysregulated HCC. Targeting the ASB11–ERLIN1 axis with the clinically approved ZoA reestablishes cholesterol homeostasis and offers a promising therapeutic strategy to overcome the current limitations of cholesterol-targeted HCC therapies.
-
Keywords: Cholesterol metabolism reprogramming; Hypoxia; ERLIN1; Factor inhibiting hypoxia inducible factor
Study Highlights
• The downregulation of the cholesterol metabolism regulator ERLIN1 in HCC tissues is a negative prognostic factor in HCC.
• ERLIN1 downregulation promotes the development of HCC by reshaping cholesterol metabolism.
• Hypoxia regulates ERLIN1 degradation via the FIH–ASB11 axis.
• The clinically approved zoledronic acid restricts HCC growth by weakening the interaction between ASB11 and ERLIN1.
Graphical Abstract
INTRODUCTION
Hepatocellular carcinoma (HCC) is a leading cause of cancer mortality [
1]. Clinical guidelines emphasize that molecular subtyping can improve HCC management [
2]. Recent molecular profiling has revealed cholesterol-dysregulated HCC, marked by cholesterol accumulation and high SOAT1 expression, which is associated with aggressive disease and poor prognosis [
3,
4]. While tyrosine kinase inhibitors and immune checkpoint inhibitors have expanded treatment options, their efficacy in metabolic HCC remains limited, underscoring the necessity of targeting metabolic vulnerabilities in the tumor tissue [
5,
6].
Cholesterol metabolic dysregulation represents a critical hallmark of aggressive HCC, in which tumor cells accumulate cholesterol and cholesterol esters to facilitate membrane biogenesis and oncogenic signaling; this dysregulation is primarily driven by sterol regulatory element-binding protein 2 (SREBP2) hyperactivation [
7]. Normal hepatocytes regulate cholesterol homeostasis through the INSIG1–SCAP–SREBP2 regulatory axis, a feedback mechanism that involves monitoring intracellular cholesterol levels and decreasing SREBP2 activity when cholesterol stores are sufficient [
8]. Conversely, HCC cells circumvent this control. While statins targeting downstream HMGCR inhibit cholesterol biosynthesis and show some therapeutic potential for HCC in preclinical models [
9], they demonstrate limited clinical efficacy [
10,
11]. Statin-induced cholesterol depletion paradoxically triggers compensatory SREBP2 hyperactivation, thereby re-establishing pathogenic cholesterol overload. This dilemma underscores that not all cholesterol-related proteins are viable HCC targets; only those that bypass feedback reactivation and restore homeostasis through physiological regulation hold therapeutic potential. A critical question remains: Which known or candidate cholesterol metabolic regulator (e.g., HMGCR or NPC1) [
3,
12] drives HCC cholesterol dysregulation? Identifying the mediator that disrupts cholesterol homeostasis within the cell’s intrinsic finetuned machinery and deciphering its role in HCC-specific dysregulation are essential for treating cholesterol-metabolic disorder-associated HCC.
To address this, we integrated a global cholesterol metabolic network with proteomic data from our HCC cohort, who were stratified into SⅠ/SⅡ/SⅢ subtypes, with SⅢ indicating significant cholesterol reprogramming [
3]. By applying three strict selection criteria—enrichment in SⅢ/SOAT1-high tumors, tumor (T) to paratumor (P) differential expression, and survival association—eight molecules, including ER lipid raft-associated 1 (ERLIN1), were identified. ERLIN1 had the greatest prognostic value for overall survival (OS) and disease-free survival (DFS) and by being located upstream of the INSIG1-SCAP-SREBP2 axis, it can theoretically circumvent the feedback reactivation of downstream targets (e.g., HMGCR and NPC1). We subsequently investigated three hypotheses regarding the role of ERLIN1: its function as a master regulator of the cholesterol sensing threshold, its dysregulation in HCC, and potential novel interventions targeting its molecular axis for patients with cholesterol-dysregulated HCC.
Our findings establish ERLIN1 as an independent favorable prognostic marker that inhibits HCC progression by stabilizing the INSIG1–SCAP–SREBP2 axis. Hypoxia impairs factor-inhibiting hypoxia-1 (FIH)-dependent ASB11 hydroxylation, promoting ASB11-mediated ERLIN1 degradation. Notably, clinically approved zoledronic acid (ZoA) mitigates this dysfunction by weakening the ASB11–ERLIN1 interaction, restoring cholesterol homeostasis, and suppressing HCC progression.
MATERIALS AND METHODS
RESULTS
ERLIN1 functions as a clinically relevant tumor suppressor within the cholesterol regulatory network in HCC
To identify prognostically relevant nodes involved in cholesterol regulation (
Fig. 1A), we mapped the global cholesterol metabolic regulatory network (
Supplementary Table 1) onto our previously established HCC proteomic cohort [
3]. In this cohort, non-negative matrix factorization–based clustering revealed three molecular subtypes (SⅠ–SⅢ) with distinct prognoses, with SⅢ showing the worst survival and pronounced cholesterol metabolic reprogramming. Within this subtype (SⅢ vs. SⅠ & SⅡ, |fold change (FC)|>1.5,
P<0.05), we prioritized proteins that were enriched in SOAT1-high tumors (SOAT1 high vs. low, |FC|>1.5,
P<0.05; SOAT1, an acyl-CoA:cholesterol acyltransferase, catalyzes cholesterol esterification and serves as a marker of intracellular cholesterol accumulation [
13,
14]), were differentially expressed between T and P (|FC|>2,
P<0.05), and were significantly associated with OS and DFS (
P<0.05). This analysis identified eight candidates, including SOAT1 and NPC1. SOAT1 is a potential downstream therapeutic target in cholesterol-dysregulated HCC [
3], whereas NPC1 promotes HCC via cholesterol-independent mechanisms [
12]. ERLIN1 stood out for its upstream position in the SREBP2 axis and strong clinical associations (
Fig. 1B). Consistently, ERLIN1 expression was reduced in SOAT1-high tumors, SⅢ tumors, and in T regions relative to P regions, and had predictive value for both OS and DFS (
Fig. 1C,
1D).
We further validated the clinical relevance of ERLIN1 in an independent cohort, and confirmed that its expression was significantly decreased in HCC tissues. Moreover, ERLIN1 expression was significantly positively correlated with OS (
Fig. 1E).
To functionally characterize ERLIN1 in HCC, we established stable ERLIN1 overexpressing (OE) HCC cell lines featuring low ERLIN1 expression, such as MHCC-97H and Hep3B, and knockdown (KD) shERLIN1 HCC cell lines with high ERLIN1 expression, such as PLC and Huh7 (
Supplementary Fig. 1). ERLIN1 OE markedly inhibited proliferation and colony-forming ability
in vitro (
Supplementary Fig. 2A,
2B), whereas shERLIN1 exerted the opposite effect (
Fig. 2A,
2B). Furthermore, the xenograft model revealed that ERLIN1 OE significantly reduced tumorigenicity compared with empty vector (EV) controls, whereas shERLIN1 enhanced tumor growth relative to non-targeting controls (
Fig. 2C and
Supplementary Fig. 2C). In a β-catenin/c-Met–driven hydrodynamic model, ERLIN1 overexpression significantly suppressed tumor development and prolonged survival without altering the initiating oncogene expression. ERLIN1 OE also decreased the number of Ki-67-immunopositive cells, confirming reduced
in vivo proliferation (
Fig. 2D and
Supplementary Fig. 3).
ERLIN2, a member of the same protein family as ERLIN1 and likewise containing an stomatin/prohibitin/flotillin/HflK/C (SPFH) domain, can function either as a heterocomplex partner of ERLIN1 or independently [
15]. However, our analyses revealed that ERLIN2 was not involved in HCC progression. Clinically, ERLIN2 expression was reduced in tumor tissue compared with adjacent tissue, but it was not correlated with tumor malignancy in our cohort (
Supplementary Fig. 4A). While low ERLIN2 expression was nominally associated with poor OS and DFS (
Supplementary Fig. 4B), this trend was not validated in the independent IHC cohort, in which no significant associations with tumor features or patient prognosis were observed (
Supplementary Fig. 4C). Functionally, ERLIN2 OE in 97H cells with intrinsically lower ERLIN2 levels (
Supplementary Fig. 4D,
4E) did not affect cell proliferation or colony formation (
Supplementary Fig. 4F).
Previous studies have shown that silencing ERLINs leads to a significant increase in cellular cholesterol levels in other contexts [
16,
17]. The potential regulation of cholesterol metabolism by ERLIN1 specifically in HCC was investigated. Multiomics analyses of HCC cell lines and xenograft tumors revealed that ERLIN1 OE suppresses cholesterol synthesis and uptake programs, reducing intracellular cholesterol and cholesteryl ester accumulation, whereas ERLIN1 KD exerts opposite effects (
Fig. 3A,
Supplementary Figs. 5,
6, and
Supplementary Tables 2–
4). Filipin staining further revealed reduced intracellular cholesterol levels upon ERLIN1 OE (
Fig. 3B), whereas ALOD4 staining showed increased intracellular cholesterol accumulation following ERLIN1 KD (
Fig. 3C). Notably, unlike HMGCR or LDLR KD, ERLIN1 OE did not induce compensatory HMGCR or LDLR upregulation, thereby bypassing feedback regulation to reduce intracellular cholesterol levels in HCC cells (
Supplementary Fig. 7). These results suggest an inhibitory role of ERLIN1 in cholesterol metabolism within HCC.
Next, we investigated whether the impact of ERLIN1 on the HCC phenotype is mediated by intracellular cholesterol levels. While recent studies suggest that key cholesterol regulators may promote HCC via cholesterol-independent mechanisms [
12,
18], our data indicate a dominant cholesteroldependent effect. Cholesterol supplementation (CHO/MCD, in which cholesterol is incorporated into the MCD to increase cellular uptake) partially reversed the inhibition of proliferation (~46%, on day 4,
P<0.001) and colony formation (~72%,
P<0.01) in ERLIN1 OE 97H cells. Conversely, cholesterol depletion using HPBCD partially attenuated the growth-promoting effects of ERLIN1 KD, reducing proliferation (~56%, on day 4,
P<0.01) and colony formation (~47%;
P<0.05) (
Fig. 3D,
3E). Consistently, SREBP2 KD phenocopied ERLIN1 OE (
Supplementary Fig. 8). These findings suggest that ERLIN1 regulates HCC growth primarily through controlling intracellular cholesterol levels.
Having established the role of ERLIN1 in cholesterol homeostasis, we investigated its regulation of the INSIG1–SCAP–SREBP2 axis, a central cholesterol-sensing mechanism [
19]. In cholesterol-depleted HCC cells, ERLIN1 OE reduced mature SREBP2 protein levels, and this effect was enhanced by 25-hydroxycholesterol (25-HC) which traps SREBP2 in the ER (
Fig. 4A). Immunofluorescence staining confirmed the impaired nuclear translocation of SREBP2 in ERLIN1 OE Hep3B cells (
Fig. 4B). Consistently, ERLIN1 OE downregulated both the mRNA and protein levels of SREBP2 targets (HMGCR, LDLR, etc.), whereas KD had the opposite effects (
Fig. 4C and
Supplementary Fig. 9). We further investigated whether INSIG1 mediates the regulation of SREBP2 by ERLIN1, and we first confirmed that ERLIN1 stabilizes INSIG1 in HCC cells (
Fig. 4D), extending previous findings in 293T cells [
16]. Rescue experiments revealed the critical function of INSIG1, as its overexpression reversed SREBP2 hyperactivation in ERLIN1 KD cells (
Fig. 4E). Clinical correlation analysis supported these mechanistic insights, demonstrating that reduced INSIG1 expression predicts poor survival in patients with HCC (
Supplementary Fig. 10).
Building on reports that INSIG1 levels modulate cholesterol sensing in Chinese Hamster Ovary cells [
8], we examined whether ERLIN1 affects this threshold in HCC cells (
Fig. 4F). Cholesterol depletion (1% HPBCD) triggered SREBP2 cleavage (from blot Lane 1 to Lane 2), while subsequent cholesterol repletion gradually suppressed this activation (Lanes 3–6). The levels of nuclear SREBP2 decreased in ERLIN1 OE HCC cells at lower cholesterol concentrations (20 μM cholesterol/MCD) compared to those in EV controls (80 μM). Conversely, ERLIN1 KD cells required higher cholesterol concentrations (80 μM) to decrease nuclear SREBP2 levels compared with scramble controls (20 μM). Overall, ERLIN1 downregulation decreases the cholesterol sensing threshold, enabling HCC cells to sustain protumorigenic SREBP2 activity despite cholesterol overload.
Deciphering the mechanisms underlying ERLIN1 downregulation in HCC is critical for the development of targeted therapies. Reexamination of our in-house data revealed that tumor malignancy correlated specifically with ERLIN1 protein expression rather than with ERLIN1 mRNA expression (
Supplementary Fig. 11 and
Supplementary Fig. 12E), suggesting that ERLIN1 is post-translationally regulated in HCC. A recent study revealed that posttranslational modification modulates the levels of ERLIN1 [
17]. Consequently, we focused on the ubiquitin proteasome system, given its essential role in protein quality control.
Using the BioGRID database, we screened for ERLIN1-specific ubiquitin ligases and deubiquitinating enzymes, excluding ERLIN2-associated factors, which yielded seven enzymes (
Fig. 5A). Among these, ASB11 is hydroxylated by the oxygen sensor FIH [
20]. Given the hypoxic tumor microenvironment of HCC and the strong prognostic implications of high hypoxia levels in our cohort (
Supplementary Fig. 12C), we hypothesized that ASB11 mediates ERLIN1 ubiquitination in the physiological context of HCC. ASB11, an ankyrin repeat and SOCS-box containing adaptor protein of the cullin-RING E3 ligase family, forms complexes comprising a cullin scaffold, a RING finger protein, a substrate-recognition subunit, and an adaptor (
Supplementary Fig. 13) [
21]. Co-IP experiments demonstrated the selective binding of ASB11 to ERLIN1, with no detectable interaction with ERLIN2, which was subsequently confirmed at the endogenous level (
Fig. 5B). Domain mapping revealed that the fifth ankyrin repeat of ASB11 and the SPFH domain of ERLIN1 were essential for their interaction, which was further confirmed by GST pull-down assays using purified proteins (
Fig. 5C,
Supplementary Fig. 14).
We next explored the regulatory role of ASB11 as an E3 ubiquitin ligase in controlling ERLIN1 stability. ASB11 OE significantly reduced the half-life of ERLIN1 (
Fig. 5D). ASB11-induced ERLIN1 degradation was reversed upon treatment with the proteasome inhibitor MG-132 (
Fig. 5E). Ubiquitination experiments revealed that ASB11 catalyzes the K48-linked ubiquitination of ERLIN1, a modification that typically targets proteins for proteasomal degradation. Critically, deletion of the SOCS box domain—which mediates cullin-RING complex assembly—abrogated this ubiquitination activity (
Fig. 5F). These findings collectively establish ASB11 as an E3 ligase that mediates the ubiquitin-proteasomal degradation of ERLIN1.
Given the susceptibility of ERLIN1 to ubiquitin-mediated degradation, we first examined its ubiquitination status and degradation rate under hypoxic conditions. Under CoCl
2-induced chemical hypoxia, the ERLIN1 half-life was significantly decreased (
Fig. 6A). Both CoCl
2-induced chemical hypoxia and 1% O
2-mediated physiological hypoxia significantly decreased ERLIN1 protein levels without altering its mRNA expression (
Fig. 6B and
Supplementary Fig. 15). This decrease was reversed by the proteasome inhibitor MG-132 (
Fig. 6B), and ubiquitination assays confirmed that hypoxia promotes K48-linked proteasomal degradation of ERLIN1 (
Fig. 6C). Critically, pharmacological inhibition of HIF1 with HIF-IN-1 did not rescue ERLIN1 levels, indicating the HIF1-independence of this regulatory mechanism (
Supplementary Fig. 16). To assess the clinical relevance of this mechanism, we performed proteomic profiling of hypoxic regions in formalin-fixed paraffin-embedded HCC specimens, using CA9 as a hypoxia marker (
Supplementary Fig. 12A,
12B) [
22]. ERLIN1 was the sole cholesterol-regu-latory candidate, whose expression was prognostically relevant and downregulated in tumors, with further suppression in CA9-positive hypoxic regions (
Supplementary Fig. 12B). In our in-house cohort, ERLIN1 expression decreased in parallel with the malignant hypoxia gradient of HCC (
Supplementary Fig. 12D,
12E) and strongly inversely correlated with the hypoxia score (
Supplementary Fig. 12F), reinforcing the clinical significance of hypoxia-driven ERLIN1 loss. These results indicate that hypoxia promotes the proteasomal degradation of ERLIN1.
ASB11 is a hypoxia-sensitive protein that is hydroxylated by the oxygen sensor FIH, and we investigated whether or how hypoxia modulates ERLIN1 degradation through ASB11. Hypoxia significantly enhanced the ASB11–ERLIN1 interaction (
Fig. 6C) and promoted ERLIN1 degradation in an ASB11-dependent manner, as evidenced by the restoration of ERLIN1 levels upon ASB11 silencing (
Fig. 6D). These results suggest that ASB11 plays an essential role in hypoxia-induced ERLIN1 degradation.
FIH, a key oxygen sensor, hydroxylates specific asparagine residues in ‘VNVN’ and similar motifs within ankyrin repeat domains under oxygen-sufficient conditions [
20]. In ASB11, FIH hydroxylates asparagine residues at positions 90 and 92. Hypoxia markedly reduced FIH-mediated hydroxylation of ASB11, with N92 hydroxylation becoming undetectable relative to normoxia (
Supplementary Fig. 17). We investigated whether hypoxic conditions destabilize ERLIN1 by impairing FIH-mediated ASB11 hydroxylation. The overexpression of FIH under normoxic conditions rescued the reduction in ERLIN1 protein levels caused by ASB11 OE, whereas FIH silencing reduced ERLIN1 protein levels in an ASB11-dependent manner, as simultaneous ASB11 KD prevented this reduction (
Fig. 6D). To elucidate the role of posttranslational modification, we generated an ASB11 N90A/N92A mutant that mimics constitutive dehydroxylation. This mutant displayed increased ERLIN1 binding affinity under normoxia conditions and exhibited hypoxia-independent interaction patterns (
Fig. 6E). Functionally, the ERLIN1 degradation activity of the mutant significantly increased under normoxic conditions but did not further increase under hypoxic conditions (
Fig. 6F). These findings provide direct evidence that FIH-mediated hydroxylation at N90/N92 serves as a critical regulatory switch, attenuating the E3 ligase activity of ASB11 toward ERLIN1 under normoxia, whereas hypoxia relieves this inhibition through hydroxylation abrogation.
We further analyzed the relevance of ASB11 and ERLIN1 expression patterns in human HCC tissue samples. IHC CA9 staining confirmed pronounced hypoxia in tumor tissues compared with adjacent normal tissues (
Supplementary Fig. 18A). Strikingly, ERLIN1 and ASB11 expression was inversely correlated specifically in hypoxic tumor regions, but not in normoxic tissues (
Supplementary Fig. 18B,
18C). These clinical findings substantiate the hypoxiadependent regulation of ERLIN1 by ASB11 in HCC. Collectively, our results establish a hypoxia-sensitive regulatory axis in HCC wherein hypoxia reduces FIH-mediated hydroxylation of ASB11, consequently enhancing its E3 li-gase-mediated degradation of ERLIN1 (
Supplementary Fig. 19).
Initially reported in 1994 [
23], ZoA is used to treat osteoporosis, malignancy-associated hypercalcemia, multiple myeloma, bone metastases from solid tumors, and Paget’s disease. ZoA has been reported to increase ERLIN1 mRNA expression [
24]. Strikingly, under hypoxic conditions, ZoA dose-dependently increased ERLIN1 protein levels, even in the presence of CHX (
Fig. 7A), suggesting posttranslational, transcription-independent regulation.
In silico CB-DOCK analysis revealed that ZoA directly binds to a pocket formed by the SPFH domain of ERLIN1 and the fifth repeat ankyrin domain of ASB11, providing structural insight into its inhibitory effect on their interaction (
Supplementary Fig. 20). Consistently, ZoA impaired the ASB11-ERLIN1 interaction (
Fig. 7B), indicating that ZoA stabilizes ERLIN1 protein levels by attenuating ASB11-mediated ubiquitination.
Next, we assessed the anticancer efficacy of ZoA in HCC. Under
in vitro hypoxic conditions, ZoA dose-dependently reduced the viability of HCC cells. Notably, ERLIN1 KD decreased the drug sensitivity of ZoA, as evidenced by an increase in the IC
50 value (from 3.444 to 16.23 μM) (
Fig. 7C), indicating that ERLIN1 is an essential mediator of the therapeutic effect of ZoA. Consistently, depletion of intracellular cholesterol using HPBCD abrogated the inhibitory effect of ZoA on HCC cell viability (
Supplementary Fig. 21), further suggesting that ZoA exerts its antitumor activity in a cholesterol-dependent manner.
In vivo xenograft experiments demonstrated that ZoA, administered intravenously at 100 μg/kg twice weekly (equivalent to a clinical dosage of 4 mg every 3–4 weeks), significantly suppressed tumor growth without affecting animal body weight (
Fig. 7E and
Supplementary Fig. 22A). Concomitantly, ZoA treatment markedly upregulated ERLIN1 expression and downregulated SREBP2, whereas it had minimal effects on the expression of ASB11 and CA9 (
Supplementary Fig. 23). The therapeutic effect was ERLIN1-dependent, as ZoA lost efficacy in the shERLIN1 KD model (
Fig. 7E). Additionally, ZoA alleviated the tumor burden in the Hepa1-6-induced orthotopic HCC model in an ERLIN1-dependent manner, without affecting body weight (
Fig. 7F,
Supplementary Fig. 22B and
Supplementary Fig. 24). Both
in vivo and
in vitro results substantiate the potent anti-HCC efficacy of ZoA.
We concurrently examined the effects of ZoA on cholesterol levels in both cell lines and subcutaneous tumor tissues (
Fig. 7D,
7E and
Supplementary Table 5). ALOD4 staining and metabolomic profiling revealed that ZoA decreased cholesterol levels in the control group, but failed to affect these levels in the shERLIN1 group. These results indicate that ZoA regulates cholesterol metabolism in HCC in an ERLIN1-dependent manner.
Moreover, the FPPS inhibitor TH-Z93 failed to exert comparable antitumor effects
in vitro or
in vivo (
Supplementary Figs. 25,
26), suggesting that ZoA acts through ERLIN1-mediated cholesterol regulation rather than through classical FPPS inhibition.
DISCUSSION
Stringent intracellular feedback checkpoints have thwarted attempts to target cholesterol metabolism in HCC, requiring the identification of therapeutically actionable regulatory nodes that can bypass these barriers and unlock therapeutic opportunities. Here, we report that ERLIN1 safeguards cholesterol homeostasis. Its downregulation predicts poor prognosis, promotes tumor progression by destabilizing the INSIG–SCAP–SREBP2 axis, and is driven by hypoxia through the FIH–ASB11 ubiquitination cascade. Importantly, pharmacologic weakening of the ASB11–ERLIN1 interaction with the clinically approved bisphosphonate ZoA restored cholesterol homeostasis and suppressed HCC growth, unveiling a clinically actionable vulnerability. Moreover, targeting the ASB11–FIH–ERLIN1 axis is a metabolically focused strategy that is distinct from current employed immunotherapies and TKIs and may help overcome cholesterol-driven immune suppression and TKI resistance in HCC.
Tumor cells maintain elevated cholesterol levels to support membrane biogenesis and oncogenic signaling, as evidenced by the increased lipid droplet number and the upregulation of SOAT1, a key enzyme for cholesteryl ester biosynthesis [
3]. While normal cells strictly regulate cholesterol through negative feedback mechanisms [
19], this homeostatic control persists but becomes desensitized in tumor cells, as evidenced by cross-species studies [
25-
27]. Therefore, understanding the desensitization mechanism of cholesterol regulation in tumor cells is crucial. The negative feedback regulation of intracellular cholesterol metabolism is coordinated by the key transcription factors LXR and SREBP2 [
19]. While proliferative stimuli may disrupt oxysterol homeostasis through LXR inhibition to elevate intracellular cholesterol levels [
28], our findings reveal impaired upstream regulation of SREBP2 as an alternative mechanism. In hypoxic HCC models ERLIN1, which functions upstream of the INSIG1–SCAP–SREBP2 axis, governs the activation threshold of intracellular cholesterol metabolism, thereby steering tumorigenic processes. This hypoxia-sensitive regulatory paradigm has gained a broader context from recent discoveries; for example, the hypoxia-responsive circINSIG1 promotes INSIG1 ubiquitination via the CUL5-ASB6 E3 complex, driving cholesterol overproduction in colorectal cancer [
29]. Employing opposing regulatory strategies (ERLIN1 stabilization versus circINSIG1 degradation) revealed that both mechanisms converge on INSIG1 as the central integrator of hypoxic cholesterol reprogramming. The pathophysiological relevance of this axis is also substantiated by SREBP2’s role in oncogenic cholesterol accumulation, as evidenced by the dominant negative
Srebp2-mediated blockade of hepatocarcinogenesis in
Fasn knockout mice [
30]. This finding is complemented by the established cholesterol-modulating capacity of ERLIN family proteins [
16,
17,
31], together with prior findings that INSIG1 regulates the activation threshold of SREBP2 in Chinese Hamster Ovary cells [
8]. Consistently, in the present study, we demonstrated that in the context of HCC, ERLIN1 stabilizes INSIG1 and further decreases the activation threshold of SREBP2.
Hypoxia, a well-established hallmark of solid tumors, critically drives tumor metabolic reprogramming and progression [
32]. While the classical hypoxia response is primarily mediated through oxygen-dependent HIF activation via impaired hydroxylation by PHDs and FIH [
33], emerging evidence has substantially expanded our understanding of hypoxia signaling beyond the HIF-centric paradigm. PHDs and FIH can exert extensive regulatory control over non-HIF substrates: PHDs regulate NF-κB and p53 under hypoxic conditions [
34], whereas FIH mediates the hydroxylation of multiple targets including OTUB1, p105, IκBα, and RIPK4 at specific asparagine residues, thereby influencing diverse cellular processes such as tumor metabolism and cell cycle progression [
34,
35]. While a recent study revealed FIH-mediated hydroxylation of ASB11 ankyrin repeat domains at “VNVN” motifs [
20], the functional consequences remain unclear. Our work revealed that ASB11 serves as an adaptor protein in the cullin-RING E3 ubiquitin ligase complex, specifically targeting ERLIN1 for proteasomal degradation. Notably, we found that hypoxia reduces FIH-mediated hydroxylation of ASB11 at the N90/N92 residue, which in turn increases the ASB11–ERLIN1 interaction and promotes ERLIN1 ubiquitination. This discovery uncovers a new oxygen-sensitive regulatory mechanism through which FIH controls protein stability through ASB11-mediated ubiquitination. Further mechanistic studies are needed to elucidate how ASB11 hydroxylation modulates its interaction with ERLIN1.
The regulatory effect of ZoA on ERLIN1 expression has drawn our attention [
24]. Our study revealed that ZoA modulates ERLIN1 protein levels by competitively inhibiting ASB11 binding, primarily through hypoxia-dependent interactions with the SPFH domain of ERLIN1. As a clinically approved osteoporosis therapeutic, ZoA has great therapeutic potential in cancer, with emerging evidence indicating its efficacy in suppressing tumor growth and nonalcoholic steatohepatitis progression through various mechanisms [
36,
37]. Previous studies have validated the anti-HCC activity of ZoA
in vitro [
36], whereas our study further confirms its therapeutic impact in hypoxic HCC models
in vivo. Mechanistically, ZoA exerts dual effects by both inhibiting FPPS (thereby disrupting small GTPase signaling via RhoA suppression) and increasing ERLIN1 expression to maintain intracellular cholesterol homeostasis. These insights expand the potential clinical utility of ZoA, as evidenced by ongoing trials evaluating ZoAchemotherapy (NCT05866172) or ZoA-sorafenib (NCT01259193) combinations in advanced HCC. With respect to the success of p53-MDM2 disruptors, future efforts should prioritize developing ASB11–ERLIN1 interaction inhibitors (molecular compounds or peptides) to augment therapeutic precision.
FOOTNOTES
-
Authors’ contributions
Y.Z., C.T., and A.S. conceived and designed the project; Y.Z., C.T., A.S., and Y.H. interpreted and analyzed the data, Y.H. and K.X. curated the bioinformatics analysis; X.W. collected the clinical samples; Y.Z., P.J., S.W., H.K., H.Z., J.J., X.H., Z.L., S.Y., and J.L. performed the experiments; Y.Z., L.Z., C.T., and A.S. wrote and revised the original draft; and F.H., C.T., and A.S. supervised the project.
-
Acknowledgements
Funding was provided by the National Key R&D Program of China (2024YFA1307704, 2024YFA1307603), the National Natural Science Foundation of China (82372835), the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (2019-I2M-5-063), and the Specific Research Fund for TCM Science and Technology of Guangdong Provincial Hospital of Chinese Medicine (YN2022DB04).
-
Conflicts of Interest
The authors have no conflicts to disclose.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
Establishment of ERLIN1 overexpression and knockdown stable cell lines. (A) Endogenous ERLIN1 expression in HCC cell lines. (B, C) Validation ERLIN1 expression in ERLIN1 OE (B) or shERLIN1 cells (C).
Supplementary Figure 2.
ERLIN1 overexpression suppresses HCC progression in vitro and in vivo. (A) Proliferation of the indicated cells cultured in DMEM containing 5% LPDS for 4 days (n=3, mean±SD, two-way ANOVA). (B) Colony formation capacity of the indicated cells was assessed (n=3, mean±SD, unpaired student’s t-test). (C) Tumor growth analysis in 97H xenograft model(n=6–8). The schematic of the xenograft model in NOD-SCID mice, representative tumors, tumor volume curves (mean±SEM, two-way ANOVA) and final tumor weight (mean±SEM, unpaired student’s t-test) are shown. *P<0.05, **P<0.01, ***P<0.001.
Supplementary Figure 3.
Expression of Ki-67, c-Met, β-catenin and ERLIN1 in HTVi model. (A) Representative images of IHC staining. (B, C) Expression levels of the indicated molecules were analyzed by Western blot (n=3) (B) and qPCR (n=3) (C). Mean±SD, unpaired student’s t-test. **P<0.01, ***P<0.001, NS>0.05.
Supplementary Figure 4.
ERLIN2 demonstrates limited functional and clinical relevance in HCC. (A) Comparative analysis of ERLIN2 expression in HCC, stratified by P and T groups and molecular subtypes (unpaired student’s t-test). (B) Kaplan–Meier analysis of OS and DFS by ERLIN2 expression levels in our HCC cohort (log-rank test). (C) Clinical and prognostic significance of ERLIN2 in an independent HCC cohort. Representative cases, scale bar, 100 μm, distribution of cases by staining intensity (χ2 test) quantitative IHC scores (mean±SEM, unpaired student’s t-test), alteration of expression in paired samples and Kaplan–Meier survival curves of OS (log-rank test). (D) Endogenous ERLIN2 expression in HCC cell lines. (E) Validation ERLIN2 expression in ERLIN2 OE cells. (F) Proliferation and colony formation of indicated cells cultured in DMEM containing 5% LPDS for 3 days (n=3, mean±SD, two-way ANOVA or unpaired student’s t-test). ***P<0.001, NS>0.05.
Supplementary Figure 5.
ERLIN1 Overexpression enhances cholesterol metabolism in xenograft HCC tumor. (A) Tumor volume and weight of 97H ERLIN1 OE xenografts (n=7), shown as mean±SEM; *P<0.05 (Student’s t-test). (B–D) Transcriptomic (B), proteomic (C), and metabolomic (D) profiles of cholesterol metabolism in xenograft tumors.
Supplementary Figure 6.
ERLIN1 knockdown enhances cholesterol metabolism in HCC cells. (A–C) Transcriptomic (A), proteomic (B), and metabolomic (C) profiles of cholesterol metabolism in shERLIN1 versus shCtrl Huh7 cells under 5% LPDS for 24 hours.
Supplementary Figure 7.
Overexpression of ERLIN1 reduces intracellular cholesterol levels in HCC cells. (A) Representative blots of siHMGCR, siLDLR, and ERLIN1 overexpression in 97H cells. (B) Cholesterol levels in 97H cells were measured using ALOD4 labeling after culturing in DMEM with 20% FBS for 48 hours; 1% HPBCD treatment for 10 min served as a positive control (n=3, mean±SD, unpaired Student’s t-test). *P<0.05, **P<0.01, NS>0.05.
Supplementary Figure 8.
SREBP2 knockdown inhibits HCC cell proliferation. (A) Western blot validation of SREBP2 knockdown in 97H and Hep3B cell lines. (B) Cell proliferation of the cells shown in (A) was assessed by CCK-8 assay (n=3, mean±SD; two-way ANOVA). ***P<0.001.
Supplementary Figure 9.
ERLIN1 downregulates the expression of SREBP2 target genes in HCC cell lines. (A, B) qPCR analysis of expression of SREBP2 target genes in the indicated cells (depletion medium, 24 hours) (n=3). Data was shown as mean±SD. *P<0.05, **P<0.01, ***P<0.001, NS>0.05. Unpaired Student’s t-test.
Supplementary Figure 10.
Clinical relevance of INSIG1 expression in HCC. Clinical and prognostic significance of INSIG1 in an independent HCC cohort. Representative cases, scale bar, 100 μm, distribution of cases by staining intensity (χ2 test), quantitative IHC scores (mean±SEM, unpaired student’s t-test), alteration of expression in paired samples and Kaplan–Meier survival curves of OS (log-rank test). *P<0.05, ***P<0.001.
Supplementary Figure 12.
Low ERLIN1 expression positively correlates with hypoxia in HCC tissues. (A) Representative H&E and CA9/ERLIN1 IHC staining in HCC FFPE specimens. Dashed lines indicate hypoxic regions selected for proteomic analysis. Scale bar, 5 mm. (B) Workflow for candidate selection: hypoxia-associated proteins identified by FFPE proteomics were cross-analyzed with tumor-associated proteins and OS/DFS-correlated targets from our in-house HCC cohort. (C) Prognostic survival curves for OS based on stratification by hypoxia scores in our HCC cohort (Log-rank test). (D, E) Hypoxia score (D) and ERLIN1 expression (E) across molecular subtypes in our HCC cohort (mean±SEM; one-way ANOVA; ***P<0.001, NS>0.05 versus the control group). (F) Spearman correlation between hypoxia score and ERLIN1 expression in our HCC cohort.
Supplementary Figure 13.
Schematic diagram of ASB11-containing E3 complex. The complex comprises Cullin (scaffold), RING finger protein (E2 binding), and ASB11 (substrate recognition via ankyrin repeats).
Supplementary Figure 14.
Direct interaction between ASB11-ANK5 and ERLIN1-SPFH. GST pull-down assay showing the interaction between His-tagged SPFH and GST-ANK5-Flag; SPFH, stomatin/prohibitin/flotillin/HflK/C domain.
Supplementary Figure 15.
Hypoxia does not alter the mRNA level of ERLIN1. qPCR analysis of the mRNA expression of ERLIN1 and PGK1 (HIF1 target as positive control) in Huh7 cells treated with 200 μM CoCl -induced or 1% O2 culture-induced hypoxia condition for 12 hours (n=3). Data was shown as mean±SD. **P<0.01, NS>0.05. Unpaired Student’s t-test.
Supplementary Figure 16.
Hypoxia downregulates ERLIN1 protein level independent of HIF1 activation. Representative blots of ERLIN1 protein abundance in Huh7 cells after 24 hours hypoxia (1% O2) with or without 30 μM HIF-IN-1 treatment.
Supplementary Figure 17.
Identification of ASB11 hydroxylation by IP–MS. (A) Workflow for IP–MS–based identification of ASB11 hydroxylation. (B) LC–MS/MS spectra of the hydroxylated ASB11 peptide at N92 under normoxic and hypoxic conditions.
Supplementary Figure 18.
Correlation analysis of ERLIN1 and ASB11 IHC scores from the IHC HCC cohort. (A) Comparative IHC score of CA9 in P and T from an independent HCC IHC cohorts (n=74). Data are shown as mean±SEM. ***P<0.001. Unpaired Student’s t-test. (B) The Pearson correlation of the IHC score of ERLN1 and ASB11 in paracancerous liver tissues from the IHC HCC cohort (n=70). (C) Correlation analysis of IHC scores for ERLIN1 and ASB11 using the Pearson correlation test in tumor tissues stratified by CA9 expression: low-CA9 group (CA9 IHC score≤2, n=40) and high-CA9 group (CA9 IHC score>2, n=30).
Supplementary Figure 19.
Illustration of the hypoxia-regulated mechanism of ERLIN1. Schematic representation of hypoxia-induced FIH inactivation leading to ASB11-mediated ERLIN1 degradation in HCC.
Supplementary Figure 20.
ZoA binding to the ERLIN1-ASB11 complex. Molecular docking analysis of ZoA binding to the ERLIN1-ASB11 complex using CB-DOCK, with complex structure simulation performed by GRAMM web-serve.
Supplementary Figure 21.
ZoA inhibits HCC cell viability in a cholesterol-dependent manner. Cell viability of Huh7 cells treated with 10 µM ZoA in the presence or absence of 30 µM HPβCD under 1% O2 conditions in medium containing 2% FBS for 72 hours (n=3). Mean±SD, unpaired Student’s t-test. **P<0.01, ***P<0.001, NS>0.05.
Supplementary Figure 23.
Expression of SREBP2, CA9, ASB11 and ERLIN1 in response to ZoA treatment in xenograft tumors. (A) Representative images of IHC staining. (B, C) Expression levels of the indicated molecules were analyzed by Western blot (n=3) (B) and qPCR (n=3) (C). Mean±SD, unpaired Student’s t-test. **P<0.01, NS>0.05.
Supplementary Figure 24.
ZoA inhibits orthotopic HCC progression in vivo in an ERLIN1-dependent manner. (A) Western blot validation of ERLIN1 knockdown in stable shERLIN1 Hepa1-6 cell lines. (B) Schematic diagram of the animal study design. (C, D) Tumor growth analysis in the Hepa1-6 orthotopic liver cancer model (n=6). Representative liver and tumor images are shown in (C). Liver weight, final tumor volume, and body weight are summarized in (D) (mean±SEM; unpaired Student’s t-test). *P<0.05, NS>0.05.
Supplementary Figure 25.
The FPPS inhibitor TH-Z93 does not affect HCC cell viability under hypoxic conditions. (A) Western blot analysis of SQLE protein expression following TH-Z93 treatment. SQLE upregulation served as a positive control to confirm effective TH-Z93 treatment. (B) Effect of TH-Z93 on Huh7 cell viability under 1% O2 in 2% FBS-containing medium. Cell viability curves were fitted using GraphPad Prism (n=3).
Supplementary Figure 26.
ZoA, but not Lovastatin or TH-Z93, inhibits HCC progression in vivo. (A) Schematic diagram of the animal study design. (B, C) Tumor growth analysis in 97H xenograft model (n=7). Representative tumors (B), tumor volume curves (mean±SEM, two-way ANOVA), and final tumor weight, as well as animal weight (mean±SEM, unpaired Student’s) are shown (C). *P<0.05, NS>0.05.
Figure 1.Downregulation of the cholesterol-regulatory node ERLIN1 predicts poor prognosis in cholesterol-dysregulated HCC. (A) Funnel diagram depicting the identification of prognostically relevant cholesterol-regulatory proteins through integration of HCC proteomic data with the global cholesterol metabolism network. (B) Protein expression profiles of cholesterol-regulatory proteins across molecular subtypes, showing tumor (T) versus para-tumor (P) expression ratios and associated prognostic hazard ratios (HRs). (C) ERLIN1 expression was compared between SOAT1-high and SOAT1-low groups, across molecular subtypes, and between tumor (T) and para-tumor (P) tissues (unpaired Student’s t-test). (D) Kaplan–Meier analysis of overall survival (OS) and disease-free survival (DFS) stratified by ERLIN1 expression (log-rank test). (E) Clinical and prognostic significance of ERLIN1 in an independent HCC cohort. Representative cases, scale bar, 100 μm, distribution of cases by staining intensity (χ2 test), quantitative IHC scores (unpaired Student’s t-test), alteration of expression in paired samples and Kaplan–Meier survival curves of OS (log-rank test). Data shown in this figure are presented as the mean±standard error of the mean. DFS, disease-free survival; ERLIN1, ER lipid raft-associated 1; HCC, hepatocellular carcinoma; HR, hazard ratio; IHC, immunohistochemistry; OS, overall survival. *P<0.05, **P<0.01, ***P<0.001.
Figure 2.ERLIN1 inhibits HCC progression both in vitro and in vivo. (A) Indicated cell proliferation in DMEM containing 5% LPDS for 4 days (n=3, mean±standard deviation, two-way ANOVA). (B) Colony formation of the indicated cells (n=3, mean±standard deviation, unpaired Student’s t-test). (C) Huh7 xenograft tumor growth in NOD-SCID mice (n=10). The schematic, representative tumors, tumor volume curves (mean±standard error of the mean, two-way ANOVA) and final tumor weight (mean±standard error of the mean, Mann–Whitney test) are shown. (D) Autologous tumor model in C57BL/6J mice (n=7–8). The schematic, representative mice/tumors, tumor nodes (>1.5 mm diameters), MAX diameters, liver/body weight ratio (mean±standard error of the mean, unpaired Student’s t-test), and survival curve (log-rank test) are shown. ERLIN1, ER lipid raft-associated 1; EV, empty vector; HCC, hepatocellular carcinoma. **P<0.01, ***P<0.001.
Figure 3.ERLIN1-mediated inhibition of HCC progression depends on intracellular cholesterol regulation. (A) Transcriptomic, proteomic, and metabolomic profiles of cholesterol metabolism in ERLIN1 OE versus EV 97H cells under 5% LPDS for 24 hours. (B) Filipin staining of cholesterol in EV versus ERLIN1 OE Hep3B cells under serum-free conditions for 24 hours (n=3, unpaired Student’s t-test). (C) ALOD4 staining of cholesterol in shCtrl versus shERLIN1 Huh7 cells under serum-free conditions for 24 hours (n=3, unpaired Student’s t-test). (D) Proliferation of ERLIN1 OE 97H cells treated with 20 μM cholesterol complex or shERLIN1 Huh7 cells treated with 30 μM HPBCD (5% LPDS, n=3, two-way ANOVA). The right panel shows relative proliferation on day 4, normalized to the respective controls (n=3, unpaired Student’s t-test). (E) Colony formation of ERLIN1 OE 97H cells treated with cholesterol complex or shERLIN1 Huh7 cells treated with HPBCD (n=3, unpaired Student’s t-test). The right panel shows relative colony numbers normalized to the respective controls (n=3, unpaired Student’s t-test). Data shown in this figure are presented as the mean±standard deviation. ERLIN1, ER lipid raft-associated 1; EV, empty vector; HCC, hepatocellular carcinoma. *P<0.05, **P<0.01, ***P<0.001, NS>0.05.
Figure 4.ERLIN1 deficiency elevates cholesterol sensing threshold via INSIG1-SCAP-SREBP2 axis. (A) Immunoblot analysis of nuclear (N) SREBP2 levels in cells cultured under complete medium (10% FBS), cholesterol-depleted medium (5% LPDS plus 1 μM lovastatin/50 μM mevalonate), or depletion medium plus 20 μM 25-HC for 24 hours. (B) Immunofluorescence of SREBP2 nuclear localization in ERLIN1 OE versus EV Hep3B cells (depletion medium, 24 hours). Nuclei were stained with Hoechst. Scale bar: 25 μm. The ratio of nuclear to cytoplasmic signal intensity per cell was quantified and is shown in the right panel (n=20, mean±standard deviation; unpaired Student’s t-test, ***P<0.001). (C) Immunoblot of SREBP2 target proteins and SOAT1 expression levels in ERLIN1 OE or ERLIN1 KD cells (depletion medium, 24 hours). (D) Half-life of INSIG1 was determined by a CHX-chase assay. (E) Immunoblot analysis of SREBP2 and its target proteins in cells transfected with ERLIN1-specific siRNAs or an INSIG1 OE plasmid for 24 hours, followed by incubation in depletion medium for 24 hours. The right panel shows densitometric quantification of protein expression (n=3). (F) Immunoblot analysis of SREBP2 (N) levels upon cholesterol-rescued. Cells were under cholesterol rescued condition for 8 hours after pretreatment with cholesterol depletion medium plus 1% HPBCD for 4 hours. CHO, cholesterol; CHX, cycloheximide; ERLIN1, ER lipid raft-associated 1; EV, empty vector; HPBCD, hydroxypropyl-β-cyclodextrin; MCD, methyl-β-cyclodextrin; OE, overexpression.
Figure 5.ASB11 interacts with ERLIN1 and promotes its ubiquitin-proteasomal degradation. (A) Identification of ERLIN1-specific E3 ligases and deubiquitinases through BioGRID analysis, excluding ERLIN2-associated proteins. (B) Co-IP assays of ASB11’s interaction with exogenous ERLIN1, exogenous ERLIN2, and endogenous ERLIN1 in Huh7 cells. The asterisk denotes nonspecific bands. (C) Domain mapping analysis identified the interacting regions of ASB11 and ERLIN1. (D) Half-life of ERLIN1 was determined by a CHX-chase assay. (E) Immunoblot analysis of ERLIN1 expression in cells treated with 50 μM MG-132 for 12 hours. (F) Ubiquitination assay in MG-132-treated Huh7 cells identified the ASB11-mediated polyubiquitin chain of ERLIN1. Representative blots from three independent experiments are shown; densitometric quantification was performed based on three biological replicates (n=3). CHX, cycloheximide; Co-IP, co-immunoprecipitation; CT, c-terminal domain; ERLIN1, ER lipid raft-associated 1; FL, full length; SPFH, stomatin/prohibitin/flotillin/HflK/C domain.
Figure 6.The FIH-ASB11 regulatory axis represses ERLIN1 expression upon hypoxia. (A) Half-life of ERLIN1 was determined by a CHX-chase assay in Huh7 cells treated with or without 200 μM CoCl2. (B) Immunoblot analysis of ERLIN1 expression in Huh7 and 97H cells following 12 hours treatment with CoCl2 or culture under 1% O2, with or without 50 μM MG-132. (C) Co-IP analysis showed that hypoxia promoted ERLIN1 K48-linked ubiquitination and enhanced the interaction between ASB11 and ERLIN1 in Huh7 cells treated with CoCl2 and MG-132 for 12 hours. (D) Immunoblot analysis of ERLIN1 expression in Huh7 cells after (i) transfection with ASB11 siRNA or control siRNA, followed by 12 hours treatment with CoCl2; (ii) transfection with Flag-tagged FIH or specific FIH siRNAs and other indicated plasmids or siRNAs. The asterisk denotes nonspecific bands. (E) The interaction between ERLIN1 and ASB11 mutant was examined by CoIP in cells treated with MG-132 and CoCl2 for 12 hours. The asterisk denotes nonspecific bands. (F) Immunoblot analysis of ERLIN1 protein levels in Huh7 cells expressing the indicated plasmids and treated with CoCl2 for 24 hours. Representative blots from three independent experiments are shown; densitometric quantification was performed based on three biological replicates (n=3). CHX, cycloheximide; Co-IP, co-immunoprecipitation; ERLIN1, ER lipid raft-associated 1; EV, empty vector; FIH, factor-inhibiting hypoxia-1.
Figure 7.Zoledronic acid attenuates the ERLIN1-ASB11 interaction to suppress HCC progression. (A) Immunoblot analysis of ERLIN1 expression in cells treated with indicated concentration of ZoA under 50 μM CHX and 200 μM CoCl2 for 12 hours. Representative blots and quantification from three independent biological replicates are shown (n=3). (B) Co-IP analysis of the effect of ZoA on the interaction between ASB11 and ERLIN1 in cells treated with MG-132 and CoCl2 plus ZoA for 12 hours. The asterisk denotes nonspecific bands. Representative blots and quantification from three independent biological replicates are shown (n=3). (C) Cell viability of cells treated with a gradient concentration of ZoA. shCtrl or shERLIN1 Huh7 cells were exposed to ZoA under normoxic or 1% O2 conditions in 2% FBS-containing medium for 72 hours (n=3, mean±standard deviation). (D) ALOD4 staining analysis of the effect of ZoA on cholesterol levels in shCtrl or shERLIN1 Huh7 cells treated with CoCl2 (n=3, mean±standard deviation, unpaired Student’s t-test). (E) The effect of ZoA on xenograft tumor growth (n=7). Experimental design, representative tumors, tumor volume curves (mean±standard error of the mean, two-way ANOVA), final tumor weight (mean±standard error of the mean, unpaired Student’s t-test), and metabolomic analysis of cholesterol and cholesterol esters in tumors are shown. (F) The effect of ZoA on orthotopic tumor growth (n=11). Experimental design, bioluminescent imaging, and total flux quantification are shown. *P<0.05, **P<0.01, ***P<0.001, NS>0.05. CHX, cycloheximide; ERLIN1, ER lipid raft-associated 1; HCC, hepatocellular carcinoma; HIF, hypoxia inducible factor; ZoA, zoledronic acid.
Abbreviations
ER lipid raft-associated 1
formalin-fixed paraffin-embedded
factor inhibiting hypoxia inducible factor
hydroxypropyl-β-cyclodextrin
lipoprotein-deficient serum
stomatin/prohibitin/flotillin/HflK/C domain
sterol regulatory element-binding protein 2
REFERENCES
- 1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49.
- 2. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatocellular carcinoma. J Hepatol 2025;82:315-374.
- 3. Jiang Y, Sun A, Zhao Y, Ying W, Sun H, Yang X, et al. Proteomics identifies new therapeutic targets of early-stage hepatocellular carcinoma. Nature 2019;567:257-261.
- 4. Gao Q, Zhu H, Dong L, Shi W, Chen R, Song Z, et al. Integrated proteogenomic characterization of HBV-related hepatocellular carcinoma. Cell 2019;179:561-577.e22.
- 5. Jiang X, Ge X, Huang Y, Xie F, Chen C, Wang Z, et al. Drug resistance in TKI therapy for hepatocellular carcinoma: Mechanisms and strategies. Cancer Lett 2025;613:217472.
- 6. Rimassa L, Finn RS, Sangro B. Combination immunotherapy for hepatocellular carcinoma. J Hepatol 2023;79:506-515.
- 7. Liu X, Lv M, Zhang W, Zhan Q. Dysregulation of cholesterol metabolism in cancer progression. Oncogene 2023;42:3289-3302.
- 8. Radhakrishnan A, Goldstein JL, McDonald JG, Brown MS. Switch-like control of SREBP-2 transport triggered by small changes in ER cholesterol: a delicate balance. Cell Metab 2008;8:512-521.
- 9. Jiang W, Hu JW, He XR, Jin WL, He XY. Statins: a repurposed drug to fight cancer. J Exp Clin Cancer Res 2021;40:241.
- 10. Jouve JL, Lecomte T, Bouché O, Barbier E, Khemissa Akouz F, Riachi G, et al. Pravastatin combination with sorafenib does not improve survival in advanced hepatocellular carcinoma. J Hepatol 2019;71:516-522.
- 11. Kim ST, Kang JH, Lee J, Park SH, Park JO, Park YS, et al. Simvastatin plus capecitabine-cisplatin versus placebo plus capecitabine-cisplatin in patients with previously untreated advanced gastric cancer: a double-blind randomised phase 3 study. Eur J Cancer 2014;50:2822-2830.
- 12. Li S, Yan L, Li C, Lou L, Cui F, Yang X, et al. NPC1 controls TGFBR1 stability in a cholesterol transport-independent manner and promotes hepatocellular carcinoma progression. Nat Commun 2025;16:439.
- 13. Geng F, Cheng X, Wu X, Yoo JY, Cheng C, Guo JY, et al. Inhibition of SOAT1 suppresses glioblastoma growth via blocking SREBP-1-mediated lipogenesis. Clin Cancer Res 2016;22:5337-5348.
- 14. Zhang S, Fang T, He Y, Feng W, Yu Z, Zheng Y, et al. VHL mutation drives human clear cell renal cell carcinoma progression through PI3K/AKT-dependent cholesteryl ester accumulation. EBioMedicine 2024;103:105070.
- 15. Browman DT, Resek ME, Zajchowski LD, Robbins SM. Erlin-1 and erlin-2 are novel members of the prohibitin family of proteins that define lipid-raft-like domains of the ER. J Cell Sci 2006;119:3149-3160.
- 16. Huber MD, Vesely PW, Datta K, Gerace L. Erlins restrict SREBP activation in the ER and regulate cellular cholesterol homeostasis. J Cell Biol 2013;203:427-436.
- 17. Teo QW, Wong HH, Heunis T, Stancheva V, Hachim A, Lv H, et al. Usp25-Erlin1/2 activity limits cholesterol flux to restrict virus infection. Dev Cell 2023;58:2495-2509.e6.
- 18. Cai D, Zhong GC, Dai X, Zhao Z, Chen M, Hu J, et al. Targeting FDFT1 reduces cholesterol and bile acid production and delays hepatocellular carcinoma progression through the HNF4A/ALDOB/AKT1 axis. Adv Sci (Weinh) 2025;12:e2411719.
- 19. Luo J, Yang H, Song BL. Mechanisms and regulation of cholesterol homeostasis. Nat Rev Mol Cell Biol 2020;21:225-245.
- 20. Leissing TM, Hardy AP, Chan H, Wang Y, Tumber A, Chowdhury R, et al. Factor inhibiting HIF can catalyze two asparaginyl hydroxylations in VNVN motifs of ankyrin fold proteins. J Biol Chem 2022;298:102020.
- 21. Deshaies RJ, Joazeiro CA. RING domain E3 ubiquitin ligases. Annu Rev Biochem 2009;78:399-434.
- 22. Husain A, Chiu YT, Sze KM, Ho DW, Tsui YM, Suarez EMS, et al. Ephrin-A3/EphA2 axis regulates cellular metabolic plasticity to enhance cancer stemness in hypoxic hepatocellular carcinoma. J Hepatol 2022;77:383-396.
- 23. Green JR, Müller K, Jaeggi KA. Preclinical pharmacology of CGP 42’446, a new, potent, heterocyclic bisphosphonate compound. J Bone Miner Res 1994;9:745-751.
- 24. Abeltino M, Bonomini S, Bolzoni M, Storti P, Colla S, Todoerti K, et al. The proapoptotic effect of zoledronic acid is independent of either the bone microenvironment or the intrinsic resistance to bortezomib of myeloma cells and is enhanced by the combination with arsenic trioxide. Exp Hematol 2011;39:55-65.
- 25. Siperstein MD, Fagan VM. Deletion of the cholesterol-negative feedback system in liver tumors. Cancer Res 1964;24:1108-1115.
- 26. Siperstein MD, Fagan VM, Morris HP. Further studies on the deletion of the cholesterol feedback system in hepatomas. Cancer Res 1966;26:7-11.
- 27. Erickson SK, Cooper AD, Barnard GF, Havel CM, Watson JA, Feingold KR, et al. Regulation of cholesterol metabolism in a slow-growing hepatoma in vivo. Biochim Biophys Acta 1988;960:131-138.
- 28. Bovenga F, Sabbà C, Moschetta A. Uncoupling nuclear receptor LXR and cholesterol metabolism in cancer. Cell Metab 2015;21:517-526.
- 29. Xiong L, Liu HS, Zhou C, Yang X, Huang L, Jie HQ, et al. A novel protein encoded by circINSIG1 reprograms cholesterol metabolism by promoting the ubiquitin-dependent degradation of INSIG1 in colorectal cancer. Mol Cancer 2023;22:72.
- 30. Che L, Chi W, Qiao Y, Zhang J, Song X, Liu Y, et al. Cholesterol biosynthesis supports the growth of hepatocarcinoma lesions depleted of fatty acid synthase in mice and humans. Gut 2020;69:177-186.
- 31. Veronese M, Kallabis S, Kaczmarek AT, Das A, Robers L, Schumacher S, et al. ERLIN1/2 scaffolds bridge TMUB1 and RNF170 and restrict cholesterol esterification to regulate the secretory pathway. Life Sci Alliance 2024;7:e202402620.
- 32. Solanki S, Shah YM. Hypoxia-induced signaling in gut and liver pathobiology. Annu Rev Pathol 2024;19:291-317.
- 33. Paredes F, Williams HC, San Martin A. Metabolic adaptation in hypoxia and cancer. Cancer Lett 2021;502:133-142.
- 34. Strowitzki MJ, Cummins EP, Taylor CT. Protein hydroxylation by hypoxia-inducible factor (HIF) hydroxylases: unique or ubiquitous? Cells 2019;8:384.
- 35. Scholz CC, Rodriguez J, Pickel C, Burr S, Fabrizio JA, Nolan KA, et al. FIH regulates cellular metabolism through hydroxylation of the deubiquitinase OTUB1. PLoS Biol 2016;14:e1002347.
- 36. Honda Y, Takahashi S, Zhang Y, Ono A, Murakami E, Shi N, et al. Effects of bisphosphonate zoledronic acid in hepatocellular carcinoma, depending on mevalonate pathway. J Gastroenterol Hepatol 2015;30:619-627.
- 37. Tang Q, Jiang S, Jia W, Shen D, Qiu Y, Zhao Y, et al. Zoledronic acid, an FPPS inhibitor, ameliorates liver steatosis through inhibiting hepatic de novo lipogenesis. Eur J Pharmacol 2017;814:169-177.
, Chunyan Tian1
