URI alleviates tyrosine kinase inhibitors-induced ferroptosis by reprogramming lipid metabolism in p53 wild-type liver cancers.

doi:10.21203/rs.3.rs-2341124/v1

Download PDF

Article

URI alleviates tyrosine kinase inhibitors-induced ferroptosis by reprogramming lipid metabolism in p53 wild-type liver cancers.

https://doi.org/10.21203/rs.3.rs-2341124/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 07 Oct, 2023

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Tyrosine kinase inhibitors (TKIs)-based systemic therapy for advanced hepatocellular carcinoma (HCC) has only limited clinical benefit and reaches a bottleneck due to its drug resistance. Here, through whole transcriptome sequencing (RNA-Seq) and lipidomic analysis, we uncover that lipid metabolism reprogramming mediated by unconventional prefoldin RPB5 interactor (URI) endows HCC with resistance to TKIs-induced ferroptosis. Mechanistically, URI directly interacts with TRIM28 and promotes p53 ubiquitination and degradation in a TRIM28-MDM2 dependent manner. Importantly, we find that p53 binds to the promoter of stearoyl-CoA desaturase 1 (Scd1) and represses its transcription. Therefore, high expression of URI is correlated with high level of SCD1 and their synergetic expression predicts poor prognosis and TKIs resistance in HCC. The combination of SCD1 inhibitor aramchol and deuterated sorafenib derivative donafenib displays promising anti-tumor effects in xenografted models of wild-type p53 liver cancer cell lines. This combination therapy has potential significant clinical benefits for the patients with advanced HCC who have wild-type p53 and high levels of URI/SCD1.

Health sciences/Gastroenterology/Hepatology/Liver diseases/Liver cancer/Hepatocellular carcinoma

Biological sciences/Cancer/Cancer therapy/Cancer therapeutic resistance

Biological sciences/Cancer/Cancer therapy/Targeted therapies

Worldwide, primary liver cancer is the sixth most commonly diagnosed cancer and the third leading cause of cancer death. Hepatocellular carcinoma (HCC) accounts for 75–80% of all cases of liver cancer¹. Approximately 50–60% of HCC patients, particularly in advanced HCC, are estimated to receive systemic therapies in their lifespan². Currently, tyrosine kinase inhibitors (TKIs)-based systemic therapies are widely used in HCC patients with Barcelona Clinic Liver Cancer (BCLC)-B/C stage³, including sorafenib⁴, regorafenib⁵, lenvatinib⁶ and donafenib⁷. Unfortunately, these drugs only moderately improved survival as single agent. Recent studies indicated that combination therapy, such as the combination of TKIs and immune checkpoint inhibitors (ICIs), shows a bright prospect⁸. In the same way, optimal combination therapies and new biomarkers are required to improve the clinical benefits of TKIs-based therapies for HCC.

Sorafenib was the first available TKI for advanced HCC for a decade. However, the objective response rate for sorafenib in SHARP study is only 2%⁴, suggesting that most HCCs are intrinsically resistant to sorafenib. Recent studies suggest that epigenetics, transport processes, regulated cell death, and the tumor microenvironment are involved in the initiation and development of sorafenib resistance in HCC⁹. Ferroptosis, an iron-dependent form of regulated cell death procedure involving the abnormal metabolism of lipid oxides¹⁰, could be induced by sorafenib in different cancer cells, including HCC. Considering this, the strategy that making cell more vulnerable to ferroptosis might improve the therapeutic efficacy of TKIs¹¹.

Reprogrammed lipid metabolism of cancer cells not only supplies appropriated energy, but also plays a central role in ferroptosis¹². Among the metabolic process, the desaturation is of profound interest, since it is supposed to (1) prevent lipotoxicity from excess saturated fatty acyl chains, (2) inhibit ferroptosis triggered by the peroxidation of polyunsaturated fatty acyl chains, and (3) reduce membrane permeability to enhance drug resistance, which are all helpful for cancer cell survival¹³. Stearoyl-CoA desaturase 1 (SCD1) is one of the key enzymes that control the generation of monounsaturated fatty acids (MUFAs). It converts saturated fatty acids (SFAs), especially the palmitoyl-CoA (16:0) or stearoyl-CoA (18:0), to palmitoleic acid (16:1) or oleic acid (18:1), respectively¹⁴. Therefore, SCD1 activity is critical in maintaining the appropriate ratio of saturated to unsaturated fatty acids within the cell. Previous studies showed that SCD1 is highly expressed in some liver cancers, and its inhibition sensitizes cancer cells to sorafenib^15,16. However, how the abnormal SCD1 expression in HCC is regulated is still need to be explored.

Unconventional prefoldin RPB5 interactor (URI) is reported to be involved in the nicotinamide adenine dinucleotide (NAD⁺) metablism¹⁷, nonalcoholic fatty liver disease¹⁸ and glucose metabolism¹⁹. Our group has also reported that URI promotes tumorigenesis and chemoresistance of intrahepatic cholangiocarcinoma (iCCA) by modulating oxidative stress²⁰. Here, we uncovered the new role of URI in the reprogramming of lipid metabolism in HCC, partially by maintaining aberrant SCD1 expression in a p53-dependent manner, which in turn promotes HCC resistant to TKIs-induced ferroptosis. We also provided a strong rationale for testing the combination therapy with SCD1 inhibitors and TKIs for the patients with advanced HCC.

URI promotes resistance to TKIs-induced ferroptosis in liver cancer cells.

Firstly, we performed RNA sequencing (RNA-seq) analysis to characterize the URI-dependent transcriptomic alteration. We found that a total of 1676 genes were differentially expressed between shRNA (Ctrl) and shURI HepG2 cells (Extended Data Fig. 1a). Functional enrichment analyses using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway revealed that ferroptosis was one of the top altered pathways regulated by URI (Fig. 1a). The expression of ferroptosis related genes¹⁰ (Slc7a11, Sat1 and Chac1) were increased with URI knockdown (Extended Data Fig. 1b). Gene set enrichment analysis (GSEA) revealed enrichment of ferroptosis but not apoptosis signaling upon URI depletion (Fig. 1b and Extended Data Fig. 1c). These date suggest that URI might play an important role in ferroptosis. Meanwhile, pathways, including EGFR tyrosine kinase inhibitor resistance, regulation of lipolysis, p53 signaling pathway, and transcriptional misregulation in cancer, were also regulated by URI (Fig. 1a).

There have been many clinical trials evaluating TKIs for advanced HCC (Fig. 1c)^{4–7, 21,22}. Consistent with the cytotoxic effect, these TKIs effectively reduced tumor viability among various cancer cell lines tested (Fig. 1d and Extended Data Fig. 2a). Notably, sorafenib, regorafenib and donafenib, with B-RAF as their common targets, favored ferroptosis-like cell death, while other drugs tested showed little effect on cellular lipid peroxidation (Fig. 1e and Extended Data Fig. 2b).

We then tested whether URI regulates TKIs-induced cytotoxicity. JHH1 and HepG2 cells were treated with increasing concentrations of sorafenib for 48 h and cell proliferation was assessed. URI depletion significantly increased the sensitivity of JHH1 and HepG2 cells to sorafenib (Extended Data Fig. 2c), with a decreased in IC50 (causing 50% inhibition of viability) from 8.324 µM (Ctrl) to 6.480 µM (shURI) in JHH1 cells, and from 8.047 µM (Ctrl) to 4.069 µM (shURI) in HepG2 cells, respectively (Fig. 1f). Similar patterns of dose-effect curves were also found in response to regorafenib or donafenib between HepG2-shURI and control cells (Extended Data Fig. 2d). Clonal formation assays confirmed the resistant role of URI in sorafenib treatment (Fig. 1g, h). Interestingly, URI knockdown in JHH1 and HepG2 cells substantially increased sorafenib-induced lipid peroxidation (Fig. 1i), suggesting that URI might inhibit the TKIs-induced ferroptosis. Furthermore, sorafenib-induced cytotoxic effect in URI knockdown cells could be rescued when they were pretreated with the antioxidant N-acetylcysteine (NAC) or the ferroptosis inhibitor ferrostatin-1 (Fer-1; lipid peroxidation scavenger), while the inhibitors of apoptosis, necrosis, or autophagy had little effect (Fig. 1j). In addition, URI depletion enhanced RSL3 (a GPX4 inhibitor)-induced ferroptosis as well (Extended Data Fig. 2e). Together, these results suggest that URI is associated with resistance to TKIs-induced ferroptosis in liver cancer cells.

Ectopic expression of URI reprograms SCD1-associated lipid metabolism in cancer cells.

Cancer cells require higher levels of lipid metabolism than normal cells, which can determine the sensitivity of cells to ferroptosis¹². To examine URI-mediated lipid metabolism and lipidomic change, mass spectrometry-based lipidomic analysis were performed in HepG2-shURI and HepG2-Ctrl cells (Extended Data Fig. 3a-c). Our lipidomic data revealed 961 different lipid species, consisting of 219 phosphatidylcholines (PCs), 215 triglycerides (TGs), 106 phosphatidylethanolamines (PEs), and other lipid classes (Extended Data Fig. 3a). The relative content of TGs, the most abundant lipid class mainly containing monounsaturated fatty acid chain, was significantly decreased in URI depleted HepG2 cells than its control, with the decreases of phosphatidylinositol (PI(18:1/18:1)) and diglyceride (DG(16:1/18:1/0:0)) (Fig. 2a, b). In contrast, the relative content of phosphatidylglycerol (PG(16:0/16:0)) was increased in HepG2-shURI cells (Fig. 2b). Meanwhile, PG(16:0/16:0) presented a negative correlation with TG(16:1/16:1/16:1), TG(16:1/18:1/16:1), and PI(18:1/18:1) (Extended Data Fig. 3d). These results suggested that URI depletion changes lipid compositions in liver cancer cells.

Given the alternation contents of lipid species with decreased C16:1 and C18:1 acyl chain and increased C16:0 or C18:0 in URI knockdown cells, we hypothesized that fatty acid synthesis, especially C16 and C18 fatty acids, was reprogrammed by URI depletion. To explore this possibility, we first analyzed the fatty acids contents in HepG2 cells. The intracellular lipid peroxidation, a major event in ferroptosis, was determined by the ratio of MUFAs and polyunsaturated fatty acids (PUFAs)^23,24. We found that the SFAs levels were increased in HepG2-shURI cells than its controls, and the SFAs-derived metabolites MUFAs levels, especially the 16:1(n-7) and 18:1(n-9) MUFAs, were significantly decreased (Fig. 2c). These results strongly suggest that URI knockdown leads to an impaired convention from SFAs to MUFAs. SCD1 is responsible for the synthesis of 16:1(n-7) and 18:1(n-9) MUFAs from the 16:0 and 18:0 SFAs, respectively (Extended Data Fig. 3e). Notably, the16:1(n-7)/16:0 and 18:1(n-9)/18:0 ratios, which are surrogates of SCD1 activity, were significantly decreased in HepG2-shURI cells than their controls (Fig. 2d), indicating a potent association between URI expression and SCD1 activity/expression.

De novo fatty acid synthesis (FAS) involves the coordinated actions of several enzymes (Fig. 2e)²⁵. We next measured mRNA levels of the key enzymes in FAS pathway in JHH1 and HepG2 cells together with their URI depletion ones (Fig. 2f and Extended Data Fig. 3f). The transcripts of enzymes participating in saturated fatty acid synthesis or fatty acid elongation, including Scd1, Fasn, Fads2, Acaca and Elovl6, were significant reduced in JHH1-shURI and HepG2-shURI cells, comparing with their parental cells. (Fig. 2f). Altogether, these results indicate that URI depletion reprograms lipid metabolism in liver cancer cells.

URI promotes resistance to TKIs-induced ferroptosis via SCD1

Lipid metabolic reprogram is involved in cancer drug resistance²⁶. Gene ontology (GO) analysis also showed that the lipid catabolic process is enriched in sorafenib-treated HepG2 cells versus its control cells (Extended Data Fig. 3g). To explore whether the protein levels of enzymes involved in lipid metabolism would be changed by URI depletion, we analyzed their levels in various cell lines at basal status. Although significant transcriptional changes of Acaca, Fasn and Fads2 in URI-knockdown cells was observed (Fig. 2f), their protein levels were not affected by URI (Fig. 3a). Remarkably, we found that lower SCD1 protein levels in shURI cells than their controls (Fig. 3a), consisting with the reduction of Scd1 mRNA levels (Fig. 2f), suggesting that URI could regulate SCD1 activity by affecting its transcription.

By re-analysis the public transcriptional GEO datasets of sorafenib-treated HCC tumor cells and sorafenib-resistant xenografts, an overlap of 13 genes, including Scd1, was identified as the core gene set involved in sorafenib-resistant (Extended Data Fig. 4a). Western-blot assay confirmed the reduction of SCD1 in sorafenib-treated JHH1 and HepG2 cells (Extended Data Fig. 4b). These results suggested a potent role of SCD1 in sorafenib resistance. Interestingly, the ferroptosis inducer, RSL3, also slightly reduced SCD1 expression in tumor cells tested (Fig. 3b). The cystine-import-GSH-GPX4 machinery is the canonical pathway in regulating ferroptosis²⁷. However, URI depletion induced mild changes of the protein levels of SLC7A11, GCLC, GCLM and GPX4 in JHH1 and HepG2 cells treated with or without RSL3 (Fig. 3a, b and Extended Data Fig. 4b). Interestingly, we found that treatment with RSL3 or sorafenib at a lethal dosage can significantly decrease the protein levels of SCD1 but not GPX4 in wild-type and shURI cells (Fig. 3b and Extended Data Fig. 4b). Thus, we speculated that URI might promote resistance to TKIs-induced ferroptosis by upregulating SCD1 expression. We found that the combination of RSL3 and A939572 or MK8245, both are SCD1 inhibitors, showed synergistic effect in liver cancer cells (Fig. 3c). SCD1 inhibitor alone did not induce ferroptosis in Ctrl cells compared with shURI cells (Extended Data Fig. 4c, d). We further explored whether SCD1 inhibitors improved the therapeutic effects of sorafenib on liver cancer cells as well. As expected, the combination of sorafenib and A939572 or MK8245 promoted cellular ferroptosis. However, this combination therapy showed little cytotoxic effect in shURI cells (Fig. 3d and Extended Data Fig. 4e, f). Consistent with these results, SCD1 inhibitors and sorafenib showed a synergistic inhibition of tumor cell growth in long-term clonogenic assays (Fig. 3e, f and Extended Data Fig. 4g, h). To confirm these findings, we stably transfected MOCK or Flag-SCD1 plasmids into cancer cells. Our results showed that stable expression of Flag-SCD1 in URI knockdown cells was sufficient to re-resist the cells to ferroptosis (Fig. 3g, h). Further experiments confirmed that the combination sorafenib plus SCD1 inhibitors significantly elevated lipid peroxidation. On the contrary, supplementation of exogenous SCD1 declined sorafenib-induced lipid peroxidation (Fig. 3i). Altogether, our results reveal that URI promotes resistance to TKIs-induced ferroptosis, by upregulating SCD1.

URI positively regulates SCD1 transcription via wild-type p53 in liver cancer cells.

SCD1 has been shown to be regulated both at the transcriptional level by several transcription factors (TFs) and at the post-translational level through ubiquitination and subsequent degradation by the proteasome¹⁴. Transfecting with His-URI plasmids into shURI cells could rescue SCD1 protein levels (Fig. 4a). To investigate how URI upregulates SCD1 expression in cancer cells, we first examined the effect of URI on the protein stability of SCD1. MG132, a proteasome inhibitor, did not increase the protein levels of SCD1 in shURI cells (Extended Data Fig. 5a). Consistently, the ubiquitination levels of neither endogenous nor exogenous SCD1 proteins were affected by URI expression (Extended Data Fig. 5b, c). These results suggest that URI upregulates SCD1 expression independent of the ubiquitin-proteasome system. Thus, we hypothesized that URI might promote SCD1 expression by upregulating its transcription. Indeed, reporter assay showed that the fluorescence intensity of SCD1-luc in shURI cells was significantly lower than that in Ctrl cells (Fig. 4b).

To further identify which transcription factor mediates the upregulation of SCD1 by URI, we transfected siRNA targeting SREBP, ChREBP or LXRα^14,28, the reported main TFs in regulating SCD1, into cancer cells (Extended Data Fig. 5d). The URI-mediated differences of SCD1-luc fluorescence intensities were not abrogated after interfering these TFs (Fig. 4c). As ChREBP and LXRα were shown to increase SCD1 transcription in a SREBP-dependent manner^29,30, we performed chromatin immunoprecipitation (ChIP) analysis with SREBP antibody in HepG2-Ctrl and HepG2-shURI cells and revealed that multiple binding sites of SREBP in the Scd1 promoter regions (Extended Data Fig. 5e-h). However, the binding of SREBP with Scd1 promoter was not affected by URI expression, consistent with the luciferase reporter assay. Our data suggest that these transcription factors are not involved in URI regulating SCD1.

Previous study has suggested that Scd1 gene is a target of p53-mediated transcriptional repression³¹. Consistently, our CUT&Tag analysis showed the existence of the binding sites for p53 at the promotor region of Scd1 (Fig. 4e). GSEA revealed a significant positive enrichment of p53 transcription gene network upon URI depletion (Fig. 4f). To validate the role of p53-mediated effects on SCD1 expression, we treated JHH1 and HepG2 cells (which express wild-type p53) with doxorubicin (Dox), a well-known p53-activating agent. The results showed that SCD1 protein and Scd1 mRNA expression were decreased after Dox treatment (Fig. 4g, h). We next transfected Myc-p53, mutant p53-R273H, or p53-R249S into p53-null cells (Hep3B), and found that wild-type p53 dramatically downregulated the protein levels of SCD1. In contrast, neither DNA-contact mutants (p53-R273H) nor conformational mutants (p53-R249S)³² affected SCD1 protein levels (Fig. 4i). We explored the potential p53 binding locations and sequences on the human Scd1 gene (Extended Data Fig. 5f, h). Indeed, the 5’flanking region of the Scd1 gene contains one site that matches the consensus p53-binding sequence (Fig. 4j). ChIP analysis further demonstrated that endogenous p53 binds on the promoter region of the Scd1 gene (Fig. 4k). Moreover, the increased protein levels by p53-knockdown could not reversed by nutlin-3 treatment (Fig. 4l). We speculated that URI upregulated SCD1 expression was mediated by p53. URI depletion enhanced p53 binding on the promoter region of the Scd1 (Fig. 4m). We next transfected siRNA of p53 into HepG2-Ctrl and HepG2-shURI cells and found that knockdown of p53 abrogated shURI-mediated SCD1 repression (Fig. 4n). Our data showed that URI depletion significantly decreased SCD1 expression in wild-type p53 cell (JHH1 and HepG2) (Fig. 3a). Similar results were also observed in other human cancer cells expression wild-type p53 (HCT116), but not in p53-null (Hep3B) or p53 mutant cells (HUH7, PLC/PRF/5 and HT29) (Extended Data Fig. 5i). Together, these data suggest that URI promotes transcriptional levels of SCD1 mediated by repressing p53.

URI promotes ubiquitination and degradation of wild-type p53 in cancer cells.

We next transfected His-URI plasmids into JHH1 and HepG2 cells and found that this treatment led to significant reduction of p53 proteins, even in the presence of Dox (Fig. 5a). To explore the molecular mechanisms of the decreased expression of p53 mediated by URI, we first examined the mRNA levels of p53 in Ctrl and shURI cells. Our data showed that URI depletion could not affect the transcriptional levels of p53 (Fig. 5b). Comparable mRNA levels of SREBP and ChREBP were also found (Extended Data Fig. 6a). Stress-induced p53 activation largely occurs through protein stabilization³³. We further explored whether URI regulates p53 protein stability. Indeed, the half-life of p53 was prolonged in JHH1-shURI cells compared with their control cells by using cycloheximide (CHX) (Fig. 5c). URI depletion significantly decreased ubiquitination of wild-type p53 in HepG2 cells (Fig. 5d and Extended Data Fig. 6b).

Since URI is not an E3 ubiquitin ligase, we hypothesized that URI might recruit the E3 ubiquitin ligase to mediate ubiquitination of p53. We further performed liquid chromatography tandem-mass spectrometry (LC-MS/MS) analysis and identified potential URI binding proteins, including TRIM28, USP5, USP7 and USP14, which were reported to regulate ubiquitination of p53^34–37 (Fig. 5e). To validate the results of the LC-MS/MS, candidates were tested for binding with URI by an immunoprecipitation assay. The results showed an interaction between URI and TRIM28 (Fig. 5f), consistent with previous studies³⁸. Two different siRNAs targeting TRIM28 increased p53 protein levels particularly in shURI cells (Fig. 5g and Extended Data Fig. 6b). Furthermore, knockdown of TRIM28 resulted in decreased ubiquitination of p53 with MG132 treatment (Fig. 5h and Extended Data Fig. 6c). As TRIM28 binding with MDM2 contributes to p53 protein stability, we explored whether URI could affect the interaction between TRIM28 and MDM2. Immunoprecipitation revealed that URI depletion inhibited the formation TRIM28-MDM2 (Fig. 5i and Extended Data Fig. 6d, e). In contrast, overexpression of URI promotes the cooperation of MDM2 and TRIM28 (Fig. 5j). Moreover, URI depletion decreased the complex of TRIM28-MDM2-p53 (Fig. 5i, j and Extended Data Fig. 6d-f), suggesting that URI might recruit E3 ubiquitin ligase TRIM28 and promote the effect of TRIM28-MDM2 on p53 protein ubiquitination. In vitro GST-pull down assays were performed to confirm whether URI directly interact with TRIM28. The interaction between purified GST-URI and His-TRIM28 increased significantly in a concentration-dependent manner (Fig. 5k). Together, our results suggest that URI interacts with TRIM28 and promotes the formation of TRIM28-MDM2-p53 complex, subsequently ubiquitinating p53. Next, we determined whether modulation of p53 protein levels by URI/TRIM28 had a functional impact on ferroptosis. Knockdown of TRIM28 significantly promotes sorafenib-induced cell death and lipid peroxidation (Fig. 5l, m). In addition, most cancer cells with p53 mutation showed resistant to TKIs in CCLE database, suggesting that wild-type p53 might sensitize cells to TKIs (Extended Data Fig. 6g-i).

SCD1 inhibitor synergizes with donafenib in liver cancer.

Our aforementioned data showed that inhibition of SCD1 significantly reversed the resistance to TKIs of liver cancer cells. We speculated that combination of TKIs and SCD1 inhibitors might be efficient in p53 wild-type cancer. Aramchol, a novel partial inhibitor of SCD1, has been proved as a promising therapy for nonalcoholic steatohepatitis (NASH) and led to no severe adverse events in a phase 2b clinical trial, compared with other SCD1 inhibitors ³⁹. To evaluate the safety and anti-tumor activity of aramchol in vivo, we utilized a xenograft model of rodents (Extended Data Fig. 7a). Aramchol did not cause weight loss of nude mice (Extended Data Fig. 7b). Moreover, there was no therapeutic effect of aramchol alone on subcutaneous tumors (Extended Data Fig. 7c-e). Indeed, the observed changes of aspartate aminotransferase (AST) indicated a benefit of aramchol in improving liver injury (Extended Data Fig. 7f, g). In addition, the levels of serum TG were not changed on a normal diet (Extended Data Fig. 7h). Aramchol did not reduce the weight or induce obviously pathologic changes of mainly organs, including the liver, spleen, kidney, lung and heart, either (Extended Data Fig. 7i-k). Altogether, these results suggest the safety of aramchol in mice models.

Aramchol had no potent inhibitory function on cell viability in vitro (Fig. 6a and Extended Data Fig. 8a), consistent with the results in xenograft models. Importantly, aramchol enhanced sensitivity of cancer cells to donafenib, a deuterated sorafenib derivative showed superiority over sorafenib⁷, in long-term clonogenic assays (Fig. 6b, c). We next investigated whether aramchol could show a synergistic effect with donafenib in vivo (Fig. 6d). The combination of armachol and donafenib elicited a complete inhibition of tumor growth in mice xenografted with HepG2-Ctrl and HepG2-shURI cells, whereas the donafenib single treatment played an effective therapeutic role in xenografts with HepG2-shURI cells (Fig. 6e-g). In addition, the SCD1 inhibitor ararmchol, alone or in combination with donafenib, has no effect on body weight of mice (Extended Data Fig. 8b). Tumor cell viability was further assessed to evaluate the anti-tumoral effect of the treatments. The combination treatment reduced tumor viability in both of the HepG2-Ctrl and HepG2-shURI models (Extended Data Fig. 8c, d). Immunohistochemical analysis of tumour sections further revealed decreased Ki67 in tumors from mice treated with aramchol and donafenib. Moreover, combined aramchol and donafenib was found to promote synergistic loss of SCD1 and p53 (Fig. 6h, i). Together, these results showed that the SCD1 inhibitor aramchol is synergetic lethal with donafenib in liver cancer.

Combined higher expression of URI and SCD1 correlated with worsen prognosis and sorafenib-resistance in HCC

In our previous study⁴⁰, we found that URI high-expression in HCC is associated with cancer malignant and poor survival of patients. To further confirm the functional link between the URI-SCD1 axis and clinical cancer therapy, we utilized a tissue microarray of cohort consisting with 134 advanced HCC samples (cohort A). Consistent with the limited clinical responses, 5-year survival rate of patients with BCLC-B/C stage HCC in cohort A was merely 11.2%, even though they have received hepatectomy (Supplementary Table 1). We further performed multiplex immunohistochemistry/immunofluorescence (mIHC/IF) in these samples to test URI and SCD1 expression, and H-score method was employed to quantitate their expression as described in the material and method (Fig. 7a). URI expression showed a positive correlation with SCD1 expression in cohort A samples and database samples (Fig. 7b and Extended Data Fig. 9a). Both of URI and SCD1 were accumulated in tumor tissue compared with para-tumor tissues (Fig. 7c). High expression of URI was associated with high levels of tumor marker α-fetoprotein (AFP) (Extended Data Fig. 9b). Patients with high URI or SCD1 expression were correlated with shorter overall survival (OS) (Extended Data Fig. 9c, d). The 5-year OS rate of patients with high URI was 3.7% (hazard ratio (HR), 2.50; 95% CI, 1.74–3.59) compared with 22.6% (HR, 0.40; 95% CI, 0.28–0.58) for low URI (p < 0.0001, log-rank test). Similar to our results, data from Gepia verified the correlation between of URI and poor survival in different cancers (Extended Data Fig. 9e-h). Meanwhile, the 5-year OS rate of patients with high SCD1 was 10.0% (HR, 1.75; 95% CI, 1.22–2.50) compared with 13.6% (HR, 0.57; 95% CI, 0.40–0.82) for low SCD1 (p = 0.0038, log-rank test). Furthermore, we stratified patients into four groups according to the expression of URI and SCD1. Significant differences in OS among different subclasses were illustrated by Kaplan-Meier curves (p < 0.0001, log-rank test) (Fig. 7d). Patients with synergetic higher URI and SCD1 expression were at the highest risk with 5-year OS rate of 4.3% (HR, 2.65; 95% CI, 1.76–3.98) compared with 18.8% (HR, 0.38; 95% CI, 0.25–0.57) for lower URI and SCD1. These results indicate that the combination of URI and SCD1 expression is a highly effective predictor for poor prognosis of HCC.

To further explore whether the role of URI-SCD1 in sorafenib resistance is clinically relevant, the expression of URI and SCD1 was examined in a cohort of HCC patients received adjuvant sorafenib treatment (Cohort B) (Fig. 7e, Extended Data Fig. 10a and Supplementary Table 2). Our data showed that URI expression was closely associated with SCD1 expression, consistent with the results from cohort A (Extended Data Fig. 10b). As most HCC recurrences occurred within 1 year after liver resection, we next analyzed whether URI and SCD1 expression is associated with 1-year recurrence free survival rate (RFS) of patients received adjuvant sorafenib treatment (Fig. 7f, g). The 1-year RFS rate of patients with high URI was 20.9% (hazard ratio (HR), 2.21; 95% CI, 1.35–3.63) compared with 53.3% (HR, 0.45; 95% CI, 0.28–0.74) for low URI (p = 0.0051, log-rank test). Moreover, the 1-year RFS rate of patients with high SCD1 was 21.7% (HR, 1.81; 95% CI, 1.11–2.95) compared with 48.6% (HR, 0.55; 95% CI, 0.34–0.90) for low SCD1 (p = 0.0222, log-rank test). Importantly, patients with combined higher URI and SCD1 expression were at the highest risk with 1-year RFS rate of 20.0% (HR, 3.00; 95% CI, 1.72–5.22) compared with that of 64.0% (HR, 0.33; 95% CI, 0.19–0.58) for low URI and low SCD1 (Fig. 7h). Altogether, these results suggest that URI combined with SCD1 could predict which patients will benefit from sorafenib treatment.

Single-agent TKIs provide modest clinical benefits in unresectable HCC with only 9.2–24.1% objective responders according to mRECIST 1.1^4–7. To improve the TKIs therapeutic effect, it is urgent to explore the TKIs resistance mechanism and identify therapeutic targets. In this study, through RNA-seq and lipidomic analysis, we found that URI links TKIs-induced ferroptosis with lipid metabolism reprogramming. URI directly interacts with TRIM28, which leads to ubiquitination and degradation of wild-type p53, preventing the transcriptional inhibition of p53 on Scd1 and promoting the generation of desaturation products. Subsequently, SCD1-dependent desaturation decreases TKIs-induced ferroptosis and enhances intrinsic resistance of cancer cells to TKIs. Attenuation the URI-p53-SCD1 axis through URI knockdown or SCD1 inhibitors effectively enhances TKIs-induced ferroptosis in vitro and synergizes the efficacy of TKIs therapy in vivo.

Previous studies have shown that URI is involved in HCC development and various metabolic processes^17–19. Here, we present an insight into the role of URI in liver cancer cells lipid metabolism reprogramming. URI significantly decreased TKIs-induced ferroptosis without affecting GPX4 pathway, which is a classic ferroptosis regulation pathway. Unrestrained lipid peroxidation is the hallmark of ferroptosis. MUFAs and PUFAs are both participated in the regulation of lipid peroxidation and are key players in determining the sensitivity of cells to ferroptosis. Increased the ratios of MUFA to SFA or PUFA in cancer cell membranes result in reduced lipotoxicity and suppressed susceptibility to ferroptosis^13,27. We found that URI promoted the accumulation of MUFAs but not the clearance of PUFAs. Since MUFAs are mainly produced with desaturation, we propose that desaturation process upregulated by URI, followed by maintaining the fluidity of cancer cell membrane, may be a prerequisite for resistance to ferroptosis.

In humans, SCD1 is the rate-limiting enzyme of desaturation and MUFA synthesis. It introduces a cis double bond at the C9 position of SFAs to MUFAs¹⁴. Here, we found that URI promotes SCD1 activity, consistent with the elevated MUFAs. The combination of SCD1 inhibitors and TKIs can efficiently abrogate the TKIs resistance mediated by URI. In contrast, exogenous SCD1 enhanced TKIs resistance even in URI knockdown cancer cells. The findings imply that URI promotes TKIs-resistance in a SCD1-dependent manner. Indeed, SCD1 was found to be upregulated in HCC^15,16. The participation of SCD1 in ferroptosis resistance has also been established^41,42. Of note, we found that SCD1 expression is positively correlated with URI expression in HCC tissues. Simultaneous high expressions of URI and SCD1 (51.5%) predicted a much poor overall survival rate for patients with advanced HCC, compared with the low expression of the two molecules. In addition, the 1-year RFS of the patients, who underwent liver resection and followed sorafenib treatment with high-expression URI and SCD1, were three times higher than those of the low-expression group. The findings suggest that the combination of URI and SCD1 is a potential biomarker for the prediction of the sorafenib response in HCC patients as well. However, how the upregulated SCD1 is induced in cancer cells is still unclear. We found that URI promotes the transcriptional levels of SCD1. The traditional transcriptional factors of SCD1, including SREBP, ChREBP and LXRα, did not participate in this process. We next identified p53 as a transcriptional repressor of SCD1, which was significantly decreased by URI.

The wild-type p53 protein and its cellular pathways mediate tumor suppression, and activated p53 transcriptionally regulates lots of genes which are involved in multiple biological processes⁴³. Mutations in p53 gene (21%), which abrogate the tumor suppressor activities of p53 proteins, are the second most common single gene alterations in HCC^44,45. Approaches towards p53-based therapy are mainly made to develop drugs that target selectively p53 mutants. Moreover, therapeutic strategies that protect p53 from its negative regulators are also attractive⁴⁶. In this study, we identified that URI inhibits the functionality of wild-type p53 in TKIs-induced ferroptosis. It is well known that the E3 ubiquitin ligase MDM2 and p53 proteins form a central hub that receives stressful inputs via MDM2 and respond via p53. TRIM28 interacts with MDM2 and suppresses p53 activation by decreasing the half-life of the p53 protein and rapidly inhibiting it for transcription³⁴. We found that URI directly binds with TRIM28, promotes p53 ubiquitination and degradation in a TRIM28-MDM2 dependent manner and subsequently attenuates p53 transcriptional repression targeted SCD1. This may helpful to keep p53 levels low as has been detected in cancer cells. In addition, HCC with mutant p53 showed poorer clinical responses to TKIs than wild-type p53, which also suggest that strategies to restore p53 levels or improve downstream of p53 pathway might promote the therapeutic sensitivity of cancer cells to TKIs.

Restoring the functionality of p53 in tumors as a therapeutic strategy has mostly met with limited success. Considering the p53 currently undruggable and the subsequent clinical conversion, we focus on the SCD1 activity targeted by p53. The combination of atezolizumab and bevacizumab displayed an acceptable side-effect profile and further improved the objective response rate for systemic treatment of advanced HCC to 33.2%. Therefore, combination therapies are expected to change the landscape of HCC systemic treatment. Unlike other SCD1 inhibitors, aramchol does not result in serious side effects. The efficacy and safety of aramchol has been evaluated in patients with NASH by a randomized, double-blind, placebo-controlled phase2b trial³⁹. The phase 3 program is ongoing and open-label part of the study was reported to meet its objectives (NCT04104321). Aramchol is also safe in nude mice, but single-agent has no significant antitumor effect. Among the TKIs that can induce ferroptosis, donafenib is associated with a significantly longer median OS than sorafenib⁷. We found that aramchol combined with donafenib showed a dramatic inhibition of tumor in vivo. Moreover, HE staining of xenograft sections showed that the tumor necrotic area with the combination treatment was significantly enlarged, mimic with the pathological response of clinical treatment. Our results indicate that the combination of donafenib plus aramchol may especially benefit patients who have wild-type p53 HCC tumors.

In summary, URI keeps low levels of p53 in a TRIM28-MDM2 dependent manner, maintains SCD1 activity and accumulation of MUFAs, and subsequently promotes resistance to TKIs in cancer cell. URI-p53-SCD1 axis mediates resistance of TKIs and may explain why p53-wild type HCC still showed intrinsic resistant to TKIs. Moreover, the combination therapy identified here may represent a promising strategy for the approximately 41% of patients with advanced HCC who have wild-type p53 and high levels of URI/SCD1 (Fig. 8). This study provides a theoretical basis for our subsequent clinical trials.

Patients and Specimens. A total of 219 HCC tissues were obtained from surgical resections of patients in Eastern Hepatobiliary Surgery Hospital (EHBH, Shanghai, China) from 2010 to 2018. The procedure of human specimen collection was approved by the Ethics Committee of EHBH. A Tissue microarray containing 134 HCC tissues from patients with follow-up information (Cohort A) was used for survival analysis. A total of 97 patients at high risk of recurrence that received sorafenib therapy after hepatectomy in EHBH from 2015 to 2018 were included in Cohort B. Recurrence-free survival (RFS) and overall survival (OS) analysis were performed using the Kaplan-Meier method. RFS was defined as the interval between the date of surgery and recurrence. OS was defined as the interval between the date of surgery and death. If recurrence was not diagnosed, patients were censored on the date of the last follow-up.

Cell lines and cell culture. HepG2 (p53-wild), Hep3B (p53-null), Huh7 (p53-Y220C), HCT116 (p53-wild), HT29 (p53-R273H) and PLC/PRF/5 (p53-R249S) cells were purchased from Shanghai Cell resource center of Chinese Academy of Sciences. JHH1 (p53-wild) cells were kindly provided by Haojie Jin at the Shanghai Cancer Institute. HEK293T cells were prechased from the American Type Culture Collection (ATCC). HepG2, Hep3B, Huh7, PLC/PRF/5, JHH1 and HEK293T cells were cultured in DMEM (BasialMedia, with 4.5 g/L glucose, 4mM L-glutamine, and 0.11 g/L pyruvate). HCT116 and HT29 cells were cultured in RPMI-1640 (BasialMedia, with 2 g/L glucose and 300mg/L L-glutamine). Culture medium was supplemented with 10% Foetal Bovine Serum (Biological Industries, 04-001-1ACS), penicillin (10000U/mL)/streptomycin (10000ug/mL) and amphotericin B (25 ug/mL) (BasialMedia). Mycoplasma Elimination Reagent (InvivoGen, 25 mg/mL) was used for the last 3 days before freezing. Fresh cells were recovered every month to maintain cell viability and a low number of cell lines passage. Cells were maintained at 37℃ in an atmosphere of humidified air containing 5% CO₂. Cells were routinely passaged every 2 days and not allowed to grow to confluence.

Real-time PCR analysis. Total RNA was isolated from cells using TRIzol Reagent (ThermoFisher, 15596-018) and RNeasy Mini Kit (Qiagen, 74104). For RT-PCR, mRNA was reverse transcribed into cDNA using the M-MLV system (Promega, M1701). Real-time PCR was performed on LightCycler 480 II system using a SYBR Green PCR Master Mix (Roche), with 18S as a reference control. Primers are listed in Supplementary Table 3.

Cell proliferation, colony formation and viability assay. Cell proliferation was measured using a CCK-8 Cell Counting Kit (Vazyme, A311) according to the manufacturer’s instructions. For colony formation assay, cells were plated in 6-well plates at a density of 3000–5000 cells/well and treated with drugs 24 h later. Colonies were stained with crystal violet after 12–14 days and then counted by using Image J for quantification. For cell viability assessment, cells were cultured in 96-well plates and tested using the CellTiter-Glo luminescent cell viability assay (Promega, G7571).

Flow cytometry. Cells (30000–50000 cells/well) were seeded in 6-well plates and treated with drugs for 48h. For tumor cell death detection, cells were resuspended in 500µl phosphate buffer saline (PBS) and incubated with PropidiumIodide (PI) (Sigma-Aldrich, P4170) for 30 minutes and then analyzed by the flow cytometry performed with a FACSVerse flow cytometer (BD Biosciences). For lipid peroxidation assay, cells were resuspended in 500µl PBS containing 20mM C11-BODIPY 581/591 (ABclonal, RM02821) and incubated for 1h at 37℃ in a cell culture incubator. The signals from both non-oxidized C11 (wave length > 580 nm) and oxidized C11 (wave length 505–550 nm) were monitored. The data were normalized to control samples as shown by relative lipid peroxidation.

Western blotting, co-immunoprecipitation and antibodies. Cells were lysed in RIPA Lysis Buffer (Beyotime, P0013B) on ice and centrifuged at 12,000 rpm 4°C for 15 minutes. Protein concentrations were measured using a BCA protein assay kit (Thermo Scientific, 23225). Blotting was performed following the standard procedures and detected using an Odyssey fluorescence scanner (Li-Cor, Lincoln, Neb). Antibodies used are URI (1:1000, Proteintech, 11277-1-AP,), SREBP1 (1:200, Santa Cruz, sc-13551), SCD1 (1:1000, ABclonal, A16429), FASN (1:1000, ABclonal, A19050), FADS2 (1:1000, ABclonal, A10270), GPX4 (1:1000, ABclonal, A11243), GCLM (1:1000, ABclonal, A11444), GCLC (1:1000, ABclonal, A4499), SLC7A11 (1:1000, Proteintech, 26864-1-AP), ACSL4 (1:1000, Proteintech, 22401-1-AP), MDM2 (1:1000, ABclonal, A13327), Myc-Tag (1:1000, ABclonal, AE010), Flag-Tag (1:1000, ABclonal, AE005), His-Tag (1:1000, ABclonal, AE003), HA-Tag (1:1000, ABclonal, AE008), ACC (1:1000, CST, 3676), TRIM28 (1:1000, Proteintech, 15202-1-AP,), P53 (1:1000, Proteintech, 10442-1-AP), P21 (1:1000, Proteintech, 10355-1-AP), USP5 (1:1000, ABclonal, A4202), USP7 (1:1000, ABclonal, A13564), USP14 (1:1000, ABclonal, A19589), Actin (1:5000, Proteintech, 66009-1-ig), GAPDH (1:5000, ABclonal, AC001).

For co-immunoprecipitation assay, cell lysates were prepared with IP lysis buffer (Beyotime, P0013) and incubated with antibodies overnight at 4°C, followed by the addition of Mag 25K/Protein A/G (Enriching, LM220622) for another 4h. Immune complexes were isolated by centrifugation and proteins were eluted into loading buffer, followed by western blotting.

Luciferase reporter assay. To assess SCD1 gene promotor activity, cells were co-transfected with SCD1 luciferase reporter plasmid, with Renilla control reporter as an internal control. After 48 h, cells were lysed in a Passive Lysis 5X Buffer (Promega, E1941), and the enzymatic activity of luciferase were measured using a Dual-Luciferase Assay kit (Vazyme, DL101-01), according to the manufacturer's protocol.

Chromatin immunoprecipitation (ChIP) assay. ChIP assay was performed using the ChIP Assay Kit (Beyotime, P2078). Briefly, 1×10⁷ cells were cross-linked with formaldehyde/PBS for 10 min at 4°C, and then added glycine solution, reaction at room temperature for 5 min. Subsequently, samples were sonicated in lysis buffer to obtain 200-1,000 bp DNA fragments to be immunoprecipitated with 2 µg of mouse SREBP1 (Santa Cruz, sc13551), mouse P53 (Santa Cruz, sc126) or mouse IgG antibodies. The primer sequences specific to the promoter region of gene are listed in Supplementary Table 3.

Tumor xenograft experiment. All mice were treated according to protocols approved by the Navy Medical University Animal Care Facility and the National Institutes of Health guidelines. For cell line derived xenografts (CDX) models, approximately 1×10⁷ HepG2-Ctrl and HepG2-shURI cells in 0.1 mL PBS were injected subcutaneously into the right flank of 6-week-old male nude mice. Tumor volume was measured using a caliper every 3 days and calculated using the modified ellipsoidal formula: tumor volume (mm³) = (length×width²) /2. After tumour establishment, mice were randomly assigned to once per day treatment. Mice were sacrificed by cervical dislocation when the volume reached 1-1.5 cm³ and tumor tissues were harvested for further study.

Immunohistochemistry and Scoring. 10% formalin-fixed paraffin-embedded tissues were sectioned (4µm) and stained with hematoxylin and eosin (H&E) for histological analysis or used for immunohistochemistry (IHC). For IHC, endogenous peroxidases were inactivated by 3% hydrogen peroxide and nonspecific signals were blocked by 1% BSA. Sections were incubated with primary antibody at 4°C overnight, HRP-conjugated secondary antibody at 37°C for 1h, and subsequently stained with DAB substrate. Counterstaining was performed with hematoxylin, and mounted with a mounting medium. Antibodies are listed in Supplementary Table 4.

H-score assay was performed, ranging from 0 to 300. URI and SCD1 staining was scored according to four categories: 0 for ‘no staining’, 1 + for ‘light staining’, 2 + for ‘intermediate staining’; and 3 + for ‘dark staining’. The percentage of cells at different staining intensities was determined by visual assessment, with the score calculated using the formula 1 × (% of 1 + cells) + 2 × (% of 2 + cells) + 3 × (% of 3 + cells).The outcome-based discriminatory threshold IHC H-score for this analysis was set at 200 and samples were then classified as either low (H-score < 200) or high (H-score ≥ 200) for URI and SCD1 protein expression.

Multiplex immunohistochemistry/immunofluorescence (mIHC/IF) staining. To investigate the expression of URI and SCD1 in patient tissues, multiplex immunohistochemistry/immunofluorescence (mIHC/IF) was conducted using a tissue microarray. Briefly, endogenous peroxidases were blocked by 3% hydrogen peroxide and nonspecific signals were blocked by 1% BSA. The slides were immersed in Tris-EDTA buffer to perform heat-induced antigen retrieval. The slides were then incubated with anti-URI (Proteintech, 11277-1-AP), anti-SCD1 (Abcam, ab236868) antibodies. Signal detection were performed using a TSA kit (AKOYA) and nuclei were counterstained with DAPI. Slides were imaged and scanned using a slice scanner (Pannoramic MIDI: 3Dhistech, Hungary) and images were analyzed via HALO 2.0 Area Quantification algorithm (Indica Labs; Corrales, NM), at Nanjing Freethinking Biotechnology Co., Ltd. (China).

RNA sequencing. Total RNA was isolated using a RNeasy mini kit (Qiagen, Germany) from three biological replicates. TruSeq Stranded Total RNA Sample Preparation kit (Illumina, USA) was used to prepare the strand-specific libraries following the manufacturer’s instructions. Briefly, using the oligo (dT) beads to enrich mRNA. Following purification, the mRNA is fragmented into small pieces using divalent cations under 86℃ for 6 min. The cleaved RNA fragments are copied into first strand cDNA using reverse transcriptase and random primers. This is followed by second strand cDNA synthesis using DNA Polymerase I and RNase H. These cDNA fragments then go through an end repair process, the addition of a single ‘A’ base, and then ligation of the adapters. Products are then purified and enriched with PCR to create the final cDNA library. Purified libraries were quantified by Qubit 2.0 Fluorometer (Life Technologies, USA) and validated by Agilent 2100 bioanalyzer (Agilent Technologies, USA) to confirm the insert size and calculate the mole concentration. Cluster was generated by cBot with the library diluted to 10 pM and then were sequenced on the Illumina NovaSeq 6000 (Illumina, USA). The library construction and sequencing was performed at Shanghai Biotechnology Corporation.

For each sample, 33-95M RNA-seq clean reads were obtained that mapped to homo sapiens using HISAT2. Sequencing read counts were calculated using Stringtie (v.1.3.0). Then expression levels from different samples were normalized by the Trimmed Mean of M values (TMM) method. The normalized expression levels of different samples were converted to FPKM (Fragments Per Kilobase of transcript per Million mapped fragments). The edgeR package of R was used to analyze the difference between intergroup gene expression. The P-value threshold was determined by controlling the FDR (False Discovery Rate) with the Benjamini algorithm. The corrected P-value is called the q-value. Differentially expressed genes (DEGs) were defined as transcripts with a fold change in expression level (according to the FPKM value) greater than 2.0 and a q-value less than 0.05. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and GO enrichment analysis was performed with the clusterProfiler package of R and the enrichment criteria including a q-value < 0.05. Heatmaps of specific genes were generated using the pheatmap package of R. Gene set enrichment analysis (GSEA) was performed using GSEA software.

CUT&Tag library construction. HepG2 cells were treated with 5µM Nutlin-3 for 4h. Dead cells were identified by trypan blue staining. Ensure that the cell viability is greater than 95% before detection. Hyperactive™ In-Situ ChIP Library Prep Kit (TD901-TD902, Vazyme Biotech) for Illumina was used to perform CUT&Tag assay⁴⁷. Briefly, concanavalin A-coated magnetic beads (ConA beads) were added to resuspended cells and incubated at room temperature to bound cells. Non-ionic detergent Digitonin was used to permeate cell membrane. Then, mouse p53 antibody (sc126, Santa), secondary antibody and the Hyperactive pA-Tn5 Transposase were incubated with the cells that were bounded by ConA beads in order. Therefore, the Hyperactive pA-Tn5 Transposase can exactly cut off the DNA fragments that were bound with target protein. In addition, the cut DNA fragments can be ligated with P5 and P7 adaptors by Tn5 transposase and the libraries were amplified by PCR with the P5 and P7 primers. The purified PCR products were evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Finally, these libraries were sequenced on the Illumina NovaSeq6000 platform and 150 bp paired-end reads were generated for the following analysis by Shanghai OEbiotech company.

Lipidomics by liquid chromatography coupled to mass spectrometry (LC-MS). Lipid extraction and mass spectrometry-based lipid detection were performed by BioNovaGene (Suzhou, China). Briefly, take an appropriate amount of sample in a 2 mL EP tube, add 750 µL of Chloroform methanol mixed solution (2:1), vortex for 30 s. Add 2 steel beads, put them into the tissue grinder, and grind at 50Hz for 60s, repeated 2 times. Put on the ice for 40 min, add 190 µL H₂O, vortex for 30 s, put on the ice for 10min. Centrifuged at 12000 rpm for 5min at room temperature and transfer 300 µL lower layer fluid into a new centrifuge tube. Then add 500 µL of Chloroform methanol mixed solution (2:1), vortex for 30 s. Centrifuged at 12000 rpm for 5min at room temperature and transfer 400µL lower layer fluid into the same centrifuge tube above. Samples were concentrated to dry in vacuum. Dissolve samples with 200µL isopropannol, and the supernatant was filtered through 0.22 µm membrane to obtain the prepared samples for LC-MS.

Chromatographic separation was accomplished with an ACQUITY UPLC® BEH C18 (100 × 2.1 mm, 1.7 µm, Waters) column maintained at 50°C. The temperature of the autosampler was 8°C. Gradient elution of analytes was carried out with acetonitrile:water = 60:40(0.1%formic acid + 10mM ammonium formate)(C) and isopropanol: acetonitrile = 90:10(0.1%formic acid + 10mM ammonium formate)(D) at a flow rate of 0.25 mL/min. Injection of 2 µL of each sample was done after equilibration. An increasing linear gradient of solvent C (v/v) was used as follows: 0–5 min, 70%-57% C; 5-5.1 min, 57%-50% C; 5.1–14 min, 50%-30% C; 14-14.1 min, 30% C; 14.1–21 min, 30%-1% C; 21-24min, 1% C; 24-24.1 min, 1%-70% C; 24.1–28 min, 70% C. The electrospray tandom mass spectrometry (ESI-MSn) experiments were used with the spray voltage of 3.5 and − 2.5 kV in positive and negative modes, respectively. Sheath gas and auxiliary gas were set at 30 and 10 arbitrary units, respectively. The capillary temperature was 325°C. The Orbitrap analyzer scanned over a mass range of m/z 150–2000 for full scan at a mass resolution of 35,000. Data dependent acquisition (DDA) MS/MS experiments were performed with HCD scan. The normalized collision energy was 30 eV. Dynamic exclusion was implemented to remove some unnecessary information in MS/MS spectra.

Targeted medium- and long-chain fatty acids quantitation. LC-MS analysis of medium- and long-chain fatty acids in HepG2-Ctrl and HepG-shURI cells was performed by Shanghai Sensichip Infotech Co. (Shanghai, China). Briefly, all the samples were mixed with 600 µL 50% acetonitrile, vortex for 1min, and centrifuged at 12000 rpm, 4℃ for 15 min. Then, 200ul of supernatant was added with 100 µL 200 mM 3-NPH and 100 µL 120 mM EDC (6% pyridine, 400 ng/mL acetic acid-D3) and was incubated at 40℃ for 1min. Samples were centrifuged at 12000 rpm, 4℃ for 15 min and their supernatants were transferred to tubes for LC-MS analysis. LC-MS data was acquired on AB SCIEX 5500 QQQ (Applied Biosystems, Foster City, CA, USA) mass spectrometer coupled with high-performance liquid chromatography (HPLC) system ACQUITY UPLC. (Waters, Milford, MA, USA). The column for chromatographic separation was a ACQUITY UPLC BEH Amide Column (2.1 x 100 mm, 1.7 µm). For determination of relative metabolite abundances, the total abundances were normalized to an internal standard (acetic acid-D3) and the weight for cell extracts.

Proteins LC-MS/MS Analysis. For in-gel tryptic digestion, gel pieces were destained in 50 mM NH4HCO3 in 50% acetonitrile. Gel pieces were dehydrated with 100 µL of 100% acetonitrile for 5 min, the liquid removed, and the gel pieces rehydrated in 10 mM dithiothreitol and incubated at 37°C for 60 min. Gel pieces were again dehydrated in 100% acetonitrile, liquid was removed and gel pieces were rehydrated with 55 mM iodoacetamide. Samples were incubated at room temperature, in the dark for 45 min. Gel pieces were washed with 50 mM NH₄HCO₃ and dehydrated with 100% acetonitrile. Gel pieces were rehydrated with 10 ng/µL trypsin resuspended in 50 mM NH₄HCO₃ on ice for 1 h. Excess liquid was removed and gel pieces were digested with trypsin at 37°C overnight. Peptides were extracted with 50% acetonitrile/5% formic acid, followed by 100% acetonitrile. Peptides were dried to completion and resuspended in 2% acetonitrile/0.1% formic acid.

LC-MS/MS Analysis

The tryptic peptides were dissolved in 0.1% formic acid (solvent A), directly loaded onto a home-made reversed-phase analytical column (15-cm length, 75 µm i.d.). The gradient was comprised of an increase from 6–23% solvent B (0.1% formic acid in 98% acetonitrile) over 16 min, 23–35% in 8 min and climbing to 80% in 3 min then holding at 80% for the last 3 min, all at a constant flow rate of 400 nl/min on an EASY-nLC 1000 UPLC system.

The peptides were subjected to NSI source followed by tandem mass spectrometry (MS/MS) in Q ExactiveTM Plus (Thermo) coupled online to the UPLC. The electrospray voltage applied was 2.0 kV. The m/z scan range was 350 to 1800 for full scan, and intact peptides were detected in the Orbitrap at a resolution of 70,000. Peptides were then selected for MS/MS using NCE setting as 28 and the fragments were detected in the Orbitrap at a resolution of 17,500. A data-dependent procedure that alternated between one MS scan followed by 20 MS/MS scans with 15.0s dynamic exclusion. Automatic gain control (AGC) was set at 5E4.

The resulting MS/MS data were processed using Proteome Discoverer 1.3. Trypsin/P (or other enzymes if any) was specified as cleavage enzyme allowing up to 2 missing cleavages. Mass error was set to 10 ppm for precursor ions and 0.02 Da for fragment ions. Carbamidomethyl on Cys were specified as fixed modification and oxidation on Met was specified as variable modification. Peptide confidence was set at high, and peptide ion score was set > 20.

DNA electrophoresis. 10× TAE buffer were diluted into 1× TAE for working concentration. 2% agarose gel was made using 1× TAE buffer with additional 10µl SolarRed (10000×, Solarbio, G5560). The separation of DNA fragments were achieved in 2% agarose gels in 1× TAE buffer using a DNA marker (Vazyme, DL5000) at 120V for 25 minutes.

Recombinant Myc-URI and His-TRIM28 production. The recombinant Myc-URI and His-TRIM28 were obtained from AtaGenix Laboratories Co. (Wuhan, China). Briefly, the URI and TRIM28 cDNA was obtained from the National Center of Biotechnology Information. The cDNA was cloned in a plasmidic expression vector (pET28b) harboring a histidine tag. After DNA sequence verification, the plasmid was transferred into Escherichia coli. The bacteria starter was obtained by incubation at 37°C for 4 h. Recombinant protein was obtained using a Ni-NTA Superflow column (Qiagen, Hilden, Germany).

Data collection. Transcriptome data were downloaded from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), and the dataset of GSE96794, GSE143233 and GSE25097 was used in this study.

Statistical analyses. All the statistical analyses were performed using GraphPad Prism8 software (GraphPad Software, Inc.). Two tailed t-test and Mann-Whitney test were used to compare shURI vs. Ctrl groups. ANOVA models were used to compare continuous outcomes across multiple experimental groups. Kaplan- Meier analysis with log-rank tests was used to determine disease-free survival (DFS) and overall survival (OS). The data represent mean values of at least three independent experiments. Error bars represent mean ± SEM or median ± SEM, as appropriate. Statistical significance was set at p < 0.05.

Availability of data and materials

The data and materials used to support the findings of this study are available from the corresponding author upon request.

Ethics approval and consent to participate

This study was approved by the Committee on Ethics of Medicine, Eastern Hepatobiliary Surgery Hospital, Shanghai, China. All animals were kept in a pathogen-free environment and fed ad lib. The procedures for care and use of animals were performed according to the Naval Medical University Animal Care Facility and the National Institutes of Health guidelines. All applicable institutional and governmental regulations concerning the ethical use of animals were followed.

Consent for publication

The authors declare that they agree to submit the article for publication.

Competing interests

The authors disclose no competing interests.

Authors’ Contributions:

Hongyang, Wang, Liwei Dong, and Lu Wang conceived and designed the study; Zhiwen Ding, Yufei Pan, Taiyu Shang, Tianyi Jiang, Yunkai Lin, Chun Yang, Shujie Pang, Xiaowen Cui, Yixiu Wang, Xiao fan Feng, Mengyou Xu, Mengmiao Pei, Yibin Cheng, Xin Li performed the experiments; Zhiwen Ding, Yufei Pan, Taiyu Shang analyzed the date. Zhiwen Ding, Yufei Pan, Taiyu Shang and Liwei Dong prepared the manuscript; All authors read and approved the final manuscript.

Acknowledgements:

We thank Dan Cao for preparing of paraffin encrusting; Yuyao Zhu for sectioning of paraffin encrusting; Haojie Jin for providing JHH1 cell lines.This work was supported from the National Natural Science Foundation of China (No. 81802311; 82172895; 81874182; 82073228), the Youth Program from Shanghai Municipal Health Commission (No. 20184Y0125), Shanghai Shenkang clinical science and technology innovation project (No. SHDC2020CR2011A, SHDC2020CR3031B, SHDC2020CR5007).

Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71, 209-249, doi:10.3322/caac.21660 (2021).
Vogel, A., Meyer, T., Sapisochin, G., Salem, R. & Saborowski, A. Hepatocellular carcinoma. Lancet 400, 1345-1362, doi:10.1016/S0140-6736(22)01200-4 (2022).
Reig, M. et al. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. J Hepatol 76, 681-693, doi:10.1016/j.jhep.2021.11.018 (2022).
Llovet, J. M. et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 359, 378-390, doi:10.1056/NEJMoa0708857 (2008).
Bruix, J. et al. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 389, 56-66, doi:10.1016/S0140-6736(16)32453-9 (2017).
Kudo, M. et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 391, 1163-1173, doi:10.1016/S0140-6736(18)30207-1 (2018).
Qin, S. et al. Donafenib Versus Sorafenib in First-Line Treatment of Unresectable or Metastatic Hepatocellular Carcinoma: A Randomized, Open-Label, Parallel-Controlled Phase II-III Trial. J Clin Oncol 39, 3002-3011, doi:10.1200/JCO.21.00163 (2021).
Finn, R. S. et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. N Engl J Med 382, 1894-1905, doi:10.1056/NEJMoa1915745 (2020).
Tang, W. et al. The mechanisms of sorafenib resistance in hepatocellular carcinoma: theoretical basis and therapeutic aspects. Signal Transduct Target Ther 5, 87, doi:10.1038/s41392-020-0187-x (2020).
Stockwell, B. R. Ferroptosis turns 10: Emerging mechanisms, physiological functions, and therapeutic applications. Cell 185, 2401-2421, doi:10.1016/j.cell.2022.06.003 (2022).
Lei, G., Zhuang, L. & Gan, B. Targeting ferroptosis as a vulnerability in cancer. Nat Rev Cancer 22, 381-396, doi:10.1038/s41568-022-00459-0 (2022).
Li, D. & Li, Y. The interaction between ferroptosis and lipid metabolism in cancer. Signal Transduct Target Ther 5, 108, doi:10.1038/s41392-020-00216-5 (2020).
Hoy, A. J., Nagarajan, S. R. & Butler, L. M. Tumour fatty acid metabolism in the context of therapy resistance and obesity. Nat Rev Cancer 21, 753-766, doi:10.1038/s41568-021-00388-4 (2021).
Kubota, C. S. & Espenshade, P. J. Targeting Stearoyl-CoA Desaturase in Solid Tumors. Cancer Res 82, 1682-1688, doi:10.1158/0008-5472.CAN-21-4044 (2022).
Ma, M. K. F. et al. Stearoyl-CoA desaturase regulates sorafenib resistance via modulation of ER stress-induced differentiation. J Hepatol 67, 979-990, doi:10.1016/j.jhep.2017.06.015 (2017).
Budhu, A. et al. Integrated metabolite and gene expression profiles identify lipid biomarkers associated with progression of hepatocellular carcinoma and patient outcomes. Gastroenterology 144, 1066-1075 e1061, doi:10.1053/j.gastro.2013.01.054 (2013).
Tummala, K. S. et al. Inhibition of de novo NAD(+) synthesis by oncogenic URI causes liver tumorigenesis through DNA damage. Cancer Cell 26, 826-839, doi:10.1016/j.ccell.2014.10.002 (2014).
Gomes, A. L. et al. Metabolic Inflammation-Associated IL-17A Causes Non-alcoholic Steatohepatitis and Hepatocellular Carcinoma. Cancer Cell 30, 161-175, doi:10.1016/j.ccell.2016.05.020 (2016).
Buren, S. et al. Regulation of OGT by URI in Response to Glucose Confers c-MYC-Dependent Survival Mechanisms. Cancer Cell 30, 290-307, doi:10.1016/j.ccell.2016.06.023 (2016).
Wan, Z. H. et al. RPB5-Mediating Protein Promotes Cholangiocarcinoma Tumorigenesis and Drug Resistance by Competing With NRF2 for KEAP1 Binding. Hepatology 71, 2005-2022, doi:10.1002/hep.30962 (2020).
Qin, S. et al. Apatinib as second-line or later therapy in patients with advanced hepatocellular carcinoma (AHELP): a multicentre, double-blind, randomised, placebo-controlled, phase 3 trial. Lancet Gastroenterol Hepatol 6, 559-568, doi:10.1016/S2468-1253(21)00109-6 (2021).
Zhu, A. X. et al. SEARCH: a phase III, randomized, double-blind, placebo-controlled trial of sorafenib plus erlotinib in patients with advanced hepatocellular carcinoma. J Clin Oncol 33, 559-566, doi:10.1200/JCO.2013.53.7746 (2015).
Yang, W. S. et al. Peroxidation of polyunsaturated fatty acids by lipoxygenases drives ferroptosis. Proc Natl Acad Sci U S A 113, E4966-4975, doi:10.1073/pnas.1603244113 (2016).
Magtanong, L. et al. Exogenous Monounsaturated Fatty Acids Promote a Ferroptosis-Resistant Cell State. Cell Chem Biol 26, 420-432 e429, doi:10.1016/j.chembiol.2018.11.016 (2019).
Batchuluun, B., Pinkosky, S. L. & Steinberg, G. R. Lipogenesis inhibitors: therapeutic opportunities and challenges. Nat Rev Drug Discov 21, 283-305, doi:10.1038/s41573-021-00367-2 (2022).
Cao, Y. Adipocyte and lipid metabolism in cancer drug resistance. J Clin Invest 129, 3006-3017, doi:10.1172/JCI127201 (2019).
Jiang, X., Stockwell, B. R. & Conrad, M. Ferroptosis: mechanisms, biology and role in disease. Nat Rev Mol Cell Biol 22, 266-282, doi:10.1038/s41580-020-00324-8 (2021).
AM, A. L., Syed, D. N. & Ntambi, J. M. Insights into Stearoyl-CoA Desaturase-1 Regulation of Systemic Metabolism. Trends Endocrinol Metab 28, 831-842, doi:10.1016/j.tem.2017.10.003 (2017).
Zhang, X. et al. Liver X receptor activation increases hepatic fatty acid desaturation by the induction of SCD1 expression through an LXRalpha-SREBP1c-dependent mechanism. J Diabetes 6, 212-220, doi:10.1111/1753-0407.12081 (2014).
Benhamed, F. et al. The lipogenic transcription factor ChREBP dissociates hepatic steatosis from insulin resistance in mice and humans. J Clin Invest 122, 2176-2194, doi:10.1172/JCI41636 (2012).
Mirza, A. et al. Global transcriptional program of p53 target genes during the process of apoptosis and cell cycle progression. Oncogene 22, 3645-3654, doi:10.1038/sj.onc.1206477 (2003).
Sabapathy, K. & Lane, D. P. Therapeutic targeting of p53: all mutants are equal, but some mutants are more equal than others. Nat Rev Clin Oncol 15, 13-30, doi:10.1038/nrclinonc.2017.151 (2018).
Hafner, A., Bulyk, M. L., Jambhekar, A. & Lahav, G. The multiple mechanisms that regulate p53 activity and cell fate. Nat Rev Mol Cell Biol 20, 199-210, doi:10.1038/s41580-019-0110-x (2019).
Wang, C. et al. MDM2 interaction with nuclear corepressor KAP1 contributes to p53 inactivation. EMBO J 24, 3279-3290, doi:10.1038/sj.emboj.7600791 (2005).
Dayal, S. et al. Suppression of the deubiquitinating enzyme USP5 causes the accumulation of unanchored polyubiquitin and the activation of p53. J Biol Chem 284, 5030-5041, doi:10.1074/jbc.M805871200 (2009).
Li, M. et al. Deubiquitination of p53 by HAUSP is an important pathway for p53 stabilization. Nature 416, 648-653, doi:10.1038/nature737 (2002).
Zhou, P. et al. Matrine derivate MASM protects murine MC3T3-E1 osteoblastic cells against dexamethasone-induced apoptosis via the regulation of USP14/p53. Artif Cells Nanomed Biotechnol 47, 3720-3728, doi:10.1080/21691401.2019.1664563 (2019).
Mita, P. et al. URI Regulates KAP1 Phosphorylation and Transcriptional Repression via PP2A Phosphatase in Prostate Cancer Cells. J Biol Chem 291, 25516-25528, doi:10.1074/jbc.M116.741660 (2016).
Ratziu, V. et al. Aramchol in patients with nonalcoholic steatohepatitis: a randomized, double-blind, placebo-controlled phase 2b trial. Nat Med 27, 1825-1835, doi:10.1038/s41591-021-01495-3 (2021).
Zhang, J. et al. RMP promotes venous metastases of hepatocellular carcinoma through promoting IL-6 transcription. Oncogene 34, 1575-1583, doi:10.1038/onc.2014.84 (2015).
Tesfay, L. et al. Stearoyl-CoA Desaturase 1 Protects Ovarian Cancer Cells from Ferroptotic Cell Death. Cancer Res 79, 5355-5366, doi:10.1158/0008-5472.CAN-19-0369 (2019).
Yi, J., Zhu, J., Wu, J., Thompson, C. B. & Jiang, X. Oncogenic activation of PI3K-AKT-mTOR signaling suppresses ferroptosis via SREBP-mediated lipogenesis. Proc Natl Acad Sci U S A 117, 31189-31197, doi:10.1073/pnas.2017152117 (2020).
Levine, A. J. p53: 800 million years of evolution and 40 years of discovery. Nat Rev Cancer 20, 471-480, doi:10.1038/s41568-020-0262-1 (2020).
Guichard, C. et al. Integrated analysis of somatic mutations and focal copy-number changes identifies key genes and pathways in hepatocellular carcinoma. Nat Genet 44, 694-698, doi:10.1038/ng.2256 (2012).
Calderaro, J. et al. Histological subtypes of hepatocellular carcinoma are related to gene mutations and molecular tumour classification. J Hepatol 67, 727-738, doi:10.1016/j.jhep.2017.05.014 (2017).
Hassin, O. & Oren, M. Drugging p53 in cancer: one protein, many targets. Nat Rev Drug Discov, doi:10.1038/s41573-022-00571-8 (2022).
Kaya-Okur, H. S. et al. CUT&Tag for efficient epigenomic profiling of small samples and single cells. Nat Commun 10, 1930, doi:10.1038/s41467-019-09982-5 (2019).

There is NO Competing Interest.

Download PDF

Journal Publication

published 07 Oct, 2023

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

URI alleviates tyrosine kinase inhibitors-induced ferroptosis by reprogramming lipid metabolism in p53 wild-type liver cancers.

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

URI promotes resistance to TKIs-induced ferroptosis via SCD1

Combined higher expression of URI and SCD1 correlated with worsen prognosis and sorafenib-resistance in HCC

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1