Inhibition of AMPK activity by TRIM11 facilitates cell survival of hepatocellular carcinoma under metabolic stress

AMP-responsive protein kinase (AMPK) is a master nutrient and energy sensor, keeping the cellular energy homeostasis during metabolic stress.1, 2 Loss of AMPK or deregulation of its activity has been detected in multiple human cancers including hepatocellular carcinoma (HCC).3, 4 However, the underlying molecular mechanism of dysregulation of AMPK activity is still largely unclear. We previously demonstrated that an E3 ubiquitin-ligase, TRIM11, is a key player during various stress conditions and tumourigenesis.5, 6 In this report, we clarified TRIM11 as a new mechanism for negatively regulating AMPK activity during glucose starvation, and suggest that TRIM11-AMPK axis is crucial for HCC survival and progression. We determined that TRIM11 is liked with metabolic reprogramming and its expression was induced upon glucose starvation (Figure 1A and Figure S1A–C). As glucose deprivation would be detrimental for tumour cell survival, we examined the role of TRIM11 in this process and found it positively regulated tumour cells viability upon glucose deprivation (Figure 1B,C and Figure S1D–F), and similar results could be found in the role of TRIM11 in vivo (Figure 1D–F). Gene set enrichment analysis (GSEA) revealed that several metabolic pathways gene set were significantly enriched in the TRIM11_low expression HCC patients (Figure 1G and Figure S1G and Table S1). Then, we indeed found that TRIM11 could regulate glucose metabolism in the HCC cells and patient tissues (Figure 1H–J and Figure S1H). Together, these results suggest that the upregulation of TRIM11 serves as a novel protecting mechanism to avoid HCC cells death upon metabolic stress. Next, we analysed the protein interaction network of TRIM11 and demonstrated that TRIM11 mainly interacts with AMPKβ2 and also shows a weak interaction with AMPKα and AMPKγ2 subunits (Figure 2A and Figure S2A–D). AMPK consists of catalytic α, regulatory β and γ subunits, and all of these subunits are closely linked to the activating AMPK,7, 8 implying TRIM11 may regulate AMPK activity by selective modulation of AMPK regulatory subunits. As expected, TRIM11 significantly enhanced ubiquitination level of AMPKβ2 and accelerated its protein degradation (Figure 2B,C and Figure S2E), and negatively controlling AMPK signaling activity in the HCC cells and tissues (Figure 2D,E and Figure S2F–L). To map the binding domain between TRIM11 and AMPKβ2, we constructed their corresponding deletion mutants (Figure S2M,N), and determined that the RING domain of TRIM11 is required for the interaction with AMPKβ2 (Figure 2F). As we previously reported,5, 6 a TRIM11 mutant (TRIM11-2CA) that lost its ubiquitination activity was used (Figure S2M). TRIM11-2CA displayed a reduced interaction with AMPKβ2, decreased its ubiquitination, increased stability of AMPKβ2, and also impaired its function in controlling AMPK activity and tumour cell viability (Figure 2G–K). Meantime, we found that the β-CTD of AMPKβ2 was crucial for its interaction with TRIM11 and that the AMPKβ2-K260R mutant could suppress the ubiquitination level of AMPKβ2 and enhance its stability (Figure 2L,M and Figure S2O), suggesting that TRIM11 directly targets K260 of AMPKβ2 to mediate its degradation. Collectively, these data demonstrate that TRIM11 destabilizes AMPKβ2 through directly promoting its protein degradation, which is required for its effects on AMPK activity and HCC cell survival. AMPK can be activated upon glucose deprivation, which results in starvation-induced autophagy, this trigging autophagic cell death,7, 9 suggesting that TRIM11 may act as an upstream regulator of AMPK/autophagy pathway. We analysed that autophagy activation was inversely linked with TRIM11 level in HCC (Figure S3A). Then, we observed the formation of autophagosomes and evaluated the localisation of LC3B, a marker protein for autophagosomes,10 as well as the expression level of the autophagy markers (LC3-II and p62), confirming that TRIM11 negatively regulates the induction of autophagy during metabolic stress (Figure 3A–F and Figure S3B,C). While TRIM11-2CA was less effective in regulating the autophagy compared with TRIM11-WT (Figure S3D). In addition, autophagic flux inhibitor chloroquine (CQ) could abrogate the effective knockdown of TRIM11-mediated activation of autophagy (Figure S3E). Of note, TRIM11-mediated autophagy regulation and tumour cell survival were diminished in AMPK-knockdown cells or treated with AMPK inhibitor compound C (Figure S3F–N), revealing that TRIM11-mediated negative regulation of autophagy depends on AMPK. Similarly, we confirmed this conclusion in MEFs (Figure 3G–J and Figure S3O,P). Together, these data demonstrate that TRIM11 negatively regulates autophagy depending on its controlling AMPK activity. Next, we explored whether AMPK activity is required for TRIM11-mediated protection of HCC cell survival. As expected, TRIM11 indeed significantly enhanced HCC cell viability upon the AMPK activator AICAR treatment, but TRIM11-mediated HCC cell survival was abrogated in the AMPK-knockdown cells (Figure S4A,B). Consistently, this effect was also confirmed in MEFs (Figure 4A). And TRIM11 also obviously abrogate metformin-mediated HCC therapeutic effects in vivo (Figure 4B,C and Figure S4C–E),suggesting that inhibiting TRIM11-AMPK axis helps effective treatment of HCC. For its clinical significance in HCC patients, we found that the expression level of TRIM11 negatively correlated with AMPKβ2 and pAMPK (Figure 4D and Figure S4F,G). In addition, we found an increased TRIM11 staining intensity in HCC tissues relative to adjacent counterparts (Figure S5A,B), and the transcription levels of TRIM11 were also significantly upregulated in many cancerous tissues (Figure S5C–J). Of note, survival analysis showed that TRIM11High predicts worse overall survival than those with TRIM11Low (Figure 4E,F). Similar results were found in the TCGA cohort (Figure 4G and Figure S5K–N). Consistently, TCGA pan-cancer cohort also revealed TRIM11High in the tumour patients displayed shorten overall survival (OS) and disease/progression-free survival time (DFS/PFS) compared with TRIM11Low tumour patients (Figure 4H and Figure S5O,P). Thus, these results demonstrate that upregulation of TRIM11 promotes tumour progression and can be served as a crucial indicator for poor prognosis in the pan-cancer cohort. Together, our data demonstrated that TRIM11 was found to be significantly induced to mediate cellular metabolic reprogramming and inhibited the stimulation of autophagy via directly targeting AMPK signaling pathway to promote HCC cell survival (Figure 4I). Our study highlights a crucial promoter implicated in metabolic stress, the TRIM11-AMPK axis, which will provide the theoretical basis and intervention targets for developing more effective means of clinical treatment of HCC. We thank Dr. Yi Xu at University of Pennsylvania for critical suggestions and reading of the manuscript. This work was supported by National Key R&D Program of China (2020YFA0710802), National Natural Science Foundation of China (31701005, 81874174, 82073190 and 81572832), SIAT Innovation Program for Excellent Young Researchers (201801), SIAT-GHMSC Biomedical Laboratory for Major Human Diseases, Shanghai Rising-Star Program (18QA1402600), Shanghai Municipal Commission of Health and Family Planning (2018YQ12) and School of Medicine, Shanghai Jiao Tong University (Excellent Youth Scholar Initiation GrantNumbers 17XJ11015 and 18XJ11006). The authors declare that there is no conflict of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.