IntroductionThe von Hippel-Lindau (VHL) tumor suppressor protein (pVHL) exerts its anti-tumorigenic functions through two distinct mechanisms. First, it acts as a core component of an E3 ubiquitin ligase complex, targeting hydroxylated substrates including HIFα [1], ZHX2 [2], and SFMBT1 [3] for proteasomal degradation. Second, pVHL serves as a scaffolding protein that negatively regulates oncogenic signaling pathways, including Akt and NF-κB pathways [4, 5]. Despite these critical roles, pVHL expression is frequently downregulated in cancers harboring wild-type VHL. However, the underlying mechanism responsible for this downregulation remains poorly understood.Aberrant glucose metabolism is a hallmark of human cancers, characterized by the upregulations of key glycolytic enzymes [6] and glucose transporters such as GLUT1 and GLUT3 [7, 8]. These metabolic dysregulations result in increased intracellular glucose levels and enhanced glucose utilization in solid tumors, which supports the energetic and biosynthetic demand of solid tumors, driving tumor progression and contributing to therapeutic resistance [9, 10]. Emerging evidence suggests that hypoxia-inducible factor alpha (HIFα) plays a vital role in modulating metabolic reprogramming, activating oncogenic signaling pathways, and promoting cancer stem cell (CSC) phenotypes as well as therapeutic resistance [11]. Furthermore, hyperactivation of the Akt pathway, which is commonly observed in human cancers, regulates metabolism by inducing or activating glucose transporters and key metabolic enzymes, thereby facilitating tumor progression [12, 13]. Given its regulatory effects on substrates such as HIFα and Akt, pVHL has been implicated in the modulation of cancer metabolism. However, it remains unclear whether and how aberrant metabolism affects pVHL turnover and its suppressive activity in human cancers.In this study, we uncover a novel mechanism of metabolic regulation affecting pVHL turnover in pancreatic ductal adenocarcinoma (PDAC), colorectal cancer, and ovarian cancer cells. We demonstrate that the activation of AMP-activated protein kinase (AMPK) is crucial for stabilizing pVHL under energy stresses conditions induced by glucose starvation, 2-deoxy-D-glucose (2-DG), and metformin. Mechanistically, AMPK directly phosphorylates BAP1, enhancing its deubiquitinase activity towards pVHL, which in turn suppresses tumor progression both in vitro and in vivo. Importantly, a strong positive correlation between p-AMPK, pSer123-BAP1, and pVHL levels was observed in PDAC, colorectal cancer, and ovarian cancer specimens. Collectively, our findings suggest that glucose-mediated energy homeostasis is an upstream regulator of the tumor-suppressive activity of the BAP1-pVHL axis. This highlights the therapeutic potential of targeting this axis for the management of cancers characterized by downregulated wild-type VHL and aberrant glucose metabolism.Materials and methodsCell culture, plasmids, and antibodiesHEK293T, PANC-1, BxPC3, AsPC1, MIAPaCa-2 cells and other lines were obtained from ATCC (American Type Culture Collection). All cell lines were mycoplasma-free and authenticated by short tandem repeat DNA profiling analysis. Pan02 cells were kindly shared by Dr Jihui Hao (Tianjin Medical University Cancer Institute and Hospital). The AMPKα wild-type and AMPKα DKO MEFs cells were kindly shared by Dr Wei Liu (Zhejiang University). HEK293T, PANC-1, Pan02, MIAPaCa-2 cells were cultured in DMEM (Gibco) medium supplemented with 10% FBS (TransGen Biotech). AsPC1 and BxPC3 cells were cultured in RPMI-1640 (Gibco) medium supplemented with 10% FBS (TransGen Biotech). All cells were maintained in a humidified cell incubator with 5% CO2 at 37 °C.BAP1, VHL, AMPKα1, and AMPKα2 were cloned into pIRES-EGFP, pLV.3-FLAG, pCMV-HA, pET28a and pGEX4T-1 vectors, respectively. All site mutants were generated by site-directed mutagenesis and identified by sequencing. BAP1, VHL, and AMPKα short hairpin RNAs (shRNAs) were cloned following the protocol-pLKO.1 -TRC cloning vector from Addgene. Sequences for VHL shRNA are 5′-CCCTATTAGATACACTTCTTA-3′. The sequences for AMPKα shRNA are 5′- ATGATGTCAGATGGTGAATTT-3′, which could specifically deplete both homo sapiens and Mus musculus AMPKα1/α2. The sequences for homo sapiens BAP1 shRNA#1 and #2 are 5′-CGTCCGTGATTGATGATGATA-3′ and 5′-CCACAACTACGATGAGTTCAT-3′. The sequences for Mus musculus BAP1 shRNA#3 and #4 are 5′-CCACAACTATGACGAGTTTAT-3′ and 5′-CGTCTGTGATTGATGATGATA-3′.Antibodies anti-pVHL (68547, dilution: 1:1000), anti-AMPKα (D5A2) (5831, dilution: 1:1000), anti-phospho-AMPK substrate (5759, dilution: 1:1000), anti-phospho-AMPKα (Thr172) (2535, dilution: 1:1000), anti-phospho-ACC1 (Ser79) (11818, dilution: 1:1000), anti-ACC1 (4190, dilution: 1:1000), anti-Akt (pan) (2920, dilution: 1:1000) and anti-phospho-Akt (Thr308) (13038, dilution: 1:1000) were purchased from Cell Signaling Technology. Antibodies anti-BAP1 (C-4) (sc-28383, dilution: 1:500) and anti-ubiquitin (sc-8017, dilution: 1:500) were purchased from Santa Cruz Biotechnology. Antibodies anti-FLAG (F3165, dilution: 1:1000), anti-HA (H3663, dilution: 1:1000) and anti-β-actin (A1978, dilution: 1:5000) were purchased from Sigma-Aldrich. Antibodies anti-HIF1α (A300-286A, dilution: 1:500) and anti-HIF2α (BL-95-1A2, dilution: 1:500) were purchased from Bethyl Laboratories. Antibody anti-pSer123-BAP1 (4443-M, custom antibody, dilution: 1:1000) was generated from Shanghai YouKe Biotechnology Co., Ltd.Denaturating Ni-NTA pulldownCells were transfected with indicated constructs and collected cell pellets were lysed in 8 M urea, 0.1 M NaH2PO4, 300 mM NaCl, and 0.01 M Tris (pH 8.0). Lysates were briefly sonicated to shear DNA and incubated with Ni-NTA agarose beads (Invitrogen, CA, USA) for 2 h at 4 °C. Beads were washed for 4 times with 8 M urea, 0.1 M NaH2PO4, 300 mM NaCl, and 0.01 M Tris (pH 8.0). Input and beads were boiled in loading buffer and subjected to SDS–polyacrylamide gel electrophoresis and immunoblotting.Denaturing immunoprecipitation for ubiquitinationCells were lysed in 100 μl 62.5 mM Tris-HCl (pH 6.8), 10% glycerol, 2% SDS, 1 mM iodoacetamide and 20 mM NEM, boiled for 15 min, diluted 10 times with NETN buffer containing protease inhibitors, 20 mM NEM and 1 mM iodoacetamide, then centrifuged to remove cell debris. Cell extracts were subjected to immunoprecipitation with the indicated antibodies and blotted.Animal studiesFemale BALB/c nude mice (5–7-week-old) were obtained from Jicui Yaokang Biotechnology Co., Ltd., Jiangsu, China and were housed under specific-pathogen-free condition in the Animal Center of Jinan University. For subcutaneous xenografting, PANC-1 cells (1 × 106) were injected subcutaneously in mouse flanks (n = 6). Tumor volumes were measured three times weekly by using a vernier caliper to measure the short diameter and long diameter of the tumor. Tumor volumes were calculated using the following formula: width2 × length × 0.4 (mm3). When tumor volumes reached about 100 mm3, mice were administered saline or metformin (100 mg/kg) every two days. Gemcitabine (50 mg/kg) was administered for three times weekly. After tumors had grown for designated time, all mice were euthanized and tumors were harvested. For the liver metastasis study, PANC-1 cells (5 × 106) were transfected as indicated and injected into the pancreatic tail of female nude mice orthotopically (n = 6). Mice were sacrificed after 6 weeks, and the number of metastatic liver nodules was counted and quantified. For patient-derived tumor xenografts (PDXs), pancreatic tumors for the fifth generation of mice (P5) were purchased from Nanchang Royo Biotech Co., Ltd. PDAC tumors were cut into pieces of 3 × 3 × 3 mm3, and tumor tissues were pushed under skin of mice by trochar. Tumor volumes were measured three times weekly by using a vernier caliper. When tumor volumes reached 30 mm3, mice were randomly divided into different groups. Lentiviruses were produced in HEK293T cells, filtered through a 0.45 µm filter, and concentrated using a PEG-8000 (DH230-1). The xenograft tumors were intratumorally injected with control, shBAP1 #1, shBAP1 #2, shBAP1 #2 + FLAG-VHL or shBAP1#2 + FLAG-BAP1 WT, shBAP1#2 + FLAG-BAP1 3A lentivirus at a dose of 1 × 108 pfu/100 μL per mouse every three days for three times to knockdown or overexpress BAP1 expression in the tumor [14,15,16]. Mice were administered saline or metformin (100 mg/kg) every two days. Gemcitabine (50 mg/kg) was administered for three times weekly. After tumors had grown for the designated time, all mice were euthanized and tumors were harvested. All animal experiments were performed in accordance with a protocol approved by the Institutional Animal Care and Use Committee of the Jinan University (20210330-04, 20220308-06, and 20230205-08).Quantification and statistical analysisFor cell proliferation and cell viability assays, all data are analyzed by GraphPad Prism 9.3. Each experiment was independently repeated for three times, following the principle of repeatability. In the animal study, data represent as the mean ± s.d. of six mice. Statistical differences between two groups were assessed using a two-tailed Student’s t-test, while comparisons among multiple groups were performed using one-way ANOVA followed by Tukey’s post hoc test. All figures report p-values, effect sizes, and 95% confidence intervals. *p