Summary: Acute myeloid leukaemia (AML) cells possess metabolism profiles, such as higher rates of oxidative phosphorylation and dependence on fatty acid oxidation for survival, and are dependent on the sophisticated regulation of reactive oxygen species (ROS) generation for survival, drug resistance and stemness maintenance. We found that sensitivity of primary AML cells to cytarabine correlated with SOD2 acetylation and the ability of the drug to induce mitochondrial ROS. The SOD2 deacetylase, SIRT3, protected AML cells from chemotherapy as shown by inhibited apoptosis via inhibited drug‐induced production of mitochondrial ROS. SIRT3 significantly decreased nicotinamide adenine dinucleotide phosphate (NADP)/reduced NADP ratio and increased reduced glutathione/oxidized glutathione ratio. Furthermore, SIRT3 enhanced oxidative phosphorylation (OxPhos) in AML cells under both basic and cytarabine‐treated conditions. A xenograft mouse model showed that SIRT3 overexpressing AML cells and patient‐derived xenograft mice bearing high SIRT3 deacetylase activity were more resistant to chemotherapy in vivo. SIRT3 inhibitor displayed synergy with cytarabine to ablate AML cells in vitro and in mouse models. Taken together, our study showed that SIRT3 is capable of reprograming mitochondrial metabolism towards OxPhos and downregulating ROS generation, which contribute to the chemoresistance of AML cells. SIRT3 can be utilized as a potential therapeutic target to improve the anti‐leukaemic efficacy of standard chemotherapeutic agents for AML.

Although oncogenicity of the stem cell regulator SOX9 has been implicated in many solid tumors, its role in lymphomagenesis remains largely unknown. In this study, SOX9 was overexpressed preferentially in a subset of diffuse large B-cell lymphomas (DLBCLs) that harbor IGH-BCL2 translocations. SOX9 positivity in DLBCL correlated with an advanced stage of disease. Silencing of SOX9 decreased cell proliferation, induced G1/S arrest, and increased apoptosis of DLBCL cells, both in vitro and in vivo. Whole-transcriptome analysis and chromatin immunoprecipitation–sequencing assays identified DHCR24, a terminal enzyme in cholesterol biosynthesis, as a direct target of SOX9, which promotes cholesterol synthesis by increasing DHCR24 expression. Enforced expression of DHCR24 was capable of rescuing the phenotypes associated with SOX9 knockdown in DLBCL cells. In models of DLBCL cell line xenografts, SOX9 knockdown resulted in a lower DHCR24 level, reduced cholesterol content, and decreased tumor load. Pharmacological inhibition of cholesterol synthesis also inhibited DLBCL xenograft tumorigenesis, the reduction of which is more pronounced in DLBCL cell lines with higher SOX9 expression, suggesting that it may be addicted to cholesterol. In summary, our study demonstrated that SOX9 can drive lymphomagenesis through DHCR24 and the cholesterol biosynthesis pathway. This SOX9-DHCR24-cholesterol biosynthesis axis may serve as a novel treatment target for DLBCLs.

Sirtuin 3 (SIRT3) deacetylase is a key regulator for chemoresistance in acute myeloid leukemia (AML) cells due to its capability of modulating mitochondrial metabolism and reactive oxygen species (ROS). SIRT3 is de-SUMOylated by SUMO-specific peptidase 1 (SENP1), which enhances its deacetylase activity. Therefore, dysregulation of SIRT3 SUMOylation may lead to fortified chemoresistance in AML. Indeed, SIRT3 de-SUMOylation was induced by chemotherapeutic agents, which in turn, exacerbated resistance against chemotherapies in AML by activating SIRT3 via preventing its proteasome degradation. Furthermore, RNA-seq revealed that expression of a collection of genes was altered by SIRT3 de-SUMOylation including inhibition of transcription factor Hes Family BHLH Transcription Factor 1 (HES1), a downstream substrate of Notch1 signaling pathway, leading to increased fatty acids oxidation (FAO). Moreover, the SENP1 inhibitor momordin-Ic or HES1 overexpression synergized with cytarabine to eradicate AML cells in vitro and in xenograft mouse models. In summary, the current study revealed a novel role of SIRT3 SUMOylation in the regulation of chemoresistance in AML via HES1-dependent FAO and provided a rationale for SIRT3 SUMOylation and FAO targeted interventions to improve chemotherapies in AML.

Background: Diffuse large B-cell lymphoma (DLBCL), the most common type of Non-Hodgkin’s Lymphoma (NHL), has a lethal nature. Thus, the establishment of a novel model to predict the prognosis of DLBCL and guide its therapy is an urgency. Meanwhile, pyroptosis is engaged in the progression of DLBCL with further investigations required to reveal the underlying mechanism. Methods: LASSO regression was conducted to establish a risk model based on those PRGs. External datasets, RT-qPCR and IHC images from The Human Protein Alta (HPA) database were utilized to validate the model. ssGSEA was utilized to estimate the score of immune components in DLBCL. Results: A model based on 8 PRGs was established to generate a risk score. Validation of the model confirmed its robust performance. The risk score was associated with advanced clinical stages and shorter overall survivals. Two novel second-line chemotherapies were found to be potential treatments for high-risk patients. The risk score was also found to be correlated with immune components in DLBCL. Conclusion: This novel model can be utilized in clinical practices to predict the prognosis of DLBCL and guide the treatment of patients at high risk, providing an overview of immune regulatory program via pyroptosis in DLBCL.

Although the cellular role of uncoupling protein 2 (UCP2) in tumorigenesis has been reported in various solid tumor models, its role in leukemogenesis remains elusive. Herein, we demonstrated that UCP2 was highly expressed in AML and significantly associated with poor prognosis and chemoresistance, suggesting that UCP2 can be used as a potential biomarker in acute myeloid leukemia. Mechanistically, in vitro and in vivo silencing of UCP2 significantly impairs acute myeloid leukemia cell growth and survival, accompanied by the disruption of mitochondrial homeostasis. Interestingly, RNA-sequencing analysis and metabolic mass spectrometry revealed that silencing UCP2 resulted in accumulated branched-chain amino acids (BCAAs), which induced oxidative stress through the PI3K/AKT/mTOR signaling pathway. Additionally, the lack of BCAAs restored leukemic cell growth and survival and decreased mitochondrial ROS production induced by inhibiting UCP2. More importantly, supplementation of BCAA enhanced the anti-tumor activity of genipin, a selective inhibitor that targets UCP2, resulting in significantly reduced acute myeloid leukemia blasts, increased mouse survival, and magnified oxidative stress. Taken together, our study elucidates the rationale of targeting the UCP2-BCAA-PI3K/AKT/mTOR signaling axis in leukemogenesis and provides a novel strategy for leveraging the metabolic dependencies of leukemic cells.