Histone modifications, such as the frequently occurring lysine succinylation1,2, are central to the regulation of chromatin-based processes. However, the mechanism and functional consequences of histone succinylation are unknown. Here we show that the α-ketoglutarate dehydrogenase (α-KGDH) complex is localized in the nucleus in human cell lines and binds to lysine acetyltransferase 2A (KAT2A, also known as GCN5) in the promoter regions of genes. We show that succinyl-coenzyme A (succinyl-CoA) binds to KAT2A. The crystal structure of the catalytic domain of KAT2A in complex with succinyl-CoA at 2.3 Å resolution shows that succinyl-CoA binds to a deep cleft of KAT2A with the succinyl moiety pointing towards the end of a flexible loop 3, which adopts different structural conformations in succinyl-CoA-bound and acetyl-CoA-bound forms. Site-directed mutagenesis indicates that tyrosine 645 in this loop has an important role in the selective binding of succinyl- CoA over acetyl-CoA. KAT2A acts as a succinyltransferase and succinylates histone H3 on lysine 79, with a maximum frequency around the transcription start sites of genes. Preventing the α-KGDH complex from entering the nucleus, or expression of KAT2A(Tyr645Ala), reduces gene expression and inhibits tumour cell proliferation and tumour growth. These findings reveal an important mechanism of histone modification and demonstrate that local generation of succinyl-CoA by the nuclear α-KGDH complex coupled with the succinyltransferase activity of KAT2A is instrumental in histone succinylation, tumour cell proliferation, and tumour development.

SUMMARY Overcoming metabolic stress is a critical step in tumor growth. Acetyl-CoA generated from glucose and acetate uptake is important for histone acetylation and gene expression. However, how acetyl-CoA is produced under nutritional stress is unclear. We demonstrate here that glucose deprivation results in AMPK-mediated acetyl-CoA synthetase 2 (ACSS2) phosphorylation at S659, which exposed the nuclear localization signal of ACSS2 for importin α5 binding and nuclear translocation. In the nucleus, ACSS2 binds to transcription factor EB and translocates to lysosomal and autophagy gene promoter regions, where ACSS2 incorporates acetate generated from histone acetylation turnover to locally produce acetyl-CoA for histone H3 acetylation in these regions and promote lysosomal biogenesis, autophagy, cell survival, and brain tumorigenesis. In addition, ACSS2 S659 phosphorylation positively correlates with AMPK activity in glioma specimens and grades of glioma malignancy. These results underscore the significance of nuclear ACSS2-mediated histone acetylation in maintaining cell homeostasis and tumor development.

Metastasis is a complex pathophysiological process. As the main cause of cancer mortality in humans it represents a serious challenge to both basic researchers and clinicians. Here we report the design and construction of a multi-organ microfluidic chip that closely mimics the in vivo microenvironment of lung cancer metastasis. This multi-organs-on-a-chip includes an upstream "lung" and three downstream "distant organs", with three polydimethylsiloxane (PDMS) layers and two thin PDMS microporous membranes bonded to form three parallel microchannels. Bronchial epithelial, lung cancer, microvascular endothelial, mononuclear, and fibroblast cells were grown separated by the biomembrane in upstream "lung", while astrocytes, osteocytes, and hepatocytes were grown in distant chambers, to mimic lung cancer cell metastasis to the brain, bone, and liver. After culture in this system, lung cancer cells formed a "tumor mass", showed epithelial-mesenchymal transition (with altered expression of E-cadherin, N-cadherin, Snail1, and Snail2) and invasive capacity. A549 cells co-cultured with astrocytes overexpressed CXCR4 protein, indicating damage of astrocytes after cancer cell metastasis to the brain. Osteocytes overexpressed RANKL protein indicates damage of osteocytes after cancer cell metastasis to the bone, and hepatocytes overexpressed AFP protein indicates damage to hepatocytes after cancer cell metastasis to the liver. Finally, in vivo imaging of cancer growth and metastasis in a nude mice model validated the performance of metastasis in the organs-on-chip system. This system provides a useful tool to mimic the in vivo microenvironment of cancer metastasis and to investigate cell-cell interactions during metastasis.

Monoclonal antibodies (mAbs) blocking immune checkpoint molecules, especially programmed cell death 1 (PD-1) and its ligands programmed cell death 1 ligand 1 (PD-L1) and ligand 2 (PD-L2), are currently been investigated for treatment of various tumors [1-3]. PD-L1 and PD-L2 are usually upregulated on the surface of multiple tumor cells to mediate immune tolerance through the interaction with inhibitory PD-1 molecule [4]. Thus, blocking PD-1/PD-Ls interaction has brought promising future for tumor immunotherapy. To date, several PD-1/ PD-L1 blockade antibodies have been approved for clinical use or under phase III clinical trials (e.g., nivolumab, pembrolizumab, atezolizumab, avelumab, durvalumab and BMS-936559, etc.) [4]. The PD-1 targeting therapeutic antibodies block the PD-1/PD-L1 or PD-1/PD-L2 interaction to restore tumor-specific T cell reactivity, without mediating antibody-dependent cell-mediated cytotoxicity (ADCC). Recently, the structural basis of hPD-1/pembrolizumab (a PD-1 targeting therapeutic antibody developed by Merck & Co., Inc., USA) has been revealed, providing a molecular insight into blocking PD-1-mediated immune suppression by antibody [5]. PD-L1 targeting therapeutic antibodies possess PD-1/ PD-L1 blockade activity with or without ADCC activity. As one of the PD-L1 targeting antibodies, avelumab is a human IgG1 antibody with ADCC activity developed by Merck (Darmstadt, Germany) and Pfizer, which is now in multiple phase III clinical trials against non-small cell lung cancer (NCT02395172), advanced renal cell cancer (NCT02684006) and gastric cancer (NCT02625610) [6]. The crystal structures of PD-L1 couplexed with its receptor PD-1 have been extensively studied, including human PD-L1 (hPD-L1) alone, mouse PD-1 (mPD-1) complexed with hPD-L1 and human PD-1 (hPD-1) complexed with hPD-L1 [7-9]. Though the complex structure of hPD-1 with a commercial mAb pembrolizumab has been solved very recently [5], hPD-L1/mAb complex structure has not been investigated. In this study, we expressed the single chain Fv frag