Abstract
The histone 3 methyltransferase NSD3 was an oncogenic driver of lung squamous cell carcinoma.
Major Finding: The histone 3 methyltransferase NSD3 was an oncogenic driver of lung squamous cell carcinoma.
Concept: Functional and structural studies showed NSD3 methyltransferase activity promoted tumorigenicity.
Impact: This work defines NSD3 as a critical oncogene and reveals an actionable therapeutic vulnerability.
Lung squamous cell carcinomas (LUSC) commonly harbor amplification of chromosomal region 8p11–12, which includes the gene encoding NSD3, a histone methyltransferase that dimethylates lysine residue 36 of histone 3 (H3K6me2). Analyzing data from patients with LUSC, Yuan, Flores, and colleagues discovered that gene amplification of NSD3 was correlated with increased mRNA, in contrast to FGFR1, the previously suspected candidate oncogenic driver on 8p11–12. In a mouse xenograft model harboring NSD3-knockout, 8p11–12-amplified LUSC, tumor growth was substantially slowed, suggesting that NSD3 may be an oncogenic driver. Similarly, a genetically engineered mouse model of LUSC showed that lung-specific conditional deletion of Nsd3, but not Fgfr1, reduced lung tumor growth and extended survival. To understand the potential oncogenic effects of increased NSD3 activity, recombinant versions of NSD3's catalytic SET domain with various cancer-associated mutations were tested for enhanced in vitro histone methylation activity. The T1232A-mutant variant of NSD3 increased H3K36 methylation compared with wild-type NSD3, and structural information from high-resolution nuclear magnetic resonance spectroscopy suggested that this hyperactive variant increased H3K36me2 catalysis by destabilizing an autoinhibited conformation of NSD3 and increasing accessibility to the H3 substrate, a finding supported by biochemical evidence. In vivo, expression of the hyperactive T1232A-mutant variant increased the volume of lung tumors, reduced survival, and was correlated with upregulation of members of oncogenic pathways, including MYC targets and mTOR signaling. The genomic distribution of H3K36me2 localization exhibited a distinct pattern across loci that were downregulated upon NSD3 depletion via CRISPR-mediated mutagenesis. Further, increased NSD3 histone methylation activity likely preferentially affected genes with robust basal expression. Xenografts in mice with NSD3-knockout human LUSC cell lines with 8p11–12 amplification or NSD3 overexpression had reduced growth, while ectopic expression of NSD3 enhanced SOX2-mediated transformation of human tracheobronchial epithelial cells. Notably, NSD3 dependency rendered patient-derived xenograft mouse models sensitive to bromodomain inhibitors, possibly due to previously discovered interactions between the bromodomain protein BRD4 and NSD3. In summary, these studies reveal NSD3 as an oncogenic driver of LUSC and highlight potential therapeutic implications of NSD3 dependency.
Note: Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.