Antibody affinity maturation, which is an antigen-based selection process for B cells, occurs in germinal centers (GCs). GCB cells must efficiently recognize, acquire, and present antigens from follicular dendritic cells (FDCs) to receive positive selection signals from T helper cells. Previous studies showed that GCB cells undergo adhesive interactions with FDCs, but the regulatory mechanisms underlying the cell adhesions and their functional relevance remain unclear. Here, we identified H3K36me2 methyltransferase Nsd2 as a critical regulator of GCB cell-FDC adhesion. Nsd2 deletion modestly reduced GC responses but strongly impaired B cell affinity maturation. Mechanistically, Nsd2 directly regulated expression of multiple actin polymerization-related genes in GCB cells. Nsd2 loss reduced B cell adhesion to FDC-expressed adhesion molecules, thus affecting both B cell receptor (BCR) signaling and antigen acquisition. Overall, Nsd2 coordinates GCB positive selection by enhancing both BCR signaling and T cell help.

The level of SETD2-mediated H3K36me3 is inversely correlated with that of EZH2-catalyzed H3K27me3. Nevertheless, it remains unclear whether these two enzymatic activities are molecularly intertwined. Here, we report that SETD2 delays prostate cancer (PCa) metastasis via its substrate EZH2. We show that SETD2 methylates EZH2 which promotes EZH2 degradation. SETD2 deficiency induces a Polycomb-repressive chromatin state that enables cells to acquire metastatic traits. Conversely, mice harboring nonmethylated EZH2 mutant or SETD2 mutant defective in binding to EZH2 develop metastatic PCa. Furthermore, we identify that metformin-stimulated AMPK signaling converges at FOXO3 to stimulate SETD2 expression. Together, our results demonstrate that the SETD2-EZH2 axis integrates metabolic and epigenetic signaling to restrict PCa metastasis.

As is the case with neurodegenerative diseases, abnormal accumulation of aggregated proteins in neurons and glial are also known to implicate in the pathogenesis of ischemic stroke. However, the potential role of protein aggregates in brain ischemia remains largely unknown. Fused in Sarcoma (FUS) protein has a vital role in RNA metabolism and regulating cellular homeostasis. FUS pathology has been demonstrated in the formation of toxic aggregates and critically affecting cell viability in neurodegenerative diseases including amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), but whether this also applies to neurological injury following cerebral ischemia is unclear. Herein, we demonstrated a critical role of aggregated FUS in astrocyte activation caused by cerebral ischemia and a possible underlying molecular mechanism. Cerebral ischemic injury significantly induced the formation of cytoplasmic FUS aggregates in reactive astrocytes and injured neurons, thereby aggravating neurofunctional damages and worsening stroke outcomes. Further analysis revealed that extranuclear aggregation of FUS in astrocytes was involved in the induction of excessive autophagy, which contributes to autophagic cell injury or death. In conclusion, our results reveal the important contribution of FUS aggregates in promoting astrocyte activation in stroke pathology independent of its transcriptional regulation activity. We thus propose that aggregation of FUS is an important pathological process in ischemic stroke and targeting FUS aggregates might be of unique therapeutic value in the development of future treatment strategies for ischemic stroke.