Elevated ASCL1 activity creates de novo regulatory elements associated with neuronal differentiation

Abstract

Background The pro-neural transcription factor ASCL1 is a master regulator of neurogenesis and a key factor necessary for the reprogramming of permissive cell types to neurons. Endogenously, ASCL1 expression is often associated with neuroblast stem-ness. Moreover, ASCL1-mediated reprogramming of fibroblasts to differentiated neurons is commonly achieved using artificially high levels of ASCL1 protein, where ASCL1 acts as an “on-target” pioneer factor. However, the genome-wide effects of enhancing ASCL1 activity in a permissive neurogenic environment has not been thoroughly investigated. Here, we overexpressed ASCL1 in the neuronally-permissive context of neuroblastoma (NB) cells where modest endogenous ASCL1 supports the neuroblast programme. Results Increasing ASCL1 in neuroblastoma cells both enhances binding at existing ASCL1 sites and also leads to creation of numerous additional, lower affinity binding sites. These extensive genome-wide changes in ASCL1 binding result in significant reprogramming of the NB transcriptome, redirecting it from a proliferative neuroblastic state towards one favouring neuronal differentiation. Mechanistically, ASCL1-mediated cell cycle exit and differentiation can be increased further by preventing its multi-site phosphorylation, which is associated with additional changes in genome-wide binding and gene activation profiles. Conclusions Our findings show that enhancing ASCL1 activity in a neurogenic environment both increases binding at endogenous ASCL1 sites and also results in additional binding to new low affinity sites that favours neuronal differentiation over the proliferating neuroblast programme supported by the endogenous protein. These findings have important implications for controlling processes of neurogenesis in cancer and cellular reprogramming.

Ascl1 phospho-site mutations enhance neuronal conversion of adult cortical astrocytes in vivo

“…Ascl1 is phosphorylated by proline-directed serine/threonine kinases in the adult cortex in vivo Phosphorylation of Ascl1 on six SP sites (S62, S88, S185, S189, S202, and S218) by proline-directed serine-threonine kinases (e.g., ERK, GSK3, and CDK1; Figure 1A) has so far only been demonstrated in transfected HEK293 (Li et al, 2014), neuroblastoma (Woods et al, 2022), and glioblastoma (Azzarelli et al, 2022) cells in vitro. To determine whether Ascl1 is phosphorylated when expressed in adult cortices in vivo, we transduced both motor cortex hemispheres of adult C57Bl/6 mice using a stereotax to guide viral delivery (AP: + 2.15, L/M: ± 1.7, DV: -1.7).…”

“…As the phospho-Ascl1 antibody had previously only been validated in vitro ( Li et al, 2014 ), to provide further support for Ascl1 phosphorylation in the adult brain in vivo , we subjected protein lysates to phosphatase treatment, and ran treated and untreated samples on a Phos-tag impregnated gel ( Woods et al, 2022 ) ( Figure 1F ). Two prominent proteins between ∼28 and 37 kDa were labeled with anti-Ascl1 (Abcam) in the adult brain, running just above the predicted MW of 25 kDa for Ascl1 ( Supplementary Figure S1D ), and matching the MW of labeled proteins in embryonic day (E) 14.5 ventral telencephalic lysates (positive control; Supplementary Figure S1C ).…”

“…Protein lysates collected for Western blotting were de-phosphorylated by incubating with 400 units of Lambda Protein Phosphatase, with 1X NEBuffer for Protein MetalloPhosphatases (PMP) and 1 mM Mncl2 (NEB, cat# P0753S) at 30°C for 30 min. 10 μg of untreated and phosphatase-treated protein was then run at 50 mA constant current at 4°C on 10% acrylamide gels containing 20 μM Phos-tag ™ reagent (#AAL-107, FUJIFILM Wako Chemicals, United States Corporation) and 40 μM MnCl 2 ( Woods et al, 2022 ). Gels were washed 3 × 10 min in transfer buffer containing 10 mM EDTA followed by a final 10 min wash in transfer buffer without EDTA before transfer to PVDF membranes and Western blotting with 1/1000 rabbit pAb (#ab74065) anti-ASCL1.…”

Low-efficiency conversion of proliferative glia into induced neurons by Ascl1 in the postnatal mouse cerebral cortex in vivo

“…The overall very low conversion efficiency suggests that not only astroglia, but also cells of the oligodendroglial lineage possess effective safeguarding mechanisms that protect against acquiring a neurogenic fate. In fact, these safeguarding mechanisms are effective even when confronted with a powerful transcription factor with pioneer factor activity, such as Ascl1 (Wapinski et al, 2013; Raposo et al, 2015; Park et al, 2017) or a mutant variant with even increased neurogenic capacity (Woods et al, 2022). Thus, despite being in a proliferative state, astroglia and oligodendrocyte progenitor cells could be potentially less plastic than their adult reactive counterparts in the injured adult brain (Sirko et al, 2013; Magnusson et al, 2014; Faiz et al, 2015; Nato et al, 2015).…”

Identification of signaling pathways associated with achaete-scute homolog 1 in glioblastomas through ChIP-seq data bioinformatics

“…Three methods were used to evaluate the centrality of the complex network, including degree centrality ( Ackermann et al, 2019 ), betweenness centrality ( Nager et al, 2018 ), and closeness centrality ( Chen et al, 1998 ). CytoNCA was a cytoscape plugin for the calculation of three topology properties (parameter setting: network is without weight) ( Mahesparan et al, 1997 ; Woods et al, 2022 ). In the CytoNCA output, the node score represented the role of the protein in the network.…”