The COP9 signalosome is an evolutionarily conserved hetero-octameric complex that functions as an isopeptidase for NEDD8 and thereby fine-tunes the activity of Cullin-RING ubiquitin ligases within the ubiquitin-proteasome system. Its subunit-specific roles vary by tumor context, with prior studies linking individual COP9 subunits to either oncogenic or tumor suppressor programs in different cancers. Before this study, only CSN6 had been investigated in glioblastoma where it was implicated in stabilizing EGFR and promoting tumor cell proliferation. The broader contribution of other COP9 subunits to glioblastoma stem cell (GSC) biology remained largely unknown.
A previously published genome-wide essentiality screen across 11 patient-derived GSC cultures identified only two COP9 subunit genes whose knockout conferred a fitness defect across all lines: GPS1 (encoding CSN1) and COPS6 (encoding CSN6). Given prior work on CSN6, the present study focused on elucidating the role of CSN1. CSN1 protein expression was confirmed across a diverse panel of patient-derived GSC cultures, supporting its relevance to GSC biology.
CRISPR-Cas9 targeting of GPS1 with three guide RNAs in two independent patient-derived GSC lines produced stable lines with either ~20% residual CSN1 (GPS1–1) or near-complete knockout (GPS1–2, referred to as CSN1-KO). Tracking growth over a 14-day period showed that reduced CSN1 expression significantly impaired GSC proliferation.
Flow cytometric EdU incorporation analysis revealed that CSN1-KO disrupted normal cell cycle distribution: CSN1-KO cultures had a reduced percentage of cells in S and G2+M phases and an accumulation of cells in G0/G1 relative to AAVS1 controls. Limiting dilution assays demonstrated a significant reduction in sphere-initiating frequency for CSN1-KO GSCs, and Western blotting showed a dramatic reduction in the stemness marker SOX2.
To assess in vivo relevance, orthotopic xenografts of two independent CSN1-KO GSC lines into SCID mice were performed. Mice implanted with CSN1-KO GSCs exhibited a significant survival advantage compared with AAVS1 controls; Kaplan–Meier analysis indicated p < 0.0001 in both tested GSC models.
RNA-sequencing of CSN1-KO versus AAVS1 GSCs followed by gene set enrichment analysis (GSEA) showed consistent transcriptional alterations. Biological processes and reactome pathways involved in cell cycle progression, including G1–S transition programs, were negatively enriched in CSN1-KO GSCs. Conversely, apoptosis-related gene sets were positively enriched in CSN1-KO cells. These transcriptional signatures align with the observed reductions in proliferation and stemness and the increase in cell cycle arrest.
At the protein level, CSN1-KO led to loss of multiple COP9 subunits: CSN3, CSN5, CSN6, CSN7A, CSN7B, and CSN8 were reduced or lost in CSN1-KO GSCs, whereas CSN2 and CSN4 expression was maintained or upregulated. Transcript-level analysis showed GPS1 as the only significantly downregulated COP9-related transcript in CSN1-KO cells, indicating that the widespread loss of COP9 proteins after CSN1 deletion is not explained by transcriptional downregulation of those subunits. The data support a model in which CSN1 is essential for proper COP9 signalosome assembly and stability; its loss likely triggers post-translational destabilization and degradation of multiple subunits.
To test effects on signaling networks, a phospho-kinase array assessing 37 phosphoproteins was performed. Quantification and subsequent validation by Western blot demonstrated increased phosphorylation of STAT3 (Y705, S727), JNK1/2/3 (T183/Y185, T221/Y223), and c-Jun (S63) in CSN1-KO GSCs without corresponding increases in total protein levels. These changes indicate an altered phospho-proteomic state following COP9 disruption.
Given CSN1's previously reported interactions with JNK/c-Jun/AP-1 signaling in other contexts, the authors evaluated transcription factor target enrichment. AP-1 targets were enriched in CSN1-KO GSCs, consistent with increased JNK and c-Jun phosphorylation and suggesting AP-1 activation contributes to downstream transcriptional changes.
Because STAT3 phosphorylation is typically linked to increased proliferation and survival in GSCs, the authors tested whether CSN1-KO cells were differentially sensitive to the JAK2/STAT3 inhibitor pacritinib at IC50 and IC75 concentrations. No difference in viability was observed between AAVS1 and CSN1-KO GSCs at the tested concentrations after 7 days, suggesting that increased STAT3 phosphorylation in CSN1-KO cells does not render them more dependent on JAK2/STAT3 signaling under these conditions.
The study notes that while AP-1 activation and STAT3 phosphorylation are observed, the precise molecular mechanisms linking COP9 disruption to these signaling events and the full set of downstream programs remain unclear and warrant further investigation.
Collectively, the data demonstrate that CSN1 is essential for COP9 signalosome assembly in GSCs and that its loss destabilizes the complex, alters the phospho-proteome (notably JNK/c-Jun/AP-1 and STAT3 phosphorylation), represses cell cycle programs, activates apoptotic pathways, reduces stemness markers such as SOX2, and prolongs survival in orthotopic xenograft models. These results identify CSN1 as a critical regulator of GSC proliferation and tumorigenesis and provide mechanistic insight into how COP9 complex integrity sustains key signaling and transcriptional programs in glioblastoma stem cells.
Data availability reported by the authors: RNA-sequencing data are deposited in the NCBI SRA under PRJNA1497041; representative original Western blots are deposited at Mendeley Data (doi:10.17632/jyc4j3vkfk.1). The study was supported by the Canadian Institutes for Health Research and the authors declared no competing interests.