Abstract

Active NOTCH1 signaling is present in approximately half of chronic lymphocytic leukemia (CLL) patients and was shown to be a marker for disease outcome. The NOTCH1 signaling pathway can be aberrantly activated through mutations in the NOTCH1 gene itself or co-factors such as FBXW7, resulting in protein stabilization of cleaved NOTCH1 (NICD). However, 50% of CLL cases with NOTCH1 activation do not harbor alterations in these genes, suggesting additional mechanisms for NOTCH1 hyperactivation. Since NICD can be stabilized due to defects in the ubiquitination/degradation machinery, we aimed to elucidate the role of the on chromosome 11q23 located deubiquitinase USP28 in the regulation of NOTCH1 activity and as a novel therapeutic target for NOTCH1-dependent CLL patients. First, we evaluated USP28 expression levels in a previously published cohort of 285 untreated CLL patients characterized via SNP array analysis and gene expression profiling (Edelmann et al, Blood 2012; Bloehdorn et al, Nat Commun 2021). The presence of monoallelic USP28 deletion was identified in 90% of CLL patients harboring a heterozygous deletion of chromosome 11q (del(11q); n=96), resulting in a significant decrease of USP28 RNA expression in comparison to non-del(11q) cases (n=199, P<0.0001). In non-del(11q) cases, USP28 expression levels were variable and assessment of protein levels of USP28 and its target proteins by western blotting in primary CLL cells (n=8) revealed a correlation between high protein levels of USP28 and remarkably high levels of NICD and other USP28 targets such as c-MYC or c-JUN. In addition, interaction of USP28 with NICD was shown by immunoprecipitation assays. In parallel, we implemented the CRISPR/Cas9 system to generate HG3 CLL-derived cell lines harboring USP28 biallelic deletion (HG3-USP28KO, n=5), monoallelic 11q deletion (HG3-del(11q), 17 Mb size including both ATM and USP28 genes, n=3), biallelic ATM truncation (HG3-ATMKO, n=3) or the combination of both events (HG3-del(11q) ATMKO, n=3). Overexpression of USP28 in HG3WT and HG3-USP28KO cells increased NOTCH1 activity measured by luciferase reporter assays (P<0.05, P<0.001, respectively). In contrast, in HG3-del(11q) cells decreased USP28 RNA and protein expression resulted in reduced NOTCH1 activity compared to HG3WT cells (P<0.05). Notably, HG3-del(11q) ATMKO cell lines showed even lower levels of NOTCH1 activity (P<0.01). Given that ATM phosphorylates USP28 (Zhang et al, Cell 2006), we hypothesized that in CLL cells ATM regulates USP28 activity and thereby also downstream NOTCH1 signaling. Indeed, NOTCH1 activity was reduced in HG3-ATMKO cells (P<0.01). However, USP28 overexpression in HG3-ATMKO cells did not lead to increased NOTCH1 signaling activity as observed in HG3WT cells. Moreover, pharmacological ATM inhibition (5 µM of KU-55933) also resulted in reduced NOTCH1 activity in HG3WT cells (P<0.0001) and reduced NICD levels in primary CLL cells (n=12; P<0.05). We finally explored the impact of the small molecule USP28 inhibitor AZ1 (alone or in combination with ibrutinib or venetoclax) on NICD protein levels and viability of primary CLL cells (n=15). AZ1 treatment (10 µM) resulted in decreased NICD levels, independent of mutations within NOTCH1 or FBXW7 (P<0.01). ATP quantification and 7-AAD staining revealed that increasing doses of AZ1 significantly reduced viability of NOTCH1/FBXW7-MUT primary cells (n=5) as well as of NOTCH1/FBXW7-WT cells (n=6) expressing high protein levels of USP28 and NICD (P<0.01; P<0.05, respectively). In contrast, AZ1 treatment did not affect viability of CLL patient cells expressing low NICD levels (n=4; P=0.56) Furthermore, AZ1 combined with venetoclax potentiated the reduction of cell viability especially in NOTCH1/FBXW7-MUT or patient cells expressing high USP28 and NICD levels (P<0.001; P<0.05). We propose that high levels of the deubiquitinase USP28 are responsible for hyperactivation of NOTCH1 signaling in CLL. This hypothesis is further confirmed in del(11q) CLL cells, where USP28 heterozygous deletion resulted in low NOTCH1 activity. In addition, our data reveal that USP28 activity is regulated by ATM, uncovering a novel mechanism of NOTCH1 modulation via the ATM-USP28-NOTCH1 axis (Figure 1). Finally, we demonstrate that USP28 inhibition by AZ1 suppresses USP28-mediated NOTCH1 activation and could represent a novel therapeutic strategy in CLL. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call