Background: Previous work in our lab identified that ixazomib combined with the histone deacetylase inhibitor, belinostat, or the CHK1 inhibitor, AZD7762, were synergistic in lymphoma. Our goals were to explore the biologic effects of rational combinations of ixazomib with other novel agents in TCL and HL cells.
Methods: Transcriptome analysis used the Human HT 12 Genechip Illumina with RNA isolated from Jurkat cells treated with vehicle, ixazomib, belinostat, or AZD7762 as single agents, and in combinations: ixazomib + belinostat or ixazomib + AZD7762. Data were background adjusted and quantile normalized using RMAExpress, and statistically relevant genes were determined by applying LIMMA with FDR <0.05. Common key genes were identified using Ingenuity Pathway Analysis (IPA) and gene set enrichment analysis (GSEA) C5 Gene Ontology databases. In Jurkat and L428 cells, viability and apoptosis were assessed using MTT and annexin/PI flow cytometry assays. Western Blots were done for cleaved caspase and PARP, Myc, acetylated H3, NRF2 and cathepsin D.
Results: Transcriptome analysis of biologic pathways with GSEA showed similar activation and inhibition of pathways in Jurkat with ixazomib + belinostat vs belinostat alone; however, there was more inhibition of immune system processes with belinostat + ixazomib (Fig. 1). In GSEA, PD1 signaling and nucleotide metabolism were downregulated by belinostat + ixazomib, with upregulation of protein folding, transferrin endocytosis and recycling. 67 key genes were differentially expressed with ixazomib + belinostat vs either agent alone. These genes were predicted to have an overall tumor suppression effect. Genes upregulated by ixazomib alone in Jurkat included: oxidative stress response (NQO1, GCLC), lysosomal degradation of proteins (SQSTM1, Lamp2), and proteasome components (PSMC2, PSMC3, PSMC4, PSMC5, PSMD2, PSMD4). Transcriptome analysis showed these genes were downregulated in Jurkat with belinostat +/- ixazomib, suggesting that belinostat may abrogate these pro-survival pathways. Confirmatory analyses for cathepsin D, a marker of lysosomal activation, and total cellular NRF2 (involved in protective response against oxidative stress) showed decreased levels for Jurkat cells treated with belinostat + ixazomib vs ixazomib alone. Similarly, in L428, decrease in total cellular NRF2 protein was seen with addition of belinostat; however, with increased dose of ixazomib, this effect was mitigated. Cleaved caspase 3 and cleaved PARP were increased in Jurkat and L428 cells treated with combined ixazomib + belinostat. The combination of ixazomib + belinostat in Jurkat and L428 cells was synergistic by MTT assay, with combination index (CI) value < 1. GSEA with ixazomib + AZD7762 in Jurkat revealed contrasting effects on biologic processes of mitotic cell cycle and chromatin modification, with these pathways upregulated by AZD7762, but downregulated by ixazomib. The effect of ixazomib + AZD7762 showed cell cycle genes were downregulated vs control. Of 103 differentially expressed genes in combination, there was overall balance between genes predicted to contribute to tumor growth and suppression. In L428, AZD7762 + ixazomib induced apoptosis and decreased cell viability in a synergistic manner (CI < 1). Finally, AZD7762 + ixazomib led to decreased Myc and acetylated histone H3 with ChIP PCR assay for MYC showing decreased promoter occupancy by acetylated-Histone vs ixazomib alone in L428.
Conclusions: Ixazomib + belinostat was synergistic and this combination resulted in predicted downregulation of immune system responses on transcriptome analysis with abrogation of pro-survival oxidative stress and lysosomal degradation pathways upregulated by single-agent ixazomib. These findings suggest combination treatment with belinostat improves efficacy of ixazomib in TCL and HL cells by targeting alternate pathways. AZD7762 + ixazomib had contrasting effects on cell cycle and chromatin modification with Myc playing a critical role. Further investigation into effects of combinatorial treatment on biologic pathways and tumor effect is warranted, including delineation of potential predictive biomarkers.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.