IRF4 Is Reciprocally Dysregulated Via NF-Κb-Dependent Expression and Loss-of-Function Mutations in Multiple Myeloma

Introduction: Interferon regulatory factor 4 (IRF4) is a transcription factor involved in B cell activation and differentiation into plasma cells, as well as B-cell lineage neoplasia. Diverse genetic abnormalities and dysregulated expression of IRF4 have been documented in various hematologic malignancies, including multiple myeloma (MM). IRF4 is also identified as a major downstream target of CRBN/IKZF to regulate MYC expression, representing a primary mechanism underlying anti-MM action of IMiDs. However, it remains unclear how the expression and function of IRF4 are regulated in MM cells.

Materials and Methods: DNAseq, whole exome RNAseq (WES), and label-free analyses of quantitative proteomics and post-translational modifications (PTMs) mapping were conducted to examine genetic aberrations, transcriptional dysregulation, or protein and PTM alterations in primary patient samples and drug-naive vs -resistant cell lines. Flow cytometry, qPCR, and Western blot analysis were utilized to monitor apoptosis or mRNA and protein levels of target genes. Stable transfection with wild-type or mutated IRF4 was carried out to evaluate its function. Analysis of the MM genome-wide GEP databases (R2: Genomics Analysis and Visualization Platform) was performed to validate the clinical significance of IRF4 and its target genes in MM patients.

Results: MM cells that acquired bortezomib (btz) resistance (PS-R) exhibited a marked increase in protein levels of multiple key components of the canonical and non-canonical NF-κB signaling cascades, in association with increased NF-κB transcriptional activity. Pharmacological inhibition of NF-κB using several compounds (e.g., Bay 11-7082, parthenolide, triptolide) significantly increased btz sensitivity in PS-R cells and their parental U266 cells, as well as other MM cell lines. These events were accompanied by down-regulation of various genes, including NF-κB-dependent genes and those whose relationship with NF-κB remains uncertain. The latter included IRF4, MYC, MCL1, BIRC2/cIAP1, SQSTM1/p62, among others. Of note, overexpression of wild-type IRF4 sharply reduced sensitivity of drug-naive cells to btz alone or in combination with NF-κB inhibitors. However, IRF4 overexpression did not prevent inactivation of NF-κB and down-regulation of the NF-κB-dependent genes (e.g., Bcl-xL, TNFAIP3/A20), suggesting that IRF4 acts downstream of NF-κB. Surprisingly, IRF4 overexpression failed to restore expression of MYC, a well-established target of IRF4, while largely reversed expression of cIAP1 and Mcl-1 at both mRNA and protein levels. Moreover, inhibitors of either IAP (birinapant/TL32711) or Mcl-1 (S63845) overcame btz-resistance conferred by IRF4 overexpression. DNAseq unveiled multiple IRF4 SNPs located in introns and exons (both 3'-UTR and coding region) of chromosome 6. WES revealed that IRF4 was highly expressed in primary bone marrow and extramedullary samples of MM patients (2~10 folds over normal donors). But no difference at mRNA and protein levels of IRF4 was observed between drug-naive and -resistant cell lines. The global PTM mapping identified two lysine residues (K59 and K399) that were highly ubiquitinated in PS-R cells (p = 0.0447 and p = 0.0004, vs U266 cells). Interestingly, ectopic expression of IRF4 mutants (e.g., K59N, K123R) lost the cytoprotective capability of their wild-type counterpart, in association with failure to reverse down-regulation of cIAP1 and Mcl-1. This observation argues that these point mutations are loss-of-function, which might explain better prognosis of patients carrying IRF4 mutations (including K59N and K123R) than wild-type IRF4 as reported previously. Last, the MM GEP database analysis validated the correlation between IRF4 and Mcl-1 or cIAP expression, as well as the adverse effect of their dysregulation on overall survival of MM patients.

Conclusion: IRF4 is highly expressed in MM cells, likely via an NF-κB-dependent mechanism. IRF4 acts downstream of the NF-κB signaling pathways, which in turn directly targets cIAP1 and Mcl-1, rather than MYC, to confer drug resistance. IRF4 could be mutated in MM patients, while the point mutations, at least at certain lysine sites, seem to be loss-of-function and thus predict better outcome. Together, these findings suggest that the oncogenic function of IRF4 might be governed at genetic, transcriptional, and PTM levels in MM cells.