The close chemical analogues lenalidomide and pomalidomide are immunomodulating drugs (IMiDs) that possess antineoplastic activity in multiple myeloma (MM) and other hematologic malignancies. IMiDs exert intrinsic antiproliferative effects on MM cells at least in part through direct interaction with intracellular cereblon. Cereblon (CRBN) is a component of the E3 ubiquitin ligase complex that also includes CUL4, RBX1, and DDB1 – together referred to as CRL4CRBN. Direct binding of IMiDs to a small hydrophobic pocket of the CRBN carboxy-terminal domain induces altered ubiquitinylation activity, including disrupted autoubiquitinylation of CRL4CRBN (Ito et al. Science, 2010; Fischer et al. Nature, 2014). Recently, we and others described the zinc finger transcription factors Ikaros (IKZF1) and Aiolos (IKZF3) as endogenous targets of CRL4CRBN ubiquitinylation in MM cells (Lu et al. Science, 2014; Kronke et al. Science, 2014). We found that treatment of MM cells with lenalidomide (LEN) causes increased ubiquitinylation of IKZF1 and IKZF3 by CRL4CRBN, and targets them for degradation by the proteasome. This loss of IKZF1/3 was both necessary and sufficient to observe the cellular antiproliferative effects of LEN, suggesting that this a primary mechanism of IMiD activity in MM. However, it remains unclear how depletion of these transcription factors ultimately leads to reduced proliferation of myeloma cells. Several gene products are known to be perturbed in MM cells upon IMiD treatment, including transcriptional downregulation of the master regulator transcription factor IRF4 (Zhu et al. Blood, 2011; Lu et al. Science, 2014). Yet rescue experiments with exogenous expression of IRF4 cannot fully rescue effects of IMiD treatment in MM cells, suggesting that other IKZF1/3 target genes play a role in the antiproliferative effects of IMiDs. Here we use complementary gene expression and genomic approaches to discern the global effects of IMiD treatment on MM cells. Using chromatin-immunoprecipitation followed by high-throughput sequencing (ChIP-seq), we have determined the genome-wide binding profile of both IKZF1 and IKZF3 in MM cells. We find both IKZF1 and IKZF3 occupy genomic regions including promoters, gene bodies, and distal enhancer elements. In MM cells, IKZF1 and IKZF3 enrichment almost always co-occurs, corresponding to reports of IKZF1/3 heterodimers facilitating transcriptional programs in lymphoid cells (Morgan et al. EMBO, 1997). Treatment of MM cells with LEN results in a dramatic decrease of both IKZF1 and IKZF3 binding to the genome at promoters and enhancers. Additionally, we performed ChIP-seq on RNA polymerase II (RNAPII) after LEN treatment in order to determine how IMiD-mediated IKZF1/3 depletion affects cellular transcriptional activity. Loss of IKZF1/3 binding to target genes most often correlates with increased density of RNAPII in gene bodies, suggesting transcriptional derepression of IKZF1/3 target genes. This observation was confirmed with genome-wide expression analysis by microarray. Among the derepressed target genes were genes with known tumor suppressor activity including CDKN1A, KLF6, and TXNIP. The IRF4 locus was also found to be a direct target of IKZF1/3, including binding to a large distal enhancer region upstream of the IRF4 coding region. Yet unlike most other IKZF1/3 target genes, RNAPII density within the IRF4 gene body is significantly decreased upon LEN treatment, suggesting that unique transcriptional regulatory mechanisms function at this locus that are distinct from other IKZF1/3 targets. Knockdown of IKZF1 and IKZF3 expression by shRNAs results in increased TXNIP and decreased IRF4 mRNA and protein expression, further suggesting that IMiD-mediated degradation of IKZF1/3 leads to opposite effects on these genes. TXNIP encodes the thioredoxin binding protein, which increases cellular reactive oxygen species and promotes G0/G1 cell cycle arrest. Forced exogenous overexpression of TXNIP inhibits MM cell growth, indicating a potential additional mechanism of IMiD activity. These studies define the IKZF1/3-mediated transcription program in MM cells and detail its perturbation by IMiDs. Ultimately these data will be informative for understanding the downstream effectors of intrinsic IMiD activity in hematologic malignancies, and to further understand mechanisms for acquired or innate resistance to these therapies.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution