Repositioning the Repurposed Drug, a Structural Study of Thalidomide Analogs

Advances in technology and understanding the molecular profile of tumors have resulted in the identification of new therapeutic targets in cancer. However, the process of drug discovery and development usually takes 10 to 15 years and is costly and risky. Given the huge demand for new cancer treatments and the lengthy process for novel drug development, some scientists turn to drug repurposing — namely, the identification of new uses for U.S. Food and Drug Administration (FDA)–approved or investigational drugs. Thalidomide is a poster child for drug repurposing in cancer, as it was developed in the 1950s as a sedative for pregnant women, subsequently used as a therapy for leprosy, and approved by the FDA in 2006 for treatment of multiple myeloma (MM).¹ 

MM is a genetically complex and heterogeneous malignancy of plasma cells that is characterized by progression of disease from distinct precursor states: monoclonal gammopathy of undetermined significance and smoldering multiple myeloma.^2,3  MM is the second most common hematologic malignancy in the United States and remains an incurable disease. Frontline therapy for MM differs across countries depending on drug availability; the most common regimens are either double-agent therapy with lenalidomide and dexamethasone or triple-agent therapies with bortezomib, lenalidomide, and dexamethasone; bortezomib, thalidomide, and dexamethasone; and bortezomib, cyclophosphamide, and dexamethasone.⁴ 

The immunomodulatory drug (IMiD) thalidomide and its structural analogs lenalidomide and pomalidomide induce the ubiquitination and proteosomal degradation of key MM transcription factors Ikaros (IKZF1) and Aiolos (IKZF3) via binding to Cereblon (CRBN). CRBN is the substrate receptor for the E3 ubiquitin ligase, and binding of those drugs to CRBN alters its selectivity to substrates such as IKZF1 and IKZF3, and subsequently leads to their degradation through proteolysis.^5,6  IKZF1 and IKZF3 are members of the largest family of C2H2 zinc finger (ZF) proteins in the human proteome, which includes many putative transcription factors — usually challenging targets for drug development due to the lack of targetable active sites.

In a recent functional and computational study, Dr. Quinlan L. Sievers and colleagues defined novel protein targets of thalidomide analogs, drawing from approximately 800 C2H2 ZF-containing proteins. Screening 6,572 C2H2 ZF motifs, they discovered 11 ZFs that were degraded in the presence of thalidomide, lenalidomide, and pomalidomide, while six of those were able to mediate the degradation of their corresponding full-length protein, comprising known targets IKFZ1/IKZF3 as well as five novel targets, ZNF692, ZFP91, ZNF276, ZNF653, and ZNF827.

Although the 11 ZF motifs discovered in the screen did not show major similarities in their amino acid sequences, those that could mediate the degradation of their full-length protein had an additional ZF protein motif that was associated with high-affinity binding to the drug-CRBN complex. A structural study of IKZF1 and pomalidomide further suggested that it is the overall shape of the C2H2 ZF domain, rather than its amino acid sequence, that determines whether a ZF motif will fit and bind the complementary groove on the drug-CRBN complex surface. Supporting this notion, Dr. Sievers and colleagues performed an analysis on structurally similar ZFs and showed that approximately 50 to 150 additional C2H2 ZFs were capable of binding to the drug-CRBN complex with similar or better scores than the 11 ZFs discovered in the screen. Functional validation of biologically relevant ZFs in vitro further proved that the drug-CRBN interface is prone to interacting with higher numbers of ZF proteins than discovered in the screen and introduced new target ZF proteins for therapeutic intervention, such as the lymphoma onco-protein BCL6, the transcription factor EGR1 that regulates multiple tumor suppressor genes, and the oncofetal protein SALL4. Importantly, chemical modification of thalidomide analogs at the drug-ZF interface resulted in the degradation of particular groups of ZFs, suggesting that thalidomide analogs can be modified to selectively target different proteins, including transcriptional factors that are traditionally considered to be “undruggable.”