Genetic Modulation of Adenosine-to-Inosine RNA Editing Selectively Disrupts Inflammasome and Extracellular Matrix Genes in Multiple Myeloma

Introduction

As the second most common blood cancer in the U.S., multiple myeloma (MM) is typified by clonal plasma cell proliferation in the bone marrow (BM) and may progress to therapy-resistant plasma cell leukemia (PCL). Despite many novel therapies, relapse rates remain high as a result of malignant regeneration (self-renewal) of MM cells in inflammatory microenvironments. In addition to recurrent DNA mutations and epigenetic deregulation, inflammatory cytokine-responsive adenosine deaminase associated with RNA (ADAR1) mediated adenosine to inosine (A-to-I) RNA editing has emerged as a key driver of cancer relapse and progression. In MM, copy number amplification of chromosome 1q21, which contains both ADAR1 and interleukin-6 receptor (IL-6R) gene loci, portends a poor prognosis. Thus, we hypothesized that ADAR1 copy number amplification combined with inflammatory cytokine activation of ADAR1 stimulate malignant regeneration of MM and therapeutic resistance.

Methods and Results

To evaluate the contribution of ADAR1 modulation to global RNA editing changes and differential gene expression, we performed whole transcriptome sequencing of primary MM and PCL samples that were lentivirally transduced with shRNA targeting human ADAR1 compared to vector control shRNA. Analysis at the single nucleotide level revealed that 20-50% knockdown of ADAR1 was sufficient to induce substantial downregulation of A-to-I editing rates throughout the transcriptome, particularly at 3'UTR loci, along with specific modulation of extracellular matrix and inflammasome-associated gene expression patterns. Moreover, in contrast to lentiviral ADAR1 shRNA knockdown and overexpression of an editase defective ADAR1 mutant (ADAR1^E912A), lentiviral wild-type ADAR1 overexpression enhanced editing of GLI1, a Hedgehog (Hh) pathway transcriptional activator and self-renewal agonist. Editing of GLI1 transcripts enhanced GLI transcriptional activity in luciferase reporter assays, and promoted lenalidomide resistance in vitro. Finally, lentiviral shRNA ADAR1 knockdown reduced regeneration of high-risk MM in humanized serial transplantation mouse models, indicative of reduced malignant self-renewal capacity. Notably, in human myeloma cells analyzed using NanoString nCounter assays, overexpression of wild-type versus edited GLI1 elicited distinct gene expression changes also in extracellular matrix and immune response genes. These data demonstrate that ADAR1 promotes malignant self-renewal of MM and if selectively inhibited may prevent progression and relapse through modulation of global A-to-I RNA editing and extracellular and immune response gene expression.

Conclusions

Deregulated RNA editing, driven by aberrant ADAR1 activation, represents a unique source of transcriptomic and proteomic diversity, resulting in self-renewal of MM cells in inflammatory microenvironments. In summary, both genetic (1q21 amplification) and microenvironmental factors (inflammatory cytokines, immunomodulatory drugs) combine to drive A-to-I RNA editing-dependent malignant regeneration in MM through inflammasome and extracellular matrix gene pathway remodeling. These effects can be down-modulated even with small reductions in ADAR1 expression and activity. Thus, ADAR1 represents both a vital prognostic biomarker and therapeutic target in MM.