Dysregulated Aid/Apobec Family Proteins Promote Genomic Instability in Multiple Myeloma

The AID/APOBEC family of cytidine deaminase proteins includes AID (activity induced deaminase), and 10 related APOBEC enzymes (A1, A2, A3A, A3B, A3C, A3D, A3F, A3G, A3H and A4). AID has been well-studied for its role in somatic hyper mutation and class switch recombination of immunoglobulin genes whereas APOBECs (apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like) have been shown to have roles in mRNA editing and in antiviral immunity. Dysregulated activity of APOBECs causes C >T transitions or C>G, C>A transversions in DNA. We have recently shown APOBEC signature mutation pattern in multiple myeloma (MM) genomes (Bolli et al Nat. Comm. 2014), and interestingly, the APOBEC mutation signature correlates with sub clonal diversity in myeloma. A role for the AID/APOBECs in generation of somatic mutations has also been proposed in a variety of other cancers based on identification of APOBEC signature mutations

In order to understand which APOBECs are dysregulated in myeloma, we performed RNA sequencing analysis of primary myeloma cells from 409 newly-diagnosed MM patients and myeloma cell lines. Our analysis showed elevated expression of several APOBEC family members; mainly A3A, A3B, A3C, and A3G. We then optimized a plasmid-based functional assay and found high cytidine deaminase activity in extracts from a number of myeloma cell lines and patient derived CD138+ cells compared to CD138+ cells from healthy donors, suggesting that APOBECs are dysregulated in myeloma. We then investigated the impact of elevated APOBEC expression/function on overall genome maintenance and acquisition of genomic changes (such as amplifications, deletions) overtime. We used shRNA-mediated knockdown of specific APOBEC proteins in myeloma cell lines and investigated the acquisition of genomic changes in control and knockdown cells during their growth in culture, using SNP (Single Nucleotide Polymorphism) arrays and WGS (whole genome sequencing) platforms. Our results with both approaches showed significant reduction in the accumulation of copy number changes (both amplifications and deletions) and overall mutation load after APOBEC knockdown. Evaluation with both the SNP and WGS showed that when control and APOBEC knockdown cells were cultured for three weeks, the acquisition of new copy number and mutational changes throughout genome were reduced by ~50%.

We next investigated the relationship between APOBEC expression/activity in MM and other DNA repair pathways. Using an in vitro HR activity assay, we measured HR activity in extracts from control and APOBEC knockdown cells. Depletion of APOBEC proteins resulted in 50-80% reduction in in vitro HR activity of the extracts. We also evaluated correlation between HR activity and gene expression using RNA-seq data from myeloma cells derived from 100 patients at diagnosis and identified the genes whose expression correlated with HR activity. Elevated expression of APOBECs 3D, 3G and 3F significantly correlated with high HR activity (R=0.3; P≤0.02), suggesting their relevance to HR. Analyzing genomic copy number information for each patient we have also observed significant correlation between higher expression of A3G and increased genomic instability in this dataset (P=0.0045).

In summary, our study shows that dysregulated APOBECs induce mutations and genomic instability, and inhibiting APOBEC activity could reduce the rate of accumulation of ongoing genomic changes. This data sheds light on biology of the disease as well as clonal evolution.