Whole genome and exome sequencing reveals the genetic landscape of burkitt lymphoma

Abstract

Burkitt lymphoma (BL) is a relatively uncommon lymphoma, but is clinically important because it is curable when diagnosed properly. BL is also an important model disease for studying cancer. Chromosomal translocations of the MYC gene are a defining feature of BL. Diffuse large B cell lymphoma (DLBCL) is the most common form of lymphoma in adults and demonstrates overlapping morphology, immunophenotype and clinical behavior with BL. The genetic causes and the role of specific mutations in BL are largely unknown. The decoding of the human genome and the advent of high-throughput sequencing have provided rich opportunities for the comprehensive identification of the genetic causes of cancer. We began by sequencing 2 complete lymphoma tumor genomes (and paired normal tissue) derived from DLBCL and BL respectively. The pattern of somatic base alterations in both DLBCL and BL genomes indicated a predominance of G A/C T and A G/T C transitions (P<10–6) suggesting that the majority of these mutations arise from endogenous processes rather than environmental exposures, as has been observed with lung cancer and tobacco. In order to comprehensively identify genes that are recurrently mutated in DLBCL and BL, we obtained a total of 95 cases of DLBCLs and 60 cases of BL. The DLBCL cases were divided into a discovery set (N=34) and a prevalence set (N=61). The Burkitt cases were also divided into discovery and prevalence sets (N=15, N=45 respectively). For each of the discovery set cases we also obtained paired normal tissue. We performed whole-exome sequencing for all of these using the Agilent solution-based system of exon capture, which uses RNA baits to target all protein coding genes (CCDS database), as well as 700 human miRNAs from miRBase (v13). In all, we generated over 6 GB of sequencing data using high throughput sequencing on the Illumina platform. We identified 525 candidate cancer genes that were recurrently somatically mutated in DLBCL and BL. We found that each tumor had an average of 20 gene alterations, which is fewer than most other solid tumors sequenced to date. Commonly implicated biological processes comprising these genes included signal transduction (e.g. PIK3CD, PDGFRA) and chromatin modification (e.g. MLL3, SETD2), affecting 17.2% and 14.8% of the total genetic events respectively. We found several genes related to cancer that were commonly mutated in both BL and DLBCL, including MYC, BTG1 and SETD2. Mutations in MYC were much more common in BL compared to DLBCL, suggesting that mutation of MYC might serve as an independent oncogenic mechanism in BL, in addition to chromosomal translocations. Many known cancer genes were found to be exclusively mutated in BL including SMARCA4, a gene known to regulate the expression of CD44 which is implicated in tumorigenesis. This study represents one of the first in-depth analyses of a BL genome and one of the largest applications of exome sequencing in cancer. Our data provide the most comprehensive genetic portrait of human BL to date, and provides a significant first step to identifying the genetic causes of the disease.