The mechanisms of disease we describe here are not “strictly” epigenetic, insofar as they are all predicated on changes in genome sequence or structure (genetic mutations). Nonetheless, our insights into dis ease pathogenesis and development of novel therapeutic targets have been vastly informed by understanding the ways in which these genetic changes drive aberrant chromatin regulation and gene expression. The examples given below represent only a subset of the known epigenetic drivers of disease.

Sickle cell anemia has long been known to result from a point mutation in the hemoglobin beta gene. The severity of this often life-threatening hemoglobinopathy is attenuated in patients having increased expression of the fetal gamma hemoglobin variant, a trait known as hereditary persistence of fetal hemoglobin (HFPH). Genome-wide association studies in patients with HFPH identified frequent single nucleotide polymorphisms (SNPs) in a small num ber of noncoding regions, near the BCL11A gene on chromosome 2. Subsequent studies have elegantly demonstrated that these SNPs are in erythroid-specific enhancers, modulating BCL11A expression. The HFPH-associated SNPs diminish binding of transcription factors GATA1 and TAL1, which results in decreased expression of BCL11A. Because BCL11A is required for efficient silencing of fetal hemoglobin expression, sickle cell anemia patients having these common variant SNPs demonstrate elevated fetal hemoglobin throughout adulthood and are often protected from the most severe manifestations of the disease. Just as sickle cell anemia is among the most striking examples of disease, caused by a point mutation in the coding region of a gene, these BCL11A enhancer SNPs demonstrate the power of gene-regulatory elements to modulate the sickle cell disease phenotype.

Chromosomal translocations that result in aberrant expression of oncogenes or leukemogenic transcription factors are another common mechanism of disease. The classical example of this is Burkitt’s lymphoma, in which t(8;13) translocations juxtapose the highly active immunoglobulin heavy chain enhancers and the c-myc oncogene, driving myc overexpression and oncogenic transformation of mature B cells. Similarly, many different translocations have been identified in T acute lymphoblastic leukemia (T-ALL), whereby overexpression of master transcription factors such as TAL1, LMO1, LMO2, and HOX11 is driven by chromosomal rearrangements involving the T-cell receptor loci.

An alternate mechanism driving TAL1 overexpression in T-ALL has recently been described, in which small genomic insertions (2 to 18 bp) upstream of the TAL1-coding region introduce novel binding sites for the MYB transcription factor. This aberrant MYB binding recruits additional transcription factors RUNX1, GATA-3, and TAL1, as well as the HAT CBP and forms a super-enhancer driving leukemogenic TAL1 overexpression.

Many different translocations resulting in fusion of the mixed lineage leukemia (MLL1/KMT2A) gene, located on chromosome 11q23, with over 70 different partner proteins have been identified in infant ALL and therapy-associate acute myeloid leukemia (AML). Only recently have the mechanisms underlying the leukemogenic nature of these translocations been elucidated. Leukemogenic MLL1 fusion proteins fuse the N-terminal targeting domain with a transcription elongation factor, such as ENL or AF9. The resulting fusion protein drives overexpression of common MLL1 targets by recruiting the DOT1L complex (having H3K79 methyltransferase activity) and the positive transcription elongation factor b (P-TEFb) complex (containing CDK9 and phosphorylating RNA PolII). Moreover, a subset of leukemogenic MLL1 fusions can inhibit the transcriptional repressive activity of PRC1. In summary, MLL translocations in ALL and AML define a paradigm of leukemia development, based upon transcriptional dysregulation through aberrant targeting and control of transcription elongation activity.

As noted earlier, inactivating mutations in components of chromatin remodeling complexes, such as SWI/SNF, have been identified in a wide variety of human cancers. For example, a recent study found mutations in the ARID1A subunit of SWI/SNF in 17% of Waldenström’s macroglobulinemia cases, and patients with ARID1A mutations had more aggressive disease features. In addition to their nucleosome remodeling activities, chromatin remodeling complexes contribute to three-dimensional chromatin structure, participate in DNA damage repair, modulate transcription factor binding, and recruit histone-modifying enzymes. Precisely how disruption of these many chromatin regulatory activities contributes to disease is an extremely active area of research.

In addition to these epigenetic contributions to disease development, much interest has evolved in potential epigenetic mechanisms of resistance to existing cancer therapies. One example of this is resistance of T-ALL to γ-secretase inhibitors (GSIs), used to target abnormal NOTCH1 activation. In vitro, treatment of T-ALL cell lines with GSIs kills a large proportion of cells but leaves behind a “persister” population of GSI-resistant cells. If GSI treatment is removed, these persister cells revert to their prior GSI-sensitive state, suggesting an epigenetic mechanism of drug resistance. A screen of chromatin regulators, required for persister cell viability, identified the bromodomain containing 4 protein (BRD4), a key factor in activating transcriptional elongation. This study, and many others, has ignited broad interest in other potential epigenetic mechanisms of therapy resistance, as well as BRD4 as a specific therapeutic target .