Epigenomics in Hematology

Epigenetics can be defined as inheritance of variation, above and beyond changes in the DNA sequence. In other words, epigenetics comprises the study of how cells sharing the same exhaustive DNA blueprint can appear and function so distinctly as white blood cells, hepatocytes, neurons, etc. Whereas the genome contains all of the vital information to direct the development of an organism, the epigenome dynamically filters and organizes that information into highly coordinated programs of gene expression.

Within the nucleus, DNA interacts with histone and non-histone proteins to form chromatin, which can be broadly classified as highly compacted and transcriptionally silent (heterochromatin) versus loosely compacted and transcriptionally active (euchromatin). Heterochromatin comprises two distinct classes of DNA: (1) noncoding, often repetitive, “structural,” DNA of centromeres and telomeres (constitutive heterochromatin), and (2) gene-encoding and gene-regulatory “functional,” DNA that is selectively rendered inactive in different cell types (facultative heterochromatin). When euchromatin is described as loosely compacted, the information content of its DNA is readily accessible to binding the protein and RNA machinery that regulate gene expression. Therefore, the study of epigenetics and chromatin aims to describe and understand the chromatin dynamics that orchestrate the four-dimensional symphony of molecular and cellular biology, from the (seemingly) one-dimensional score that is the genome.

The information contained within chromatin can be grossly divided into two main categories: (1) the structural genes themselves, which are transcribed and translated into proteins or act as functional RNAs, and (2) gene-regulatory regions, which control the timing and amount of transcription ( Fig. 2.1A ). The information contained in transcribed and translated regions can be interpreted using the “genetic code,” wherein the DNA sequence of the gene specifies, through a messenger RNA intermediate, the amino acid sequences of resulting proteins. While there is no universal genetic code to decipher the function of RNAs that are not translated into proteins, some such as ribosomal-RNA and transfer-RNA genes have well understood functions. In addition, several other classes of non-protein coding RNA genes with known functions exist, including small nuclear RNA (snRNA) involved in RNA splicing, Piwi-interacting RNA (piRNA) involved in silencing of transposable elements, small nucleolar RNA (sno-RNA) involved in directing the chemical modification of other RNA, and micro-RNA (miRNA) involved in translational silencing. A growing class of long noncoding RNA (lncRNA) have been identified, with a variety of proposed functions. Interestingly, these lncRNA genes appear to be regulated in much the same way as protein-coding genes. Protein-coding regions comprise approximately 1% to 2% of the genome. In contrast, the information contained in gene-regulatory regions is the “epigenetic code,” which has yet to be fully deciphered and is based on the accessibility of those regions to dynamic protein-DNA interactions, the identity of those interacting proteins, and the identity of the gene(s) whose expression is being modulated.

Open full size image

CHROMATIN STRUCTURE.

(A) Functional chromatin domains and their characteristic histone modifications and protein-binding features. (B) Higher-order chromatin structure, from least condensed (top) to most condensed (bottom). (C) Schematic of nucleosome with DNA (light blue) wrapped around histone octamer (H2A, H2B, H3, H4) having protruding histone tails. (D) Three-dimensional chromatin looping brings enhancers into close proximity with promoters via interactions with Cohesin and Mediator protein complexes.

The most dramatic example of chromatin compaction is the condensation that occurs during mitosis, making individual chromosomes visible by light microscopy and allowing segregation of replicates equally among daughter cells. A condensed or compacted chromosome is folded many times upon itself and is highly protein-bound, affording little or no access to genomic information and remaining transcriptionally silent (see Fig. 2.1B ). Contrast this with the “decondensed,” chromatin state that is necessary for DNA replication, during the synthesis phase of the cell cycle. DNA replication requires unfolding of chromatin, disruption of its protein-DNA interactions, and “unzipping,” the double helix to allow every base in the genome to be copied. When not dividing, cells maintain their chromatin in intermediate states of compaction. Actively transcribed genes and their associated regulatory chromatin regions are “open,” and “accessible,” insofar as the underlying protein-DNA interactions are readily modified and disrupted to accommodate binding of transcription factors, cofactors, RNA polymerases, and the totality of functional components underlying gene expression.

It is important to remember some key differences between genomic and epigenomic research. Whereas the genome is essentially an unvarying feature of every cell in an organism (with the important exception of T and B cells that rearrange and mutate their antigen receptor genes), the epigenome of each cell within that organism is unique. Moreover, epigenomes are fluid throughout a cell’s life span, integrating intrinsic cellular “identity,” with contextual signals to specify a program of gene expression. Finally, the mechanics of DNA replication and cell division necessarily disrupt the protein-DNA interactions that comprise the epigenome. How cells re-establish their epigenetic identity, after cell division, is not well understood.