Epigenetic regulation of normal, aged and malignant hematopoiesis
Figure 2: A number of differentially methylated regions (DMRs) of the genome can be identified between long-term HSCs (LT-HSCs) and their immediate progeny within the various multipotent progenitor compartments (MPPs). Genome-wide analysis of these DMRs provides a picture of how these loci either progressively gain DNA methylation during differentiation or, as is the case in the figure above, progressively lose DNA methylation. Although DNA methylation status does not absolutely correlate with the expression level of a given gene locus, it enforces a permissive or restrictive state that we believe is responsible for the unidirectionality of the differentiation process under normal physiologic conditions.
© dkfz.de
Hematopoiesis is a tightly regulated process which facilitates the production of all mature blood cell types throughout an individual’s life-time. This process is regulated at the epigenetic level and hematopoietic differentiation is characterized by progressive epigenetic changes, which enforce programs of cellular differentiation. Therefore the analysis of the hematopoietic stem cell (HSC) and hematopoietic progenitor epigenome holds the potential to give a new level of mechanistic insight into the process of normal hematopoietic differentiation as well as into abnormal hematopoiesis across a range of pathologic settings. However, such analyses are severely restricted by the low abundance of HSC and progenitor populations within the bone marrow, with approximately 1 in 100,000 murine bone marrow cells being an HSC.
We have established an extremely productive collaboration with the group of Dr. Michael Milsom (Division of Experimental Hematology, DKFZ) in order to interrogate epigenetic changes in HSCs and hematopoietic progenitors, using new technological approaches that allow analysis of the epigenome using the limited amounts of starting material that can be harvested from HSCs and their immediate progeny. Together, we have been successful in generating genome wide DNA methylation data for highly purified murine HSCs and several MPP populations using a tagmentation-based whole genome bisulphite sequencing approach which allows increased efficiency of sequencing library preparation compared to conventional bisulphite sequencing (Figure 1). As part of a multi-group consortium, this data was combined with gene expression and proteome data from the same cell populations and published as a comprehensive resource article (Cabezas-Wallscheid et al., Cell Stem Cell 2014). Along with a follow up publication in Cell Cycle (Lipka et al., Cell Cycle 2014), this work provides a unique insight into the differential regulation of gene expression across several highly related cell populations that together form the apex of the murine hematopoietic system.
The process of aging is characterized by a progressive attrition of the hematopoietic stem cell (HSC) compartment, leading to the development of age-associated pathologies such as anemia, or the evolution of clonal hematopoiesis, which can then progress to myelodysplastic syndrome (MDS) and to leukemia. It has been demonstrated that aging is associated with changes in chromatin structure which can enforce stable changes in gene expression programs, that can in turn lead to altered cell biology and disease. Indeed, abnormal functionality in enzymes responsible for regulating the levels of DNA methylation and/or histone methylation/acetylation have been identified as causative for cellular transformation and, in the setting of the hematopoietic system, an age-associated emergence of non-malignant dominant HSC clones. The gradual evolution of normal HSCs to their leukemic counterpart, via a pre-leukemic state, is a process that normally takes several decades in humans and is dependent on the coordinate acquisition of oncogenic mutations and epigenetic reprogramming. It would be highly desirable to gain further insight into this process from the perspective of risk assessment as well as both preventative and curative therapy for diseases such as MDS and acute myeloid leukemia. However, the extremely long time course of the development of disease makes mechanistic studies very challenging, as it is almost impossible to capture the instance where causative events occur. Surrogate models of this process must therefore be used.
Our long-term goal is to identify key-factors that are epigenetically deregulated early during the leukemogenic process, which could then serve as therapeutic targets to prevent the clonal progression of pre-leukemia.