DNA Methylation in Aging and Cancer
A central focus of the Kim lab is on the molecular mechanisms that regulate DNA methylation and other types of epigenetic change that determine self-renewal capacity, differentiation potential, and normal cellular behavior. Evidence is now emerging to suggest that epigenetic alterations, such as heavily methylated (or silenced) genomic loci, are associated with age-related disease, including cancer. DNA methylation is also strictly regulated at the other end of the spectrum, in embryonic stem cells (ESC) (Figure 1). Our group is specifically interested in studying the process of somatic cell reprogramming (induction of a pluripotent, ESC-like state) in order to better understand mechanisms of cellular aging and tumorigenesis. Moreover, by using reprogramming approaches in cells of various ages and disease states, we can learn a tremendous amount about the intracellular and/or epigenetic signals necessary for rejuvenation of senescent or aged cells and those that contribute to cancer initiation and progression.
Mouse embryonic stem cells.
Very little is currently known about the molecular underpinnings of methylation during the reprogramming process. It is clear, however, that DNA demethylation is an essential component of reprogramming somatic cells to a pluripotent state. Following the “erasure” of epigenetic memory, cells subsequently undergo a re-methylation period to restore the epigenetic landscape to one characteristic of ESC. Although poorly understood, demethylase activity present in ooplasm or ESC is thought to be responsible for the widespread epigenetic change that occurs during the early phase of reprogramming with micromanipulation techniques, such as ESC fusion with somatic cells (fESC) and nuclear transfer to ESC (ntESC). Importantly, generation of induced pluripotent stem cells (iPSC) by introduction of four reprogramming factors, Klf4, Oct4, Sox2, and Myc, also produces a pattern of epigenetic modification characteristic of an ESC-like state; however, these transcription factors have no measurable demethylase activity of their own, and we have previously shown that residual epigenetic marks are retained from the tissue of origin (i.e., blood or fibroblasts) thereby skewing differentiation fates (Figure 2). In addition, the extent of demethylation directly correlates with the age of the donor tissue: embryonic donors exhibit near-complete demethylation and tissue from older donors exhibits increasingly incomplete demethylation, resulting in restricted differentiation potential of “aged” iPSC. This implies that the aging process limits endogenous demethylation mechanisms, and that additional modulation of demethylation during iPSC generation from the tissue donor is required for complete reprogramming to occur. Hence, two scenarios exist — partial reprogramming of 1) iPSC and 2) aged cells — that allow us to model the effects of inefficient demethylation on pluripotency and fate determination and provide powerful tools to explore pathways fundamental to regulation of genome-wide methylation.
Osteogenic colonies, generated from iPSC.
To address mechanistically how global methylation patterns are determined and how changes in chromatin modification affect gene expression, we are using a number of approaches, including demethylation reporters, pyrosequencing, cDNA screens, and pharmacological approaches. Many of our studies of differentiation potential of iPSC include analysis of hematopoietic lineages, as well as other classic assays of in vitro and in vivo differentiation.
Histocompatible Pluripotent Stem Cells
Another major interest of our research program is in the development of pluripotent stem cell sources that can be used for personalized patient care. Induced pluripotent stem cells (iPSC) can be generated on an individual basis from skin biopsies or blood samples, and yet it is clear that iPSCs possess significant epigenetic memory that could impact their clinical value.
In an effort to identify alternative stem cell sources, our group has generated nuclear-transferred ESC (ntESC) and parthenogenetic ESC (pESC). Despite their consistent high quality and strong performance in differentiation assays, both ntESC and pESC require the use of oocytes (Figure 3). The limited availability of human oocytes is a significant obstacle to therapeutic application of these types of cells and to our understanding of their safety. Instead, my group aims to develop pluripotent cells adaptable in an individualized, patient-specific manner via chromosome transfer. This method uses existing embryonic stem cells (ESC) and applies targeted chromosome elimination and transfer methods to produce histocompatible, diploid somatic cell chromosome transfer ESC (ctESC). We are currently utilizing a number of functional assays for cellular potential and immune compatibility, including transplantation into a mouse model of hematological disease.
Parthenogenetic mouse oocytes.
The process of reprogramming somatic cells to a pluripotent state is still something of a black box, where very little is known about its molecular underpinnings. And so, the goal of this research is not solely to create histocompatible patient-specific stem cells, but also to develop novel tools with which to study the phenomenon of pluripotency systematically. Future research projects will involve assessing the function and epigenetic status in histocompatible ESC and differentiated tissues by monitoring methylation of imprinted and pluripotent genes.