February 1, 2011 (Vol. 31, No. 3)
Richard A. A. Stein M.D., Ph.D.
Field Is Shedding Light on Biological Processes Involved in Development and Differentiation
In his book The Triple Helix, Richard Lewontin underscores the importance of integrating the contribution of genes, the organism, and the environment, if one intends to fully understand biological systems. The past few decades, and in particular the most recent years, have revealed that genetics alone is not sufficient to explain phenotypes. Epigenetics, defined as heritable changes in gene expression that do not involve alterations in the sequence of nucleotides, has assumed increasingly important roles.
Epigenetics promises to shed light on biological processes implicated in development and differentiation, the interface between organisms and environmental perturbations, and carcinogenesis.
One of the major questions during development is not only how the state of cell differentiation is achieved but, in addition, how it is maintained once it is established. “The main focus of our work is to understand how cells organize their specific regulatory landscapes,” says Gioacchino Natoli, M.D., group leader in the department of experimental oncology at the European Institute of Oncology, Milan.
While epigenetic mechanisms are required to establish a genomic landscape during differentiation, a more interesting question is to understand their involvement once cells become differentiated. “It is not sufficient to have epigenetic mechanisms, even though they are important, but the constant and continuous supervision by lineage-specific transcription factors is required for maintenance of lineage determination,” says Dr. Natoli.
This view is supported by several pieces of evidence, which reveal that the deletion of lineage-specific transcription factors from fully differentiated cells is followed by the loss of differentiation. Moreover, the constant action of environmental perturbations modifies the epigenome, which subsequently needs to revert to the initial stage from before the perturbation.
Transcription factors are constantly binding to the DNA and continually shape the chromatin landscape, and whenever they sense an environmental change, they may activate a different set of genes or identical genes at different times. An attractive possibility is that lineage determining transcription factors act mainly by organizing the repertoire of enhancers, which are known to be strongly cell type-specific, and an emerging concept is the involvement of a specific three-dimensional landscape, in which the spatial organization of the genome ensures that enhancers correctly activate only the right promoters. “This is the new idea that is coming out and requires additional work,” explains Dr. Natoli.
“We recently identified and characterized a new type of epigenetic control, a unique type of methylation that occurs on p53 and is quite instrumental in determining cell fate,” says Nicholas La Thangue, Ph.D., professor and chair of cancer biology at the University of Oxford.
At the “Oxford Symposium on Epigenetic Mechanisms in Health and Disease,” Dr. La Thangue talked about this modification, which occurs on an arginine residue located in the p53 tumor suppressor region and is mediated by an enzyme called protein arginine methyl transferase 5.
A second topic of interest in Dr. La Thangue’s lab is the retinoblastoma protein pRb, which acts as a gatekeeper by regulating the G1 to S phase transition during cell-cycle progression and is frequently mutated in tumors. We found that a lysine methyl transferase, Set7/9, targets pRb.”
In addition to unveiling this epigenetic modification, which is unlike the one in p53 where the methylated residue is an arginine, Dr. La Thangue and collaborators made another fundamental discovery when, after mapping this specific lysine residue that becomes methylated by Set7/9, they noticed that the same residue governs another level of control, which is pRb phosphorylation. However, pRb phosphorylation is mediated by cyclin-dependent kinases, which regulate cell-cycle progression, and by phosphorylating and inactivating pRb, they release its negative regulation of the cell cycle.
In its methylated state, pRb phosphorylation is switched off and cells enter a quiescent, nonproliferating state. “The provocative thing is the interplay that we observed between different modifications on pRb,” notes Dr. La Thangue. “This is a very nice example of interplay between different types of epigenetic marks that can occur on other proteins.”
This complex interplay between different types of epigenetic modification makes these proteins become promising leads in designing therapeutic targets, particularly when targeting one protein in a pathway can impact a global cellular event. “There are some truly superb drug targets to focus on.”
Janus Kinase 2
One of the proteins implicated in many cellular processes, including cell-cycle progression, recombination, and apoptosis, is Janus kinase 2 (JAK2), an enzyme involved in physiological and leukemic hematopoiesis. Until recently, JAK2 was thought to reside in the cytoplasm and to perform its function via cytoplasmic signal transduction networks, such as the JAK-STAT pathway. However, recent findings challenged this model when they revealed the existence of a previously unrecognized pool of JAK2.
“We found that this signaling pathway extends all the way down to the chromatin, and the enzyme is also in the nucleus,” explains Tony Kouzarides, Ph.D., Royal Society professor at the University of Cambridge and deputy director at the Gurdon Institute.
Dr. Kouzarides and collaborators revealed, by in vitro and in vivo experiments, that JAK2 phosphorylates tyrosine 41 of histone H3. This specific tyrosine is found in a region on the histone tail that is important in shaping the dynamic properties of nucleosomes and influences nucleosome remodeling.
In addition, Dr. Kouzarides and colleagues showed that the phosphorylated histone prevents the binding of a heterochromatin protein, HP1α, to DNA. HP1α normally represses transcription, and when it is excluded from the DNA as a result of this histone modification, it leads to transcriptional activation, providing a fascinating mechanistic insight into the cellular and molecular consequences of this epigenetic change.
“We have shown that some of the genes that this modification regulates are cancer genes. This is a new paradigm,” explained Dr. Kouzarides. This finding has important clinical applicability, because in some leukemia patients the JAK2-STAT5 pathway becomes constitutively activated as a result of a mutation that increases JAK2 activity and causes the unregulated displacement of HP1α from the chromatin.
At the recent Oxford meeting, Kevin Struhl, Ph.D., professor in the department of biological chemistry and molecular pharmacology at Harvard Medical School, talked about an experimental model of oncogenesis that his group developed to dissect the impact of environmental signals on initiating and maintaining epigenetic states.
In this system, the ligand-binding domain of the estrogen receptor and the Src kinase oncoprotein were used to create a fusion protein that is inducible by tamoxifen and activates NF-κB. By using this construct, Dr. Struhl and collaborators revealed that, subsequent to an Src induction as short as five minutes, the nontransformed breast epithelial cell line was becoming transformed 24–36 hours later.
Most importantly, after becoming transformed, the cells could be propagated for many generations and stayed transformed even though Src was no longer induced. This finding, analogous to the one described during development, revealed that a transient inflammatory reaction was able to activate a positive feedback loop and maintain an epigenetic state for several generations in the absence of the originating event, providing a mechanistic link between inflammation and malignant transformation.
“The inducer starts a positive feedback loop, and once the loop is activated the inducer is not needed any longer because there is a self-propagated mechanism,” explains Dr. Struhl. This phenomenon, known as an “epigenetic switch”, becomes an intriguing idea in the context of malignant transformation.
Mutations and DNA methylation are two examples of stably propagated modifications that, even though they are not sufficient as single events, they contribute to tumorigenesis when acting in combination.
“This is different; it is a switch, analogous to switching cell fates, and we think that it can be a step in tumorigenesis—not by itself, but in combination with other events,” says Dr. Struhl. More recently, research in Dr. Struhl’s lab identified several miRNAs that, upon transient upregulation, were able to induce this epigenetic switch and generate a stably transformed state. Also, this appeared to regulate tumor suppressor genes and activate the positive feedback loop induced during Src-initiated transformation.
In molecular oncology, one of the recent landmarks is the identification of epigenetic targets, which have become promising leads for therapeutic drug development. Epigenetic drug targets have been classified into “writers”, which add chemical groups, “readers”, which recognize the epigenetic modifications, and “erasers”, which remove the marks.
“There has been intense research interest in pharmaceutical companies and academia to develop prototype drugs that target the enzymatic components of the epigenetic machinery,” says James E. Bradner, M.D., assistant professor of medicine at Dana Farber Cancer Institute and Harvard Medical School.
Dr. Bradner and colleagues are focusing on epigenetic readers, which transmit signals from chromatin to the gene-expression machinery by recognizing and binding the amino acid site chains on histone proteins. “In our view, epigenetic reader proteins emerged as a potentially fantastic target for discovery chemistry.”
Recent advances in Dr. Bradner’s lab led to the development of a biochemical method to study epigenetic reader proteins, and catalyzed the development of the first potent and selective inhibitor of acetyl-lysine recognition motifs, known as bromodomains.
In collaboration with Steffan Knapp’s group from the University of Oxford, Dr. Bradner’s lab reported that the targeted use of a small molecule, JQ1, which is a competitive bromodomain inhibitor, in a rare and invariably fatal form of squamous carcinoma, displaced an oncoprotein involved in a chromosomal translocation and exhibited antiproliferative effects.
“This work firmly establishes bromodomain inhibition as a therapeutic rationale for targeted therapy and is approximately a year away from reaching patients in the clinic.” More recent work in Dr. Bradner’s lab explores JQ1 derivatives with improved therapeutic properties. “We anticipate a ligand emerging from this research for clinical development within hopefully the next one to two years.”
A topic that attracted great interest in immunology is understanding not only how transcriptional and epigenetic mechanisms are used to turn genes on but, also, how some genes are turned off in specific cells that, as a result, do not mount an immune response, a phenomenon known as immune tolerance. This is particularly important, because even though we are exposed and respond to potentially harmful antigens on a daily basis, we are at the same time surrounded by innocuous antigens.
“One of my interests over the past couple of decades has been to understand immune tolerance, and find out what keeps T cells from producing inflammatory cytokines in response to ‘self’ antigens,” says Andrew D. Wells, Ph.D., associate professor of pathology and laboratory medicine at the University of Pennsylvania School of Medicine.
Results from many labs indicate that once a T cell has become anergic, it does not respond and does not produce cytokines when subsequently provided with a co-stimulatory signal, suggesting the existence of a type of negative memory that tells the cell not to engage in an immune response. “This led to our hypothesis, years ago, that a more stable modification could be involved, and that perhaps the genes have been silenced through epigenetic mechanisms,” explains Dr. Wells.
The IL-2 and IFN-γ genes provide valuable systems to examine changes that the right combination of signals causes in the local chromatin structure. It is known that primed T cells can produce inflammatory cytokines very quickly, while naïve T cells need to go through several hours of gene remodeling before expressing them.
“In T cells receiving the right antigenic and co-stimulatory signals that after a primary proliferative phase subsequently came back to rest, we found that the promoter structure was different from the one in the naïve cells, with less DNA methylation and more acetylation of histone H3, which are modifications telling the cell that these genes should be kept accessible and turned on quickly again in response to the right signals.”
While it is known that the structure of cytokine gene regulatory regions changes in cells that become tolerant, the enzymes that mediate these processes are ubiquitously expressed, and they do not bind DNA by themselves but are instead recruited to the chromatin.
“We became interested in looking at transcription factors involved in epigenetically silenced genes,” explains Dr. Wells. Two of these transcription factors are the focus of on-going research in Dr. Wells’ laboratory. One of them, FOXP3, is expressed only in a subset of cells called regulatory T cells, which were unveiled during the last 10–15 years and are important for immune tolerance. While anergy turns off T cells when they are not supposed to mount an immune response, regulatory T cells orchestrate an additional layer of control, with unique anti-inflammatory properties.
“We found that FOXP3 binds endogenous cytokine genes in regulatory T cells and remodels chromatin to make them more silent.”
The second transcription factor that Dr. Wells’ lab is focusing on, a protein called Ikaros, is expressed in all T cells. This protein was discovered many years ago and functions in the development of leukocytes in the bone marrow and thymus. “But quite a lot of Ikaros is expressed in mature T cells, and nobody looked carefully at what it is doing there.”
While examining the sequence of the IL-2 promoter, which is well characterized, Dr. Wells and colleagues found a few regions that could function as Ikaros binding sites. “We knew that Ikaros can be a powerful transcriptional repressor, and we reasoned that it could perhaps have a role in mature T cells, in keeping IL-2, and perhaps other genes, turned off under circumstances when a T cell should not make these factors.” Indeed, the Wells lab went on to show that T cells with reduced Ikaros activity tend to overproduce cytokines like IFN-γ and TNF-α, and can cause fatal inflammatory colitis when transferred into mice.
Findings from many labs that use various experimental approaches, and different model systems, reveal that epigenetic modifications play fundamental roles not only during development, but also in shaping biological processes in differentiated adult cells and organisms. As a result of recent advances in this field, old paradigms are replaced by new concepts, and well-known phenomena find mechanistic explanations, underscoring the dynamics of scientific inquiry, and promising attractive prophylactic, diagnostic, and therapeutic applications.