 |
Gene Activation and Silencing in the Immune System
Summary: Stephen Smale uses the mouse immune system as a model for studying fundamental mechanisms of gene regulation in a physiological setting.
Like many biological processes, the development of the mammalian immune system and its response to host infection by microbial pathogens rely on the proper regulation of a wide variety of genes. For example, the development of a T lymphocyte involves numerous gene activation and silencing events that occur in a tightly ordered cascade, with the ultimate identity of the mature cell dictated by its developmental history and by the genes that remain active or poised for activation in response to an appropriate stimulus. A major goal of our laboratory is to elucidate the specific events that regulate gene activation and silencing during immune development and immune responses. An equally important goal is to use the immune system as a model for advancing our basic knowledge of gene regulation mechanisms. Cells of the immune system provide several advantages for basic studies of gene regulation during development and in response to acute stimulation, as primary cells are relatively easy to isolate in large quantities at defined developmental stages.
One focus is on the mechanisms of heritable gene silencing during T cell development. (These studies are funded by a grant from the National Institutes of Health.) Our interest in gene silencing began with an observation made in collaboration with Amanda Fisher and Matthias Merkenschlager (Imperial College School of Medicine, Hammersmith Hospital, London) that a key regulator of lymphocyte development, Ikaros, is localized to foci of pericentromeric heterochromatin. These foci contain the pericentromeric repeat sequences from all murine chromosomes, which assemble into a repressive chromatin environment in interphase nuclei. The Fisher and Merkenschlager laboratories showed that lymphocyte development is accompanied by the repositioning of the silent genes to the Ikaros-containing pericentromeric foci. Studies are under way to determine whether Ikaros is responsible for the pericentromeric recruitment of developmentally regulated genes, and whether pericentromeric recruitment is truly important for heritable gene silencing.
During these studies, it became apparent that our ability to understand the precise functions of Ikaros and the significance of pericentromeric repositioning was hindered by a general deficiency in knowledge of the assembly of silent chromatin. Although much is known about the hallmarks of silent chromatin, technical hurdles have slowed efforts to define the pathways involved in assembling silent chromatin at developmentally regulated genes. In particular, to study the temporal order of events involved in silent chromatin assembly, we need large quantities of primary cells maintained at an immature developmental stage. Furthermore, it must be possible to stimulate cell maturation ex vivo in a highly efficient and synchronous manner.
Murine thymocytes are unusual in that they possess all of the properties required for a temporal analysis of gene silencing. First, large quantities of primary cells maintained at a specific stage of thymocyte development can be obtained from a single mouse at 90–95 percent homogeneity. Second, these immature thymocytes, known as double-positive thymocytes, mature efficiently and synchronously when stimulated ex vivo. Third, maturation is accompanied by the permanent silencing and pericentromeric repositioning of genes encoding components of the V(D)J recombinase, the Rag and Dntt genes. Another valuable tool is a transformed cell line that can undergo the same developmental step in response to the same stimulus. The Rag and Dntt genes are not, however, permanently silenced in this line and can be rapidly reactivated by removal of the stimulus.
We have used this system to study the temporal order of events associated with gene silencing, with a focus on Dntt. The earliest events include the deacetylation of the amino-terminal tail of histone H3 and, surprisingly, repositioning of the Dntt locus to pericentromeric foci. Deacetylation begins at the Dntt promoter and then spreads in a bidirectional manner. In the transformed line, efficient histone deacetylation occurs at the promoter, but spreading and repositioning are not observed. Deacetylation and repositioning are both resistant to inhibitors of protein synthesis. Next, the loss of histone H3–lysine 4 methylation occurs, followed by an increase in histone H3–lysine 9 methylation. This latter event is an established hallmark of silent chromatin.
Temporal spreading of lysine 9 methylation was observed with the nucleation site again at the promoter, but the loss of lysine 4 methylation occurred simultaneously throughout the Dntt locus. The loss of lysine 4 methylation and acquisition of lysine 9 methylation were sensitive to protein synthesis inhibitors. DNA methylation, another well-established hallmark of silent chromatin, was observed only after fully mature T cells isolated from the spleen were induced to proliferate. These results establish an order of events associated with the assembly of silent chromatin during thymocyte development and demonstrate that the histone modification changes are nucleated at the promoter, rather than at repetitive DNA elements, as has been observed in recent studies of constitutively silent chromatin. Importantly, these nucleation events can be recapitulated in stable transfection experiments with promoter-reporter plasmids, providing a physiological assay for the definitive identification of control elements and proteins responsible for the sequential recruitment of histone modification complexes involved in developmental gene silencing.
A second focus of our laboratory is the regulation of genes that contribute to inflammation, which can be initiated by an interaction between a microbial pathogen and host macrophages. This interaction leads to the transcriptional induction of several proinflammatory cytokine genes and genes encoding other proteins that contribute to an inflammatory response. Toll-like receptors (TLRs) comprise a family of mammalian cell-surface proteins that stimulate proinflammatory gene transcription in response to lipopolysaccharide (LPS) and other bacterial products. TLRs stimulate signal transduction pathways that ultimately induce specific transcription factors. Although transcription factors involved in proinflammatory gene regulation have been identified, our knowledge of the regulatory mechanisms for these genes remains at a preliminary stage. For example, several important agents that enhance and suppress inflammation by modulating proinflammatory gene transcription have little or no impact on the known regulators of these genes.
Recently we have attempted to understand the logical organization of the genome with respect to the regulation of genes induced by LPS in macrophages. We are particularly interested in defining consistent rules that govern the contributions of chromatin structure to acute gene activation. Previous studies suggested that some inducible genes possess chromatin structures that are accessible to transcription factors and transcriptional activation in unstimulated cells and appear to require no further alterations in chromatin structure upon stimulation. In contrast, we have observed dramatic changes in the covalent modifications of histones and in the remodeling of nucleosomes by ATP-dependent nucleosome-remodeling complexes at other inducible genes following cell stimulation. By examining cells in which specific nucleosome-remodeling complexes have been depleted, we have found that LPS-induced genes can be divided into three distinct categories on the basis of their time courses of activation, requirements for protein synthesis, and requirements for specific nucleosome-remodeling complexes.
Current studies are directed toward identification of the LPS-induced factors that are responsible for, or dedicated to, the recruitment of remodeling complexes to the two classes of genes that require nucleosome remodeling prior to activation. We are also attempting to understand how an accessible chromatin structure is acquired during macrophage development at the class of genes that exhibits no remodeling requirement for gene activation. Finally, our analysis has revealed that one nucleosome-remodeling complex, the Mi-2/NuRD complex, tempers the induction of the two classes of LPS-induced genes that require remodeling for gene activation; elimination of the Mi-2/NuRD complex results in the superactivation of these genes. The Mi-2/NuRD complex therefore provides a potent anti-inflammatory function during the LPS response. We are exploring the biological relevance of this anti-inflammatory effect and the mechanism by which the Mi-2/NuRD complex is recruited to the control regions of LPS-induced genes.
Last updated March 09, 2006
|
|
|
|
