| Biography: Dr. Sen received his Ph.D., degree in chemistry from Columbia University in 1982. He made the transition to molecular biology as a postdoctoral fellow in David Baltimore's laboratory at M.I.T. and the Whitehead Institute. During this stage he developed his current interests in gene regulation. In 1987 Dr. Sen was appointed Assistant Professor in the Department of Biology and Rosenstiel Research Center at Brandies University. He earned tenure in 1991 and was promoted to Professor of Biology in 1998. He moved to his present position as Chief, Laboratory of Cellular and Molecular Biology, National Institute on Aging in 2003. |
| Research Summary: B and T cell differentiation share several common features. B lymphopoiesis takes place in the bone marrow where environmental cues commit multipotent cells to the B lineage. Close to the point of lineage commitment gene rearrangements are initiated at the immunoglobulin heavy chain (IgH) gene locus. Activation of the locus and subsequent V(D)J recombination is regulated in complex ways, and one of our objectives is to understand the molecular mechanisms that underlie this complexity. A parallel pathway operates in the thymus where multipotent cells commit to the T lineage. One important consequence is the activation and recombination of T cell receptor (TCR) b chain genes. The TCRb gene enhancer has been shown to be essential in this process and we have used it to probe this differentiation step. |
| 1. Regulatory Mechanisms in Pro-B Cells: The immunoglobulin heavy chain gene locus is spread over several megabases. Functional IgH genes are assembled in pro-B cells by gene recombination events that bring together VH, DH and JH gene segments. We have recently found that this locus is activated in discrete, independently regulated steps. An approximately 90 kb domain is activated first prior to the initiation of V(D)J recombination. This domain includes all the DH gene segments and extends till Cm. VH genes are inactive at this stage, which ensures that DH to JH recombination takes place first. Our analysis suggests that DJH recombination activates VH genes that lie closest to the DH/Cm regions. Other parts of the VH locus are activated independently: the 5'VH J558 family requiring IL-7 and the intermediate VH10 genes responding to tyrosine kinase signals. |
| The problem of IgH locus activation can therefore be broadly divided into two parts. First, regulation of the 90 kb DH/Cm domain and second, the regulation of VH genes. Our objective is to understand the molecular basis for these regulatory events. |
| 1A. The DH/Cm Locus - Within this 90 kb lies the first tissue-specific transcriptional enhancer identified, the m enhancer (mE). This regulatory element was subsequently shown to be a recombinational enhancer in artificial recombination substrates, further strengthening its importance as a regulator of IgH gene expression in pro-B cells. The presence of other recombinational enhancers in the locus was inferred from the observation that deletion of mE from the endogenous locus had little effect on DH to JH recombination. We have examined approximately 60kb of the 90kb region and found evidence for only one other regulatory sequence, which is close to DQ52. |
| Analysis of mE - We have studied this enhancer for several years from the perspective of transcriptional activation. We know the proteins that bind, the functional consequences of disrupting protein binding, proteins that interact with other mE binding proteins, and the biochemical consequences of some of these interactions. Yet, a deep understanding of the basis of enhancer function is still lacking. For example, we do not understand why certain protein binding sites need to be next to each other, or why they are spaced the way they are, or even the function of individual, or combinations of, proteins. Current studies of the m enhancer aim to address such mechanistic issues in the context of transcription and recombination. A second major theme is to study the enhancer as a modulator of chromatin structure, since this property very likely directly impacts its function as a recombination enhancer. Two examples of ongoing studies are described below. |
| a. Transcriptional Synergy between E47, Ets-1 and TFE3 To circumvent the complexity of the full enhancer (and its 18 associated proteins), we have taken the approach of functionally dissecting smaller domains of the enhancer. One such domain comprises the motifs mE2, mA and mE3 that bind the proteins E47, Ets-1 and TFE3, respectively. Several lines of evidence indicate that these motifs work together. To address this we are reconstituting Ets-1 dependent synergy between E47 and TFE3 using purified factors and looking for additional proteins biochemically and genetically. The motivation for such studies comes not only for their relevance to m enhancer function, but also to understand the molecular basis of combinatorial control. |
| b. m Enhancer and Chromatin Structure The goals are to determine the effects of individual, or combinations of proteins, on chromatin structure. Towards this end, enhancer-containing plasmids are assembled into chromatin in vitro using a fully reconstituted system (in collaboration with Mike Pazin, Massachusetts General Hospital), in the presence or absence of purified enhancer binding proteins. We use structural assays such as nuclease digestion, nucleosome positioning and restriction enzyme accessibility, and functional assays such as in vitro transcription and RAG cleavage (in collaboration with David Schatz, Yale). Recent results show that TFE3 alone can find its site and bind to nucleosome assembled plasmids. This results in nucleosome positioning and induction of a nuclease hypersensitive site. Our immediate objectives are to understand the contribution ETS-domain proteins and E47 in this context, and to identify chromatin-remodeling activities in B cell extracts that are required for structural alterations. |
| Other Sequences that Activate DH-Cm - We recently identified a possible regulatory site close to DQ52. We are currently characterizing the region for transcriptional and recombination enhancer activity, as well as identifying proteins that mediate the effects. In collaboration with Gene Oltz (Vanderbilt), we will analyze the chromatin structure of alleles deleted for this sequence, or a combined deletion of this sequence and the m enhancer, to determine their contribution to the overall structure of the DH/Cm locus. The double deletion will also indicate whether these sequences are sufficient to activate the 90kb domain, or whether additional sequences are likely to contribute. In parallel, we are continuing to map nuclease hypersensitive sites in the remaining 30kb that we have not yet analyzed. If additional sites are identified, we will evaluate their role in locus activation by deleting them, and testing their function in artificial substrates. |
| 1B. The VH Locus - We have evidence for three independently regulated domains of VH genes: the 5' VH J J558 genes are IL-7 responsive, the 3' VH 7183 and SM7 genes are activated by DJH recombination, and the intermediate VH10 genes are activated by the v-abl tyrosine kinase. Our immediate objectives are to i) confirm the model that DH-proximal VH genes are activated in response to DJH recombination, ii) to identify the normal signals that activate VH10 and co-regulated genes and iii) to study the mechanism of VH allelic exclusion. An example of ongoing studies is described below. |
| Implications for Allelic Exclusion of VH Genes - Allelic exclusion refers to the phenomena that B and T lymphocytes express only one antigen receptor. Though this could result from low probability of generating two functional rearrangements, it has been convincingly demonstrated that allelic exclusion at IgH (and TCRb) is actively regulated by a feedback mechanism. Cells sense IgH protein via the pre-B cell receptor and terminate further VH to DJH recombination. Based on our recent insights into the activation of of VH genes, we proposed the simple hypothesis that allelic exclusion is the opposite of VH gene activation. For example, since IL-7 activates VHJ558 genes, according to our model loss of IL-7 signals results in allelic exclusion of this family. We are currently testing several predictions of this model as well as investigating the mechanism of VH gene inactivation. |
| 2. Regulatory Mechanisms in Pro-T Cells: TCRb chain gene recombination and expression requires an enhancer located several kilobases 3' of the Cb2 exons. We have initiated a systematic analysis of this enhancer with the goal of identifying critical motifs (and associated DNA binding proteins) that are responsible for activating it at the earliest stages of T cell development. The working hypothesis is that thymic environmental signals that commit a multipotent cell to the T cell lineage also activate the TCRb enhancer. Thus, working back from the enhancer provides one route to identifying the signaling pathways that operate in the earliest thymocytes. We identified two novel sequence motifs in the TCRb enhancer that lie between two composite ETS/CBF elements. We plan to identify these proteins biochemically and/or genetically. Their role in early thymocytes will be further addressed once we have antibodies and gene sequences. |
| A second interesting aspect of these studies is the relationship of the enhancer to MAP kinase signals. Activation of MAP kinases further increases TCRb enhancer activity. Interestingly, under these conditions the intervening sequence motifs are no longer necessary for enhancer activity; that is, enhancer activity is only dependent on the ETS and CBF elements. Because ETS and CBF proteins are expressed in multipotent hematopoeitic precursors, this leads to a model for the initiation and maintenance of enhancer activity. We propose that an early thymocyte may receive a MAP kinase-activating signal from the thymic microenvironment, which turns on the TCRb enhancer using pre-existing ETS and CBF proteins in these cells. Once the MAP kinase signal has terminated, TCRb enhancer activity may be maintained by newly expressed bE5 binding proteins. This model provides a pathway by which gene expression mediated by transient differentiation signals may be maintained at subsequent stages. |
| 3. Function of NF-kB Proteins: NF-kB proteins are a family of inducible transcription factors that allow cells to respond to extracellular stimuli. The diverse stimuli that activate NF-kB and the distinct cellular responses that ensue raise the question as to how specificity of the response is regulated. This complexity is most likely a reflection of the several different Rel proteins that constitute the NF-kB family and the several different IkB proteins that inactivate them. For example, there may be differences in the way Rel proteins are sequestered in the cytoplasm, different signals may target different IkBs, and different family members may activate different genes. However, there are very few well characterized examples of such differences and even fewer molecular mechanisms to explain them. Our long-term interest is to attempt to unravel some of this complexity, particularly in cells of the immune system. Current research interests are summarized below. |
| 3A. Subcellular Dynamics of NF-kB Proteins - We recently showed that NF-kB/IkBa complexes in cells are in constant flux due to entry and exit from the nucleus. This is because nuclear localization signals (NLS) in NF-kB (particularly the p50 component) are not well hidden by association with IkBa. This permits the complex to enter the nucleus. Once in the nucleus, a strong nuclear export sequence (NES) in IkBa takes over and directs the complex back to the cytoplasm. We have proposed that the nuclear export function of IkBa ensures against spurious NF-kB-dependent gene activation that may occur due to inappropriate nuclear entry of NF-kB. IkBa-dependent export may also participate in terminating an NF-kB response when the stimulus that activated it is no longer present. |
| This dynamic state is cell-type specific. We have found that c-Rel is associated with IkBa only in B cells, but not in pre-B cells or T cells. This means that c-Rel constitutively transits in and out of the nucleus only in B cells. Yet, at the end of a TNFa signal in T cells, c-Rel transiently associates with IkBa presumably to be retrieved from the nucleus. This IkBa associated c-Rel must be ultimately transferred to IkBb to re-establish the resting state of the cell; how this is brought about remains unclear. Interestingly, IkBb and IkBe do not have export potential; instead they hide the NLS of Rel proteins more effectively, thereby preventing nuclear entry. Ongoing effort is directed at understanding how IkBa synthesis and post-translation modifications help to time the end of NF-kB-dependent transcription, and the functional consequences thereof. We are also investigating the functional significance of the B cell-specific c-Rel/IkBa complex. |
| 3B. Functional Differences Between p65 and c-Rel - p65 and c-Rel are two closely related, but functionally different, Rel family members. One of the striking differences pertinent to immune function is that T cells in c-Rel-/- mice do not make IL-2 in response to TCR signals. However, IL-2 dependent proliferation is normal when these cells are provided with exogenous IL-2, as is p65 induction in response to TCR. The question is where is the crucial difference between the two proteins that allows only one to activate IL-2; conversely, where is the c-Rel responsive element in IL-2 that does not respond to p65, and what is the basis for the difference? (The DNA binding specificity of the two proteins is quite similar, suggesting that target specificity is unlikely to be a major factor in their distinctive properties). This problem also has some practical appeal because the part of c-Rel that distinguishes it from p65 is a potential target for immunosuppressive drugs that will block T cell activation (by blocking IL-2) without affecting p65-dependent transcription. |
