Back to search

National and Kapodistrian University of Athens

Μελέτη της δυναμικής μεταγραφικών συμπλόκων in vivo

Abstract

dc:description

The first line of antiviral response is laying in the production and secretion of the interferon type I proteins, which, subsequently bind to their specific receptors at the surface of healthy and infected cells in order to activate a large number of antiviral genes. The products of those genes (transcription factors, cytokines etc) act by inhibiting the penetration or/and replication of the virus inside the cells. The infection of human cells with different types of viruses induces the transcription of the interferon beta (IFN-β) gene, through the coordinate activation of three distinct groups of transcription factors (NF-kB, ATF-2/c-JUN, IRFs). The latter bind to the enhancer of the IFN-β gene in a cooperative manner and they create, in combination with the architectural protein HMGI(Y), a nucleoprotein complex, known as enhanceosome. The enhancesome, after its formation the nucleosome-free enhancer, instructs the ordered recruitment of several so-activators and basal transcription factors at the adjacent promoter. The recruitment program culminates with the sliding of the nucleosome masking the promoter, allowing the RNA polymerase II binding and the subsequent transcriptional initiation. Both the cooperativity in the binding of transcription factors to the enhancer of the gene and the synergy in the recruitment of the co-activators at the enhanceosome, imply an extremely complex and accurate mechanism of combinatorial control of the transcriptional activation of the IFN-β gene. Another striking feature of the IFN-β gene expression, as well as many other cytokine genes, is that even under optimal conditions, only a fraction of the cells in the population (~22% for the IFN-β) expresses the cytokine gene at any given moment. This heterogeneity is known as stochasticity and it is observed, usually, in biological systems or procedures, characterized by high complexity and low concentrations of the implicated molecules. One attractive model proposes that the stochasticity is correlated with the transcription process itself and therefore is likely associated with the inherent complexity of enhanceosome assembly on the IFN-β enhancer/promoter. Since the enhanceosome assembly is a cooperative process, we wondered whether limiting concentrations of one or some of the IFN-β activators account for the stochastic assembly of enhanceosomes, thus leading to binary transcriptional switches. Our results showed that at least two of the transcription factors bound to the IFN-β enhancer, specifically NF-κΒ and IRF-7, play a true limiting role in the possibility of the IFN-β gene expression. This conclusion derived from the fact that exogenous increase of their concentration induces a significant increase at the percentage of cells that finally respond to the virus infection. The low concentration of those factors arise a reasonable question regarding the mechanism by which these factors locate their special binding sites nearby their target genes within the genomic “chaos” in the 3D nucleus. We focused, mainly, to the mechanism of the limiting NF-κΒ targeting to its target genes, and specifically to the IFN-β, because, firstly NF-κΒ is the first transcription factor nucleating the enhanceosome assembly and secondly, there have been found (by genome-wide methods or bioinformatics analysis) millions putative NF-κΒ binding sites within the human genome. We form a working hypothesis that some of these binding sites are specific in receiving immediately NF-κΒ and transferring it at the target genes through interchromosomal interactions. By applying several technically advanced molecular techniques, such as 4C-ChIP and DNA FISH, we identified three genomic regions located in different chromosomes, interacting specifically with the IFN-β gene through NF-κΒ. All these regions are localised away from transcribed regions contain a specialized NF-kB binding site as a part of an Alu repetitive element and they have previously, unknown function. Our experiments showed that these interchromosomal interactions (involving only one allele of each interacting locus) are virus inducible and occur at high frequencies during enhanceosome formation, whereas they are significantly reduced at the transcriptional initiation or extension. In order to study the functional role of these interactions on the transcriptional activation of the IFN-β gene, we performed the RNA/DNA FISH technique. Surprisingly, this method revealed that the IFN-β gene is expressed monoallelically. Importantly, the single transcriptionally active IFN-β allele is the one interacting with -at least- one of the 3 genomic regions, described above. These results lead to the reasonable hypothesis that an increase the frequency of the inter-chromosomal associations would increase the probability of the IFN-β expression. Indeed, transient transfection of HeLa cells with multiple copies of each of the 3 genomic regions resulted to significant increase to both the percentage of the IFN-β expressing cells and the total IFN-β mRNA levels. Additionally, we found that the positive effect of the multiple copies on the IFN-β expression levels is dependent on the specialized NF-κΒ binding site present in these regions. This phenomenon is explained by the fact that NF-κΒ associates first with the 3 genomic regions and then is transferred to the IFN-β locus, as revealed by ChIP experiments with chromatin isolated by cells infected with virus in time course manner. Furthermore, the observation that the 3 genomic regions exist in physical proximity even before the entrance of NF-κΒ in the nucleus suggests that probably they form an “NF-kB receptor center”, which can receive immediately the NF-κB and distribute the factor to the promoters of selected genes via interchromosomal interactions. We do not know the mechanism assuring the immediate binding of NF-κB to the genomic regions of interest and the specificity of these interchromosomal interactions since not all the NF-κB regulated genes (e.g. IL-8) are recruited to the NF-κB center. We can imagine that the NF-κB receptor center is localized in specialized subnuclear regions to which the incoming NF-κB can have instant access. Support for such a model comes from the fact that the affinity of NF-κB for these specialized sites is 2-3 times lower than that of the IFN-β gene, implying that its preference for the receptor center could be due either the local nuclear architecture, the DNA-induced allostery and/or other proteins localized in this microenvironment. In parallel, we studied the mechanism driving the long-range movements of the described genomic regions. Our preliminary results showed that the observed translocations are ATP-dependent and, furthermore they require the presence of some co-activators, like the CBP/p300 and the chromatin remodelling complex SWI/SNF. Certainly, the clarification of the underlying mechanism demands additional extensive research using molecular, biochemical and cytological methods. All the previous results prove that the stochastic monoallelic expression of the IFN-β gene as response to virus infection is dependent on interchromosomal interactions with 3 genomic regions. These interactions mediate the binding of the limiting factor NF-κB to the enhancer of a single IFN-β allele, which is selected probabilistically. The bound NF-κΒ orchestrates the enhanceosome assembly in this allele leading to the transcriptional activation. Furthermore, we studied the IFN-β expression at the later time-points of viral infection (8-10 hours after virus). We observed that the IFN-β production is significantly enhanced, because both the percentage of IFN-β expressing cells and the number of expressing alleles per cell are increased. The gradual conversion of mono-allelic expression to multi-allelic is independent from interchromosomal associations. In order to elucidate the underlying mechanism, we first tested the possible role of the secreted IFN-β protein in a mean of positive feedback loop. Our experiments, validated this assumption, since they proved that the initially and mono-allelically produced IFN-β triggers the transcriptional activation of more alleles in additional cells, through the IFN-β signal transduction pathway and specifically, through the activation of the IRF-7 factor. Since, the IRF-7 is one of the limiting transcription factors of the IFN-β gene, the increased intracellular concentration probably ensures its binding to the promoter of initially non- expressing alleles, where it organizes the enhanceosome assembly, triggering a second wave of IFN-β production by additional cells and additional alleles per cell. Summarizing, in our current study we present strong evidence for a model of biphasic IFN-β gene expression. According to this model, at the early phase virus infection the IFN-β gene is expressed stochastically from a single allele in a small proportion of cells. The choice of the allele to be expressed is dependent on interchromosomal associations between the IFN-β gene and at least one of the three distinct genomic loci that could mediate binding of the NF-κΒ to the IFN-β enhancer. Then, the limiting NF-κΒ nucleates enhanceosome assembly, a prerequisite for chromatin remodeling and activation of transcription from this allele. The secreted IFN-β protein induces increased intracellular levels of the IRF-7 factor, which in turn orchestrates the enhanceosome formation in the remaining alleles in additional cells. Thus, the IFN-β enhancer functions in a non-linear fashion by working as a signal amplifier. Finally, in the current study we showed that it is possible to manipulate experimentally the probability of the IFN-β gene expression by two distinct ways. The first is lying on the direct increase of the concentration of the limiting transcription factors. As a result, the artificially high concentration factor has high probability to find and bind randomly to the enhancer of any allele in every cell. Then, the bound factor will recruit cooperatively the rest components of the enhanceosome, leading to the transcriptional activation of multiple alleles in many more cells. The second mechanism exploits the role of the interchromosomal interactions. It is the first time that someone reveals that the expression probability of a gene could be raised by increasing the copy number of an intergenic (noncoding) region, which is capable to receive immediately a low abundance factor and deliver it to the target gene. Remarkably, the construction of stable cell lines with multiple copies of such a genomic region resulted to: 1) increased expression levels of the IFN-β gene, 2) earlier and more immediate response to the virus infection and finally, 3) higher sensibility of the cells, which become able to “sense” and respond to lower viral titles.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2009

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Apostolou, Effie
  • Αποστόλου, Ευτυχία

Subjects

dc:subject × 20

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/22915

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Apostolou, Effie; Αποστόλου, Ευτυχία. Μελέτη της δυναμικής μεταγραφικών συμπλόκων in vivo. National and Kapodistrian University of Athens, 2009. http://hdl.handle.net/10442/hedi/22915