The “EPI-B-PLASMADIFF” project (36 months, 4 teams) will address scientific issues pertinent to the B-cell-lineage and their implications in immunopathology, with specific focus on antibody-secreting plasma cells. Antibody secreting cells are critical effector cells and long-lived sentinels for immune memory. B cell maturation should be tightly regulated to ensure efficient immune response without autoimmunity or immune deficiency. On the transcriptional level, the differentiation of B cells into plasma cells is associated with substantial and coordinated changes in the gene expression profile, which fall into two main categories: the loss of B cell-associated transcripts and the acquisition of plasma cell gene expression program. Although the role of the complex network of transcription factors involved in PCD has been investigated, the mechanisms regulating key plasma cell differentiation transcription networks remain poorly known. Little is known about the role of epitranscriptomic modifications in B to plasma cell differentiation and how it could regulate fundamental processes during normal plasma cell differentiation. In this proposal, we seek to characterize and understand the epitranscriptomic remodeling and the transcriptional, translational and epigenetic impacts during B to plasma cell differentiation with a particular focus on m6A modifications. We would like to reveal both the pathways involved in this remodeling and their downstream effects. The rationale of our project is in line with the identification of major epitranscriptomic changes during plasma cell differentiation, which was identified by mass spectrometry, and preliminary data supporting a role of the enzymes involved in m6A modification in the biology of plasma cells. We plan to identify upstream pathways mediating these changes and their downstream transcriptional and translational impacts. These studies will take advantage of a unique combination of powerful models and new technologies that are fully mastered by the four partners of the project.

Duplication of a complete genome is a formidable task that requires to be perfectly controlled to avoid the transmission of mutations or rearrangements to the daughter cells. However this process is a bumpy road: the replisome, the huge proteic machinery required for copying the DNA, is facing many impediments that interfere with its progression and increase replication stress. Other the years many sources from endogenous origin have been shown to perturb replication such as genomic features (tandem repeats, G-quadruplexes, highly transcribed genes, topological constraints…) or by-products of cellular metabolism (formaldehyde, reactive oxygen species…). In addition exogenous sources from natural origin (ultraviolet light, ?-irradiation…) or non-natural origin (pollution, food toxins, cigarette smoke, chemotherapeutic treatments…) are also perturbing replisome advancement. Previous work from many labs shown that cells have evolved a plethora of mechanisms to stabilize and process stalled replication forks in order to prevent forks collapsing that could lead to subsequent genome rearrangements. However several questions remain unanswered such as the exact nature and dynamic of proteins recruited at stalled forks as well as their mechanisms of recruitment. The main objective of the project is to explore the molecular mechanisms that allow eukaryotic cells to respond to replication forks impediments. Our main hypothesis is that proteins recruited or removed from the replisome in response to molecules that induce replicative stress should be key determinants of the response to stalled replication forks. Systematic analysis of protein recruited to stalled forks was a nearly impossible task few years ago, however with the appearance of methods to capture nascent chromatin it is now a feasible challenge. We set-up the iPOND (isolation of proteins on nascent DNA) method coupled with mass spectrometry (iPOND-MS). Thanks to this method we are able to analyze the complete landscape of proteins recruited at replication forks under basal conditions and in response to replicative stress induced by exogenous molecules. We propose to intensively use the iPOND-MS to analyze systematically the dynamic of proteins at replication forks in response to a panel of exogenous agents that induce different kind of replication impediments. The panel of molecules will contain replication inhibitors, DNA damaging agents, topoisomerases inhibitors, specific inhibitors and G-quadruplexes ligands. The use of molecules with really different mechanisms of action should allow the discovery of a large variety of proteins. This first step should lead to a list of candidates that will be validated using and orthogonal method based on high-throughput microscopy (Celigo). The roles of the proteins identified with respect to replicative stress will be addressed by methods such as DNA combing or iPOND coupled with innovative approaches like inducible degradation of proteins (degron) or CRISPR-Cas9. We are convinced that our approach will lead to the discovery of new key players of forks response to replicative stress. However, in case we would fail to discover new proteins, we will focus on the deep study of BAZ1B (a regulator of Topoisomerase 1 during replication) and on RIF1 (a protein involved in DNA repair, replication and telomeres protection), two proteins we recently identified using iPOND-MS. This project should lead to benefits in basic science by increasing the knowledge on mechanism of DNA replication but also in public health. Indeed proteins identified as key players of response to stalled replication forks could be subsequently validated as biomarkers of aging or exposure to DNA damaging agents found in the environment.

Alternative splicing affects up to 95% of multi-exonic genes in humans. There are three main types of alternative splicing: exon skipping, alternate 5` or 3` usage and intron retention (IR). While the first two are well described in normal development and disease, the role of IR in these processes remains to be definitively determined. IR occurs when an intron is included in a mature mRNA. Previously regarded as a byproduct of faulty splicing, transcripts with retained introns are often rapidly degraded by a surveillance mechanism called nonsense-mediated decay (NMD). We discovered that normal granulocytic blood cells make use of this mechanism by increasing the amount of transcripts with retained introns of dozens of genes that are essential to granulocytic differentiation (Cell, 2013). We then found a crucial role for IR in the the regulation of pluripotent stem cells (Nature, 2014) and in erythrocyte differentiation (Blood, 2016). Since, we and others have shown that regulation through IR coupled with NMD is evolutionarily conserved, widespread in normal and diseased tissues, and occurs in over 80% of coding genes. Because IR could not previously be correctly identified, numerous studies have overlooked potential biomarkers and therapeutic targets linked to this novel type of gene regulation. The importance of IR in disease is further underlined by the recent finding that tumour suppressor genes are modulated by IR in many different cancers (Nature Genetics, 2015). As such, a greater understanding of regulation through IR could positively impact ongoing and future research in numerous fields that contribute to human health. In this proposal we will further investigate the molecular mechanisms that lead to IR and develop bioinformatics tools to allow other teams to investigate this mode of regulation in their own data. The model we will use to investigate IR coupled with NMD is that of mouse granulocytic differentiation. These cells can be sorted with over 95% purity and we showed that their fate is orchestrated in part by dramatic changes in IR. They are thus an ideal model to investigate the molecular mechanisms that regulate IR. We will dissect 2 main mechanisms that cause IR : cis-regulatory motifs that activate or repress IR and epigenetic modifications that change polII elongation rates and thus affect intron recognition by the splicing machinery. To precisely detect motifs that enhance IR, we will first develop a novel approach to more accurately identify and characterize IR events. This approach will combine a third generation sequencing technology (PacBio) that generates ultra-long sequences from single molecules with high throughput Illumina sequencing. From this set of high confidence IR we will detect DNA motifs that are enriched in frequently retained introns and test them using CRISPR-Cas9 genome editing in the MPRO granulocytic cell line. We will then focus on epigenetic changes that may modulate IR. Nucleosome occupancy and DNA methylation through the intermediary of MeCP2 and CTCF proteins can modulate polII elongation and affect mRNA splicing. By assessing nucelosome occupancy, DNA methylation, the binding of MeCP2 and CTCF and the precise identification of IR, we will be able to discover associations between epigenetic marks and IR. We will then test this association by knocking out MeCP2 and CTCF then by altering the methylation status of regions associated with IR using a TALE-Tet1 and a dCas9-DNMT3A system. This project will develop methods to precisely detect IR events and will allow us to decipher the molecular mechanisms of a this novel mode of gene regulation that has thus far been linked with numerous disorders and normal differentiation processes.

MicroRNAs ("miRNAs") are important gene regulators: they recognize specific target mRNAs through sequence complementarity. Their expression level depends on their transcription and maturation (which are well known), as well as on their stability. Several observations suggest that miRNA stability is actively regulated, in particular in neurons, but the mechanism of such regulated degradation has long remained mysterious. A phenomenon of miRNA stability control by long RNAs which exhibit sequence complementarity to the miRNA (hence: similar to miRNA targets) has recently been revealed: TDMD ("target-directed microRNA degradation"). TDMD redefines the expression pattern of miRNAs, thereby controlling neuronal and behavioral phenotypes. While the biochemistry of TDMD is starting to be understood, many aspects of that process remain mysterious: for the moment, only two cellular RNAs, and a few viral RNAs, are known to induce TDMD; rules distinguishing TDMD inducers from classical miRNA targets are imprecisely described; and it seems that TDMD is neuron-specific, while its known molecular effectors are ubiquitous. Combining our expertise (high-throughput molecular biology, advanced computational analyses), we want to characterize that phenomenon more precisely. First, using the few known rules for RNA/RNA pairing that trigger TDMD, and complementing them with phylogenetic analyses, we established a list of candidate TDMD inducers in mouse cortical neurons, whose activity we want to verify experimentally. We will extend that work (predictions and experimental validation) to Drosophila, and we will perform a comprehensive prediction of TDMD inducers in mammals (that resource will be made available freely on the Internet). Second, we will clarify the recognition rules between miRNAs and TDMD inducers, with a high-throughput CRISPR screen followed by a machine-learning analysis. Our approach will use a randomized sequence library, allowing us to interrogate the TDMD inducing activity of dozens of millions of distinct sequences, for 8 different miRNAs, exhaustively exploring every possible pairing geometry and nucleotide identity. Finally, we will determine whether TDMD is more active in neurons, as has been apparent so far (or: whether this is a mere coincidence, on the few known examples). If neuronal tropism is confirmed, we will search for neuronal factors responsible for such a specificity, by a complementation experiment in non-neuronal cells. That work will therefore explore an emerging field, it will be articulated with the most recent international discoveries. It will also allow to better understand cross-dependency relationships in gene expression, making our list of predicted mammalian TDMD inducers accessible to all. Such a mechanistic understanding of inter-relations in the transcriptome is essential in order to make sense out of "omics" datasets which are currently descriptive and correlative. In particular, the development of personalized medicine can only become fertile if one can understand causal links among gene expression levels, and give some biological sense to "big data" which is currently piling up.

Project abstract Transcriptional control is essential for plants to adapt to ambient fluctuations. By implementing the latest MS2 RNA imaging system combined with microfluidics, we achieved real-time quantitative measurements of the transcriptional activity of single loci, in all cells of entire organs of living Arabidopsis. Here, we will bring these technologies to a new level, optimized for whole organism imaging. We will develop: (i) brighter and multicolor live cell RNA imaging technologies, using several strategies and in particular nanobody signal amplification; (ii) new software based on machine-learning to automatically and accurately analyze image data. These tools will be combined with genetics to unravel the complex transcriptional control triggered by an essential plant macronutrient: phosphate. We will address the fundamental question of how single cell responses, known to be noisy and heterogenous, are integrated into a coordinated response at the organism level. The progress brought by this project will provide tools and concepts that will renew our vision of transcriptional regulation from plants to metazoans.