Chromatin and cell biology

  • The genome is more than a linear string of genes. It has secondary, tertiary and quaternary higher-order structures. Strings of genes organize in chromosomal domains that are characterized by one of relatively few different types of chromatin, either repressive or activating. In the 3D space, each of the elements in one domain makes frequent contacts with other elements in the same domain. These entities have been called physical- or topologically associated- domains (TADs). At a chromosome level, individual TADs form contacts with other TADs, preferentially of the same type, in order to build ordered 3D architectures that are called chromosome territories. Finally, different chromosomes organize non randomly in the nuclear space. Therefore, the eukaryotic genome is highly organized in 3D and this regulation can be transmitted or modulated during the life of cells and organisms. This sophisticated chromosomal choreography involves thousands of different players – DNA sequences, RNAs and proteins – but rather than combining in infinite numbers of ways, these components organize a relatively limited number of types of chromatin, either active or repressive. In particular, two main groups of genome regulatory components are proteins of the Polycomb Group (PcG) and of the trithorax Group (trxG). PcG proteins are maintain the memory of silent states of gene expression through cell physiology and multiple cell divisions, while trxG members maintain active chromatin states. These proteins are able to recognize regulatory states of their target genes and to maintain these states even after disappearance of the primary transcriptional regulators that have induced them in the first place. Remarkably, these states can also be transmitted to a fraction of the progeny over multiple generations. In our lab, we aim at understanding the principle governing 3D genome organization, its functional implications, and the molecular mechanisms by which PcG and trxG proteins regulate their target genes, convey inheritance of chromatin states and orchestrate development. To reach this goal, we employ a variety of complementary approaches and techniques in the areas of molecular, cellular and developmental biology, genomics and bioinformatics.


    To date, our research is funded by:

    logo anr
    logo cnrs 80
    logo inserm
    logo frm
    logo epigenmed
    logo europe com
    logo inc
    logo arc

      Learn more


    Frédéric Bantignies
    Bantignies Frédéric
    Giacomo Cavalli
    Cavalli Giacomo
    Thierry Cheutin
    Cheutin Thierry
    Anne-Marie Martinez
    Martinez Anne-Marie
    Bernd Schuttengruber
    Schuttengruber Bernd
    Lauriane Fritsch
    Fritsch Lauriane
    Axelle Donjon
    Donjon Axelle
    Marco Di-Stefano
    Di-Stefano Marco
    Max Fitz-James
    Fitz-James Max
    Ivana Jerkovic
    Jerkovic Ivana
    Flora Paldi
    Paldi Flora
    Giorgio-Lucio Papadopoulos
    Papadopoulos Giorgio-Lucio
    Gonzalo Sabaris
    Sabaris Gonzalo
    Nazli Akilli
    Akilli Nazli
    Sandrine Denaud
    Denaud Sandrine
    Victoria Parreno
    Parreno Victoria
    Hadrien Reboul
    Reboul Hadrien
    Michael Szalay
    Szalay Michael
    Ana-Maria Popmihaylova
    Popmihaylova Ana-Maria


    Regulation of single-cell genome organization into TADs and chromatin nanodomains

    Szabo, Q., Donjon, A., Jerkovic, I., Papadopoulos, G.L., Cheutin, T., Bonev, B., Nora, E., Bruneau, B.G., Bantignies, F., and Cavalli, G.

    2020 - Nature Genetics 52, 1151–1157. (2020)

    Request for full articleMore informations

    4D Genome Rewiring during Oncogene-Induced and Replicative Senescence

    Sati, S., Bonev, B., Szabo, Q., Jost, D., Bensadoun, P., Serra, F., Loubiere...

    2020 - Mol Cell, 10.1016/j.molcel.2020.03.007

    Request for full articleMore informations

    Widespread activation of developmental gene expression characterized by PRC1-dependent chromatin looping.

    Loubiere V, Papadopoulos GL, Szabo Q, Martinez AM, Cavalli G

    2020 - Sci Adv, 6(2):eaax4001

    Request for full article31950077

    Advances in epigenetics link genetics to the environment and disease

    Giacomo Cavalli and Edith Heard

    2019 - Nature, 10.1038/s41586-019-1411-0

    Request for full article
    Display all the publications

    Publications of the team

  • 3D organization and function of the genome

    The information stored in our genome is intertwined with its function, such that, when cells are submitted to specific sets of conditions, they may pass on to their progeny their functional state. Since DNA has been identified as a critical carrier of genetic information and since the same DNA can correspond to alternative, heritable functional states in certain cases, this transmission of cellular memory has been dubbed epigenetic inheritance. In the most spectacular way, this extends to inheritance of a phenotypic trait into subsequent generation, a phenomenon for which Conrad H. Waddington provided evidence some sixty years ago and which is well documented in plants. However, to which extent epigenetic inheritance operates in animals is hotly debated. Chromatin and its higher-order organization are epigenetic components that play an essential role in genome regulation. Both the DNA molecule and the nucleosomal histones can be extensively modified in a way that impinges on gene expression and may be inherited as well as erased upon specific regulatory cues. Furthermore, chromatin fibers can be folded into yet higher-order structures and chromosomes are confined in discrete “territories”.

    We and others have discovered that metazoan chromosomes share a modular organization of their chromatin in structures called “Physical domains” or “Topologically Associating Domains” (TADs). TADs can be defined as linear units of chromatin that fold as discrete three-dimensional (3D) structures tending to favor internal, rather than external, chromatin interactions. TADs are delimited by boundaries, which contain housekeeping genes and insulator sites. They are detected by methods such as Hi-C, which allows genome-wide identification of chromatin contacts, and they correspond to Chromosomal Domains (CDs), previously identified by microscopy. The investigation of chromatin landscapes in metazoa through genome-wide association studies proved to be a fruitful approach. Theoretically, a huge number of chromatin types based on different combinations of chromatin-associated marks would be possible but, in fact, every report basically recapitulated the presence of an active chromatin environment, sometimes further subdivided, and of three major types of repressive chromatin: a Polycomb-repressed environment, a null environment and a heterochromatic environment. Strikingly, TADs were found to overlap with linear chromatin domains, indicating that epigenomic labeling of chromosome domains is intimately linked to their 3D folding.

    We are trying to understand the principles governing 3D folding of the genome, from establishment of chromatin loops to the generation of chromosome domains, compartments, territories and the establishment of interchromosomal interactions. We use Drosophila, but also mouse and human cells, and state of the art molecular, genomic, computational and imaging approaches, in order to reach an integrated understanding of these different levels of genome organization

    figure 1
    Hierarchical organization of the genome in chromatin fibers, loops, domains and chromosome territories involving multiple regulatory 3D chromatin contacts

    Mechanisms of Polycomb-mediated genome regulation

    Polycomb group (PcG) and trithorax group (trxG) proteins are key regulators of the expression of major developmental genes. PcG proteins are able to silence gene expression, while trxG proteins counteract gene silencing in the appropriate cells. The current model proposes that a sequence-specific DNA binding protein called PHO binds at so-called Polycomb response elements (PREs). PHO might recruit the PcG complex called PRC2, which contains the core subunits E(z), a histone methyltransferase that trimethylates histone H3 lysine 27 (H3K27me3), Su(z)12, Esc and Nurf55. H3K27me3 might then be recognized by the chromo domain of the PC subunits of PRC1, which also contains Ph, PSC and Sce/dRing. Once recruited, PcG complexes can propagate silencing through cell division. Genome-wide mapping studies have shown that PcG target genes encode for components controlling major signalling pathways and, importantly, PcG misexpression has also been associated with many cancer types, including breast and prostate cancer.

    In addition to their role in cellular memory, PcG proteins participate in dynamic gene regulatory processes. In flies, different cell lines have a partially different set of PcG bound sites and H3K27me3-marked genomic regions change during development. In mammalian embryonic stem cells, many PcG target genes have been reported to bear both repression- and activation-associated marks. Upon differentiation, these “bivalent states” are resolved into fully active or fully repressed. In some instances, PcG components may even activate transcription, although it is unclear whether this phenomenon is widespread or rare. Importantly, PcG proteins regulate the organization of their target genes in the three-dimensional space of the nucleus, and this regulatory function is involved in the maintenance of cellular memory.

    We would like to understand the molecular mechanisms of action of these factors, the role of regulation of higher order chromatin structure and nuclear organization in gene regulation, and the key molecular pathways that are mobilized by these proteins to coordinate the regulation of cell differentiation with that of cell proliferation. In particular, our research aims at (1) understanding, on a genome-wide scale, how these proteins are targeted to DNA and what are the consequences of this targeting on chromatin structure (2) understanding the effect of PcG proteins on cell proliferation, cell differentiation and cell polarity, and to dissecting the key components regulated by PcG proteins to modulate these pathways in specific tissues and developmental processes; (3) identifying the rules governing the distribution of their target genes in the cell nucleus and the effect of this organization on gene expression.

    figure 2
    The many roles of PcG proteins. The scheme illustrates the biological processes implicating PcG component on the outside and the genes targeted by these proteins to regulate these processes (on the inside of the light blue oval).

    The role of Polycomb proteins in development and tumorigenesis

    Coordination between cellular proliferation and differentiation ensures proper tissue morphogenesis and maintains homeostasis in multicellular organisms. Appropriate numbers of undifferentiated cells must be generated at specific developmental stages and these cells must exit the cell cycle in a tightly regulated manner to ensure proper cell fate specification and pattern formation.

    The prevailing paradigm posits that Polycomb Group (PcG) proteins maintain stem cell identity by repressing differentiation genes. Mutation or misexpression of PcG genes has been associated with several types of human cancer. Polycomb group (PcG) proteins form two main epigenetic transcription repressor complexes, PRC2 and PRC1, highly conserved from fly to humans, which generally coregulate their target genes. They colocalize almost perfectly during embryogenesis, and their embryonic phenotypes are similar, with posterior homeotic transformations due to misexpression of homeotic Hox genes. Later in development, alterations in PcG components induce cancer (Figure 3), suggesting that PcG proteins may be dynamically recruited to new target genes. PcG proteins additionally bind and regulate genes implicated in major signaling pathways and therefore also participate in dynamic gene regulatory processes.

    Using as a model the Drosophila larval imaginal eye disc, which shares critical features with mammalian epithelial tissues, we demonstrated that mutations affecting PRC1 subunits, but not PRC2, trigger neoplastic tumours in the larval imaginal discs. PRC1 components act as neoplastic tumor suppressors independently of PRC2 function.

    by specifically targeting a thousand of new genes during larval stages of fly development. We named these non-canonical genes, “Neo-PRC1”; they massively outnumber canonical targets, are devoid of the H3K27me3 epigenetic mark and carry instead the active mark H3K27Ac. Remarkably, neo-PRC1 genes are mainly involved in the regulation of cell proliferation, differentiation, signaling and polarity. Alterations in PRC1 components specifically deregulate this set of genes, whereas canonical targets are derepressed in both PRC1 and PRC2 mutants. Together, these results suggest that the mechanism of recruitment of PRC1 on its neo-sites is independent of PRC2 and depends on new molecular mechanisms that remain to be determined. The search for these mechanisms and of their molecular significance in flies and mammals is the basis of our current interest.

    Figure 3
    Figure3: Tumor suppressor function for PcG genes. Left: the “+/+” panel shows control clones that do not carry mutations, which are positively labelled by green fluorescent protein. The ph0/ph0 panel shows clones corresponding to a knock out of the ph locus in order to eliminate the PH protein of PRC1. This induces massive overgrowth compared to the +/+ control disc containing neutral clones. The “Control” panel shows a, transplantation from wt GFP tissue into a wt host fly. The black stain in the abdomen shows the scar of the transplanted material. The “ph-/-“ panel shows an F1 generation host in which mutant tissue has been transplanted. The GFP labelled, ph mutant tumor grows in the body and kills the transplanted hosts.

    Understanding transgenerational epigenetic inheritance

    Epigenetic inheritance entails transmission of phenotypic traits not encoded in the DNA sequence and, in the most extreme case, transgenerational Epigenetic Inheritance (TEI) involves transmission of memory through multiple generations. Very little is known on the principles and the molecular mechanisms governing TEI. Chromatin has been linked to TEI but little is known on how chromatin modifications might be transmitted across generations. Polycomb components are involved at each level of chromatin folding, from post-translational histone modification all the way up to regulation of global chromosome architecture.

    In our previous work, we established that perturbation of 3D chromatin architecture affected Polycomb function in a heritable way. In Drosophila lines carrying a transgene with a PRE from the Fab-7 region of the Hox cluster called bithorax complex (BX-C), we showed that the transgene establishes 3D contacts with the endogenous BX-C locus. The transgene is flanked by a reporter gene called mini-white, which is responsible for eye pigmentation. In the Fab2L line, the PRE in the transgene partially represses the reporter gene. Therefore flies have low levels of eye pigmentation that are variable between individuals, a phenomenon that reflects metastable silencing in somatic cells. These somatic epigenetic differences are not transgenerationally heritable, as self-crossing of flies with the most repressed or the most derepressed eye phenotypes does not cause any phenotypic shift in the progenies. However, we found that a transient perturbation of nuclear organization can induce TEI. We transiently removed one copy of the endogenous Fab-7 element by crossing Fab2L flies with a derivative carrying the transgene in the absence of the endogenous Fab-7 copy. This induces an increase in the contact frequency between the transgenic locus and the single remaining copy of the endogenous Fab-7 in the F1. We then restored the missing endogenous Fab-7 and obtained an F2, which is genotypically and phenotypically the same as the P0 parental generation. F2 flies with the most repressed and the most derepressed eye color were then segregated in two distinct groups, whose progenies were subjected to selection based on the eye color. Strikingly, starting at the F3, we observed the appearance of more pigmented flies in the selection of the active state and of more reduced pigment levels in the selection of the repressed state. This trend further amplified across generations, finally reaching the establishment of two “epilines”, which could stably maintain their different phenotypic trait.

    We showed that this form of inheritance applies to multiple transgenic lines, to endogenous genes, that it involves components of the PRC2 complex and that it can be modulated by environmental conditions. Our future research aims at understanding the mechanisms of TEI and to uncover its role in natural processes.

    Figure 4
    Establishment of stable Drosophila epilines that can transmit Polycomb-dependent chromatin states through multiple generations