Thèse Relier la Division et la Croissance Cellulaire Via la Ségrégation Mitotique du Matériel Nucléolaire H/F

Doctorat.Gouv.Fr

Établissement : Université de Montpellier
École doctorale : Sciences Chimiques et Biologiques pour la Santé
Laboratoire de recherche : IGMM - Institut de Génétique Moléculaire de Montpellier
Direction de la thèse : Liliana KRASINSKA ORCID 0000000268580852
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-05-11T23:59:59

Ki-67 est une protéine nucléaire exprimée dans toutes les cellules en prolifération et constitue le marqueur de prolifération cellulaire le plus utilisé en histopathologie du cancer. Nous avons montré que bien qu'elle ne soit pas indispensable à la prolifération cellulaire, Ki-67 est nécessaire à la formation correcte de la couche périchromosomique (PCL) des chromosomes mitotiques, composée en grande partie de matériel nucléolaire (1, 2). Ce mécanisme permet une distribution équitable du matériel nucléolaire entre les cellules filles. Dans les cellules dépourvues de Ki-67, la PCL ne se forme pas correctement ; nous émettons alors l'hypothèse que le matériel nucléolaire est hérité de manière stochastique, entraînant une variabilité accrue de la capacité nucléolaire et, par conséquent, une biogenèse et une croissance ribosomiques plus variables. Nous proposons également que la « décision prolifération-quiescence contrôlée par une bifurcation de l'activité de CDK2 » (3) dépend étroitement de la capacité de biogenèse ribosomique, et que, de ce fait, les cellules invalidées pour Ki-67 pourraient présenter une proportion accrue de cellules quiescentes ou des durées de cycle cellulaire plus variables. Le nucléole organisant également l'hétérochromatine, et ayant montré que cette dernière est réorganisée dans les cellules dépourvues de Ki-67 (1), il est possible que les cellules filles présentent une hétérochromatine plus variable. Par ailleurs, ayant également démontré que Ki-67 facilite le remodelage du transcriptome et est essentiel à la carcinogenèse (4), la question se pose de l'implication d'une répartition inégale du matériel nucléolaire dans ces phénotypes. Ce projet de thèse vise à répondre à ces questions. À partir de nos lignées cellulaires cancéreuses humaines et murines de type sauvage (WT) et Ki-67-/- générées précédemment, couplées à des marqueurs fluorescents du matériel nucléolaire et de l'activité CDK, nous déterminerons : i) si une transmission asymétrique du matériel nucléolaire résulte de la perte du PCL ; ii) quels domaines de Ki-67 sont nécessaires/suffisants pour la formation du PCL, et comment cela se produit ; iii) comment la transmission du matériel nucléolaire est corrélée à la dynamique de l'activité CDK et à la durée du cycle cellulaire. iv) si la restauration du PCL avec un sous-domaine Ki-67 minimal restaure l'apparence normale de l'hétérochromatine et la capacité tumorigène des cellules.

Ki-67 is the most widely used cell proliferation marker in cancer histopathology, but its physiological roles are not well understood, and whether it is involved in carcinogenesis has been unclear. We have shown that Ki-67 is dispensable for cell proliferation and mouse development, but is required for formation of the mitotic perichromosomal layer (PCL) and organises heterochromatin (1). The PCL, which comprises 30-50% of the volume of condensed mitotic chromosomes in cells, is largely composed of ribosomal RNA and nucleolar material and depends on Ki-67 for its formation (1, 6). This ensures equal distribution of nucleolar material, which assembles into nucleoli in G1 phase of the cell cycle by multi-phase condensation on nucleolar organising regions of chromosomes (7). The principal function of the nucleolus is ribosomal biogenesis, and protein translation rates integrate mitogen signaling history to control cell cycle progression (8). Therefore, we hypothesise that equal partitioning of nucleolar material into daughter cells gives rise to cells with similar proliferation rates. Conversely, disruption of this equal partitioning by loss of Ki-67 should increase cell to cell variability of growth and cell cycle progression. Given that recent data from mouse genetic cancer models suggests that cancer cells also arise from the most rapidly proliferating cell types in a given tissue (9), and our previous work implicating Ki-67 in multiple steps of carcinogenesis including transformation (5), it is possible that one role of Ki-67 in cancer is to promote homogeneity of cell proliferation.
The mechanistic basis by which Ki-67 promotes PCL formation is unknown, and the domain(s) of Ki-67 required for PCL formation have not been established. Ki-67 is a very large nuclear protein, containing an N-terminal FHA domain of about 100 amino acids while the remainder of the protein is predicted to be intrinsically disordered and contains the characteristic Ki-67 repeat domain. Our recent work has shown that mitotic Ki-67 phosphorylation triggers its phase separation in cells and that it is highly dynamic in the PCL, suggesting that the PCL is also a liquid-like phase (10). We also found that overexpression of full length Ki-67 promotes ectopic heterochromatin formation, while its deletion causes loss of the most compact heterochromatin (as measured by FRET-FLIM) and of typical heterochromatin histone marks around chromocenters and nucleoli (1). Whether and how nucleolar formation is coupled with heterochromatin establishment after cell division remains unknown.
In this project we will use our established constructions and Ki-67 knockout cell lines to investigate whether and how Ki-67 promotes homogeneity of cell cycle progression via nucleolar material inheritance and heterochromatin establishment, and the implications for tumourigenic capacity of cells.

In Aim 1, we will use single-cell live imaging assays to determine whether CDK activity dynamics are related to the level of nucleolar material and whether Ki-67 knockout results in more variable daughter cell inheritance of the latter.

In Aim 2, we will express different deletion mutants of Ki-67 to determine which subdomain(s) of Ki-67 are required and sufficient for PCL and heterochromatin formation, and define candidate mechanisms by determining their interacting proteins.

In Aim 3, we will explore whether the PCL or heterochromatin are important for Ki-67-dependent tumourigenic capacity.

Aim 1: We will perform live cell imaging on WT and Ki-67 KO cell lines expressing single cell reporter probes for CDK activity (11) and BFP-nucleolin to quantify nucleolar material inheritance in exponentially growing cells. Single cell traces will be computationally synchronised (aligned on the meta-anaphase transition). This will establish the distribution of cell cycle times and nucleolar material and its dependency on Ki-67 in a variety of cell lines.

Aim 2: In Ki-67 KO cell lines, we will complement the deletion with inducible constructs expressing either GFP-tagged full-length Ki-67 or a series or deletion mutants that we have already generated (lacking FHA domain, repeats, C-terminus, several motif mutants, or the C-terminus or FHA domain alone). The PCL will be analysed by immnuofluorescent (IF) microscopy of nucleolar components in mitotic cells, while heterochromatin will be analysed by IF and ChIP-seq of heterochromatic histone marks, as well as visualisation of major satellite DNA using a fluorescent TALE probe. This will establish which domains of Ki-67 are required for establishing the PCL and for heterochromatin formation / organisation. Proteins interacting with each of the domains will be purified by affinity chromatography (GFP-trap) and identified by mass spectrometry. Candidate interactors will be screened bi siRNA for their effects on the PCL and heterochromatin.

Aim 3: Our previously published work and preliminary results show that Ki-67 KO strongly affects tumourigenesis in syngeneic mouse cancer models of breast and colon cancer. We will evaluate whether efficient tumourigenesis is restored to cell lines stably complemented with different domains of Ki-67 and correlate the results with their effects on the PCL and heterochromatin. This will initially by done in 4T1 cells which will be orthotopically injected into mammary fat pads of nude or syngeneic (BALB/c) mice. Tumourigenesis will be followed by electronic caliper measurements, and lung metastasis quantified at the end-point by counting colonies derived from lung material that grow under 4-thioguanine selection. Results will be validated in the CT26 colon cancer line.