Structure-based discovery of mitochondrial protein-protein interaction modulators of a kinase/ATP5F1 complex.

The INSERM U1242 laboratory is a translational research structure bringing together biologists, chemists and clinicians to better understand oncogenesis in brain and gynecological cancers. By associating biologists working on cancer stress signaling pathways, chemists developing molecular tools and cancer clinicians, the laboratory aims at deciphering how tumor cells cope with stresses to resist to death and how they could be sensitized to treatments. The candidate will join an exciting and multidisciplinary scientific environment with on-site access to cutting-edge facilities in biochemistry, biophysics, molecular biology, cell biology and structural biology. The INSERM U1242 laboratory Is affiliated to the Université of Rennes, a top French university that offers a dynamic and lively international campus with an ideal location less than 2 hours from Paris and 1 hour from Saint Malo and the beautiful Britany coast and its beaches. 

Supervisor: Dr X. Guillory (PI); Remuneration: 580€ net/month; Duration: 6 months between January & July 2027

Scientific background – Mitochondria produce most of the cell’s ATP through oxidative phosphorylation (OXPHOS), centered on Complex V, the mitochondrial ATP synthase. Our collaborator, Dr G. Bertolin (IGDR, UMR CNRS 6290), has shown that a particular kinase localizes inside mitochondria, where it physically and functionally interacts with the ATP5F1A and ATP5F1B subunits of Complex V to sustain ATP production and cancer cell viability.^[1] Current inhibitors targeting this kinase act on the whole cellular pool of the protein and lack organelle selectivity, which drives toxicity and resistance. There is therefore a strong need for molecules that selectively disrupt the mitochondrial protein-protein interaction (PPI) between this kinase and the ATP5F1 complex. Our group has a strong expertise in drug discovery, from in silico drug screening to in vivo evaluation, and in particular in the development of kinase inhibitors and small-molecule PPI modulators.^[2,3,4] Building on a first high-confidence structural model of the complex obtained with AlphaFold3 and further evaluated through molecular dynamic simulations, this internship will consolidate the model and exploit it to screen for first-in-class modulators of this interface.

Description of the project & duties - This internship is part of MITOCORE, a collaborative program funded by the Cancéropôle Grand-Ouest, aiming to validate and pharmacologically target the kinase/ATP synthase interaction in breast cancer. You will drive the computational component of the project: refining the structural model of the complex, characterizing its druggable interface, and running a structure-based virtual screening campaign to prioritize candidate modulators for experimental validation by our partner team.

As part of this internship, you will focus on:

1 – Refining the structural model of the kinase/ATP5F1 interface.

2 – Running and analyzing molecular dynamics simulations to assess the stability of the complex and to map druggable hot-spots and key interface residues.

3 – Performing structure-based virtual screening of commercially available, PPI-focused chemical libraries (e.g. ChemSpace, Fr-PPIChem^[5]).

4 – Prioritizing candidate modulators according to predicted binding affinity, binding mode and physicochemical/ADMET properties.

5 – Preparing a shortlist of compounds and interfacing with the experimental team for split-FAST and biochemical validation.

Main activities and techniques involved –

• Structure prediction and modelling of protein complexes (AlphaFold3, protein-protein docking).

• Molecular dynamics simulations and trajectory analysis (GROMACS, MDAnalysis, etc).

• Analysis of protein-protein interfaces and identification of druggable hot-spots.

• Structure-based virtual screening: compound library preparation, ADMET/physicochemical filtering, docking and candidate prioritisation.

• Scientific computing on HPC resources (Eskemm Data mesocentre); Linux environment and scripting (e.g. bash; python).

Applicant profile – We encourage applications from highly motivated Master/M2 students from the national and international community, with a background in molecular modelling, chemoinformatics, drug discovery, structural biology or related life sciences, and with outstanding qualifications. Familiarity with molecular modelling software, a Linux environment and scripting is appreciated but not required.