Résumé : Major Facilitator Superfamily (MFS) transporters are ubiquitous and play an essentialrole in multiple physiological processes. Some multidrug transporters (MDR) form asub-family of these MFS transporters and are capable to recognize and extrude out ofthe cell an unusually wide range of chemically dissimilar noxious molecules thereforeconferring multidrug resistance to their host. Although they share low sequence similarity,crystal structures reveal a conserved three-dimensional fold with the rest of the family.These observations led to the proposal of a common resistance mechanism for MDR-MFStransporter. Despite intense investigation, the transport mechanism has not been yetdeciphered as only two MFS multidrug transporter have been structurally elucidated.Therefore, in order to address this lack, we decided to solve the crystal structure ofLactococcus lactis multidrug resistance Protein. LmrP is a multidrug transporter from L.lactis belongings to the Major Facilitator Superfamily (MFS). The proteins is poweredby the proton motive force and is capable to recognize and transport a wide range oflipophilic compounds including well known classes of antibiotic.During this work, we have intensively explored the crystallization space of the proteinfrom amino acid sequence through the degree of lipidation to the composition of thecrystallization drop. By honing our lipidation protocol, we were able to produce a largeamount of crystals in various conditions that diffracted X-rays up to 5 Å. We removed fourresidues in the loop connecting the two halves of the protein. Engineered LmrP crystalsdiffracted X-rays beyond 3.0 Å and they had a completely different morphology that hadnot been observed for wild type LmrP crystals. This work reports the crystal structureof LmrP at 2.9 Å in complex with a Hoechst 33342 molecule, a prototypical substrate ofmultidrug transporter. The protein was crystallized in the inward-close/outward-openconformation. The crystal structure reveals another putative conformational switch,consisting of D340 and R135, that has not yet been identifed so far by other techniques.Intriguingly, the structure also reveals an unidentifed density in the binding site thatexpands in the two lobes of the protein and that lines up with the substrate. Our nativemass spectrometry analysis indicates that the density could be due to a lipidic molecule.To our knowledge, this is the frst time that a lipid is reported in the binding cavity ofa multidrug transporter. We, therefore, propose that the lipid plays a modulation rolein the recognition of the substrate. If true, this could shed light on a completely newmechanism of secondary multidrug transport.