First, we performed a lipid dot-blot analysis in which recombinant Drp1 was incubated with a nitrocellulose membrane containing some of the most common glycerolipids and sphingolipids present in mammalian cells. Among the lipid species examined, Drp1 bound most strongly to CL, less strongly to the anionic lipids phosphatidylserine and phosphatidic acid and interestingly not detectably to the polyanionic lipid phosphatidylinositol -trisphosphate even though this lipid displays higher net negative charge than CL. Dose-response experiments indicated that recombinant Drp1 bound to CL with,5-fold,10-fold and,18-fold higher potency than to PS, PG and PI, respectively. Importantly, additional lipid dot-blot assays revealed that the endogenous Drp1 from MEF cells also recognized a variety of anionic phospholipid species and also displayed a similar preference for CL. Considering that phospholipids spread on a nitrocellulose membrane do not form a lipid bilayer, an extrapolation from these data would suggest that the Drp1:CL interaction does not rely on CL-mediated changes in the physical properties of the bilayer, such as induction of lateral segregation of membrane lipids into domains or generation of negative membrane monolayer curvature stress. To further address this point, we examined the interaction of Drp1 with PC/PE and PC/PE/ CL vesicles of different sizes ranging from 50 nm to 400 nm. We observed almost no binding of Drp1 to PC/PE vesicles, while a similar binding was detected for CL containing vesicles, independently of their size. These results support the notion that Drp1 interaction with CL-containing liposomes is not affected by the membrane intrinsic net curvature. To study the lipid-binding properties of Drp1 in real-time, we performed surface plasmon resonance studies. To this end, LUVs with different lipid compositions were immobilized on the surface of the L1 sensor chip and used as ligands to probe the kinetics of Drp1 binding. Representative SPR sensorgrams are shown in Figure 1D. Maximal binding was observed with LUVs containing CL, while Drp1 did not bind to LUVs composed of PC/PE possessing zero net negative charge. Interestingly, replacing CL with other anionic lipids while maintaining the same net negative charge on the vesicle, did not reproduce the strong binding response observed with CL containing vesicles, supporting the notion that the Drp1:CL interaction does not rely exclusively on electrostatic forces. Next, we analyzed Drp1 binding to vesicles containing increasing amounts of CL and PS. At all concentrations tested, Drp1 binding KU-0059436 responses were lower for PS-containing liposomes relative to CL-containing liposomes. Unfortunately kinetic rate constants could not be obtained by fitting sensorgrams to a first-order binding model. Similar observations were previously reported for other protein:membrane interactions analyzed by SPR. To investigate further the specificity of Drp1 interaction with CL, we examined the effect of the CLinteracting drug doxorubicin.