By use of the quartz crystal microbalance technique, the interaction of the Raf-Ras binding domain (RafRBD) and the cysteine-rich domain Raf-C1 with lipids was quantified by using solid-supported bilayers immobilized on gold electrodes deposited on 5 MHz quartz plates. Solid-supported lipid biloyers were composed of an initial octanethiol monolayer chemisorbed on gold and a physisorbed phospholipid monolayer varying in its lipid composition as the outermost layer. The integrity of bilayer preparation was monitored by impedance spectroscopy. For binding experiments, a protein construct comprising the RafRBD and Raf-C1 linked to the maltose binding protein and a His tag, termed MBP-Raf-C1, was used. Dissociation constants and rate constants of the association and dissociation were obtained for various 1,2-dimyristoyl-sn- glycero-3-phosphocholine (DMPC)/1,2-dimyristoyl-sn-glycero-3- phosphoserine (DMPS) lipid mixtures. Independently of the phosphatidylserine (PS) content, the dissociation constants were in the order of 5 x 10(-7) M, while the on-rate constants were in the range of 2 x 10(3) (MS)(-1) and the off-rate constants in the range of 1 x 10(-3) s(-1). The maximum frequency shirt increased significantly with increasing amounts of DMPS; this indicates that this negatively charged lipid is the primary binding site for MBP-Raf-C1. Exchange of DMPS for 1, 2-dimyristoyl-sn-glycero-3-phosphoglycerol (DMPG) did not after the thermodynamics and kinetics of protein binding, which implies that the protein interaction is mainly electrostatically driven. Scanning force microscopy (SFM) was employed to render protein adsorption visible and to confirm the assumption of a protein monolayer on the lipid layer, SFM images clearly revealed that the protein binds preferentially, but not solely, to negatively charged phosphotidylserine heodgroups. We hypothesize that PS-enriched domains are initial binding sites with high affinity for Raf-C1, but that lateral interactions may account for protein domain growth.