Protein-mimetic materials are of great interest for biotechnology to grant protein-like properties to artificial systems. Additionally, these materials can be used to shed light on the fundamental properties of proteins in many environments. Nanoparticles, in particular, offer a wide choice of parameters like size, core material, and surface chemistry that can be modified to better mimic proteins. Recently, it was shown that amphiphilic gold nanoparticles are able to fuse with and penetrate cell membranes in an energy independent manner. This behavior is similar to that of cell penetrating peptides. The objective of this thesis was to investigate the key parameters that allow amphiphilic nanoparticles to fuse with lipid bilayers and to generate nanoparticles capable of performing other protein-like functions. The thesis start with a systematic study on the interaction between amphiphilic gold nanoparticles and liposomes. All particles studied had a gold core, whose size varied from 1.7 to 6 nm. Amphiphilic ligand shells were achieved by using mixtures of a hydrophobic and hydrophilic ligand, with the hydrophobic one limited at 30% of the mixture because of solubility limitations. The liposomes studied had a size range of 60 to 300 nm. We found many different interactions. Sulfonated particles interacted with small vesicles (smaller than 60 nm) mediating clustering of vesicles, similar to adhesive proteins. The same particles, once combined with larger liposomes, behaved like curvature inducing proteins, that is they generated flat areas terminating in high curvature joints. Particles at the highest degree of amphiphilicity (i.e. a loading of the hydrophobic molecules at the limit of solubility), however, showed an even larger variety in interactions. The largest particles (~5 nm) fused with lipid membranes (independent on the liposome size) like cell penetrating peptides. Particles of intermediate size (~3 nm) fused with the membranes and brought liposomes in close proximity (like the less amphiphilic ones). These particles had an additional important property, they were able to mediate a fusion event upon Ca2+ addition between the adjacent liposomes. Hence these particles behaved as SNARE proteins and can be considered as the first synthetic material to be fusogenic. The smallest particles in this category fused with the lipid bilayer (as all others in this family), were fusogenic (as the middle sized ones) and spontaneously generated ring patterns on the liposomes. These rings where found to delimit flat vesicle regions. This latter phenomenon does not seem to have a biological analog. Overall, we find that amphiphilic nanoparticle can be considered as protein analogs, especially proteins that interact with membranes. In particular when the degree of amphiphilicity is high we achieved a series of particles that behave as membrane proteins, and depending on their size show different functions. These findings offer insight to decode the key factors in the interaction between nanoparticles and lipid bilayers and provide the first hints on how to use nanoparticles model systems to gain a better understanding on some key features of protein and protein functions.