Silica nanomaterials have been proposed as promising candidates for the delivery of drugs, including purified proteins, thus a systematic investigation on the role of silica surface porosity, for optimal protein loading and release, represents an urgent need. Herein, we report the studies of the loading and releasing processes of different volume-proteins, namely Myoglobin, Bovine Serum Albumin and Factor VIII (FVIII), performed using three porous silica samples. To this aim, silica nanoparticles are prepared through modified sol-gel methods, in which suitable templating and swelling agents are used to enlarge the surface pores from few nm to 25 nm, but retaining the average dimensions of the nanoparticles (80 –90 nm). TEM imaging and Zeta-Potential measurements are used to characterize the silica samples. In this study, the loading process is carried out at pH values close to the isoelectric point of each protein, and the amount of adsorbed protein is quantified and values in the range of μgprotein per mgsilica are obtained. Then, the release profiles are monitored at physiological conditions; for all systems, the release of the proteins is detected, although different behavior are observed. The release profiles are analyzed through the power law equation and through a modular exponential function to achieve information on the mechanism and on kinetics of the release profile. The data show the release from silica particles with small pores follows a diffusion-controlled mechanism, while the kinetic analysis enables to distinguish two-components associated to superficially adsorbed protein and to the portion of protein allocated inside the pores. On the other hand, the particles characterized by the largest pore size (ca. 25 nm) display non-Fickian release profile and a mono-exponential dependency, indicating a mono-modal distribution of the proteins on the silica matrix. Moreover, the conformation or the biological activity of the released proteins has been evaluated through in-vitro methods and the data demonstrate the excellent quality of the recovered proteins. The loading and release data were rationalized in terms of mesoporous structures of the silica nanoparticles in relation to the protein dimensions.