It is well understood that when nanoparticles (NPs) enter a biological medium, their surface is coated by various proteins; thus, the interaction of the living systems with the NPs depends on the composition of the protein layer, rather than the surface characteristics of the nanoparticle itself. However, there are several neglected parameters in protein-NP interactions (e.g., the key role of the protein source) that should be addressed. The composition of the protein corona is recognized as having a crucial influence on the delivery of NPs into cells, which is important in therapeutic applications and in nanotoxicology; however, the effect of "cell observer" (cell type) is poorly understood. This study probed the effects of different protein sources (fetal bovine serum [FBS] and human plasma [HP]) on the composition and protein thickness of the hard corona formed at the surface of superparamagnetic nanoparticles (SPIONs) with various sizes and surface chemistries. The results show that the hard corona can change quite considerably as one passes from the biophysicochemical properties of nanoparticles and protein sources (e.g., FBS and HP) appropriate to in vitro cell/tissue studies to those appropriate for in vivo studies. These changes in the hard corona have deep implications for in vitro-in vivo extrapolations. In addition, we probed the "cell observer" effect on the uptake and toxicity of SPIONs with the same protein corona composition to highlight the effect of cell type in nanobiosciences. The particles interacted with various cell lines. We find that without consideration of the "cell observer" effect, the cellular targeting/toxicity of NPs is inherently imprecise; thus, a deep understanding of both the protein corona composition and the "cell observer" effect offer a way to predict NP dosage for therapy and for the study of nanotoxin. (c) 2012 Elsevier Inc. All rights reserved.