Characterization of the protein corona of gold nanoparticles by an advanced treatment of CE-ICP-MS data

CE is well known not only as an efficient separation method, but also as a viable tool for studying chemical reactions, including kinetic assaying and analysis of chemical equilibria. In this communication, the latter feature of CE interfaced with ICP-MS was exploited to determine the stoichiometric composition of the protein corona of gold nanoparticles (AuNPs) at equilibrium conditions. For both individual albumin and human serum involved in binding, the number of protein molecules bound per AuNP (n) was calculated. Since the time scale of the corona formation was previously found to be dependent on the particle size, two calculation algorithms were adopted here. In the case of 5-nm AuNPs, rather slowly associating with the protein, the peak areas measured for the conjugated and free particles were taken in computation (the S-34 signal due to bound protein was also monitored simultaneously to confirm that equilibrium is reached). In binding labile systems (10-50 nm AuNPs), the particles are converted into the protein-bound form relatively fast due mostly to the favor of a much greater excess of the protein so that no peak of the free particles interacting with serum being recorded. Therefore, the n value was estimated by relating the sulfur peak area of each of these conjugates to that of 5-nm AuNPs to calculate the number of bound albumin molecules that was then divided by the number of AuNPs. The AuNPs were found to react with from 13 to 292 albumin molecules that is in good agreement with the literature data.