The aggregation state of NP has been a significant source of difficulty for assessing their toxic activity and great efforts have been done to reduce aggregation of and/or to disperse NP in experimental systems. The exact impact of aggregation on toxicity has, however, not been adequately assessed. Here we compared in vitro the cytotoxic activity of stable monodisperse and aggregated silicon-based nanoparticles (SNP) without introducing a dispersing agent that may affect NP properties. SNP aggregates (180 nm) were produced by controlled electrostatic aggregation through addition of KCl to a Ludox SM sol (25 nm) followed by stabilization and extensive dialysis. The size of the preparations was characterized by TEM and DLS; specific surface area and porosity were derived from N(2) sorption measurements. Macrophage (J774) and fibroblast (3T3) cell lines were exposed to monodisperse or aggregate-enriched suspensions of SNP in DMEM in absence of serum. The cytotoxic activity of the different preparations was assessed by the WST1 assay after 24h of exposure. Parameters that determined the cytotoxic activity were traced by comparing the doses of the different preparations that induced half a maximal reduction in WST1 activity (ED(50)) in both cell lines. We found that ED(50) (6-9 μg/ml and 15-22 μg/ml, in J774 and 3T3, respectively) were hardly affected upon aggregation, which was consistent with the fact that the specific surface area of the SNP, a significant determinant of their cytotoxic activity, was unaffected upon aggregation (283-331 m(2)/g). Thus studying small aggregated NP could be as relevant as studying disperse primary NP, when aggregates keep the characteristics of NP, i.e. a high specific surface area and a nanosize dimension. This conclusion does, however, not necessarily hold true for other toxicity endpoints for which the determinants may be different and possibly modified by the aggregation process.
Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.