Search for content and authors
 

Nano-QSAR modelling - efficient way of predicting the toxicity of metal oxide nanoparticles to human keratinocyte cell line

Agnieszka K. Gajewicz 1Jerzy Leszczynski 2Tomasz Puzyn 1

1. University of Gdańsk, Faculty of Chemistry, Laboratory of Environmental Chemometrics (UG), Wita Stwosza 63, Gdańsk 80-952, Poland
2. Interdisciplinary Center for Nanotoxicity Jackson State University, 1400 Lynch St., Jackson, MS 39217, United States

Abstract

Together with an increasing role of nanotechnology in our every-day-life, we can also expect increasing emissions and, consequently, increasing levels of nanoparticles (NPs) in the environmental compartments. However, there is an increasing number of contributions that report toxicity and/or ecotoxicity of selected NPs and highlight the potential risk related to the development of nanoengineering. Therefore, novel, fast and inexpensive procedures for identifying potentially hazardous NPs without necessity of extensive empirical and animal testing are needed. Toxicity of NPs can be predicted in alternative way by applying Quantitative Structure-Activity Relationships methods (QSAR). QSAR methods are based on the assumption that the variance in a given physico-chemical or biological (e.g. toxicity) property in a set of compounds (so-called endpoint) is determined by the variance in their molecular structures, encoded by so-called descriptors. Consequently, when the values of the endpoint are available only for a part of the group, it is possible to interpolate lacking data from an appropriate mathematical model.

The main purpose of the study was to illustrate the thesis above by developing a Nano-QSAR model that describes the relationship between the structure and toxicity of 22 nano-metal oxides to human keratinocyte (HaCaT) cell line. We have investigated the changes in cell viability and the process of generating reactive oxygen species by employing a human keratinocyte cell line as a model for dermal exposure. The experimental results expressed in term of LD50 values indicated that ZnO was the most toxic one in the set of the NPs tested. The LD50 value determined for ZnO was 27 µg/ml, while the values of LD50 for the majority of the other NPs (In2O3, La2O3, SnO2) were higher than 250 µg/ml. Additionally, in order to find the best structural parameters, reflecting the essential properties of the studied nanomaterials (shape, porosity, surface area, the electronic states resulting from quantum effects, etc.) we proposed a set of (i) image descriptors (based on images taken from Transmission Electron Microscopy) and (ii) quantum-mechanical descriptors (based on quantum-chemical calculations). The combined experimental-theoretical study allowed us to develop a Nano-QSAR model that reliably predicts the toxicity of all considered compounds. Such a model could be applied not only to NPs investigated in the current work, but also to unexplored related species.

 

Legal notice
  • Legal notice:
 

Related papers

Presentation: Poster at Nano and Advanced Materials Workshop and Fair, by Agnieszka K. Gajewicz
See On-line Journal of Nano and Advanced Materials Workshop and Fair

Submitted: 2013-08-28 13:39
Revised:   2013-08-28 13:39