Ecotoxicology of Manufactured Nanoparticles - Environmental and Human Health Impacts of Nanotechnology - page 280

7.4 Development of Valid/Realistic Toxicity Testing Protocols

The screening of new chemicals for ecotoxicity by using a battery of standardised
tests is a cornerstone of our current approach to hazard assessment. Organisations
such as the US EPA and the OECD play a central role in developing standardised
testing protocols that are used internationally. Research studies which deviate from
the standard toxicity testing approaches are a vital complimentary facet of ecotoxi-
cology, as these provide information on the effects of environmental variables on
toxicity and also the effects of toxicants on other organisms that are not stan-
dardised test species. Typically, important environmental variables that infl uence
toxicity include pH, dissolved organic mater concentration, ionic strength and the
concentration of inorganic ions such as calcium and magnesium. Such information
is not accessed from routine toxicity testing and provides vital information for the
development of predictive models of toxicity.
An immediate problem is how relevant standardised testing protocols are
for nanoparticle testing. Key concerns are for the chemical and physical form
of the nanoparticles (size, aggregation, solubility), the use of appropriate dose
metrics, compounding effects from other contaminants (natural organic matter,
surfactants) and photocatalytic effects. In research studies appropriate controls
comprise micron-sized versions of the nanomaterial (e.g. particulate zinc oxide,
gold, silver, etc.) and where possible a dissolved control (e.g. ionic silver for
silver nanoparticle toxicity studies). For routine toxicity testing, control solutions
comprising a ' standard ' nanoparticle with known toxicity would be highly
desirable.
As already discussed, nanoparticle aggregation in some cases can be minimised
by the addition of compounds that change the surface properties, by changing solu-
tion pH, by coating or by altering surface charge. In practice, this may not be
entirely feasible as many toxicity tests may be affected by the additives used to
disperse the nanoparticles. Any toxicity testing in the presence of such compounds
must include controls that determine the toxicity of the additives alone. Even so,
the role of the additives in facilitating interactions of nanoparticles with cell mem-
branes will not be distinguished by such controls, so data interpretation will be
diffi cult. Similarly, as will be discussed later in relation to carbon nanotubes, addi-
tives in the form of solvents used in the preparation of nano-sized particles can
have toxic effects that can be interpreted as being due to the nanoparticles (Zhu
et al. , 2006 ).
A number of chemical measurements are required in order to fully understand
the dynamics of nanoparticle interactions during a toxicity test. The determination
of particle size at the start and fi nish of the test is highly desirable, as this will
characterise any aggregation that may occur over the time course of the toxicity
test. It will be important in assessing nanoparticle toxicity to determine the extent
to which dissolution of particles contributes to any observed effects. Size-based
separation methods such as ultrafi ltration and dialysis are appropriate procedures
to measure the soluble fractions. Franklin et al. (2007) showed that nanoparticulate
zinc oxide had appreciable solubility in water, despite literature reports that it was