reduce aggregation and counteract aggregation effects, and even organisms may
cause dispersion by producing organics that stabilize nanoparticles in suspension
(Unrine et al. 2012).
28.3.1 Endpoint Selection
An ecotoxicological endpoint is the parameter measured as a response to a potentially toxic substance. Numerous endpoints may be used when assessing the effects
on a test organism. For many organisms, mortality or growth rate are rather coarse
endpoints used for testing acute toxicity, while enzymatic activities (e.g., of antioxidative enzymes), genetic mutations, or expression of genes related to damage
repair or stress are gradually more sensitive test endpoints, permitting detection of
more subtle and chronic adverse effects at lower concentrations (Walker et al. 2001).
The interpretations of the responses to the endpoints with less obvious toxicity
functions are however a minefield. Are they, e.g., altered behavior, avoidance,
expression of stress-related genes, or enhanced frequencies of apoptosis indicators
of toxicity? For example, if nZVI is introduced into soil or a test system, the redox
conditions may change rapidly as to cause oxidative stress (or irreversible organismal damage) due to reactions of free O 2 with nZVI. But frequently such conditions
are short-lived (El-Temsah et al. 2013), and oxygen will diffuse in from the border
zones and re-establish oxygenated conditions and alleviate the stress caused by the
reduced O 2 availability. For those organisms that have survived the period of
reduced O 2 availability, the effects may be fully reversible, with no negative impacts
on populations or communities (Nguyen et al. 2018a). Thus, if the exposure ceases
and the organism has avoided it or only passed through a period with sub-optimal
living conditions (stress) due to the nanoparticle exposure, it is not appropriate to
interpret this as toxicity.
A particular consideration to make when it comes to testing of nZVI and other
nanoparticles for remediation that may affect oxygen availability to organisms is the
fact that these nanoparticles may cause a lack of oxygen needed by aerobic organisms during respiration. nZVI may, e.g., react with the available O 2 in the test
medium as to render the conditions anoxic, thus asphyxiating the test organisms.
This is particularly relevant for exposure in water and wet soil where O 2 diffusion
and replenishment is slow. Such induction of anaerobic conditions and its detrimental effects on aerobic organisms is not a nano-specific effect (though the dynamics of
O 2 consumption may differ between nanoparticles and similarly reductive
chemicals/bulk-size particles due to the specific surface area and chemical reactivity). The effects nZVI may have on alternative electron acceptors (NO 3 , SO 4 ,
oxidized forms of Mn, etc.) can similarly preclude the use of anaerobic test organisms to circumvent the need for O 2 during testing.
Testing for nanoparticle-specific effects when the nanoparticles to be tested cause
changes in the availability of O 2 or other electron acceptors require the use of control
treatments that have comparable redox conditions, or the use of test systems or
578
C. Coutris et al.
cause dispersion by producing organics that stabilize nanoparticles in suspension
(Unrine et al. 2012).
28.3.1 Endpoint Selection
An ecotoxicological endpoint is the parameter measured as a response to a potentially toxic substance. Numerous endpoints may be used when assessing the effects
on a test organism. For many organisms, mortality or growth rate are rather coarse
endpoints used for testing acute toxicity, while enzymatic activities (e.g., of antioxidative enzymes), genetic mutations, or expression of genes related to damage
repair or stress are gradually more sensitive test endpoints, permitting detection of
more subtle and chronic adverse effects at lower concentrations (Walker et al. 2001).
The interpretations of the responses to the endpoints with less obvious toxicity
functions are however a minefield. Are they, e.g., altered behavior, avoidance,
expression of stress-related genes, or enhanced frequencies of apoptosis indicators
of toxicity? For example, if nZVI is introduced into soil or a test system, the redox
conditions may change rapidly as to cause oxidative stress (or irreversible organismal damage) due to reactions of free O 2 with nZVI. But frequently such conditions
are short-lived (El-Temsah et al. 2013), and oxygen will diffuse in from the border
zones and re-establish oxygenated conditions and alleviate the stress caused by the
reduced O 2 availability. For those organisms that have survived the period of
reduced O 2 availability, the effects may be fully reversible, with no negative impacts
on populations or communities (Nguyen et al. 2018a). Thus, if the exposure ceases
and the organism has avoided it or only passed through a period with sub-optimal
living conditions (stress) due to the nanoparticle exposure, it is not appropriate to
interpret this as toxicity.
A particular consideration to make when it comes to testing of nZVI and other
nanoparticles for remediation that may affect oxygen availability to organisms is the
fact that these nanoparticles may cause a lack of oxygen needed by aerobic organisms during respiration. nZVI may, e.g., react with the available O 2 in the test
medium as to render the conditions anoxic, thus asphyxiating the test organisms.
This is particularly relevant for exposure in water and wet soil where O 2 diffusion
and replenishment is slow. Such induction of anaerobic conditions and its detrimental effects on aerobic organisms is not a nano-specific effect (though the dynamics of
O 2 consumption may differ between nanoparticles and similarly reductive
chemicals/bulk-size particles due to the specific surface area and chemical reactivity). The effects nZVI may have on alternative electron acceptors (NO 3 , SO 4 ,
oxidized forms of Mn, etc.) can similarly preclude the use of anaerobic test organisms to circumvent the need for O 2 during testing.
Testing for nanoparticle-specific effects when the nanoparticles to be tested cause
changes in the availability of O 2 or other electron acceptors require the use of control
treatments that have comparable redox conditions, or the use of test systems or
578
C. Coutris et al.