Published by Oxford University Press on behalf of the Society of Toxicology 2013.This work is written by (a) US Government employee(s) and is in the public domain in the US.

Review

Developmental Neurotoxicity of engineered Nanomaterials: identifying Research Needs to Support Human Health Risk Assessment

Christina M. Powers,* Ambuja S. Bale,† Andrew D. Kraft,† Susan L. Makris,† Jordan Trecki,† John Cowden,* Andrew Hotchkiss,* and Patricia A. Gillespie*,1

*Office of Research and Development, National Center for Environmental Assessment, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711; and †Office of Research and Development, National Center for Environmental Assessment, U.S. Environmental Protection Agency,

Washington, DC 20460

1To whom correspondence should be addressed at National Center for Environmental Assessment, U.S. Environmental Protection Agency, MD B243-01, Research Triangle Park, NC 27711. Fax: (919) 541-2985. E-mail: gillespie.patricia@gmail.com.

Received January 15, 2013; accepted April 23, 2013

Increasing use of engineered nanomaterials (ENM) in con-sumer products and commercial applications has helped drive a rise in research related to the environmental health and safety (EHS) of these materials. Within the cacophony of information on ENM EHS to date are data indicating that these materials may be neurotoxic in adult animals. Evidence of elevated inflammatory responses, increased oxidative stress levels, alterations in neu-ronal function, and changes in cell morphology in adult animals suggests that ENM exposure during development could elicit developmental neurotoxicity (DNT), especially considering the greater vulnerability of the developing brain to some toxic insults. In this review, we examine current findings related to develop-mental neurotoxic effects of ENM in the context of identifying research gaps for future risk assessments. The basic risk assess-ment paradigm is presented, with an emphasis on problem for-mulation and assessments of exposure, hazard, and dose response for DNT. Limited evidence suggests that in utero and postpartum exposures are possible, while fewer than 10 animal studies have evaluated DNT, with results indicating changes in synaptic plas-ticity, gene expression, and neurobehavior. Based on the avail-able information, we use current testing guidelines to highlight research gaps that may inform ENM research efforts to develop data for higher throughput methods and future risk assessments for DNT. Although the available evidence is not strong enough to reach conclusions about DNT risk from ENM exposure, the data indicate that consideration of ENM developmental neurotoxic potential is warranted.

Key Words: developmental neurotoxicity; engineered nanoma-terials; risk assessment research.

Engineered nanomaterials (ENM) are increasingly used in consumer products and commercial processes, raising concern over the potential health impacts of increased human exposure (Hansen et al., 2008; Hendren et al., 2011). The definition of ENM continues to evolve as the nanotechnology field itself grows (Grieneisen and Zhang, 2011; Maynard, 2011). In this review, we use the National Nanotechnology Initiative defini-tion: ENM are materials intentionally synthesized or manufac-tured with at least one external dimension of 1 to ~100 nm and that possess unique properties due to their size (NSTC, 2011). Products in which manufacturers claim to use ENM include consumer products such as cosmetics, fabrics and clothing, per-sonal care items, cleaning solutions, sporting equipment and electronics, and goods specifically intended for use by children (e.g., toys) (Hansen et al., 2008). Within these various products is a diversity of ENM types, including nanoscale silver (nano-Ag), titanium dioxide (nano-TiO

2), and carbon (nano-C).

Based upon the variety of ENM products and applications, coupled with variability in ENM production methods and measurement techniques, it has been suggested that environmental health and safety testing of ENM could be both time consuming and expensive. In the United States, environmental health and safety testing could take 34–53 years and cost an estimated $0.25–$1.18 billion, depending on the extensiveness of testing required to determine safety (i.e., minimal testing required vs. long-term in vivo testing to establish safety; see Choi et al., 2009 for greater details). Notably, this estimate is based on a 2009 analysis of U.S. firms producing or using ENM and thus does not reflect resources that might be necessary to test new ENM coming into the market. Recognizing that the use and development of novel ENM applications will outpace environmental health and safety research, several

DNT ENM RISK ASSESSMENT RESEARCH

POWERS ET AL.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency.

toxicological sciences 134(2), 225–242 2013doi:10.1093/toxsci/kft109Advance Access publication May 24, 2013

preliminary evaluations have used currently available data to highlight data gaps for assessing human health risks from ENM exposure (Aschberger et  al., 2011; Savolainen et  al., 2010; Wijnhoven et al., 2009). Such data gaps include exposure data, hazard identification and characterization information, quantitative dose-response data, standards related to metrics for ENM testing, and analytical approaches for reliable detection of ENM in biological and environmental milieus.

A variety of toxicity outcomes are associated with ENM expo-sure in animal models (Savolainen et al., 2010). Of these out-comes, of particular note for this article are indications that some of these materials translocate to the brain and evoke biological effects in the nervous system of animal models (Boyes et al., 2012; Hu and Gao, 2010; Oberdörster et al., 2009; Sharma and Sharma, 2010; Simkó and Mattsson, 2010; Yang et al., 2010). Biological changes in the brain of adult animals following ENM exposures include elevated inflammatory responses, increased oxidative stress levels, alterations in neuronal function, and changes in cell morphology (Simkó and Mattsson, 2010). These types of effects on the nervous system could result in functional neurotoxicity, and indeed, early evidence shows behavioral changes (e.g., decreased motility in open field tests) (Simkó and Mattsson, 2010). It remains unclear, however, whether the behavioral effects are ENM specific or not (e.g., conflict-ing findings on whether nanoscale vs. ionic Mg led to changes in open field motility as detailed in Oszlánczi et al., 2010a, b). However, neurotoxic effects of ENM in the adult brain suggest the potential for developmental neurotoxicity (DNT) from ENM exposure due to the greater vulnerability of the developing brain to toxic insults (reviewed by Bondy and Campbell, 2005 and Rice and Barone, 2000), as discussed further below.

Here, we highlight several characteristics of ENM and the developing brain that suggest a particular concern for DNT from ENM exposure. First, production volume estimates indi-cate that metal or metal-oxide ENM are widely used (Hendren et al., 2011), and some of these metals (e.g., silver, cadmium, and manganese) are associated with adverse neurological effects prior to being manufactured at the nanoscale (Burton and Guilarte, 2009; Lansdown, 2007; Méndez-Armenta and Rios, 2007). Secondly, even materials without known neu-rotoxic effects (e.g., carbon-based materials) are increas-ingly produced as ENM (Innovative Research and Products Incorporated, 2011) and may disrupt neurodevelopment when manufactured into ENM, as the higher surface area to volume ratio makes them more biologically reactive compared with larger size materials (i.e., those with dimensions greater than 100 nm) (Oberdörster et  al., 2005). Thirdly, the blood-brain barrier (BBB) continues to mature up to about 6 months of age in humans, which when coupled with the small size of ENM, increases the likelihood of ENM reaching the developing brain (Watson et al., 2006). Finally, the immature status of several systems may lead to greater effects of ENM in the develop-ing fetus. For instance, metabolic systems capable of protect-ing the body from toxicant exposures, such as cytochrome

P450 enzymes, are still developing in the fetus (Bondy and Campbell, 2005). As a result, the ability to metabolize ENM or ENM components (e.g., surface coatings on ENM) will likely differ across life stages, and studies to determine such potential differences will need to account for the ontogeny of expres-sion of metabolic enzymes (Wrighton and Stevens, 1992). Such studies might focus on determining whether there are particular types of ENM for which the fetus could serve as a “sink” for ENM in maternal circulation, suggesting a period of prolonged exposure to the developing brain should ENM cross the placen-tal barrier. In fact, recent evidence indicates that certain sizes of ENM pass through the human placental barrier (Wick et al., 2010) and accumulate in fetal tissues, including the brain, of animal models (Gao et al., 2011; Yamashita et al., 2011).

In summary, the greater susceptibility of the developing brain, unique size and reactivity of ENM compared with larger sized materials, and differences in maturation of key systems between adults and young children, all suggest that attention to the poten-tial for DNT is warranted. Here, we review current data through March 2013 related to the potential for ENM DNT and then use established testing guidelines (OECD, 2007; U.S. EPA, 1998) to highlight areas of research that could inform future ENM risk assessment efforts related specifically to DNT. We further note how current DNT guidelines could be used to help develop a tiered testing approach for the more rapid, inexpensive testing and characterization of ENM that will likely be necessary given the increase in ENM production (Hendren et al., 2011).

RISk ASSESSMENT of ENM

A standard paradigm for the assessment of risks to human health resulting from environmental chemical exposures has been described by the National Research Council of the National Academy of Science as consisting of the processes of hazard identification, dose-response assessment, and expo-sure assessment (NRC, 1983). These components contribute to the development of a risk characterization that provides critical information for risk management decisions and actions within a broader social and regulatory context. The value and impor-tance of approaching the risk assessment process with a clear understanding of the information that will be needed to charac-terize the risks have been recognized and formalized in a prob-lem formulation and scoping step that precede the actual risk assessment (NRC, 1994, 2009) (Fig. 1).

In the context of human health risk assessment, DNT testing is a unique and important tool for hazard identification and dose-response assessment of potentially neurotoxic chemicals providing insight on functional, behavioral, and anatomical changes to the nervous system, which result from exposures during critical periods of neurological development and which are not evaluated in other testing procedures (Makris and Raffaele, 2009). The U.S. Environmental Protection Agency (EPA) DNT screening guideline (shown in Fig.  2) has been evaluated for

226 POWERS ET AL.

sensitivity in detecting changes in neurological development for a variety of neurotoxic chemicals (Makris et al., 2009; U.S. EPA, 1998). Included in this screening guideline, which is generally conducted in rats, are measures of (1) developmental landmarks (e.g., pup weight gain and sexual maturation), (2) neurobehavior (e.g., functional observations, auditory startle habituation, motor activity testing, and cognitive testing), (3) neuropathology, and (4) brain weight (U.S. EPA, 1998). The Organisation for Economic Co-operation and Development (OECD) guideline for DNT testing (GL 426) (OECD, 2007) is harmonized with the EPA DNT guideline but includes a number of refinements, such as a longer postnatal exposure period (see Table 2 in Makris and Raffaele, 2009). As discussed below, both of the EPA and OECD guidelines could help identify ENM DNT research that would support risk assessment work in this area; however, challenges specific to ENM necessitate that the study design will need to be carefully crafted. Thus, we note areas in which studies with ENM might need to expand upon current DNT testing guidelines.

AvAILABLE EvIDENcE foR coNDucTINg ENM DNT PRoBLEM foRMuLATIoN

As discussed above, problem formulation is critical to develop a clear understanding of what information to include in subsequent steps of the risk assessment process. This section

thus steps through problem formulation for potential ENM exposure and hazard in preparation for subsequent sections that summarize available data on ENM DNT and related data gaps that could inform future assessments.

Regarding potential exposure scenarios for ENM, many ENM-containing products are consumer goods as outlined earlier (Hansen et al., 2008; Hendren et al., 2011). In turn, the production, use, and disposal of these products could result in a number of human exposure scenarios through either direct contact with ENM or ENM products, or environmental contact with ENM released during any of these stages of the product life cycle. Some of these scenarios could lead to ENM exposure to the developing nervous system, of which two primary potential exposure scenarios emerge: (1) in utero exposures following maternal inhalation, ingestion, or dermal contact of ENM, and (2) postpartum exposures from the same portals of entry in the young child (Fig. 3). Although biomedical applications (e.g., pharmaceuticals) could also result in iv exposures, the potential impacts on the developing nervous system are expected to be discussed with a medical professional. In turn, the focus here is on incidental exposures to ENM. Although exposure at conception is not expected to have significant latent effects on CNS development, the possibility of ENM persisting in the mother and affecting neural tube development or related processes several weeks following conception

fIg. 1. Framework laid out by the National Research Council for conducting risk assessment with an increased focus on problem formulation (phase 1).

DNT ENM RISK ASSESSMENT RESEARCH 227

cannot be completely ruled out. In utero and postpartum ENM exposure could result in distinct neurological effects, owing to differences in the stage of nervous system developmental and the biological barriers that ENM need to cross to reach the developing nervous system.

The nervous system, and particularly the brain, undergoes a variety of spatiotemporally regulated developmental pro-cesses from around 2 weeks of gestation, when the neural plate forms in humans, to adolescence (Makris and Raffaele, 2009; Rice and Barone, 2000). Given the diversity of developmental changes occurring during this time period, the developing nerv-ous system is particularly sensitive to toxicant exposures, and the outcome(s) of exposure is heavily influenced by the timing of exposure (Rice and Barone, 2000). For instance, although apoptosis and synaptogenesis continue from mid-gestation to the postnatal period, myelination occurs mostly in the late ges-tation and postpartum periods (see Fig. 1 in Rice and Barone, 2000). Thus, the dynamic maturation of the brain means that postpartum exposures could affect neurodevelopment in ways distinct from in utero exposures because different neurodevel-opmental processes are operant at different times during gesta-tion and postpartum development.

Throughout brain development, there are three categories of exposure sources to consider for ENM. They are (1) occu-pational exposures in which individuals manufacturing ENM come into contact with the material, either before or after incor-poration into a product, (2) exposure from using consumer products containing ENM, and (3) environmental exposure to ENM released during product manufacturing, use, or disposal. Levels of actual ENM exposures in each of these categories are generally poorly understood, but some estimates from recent modeling efforts are summarized in Table 1. In general, occupational ENM exposures are expected to be higher than environmental ENM exposures, but as the application of ENM continues to increase, so does the likelihood of aggregate and cumulative exposures to multiple types of ENM via multiple routes in consumer products or other applications.

During in utero development, the large surface area of the placenta combined with the increased surface area to volume ratio of nanoparticles compared with larger sized materials increases the likelihood of fetal ENM exposure (Saunders, 2009). Further, maternal differences in absorption, distribution, metabolism, and excretion during pregnancy may influence prenatal ENM exposure levels. For instance, higher pulmonary

fIg. 2. The DNT study design as specified in U.S. EPA guidelines is unique among standard regulatory testing protocols in that it includes exposures to test subjects (generally conducted in rats) during critical periods of pre- and postnatal nervous system development and calls for assessments of effects on functional and structural outcomes at multiple time points and life stages.

228 POWERS ET AL.

function during pregnancy results in an approximately 72% increase in air exchange volume over an 8-h period (Selevan et al., 2000). Although this increased air exchange could result in greater exposure levels, ENM exposure could itself impact respiratory function (e.g., decrease respiration rate due to inflammatory response; Mühlfeld et  al., 2008), which would need to be taken into account as well when evaluating potential exposure levels during pregnancy.

Consideration of postpartum ENM exposure may be particularly relevant due to several factors unique to infants and young children (Selevan et al., 2000), including (1) greater surface area to body mass ratio (i.e., ~2.7-fold greater ratio in infants than adults), (2) higher ventilation rates (i.e., 4 times higher in infants [~44 breaths/min] compared with adults [12 breaths/min]; Besson et  al., 2010; Sherwood, 2013), and (3) increased drinking water intake (i.e., 4 times greater intake in infants compared with adults per body weight; U.S. EPA, 2011). These characteristics of young children combined with the small size and increased reactivity of ENM compared with larger size materials suggest a greater potential for DNT as a result of ENM exposure throughout neurodevelopment. Notably, lactational exposure primarily occurs during the early postpartum developmental period with some rare exceptions (e.g., mothers choosing to nurse children past 1 year of age) and highlights the potential for direct exposure to the developing

brain via particle uptake by afferent sensory fibers in the mouth, followed by axonal transport to the brain. Although such uptake has not been demonstrated for ENM to date, transport up afferent olfactory fibers has been demonstrated in adult rodents for nanosized gold and other types of ENM (Simkó and Mattsson, 2010). For lipophilic ENM (e.g., fullerenes, carbon nanotubes), lactational exposure might be a particularly important route of exposure to consider based on evidence of elevated levels of lipophilic compounds such as polychlorinated biphenyls in breast milk (Schwartz et al., 1983).

ENM characteristics other than lipophilicity (e.g., production processes, size, surface, area, shape) will similarly influence potential exposures throughout in utero or postpartum develop-ment (see Yokel and Macphail, 2011 for discussion on how these characteristics might influence occupational exposures). The propensity of ENM to agglomerate or adhere to other materials means that individuals are likely exposed to ENM with different sizes, surface charge, or other properties than the materials as produced. In turn, these considerations need to be accounted for in efforts to understand the potential for exposure and effects in the developing nervous system. Importantly, the likelihood and level of any of the exposure scenarios discussed here remain to be determined. The references listed in Table 1 and others (Benn and Westerhoff, 2008; Dahm et al., 2011; Hendren et al., 2011; Wijnhoven et al., 2009) provide additional information.

fIg. 3. The potential for exposure to engineered nanomaterials exists during both in utero and postpartum development. Differences in the types and devel-opmental status of biological barriers and the types of exposure routes are depicted here. Some sources of exposure are likely specific to developmental stages as well, for instance, children might mouth their hands or other body parts, which could lead to oral exposure to sunscreen but this would be unlikely to be a source of oral exposure for mothers. Similarly, oral or dermal contact with toys is more likely for young children than mothers.

DNT ENM RISK ASSESSMENT RESEARCH 229

Another consideration for problem formulation of ENM DNT is the vulnerability of the developing brain. As mentioned ear-lier, recent evaluations of ENM exposures in adult animal mod-els show neurotoxic outcomes, which may have greater impact in the developing brain. Although the physiological relevance of many of these observations in adults is unknown due to the study design (i.e., relatively high-dose concentrations, lack of mate-rial characterization, methodological questions), the findings may have implications for the more vulnerable developing brain. For instance, exposure to 80 or 155 nm TiO

2 resulted in signifi-

cant cell loss in the hippocampus of adult animals (Wang et al., 2008a). Given that cell death is programmed to occur in specific areas of the developing brain at certain times (for a review, see Costa et al., 2004b), alterations in apoptosis resulting from ENM exposure could interfere with neurological function later in life.

Similarly, neuronal cell migration is a highly orchestrated process that results in specific cells in particular places at selected times during brain development (Costa et al., 2004a). Data showing altered neuronal morphology in the adult brain after ENM exposure suggest that such migration events may be altered by ENM exposure during neurodevelopment (Ma et al., 2010; Wang et al., 2008a, b). Reports also show alterations in oxidative stress (Wang et al., 2008b) and indications of glial cell activation in the adult brain (Elder et al., 2006; Maysinger et al., 2007). Glial cells protect neurons from oxidative stress (Tanaka et al., 1999) and are a target of other known develop-mental neurotoxicants (e.g., nicotine, alcohol, methylmercury) (Abdel-Rahman et  al., 2005; Guizzetti et  al., 1997; Shanker et al., 2003). Importantly, the major period of glial cell prolif-eration does not begin until the third trimester of pregnancy and continues well into postnatal maturation (Costa et al., 2004b), leaving the developing brain vulnerable to elevations in oxi-dative stress that might result from ENM exposure. A  num-ber of factors, including higher availability of iron to catalyze free radical formation and lower levels of antioxidant, in the developing brain amplify vulnerability to oxidative stress from ENM (Blomgren and Hagberg, 2006). Data further indicate altered neurotransmitter levels in the adult brain after ENM

exposure (Ma et al., 2010). Because neurotransmitters are criti-cal to regulating cell proliferation, migration, and differentia-tion in the developing brain (Costa et al., 2004b), such effects in the developing brain could have long-term consequences on neurobehavior and neuronal function. Finally, observations in adult animal models show inflammation in the brain after ENM exposure (Elder et al., 2006; Maysinger et al., 2007; Tang et al., 2009a). Recent evidence suggests that systemic inflammation negatively affects neurodevelopment and may have long-term consequences on behavior and psychiatric health (Hagberg and Mallard, 2005), again suggesting that ENM exposure in the developing brain could lead to persistent changes later in life. As discussed in the Hazard Identification section below, initial data suggest that several observations noted in the brain of adult animal models also occur in the developing brain after in utero exposure or lactational exposure. Presumably, these exposures resulted in lower concentrations in the developing brain than the adult brain because the offspring were not directly dosed. The similarity in results thus supports the potential for greater vulnerability to ENM in the developing brain.

AvAILABLE EvIDENcE foR coNDucTINg ENM DNT ExPoSuRE ASSESSMENT

As depicted in Figure 3, in utero exposures lead to the poten-tial for indirect exposure to the developing nervous system as the material has to cross several barriers (e.g., maternal skin, diges-tive tract lining, and placenta). In contrast, ENM exposure during the postpartum period would need to cross fewer barriers (i.e., no placental barrier) to reach the developing nervous system; how-ever, these barriers are more developed and thus likely less pen-etrable by ENM (Rice and Barone, 2000). Of these barriers, two are most relevant when discussing the potential for ENM to enter the developing fetal brain: the placental barrier and the BBB.

Early evidence in animal models shows that ENM with certain sizes and coatings can cross the placental barrier (Yamashita et al., 2011). Although differences between rodent

TABLE 1Example Exposure Levels

Exposure route Exposure scenario Exposures References

Inhalation Occupational Indirect exposure, prenatal 0.001–0.46 mg/m3 (Gangwal et al., 2011)Environmental, Soil Indirect exposure, prenatal and

postnatal; direct exposure, postnatal0.024–8.3 Δµg/kg/year (Gottschalk et al., 2009)

Environmental, Air ≤ 0.0005–0.0020 µg/m3 (Gottschalk et al., 2009)Inhalation and dermal Indoor spray product (e.g., cleaning

solution applied twice a year)Indirect exposure, prenatal and

postnatal43.8 µg/kg bw/year in

adult woman(Hansen et al., 2008)

Oral Environmental, Soil Indirect exposure, prenatal and postnatal; direct exposure, postnatal

0.024–8.3 Δµg/kg/year (Gottschalk et al., 2009)

Dermal Consumer product, Sunscreen Direct exposure, postnatal 56.7 mg/kg bw/day in child (Hansen et al., 2008)

Note. A number of assumptions were used to reach these estimates, including homogenous regional distribution, well-mixed environmental compartments, and physicochemical differences in individual ENM (e.g., singlewalled vs. multiwalled carbon nanotubes), and thus caution is warranted in reviewing the results. Δ, change; bw, body weight. mg/m3 and mg/kg-bw/time, refers to predicted exposure concentrations; Δµg/kg/year, refers to the annual increase in ENM levels.

230 POWERS ET AL.

and human placenta structures necessitate caution in interpret-ing such findings, a recent human ex vivo study demonstrated that ENM ≤ 240 nm can cross this barrier (Saunders, 2009; Wick et  al., 2010). Animal models also indicate that metal and metal oxide ENM can cross the developing BBB (Takeda et al., 2009; Yamashita et al., 2011). This is in contrast to stud-ies of the developed BBB in adult animals, which shows that similar types of ENM cross at low concentrations or not at all (Cho et al., 2009; De Jong et al., 2008; Hardas et al., 2010). Moreover, recent data show that ENM can disrupt the BBB integrity through elevating inflammatory responses or reduc-ing expression of tight junction proteins (e.g., occludin-, clau-din-5) in the BBB endothelium (Chen et al., 2008; Sharma and Sharma, 2007; Sharma et al., 2009; Tang et al., 2010; Trickler et al., 2010, 2011). Efforts to design therapeutics for the brain show that specific surface coatings and sizes can enhance ENM delivery to the brain (Kreuter, 2004). Importantly, the mechanism(s) by which ENM cross biological barriers such as the placenta and BBB is poorly understood and may involve passive diffusion, active transport, or more likely, a combina-tion of the two depending on the ENM physical and chemical characteristics (Chithrani et al., 2006; Lockman et al., 2002).

Whether ENM cross these barriers will likely depend on the route of exposure (e.g., inhalation, oral, dermal). For example, inhalation exposure in animal models has demonstrated that ENM can bypass the BBB via anterograde axonal transport to the olfactory bulb (Oberdörster et al., 2009). Whether dermal exposure leads to systemic circulation of ENM is contested (Gulson et al., 2010; Popov et al., 2010), and no studies have specifically evaluated translocation to the developing brain after this type of exposure. Importantly, the extent to which ENM reaches the developing brain is expected to also be heavily dependent on the specific ENM physical and chemical charac-teristics because properties such as size, surface charge, and sur-face coating all influence biodistribution (Maynard et al., 2011).

AvAILABLE EvIDENcE foR coNDucTINg ENM DNT HAzARD IDENTIfIcATIoN

As discussed further in the subsections that follow, in vivo and in vitro studies have examined the DNT potential of ENM expo-sure. Perinatal exposure in animal models has resulted in electro-physical, histopathological, and behavioral effects in offspring. In vitro studies have provided mechanistic data for the DNT effects observed in the in vivo studies. Together, these studies suggest that gestational exposure to materials not characterized as DNT agents may in fact lead to DNT effects when they are synthesized at the nanoscale; however, as discussed in greater detail below, there are a number of limitations to the currently available data.

In Vivo Evaluations of Hazard

A limited number of in vivo studies evaluating potential haz-ardous effects of ENM in the developing nervous system were

identified (See Fig. 4 and Table 2), and none of these studies utilized a standard guideline DNT testing approach. In addi-tion, several identified studies used sc exposures to determine potential ENM effects, which is not expected to be a relevant exposure route for ENM products; however, due to the limited number of studies available, these studies are included here to help guide future research.

DNT Observations Following Oral ENM Exposure

Only one study examined DNT effects of oral exposure to ENM. Pregnant or lactating rats were treated orally with a nano-TiO

2 suspension in distilled water (final dosage 100 mg/kg

body weight) from gestational day (GD) 21 to day 2 or post-natal day (PND) 2–21, respectively. Results indicated that the exposure during pregnancy or lactation impaired offspring short-term synaptic plasticity in the hippocampal dentate gyrus (DG) area (Gao et  al., 2011). In addition, after lacta-tional exposure, both long-term synaptic plasticity of the hippocampal DG area and basic synaptic transmission were weakened, effects which were not seen in offspring exposed during pregnancy. These results occurred in the presence of significant increases in nano-TiO

2 in the hippocampus and

indicate that developmental nano-TiO2 exposure could impair

synaptic plasticity in rats’ hippocampal DG area, especially during lactational exposure.

DNT Observations Following sc ENM Exposure

Three studies examined the effects of sc exposure to nano-TiO

2 (100  µl of 1 mg/ml TiO

2) in pregnant mice during ges-

tation (Shimizu et  al., 2009; Takahashi et  al., 2010; Takeda et al., 2009). In a gene expression study, Shimizu et al. (2009) exposed dams on GD 6, GD 9, GD 12, and GD 15 and then collected brain tissue from male offspring on GD 16 and on PND 2, PND 7, PND 14, and PND 21. Based on microarray data, the authors reported significant changes in gene expres-sion at each time point examined, with the largest number of significant changes observed at PND 21. Authors categorized the gene expression changes using gene ontology (GO) and medical subject headings (MeSH) related to neurodevelop-ment and determined statistical significance using an enrich-ment factor (ratio of differently expressed genes in a category to ratio of differentially expressed genes in the microarray) for each GO or MeSH category. In general, data showed altered gene expression associated with apoptosis, brain development, oxidative stress, inflammation, and neurotransmitters (epineph-rine, norepinephrine, serotonin, and glutamic acid) in what could be an age-dependent manner. Results suggest that mater-nal nano-TiO

2 exposure impacted prenatal gene expression and

that alterations may persist after cessation of maternal exposure although data are difficult to interpret without accompanying functional measurements or accounting for basal expression differences in different gene categories.

A similar exposure paradigm elevated apoptotic cells in the olfactory bulb of 6-week-old male offspring and increased

DNT ENM RISK ASSESSMENT RESEARCH 231

levels of TiO2 in the olfactory bulb and cerebral cortex compared

with controls (Takeda et al., 2009). Microscopy combined with spectroscopy data demonstrates translocation of nanomaterials to the developing nervous system following maternal exposure although the authors did not indicate whether TiO

2 nanomateri-

als were observed at an earlier time point on PND 4 and did not examine other areas of the brain for apoptotic markers. The observation of nanomaterials and increased apoptosis in the olfactory bulb at 6 weeks of age suggests that both nanomateri-als and the neurotoxic effects of nanomaterials may persist in the developing brain even after cessation of maternal exposure; however, the authors did not provide a clear rationale regarding the mechanism through which nano-TiO

2 might have resulted

in the observed effects after sc exposure.Takahashi et al. (2010) exposed pregnant mice to 100 µl of

1 mg/ml nano-TiO2 sc at similar time points in gestation. Of the

10 offspring brain regions examined, statistically significant changes in dopamine and dopamine metabolites were only observed in the prefrontal cortex and the neostriatum. In the prefrontal cortex, maternal nano-TiO

2 exposure significantly

decreased the monoamine levels. Conversely, monoamine lev-els in the neostriatum were significantly increased following maternal exposure. Interestingly, the authors did not observe

any monoamine level changes in the olfactory bulb, despite previous reports from this lab, which identified the olfactory bulb as a target of prenatal nanomaterial exposure (Takeda et al., 2009). This study demonstrated that maternal nano-TiO

2

exposure impacted monoamine levels after cessation of mater-nal dosing. However, the lack of an exposure assessment or characterization of the functional impact of monoamine level changes limits the ability of this study to inform the poten-tial neurodevelopmental impact of gestational exposure to nanomaterials.

DNT Observations Following Inhalation ENM Exposure

One study evaluated DNT after inhalation exposure to nano-TiO

2 in pregnant mice (Hougaard et  al., 2010). After mater-

nal exposure for 1 h/day to 42 mg/m3 aerosolized powder from GD 8–18, a number of endpoints were examined including maternal lung inflammation, gestational and litter parameters, offspring neurofunction, and fertility. In addition, particle physicochemical properties and exposure were characterized. Gestationally exposed offspring displayed moderate neurobe-havioral alterations as young adults, including altered open field behavior (males and females) and enhanced prepulse inhi-bition (females). No effect was observed on cognitive function

fIg. 4. Recommended exposures in DNT guidelines (A), exposures tested to date (from Table 2) in ENM studies (B), and endpoints to be assessed accord-ing to DNT guidelines with corresponding information in ENM studies (C); √ = age(s) recommended by EPA or OECD DNT guidelines for testing the specified endpoint(s); 1growth includes endpoints such as eye opening and incisor eruption; 2cage-side observations for evidence of deficits in functional development; 3age for acquisition of competence, for example, surface righting and edge avoidance; 4these endpoints can be alternatively assessed at PND 11 or PND 21.

232 POWERS ET AL.

TA

BL

E 2

Sum

mar

y of

Haz

ard

Iden

tific

atio

n St

udie

s W

ith

In V

ivo

EN

M E

xpos

ures

EN

MSi

ze (

nm)

[EN

M]

Exp

osur

eM

ain

findi

ngs

Ref

eren

ces

TiO

2 ana

tase

< 2

510

0 m

g/kg

SD r

ats

Ora

lG

D 2

1–PN

D 2

; PN

D 2

–21

10 o

ffsp

ring

/gro

up: E

NM

in h

ippo

cam

pus

↓hip

poca

mpa

l DG

I/O

(PN

D)

↓pai

red-

puls

e re

actio

n (P

ND

)

(Gao

et a

l., 2

011)

TiO

2 rut

ile10

–100

40 m

g/m

3C

57B

L/6

Bom

Tac

mic

eIn

hala

tion

GD

8–1

8, 1

h/d

ay

10–1

2 of

fspr

ing/

grou

p:14

wee

ks, o

pen

field

: ↓vi

sits

(M

), ti

me

(F)

↑pre

puls

e in

hibi

tion.

(F)

Mat

erna

l lun

g in

flam

mat

ion

(Hou

gaar

d et

 al.,

201

0)

TiO

2 ana

tase

27, 2

,429

100 

μl o

f 1

mg/

ml i

n sa

line

+ 0

.05%

Tw

een8

0IC

R m

ice

sc GD

6, 9

, 12,

15,

and

18

8 m

ales

/gro

up:

6 w

eeks

: ↑D

A a

nd m

etab

olite

(DO

PAC

, HV

A, 3

-MT

) le

vels

:N

eost

riat

ium

and

pre

fron

tal c

orte

x

(Tak

ahas

hi e

t al.,

201

0)

TiO

2 ana

tase

25–7

010

0 μ

l of

1 m

g/m

l in

salin

e +

0.0

5% T

wee

n80

Slc:

ICR

mic

esc G

D 3

, 7, 1

0, a

nd 1

4

8 m

ales

/gro

up:

4 an

d 6

wee

ks: 2

2%↓b

w@

6 w

eeks

: EN

M in

olf

acto

ry b

ulb

and

cere

bral

cor

tex

↑apo

ptot

ic m

arke

rs, o

lfac

tory

bul

b

(Tak

eda

et a

l., 2

009)

TiO

2 ana

tase

2570

100 

μl o

f 1 

μg/

μl i

n sa

line

+ 0

.05%

Tw

een8

0IC

R m

ice

sc GD

6, 9

, 12,

and

15

8–10

mal

es/g

roup

:G

ene

expr

essi

on c

hang

e, e

.g.—

Cel

l dea

th (

PND

2, 7

, 14,

and

21)

Bra

in d

evel

opm

ent (

PND

2, 7

, and

14)

Oxi

dativ

e st

ress

(G

D 1

6, P

ND

2, 7

, 14,

and

21)

Neu

rotr

ansm

itter

act

ivity

(PN

D 1

4 an

d 21

)M

otor

act

ivity

(PN

D 2

, 7, a

nd 1

4)

(Shi

miz

u et

 al.,

200

9)

Car

bon

blac

k14

0a11

, 54,

and

268

 μg/

anim

alM

ice

Intr

atra

chea

lG

D 7

, 10,

15,

and

18

Fem

ales

(hig

h do

se o

nly)

habi

tuat

ion

chan

ge (

Act

ivity

↑,↓

)M

ater

nal i

nflam

mat

ion

(Jac

kson

et a

l., 2

011)

TiO

2

rutil

e35

100 

μl o

f 0.

8 m

gB

AL

B/c

mic

eiv

, tai

lG

D 1

6 an

d 17

TiO

2 in

plac

enta

, fet

al li

ver

and

brai

n↓u

teri

ne w

eigh

tM

ater

nal w

eigh

t ↓

(Yam

ashi

ta e

t al.,

201

1)

C60

143

100 

μl o

f 0.

8 m

gB

AL

B/c

mic

eiv

, tai

lG

D 1

6 an

d 17

No

effe

ct(Y

amas

hita

et a

l., 2

011)

Not

e. 3

-MT,

3-m

etho

xyty

ram

ine

hydr

ochl

orid

e; b

w, b

ody

wei

ght;

DO

PAC

; 3,4

-dih

ydro

xyph

enyl

acet

ic a

cid;

F, f

emal

e; H

VA

, hom

ovan

illic

aci

d; I

CR

, im

prin

ting

cont

rol r

egio

ns; I

/O, i

nput

/out

put;

M,

mal

e; S

D r

ats,

Spr

ague

Daw

ley

rats

.a M

ajor

mod

e as

det

erm

ined

by

hydr

odyn

amic

num

ber

size

dis

trib

utio

n w

as n

oted

by

the

stud

y au

thor

s as

50–

60 n

m.

DNT ENM RISK ASSESSMENT RESEARCH 233

(in a Morris water maze test). Results also showed persistent maternal lung inflammation (26–27  days postexposure) after exposure termination, which could increase offspring vulner-ability to neurodevelopmental effects (e.g., impaired learning and memory, similar to observations in offspring after maternal lipopolysaccharide exposure; Golan et al., 2005). The authors of the study concluded that these results suggest that inhalation of nano-TiO

2 can induce long-term (i.e., greater than 20 days

postexposure) lung inflammation in pregnant mice and result-ing neurobehavioral alterations in gestationally exposed off-spring. The authors further noted that the effects in offspring likely derived from a combination of indirect and direct path-ways and were related to the observed maternal inflammation.

Another study examined effects of pulmonary exposure in pregnant mice on offspring neurodevelopment. Jackson et al. (2011) intratracheally instilled animals with Carbon Black (Printex 90) dispersed in Millipore water on GD 7, GD 10, GD 15, and GD 18, with total doses of 11, 54, and 268 µg Printex 90 per animal. Similar to findings described above from the same lab with nano-TiO

2 (Hougaard et  al., 2010), maternal

lung inflammation was observed after prenatal exposure to Printex 90. In offspring, no treatment effects were observed for acoustic startle or prepulse inhibition. However, the authors did observe an altered habituation pattern in open field testing for female offspring exposed to 268 µg of Printex 90. Similarly, the authors of the study suggested that the change in neurobehav-ior of exposed offspring was likely an indirect effect resulting from maternal pulmonary inflammation. The mechanism(s) by which such indirect effect occurs is unknown.

In Vitro Evaluations of Mechanisms Relevant to Neurotoxicity and DNT

A larger body of evidence exists using in vitro models to evaluate potential ENM effects, and some of those that might inform DNT evaluations are discussed here (Table 3). Questions about the relationship between in vitro and in vivo dosimetry and alterations in ENM behavior in the in vivo environment may limit the utility of in vitro studies in risk assessment efforts; however, a few findings are briefly reviewed here to highlight the mechanistic findings that may support the observed in vivo DNT effects, namely, cell death, oxidative stress, alternations in neurotransmitters, and cell type–specific effects.

Mechanisms

In vitro studies have shown elevations in oxidative stress and decreases in cell viability after ENM exposure, support-ing the observed increase in apoptotic cells in vivo by Takeda et al. (2009). Powers et al. (2011) found that silver nanomateri-als (different sizes and coatings) elicited oxidative stress and decreased cell viability in undifferentiated and neuronal PC12 cells, but at a lower potency than Ag ions. Nano-Ag and Ag ions were prepared in stock solutions equivalent to a nomi-nal concentration of 1mM Ag and diluted to achieve nominal

concentrations of Ag between 0–100µM (See Powers et  al., 2011 for additional study details). Treatment with ascorbate, an agent that blocks ionic Ag-mediated oxidative stress, was inef-fective in blocking oxidative stress from 10µM nano-Ag. These results suggest that observed effects may be due to a combina-tion of the ENM form of the material and the ENM core com-position (silver) because negative controls (silica ENM and particle-coating material alone) had no effect. A more recent adult neurotoxicity study (Zhang et  al., 2013) extended the findings from Powers et al. (2011) by showing that neurotoxic-ity is associated with Ag ENM exposure in adult male rats.

Similarly, Phenrat et  al. (2009) evaluated oxidative stress and cell viability with several types of iron ENM (1–20 µg/ml) in different cell lines. In immortalized mouse microglia (BV2 cells), fresh and aged (> 11 months) nanoscale zerovalent iron (nZVI) elevated intracellular H

2O

2 (a possible indicator of oxi-

dative stress and/ or microglial activation), but magnetite (end oxidative product of nZVI) and polyaspartate surface modified nZVI (SM-nZVI) elicited no effect at the concentrations tested. In immortalized rat dopaminergic neuron (N27) cells, levels of intracellular adenosine triphosphate (ATP) were reduced by all nZVI species (fresh [≥ 5 µg/ml, 1 h], SM-nZVI [20 µg/ml, 1 h], aged and magnetite [20 µg/ml, 6 h]). In general, findings show that nZVIs may exhibit differential toxicity to neural cell subtypes (BV2 vs. N27), with the fresh nZVI representing the most toxic species tested. Differences in particle chemistry and aging of the particle were characteristics that influenced toxic-ity (e.g., fresh nZVI was more toxic than aged nZVI). Fresh nZVI may have been more toxic because of the higher content of nanosized iron particles and thus a potential higher “redox” activity than aged nZVI. This would align with previous find-ings that nanosized particles exhibit higher reduction potential compared with their larger, non-nanoscale size counterparts (Buzea et al., 2007; Stone et al., 2007). Even though SM-nZVI had more iron than aged nZVI, the surface coating might have prevented reactive oxygen species generation in BV2 cells. These results suggest that there is neurotoxicity associated with exposure to nZVI particles with a potential for developmental neurotoxicity. These in vitro evaluations of ENM suggest that ENM characteristics such as surface coating, charge, and parti-cle size influence the effects of ENM exposure. This highlights the need for future in vitro studies to directly compare different types of ENM and subsequently translate these findings to in vivo animal models.

Additionally, both Wang et  al. (2009) and Powers et  al. (2011) reported dopaminergic alterations in undifferentiated and differentiated PC12 cells, respectively, after ENM expo-sure, supporting the in vivo changes in this neurotransmitter discussed above (Takahashi et al., 2010). Wang et al. (2009) showed that 10  μg/ml of copper (Cu-90), silver (Ag-15), or manganese (Mn-40) nanoparticles significantly decreased the content of all DA metabolites measured (DA, 3,4-dihydroxy-phenylacetic acid, and homovanillic acid). They subsequently evaluated gene expression changes at this concentration and

234 POWERS ET AL.

TA

BL

E 3

Sum

mar

y of

Haz

ard

Iden

tific

atio

n St

udie

s W

ith

In V

itro

EN

M E

xpos

ures

EN

MSi

ze (

nm)

[EN

M]

Exp

osur

eM

ain

findi

ngs

Ref

eren

ces

Ag

citr

ate

coat

ed10

1, 3

, 10,

30,

and

10

00µM

Pheo

chro

moc

ytom

a (P

C12

) ce

lls, u

ndif

fere

ntia

ted

and

diff

eren

tiate

d

Und

iffe

rent

iate

d ce

lls: I

mpa

ired

DN

A a

nd p

rote

in

synt

hesi

sD

iffe

rent

iate

d ce

lls: O

xida

tive

stre

ss, i

mpa

ired

di

ffer

entia

tion

into

the

Ach

phe

noty

pe

(Pow

ers

et a

l., 2

011)

Ag

poly

viny

lpyr

rolid

one

coat

ed10

and

50

10, 3

0, a

nd 5

0µM

PC12

cel

ls, u

ndif

fere

ntia

ted

and

diff

eren

tiate

dU

ndif

fere

ntia

ted

cells

: Im

pair

ed D

NA

syn

thes

isD

iffe

rent

iate

d ce

lls: O

xida

tive

stre

ss, i

mpa

ired

di

ffer

entia

tion

into

the

Ach

phe

noty

pe,

augm

ente

d di

ffer

entia

tion

into

the

DA

ph

enot

ype

(10

nm o

nly)

(Pow

ers

et a

l., 2

011)

Zer

o-va

lent

Fe

bare

, ox

idiz

ed, t

wo-

surf

ace

mod

ified

and

mag

netit

e

< 1

000

to >

200

01–

20 µ

g/m

lC

ultu

red

rode

nt m

icro

glia

(B

V2)

an

d ne

uron

s (N

27)

BV

2 ce

lls: O

xida

tive

stre

ss, m

orph

olog

ical

evid

ence

of

mito

chon

dria

l sw

ellin

g an

d ap

opto

sis

N27

cel

ls: R

educ

ed A

TP

leve

ls, m

orph

olog

ical

ev

iden

ce o

f pe

rinu

clea

r flo

ccul

ar m

ater

ial,

cyto

plas

mic

gra

nula

rity

, and

par

ticle

dep

ositi

on.

(Phe

nrat

et a

l., 2

009)

Mn,

Ag,

and

Cu

Mn-

40, A

g-15

, Cu-

9010

 μg

/ml

PC12

cel

lsD

opam

ine

depl

etio

n (C

u an

d M

n), a

ltere

d ge

ne

expr

essi

on a

ssoc

iate

d w

ith th

e do

pam

iner

gic

syst

em

(Wan

g et

 al.,

200

9)

Poly

styr

ene

unm

odifi

ed,

carb

oxyl

ated

, and

am

ino-

mod

ified

po

lyst

yren

e pa

rtic

les.

606,

25,

or

100 

μg/

ml

depe

ndin

g on

cel

l ty

pe a

nd a

ssay

Mac

roph

age

(RA

W 2

64.7

) an

d ep

ithel

ial (

BE

AS-

2B)

cells

, hum

an m

icro

vasc

ular

en

doth

elia

l (H

ME

C),

hep

atom

a (H

EPA

-1),

and

PC

12 c

ells

EN

M u

ptak

e, in

trac

ellu

lar

loca

lizat

ion,

and

to

xici

ty(X

ia e

t al.,

200

8)

Not

e. A

g, s

ilver

; Ach

, Ace

tych

olin

e; C

u, c

oppe

r; M

n, M

anga

nese

; ppm

, par

ts p

er m

illio

n.

DNT ENM RISK ASSESSMENT RESEARCH 235

found that nano-Mn and nano-Cu may contribute to patho-logical processes related to altered DA metabolism, impaired proteasomal activity, and, in the case of nano-Cu, disrupted redox balance, which leads to dopaminergic neurotoxicity. Powers et al. (2011) observed that 10 nm polyvinylpyrrolidone (PVP)–coated nano-Ag augmented differentiation of cells into the dopaminergic phenotype. In contrast, all three types of Ag ENM tested (10 and 50 nm PVP-Ag ENM and 10 nm citrate-Ag ENM) impaired differentiation into the acetylcholine phe-notype. Together, these studies suggest that ENM of different compositions can perturb dopamine production and/or metabo-lism and alter neurotransmitter phenotypes in this cell line.

As suggested by the findings of Phenrat et al. (2009), differ-ences in cell type may further influence ENM mechanisms in the developing nervous system, in addition to the influence of ENM characteristics. Xia et al. (2008) demonstrated that ENM uptake, intracellular localization, and toxicity depended on cell type in a study comparing effects of cationic polystyrene (PS) ENM (6, 25, or l00 μg/ml depending on cell type and assay) in cell lines representing five different cell types. In this study, PS ENM did not affect PC12 cell viability, unlike nano-Ag (Powers et al., 2011), but PS ENM did significantly increase cytotoxicity in both macrophage and epithelial cell lines (Xia et al., 2008). Effects in the later two cell lines were tied to mitochondrial dam-age and ATP depletion, similar to results by Phenrat et al. (2009) that showed decreased ATP levels in immortalized rat dopamin-ergic cells after exposure to Fe ENM. Together, data point to nuanced effects in different cell types that will be informed by pairing mechanistic, in vitro studies with in vivo studies evaluat-ing effects in the multicellular matrix of the developing brain.

In vitro studies of EMN toxicity in other culture systems could provide additional mechanistic information that may have particular relevance to in vivo human exposures. For example, induced pluripotent stem cells (iPSCs) could be used as a poten-tial platform to study the effects of ENM on either undifferenti-ated or differentiated neural cells, and because iPSCs can be derived from human tissues, these cultures may provide relevant information on the potential for early or late DNT effects in humans (reviewed in Kumar et al., 2012). Similarly, zebrafish cultures (reviewed in de Esch et al., 2012) could be used as a platform to evaluate neural cell responses to ENM within intact organisms in a more “high throughput” fashion.

DoSE-RESPoNSE INfoRMATIoN

Although the early evidence discussed above suggests that the developing brain may be exposed to ENM and that ENM exposures can cause DNT effects, there is little information on the dose-response relationship between ENM and neurodevel-opmental effects. Of the identified in vivo studies, only Jackson et al. (2011) used multiple doses to evaluate effects of mater-nal exposure to ENM; however, no dose-response relation-ship was observed because offspring neurobehavioral function was only tested at the highest concentration (268 μg Printex

90/animal). Notably, this concentration also caused maternal inflammation, which confounds findings related to neurodevel-opmental effects because maternal inflammation has also been shown to influence neurodevelopment (Golan et al., 2005 and reviewed in Hagberg and Mallard, 2005). As discussed above, differences in ENM coating, size, and core composition can result in differences in effects; thus, these parameters need to be accounted for in future dose-response studies. As discussed further below, additional dose-response information will help to fill important gaps in current understanding of the potential for DNT from ENM exposures.

DIScuSSIoN

As the use of ENM in a variety of applications, including cosmetics, clothing, and children’s toys, continues to rise, the likelihood of human exposures may increase as well. In turn, an understanding of potential human health risks would inform decisions about possible trade-offs between the economic and societal value of using ENM applications and impacts on human health. Evidence suggests that ENM evoke biologi-cal changes in the brain of adult animal models (Hu and Gao, 2010; Oberdörster et  al., 2009; Sharma and Sharma, 2010; Simkó and Mattsson, 2010; Yang et  al., 2010). As discussed above, the greater vulnerability of the developing brain to toxic insult combined with unique characteristics of ENM suggests that DNT may be a potential outcome of ENM exposure. We discussed the potential for ENM exposure in the developing brain, highlighted recent findings on ENM DNT endpoints, and discussed the need for dose-response information.

Although the identified hazard identification studies pro-vide initial evidence of effects on the developing brain after early life exposures, there are several limitations to their abil-ity to inform future evaluations of ENM DNT. In general, authors did not provide detailed ENM characterization data from their own analyses but rather relied on manufacturer sup-plied specifications. Such independent evaluations are critical for understanding the relationship between specific types of ENM and observed effects (Oberdörster et al., 2005). In some studies, maternal effects were observed that may obscure any direct effects of ENM on neurodevelopment. Without exposure assessments, it is challenging to determine whether neurobe-havioral or other endpoints reported are due to direct ENM effects or result from indirect effects (e.g., maternal inflam-mation). We would recommend minimally invasive techniques such as measurement of maternal and infant blood or urine samples in humans. With regards to animal studies, it would be ideal to measure maternal and fetal blood and brain ENM levels. Additionally, most studies excluded exposure during lactation, and this early postnatal period includes a number of developmental milestones for the brain (Rice and Barone, 2000). The lack of functional characterization of the devel-oping nervous system in most studies limits the ability to tie changes in gene expression or cell viability to DNT endpoints.

236 POWERS ET AL.

Almost all of the studies exclusively evaluated males on the premise that a number of neurobehavioral disorders are seen to a greater extent in males; however, evaluating both males and females, as recommended by current guidelines (OECD, 2007; U.S. EPA, 1998), would be informative for understanding any sex differences in susceptibility. Further, the majority of studies investigated effects of nano-TiO

2, which is only one of many

types of available ENM that would be informative to evaluate. Finally, most available studies did not include comparisons to materials that do not have nanoscale dimensions, which would provide insight on whether effects are due to specific character-istics of the nanosized material or to the material itself (Powers et al., 2011). Table 4 summarizes the research needs for ENM with respect to developmental neurotoxicity. In the following section we provide a strategy for addressing these data gaps for ENM exposure and developmental neurotoxicity.

Strategy for Planning ENM Research Protocols for DNT Testing

Exposure Research Needs

Although measurements of actual child exposure via mul-tiple routes are necessary to inform future risk assessment efforts, the time and resources required to gather such data, coupled with the rapid rise in ENM use, demand another, paral-lel approach to study potential ENM DNT. As indicated above, the EPA and OECD DNT Study Guidelines present a robust battery of tests to query potential effects during the most rapid period of nervous system development (Makris et  al., 2009; OECD, 2007; U.S. EPA, 1998). Such an approach lends itself to identifying the most sensitive window(s) during which to evaluate dose-response relationships and exposure, which

are critical aspects of risk assessment (Selevan et  al., 2000). In these guidelines, exposure extends from GD 6 to PND 11 or PND 21 to ensure that offspring are exposed to maternal circulation in utero and maternal milk in the postnatal period (Makris and Raffaele, 2009). Although the temporal rate of development of individual nervous system structures generally differs substantially between humans and rodents, the overall patterns of progression are largely the same, and the tests in this screening battery have been previously evaluated for inter-species reliability (Makris and Raffaele, 2009; Makris et  al., 2009; Rice and Barone, 2000). In turn, such a well-established, screening-level approach should provide important evidence for moving forward with future risk assessment efforts focused on ENM and neurodevelopment.

For instance, data from DNT screening efforts that include toxicokinetic data could inform the development of physi-ologically based pharmacokinetic models that account for changes in adsorption, metabolism, and excretion that occur during pregnancy and early development (Selevan et al., 2000). Pairing such models with measurements of exposure in occupa-tional, residential, and other settings relevant to early life could then support an understanding of dosimetry in the developing nervous system. Importantly, future measurements and mod-eling efforts should include the variety of parameters that are likely to influence exposure levels, such as product formula-tion (e.g., liquid vs. cream), use environment (e.g., indoors vs. outdoors), and use duration (e.g., one application vs. multiple applications per day), in addition to ENM characteristics (e.g., particle size, surface area, and charge) (Hansen et al., 2008). Further, based on the potential for ENM exposure to continue well beyond the early postnatal period, the longer exposure period included in the OECD guidelines (i.e., to PND 21) or

TABLE 4key Research gaps for ENM DNT

Research area Research gap

Exposure 1. Identification of sensitive exposure windows in offspring using DNT guideline studies with extended exposure periods for offspring (i.e., PND 21 or beyond). Comparisons of sensitive exposure windows between different types of ENM could support the prioritization of in vitro and/or in vivo follow-up studies.

2. Development of PBPK models using toxicokinetic measurements of ENM in offspring after maternal dosing or direct dosing of ENM in offspring. Models would be developed with consideration of appropriate material characteristics (e.g., size and charge) and exposure parameters (e.g., route, frequency, and direct vs. indirect contact).

3. Measurements of exposure in occupational, residential, and other settings relevant to early life. The utility of measurements increases with the inclusion of parameters that are likely to influence exposure levels, such as product formulation (e.g., liquid vs. cream), use environment (e.g., indoors vs. outdoors), and use duration (e.g., one application vs. multiple applications per day), in addition to ENM characteristics (e.g., particle size, surface area, and charge). The development of tools and approaches to accurately and quickly evaluate ENM exposure in the environment parallels this research gap.

Hazard 1. Evaluation of neuropathology and behavior (see text and DNT guideline for discussion of endpoints) at multiple time points in developing offspring (i.e., PND 4, 11, 21, 35, 45, and 60) following maternal ENM exposures.

2. Evaluations of glia cell number and function in offspring during the postnatal period following maternal ENM exposure during prenatal and lactational periods.

3. Evaluation of visual function in offspring following maternal ENM exposure or direct dosing of offspring during the early postnatal period.Dose response 1. Evaluation of dose response using dose levels suggested in DNT guidelines with well characterized ENM (e.g., data on size, surface

charge, and shape).2. Development of analytical methods capable of characterizing multiple ENM characteristics (e.g., size, surface charge, and composition).

DNT ENM RISK ASSESSMENT RESEARCH 237

perhaps even longer (to the age of puberty or adulthood) may be useful. Additionally, although direct dosing of preweanling offspring is not stipulated in either EPA or OECD guidelines, given observations of maternal inflammation by Hougaard et al. (2010) and Jackson et al. (2011), it may be informative in ENM studies to distinguish between direct and indirect ENM effects. Alternatively, toxicokinetic measurements of ENM in offspring could be used in lieu of direct dosing to establish whether the material reaches offspring and distinguish direct from indirect effects. Both direct dosing of preweanling offspring and mater-nal dosing studies would be informative for developing a better understanding of whether, and the extent to which, different types of ENM cross the blood-brain and placental barriers.

Notably, as the development of ENM and ENM applications continues to place new products on the market, efforts to develop the tools and techniques to evaluate human exposures to ENM in the environment will likely continue to be a critical research area.

Hazard Identification Research Needs

The observations included in the DNT screening guidelines provide a starting point for building upon current knowledge about neurological endpoints sensitive to ENM exposure. For instance, existing work evaluated animals for gross neurologi-cal and behavioral changes at a limited number of time points after developmental ENM exposure (Gao et al., 2011; Hougaard et al., 2010; Takeda et al., 2009); thus, follow-up studies that include multiple time points throughout development (i.e., PND 4, 11, 21, 35, 45, and 60) could be informative (U.S. EPA, 1998). Only one study noted changes in body weight after ENM exposure (Takeda et al., 2009), but recording such developmen-tal landmarks will be informative in future work (U.S. EPA, 1998). Similarly, only one of the identified studies evaluated motor activity changes, auditory startle responses, and cogni-tive (learning and memory) effects (Hougaard et al., 2010). As such, follow-up work could provide additional measurements of motor activity at multiple time points for time periods of at least 10 min, coupled with measurements of auditory star-tle, prepulse inhibition, and learning and memory at the time of weaning and around PND 60 (U.S. EPA, 1998). Notably, observations of negative effects on synaptic transmission and plasticity in the hippocampus after exposure to TiO

2 ENM

by Gao et al. (2011) are similar to those noted by Tang et al. (2009b) in adult animals after ENM (cadmium selenium [Cd/Se] and CdSe/ZnS) were applied directly to the hippocampus, suggesting that this endpoint may be informative to include in future studies. Finally, neuropathological evaluations in several areas of the brain on PND 11 or PND 21 and at the end of study would supplement data from the study by Takeda et al. (2009), which indicated elevations in apoptotic markers in the olfac-tory bulb and cortex of 6-week-old offspring. Specific obser-vations to include in investigations of neuropathology can be found in the DNT guidelines (OECD, 2007; U.S. EPA, 1998). In addition to the endpoints outlined in available guidelines, evaluations related to glial cell function may be informative for

ENM because proliferation of this cell type continues well into the postnatal period (Rice and Barone, 2000), and current data indicate that ENM are taken up by microglia and astrocytes (Hutter et  al., 2010; Luther et  al., 2011; Tang et  al., 2010). Evidence further suggests that ENM can alter glial cell differ-entiation (Shimizu et al., 2009) and activation (Sharma et al., 2009). Finally, recent in vitro data show phototoxicity of cer-tain ENM to ocular tissues (Roberts et al., 2008; Sanders et al., 2012; Wielgus et al., 2010). Photoactive compounds, such as nano-TiO

2, are excited by ultraviolet A radiation, which could

potentially reach the retina of infants or young children to a greater extent than in adults because the lens is still developing (Boyes et al., 2012). In turn, evaluating effects on visual func-tion might be particularly relevant for future hazard identifica-tion efforts for ENM DNT testing.

Dose-Response Research Needs

As noted above, only one of the identified in vivo studies included multiple dose levels. Future work should build upon current findings by using at least three dose levels in addition to control groups. Specifics on selecting dose levels can be found in the DNT guidelines, but generally, the highest dose should elicit observable maternal effects, whereas the lowest dose should not result in any observable neurological changes in dams or offspring (OECD, 2007; U.S. EPA, 1998). Intermediate doses should be equally spaced between the highest and low-est selected doses. Notably, quantitative dose-response data for ENM will need to take into account not only the concen-tration of the material but also specific ENM characteristics (e.g., surface charge, shape, size, and composition) that may cause different effects. Because a single ENM batch generally contains a mixture of individual particles with different char-acteristics (e.g., TiO

2 ENM containing 10, 50, and 100 nm par-

ticles), it is necessary to explicitly evaluate these parameters. Measuring and characterizing ENM in biological media are challenging and require a multitude of techniques and instru-ments (Johnston et al., 2010; Lehman et al., 2011). Thus, the development of analytical methods capable of characterizing multiple ENM characteristics such as those noted above would support the development of detailed dose-response studies, which would inform future assessment efforts.

In addition to applying the DNT guidelines in evaluat-ing DNT effects associated with ENM, information on dose response could also be used from adult neurotoxicological evaluations. For example, Zhang et  al. (2013) treated adult male rats with 5, 10, and 45 mg/kg-day of Ag ENM for 3 days. Significant decreases in rearing behavior were only observed at the highest dose (45 mg/kg-day). Therefore, these dose levels could also be used as a guide for setting dosing regimens in a DNT study for Ag ENM.

Key points for ENM DNT evaluations. It is unlikely that the research outlined above using DNT screening guidelines can be carried out for every available ENM type. However,

238 POWERS ET AL.

initial work to compare ENM types using established guide-lines could guide the design of higher throughput studies to show the relationship(s) between in vivo findings and in vitro or alternative model studies. Although high-throughput screen-ing approaches are beginning to be applied to predict ENM toxicity in general (i.e., for endpoints other than DNT) (Nel et al., 2012), the complexity of the developing nervous system combined with current limitations in understanding related to ENM-specific issues in vitro (e.g., dosimetry [Teeguarden et  al., 2006], assay compatibility [Monteiro-Riviere et  al., 2009], standards and methods for material characterization [Bouwmeester et  al., 2011]) suggests that high-throughput methods for ENM DNT will not likely be available in the near term. Given the rapid rate of new ENM applications, produc-ing data to inform decisions in the near term on potential ENM DNT may be best achieved through the use of established methods, while efforts to refine high-throughput approaches to inform risk assessments continue. The use of traditional DNT guidelines to plan and conduct research in the interim could not only support more near-term decision making but also help inform the development of a tiered higher through-put screening method(s) for longer term decisions about ENM DNT. This approach has been discussed in a recent report by the National Research Council: A  Research Strategy for Environmental, Health, and Safety Aspects of Engineered Nanomaterial (NRC, 2012). Importantly, data from ENM DNT guideline studies should be reported in a standardized format to increase the utility of the information for both near-term decision making and the development of higher throughput screening method(s) (see Thomas et al., 2013 for an example data-sharing approach).

Ultimately, decisions regarding the health and safety of ENM-enabled products and applications will require an exami-nation of the trade-offs between economic and societal benefits of using the product versus potential health risks. Such trade-offs will differ between individual types of ENM and each prod-uct or application they are proposed to be used in. For some, the use of ENM may decrease the use of materials with relatively well-understood health and environmental effects (e.g., carbon nanotubes replacing polybrominated flame retardants). Yet, other types of ENM may be proposed to replace chemicals for which health effects are less well understood (e.g., nano-Ag vs. conventional disinfectant products). In each instance, deci-sions will need to be made as to whether the ENM-enabled product is in fact favorable. The objective here is not to argue for or against the use of ENM but rather to encourage efforts to gather sufficient data to support knowledgeable decisions about these products. In doing so, we have the opportunity to avoid past experiences with overlooking the potential for DNT from chemicals in mass production, which has cost billions of dollars in economic terms alone (Grandjean and Landrigan, 2006). The evidence for ENM DNT to date is not strong enough to reach any conclusions about potential risk; however, it indicates that a closer look is warranted.

REfERENcES

Abdel-Rahman, A., Dechkovskaia, A. M., Sutton, J. M., Chen, W. C., Guan, X., Khan, W. A., and Abou-Donia, M. B. (2005). Maternal exposure of rats to nicotine via infusion during gestation produces neurobehavioral deficits and elevated expression of glial fibrillary acidic protein in the cerebellum and CA1 subfield in the offspring at puberty. Toxicology 209, 245–261.

Aschberger, K., Micheletti, C., Sokull-Klüttgen, B., and Christensen, F. M. (2011). Analysis of currently available data for characterising the risk of engineered nanomaterials to the environment and human health–lessons learned from four case studies. Environ. Int. 37, 1143–1156.

Benn, T. M., and Westerhoff, P. (2008). Nanoparticle silver released into water from commercially available sock fabrics. Environ. Sci. Technol. 42, 4133–4139.

Besson, S., Latzin, P., Stern, G., Strippoli, M. P. F., Kuehni, C. E., Roosli, M., and Frey, U. (2010). Influence of atmospheric parameters and pollutants on control of breathing in healthy neonates. Ann. Respir. Med. 1.

Blomgren, K., and Hagberg, H. (2006). Free radicals, mitochondria, and hypoxia-ischemia in the developing brain. Free Radic. Biol. Med. 40, 388–397.

Bondy, S. C., and Campbell, A. (2005). Developmental neurotoxicology. J. Neurosci. Res. 81, 605–612.

Bouwmeester, H., Lynch, I., Marvin, H. J., Dawson, K. A., Berges, M., Braguer, D., Byrne, H. J., Casey, A., Chambers, G., Clift, M. J., et al. (2011). Minimal analytical characterization of engineered nanomaterials needed for hazard assessment in biological matrices. Nanotoxicology 5, 1–11.

Boyes, W. K., Chen, R., Chen, C., and Yokel, R. A. (2012). The neurotoxic potential of engineered nanomaterials. Neurotoxicology 33, 902–910.

Burton, N. C., and Guilarte, T. R. (2009). Manganese neurotoxicity: Lessons learned from longitudinal studies in nonhuman primates. Environ. Health Perspect. 117, 325–332.

Buzea, C., Pacheco, I. I., and Robbie, K. (2007). Nanomaterials and nanoparti-cles: Sources and toxicity. Biointerphases 2, MR17–MR71.

Chen, L., Yokel, R. A., Hennig, B., and Toborek, M. (2008). Manufactured alu-minum oxide nanoparticles decrease expression of tight junction proteins in brain vasculature. J. Neuroimmune Pharmacol. 3, 286–295.

Chithrani, B. D., Ghazani, A. A., and Chan, W. C. (2006). Determining the size and shape dependence of gold nanoparticle uptake into mammalian cells. Nano Lett. 6, 662–668.

Cho, M., Cho, W. S., Choi, M., Kim, S. J., Han, B. S., Kim, S. H., Kim, H. O., Sheen, Y. Y., and Jeong, J. (2009). The impact of size on tissue distribu-tion and elimination by single intravenous injection of silica nanoparticles. Toxicol. Lett. 189, 177–183.

Choi, J. Y., Ramachandran, G., and Kandlikar, M. (2009). The impact of tox-icity testing costs on nanomaterial regulation. Environ. Sci. Technol. 43, 3030–3034.

Costa, L. G., Aschner, M., Vitalone, A., Syversen, T., and Soldin, O. P. (2004a). Developmental neuropathology of environmental agents. Annu. Rev. Pharmacol. Toxicol. 44, 87–110.

Costa, L. G., Yagle, K., Vitalone, A., and Guizzetti, M. (2004b). Alcohol and Glia in the Developing Brain. CRC Press, Boca Riton, FL.

Dahm, M. M., Yencken, M. S., and Schubauer-Berigan, M. K. (2011). Exposure control strategies in the carbonaceous nanomaterial industry. J. Occup. Environ. Med. 53(6 Suppl), S68–S73.

de Esch, C., Slieker, R., Wolterbeek, A., Woutersen, R., and de Groot, D. (2012). Zebrafish as potential model for developmental neurotoxicity test-ing: A mini review. Neurotoxicol. Teratol. 34, 545–553.

De Jong, W. H., Hagens, W. I., Krystek, P., Burger, M. C., Sips, A. J., and Geertsma, R. E. (2008). Particle size-dependent organ distribution of gold nanoparticles after intravenous administration. Biomaterials 29, 1912–1919.

DNT ENM RISK ASSESSMENT RESEARCH 239

Elder, A., Gelein, R., Silva, V., Feikert, T., Opanashuk, L., Carter, J., Potter, R., Maynard, A., Ito, Y., Finkelstein, J., et al. (2006). Translocation of inhaled ultrafine manganese oxide particles to the central nervous system. Environ. Health Perspect. 114, 1172–1178.

Fadri, G., Sonderer, T., Scholz, R. W., and Nowack, B. (2009). Modeled envi-ronmental concentrations of engineered nanomaterials (TiO2, ZnO, Ag, CNT, fullerenes) for different regions. Environ. Sci. Technol. 43, 9216–9222.

Gangwal, S., Brown, J. S., Wang, A., Houck, K. A., Dix, D. J., Kavlock, R. J., and Hubal, E. A. (2011). Informing selection of nanomaterial concentra-tions for ToxCast in vitro testing based on occupational exposure potential. Environ. Health Perspect. 119, 1539–1546.

Gao, X., Yin, S., Tang, M., Chen, J., Yang, Z., Zhang, W., Chen, L., Yang, B., Li, Z., Zha, Y., et al. (2011). Effects of developmental exposure to TiO2 nanoparticles on synaptic plasticity in hippocampal dentate gyrus area: An in vivo study in anesthetized rats. Biol. Trace Elem. Res. 143, 1616–1628.

Golan, H. M., Lev, V., Hallak, M., Sorokin, Y., and Huleihel, M. (2005). Specific neurodevelopmental damage in mice offspring following maternal inflammation during pregnancy. Neuropharmacology 48, 903–917.

Gottschalk F., Sonderer, T., Scholz, R. W., and Nowack, B. (2009). Modeled Environmental Concentrations of Engineered Nanomaterials (TiO2, ZnO, Ag, CNT, Fullerenes) for Different Regions. Environ Sci Technol. 43, 9216–9222.

Grandjean, P., and Landrigan, P. J. (2006). Developmental neurotoxicity of industrial chemicals. Lancet 368, 2167–2178.

Grieneisen, M. L., and Zhang, M. (2011). Nanoscience and nanotechnology: Evolving definitions and growing footprint on the scientific landscape. Small 7, 2836–2839.

Guizzetti, M., Catlin, M., and Costa, L. G. (1997). The effects of ethanol on glial cell proliferation: Relevance to the fetal alcohol syndrome. Front. Biosci. 2, e93–e98.

Gulson, B., McCall, M., Korsch, M., Gomez, L., Casey, P., Oytam, Y., Taylor, A., McCulloch, M., Trotter, J., Kinsley, L., et al. (2010). Small amounts of zinc from zinc oxide particles in sunscreens applied outdoors are absorbed through human skin. Toxicol. Sci. 118, 140–149.

Hagberg, H., and Mallard, C. (2005). Effect of inflammation on central nerv-ous system development and vulnerability. Curr. Opin. Neurol. 18, 117–123.

Hansen, S. F., Michelson, E. S., Kamper, A., Borling, P., Stuer-Lauridsen, F., and Baun, A. (2008). Categorization framework to aid exposure assessment of nanomaterials in consumer products. Ecotoxicology 17, 438–447.

Hardas, S. S., Butterfield, D. A., Sultana, R., Tseng, M. T., Dan, M., Florence, R. L., Unrine, J. M., Graham, U. M., Wu, P., Grulke, E. A., et al. (2010). Brain distribution and toxicological evaluation of a systemically delivered engineered nanoscale ceria. Toxicol. Sci. 116, 562–576.

Hendren, C. O., Mesnard, X., Dröge, J., and Wiesner, M. R. (2011). Estimating production data for five engineered nanomaterials as a basis for exposure assessment. Environ. Sci. Technol. 45, 2562–2569.

Hougaard, K. S., Jackson, P., Jensen, K. A., Sloth, J. J., Löschner, K., Larsen, E. H., Birkedal, R. K., Vibenholt, A., Boisen, A. M., Wallin, H., et al. (2010). Effects of prenatal exposure to surface-coated nanosized titanium dioxide (UV-Titan). A study in mice. Part. Fibre Toxicol. 7, 16.

Hu, Y. L., and Gao, J. Q. (2010). Potential neurotoxicity of nanoparticles. Int. J. Pharm. 394, 115–121.

Hutter, E., Boridy, S., Labrecque, S., Lalancette-Hébert, M., Kriz, J., Winnik, F. M., and Maysinger, D. (2010). Microglial response to gold nanoparticles. ACS Nano 4, 2595–2606.

Innovative Research and Products Incorporated (2011). Production and applica-tions of carbon nanotubes, carbon nanofibers, fullerenes, graphene and nano-diamonds: A global technology survey and market analysis. Technical report et-113, 531.

Jackson, P., Vogel, U., Wallin, H., and Hougaard, K. S. (2011). Prenatal expo-sure to carbon black (printex 90): Effects on sexual development and neuro-function. Basic Clin. Pharmacol. Toxicol. 109, 434–437.

Johnston, H. J., Hutchison, G., Christensen, F. M., Peters, S., Hankin, S., and Stone, V. (2010). A review of the in vivo and in vitro toxicity of silver and gold particulates: Particle attributes and biological mechanisms responsible for the observed toxicity. Crit. Rev. Toxicol. 40, 328–346.

Kreuter, J. (2004). Influence of the surface properties on nanoparticle-mediated transport of drugs to the brain. J. Nanosci. Nanotechnol. 4, 484–488.

Kumar, K. K., Aboud, A. A., and Bowman, A. B. (2012). The potential of induced pluripotent stem cells as a translational model for neurotoxicologi-cal risk. Neurotoxicology 33, 518–529.

Lansdown, A. B. (2007). Critical observations on the neurotoxicity of silver. Crit. Rev. Toxicol. 37, 237–250.

Lehman, J. H., Terrones, M., Mansfield, E., Hurst, K. E., and Meunier, V. (2011). Evaluating the characteristics of multiwall carbon nanotubes. Carbon 49, 2581–2602.

Lockman, P. R., Mumper, R. J., Khan, M. A., and Allen, D. D. (2002). Nanoparticle technology for drug delivery across the blood-brain barrier. Drug Dev. Ind. Pharm. 28, 1–13.

Luther, E. M., Koehler, Y., Diendorf, J., Epple, M., and Dringen, R. (2011). Accumulation of silver nanoparticles by cultured primary brain astrocytes. Nanotechnology 22, 375101.

Ma, L., Liu, J., Li, N., Wang, J., Duan, Y., Yan, J., Liu, H., Wang, H., and Hong, F. (2010). Oxidative stress in the brain of mice caused by translocated nano-particulate TiO2 delivered to the abdominal cavity. Biomaterials 31, 99–105.

Makris, S. L., and Raffaele, K. C. (2009). Developmental Neurotoxicity. John Wiley & Sons, New York, NY.

Makris, S. L., Raffaele, K., Allen, S., Bowers, W. J., Hass, U., Alleva, E., Calamandrei, G., Sheets, L., Amcoff, P., Delrue, N., et al. (2009). A retro-spective performance assessment of the developmental neurotoxicity study in support of OECD test guideline 426. Environ. Health Perspect. 117, 17–25.

Maynard, A. D. (2011). Don’t define nanomaterials. Nature 475, 31.

Maynard, A. D., Warheit, D. B., and Philbert, M. A. (2011). The new toxicol-ogy of sophisticated materials: Nanotoxicology and beyond. Toxicol. Sci. 120(Suppl. 1), S109–S129.

Maysinger, D., Behrendt, M., Lalancette-Hébert, M., and Kriz, J. (2007). Real-time imaging of astrocyte response to quantum dots: In vivo screening model system for biocompatibility of nanoparticles. Nano Lett. 7, 2513–2520.

Méndez-Armenta, M., and Ríos, C. (2007). Cadmium neurotoxicity. Environ. Toxicol. Pharmacol. 23, 350–358.

Monteiro-Riviere, N. A., Inman, A. O., and Zhang, L. W. (2009). Limitations and relative utility of screening assays to assess engineered nanoparticle tox-icity in a human cell line. Toxicol. Appl. Pharmacol. 234, 222–235.

Mühlfield, C., Rothen-Rutishause, B., Blank, F., Vanhecke, D., Ochs, M., and Gehr, P. (2008). Interactions of nanoparticles with pulmonary structures and cellular response. Am. J. Physiol. Lung Cell Mol. Physiol. 294, L817–L829.

Nel, A., Xia, T., Meng, H., Wang, X., Lin, S., Ji, Z., and Zhang, H. (2012). Nanomaterial toxicity testing in the 21st century: Use of a predictive toxi-cological approach and high-throughput screening. Acc. Chem Res. 46, 607–621.

NRC (1983). Risk Assessment in the Federal Government: Managing the Process. National Academies Press, Washington, DC.

NRC (1994). Science and Judgment in Risk Assessment. National Academy Press, Washington, DC.

NRC (2009). Science and Decisions: Advancing Risk Assessment. National Academies Press, Washington, DC.

NRC (2012). A Research Strategy for Environmental, Health, and Safety Aspects of Engineered Nanomaterials. National Academies Press, Washington, DC.

NSTC (2011). National Nanotechnology Initiative: Environmental, Health, and Safety Research Strategy. National Nanotechnology Initiative, Arlington, VA.

Oberdörster, G., Elder, A., and Rinderknecht, A. (2009). Nanoparticles and the brain: Cause for concern? J. Nanosci. Nanotechnol. 9, 4996–5007.

240 POWERS ET AL.

Oberdörster, G., Oberdörster, E., and Oberdörster, J. (2005). Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 113, 823–839.

OECD (2007). OECD Guideline for the Testing of Chemicals Developmental Neurotoxicity Study. Organization for Economic Co-operation and Development, Paris, France.

Oszlánczi, G., Vezér, T., Sárközi, L., Horváth, E., Kónya, Z., and Papp, A. (2010). Functional neurotoxicity of Mn-containing nanoparticles in rats. Ecotoxicol. Environ. Saf. 73, 2004–2009.

Oszlánczi, G., Vezér, T., Sárközi, L., Horváth, E., Szabó, A., Horváth, E., Kónya, Z., and Papp, A. (2010). Metal deposition and functional neurotoxic-ity in rats after 3-6 weeks nasal exposure by two physicochemical forms of manganese. Environ. Toxicol. Pharmacol. 30, 121–126.

Phenrat, T., Long, T. C., Lowry, G. V., and Veronesi, B. (2009). Partial oxida-tion (“aging”) and surface modification decrease the toxicity of nanosized zerovalent iron. Environ. Sci. Technol. 43, 195–200.

Popov, A., Zhao, X., Zvyagin, A., Lademann, J., Roberts, M., Sanchez, W., Priezzhev, A., and Myllylä, R. (2010). ZnO and TiO2 parti-cles: A  study on nanosafety and photoprotection. Proc. SPIE 7715, 77153G77151–77153G77157.

Powers, C. M., Badireddy, A. R., Ryde, I. T., Seidler, F. J., and Slotkin, T. A. (2011). Silver nanoparticles compromise neurodevelopment in PC12 cells: Critical contributions of silver ion, particle size, coating, and composition. Environ. Health Perspect. 119, 37–44.

Rice, D., and Barone, S., Jr. (2000). Critical periods of vulnerability for the developing nervous system: Evidence from humans and animal models. Environ. Health Perspect. 108(Suppl. 3), 511–533.

Roberts, J. E., Wielgus, A. R., Boyes, W. K., Andley, U., and Chignell, C. F. (2008). Phototoxicity and cytotoxicity of fullerol in human lens epithelial cells. Toxicol. Appl. Pharmacol. 228, 49–58.

Sanders, K., Degn, L. L., Mundy, W. R., Zucker, R. M., Dreher, K., Zhao, B., Roberts, J. E., and Boyes, W. K. (2012). In vitro phototoxicity and hazard identification of nano-scale titanium dioxide. Toxicol. Appl. Pharmacol. 258, 226–236.

Saunders, M. (2009). Transplacental transport of nanomaterials. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 1, 671–684.

Savolainen, K., Alenius, H., Norppa, H., Pylkkänen, L., Tuomi, T., and Kasper, G. (2010). Risk assessment of engineered nanomaterials and nanotechnolo-gies–a review. Toxicology 269, 92–104.

Schwartz, P. M., Jacobson, S. W., Fein, G., Jacobson, J. L., and Price, H. A. (1983). Lake Michigan fish consumption as a source of polychlorinated biphenyls in human cord serum, maternal serum, and milk. Am. J. Public Health 73, 293–296.

Selevan, S. G., Kimmel, C. A., and Mendola, P. (2000). Identifying criti-cal windows of exposure for children’s health. Environ. Health Perspect. 108(Suppl. 3), 451–455.

Shanker, G., Syversen, T., and Aschner, M. (2003). Astrocyte-mediated meth-ylmercury neurotoxicity. Biol. Trace Elem. Res. 95, 1–10.

Sharma, H. S., Ali, S. F., Hussain, S. M., Schlager, J. J., and Sharma, A. (2009). Influence of engineered nanoparticles from metals on the blood-brain bar-rier permeability, cerebral blood flow, brain edema and neurotoxicity. An experimental study in the rat and mice using biochemical and morphological approaches. J. Nanosci. Nanotechnol. 9, 5055–5072.

Sharma, H. S., and Sharma, A. (2007). Nanoparticles Aggravate Heat Stress Induced Cognitive Deficits, Blood-Brain Barrier Disruption, Edema Formation and Brain Pathology, Vol. 162. Elsevier B.V., New York, NY.

Sharma, H. S., and Sharma, A. (2010). Conference scene: Nanoneuroprotection and nanoneurotoxicity: Recent progress and future perspectives. Nanomedicine (Lond). 5, 533–537.

Sherwood, L. (2013). Human Physiology from Cells to Systems, 477. Brooks/Cole, Cengage Learning, Belmont, CA.

Shimizu, M., Tainaka, H., Oba, T., Mizuo, K., Umezawa, M., and Takeda, K. (2009). Maternal exposure to nanoparticulate titanium dioxide during the prenatal period alters gene expression related to brain development in the mouse. Part. Fibre Toxicol. 6, 20.

Simkó, M., and Mattsson, M. O. (2010). Risks from accidental exposures to engineered nanoparticles and neurological health effects: A critical review. Part. Fibre Toxicol. 7, 42.

Stone, V., Johnston, H., and Clift, M. J. (2007). Air pollution, ultrafine and nanoparticle toxicology: Cellular and molecular interactions. IEEE Trans. Nanobioscience 6, 331–340.

Takahashi, Y., Mizuo, K., Shinkai, Y., Oshio, S., and Takeda, K. (2010). Prenatal exposure to titanium dioxide nanoparticles increases dopamine levels in the prefrontal cortex and neostriatum of mice. J. Toxicol. Sci. 35, 749–756.

Takeda, K., Suzuki, K. I., Ishihara, A., Kubo-Irie, M., Fujimoto, R., Tabata, M., Oshio, S., Nihei, Y., Ihara, T., and Sugamata, M. (2009). Nanoparticles transferred from pregnant mice to their offspring can damage the genital and cranial nerve systems. J. Health Sci. 55, 95–102.

Tanaka, J., Toku, K., Zhang, B., Ishihara, K., Sakanaka, M., and Maeda, N. (1999). Astrocytes prevent neuronal death induced by reactive oxygen and nitrogen species. Glia 28, 85–96.

Tang, J., Xiong, L., Wang, S., Wang, J., Liu, L., Li, J., Yuan, F., and Xi, T. (2009a). Distribution, translocation and accumulation of silver nanoparticles in rats. J. Nanosci. Nanotechnol. 9, 4924–4932.

Tang, J., Xiong, L., Zhou, G., Wang, S., Wang, J., Liu, L., Li, J., Yuan, F., Lu, S., Wan, Z., et al. (2010). Silver nanoparticles crossing through and distribution in the blood-brain barrier in vitro. J. Nanosci. Nanotechnol. 10, 6313–6317.

Tang, M., Li, Z., Chen, L., Xing, T., Hu, Y., Yang, B., Ruan, D. Y., Sun, F., and Wang, M. (2009b). The effect of quantum dots on synaptic transmission and plasticity in the hippocampal dentate gyrus area of anesthetized rats. Biomaterials 30, 4948–4955.

Teeguarden, J. G., Hinderliter, P. M., Orr, G., Thrall, B. D., and Pounds, J. G. (2006). Particokinetics in vitro: Dosimetry considerations for in vitro nano-particle toxicity assessments. Toxicol. Sci. 95, 300–312.

Thomas, D. G., Gaheen, S., Harper, S. L., Fritts, M., Klaessig, F., Hahn-Dantona, E., Paik, D., Pan, S., Stafford, G. A., Freund, E. T., et al. (2013). ISA-TAB-Nano: A specification for sharing nanomaterial research data in spreadsheet-based format. BMC Biotechnol. 13, 2.

Trickler, W. J., Lantz, S. M., Murdock, R. C., Schrand, A. M., Robinson, B. L., Newport, G. D., Schlager, J. J., Oldenburg, S. J., Paule, M. G., Slikker, W., Jr, et al. (2010). Silver nanoparticle induced blood-brain barrier inflamma-tion and increased permeability in primary rat brain microvessel endothelial cells. Toxicol. Sci. 118, 160–170.

Trickler, W. J., Lantz, S. M., Murdock, R. C., Schrand, A. M., Robinson, B. L., Newport, G. D., Schlager, J. J., Oldenburg, S. J., Paule, M. G., Slikker, W., Jr, et al. (2011). Brain microvessel endothelial cells responses to gold nanopar-ticles: In vitro pro-inflammatory mediators and permeability. Nanotoxicology 5, 479–492.

U.S. EPA (1998). Health Effects Test Guidelines OPPTS 870.6300 Developmental Neurotoxicity Study. U.S. Environmentall Protection Agency, Washington, DC.

U.S. EPA (2011). Exposure Factors Handbook: 2011 Edition (Final). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-09/052F.

Wang, J., Chen, C., Liu, Y., Jiao, F., Li, W., Lao, F., Li, Y., Li, B., Ge, C., Zhou, G., et al. (2008a). Potential neurological lesion after nasal instillation of TiO(2) nanoparticles in the anatase and rutile crystal phases. Toxicol. Lett. 183, 72–80.

Wang, J., Liu, Y., Jiao, F., Lao, F., Li, W., Gu, Y., Li, Y., Ge, C., Zhou, G., Li, B., et al. (2008b). Time-dependent translocation and potential impairment on central nervous system by intranasally instilled TiO(2) nanoparticles. Toxicology 254, 82–90.

Wang, J., Rahman, M. F., Duhart, H. M., Newport, G. D., Patterson, T. A., Murdock, R. C., Hussain, S. M., Schlager, J. J., and Ali, S. F. (2009). Expression

DNT ENM RISK ASSESSMENT RESEARCH 241

changes of dopaminergic system-related genes in PC12 cells induced by man-ganese, silver, or copper nanoparticles. Neurotoxicology 30, 926–933.

Watson, R. E., Desesso, J. M., Hurtt, M. E., and Cappon, G. D. (2006). Postnatal growth and morphological development of the brain: A species comparison. Birth Defects Res. B. Dev. Reprod. Toxicol. 77, 471–484.

Wick, P., Malek, A., Manser, P., Meili, D., Maeder-Althaus, X., Diener, L., Diener, P. A., Zisch, A., Krug, H. F., and von Mandach, U. (2010). Barrier capacity of human placenta for nanosized materials. Environ. Health Perspect. 118, 432–436.

Wielgus, A. R., Zhao, B., Chignell, C. F., Hu, D. N., and Roberts, J. E. (2010). Phototoxicity and cytotoxicity of fullerol in human retinal pigment epithelial cells. Toxicol. Appl. Pharmacol. 242, 79–90.

Wijnhoven, S. W. P., Peijnenburg, W. J. G. M., Herberts, C. A., Hagens, W. I., Oomen, A. G., Heugens, E. H.  W., Roszek, B., Bisschops, J., Gosens, I., van de Meent, D., et  al. (2009). Nano-silver: A  review of available data and knowledge gaps in human and environmental risk assessment. Nanotoxicology 3, 109–138.

Wrighton, S. A., and Stevens, J. C. (1992). The human hepatic cytochromes P450 involved in drug metabolism. Crit. Rev. Toxicol. 22, 1–21.

Xia, T., Kovochich, M., Liong, M., Zink, J. I., and Nel, A. E. (2008). Cationic polystyrene nanosphere toxicity depends on cell-specific endocytic and mitochondrial injury pathways. ACS Nano 2, 85–96.

Yamashita, K., Yoshioka, Y., Higashisaka, K., Mimura, K., Morishita, Y., Nozaki, M., Yoshida, T., Ogura, T., Nabeshi, H., Nagano, K., et al. (2011). Silica and titanium dioxide nanoparticles cause pregnancy complications in mice. Nat. Nanotech. 6, 321–328.

Yang, Z., Liu, Z. W., Allaker, R. P., Reip, P., Oxford, J., Ahmad, Z., and Ren, G. (2010). A review of nanoparticle functionality and toxicity on the central nervous system. J. R. Soc. Interface 7(Suppl. 4), S411–S422.

Yokel, R. A., and Macphail, R. C. (2011). Engineered nanomaterials: Exposures, hazards, and risk prevention. J. Occup. Med. Toxicol. 6, 7.

Zhang, Y., Ferguson, S. A., Watanabe, F., Jones, Y., Xu, Y., Biris, A. S., Hussain, S., Ali, S. F. (2013) Silver nanoparticles decrease body weight and locomotor activity in adult male rats. Small. 9, 1715–1720.

242 POWERS ET AL.