Detecting Neurodevelopmental Toxicity of Domoic Acid and Ochratoxin A Using Rat Fetal Neural Stem Cells

Currently, animal experiments in rodents are the gold standard for developmental neurotoxicity (DNT) investigations; however, testing guidelines for these experiments are insufficient in terms of animal use, time, and costs. Thus, alternative reliable approaches are needed for predicting DNT. We chose rat neural stem cells (rNSC) as a model system, and used a well-known neurotoxin, domoic acid (DA), as a model test chemical to validate the assay. This assay was used to investigate the potential neurotoxic effects of Ochratoxin A (OTA), of which the main target organ is the kidney. However, limited information is available regarding its neurotoxic effects. The effects of DA and OTA on the cytotoxicity and on the degree of differentiation of rat rNSC into astrocytes, neurons, and oligodendrocytes were monitored using cell-specific immunofluorescence staining for undifferentiated rNSC (nestin), neurospheres (nestin and A2B5), neurons (MAP2 clone M13, MAP2 clone AP18, and Doublecortin), astrocytes (GFAP), and oligodendrocytes (A2B5 and mGalc). In the absence of any chemical exposure, approximately 46% of rNSC differentiated into astrocytes and neurons, while 40% of the rNSC differentiated into oligodendrocytes. Both non-cytotoxic and cytotoxic concentrations of DA and OTA reduced the differentiation of rNSC into astrocytes, neurons, and oligodendrocytes. Furthermore, a non-cytotoxic nanomolar (0.05 µM) concentration of DA and 0.2 µM of OTA reduced the percentage differentiation of rNSC into astrocytes and neurons. Morphometric analysis showed that the highest concentration (10 μM) of DA reduced axonal length. These indicate that low, non-cytotoxic concentrations of DA and OTA can interfere with the differentiation of rNSC.


Introduction
The central nervous system (CNS) is a complex structure resulting from an intricately orchestrated sequence of events, which includes cell proliferation, differentiation, migration, synaptogenesis, neurogenesis, neurite and network formation, myelination, apoptosis, and gliogenesis. These stages of development occur sequentially according to different temporal profiles within brain regions, resulting in a heterogeneous pattern of synaptic connectivity [1][2][3][4]. Significant connectivity-based changes occur during development, a process that begins during gestation and continues until adolescence. Brain changes that occur during critical periods of development can result in permanent changes in brain structure and function and must be considered in neurotoxicity testing. As a result of these critical development periods, developing brains are more susceptible to chemical exposure than the adult brain [1][2][3][4]. In addition, children have increased susceptibility to toxic chemicals compared to adults because (a) they have higher exposure to chemicals for their body weight; (b) their metabolic pathways

Neurosphere Proliferation Without Differentiation
We observed the different size of neurospheres ranging from 0.05 to >1 mm in diameter and many of them were fused to form large neurospheres within 2-10 days of culturing in both attached ( Figure 2A) and suspension neurospheres ( Figure 2B). After 10 days of suspension and immunostaining with rNSC marker, attached neurospheres appeared clearly fused (Figure 2A,B). Our observations are similar to those previously reported [24][25][26].

Neurosphere Proliferation Without Differentiation
We observed the different size of neurospheres ranging from 0.05 to >1 mm in diameter and many of them were fused to form large neurospheres within 2-10 days of culturing in both attached ( Figure 2A) and suspension neurospheres ( Figure 2B). After 10 days of suspension and immunostaining with rNSC marker, attached neurospheres appeared clearly fused (Figure 2A,B). Our observations are similar to those previously reported [24][25][26].

Neurosphere Differentiation into Oligodendrocytes
Most of the rNSC neurospheres attached to Geltrex-coated slide chambers were differentiated into oligodendrocytes ( Figure 2C-E). The cell morphology was characterized by bipolar and multipolar morphology ( Figure 2D,E) at day 4 and day 9, respectively. After immunostaining with the oligodendrocyte marker, differentiated neurospheres clearly showed fused and fusing morphology ( Figure 2B,E).

Neurosphere Differentiation into Oligodendrocytes
Most of the rNSC neurospheres attached to Geltrex-coated slide chambers were differentiated into oligodendrocytes ( Figure 2C-E). The cell morphology was characterized by bipolar and multipolar morphology ( Figure 2D,E) at day 4 and day 9, respectively. After immunostaining with the oligodendrocyte marker, differentiated neurospheres clearly showed fused and fusing morphology ( Figure 2B,E).

rNSC Monolayer-Based Models for Three Types of Differentiation Processes
For the monolayer-based system, rNSC were cultured onto chambered slides with cell-specific coating matrix and containing cell-specific directed differentiation media as described in the materials and methods.

rNSC Differentiation Directed into Oligodendrocytes
The rNSC cultured on poly-l-ornithine and laminin coated chamber slides with oligodendrocyte differentiating medium were differentiated into oligodendrocytes ( Figure 1B-E). Of the total number of cells counted, 40% of these cells were oligodendrocytes, as shown by morphology and specific antibody staining. Morphological differences were observed in phase contrast microscope, A2B5 and mGalc immunofluorescent, and DAPI overlapped images ( Figure 1B-E). Both A2B5 and mGalc gave comparable results. Differentiation into oligodendrocytes is characterized by bipolar, multipolar, and highly branched neurites in morphology, and networks with neurons and other cells.

rNSC Differentiation Directed into Astrocytes
The rNSC were cultured in Geltrex-coated chamber slides with astrocyte differentiation media containing the N-2 supplement. Morphological characteristics of astrocytes appeared within 9 days ( Figure 1D, Figure 3A, and Figure 4A-D), as observed by the intricate networks associated with astrocytes, and visualized by immunofluorescent staining of astrocyte cell-specific marker GFAP ( Figure 1D, Figure 3A, and Figure 4A-D). In the controls, the percentage of differentiation into astrocytes was 46% of the total cells.  (A) rNSC grown in astrocyte differentiation media were differentiated into astrocytes, immuno-stained with astrocyte-specific marker GFAP (green). (B) rNSC grown in neuron differentiation media were differentiated into neurons, immuno-stained with neuron-specific marker MAP2 clone AP18 (pink).
An interesting observation was that four different types of astrocytes were observed in the phase contrast microscope, GFAP immunofluorescent, and DAPI overlapped images. However, only two types, the protoplasmic ( Figure 4A) and fibrous ( Figure 4B) types have been reported in rodents. The other two types of astrocytes were the varicose projection ( Figure 4C) and the interlaminar ( Figure 4D) types. These varicose projection-type astrocytes ( Figure 4C) and interlaminar-type astrocytes ( Figure 4D) have only been reported in higher primates and humans [27][28][29]. The interlaminar astrocytes have spherical cell bodies, whereas the varicose projection astrocytes have long processes. These were observed using immunofluorescence staining with GFAP ( Figure 4C,D and seen in Figures 3A and 5A). An interesting observation was that four different types of astrocytes were observed in the phase contrast microscope, GFAP immunofluorescent, and DAPI overlapped images. However, only two types, the protoplasmic ( Figure 4A) and fibrous ( Figure 4B) types have been reported in rodents. The other two types of astrocytes were the varicose projection ( Figure 4C) and the interlaminar ( Figure 4D) types. These varicose projection-type astrocytes ( Figure 4C) and interlaminar-type astrocytes ( Figure 4D) have only been reported in higher primates and humans [27][28][29]. The interlaminar astrocytes have spherical cell bodies, whereas the varicose projection astrocytes have long processes. These were observed using immunofluorescence staining with GFAP (

rNSC Differentiation Directed into Neurons
The rNSC cultured on poly-L-ornithine and laminin coated chamber slides with neuron differentiation medium were differentiated into neurons as manifested by the multipolar

rNSC Differentiation Directed into Neurons
The rNSC cultured on poly-l-ornithine and laminin coated chamber slides with neuron differentiation medium were differentiated into neurons as manifested by the multipolar morphology and intricate neurite networks ( Figure 1F, Figure 3B-D, and Figure 6A). Neurons gradually developed into mature morphology, as observed with a cell body containing a nucleus and extending two types of cytoplasmic dendritic processes. Neurons with several types of axon morphologically including unipolar, bipolar, multipolar, and pyramidal-like neurons were observed in vitro in phase contrast, MAP2 clone AP18 ( Figures 1F and 3B), Doublecortin ( Figure 3C), and MAP2 clone M13 ( Figure 3D) immunofluorescent and DAPI overlapped images. In the controls, the percentage of differentiation into neurons was 46% of the total cells.

Effects of Domoic Acid on rNSC Differentiation Directed into Astrocytes, Neurons, and Oligodendrocytes
rNSC were cultured in specific differentiation media and treated as described in the Materials and Methods section. Subsequently, these cells were fixed and immuno-stained with their corresponding cell-specific markers-GFAP for astrocytes, MAP2 and doublecortin for neurons, and A2B5 and mGalc for the oligodendrocytes. Fluorescent images of each differentiated cell type showed a reduction in the number of differentiated cells at day 9 in 0.05 µM of DA ( Figure 5B, Figure 6B, and Figure 7B) and 5 µM of DA ( Figure 5C, Figure 6C, and Figure 7C), compared to controls in 0 µM of DA ( Figure 5A, Figure 6A, and Figure 7A). The reduced number of differentiated cells are shown for astrocytes ( Figure 5B,C), neurons ( Figure 6B,C), and oligodendrocytes ( Figure 7B,C), respectively, compared to their corresponding controls ( Figure 5A, Figure 6A, and Figure 7A).

Effects of DA and OTA on Cytotoxicity and on the Differentiation of rNSC Directed into Astrocytes
The "total number of cells % control" was used as an index of cytotoxicity and the percentage differentiation was used as an index of the degree of differentiation of rNSC into astrocytes, neurons, and oligodendrocytes. There was a statistically significant effect of DA concentration on the percentage of viable cells, F(5,30) = 40.643, p < 0.001, as determined by one-way ANOVA. Pairwise comparison of the concentration-response of DA by Tukey's HSD post-hoc test revealed that all the concentrations of DA except 0.05 µM and 0.1 µM significantly reduced the "total number of cells % control" when compared with the control ( Figure 8A). There was no statistically significant difference in the survival rate at the lowest concentrations of DA (0.05 µM and 0.1 µM) compared to controls. Concentration-response graph of DA on "total number of cells % control" clearly showed that the concentration of DA is directly proportional to the cytotoxicity. Similarly, OTA showed a significant reduction in the "total number of cells % control" F(2,15) = 62.845, p < 0.001. Tukey's HSD post-hoc test revealed that 5 µM of OTA significantly reduced the "total number of cells % control" (p < 0.001), while 0.2 µM of OTA did not reduce it when compared with the control ( Figure 8A). The data suggest that DA concentrations of 0.5 µM or greater and 5 µM of OTA are cytotoxic.
In addition, these data showed a statistically significant effect of DA concentration on the percentage differentiation of rNSC into astrocytes F(5,53) = 43.696, p < 0.001. Scheffe post-hoc tests revealed that all DA concentrations significantly reduced the percentage differentiation of rNSC into astrocytes in a concentration-dependent manner ( Figure 8B). Even DA concentrations that were shown to be non-cytotoxic (0.05 µM and 0.1 µM) significantly reduced (p < 0.001) the differentiation of rNSC into astrocytes compared to control, 25.73 ± 2.59% and 31.86 ± 2.68%, respectively ( Figure 5A,B and Figure 8B).
Similarly, OTA-treated cells showed a significant effect on the percentage differentiation of rNSC into astrocytes F(2,32) = 44.198, p < 0.001. Only 0.2 µM OTA and greater treated cells showed a significant reduction in percentage differentiation of rNSC into astrocytes (p < 0.001), while 0.2 µM OTA-treated rNSC did not show significant reduction in "total number of cells % control" (Figure 8A,B).

Effects of DA and OTA on Cytotoxicity and on the Differentiation of rNSC Directed into Neurons
There was a statistically significant effect of different concentrations of DA on the "total number of cells % control" F(7,40) = 13.335, p < 0.001, as tested by one-way ANOVA. Tukey's HSD post-hoc test revealed that all concentrations except 0.05 µM of DA significantly reduced the "total number of cells % control" when compared with the control ( Figure 8C). The concentration-response graph of DA on "total number of cells % control" clearly showed that DA increased the cytotoxicity in a dose-dependent manner.
Similarly, different concentrations of OTA-treated cells showed statistically significant effects on the "total number of cells % control" F(4,25) = 10.423, p < 0.001. Tukey's HSD post-hoc test showed that only 1 µM and 5 µM of OTA significantly reduced "total number of cells % control" and hence increased cytotoxicity compared to the control. However  Figure 8D). Although there was no effect on the survival of total cells starting at 0.05 µM of DA and 0.2 µM of OTA ( Figure 8C), there was a statistically significant reduction on the differentiation of neurons ( Figure 8D, also seen in Figure 6A,B) compared to controls.
The degree of percentage differentiation of rNSC into neurons was decreased with increasing concentrations of DA ( Figure 6A-C and Figure 8D) or OTA ( Figure 8D) when compared with controls.

Effects of DA and OTA on Cytotoxicity and on the Differentiation of rNSC Directed into Oligodendrocytes
Cytotoxicity studies showed statistically significant effects of different concentrations of DA on the "total number of cells % control" F(7,42) = 33.822, p < 0.001, as determined by one-way ANOVA. Pairwise comparison of the DA-treated groups by Tukey's HSD post-hoc test revealed that all the used concentrations of DA significantly reduced the "total number of cells % control" (p < 0.001) when compared with the control ( Figure 8E). The concentration-response graph of DA revealed that all groups of DA-treated cells showed cytotoxicity.
Similarly, OTA-treated groups showed statistically significant effects on the "total number of cells % control" F(4,25) = 68.304, p < 0.001. Pairwise comparison of OTA-treated groups by Tukey's HSD post-hoc test revealed that all the concentration of OTA-treated groups except 0.05 µM of OTA significantly reduced the "total number of cells % control" (p < 0.001).
It showed that all concentrations of DA and OTA tested significantly reduced the "total number of cells % control" except 0.05 µM of OTA ( Figure 8E). This indicates that all concentrations of DA and OTA tested are cytotoxic except 0.05 µM of OTA, suggesting that oligodendrocyte differentiation may not be susceptible to low concentrations of OTA.
The data also showed a statistically significant effect of different concentrations of DA on the percentage differentiation of rNSC into oligodendrocytes F(6,84) = 35.848, p < 0.001. Pairwise comparison of all DA-treated groups by Scheffe post-hoc tests showed a significant reduction in percentage differentiation of rNSC into oligodendrocytes in DA-treated groups compared to the control ( Figure 8F). All cytotoxic concentrations of DA ( Figure 8E) significantly reduced the degree of differentiation of rNSC into oligodendrocytes compared to the control ( Figure 7A-C and Figure 8F). Immunostaining for oligodendrocytes showed that exposure to DA also reduced the degree of differentiation of rNSC into oligodendrocytes with increasing concentration of DA ( Figure 7A-C).
Similarly, there was a statistically significant effect of different concentration of OTA on the percentage differentiation of rNSC into oligodendrocytes, F(3,44) = 20.499, p < 0.001. Scheffe post-hoc test showed a significant reduction in percentage differentiation of rNSC into oligodendrocytes in OTA-treated groups compared to the control ( Figure 8F).
A dose-response graph showed that all concentrations of both DA and OTA significantly reduced the percentage differentiation of rNSC into oligodendrocytes relative to controls ( Figure 8F). However, the non-cytotoxic concentration of 0.5 µM OTA also significantly reduced the percentage differentiation of rNSC into oligodendrocytes. These data suggest that both non-cytotoxic and cytotoxic concentrations have an effect on the differentiation of rNSC into oligodendrocytes.

Effects of DA and OTA on the Axonal Length of the Mature Neurons
The axon length of the differentiated neurons was measured at the end of the treatment ( Figure 9C). The axons were traced in magenta color in luminous images of controls and DA-treated neurons ( Figure 9A,B). Non-parametric analysis done by the Kruskal-Wallis H test revealed that there was a statistically significant difference in axonal length of mature neurons between DA-treated groups and the control group,

Discussion
CNS neural stem cells present in developing and adult brains are multipotent, and thus able to differentiate into neurons, astrocytes, and oligodendrocytes. Throughout life, neural stem cells are present in the subventricular zone of the lateral ventricles and the subgranular zone of the dentate gyrus of the hippocampus. Hence, it is important to study the effects of toxins on these cells and how they could contribute to neurotoxicity and neurobehavioral outcomes. In this study, we differentiated rNSC into astrocytes, neurons, and oligodendrocytes in the presence and absence of DA or OTA. We tested the suitability of the rNSC suspension neurosphere assay, rNSC attached neurosphere assay, and rNSC monolayer system to investigate cytotoxicity and degree of differentiation. The results of the rNSC suspension neurosphere and attached neurosphere assays were unreliable and unreproducible as fusion of neurospheres were observed. Their quantification did not correlate with the rNSC numbers. These data are in agreement with the published data [24,25,30,31]. These authors have also highlighted the limitations and accuracy of the neurosphere assay for measuring NSC frequency in relation to NSC regulation. Based on these limitations, we abandoned the use of suspension neurosphere and attached neurosphere models. Hence, we used the rNSC monolayer-based system to investigate the effects in the presence and absence of DA and OTA on the cytotoxicity and degree of differentiation of rNSC into oligodendrocytes, astrocytes, and neurons. For the astrocytes and neurons, the differentiation media has been described by others; for differentiation into oligodendrocytes, differentiation medium was supplemented with neurogenic 2% B27 (components are listed in the Materials and Methods section), which is known to enhance maximum in vitro survival, improved maturation, and functionality of pluripotent stem cell (PSC)-derived neurons [32][33][34].
In the absence of any chemical exposure, the percentage of differentiation of rNSC into astrocytes and neurons was approximately 46.0%, and 40.0% for the oligodendrocytes. Non-cytotoxic and cytotoxic doses of DA and OTA resulted in the reduction of differentiation of rNSC into all three cell types, albeit in different ratios. For DA, 0.05 µM was non-cytotoxic for astrocytes and neurons but reduced the degree of differentiation of astrocytes (25.73 ± 2.59%) and neurons (7.95 ± 2.48%) compared with the relevant control. The non-cytotoxic dose for OTA for astrocytes and neurons was 0.2 µM, and much lower for the oligodendrocytes (0.05 µM). However, these doses reduced the degree of differentiation of astrocytes (24.5 ± 2.62%), neurons (12.53 ± 2.62%), and oligodendrocytes (11.47 ± 3.66%), compared with the relevant control. Based on the reduction of the degree of differentiation, DA was more potent than OTA. Furthermore, DA led to a reduction in the axon length at 10 µM (45.00 ± 2.22 ÃµM) compared with the control (78.67 ± 3.72 ÃµM).
Another noteworthy observation was that the differentiation of rNSC into astrocytes resulted in four types of morphologically distinct astrocytes, namely the protoplasmic, fibrous, varicose projection, and interlaminar types. Only the protoplasmic and fibrous types have been reported in rodents. The varicose projection and interlaminar types have only been reported in higher primates and humans [18,19,23]. This differentiation of rNSC into these two additional types of astrocytes could be attributed to the use of some components in the differentiation media of human origin (human transferrin and recombinant human insulin). We did not quantify or characterize which types of astrocytes was affected by these toxins, as there are a limited number of markers to identify and distinguish potentially heterogeneous astrocyte subtypes in animal models and in the human brain [31]. However, we counted the total number of astrocytes based on GFAP immunostaining.
The results show that both OTA and DA reduced degrees of differentiation of the main cell types of the CNS. This reduction in the mature cell types and changes in the axonal length may lead to changes in the cytoarchitectural or neural network changes in the brain, which may lead to functional/behavioral changes [35]. Alteration of the oligodendrocyte to neuron ratio may affect cognition due to the role of oligodendrocytes in synapse formation [35]. The astrocytes are the most abundant cells, of which the main task is to maintain the physiological homeostasis of neurons by providing antioxidant protection, substrates for neuronal metabolism, and glutamate clearance [36]. In addition, astrocytes are also involved in regulating synaptic activity and neuronal circuitry. Astrocytes are also known to modulate the blood-brain barrier (BBB) development during late embryogenesis and after birth [37]. Given the close coupling between astrocytes and neurons to form the neurovascular unit in adulthood [37], it is likely that its development must also be tightly regulated to coordinate developing cerebral vascular supply to neuronal demand [31]. Therefore, it could be hypothesized that both DA and OTA could interfere with the development of the BBB in the fetus and of neurovascular units in adulthood, since there was a decrease in the degree of differentiation of astrocytes and neurons upon exposure to these toxins. The BBB is formed by cerebral blood vessel endothelial cells in concert with astrocyte endfeet and creates a barrier between blood and the brain parenchyma. In humans, protoplasmic astrocytes wrap around blood vessels much more completely than they do in mice and rats, hence playing a much more important role in keeping agents in the blood from entering the brain and in regulating blood flow [38].
Oligodendrocytes are the most vulnerable cells in the brain [39]. This is supported by our findings as well, because the degree of differentiation into oligodendrocytes was the most affected by DA and OTA. Oligodendrocytogenesis during brain development is necessary for proper brain function, as oligodendrocytes form and maintain myelin sheaths around axons in the CNS. Disturbances of oligodendrocyte development may result in demyelination diseases that would severely affect neuronal functioning. These degenerative events could lead to changes in conduction velocity, the formation of redundant myelin, and enhanced thickness of myelin sheaths. Studies have shown that adult neural stem cells differentiated into oligodendrocytes are capable of re-myelinating axons following injury and myelin damage [40,41].
DA led to a reduction in the axonal length of neurons. Case reports of DA-poisoned patients show that they suffered from diffuse axonopathy [15], diffuse axonal sensorimotor neuropathy, and diffuse axonal damage [42]. DA poisoning in laboratory animals also showed degenerated axons [43,44]. In addition, DA exerts its neurodevelopmental effects through the alteration of neurogenesis and morphology, particularly within the hippocampus, showing EEG irregularities and decreased threshold to DA-induced seizures [15]. According to previously published data, DA-epileptic disease is characterized by spontaneous recurrent seizures after weeks to months of DA poisoning. Furthermore, atypical behaviors in animals and the latent period of silent toxicity characterizes the transition between DA poisoning and epileptic disease [14,45,46].
The data from this study show the reduction in the degree of differentiation of rNSC into all three type of cells and the reduction in axonal length of neurons in the presence of low non-cytotoxic concentration of DA (0.05 µM). Data from rodents and non-human primates have shown that non-cytotoxic low dose exposure to DA during early life was associated with behavioral, memory deficits, and/or structural changes in the brain [47][48][49][50][51][52]. This memory loss may be linked to a reduction and/or the abnormal differentiation of neural stem cells. Several studies have shown that transplanted mouse NSCs differentiate into mature cell types within the brain and improve learning and memory in mouse models of Alzheimer's disease [53][54][55].
The non-cytotoxic concentrations of OTA that reduced the degree of differentiation is consistent with the previous data published by Sava et al [9]. They showed that low concentrations of OTA (0.01-100 µg/ml) caused a dose-dependent decrease in viability of both proliferating and differentiating NSC. However, these authors did not quantify rNSC differentiation into astrocytes, neurons, and oligodendrocytes. Further, non-cytotoxic concentrations of OTA (<10 nM) perturbed the homeostasis of highly differentiated neural stem cell cultures [11].
The mechanism by which DA and OTA reduce the degree of differentiation of rNSC into the different cell types is not fully known at present. The cytotoxic effect of these chemicals could be due to the activation of intracellular pathways that lead to apoptosis, even at exposure levels that do not cause the death of other neural cell types. DA is known to activate α-amino-3-hydroxy -5-methyl-4-isoxazolepropionic acid and kainate receptors which, in turn, causes glutamate release that subsequently activates the N-methyl-d-aspartic acid receptor, causing apoptotic and necrotic neuronal cell death [56,57]. The effects of OTA on the developing brain have not been fully characterized [58].
To fully elucidate the mechanisms of action of DA and OTA on neurogenesis, molecular pathways related to the developmental processes need to be deciphered.

Reagents
All reagent preparation, use, and disposal of toxic and bio-hazardous materials were performed according to the laboratory safety regulations of Health Canada. Both DA and OTA were purchased from Sigma-Aldrich (St. Louis, MO, USA). These were dissolved in ethanol, and sterile water was added to make a 1 mM stock solution (final concentration of 0.1% ethanol), which was stored at −20 Commercially available multipotent fetal rNSC extracted from Sprague-Dawley embryonic day 14 rats (Gibco) were differentiated into astrocytes, neurons, and oligodendrocytes [59,60]. rNSC were seeded (2 × 10 6 cells) in complete StemPro NSC SFM medium and incubated at 37 • C, 5% CO 2 , and 90% humidity. On the following day, the medium was replaced with an equal volume of fresh, pre-warmed complete StemPro NSC SFM. The medium was replaced every 2-3 days. The cells were passaged when culture was at 75-90% confluent; and P3 cells were subsequently used for the differentiation experiments.

rNSC Neurosphere-Based Model for Differentiation
Attached neurospheres and neurosphere suspensions were used for this experiment. For attached neurospheres, each chamber was coated with 400 µL of 1× Geltrex (Gibco, Life Technologies Corporation, Carlsbad, CA, USA) for 4 h at RT in laminar flow and the excess Geltrex solution was removed immediately before use. For neurosphere suspension, uncoated slide chambers were used. rNSC were seeded at 5.6 × 10 4 cells/chamber in complete StemPro NSC SFM medium and incubated at 37 • C, 5% CO 2 , and 90% humidity. After 2 days, the Geltrex-coated slide chambers showed attached neurospheres, while uncoated slide chambers showed suspension neurosphere culture. Medium in the coated slide chambers was replaced by the oligodendrocyte differentiation medium, whereas medium of the control chambers was changed with same complete StemPro NSC SFM medium. All chamber slides were incubated at 37 • C, 5% CO 2 , and 90% humidity, and relevant media were changed every 2 days and continued culturing for 7 days for the differentiation process. However, when attempts were made to change the media for suspension neurosphere cultures, most of the smaller suspended neurospheres were lost during aspiration.

A. Coating Chamber Slides for Differentiation into Various Cell Types
For monolayer-based assays and for differentiation into neurons and oligodendrocytes, the chamber slides were double coated. First, each chamber surface was covered with 500 µL of Poly-L-ornithine (20 µg/mL) and incubated overnight at RT in laminar flow, and subsequently washed twice with sterile water and air dried. This was followed by coating each chamber with 500 µL of laminin (10 µg/mL, Gibco, Life Technologies Corporation, Frederick, MD, USA) and incubated for 2 h at 37 • C and rinsed with D-PBS without Ca 2+ and Mg 2+ .
C. rNSC Culturing for Differentiation rNSC (P3) were seeded at 5.6 × 10 4 cells per chamber in complete StemPro NSC SFM medium in all chambers and incubated at 37 • C, 5% CO 2 , and 90% humidity. The cells were allowed to attach for 2 days, the, medium in the control chambers coated with poly-l-ornithine was replaced by the complete StemPro NSC SFM medium for culturing rNSC without differentiation. For the differentiation of rNSC to astrocytes, medium in the slide chambers coated with Geltrex was replaced with astrocyte differentiation medium. For the differentiation of rNSC into oligodendrocytes, medium in the slide chambers coated with poly-l-ornithine and laminin was replaced with oligodendrocyte differentiation medium. For the differentiation of rNSC to neurons, medium in the slide chambers coated with poly-l-ornithine and laminin was replaced with neuron differentiation medium. All chamber slides were incubated at 37 • C, 5% CO 2 , and 90% humidity and relevant media were changed every 2 days and continued culturing for 7 days for the differentiation process. After 7 days (DIV 9-10) of differentiation, process medium in each chamber was aspirated and cells were washed once with PBS and aspirated. To fix the cells, 0.5 mL of 4% PFA was added and incubated at RT for 15 min, and washed three times with PBS with Ca 2+ and Mg 2+ . Then, 0.5 mL of PBS with Ca 2+ and Mg 2+ was added to each chamber, wrapped with parafilm, and stored at 4 • C for immunostaining.

Cytotoxicity Assays (Relative Cell Count)
The relative cell count (RCC) was expressed as a "total number of cells % control" and was used as an index of cytotoxicity, which has been used in previous publications [62,63].

Effects of DA and OTA on Directed Differentiation on the Different Cell Types
The above experiments were repeated with different concentrations of freshly prepared DA or OTA, added separately to each differentiation media. Differentiation media for each cell type containing DA or OTA were replaced every two days. The concentrations of DA used were 0.05, 0.1, 0.2, 0.5, 1, 5, and 10 µM, while OTA effects were examined at 0.05, 0.2, 1, and 5 µM. Controls were treated only with the vehicle used to prepare the relevant concentration of chemicals. After 7 days of differentiation process in the presence and absence of DA or OTA, the medium was aspirated, cells were washed, fixed, and immuno-stained for enumeration. The total number of cells (% control) and percentage differentiation [64] of rNSC into astrocytes, oligodendrocytes, and neurons were measured. The RCC was expressed as "total number of cells % control" and has been used as an index of cytotoxicity previously [62,63].

Image Analysis
All images were captured using a fluorescence and transmitted light microscope, EVOS FL Color Imaging System equipped with a Sony ICX285AQ color CCD camera. Image analysis was done in 10× and 20× 24-bit color TIFF (1360 × 1024 pixels) images using EVOS FL Color Imaging software with manual assisted cell counting and FIJI ImageJ software. Cells with uniformly stained DAPI nucleus with proper morphology were counted for the quantification of a total number of cells and were compared with the relevant control. Cells stained with specific markers, complete DAPI stained nucleus, and morphology with neurites were considered in quantification. Fluorescent, DAPI, and phase contrast overlapping images were used for percentage cell differentiation quantification. More than 10 randomly selected fields of 200 × 400 µm area were selected from each slide chamber for the percentage cell differentiation quantification, whereas 400 × 800 µm area were selected for total cell quantification.

Tracing and Measurement of the Axonal Length of the Mature Neurons
Tracing and measurement of the axonal length of the mature neurons were done on rNSC directed into neuron differentiation process in the absence and presence of DA and OTA. After 7 days of treatment, the cells were fixed and stained. Subsequently, axons of treated and control neurons were traced and length measured using Neurite Tracer in the ImageJ software (National Institute of Health, Bethesda, MD, USA). The axons were traced in magenta color in luminous images of controls, in 0.05, 1, and 10 µM of DA-and OTA-treated differentiating neurons.

Statistical Analysis
All the statistical analyses were performed by one-way ANOVA using the IBM SPSS 19 program (IBM, New York, NY, USA). Both the Shapiro-Wilk test and the Kolmogorov-Smirnov test were used to check normal distribution. For concentration response experiments, total cell count, cell differentiation, and axonal length data were normalized within an experiment to corresponding control prior to statistical analysis. Tukey's HSD post-hoc test was used for a situation with equal sample sizes, and the Scheffe post-hoc test was used for unequal sample size per group. These pairwise comparison tests compare the difference between various parameters of control and the DA-and OTA-treated cells. Non-parametric statistical data were analyzed by one-way ANOVA Kruskal-Wallis H test. All data shown are means ± standard error (SE) in graphs. Statistical significance is given as follows: * p < 0.05, ** p < 0.01, *** p < 0.001 in each experiment legend.

Conclusion and Future Perspectives
The present data show that the rNSC monolayer-based system is a reliable assay system for monitoring the neurotoxic effects of DA and OTA. In addition, data show that non-cytotoxic concentrations led to a reduction in the degree of differentiation of rNSC into astrocytes, neurons, and oligodendrocytes. These changes in the cell types and numbers could lead to altered cytoarchitecture and affect processes including cell division, migration, differentiation and cell death, all of which regulate neural development.
This model system can be used to test the effects of unknown chemicals in conjunction with molecular tools such as transcriptomics, proteomics, and metabolomics to dissect pathways related to the developmental processes.
Author Contributions: S.G. and V.M.R.K. equally contributed to the conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing-original draft preparation, writing-review and editing, and visualization.