Stimulation of Neurite Outgrowth in Cerebrocortical Neurons by Sodium Channel Activator Brevetoxin-2 Requires Both N-Methyl-D-aspartate Receptor 2B (GluN2B) and p21 Protein (Cdc42/Rac)-Activated Kinase 1 (PAK1)

N-methyl-D-aspartate (NMDA) receptors play a critical role in activity-dependent dendritic arborization, spinogenesis, and synapse formation by stimulating calcium-dependent signaling pathways. Previously, we have shown that brevetoxin 2 (PbTx-2), a voltage-gated sodium channel (VGSC) activator, produces a concentration-dependent increase in intracellular sodium [Na+]I and increases NMDA receptor (NMDAR) open probabilities and NMDA-induced calcium (Ca2+) influxes. The objective of this study is to elucidate the downstream signaling mechanisms by which the sodium channel activator PbTx-2 influences neuronal morphology in murine cerebrocortical neurons. PbTx-2 and NMDA triggered distinct Ca2+-influx pathways, both of which involved the NMDA receptor 2B (GluN2B). PbTx-2-induced neurite outgrowth in day in vitro 1 (DIV-1) neurons required the small Rho GTPase Rac1 and was inhibited by both a PAK1 inhibitor and a PAK1 siRNA. PbTx-2 exposure increased the phosphorylation of PAK1 at Thr-212. At DIV-5, PbTx-2 induced increases in dendritic protrusion density, p-cofilin levels, and F-actin throughout the dendritic arbor and soma. Moreover, PbTx-2 increased miniature excitatory post-synaptic currents (mEPSCs). These data suggest that the stimulation of neurite outgrowth, spinogenesis, and synapse formation produced by PbTx-2 are mediated by GluN2B and PAK1 signaling.

Activity-dependent control of neuronal development involves calcium and neurotrophic signaling. Calcium (Ca 2+ ) signaling involves Ca 2+ influx pathways, including NMDARs and voltage-gated Ca 2+ channels (VGCCs) [8][9][10]. NMDARs belong to the family of ionotropic glutamate receptors and play a critical role in activity-dependent development
sodium-calcium exchanger inhibitor; NCXrev) affected the NMDA-induced Ca 2+ ( Figure 2B). Both nifedipine and KB-R7943 had minimal effects on the NMDA-in Ca 2+ influx, with nifedipine showing a 6% reduction ( Figure S2A) and KB-R7943 di ing a 7% increase ( Figure S2B). We next assessed the effect of PbTx-2 alone on Ca namics in DIV 10-12 cerebrocortical neurons. PbTx-2 (100 nM) produced an incre [Ca 2+ ]i, which was partially blocked by nifedipine ( Figure 2C), KB-R7943 ( Figure 2D MK-801 ( Figure 2E). All three inhibitors displayed a substantial reduction in the Pb induced Ca 2+ influx, with nifedipine showing a 32% reduction ( Figure S2C), KBshowing a 54% reduction ( Figure S2D), and MK-801 displaying a 48% reduction R7943 is an isothiourea derivative that has been shown in cardiac ventricular cells to erentially inhibit the reversed operation of the Na + /Ca 2+ exchanger (IC50 = 0.32 μM) inhibiting with lower potency voltage-gated sodium channels (IC50 = 14 μM). In th sent study, KB-R7943 was used at a concentration of 3 μM to minimize its effect o voltage-gated sodium channel function while maximizing Na + /Ca 2+ exchanger inhib At this concentration, KB-R7943 would be expected to produce a greater-than-90% tional occupancy of the Na + /Ca 2+ exchanger and a less-than-20% fractional occupa voltage-gated sodium channels [60,61]. It is, therefore, reasonable to infer that the pr effect of 3 μM KB-R7943 is to inhibit the reversed mode of the Na + /Ca 2+ exchanger tion. NMDA and PbTx-2 exposure triggers distinct Ca 2+ influx pathways. NMDA-induc Ca 2+ influx is mediated exclusively by NMDARs with an absence of an effect by the L-type V inhibitor nifedipine (A) or the reverse sodium-calcium exchange inhibitor KB-R7943 (B), wh PbTx-2-induced Ca 2+ influx involves L-type Ca 2+ channels, as shown with nifedipine (C); sod Ca 2+ exchanger, as shown with KB-R7493 (D); and NMDARs, as shown with MK-801 (E). Th Figure 2. NMDA and PbTx-2 exposure triggers distinct Ca 2+ influx pathways. NMDA-induced Ca 2+ influx is mediated exclusively by NMDARs with an absence of an effect by the L-type VGCC inhibitor nifedipine (A) or the reverse sodium-calcium exchange inhibitor KB-R7943 (B), whereas PbTx-2-induced Ca 2+ influx involves L-type Ca 2+ channels, as shown with nifedipine (C); sodium Ca 2+ exchanger, as shown with KB-R7493 (D); and NMDARs, as shown with MK-801 (E). This experiment was repeated in triplicate in DIV 10-12 cerebrocortical neurons. Data shown represent the mean ± SEM of 3 experiments.
Mar. Drugs 2022, 20, x FOR PEER REVIEW 7 o outgrowth in a biphasic pattern. A previous report has shown that ifenprodil potentia nerve-growth-factor-induced neurite outgrowth in PC12 cells [63].

PbTx-2-Induced Neurite Outgrowth Involves Rac1 but Not Rho A.
Rho family GTPases play an important role in the regulation of actin dynamics. The members of the Rho GTPase family that are well studied are Cdc42, Rac1, and RhoA. To explore their role in PbTx-2-induced neurite outgrowth, we used a Rac1 selective inhibitor

PbTx-2-Induced Neurite Outgrowth Involves Rac1 but Not Rho A
Rho family GTPases play an important role in the regulation of actin dynamics. The members of the Rho GTPase family that are well studied are Cdc42, Rac1, and RhoA. To explore their role in PbTx-2-induced neurite outgrowth, we used a Rac1 selective inhibitor NSC23766 that inhibits Rac1-GEF interaction and, thus, the activation of Rac GTPase. In vitro studies have shown that NSC23766 inhibits the Rac1 binding and activation via Rac-specific GEF Trio or Tiam 1 (IC 50~5 0 µM) without altering RhoA or Cdc42 binding or activation [68]. To block the Rho pathway, we utilized Y27632 (K i = 0.14 µM), a selective inhibitor of the Rho kinase [69]. The Rac1 inhibitor NSC23766 blocked PbTx-2-induced neurite outgrowth ( Figure 6A,B), but the Rho inhibitor Y27632 was without an effect ( Figure 6C,D).
Mar. Drugs 2022, 20, x FOR PEER REVIEW 10 temporal activation of PAK1. PbTx-2 treatment led to increased phosphorylation of T212, which peaked at 15 min post-exposure ( Figure 8).   Data shown represent the mean ± SEM of 2 experiments.

PbTx-2 Induces Release of Glutamate into Extracellular Medium
PbTx-2 (30 nM) treatment led to a functional consequence, as demonstrated by the increase in neurite outgrowth. We next examined the release of glutamate into extracellular medium follows PbTx-2 treatment. PbTx-2 (30 nM) was used to provoke glutamate release in murine cerebrocortical neurons. Extracellular glutamate levels were measured by collecting the culture medium of primary neurons that had been exposed to either vehicle or PbTx-2 (30 nM). Glutamate levels were quantified via a glutamate ELISA assay. Our result demonstrated that PbTx-2 (30 nM) caused an approximately two-fold increase in glutamate release compared to the vehicle-treated group ( Figure 9A).  D). Both experiments were repeated twice, and 25 to 30 neurons were quantified for each exposure condition. (* one-way ANOVA, followed by Dunnett's multiple comparison test, with control p < 0.0001; # oneway ANOVA, followed by Dunnett's multiple comparison test, with PbTx-2 30 nM p < 0.0001). Data shown represent the mean ± SEM of 2 experiments.

PbTx-2 Induces Release of Glutamate into Extracellular Medium
PbTx-2 (30 nM) treatment led to a functional consequence, as demonstrated by the increase in neurite outgrowth. We next examined the release of glutamate into extracellular medium follows PbTx-2 treatment. PbTx-2 (30 nM) was used to provoke glutamate release in murine cerebrocortical neurons. Extracellular glutamate levels were measured by collecting the culture medium of primary neurons that had been exposed to either vehicle or PbTx-2 (30 nM). Glutamate levels were quantified via a glutamate ELISA assay. Our result demonstrated that PbTx-2 (30 nM) caused an approximately two-fold increase in glutamate release compared to the vehicle-treated group ( Figure 9A).

PbTx-2-Induced Elevation of Miniature Excitatory Post-Synaptic Current (mEPSCs) Is Dependent on PAK1
Spontaneous glutamate mEPSCs are responses to the release of a single quantum of the excitatory neurotransmitter glutamate. Given the previous demonstration that PbTx-2 exposure enhanced synaptogenesis in immature cerebrocortical neurons, we determined the effect of PbTx-2 on mEPSCs. PbTx-2 increased mEPSCs in DIV-5 cerebrocortical neurons, reflecting accelerated synaptogenesis, and this effect was abrogated by IPA-3 ( Figure 9B,C). Based on experimental evidence, we hypothesized that exposure to PbTx-2 increases neurite outgrowth via an NMDA receptor and PAK1 mechanism. The assessment of mEPSC was to address the formation of functional synapses. Recordings of mEPSC at −70 mV would primarily assess AMPA receptor currents, which are observed at mature synapses rather than, for example, silent synapses that may contain NMDA receptors but lack AMPA receptors. PbTx-2 facilitated the formation of functional synapses, as evidenced by the increase in mEPSC events, and IPA3 inhibited this effect, suggesting dependence on PAK1.

PbTx-2 Enhancement of Dendritic Complexity Requires PAK1
To test the involvement of PAK1 in PbTx-2-induced increases in dendritic arborization, cerebrocortical neurons were treated with PbTx-2 in the presence and absence of IPA-3. In DIV-5 control neurons, Sholl analysis indicated a gradual increase in branch complexity as a function of distance from the soma, reaching a maximum of 6.5 ± 0.3 (control) and 6.4 ± 0.3 (IPA-3 1 µM) intersections per neuron at 11 µm ( Figure 10A,B). PbTx-2 (30 nM) produced a robust increase in dendritic complexity, reaching a maximum of 10.0 ± 0.7 intersections per neuron at 11 µm. IPA-3 pretreatment reduced this effect of PbTx-2 (30 nM) by 7.0 ± 0.4 intersections per neuron at 11 µm. An area under the curve (AUC) analysis of these Sholl data showed a significant increase in dendritic complexity following 30 nM PbTx-2 compared with control (ANOVA, * p < 0.05; Dunnett's post hoc test), which was blocked by IPA-3 ( Figure 10C).

PbTx-2 Enhancement of Dendritic Complexity Requires PAK1
To test the involvement of PAK1 in PbTx-2-induced increases in dendritic arb tion, cerebrocortical neurons were treated with PbTx-2 in the presence and absence o 3. In DIV-5 control neurons, Sholl analysis indicated a gradual increase in branch plexity as a function of distance from the soma, reaching a maximum of 6.5 ± 0.3 (co and 6.4 ± 0.3 (IPA-3 1 µ M) intersections per neuron at 11 µ m ( Figure 10A,B). PbTx nM) produced a robust increase in dendritic complexity, reaching a maximum of 0.7 intersections per neuron at 11 µ m. IPA-3 pretreatment reduced this effect of P (30 nM) by 7.0 ± 0.4 intersections per neuron at 11 µ m. An area under the curve ( analysis of these Sholl data showed a significant increase in dendritic complexity f ing 30 nM PbTx-2 compared with control (ANOVA, * p < 0.05; Dunnett's post hoc which was blocked by IPA-3 ( Figure 10C).

PbTx-2 Treatment Produces Phosphorylation of Cofilin and Increases F-actin Density
Next, we assessed the effect of PbTx-2 on actin dynamics. Cofilin has been impl as an important regulator of synaptic plasticity under both physiological and pathol conditions [74,75]. The phosphorylation of cofilin suppresses its activity and has shown to promote the development of mature spines [76,77]. To explore the role of 2-induced changes in actin dynamics, DIV-5 cerebrocortical neurons were treated . Histogram summary (mean ± SEM) for the area under the curve analysis of Sholl data (C). The experiment was performed two times with independent cultures, and 25 to 30 neurons were quantified for each exposure condition. (* one-way ANOVA, followed by Dunnett's multiple comparison test, with control p < 0.05; # one-way ANOVA, followed by Dunnett's multiple comparison test, with PbTx-2 30 nM p < 0.05).

PbTx-2 Treatment Produces Phosphorylation of Cofilin and Increases F-Actin Density
Next, we assessed the effect of PbTx-2 on actin dynamics. Cofilin has been implicated as an important regulator of synaptic plasticity under both physiological and pathological conditions [74,75]. The phosphorylation of cofilin suppresses its activity and has been shown to promote the development of mature spines [76,77]. To explore the role of PbTx-2-induced changes in actin dynamics, DIV-5 cerebrocortical neurons were treated with different concentrations of PbTx-2 (10, 30, and 300 nM). PbTx-2 treatment showed a modest enhancement in the phosphorylation of cofilin in DIV 5 cerebrocortical neurons ( Figure 11A). We then assessed the response of F-actin following PbTx-2, IPA-3, or both treatments in DIV-1 cerebrocortical neurons stained with phalloidin and quantified using Image J. PbTx-2 treatment led to an increase in F-actin density, which was blocked by IPA-3 ( Figure 11B,C).
Mar. Drugs 2022, 20, x FOR PEER REVIEW 1 Image J. PbTx-2 treatment led to an increase in F-actin density, which was blocked by 3 ( Figure 11B,C).

Discussion
Neuronal activity plays a key role in regulating the morphology and connectiv neurons during development. The mechanisms by which neurons decode this neu activity into the activation of signaling pathways that regulate morphological comp are not fully understood. These pathways are reported to involve local (modulat actin dynamics) as well as global (activation of nuclear signaling and gene transcri signaling mechanisms [12]. The elevation of intracellular sodium in dendrites and s as a consequence of neuronal activity affects NMDAR function and activity-depe synaptic plasticity [78]. We have previously shown that Ca 2+ entry in neurons exposed to PbTx-2 o through three primary routes: NMDA receptor ion channels, L-type Ca 2+ channel the reversal of the Na + /Ca 2+ exchanger [61]. The NMDA-receptor-dependent increm [Ca 2+ ]i represents a relatively small fraction of the total Ca 2+ load induced by PbTx-2 sure. In contrast, when exposed to an NMDA receptor agonist such as L-glutamat Figure 11. PbTx-2 treatment increased the phosphorylation of cofilin and F-actin density. PbTx-2 causes increased phosphorylation of cofilin in DIV5 cerebrocortical neurons (A). Representative images of DIV 1 cerebrocortical neurons (B). PbTx-2 treatment leads to an increase in F-actin density, which is blocked by IPA-3 (B,C). The experiment was performed two times with independent cultures. (* one-way ANOVA, followed by Dunnett's multiple comparison test, with control p < 0.05; # one-way ANOVA, followed by Dunnett's multiple comparison test, with PbTx-2 30 nM p < 0.05). Data shown represent the mean ± SEM of 2 experiments.

Discussion
Neuronal activity plays a key role in regulating the morphology and connectivity of neurons during development. The mechanisms by which neurons decode this neuronal activity into the activation of signaling pathways that regulate morphological complexity are not fully understood. These pathways are reported to involve local (modulation of actin dynamics) as well as global (activation of nuclear signaling and gene transcription) signaling mechanisms [12]. The elevation of intracellular sodium in dendrites and spines as a consequence of neuronal activity affects NMDAR function and activity-dependent synaptic plasticity [78].
We have previously shown that Ca 2+ entry in neurons exposed to PbTx-2 occurs through three primary routes: NMDA receptor ion channels, L-type Ca 2+ channels, and the reversal of the Na + /Ca 2+ exchanger [61]. The NMDA-receptor-dependent increment in [Ca 2+ ] i represents a relatively small fraction of the total Ca 2+ load induced by PbTx-2 exposure. In contrast, when exposed to an NMDA receptor agonist such as L-glutamate, the increase in [Ca 2+ ] i appears to be almost entirely through the NMDA receptors [61].
PbTx-2 treatment mimics activity-dependent structural plasticity in that it produces an increase in dendritogenesis, spinogenesis, and synaptogenesis [6,64]. Given the dependence of PbTx-2-induced structural plasticity on NMDAR function, here, we have addressed the NMDAR subtype and downstream signaling mechanisms by which sodium channel activators influence neuronal morphology in DIV 1 and 5 cerebrocortical neurons. Inasmuch as synaptic activity has been shown to elevate [Na + ] i [79] and sodium acts as a positive regulator of NMDAR function [78], we used the sodium channel activator PbTx-2 as a probe to explore the influence of sodium on NMDAR signaling.
Although NMDARs represent a key source of Ca 2+ entry following PbTx-2 exposure, the NMDAR subtype responsible for this response was unknown. In this report, we show that PbTx-2-induced Ca 2+ influx and neurite length were blocked by a GluN2B subtypeselective antagonist in cerebrocortical neurons (Figures 1-4). Notably, GluN2B-containing NMDARs have higher surface mobility than GluN2A-containing receptors [80]. We have previously reported that PbTx-2 exposure leads to increased surface expression of GluN2B, suggesting the involvement of this subtype of NMDAR in Ca 2+ influx [3]. GluN2B receptors have been shown to increase spine and filopodia motility [81,82], which are important for establishing synaptic connections [83]. Moreover, the GluN2B antagonist ifenprodil has been shown to prevent NMDA-induced filopodia formation [84]. A reduction in the number of functional synapses was indicated by a significant reduction in the frequency of mEPSCs and significantly lower spine density [82].
Many of the effects of neuronal activity on synaptic plasticity are mediated by Ca 2+dependent signaling events [85]. Ca 2+ /CaM dependent protein kinases (CaMKs) appear to be key mediators of Ca 2+ -dependent neurite outgrowth [86]. Members of the CaMK family are activated in response to GluN receptor stimulation and following neuronal activity [40]. Among the CaMKs, CaMKII is abundantly present in the CNS, with a wide repertoire of substrates, including cytoskeletal proteins. Additionally, GluN2B is reported to be a key activity-dependent recruiter of CaMKII to post-synaptic sites [87]. Such activity-dependent incorporation of CaMKII into post-synaptic sites may play a role in structural and functional synapse maturation during development as well as in learning and memory [88]. Our results ( Figure 5) are consistent with previous findings where Ca 2+ /calmodulin-dependent protein kinase II was shown to promote neurite outgrowth in Neuro-2a cells [89,90]. Likewise, inhibition of the CaMK family with KN-93 has been associated with a reduction in synaptic plasticity [65,67].
The Rho family of small GTPases, which includes RhoA, Rac1, and Cdc42, promote morphological changes during neuronal development, including neurite outgrowth, axonal guidance, and dendritic development [91,92]. Rac1 activation is closely coupled to the activation of Cdc42 [93], allowing for the coincident and coordinated formation of filopodia and lamellopodia. NMDAR stimulation is able to regulate Rac1-dependent actin remodeling, which is important for the development and structural remodeling of dendritic arbors and spines [94,95]. Here, we show that PbTx-2 feeds into the Rac1 pathway to induce neurite outgrowth and that blocking RhoA does not inhibit PbTx-2-induced neurite outgrowth ( Figure 6). Our results are consistent with previous findings, where it has been generally demonstrated that RhoA inhibits, whereas Rac1 and Cdc42 promote, the growth and stability of dendritic spines [96]. One study using Rac1 inhibitor NSC23766 in rat cortical neurons showed this inhibitor might function as an NMDAR antagonist instead of inhibiting cytosolic Rac1; however, in that study, the authors used a considerably higher concentration of NSC 23766 (100 µM) [97], whereas we used 1 and 10 µM NSC23766 concentrations. Using higher concentrations of this inhibitor might contribute to off-target effects of this Rac1 antagonist.
PAK1 is a key downstream effector of Rac1 [70]. PAK1 was originally identified in a Rho-GTPase screen, where it was complexed with activated GTP-Rac1 [70]. PAK1 is a serine/threonine kinase involved in cellular activities such as cytoskeletal dynamics, cell migration, neurogenesis, angiogenesis, mitosis, apoptosis, and transformation [98][99][100][101][102][103]. PAK kinases are effectors of Rac1 and have been shown to play an important role in neurite initiation and outgrowth [104]. Inhibition of PAK1 leads to a decrease in basal NMDAR currents [105]. Our result suggests that PbTx-2-induced neurite outgrowth involves PAK1 since inhibition by both a non-competitive inhibitor of PAK1 (IPA-3) and PAK1 siRNA blocks PbTx-2-induced neurite outgrowth (Figure 7). Further, PAK1 has a phosphorylation site on T212, a site absent in PAK2 and PAK3 [106]. Areas of embryonic mouse brain demonstrate differential expression of phosphorylated PAK-T212, which is present primarily in the cerebral cortex, hippocampus, and thalamus. Increased PAK1-T212 phosphorylation plays a specific role in neurons when they are undergoing extensive cytoskeletal changes, such as during axonal outgrowth [107]. Furthermore, it has been demonstrated that mice treated with the Rac1 inhibitor showed a substantial reduction of PAK1-T212 phosphorylation levels, providing direct evidence that Rac1 regulates PAK1-T212 phosphorylation [108]. We found that PbTx-2 treatment increases the phosphorylation of PAK1 at T212, which implicates a role for this phosphorylation site in PbTx-2-induced neurite outgrowth (Figure 8). Future studies are warranted to explore additional phosphorylation sites on PAK1 that may be involved in response to PbTx-2 exposure.
Dissociated neurons in culture form a network of synaptically connected cells, and spontaneous Ca 2+ oscillations that are thought to be due to the rhythmic release of glutamate. Such synchronized Ca 2+ oscillations in primary neuronal cultures measured at the population level with Ca 2+ -sensitive fluorescent probes have been strictly associated with bursts of action potentials [109]. Previously, we reported that PbTx-2 exposure in cerebrocortical neurons produces an increase in [Na + ] i , an up-regulation of NMDAR function, acceleration of the emergence of spontaneous Ca 2+ oscillations, and the engagement of downstream Ca 2+ dependent signaling pathways [64]. Our results herein demonstrate that PbTx-2 treatment provoked the release of glutamate ( Figure 9A), and this increment in extracellular glutamate was sufficient to cause an increase in spontaneous Ca 2+ oscillations and neurite outgrowth in cerebrocortical neurons.
One of the hallmarks of synaptic scaling is the frequency of mEPSCs [110]. Previously, we reported that PbTx-2 treatment increased synaptic density [64]. In this report, PbTx-2 treatment led to an increase in mEPSCs events, which were blocked by IPA-3, confirming that PbTx-2 treatment caused increased synapse formation with the involvement of PAK1 ( Figure 9B,C). We have also shown that PbTx-2 30 nM treatment increased dendritic arborization and filopodia formation [64]. Our results suggest that PbTx-2-induced neurite outgrowth involves PAK1 (Figure 10), inasmuch as IPA-3 blocks dendritic arborization.
A major downstream effector of PAK1 is cofilin, an actin-binding protein essential for controlling the equilibrium between filamentous and monomeric actin [111]. Cofilin is inactivated by LIM-kinase-mediated phosphorylation (LIM kinase is a substrate of PAK) and is reactivated by cofilin phosphatase [112]. The dephosphorylated cofilin binds to F-actin, leading to actin-severing depolymerization [75,113]. Since the Ser3 residue of cofilin acts as a switch for actin assembly (F-actin stabilization) and disassembly (F-actin severing) [114], we used a Ser3-phosphorylated antibody to probe for PbTx-2-induced p-cofilin levels. We found that PbTx-2 causes a significant increase in p-cofilin expression in DIV-5 neurons ( Figure 11A) and increased density of F-actin throughout the dendritic arbor and soma ( Figure 11B,C) in DIV-1 neurons.
Based on these data, we proposed a model for PbTx-2-induced neurite outgrowth ( Figure 12) that involves GluN2B-NMDARs. The involvement of this NMDAR subtype leads to the activation of CaMKII, Rac1, and its effector PAK1, causing changes in cofilin activity and the stabilization of F-actin, resulting in the stabilization of the actin cytoskele-ton. Thus, VGSC activators may represent a novel pharmacological strategy to promote neuronal plasticity through the NMDAR-GluN2B-CaMKII-Rac1-PAK1-dependent pathway. Together, these data may contribute to the neural repair mechanisms occurring in the PbTx-2-treated post-ischemic stroke mouse model that corresponds with improvements in motor function [7], suggesting a novel pharmacological strategy in the treatment of neurodegenerative and neurological disorders.

Reagents
Trypsin, penicillin, streptomycin, heat-inactivated fetal bovine serum, horse serum, and soybean trypsin inhibitor were obtained from Atlanta Biologicals

Cerebrocortical Neuron Culture
Primary cultures of cerebrocortical neurons were harvested from embryos of Swiss-Webster mice on embryonic day 16 and cultured as described previously [2]. Cells were plated onto poly-l-lysine-coated (Sigma-Aldrich, St. Louis, MO, USA) 96-well clearbottomed, black-well culture plates (MidSci, St. Louis, MO, USA) at a density of 1.5 × 10 5 cells/mL (150 µL/well), 24-well (15.6 mm) culture plates at a density of 0.05 × 10 6 cells/mL (0.5 mL/well), or 6-well (35 mm) culture dishes at a density of 2.25 × 10 6 cells/mL (2 mL/well), respectively, and incubated at 37 • C and 95% humid atmosphere in 5% CO 2 . Cytosine arabinoside (10 µM) was added to the culture medium on day 1 after plating to prevent the proliferation of non-neuronal cells. The culture media was changed on days 4 and 7 using a serum-free growth medium containing Neurobasal medium supplemented with B-27, 100 IU/mL penicillin, 0.1 mg/mL streptomycin, and 0.2 mM L-glutamine. All animal-use protocols were approved by the Creighton University Institutional Animal Care and Use Committee (IACUC).

Intracellular Ca 2+ Monitoring
Cerebrocortical neurons grown in 96-well plates were used for intracellular Ca 2+ concentration ([Ca 2+ ] i ) measurements as previously described [6]. Briefly, the growth medium was removed and replaced with dye loading medium (100 µL per well) containing 4 µM Fluo-3 AM and 0.04% pluronic acid in Locke's buffer. After 1 h of incubation in the dye loading medium, the neurons were washed four times in fresh Locke's buffer (150 µL per well, 22 • C) using an automated microplate washer (Bio-Tek Instruments Inc., Winooski, VT, USA) and transferred to a FLEX Station™ II (Molecular Devices, Sunnyvale, CA, USA) benchtop scanning fluorometer chamber. Fluorescence measurements were carried out at 37 • C. The neurons were excited at 488 nm, and a Ca 2+ -bound Fluo-3 emission was recorded at 538 nm at 1.2 s intervals. After recording baseline fluorescence for 60 s, 50 µL of a 4X concentration of NMDA and PbTx-2 or 50 µL of a mixture of antagonist and agonist was added to the cells at a rate of 26 µL/s, yielding a final volume of 200 µL/well; the fluorescence was monitored for an additional 140-240 s. The Fluo-3 fluorescence was expressed as (F max −F 0 ), where F max is the maximum and F 0 the baseline fluorescence measured in each well.

Determination of Total Neurite Length and Diolistic Labeling
Cells were plated on poly-lysine-coated 12 mm glass coverslips (Thermo Fisher Scientific, Waltham, MA, USA) and placed inside 24-well culture plates at a low density of 0.05 × 10 6 cells/mL (0.5 mL/well). To assess the influence of PbTx-2 on neuritogenesis, primary cultures of immature cerebrocortical neurons were exposed to 30 nM PbTx-2 for 24 h, beginning 3 h after plating, and total neurite outgrowth was measured. In some experiments, 30 nM PbTx-2 were co-incubated with ifenprodil, NSC 23766, Y27632, and IPA-3 (Sigma-Aldrich, St. Louis, MO, USA). At 24 h after plating, cultures were fixed at room temperature for 15 min using 1.8% paraformaldehyde in phosphate-buffered saline (PBS). After fixation, neurons were diolistically labeled with DiI [115]. The dye particles were allowed to spread across the neuronal membrane overnight, and coverslips were then mounted for imaging on an Olympus IX 71 inverted microscope with a Himamatsu camera. Digital images of individual neurons were captured, and total neurite length was quantified [64]. To reduce the effect of paracrine neurotrophic factors on neurite growth, only those neurons that were separated from surrounding cells by approximately 150 µm were digitally acquired and analyzed. Digital images of individual neurons were captured and exported as 16-bit images. All neurites in a single neuron, including those from secondary branches, were semi-automatically traced, and the length was measured by using the using Filament Tracer module of IMARIS 6.4.0 software (Bitplane, South Windsor, CT, USA). At least 25-30 randomly chosen neurons from two different cultures were evaluated for each treatment group.

Plasmids and Nucleofection
The PAK1 pSUPER-GFP plasmid and mutated control PAK1 pSUPER-GFP plasmid [73] were generous gifts from Dianqing Wu (University of Connecticut, Farmington). The primary cultures of immature cerebrocortical neurons were transfected with Lipofectamine 2000 (Life Technologies, Grand Island, NY, USA). Dissociated cortical neurons obtained from E16 pups were plated at a density of 0.5 × 10 6 neurons/well. Two hours post-plating, cells were transfected with 1.0 µg of plasmids containing the gene of interest or mutated control. Three hours post-transfection, cells were treated with 30 nM PbTx-2 or vehicle control. In order to give more time for the expression of genes of interest and to access the influence of PbTx-2 on neuritogenesis, DIV-2 neurons were imaged in experiments involving transfection.

Measurement of Dendritic Complexity
Neurons grown on poly-D-lysine-coated glass coverslips placed inside 24-well culture plates (0.1 × 10 6 cells per well) were used. To assess the influence of PbTx-2 and PAK1 on dendritic complexity, primary cultures of immature cerebrocortical neurons were exposed to either 30 nM PbTx-2, 1 µM IPA-3, or both for 5 days, beginning 3 h after plating. At 5 days after plating, cultures were fixed at room temperature for 15 min using 1.8% paraformaldehyde in phosphate-buffered saline (PBS). After fixation, neurons were diolistically labeled with DiI [115]. Z-stacked images were acquired using an Olympus spinning-disk confocal microscope, and each neuron was scanned at 0.2 µm intervals along the z-axis with a depth of 5 µm (25 planes). For quantitative analysis, a 3D perspective was rendered by the surpass module of IMARIS software (Bitplane, South Windsor, CT, USA). For the analysis of dendritic arbor complexity, the dendritic tracings were quantified by an automated 3D Sholl analysis [64].

Western Blot
Western blot analysis was performed by using cells grown in six-well plates. For acute experiments, DIV-1 cells were exposed to 30 nM PbTx-2 for 30 min at 37 • C. For pharmacological experiments, cultures were pre-incubated either in the presence or absence of specific antagonists or vehicle for 15 min. At the end of the time period, cultures were transferred onto ice slurry to terminate drug exposure and washed three times with icecold PBS. Cells were lysed using ice-cold lysis buffer (50 mM Tris, 50 mM NaCl, 2 mM EDTA, 2 mM EGTA, 1% Nonidet P40, 0.1% SDS, 2.5 mM sodium pyrophosphate, and 1 mM sodium orthovanadate). Phenylmethylsulfonyl fluoride (1 mM) and 1× protease inhibitor mixture (Sigma-Aldrich, St. Louis, MO, USA) were then added, and the lysate was incubated for 30 min at 4 • C. Cell lysates were sonicated and then centrifuged at 13,000× g for 15 min at 4 • C. The supernatant was assayed by the Bradford method [116] to determine protein content. Equal amounts of protein were mixed with the Laemmli sample buffer and heated for 5 min at 75 • C. The samples were loaded onto a 10% SDS-polyacrylamide gel electrophoresis gel, transferred to a PVDF membrane, and immuoblotted with specific antibodies. Blots were developed with a Pierce ECL kit (Thermo Fisher Scientific, Rockford, IL, USA) for 2 min. Blots were subsequently stripped (63 mM Tris base, 70 mM SDS, 0.0007% 2-mercaptoethanol, pH 6.8) and reprobed for further use. Western blot densitometry data were obtained using Image J software (NIH, http://imagej.nih.gov/ij/).

Glutamate Release Assay
For the in vitro glutamate release assay, cerebrocortical neurons were grown for 5 days in 6-well plates as described previously [64]. After 5 days in culture, the cells were washed (2×) in Locke's buffer and incubated in Locke's for 30 min. Cells were then treated with either vehicle or PbTx-2 (30 nM) for an additional 30 min. Supernatant was collected and immediately processed for glutamate. The glutamate concentration was determined by an enzyme-linked immunosorbent assay (ELISA; Labor Diagnostika Nord, Nordhorn, Germany) in accordance with the manufacturer's instructions. Samples were analyzed in duplicate on a plate reader (Bio-Tek, Winooski, VT, USA) by measuring the absorbance at 450 nm.

F-Actin Staining
Cells were plated on poly-lysine-coated 12-mm glass coverslips (Thermo Fisher Scientific, Waltham, MA, USA) and placed inside 24-well culture plates at a low density of 0.05 × 10 6 cells/mL (0.5 mL/well). To assess the influence of PbTx-2 and PAK1 on cytoskeleton changes, primary cultures of immature cerebrocortical neurons were exposed to either 30 nM PbTx-2, 1 µM IPA-3, or both for 24 h, beginning 3 h after plating, and F-actin visualization was performed using an F-actin visualization kit (Cytoskeleton Inc., Denver, CO, USA). The DIV-1 cultured cortical neurons were fixed, and F-actin was labeled strictly following the manufacturer's protocol. Imaging was performed on an Olympus IX 71 inverted microscope with a Himamatsu camera. The density of individual neurons was quantified using Image J software (NIH, http://imagej.nih.gov/ij/).

Statistical Analyses and Graphical Illustration
Data were analyzed and graphical illustrations were generated using GraphPad Prism (La Jolla, CA, USA). Calcium influx analyses were plotted as fluorescence reads over time. The AUC was plotted, and a one-way ANOVA with Dunnet's multiple comparisons was used to determine statistical significance. Neurite outgrowth was quantified using IMARIS software and then plotted in GraphPad Prism. A one-way ANOVA with Dunnet's multiple comparisons was used to determine statistical significance, with comparisons to both the positive and negative controls denoted. For mESPCs, a histogram summary was used, showing the mean ± SEM.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/md20090559/s1. Figure S1: Area under the curve (AUC) was assessed and a one-way ANOVA with multiple comparisons were performed to determine statistical significance for the effects of pretreatment with MK-801, TC-201, ifenprodil, and QNZ-46 on NMDAinduced Ca 2+ influx. Figure S2: AUC was assessed and a one-way ANOVA with multiple comparisons were performed to determine statistical significance for the effects of pretreatment with nifedipine and KB-R7943 on NMDA-induced Ca 2+ influx, and nifedipine, KB-R7943, and MK-801 on PbTx-2induced Ca 2+ influx. Figure S3. Raw data for p-PAK Western Blot. Figure S4. Raw data for p-cofilin Western Blot.