Inducible Nitric Oxide Synthase Inhibitor, 1400W, Mitigates DFP-induced Long-term Neurotoxicity in the Rat Model

Abstract

Chemical nerve agents (CNA) are increasingly becoming a threat to both civilians and military personnel. CNA-induced acute effects on the nervous system have been known for some time and the long-term consequences are beginning to emerge. In this study, we used diisopropylfluorophosphate (DFP), a seizurogenic CNA to investigate the long-term impact of its acute exposure on the brain and its mitigation by an inducible nitric oxide synthase (iNOS) inhibitor, 1400W as a neuroprotectant in the rat model. Several experimental studies have demonstrated that DFP-induced seizures and/or status epilepticus (SE) causes permanent brain injury, even after the countermeasure medication (atropine, oxime, and diazepam). In the present study, DFP-induced SE caused a significant increase in iNOS and 3-nitrotyrosine (3-NT) at 24h, 48h, 7d, and persisted for a long-term (12 weeks post-exposure), which led to the hypothesis that iNOS is a potential therapeutic target in DFP-induced brain injury. To test the hypothesis, we administered 1400W (20 mg/kg, i.m.) or the vehicle twice daily for the first three days of post-exposure. 1400W significantly reduced DFP-induced iNOS and 3-NT upregulation in the hippocampus and piriform cortex, and the serum nitrite levels at 24h post-exposure. 1400W also prevented DFP-induced mortality in <24h. The brain immunohistochemistry (IHC) at 7d post-exposure revealed a significant reduction in gliosis and neurodegeneration (NeuN+ FJB positive cells) in the 1400W-treated group. 1400W, in contrast to the vehicle, caused a significant reduction in the epileptiform spiking and spontaneous recurrent seizures (SRS) during 12 weeks of continuous video-EEG study. IHC of brain sections from the same animals revealed a significant reduction in reactive gliosis (both microgliosis and astrogliosis) and neurodegeneration across various brain regions in the 1400W-treated group when compared to the vehicle-treated group. A multiplex assay from hippocampal lysates at 6 weeks post-exposure showed a significant increase in several key pro-inflammatory cytokines/chemokines such as IL-1α, TNFα, IL-1β, IL-2, IL-6, IL-12, IL-17a, MCP-1, LIX, and Eotaxin, and a growth factor, VEGF in the vehicle-treated animals. 1400W significantly suppressed IL-1α, TNFα, IL-2, IL-12, and MCP-1 levels. It also suppressed DFP-induced serum nitrite levels at 6 weeks post-exposure. In the Morris water maze, the vehicle-treated animals spent significantly less time in the target quadrant in a probe trial at 9d post-exposure compared to their time spent in the same quadrant 11 days previously (i.e., 2 days prior to DFP exposure). Such difference was not observed in the 1400W and control groups. However, learning and short-term memory were unaffected when tested at 10–16d and 28–34d post-exposure. Accelerated rotarod, horizontal bar test, and the forced swim test revealed no significant changes between groups. Overall, the findings from this study suggest that 1400W may be considered as a potential therapeutic agent as a follow-on therapy for CNA exposure, after controlling the acute symptoms, to prevent mortality and some of the long-term neurotoxicity parameters such as epileptiform spiking, SRS, neurodegeneration, reactive gliosis in some brain regions, and certain key proinflammatory cytokines and chemokine.

Graphical abstract

This study was designed based on the hypothesis that DFP exposure induces hyperexcitability of neurons to cause seizures/SE and activates glia to produce excessive inducible nitric oxide synthase (iNOS/NOS-II) and proinflammatory cytokines and chemokines to cause long lasting brain injury. Therefore, inhibiting iNOS with a highly selective pharmacological inhibitor, 1400W, we demonstrate a significant suppression/reduction in: iNOS, 3-NT, and serum nitrite at 24h; mortality in <24h post-exposure; gliosis and neurodegeneration at both 7d and 12 weeks in certain brain regions; key proinflammatory cytokines and chemokine in the hippocampus, and; SRS and epileptiform spikes during 12 weeks of continuous monitoring.

1. Introduction

The organophosphate (OP) chemical nerve agents (CNA) such as soman, sarin, and VX are considered a threat to the public and military personnel by the US government (Coupland and Leins, 2005; Jett, 2012, 2010; Jett and Spriggs, 2018; Stone, 2018). CNAs are likely to be abused to attack civilians (Jett and Yeung, 2010; OPCW Report, 2017; Therrien and Roxby, 2018; UN Report, 2013; Zarocostas, 2017). CNA attacks in Tokyo, Syria, Kuala Lumpur, and England suggest increased likelihood of future CNA threat elsewhere in the world (Chai et al., 2018, 2017; Doyle, 2017; Okumura et al., 2005; Rosman et al., 2014; Tu, 2007; Vale et al., 2018). Acute CNA exposure at high concentrations is likely to cause death but sublethal exposure has the potential to precipitate seizures of varying degree, which can cause long-lasting or permanent brain injury in both humans and experimental models (Figueiredo et al., 2018; Miyaki et al., 2005; Morita et al., 1995; Murata et al., 1997; Nishiwaki et al., 2001; Okumura et al., 1996; Pouliot et al., 2016; Suzuki et al., 1995; Yanagisawa et al., 2006). The survivors of the past and recent CNA attacks in the affected demographic areas are likely to develop cognitive dysfunction and psychiatric disorders (Dolgin, 2013; Murata et al., 1997; Rosman et al., 2014; Yanagisawa et al., 2006).

OP CNAs are potent seizurogenic neurotoxins and they irreversibly inhibit acetylcholinesterase (AChE) to cause cholinergic toxidrome (Jett, 2012; Jett and Spriggs, 2018; Millard et al., 1999). CNA exposure in rats and mice have provided evidence for long-term consequences of neurological problems such as the development of epilepsy, cognitive dysfunction, and depression-like behavior despite treating the acute symptoms with medical countermeasures (MCM) (de Araujo Furtado et al., 2010; Flannery et al., 2016; Reddy and Kuruba, 2013; Wright et al., 2010). Reports on the long-term consequences of CNA exposure in humans are only beginning to emerge in recent years (Dolgin, 2013; Miyaki et al., 2005; Morita et al., 1995; Murata et al., 1997; Nishiwaki et al., 2001; Okumura et al., 1996; Rosman et al., 2014; Suzuki et al., 1995; Yanagisawa et al., 2006).

The MCMs such as atropine, 2-pyridine aldoxime methyl chloride (2-PAM), and diazepam (DZP) or midazolam do control acute clinical signs of CNA exposure in animal models and can prevent the long-term consequences only if administered immediately after the exposure (Kuruba et al., 2018; McDonough et al., 1995; Shrot et al., 2014; Wu et al., 2018). However, if the intervention is delayed, as it would be in cases of chemical warfare in humans, none of these drugs fully prevent mortality and long-term neurotoxicity such as neurodegeneration, spontaneous recurrent seizures (SRS- both convulsive and non-convulsive types), neuroinflammatory responses, and associated behavioral deficits (de Araujo Furtado et al., 2010; Flannery et al., 2016; Kadriu et al., 2011; Sisó et al., 2017; Wu et al., 2018). Moreover, currently, there are no drugs that can effectively prevent CNA-induced long-term changes if the treatment is delayed (Jett and Spriggs, 2018; Jett and Yeung, 2010). There is an urgent need for investigating an effective countermeasure for the long-term neuroprotection when a drug is administered as a follow-on therapy after a gap of several hours of CNA exposure to target the subjects after-field evacuation and in-hospital scenarios (Jett and Spriggs, 2018). Therefore, alongside the MCM, we have tested a novel neuroprotectant, 1400W, an inducible nitric oxide synthase (iNOS/NOS-II) inhibitor as a follow-on therapy two hours after administering diazepam in the rat diisopropylfluorophosphate (DFP) model.

DFP is a CNA and was considered a chemical weapon during World War II (Jett and Spriggs, 2018; Renshaw, 1946; Sisó et al., 2017). Seizure induction and the subsequent brain pathology caused by acute DFP exposure are somewhat similar to those caused by other CNAs, OP pesticides, and other chemoconvulsants (Curia et al., 2008; de Araujo Furtado et al., 2010; Deshpande et al., 2014; Flannery et al., 2016; Sharma et al., 2018b; Todorovic et al., 2012; Zolkowska et al., 2012). DFP, like other CNAs, irreversibly inhibits AChE, which is clinically manifested as neurological, respiratory, and cardiac symptoms in addition to seizures or status epilepticus (SE) (Lallement et al., 2002; Lemercier et al., 1983; Miller et al., 1993; Peter et al., 2014). SE-induced brain injury, irrespective of causative agent, increases the production of reactive oxygen/nitrogen species (ROS/RNS), hyperexcitability of neurons (evident from epileptiform activity), neuroinflammation, and neurodegeneration (Kim et al., 1999; Kuruba et al., 2018; Li et al., 2011; Pauletti et al., 2017; Pearson and Patel, 2016; Pearson-Smith et al., 2017; Sharma et al., 2018a; Sisó et al., 2017; Vanova et al., 2018; Zaja-Milatovic et al., 2009).

We have previously demonstrated the upregulation of both neuronal NOS (nNOS) and iNOS in the hippocampus after the induction of SE, and the beneficial effects of selective NOS inhibitors in the rodent kainate (KA) models of epilepsy (Beamer et al., 2012; Cosgrave et al., 2008; Puttachary et al., 2016b). Importantly, nNOS upregulation in neurons is transient in response to SE while iNOS upregulation persists in glia during both early and late stages of epileptogenesis and in chronic epilepsy (Beamer et al., 2012; Cosgrave et al., 2008; De Simoni et al., 2000). Therefore, we hypothesized that targeting iNOS soon after CNA exposure could be therapeutically beneficial. Indeed, we had demonstrated a significant modification in epileptogenesis in the rat KA model of chronic epilepsy when 1400W, a highly selective iNOS inhibitor, was administered at 20 mg/kg twice daily for the first 3 days of post-SE (Puttachary et al., 2016b). Based on this premise, we tested 1400W in the rat DFP-induced long-term neurotoxicity model. In this study, we demonstrate that 1400W prevents mortality and/or modifies the long-term DFP-induced neurotoxicity by suppressing epileptiform spiking, SRS, gliosis, some key proinflammatory cytokines and chemokine (IL-1α, TNFα, IL-2, IL-12, and MCP-1), and neurodegeneration.

2. Materials and Methods

2.1. Animal source, care, and ethics statement

Young adult male Sprague Dawley rats (8 weeks old) were purchased from Charles River (MA, USA) and maintained in the Laboratory of Animal Resources, Iowa State University (ISU) under controlled environment (19°C–23°C, 12 h light: 12 h dark), with ad libitum access to food and water. Animals were used for experiments after four days of quarantine and experiments were conducted as per the approved protocols by the ISU Institutional Animal Care and Use Committee (IACUC; protocol numbers 18–159 and 18–160). All surgical procedures were carried out in sterile and aseptic conditions under general anesthesia. Pre- and post-operative care were given to all animals to minimize pain and discomfort. Animals were monitored and weighed daily after surgery and after exposure to DFP. At the end of each experiment, the animals were euthanized by intraperitoneal (i.p.) administration of 100 mg/kg pentobarbital sodium as per the recommendations of the American Veterinary Medical Association Guidelines for Euthanasia of Animals. All experiments conducted in this study complies with ARRIVE guidelines (Kilkenny et al., 2010). The animals were housed individually throughout the course of the experiment.

2.2. Chemicals and reagents, and authentication

DFP (93% pure, Sigma Aldrich, USA) was purchased as a solution (1g/L) and immediately aliquoted in small quantities and stored at −80°C as its stability decreases over time even at 4°C (Heiss et al., 2016). Before used in animals, DFP was always diluted in cold 0.1M phosphate buffer saline (PBS), at pH 7.2. Atropine sulfate and 2-PAM were purchased from Acros Organics and Sigma Aldrich, USA, respectively. 1400W (99.6% pure, Tocris Bioscience, USA) was dissolved in sterile distilled water (DW) at a concentration of 10 mg/mL. For perfusion, 4% paraformaldehyde (PFA) solution (Acros Organics, USA) in PBS at pH 7.2 was used. The primary antibodies and the concentrations used for IHC and/or the Western blot (WB) were as follows: Ionized calcium-binding adaptor molecule 1 (IBA1) (goat polyclonal, 1:500 for IHC, Abcam, MA, USA); β-actin (rabbit polyclonal, 1:10,000 for WB, Sigma Aldrich, USA); NeuN (rabbit polyclonal, 1:400 for IHC, EMD Millipore, USA); iNOS (rabbit polyclonal,1:100 for IHC, Abcam, MA, USA); iNOS (rabbit polyclonal, 1:500 for WB, Santa Cruz, USA); glial fibrillary acidic protein (GFAP) (mouse monoclonal, 1:400 for IHC, Sigma Aldrich, MO, USA); CD68 (rabbit polyclonal, 1:400 for IHC, Abcam, MA, USA); S-100β (rabbit polyclonal, 1:400 for IHC, Sigma Aldrich, MO, USA; 3-nitrotyrosine (3-NT) (mouse monoclonal, 1:1000 for WB, Abcam, MA, USA). Fluoro-Jade B (FJB) was purchased from Histochem Inc., Jefferson, AR, USA. The secondary antibodies tagged with either a fluorescent dye (CY3 conjugated 1:200 or FITC conjugated 1:80) or biotin (1:400) were used for IHC. These were purchased from Jackson ImmunoResearch Laboratories, PA, USA. All primary and secondary antibodies were prepared in 2.5% donkey serum (neutral species) to prevent cross-reactivity, 0.25% sodium azide and 0.1% Triton in 0.1M PBS. Streptavidin conjugates were diluted in PBS alone. For WB, IRDye 680LT and 800CW donkey anti-goat or anti-mouse or anti-rabbit secondary antibodies (LI-COR Biosciences, NE, USA) were used at the dilution of 1:10,000. Appropriate neutralizing IgGs for primary antibodies were purchased from the same source as the primary antibodies. The Bradford protein assay kit was purchased from Bio-Rad, CA, USA for protein estimation. We used bead-based rat multiplex assay (Rat Cytokine Chemokine Array 27 Plex (RD27) from Eve Technologies, Canada) for hippocampal lysates, serum and CSF samples. For nitrite assays, we used the Griess reagent kit from Sigma Aldrich, USA. In all assays, appropriate positive and negative controls were used.

The DFP and 1400W chemical identity and purity were determined by GS-MS and UHPLC-MS, respectively (Fig. 1S, 2S). Before 1400W was used in the animals, its bioactivity was confirmed in the lipopolysaccharide (LPS)-stimulated microglia cell line by determining the suppression of LPS-induced nitrite levels in the media. The methodological details of cell culture, cell line, and drugs treatment are included in supplementary figure (Fig. 3S). Nitrite assay is described in section 2.11.

2.3. Experimental groups, drug treatment, and euthanasia

We used 101 male rats in this study. The animals were randomized, grouped, and coded before they were used in the experiments. However, the animals were not further randomized, but the groups were blinded, after exposure to DFP. The rats were distributed into three groups; naïve control (referred as control in the figures), DFP+ vehicle, and DFP+ 1400W. The number of animals used in each group is given in Table 1. The naïve control group received nothing. The dose and the route of DFP and MCM administration were as follows: DFP at 4 mg/kg in 300–400 μL of cold PBS, s.c. (telemetered rats received 3 mg/kg), immediately followed by the administration of atropine sulfate (ATS, 2 mg/kg, i.m), and 2-PAM (25 mg/kg, i.m) to counteract the peripheral effects of excess acetylcholine and to reactivate AChE (Kim et al., 1999). Diazepam (5 mg/kg, i.m.) was administered at 2h after the first onset of convulsive seizure to prevent SE-induced death. DFP alone without MCM has been shown to cause >50% mortality (Deshpande et al., 2010), therefore, DFP alone without MCM treatment as an additional control was not considered. Two hours after the administration of diazepam, the rats were treated with 1400W (20 mg/kg, i.m; the required volume, approximately 400 μL, was split between the two hind limbs) or the equal volume of the vehicle twice daily (i.e., at 12h intervals) for the first three days only. However, in the 24h post-exposure group, two doses of 1400W or the vehicle (at 12h interval) were administered to investigate the effects of iNOS inhibition on 3-NT levels and iNOS itself. Tissues were collected approximately 8 hours after the second injection. To understand whether the iNOS and 3-NT levels also persisted beyond 24h, some animals (n=4) treated only with DFP and MCM were euthanized at 48h and 7d post-exposure. The remaining non-telemetry animals, treated with DFP and vehicle or 1400W, were euthanized at 7 days to determine gliosis and neurodegeneration or used for the behavioral tests and euthanized at 6 weeks. The tissues from 6 weeks were used for the multiplex assay. In telemetered rats, DFP was administered at 10 days post-surgery, subjected to continuous video-EEG monitoring, and euthanized at 12 weeks post-DFP. All the DFP exposed animals received 1 mL of Ringer’s lactate solution (s.c.) twice daily for the first three days to facilitate recovery from DFP-induced dehydration and to gain weight.

Table 1.

The number of animals used in different experiments. Fourteen animals died during the course of the experiment are not included in the table.

Animal Groups	Naïve Control	DFP+ Vehicle	DFP+ 1400W
Telemetry (12 weeks)	6	8	8
Non telemetry groups
24h	6	4	5
48h		4	0
7d		10	6
6 weeks	6	12	12
Total	18	38	31

2.4. Quantification of the severity of DFP-induced SE and further grouping of animals for 1400W or vehicle treatment

Following administration of DFP, animals were directly observed and scored for SE severity by an experimenter. Video recordings were used for a secondary validation as in our previous studies in the rat KA models (Puttachary et al., 2016b; Sharma et al., 2018a). A different set of criteria than the KA model were used to determine the stages of SE in the DFP model, somewhat similar to studies from other laboratories (for example, (Sisó et al., 2017). The stages were classified as follows: stage 1- excessive salivation, lacrimation, urination and defecation (SLUD), mastication, chewing; stage 2- tremors, wet-dog shakes, head nodding, neck jerks, kyphosis, and opisthotonus; stage 3- forelimb clonus, Straub tail, rearing and rigid extension of forelimbs; stage 4- rearing, forelimb clonus and loss of righting reflex; and stage 5- abducted limbs clonus/repeated rearing and generalized seizures. An example of the behavioral and EEG correlates of different stages of DFP-induced SE and the corresponding power spectrum are illustrated in Figure 1 (A–D). Stage 1 and 2 were considered non-convulsive seizures (NCS) and ≥ stage 3 were considered convulsive seizures (CS). We calculated the duration of CS (≥3 stage) in each animal to further classify the severity of SE as mild (10–15 min), moderate (15–30 min), and severe (>30 min) (Table 2). We grouped the animals in such a way that both DFP+ vehicle and DFP+ 1400W groups had almost equal time spent in CS during SE across each group (Fig. 1E, 1F). Therefore, after exposure to DFP, the animals were not randomized. It is important to note that in this study the mild, moderate, and severe terminologies refer to the duration of CS and not the stages of SE.

Figure 1.

Overview of the integrated video-EEG telemetry and the comparison of the severity of DFP-induced SE response between groups prior to the vehicle or 1400W treatment. A-D: The behavioral seizure correlates of EEG are represented in panels A and B. The numbers from 1 to 5 in panel A represent the stages of SE. The SE stages considered were: Stage 1, SLUD, mastication, chewing; Stage 2, tremors, wet-dog shakes, head nodding, neck jerks, kyphosis, and opisthotonus; Stage 3, forelimb clonus, Straub tail, rearing and rigid extension of forelimbs; Stage 4, rearing, forelimb clonus and loss of righting reflex; Stage 5, abducted limbs clonus or repeated rearing or falling, with generalized seizures. A representative 9 minutes EEG trace (C-ii) and the corresponding power spectrum (C-i), and the expanded EEG traces (C- iii to v) are illustrated. In panel D, an expanded segment of the stage 5 power spectrum (0–80 Hz) at 10s epochs is shown. E-F: Initial SE severity and grouping were based on the duration of CS, during the 2h SE, for the vehicle or 1400W treatment. Panel E shows the average stage of behavioral seizures over the period of SE. Panel F shows the average number of minutes rats spent in a convulsive seizure (CS; ≥ stage 3) during SE. There were no significant differences in the initial SE severity between the two groups. Data presented as a group mean ±SEM, ns= non-significant, n=8 per group.

Table 2.

The response to DFP exposure. The duration of CS is calculated based on the amount of time the animals spent in ≥ stage 3 during the 2h between DFP and diazepam injections. Seizures severity presented as mean±SEM. *p<0.05 vs. DFP+Vehicle (Fisher Exact’s test)

Experimental groups	% NCS	% Convulsive seizures			% Mortality
		10–15 min (mild)	15–30 min (moderate)	>30 min (severe)
Telemetry	4	30	9	57	30.4
Non telemetry	14	16	9	61	12
Non telemetry	CS >90 min	Treatments
		DFP+Vehicle		DFP+1400W
	Seizures Severity (CS in min)	110.3±3.19		100±7.497
	24h Mortality (%)	6/20 (33%)		0/14 (0%) *

2.5. Transmitter device implantation and video-EEG acquisition

Initially, 23 rats were implanted with the CTA-F40 PhysioTel™ telemetry device (Data Science International, Minneapolis, USA) for video-EEG acquisition, 10 days prior to DFP exposure. Initially, we tested 4 mg/kg of DFP which caused high mortality during SE or in less than 3 day of post-exposure, therefore, we reduced the dose to 3 mg/kg. The animals were administered with analgesic buprenorphine (0.3 mg/kg, s.c.) as a pre-operative medication prior to the induction of anesthesia with 4.0% isoflurane (flow rate at 1L/min O₂) and maintained at 2.25% during the surgical procedure, which was performed in a sterile condition as described in our previous publications (Puttachary et al., 2016b; Sharma et al., 2018a). To prevent dry eyes and corneal ulceration, eyes were lubricated with artificial tears ointment during surgery. The electrodes were placed on the surface of the dura mater, overlying the cortical hemispheres, and the telemetry device was placed in the subcutaneous pouch. The electrodes were secured to the skull with dental cement (A-M systems, WA, USA) and the incision was closed with sterile surgical clips. The triple antibiotic ointment, Vetropolycin, was applied to the surgical site. Systemic antibiotic, Baytril (5 mg/kg, s.c., Bayer Pharma, PA), and 1 mL of normal dextrose saline were administered subcutaneously before the animals recovered from anesthesia. The animals were individually caged and placed on PhysiolTel receivers (RPC-1) connected to the Data Exchange Matrix (DSI Dataquest A.R.T. system) for continuous video-EEG acquisition. During the 10 days period of post-surgery, EEG was recorded to cover both day and night cycles to evaluate the impact of surgery on spontaneous epileptiform spiking activity as well as to generate a baseline for each animal. The telemetry device has a sensor to record body temperature and locomotor activity. Out of the 16 telemetry implanted rats that received 3 mg/kg DFP, 8 rats served as vehicle control and the other 8 rats were treated with 1400W. The rats were subjected to continuous video-EEG monitoring for 12 weeks following DFP exposure. Epileptiform spikes and SRS were quantified to determine the impact of 1400W as a follow-on treatment.

2.6. Quantification of epileptiform spikes and SRS

Baseline EEG was used to normalize post-DFP EEG for accurate detection of epileptiform spiking activity. Electrical noise, exploratory behavior, and grooming were identified and excluded from epileptiform spike analysis as described previously for the rat and mouse KA models (Puttachary et al., 2015, 2016b; Sharma et al., 2018a; Tse et al., 2014). The epileptiform spikes quantified included the spikes in the spike trains and electrographic NCS. The spike trains consisted of both spike clusters and individual epileptiform spikes, including isolated pre-ictal and inter-ictal spikes. All epileptiform spikes were identified using the NeuroScore 3.2.0 software and were summed across animal groups at different time points to evaluate the effects of 1400W. SRS was identified on the EEG as high amplitude and high-frequency spikes and was verified against the integrated real-time video and power spectrum, which was generated via fast Fourier transformation. Episodes of SRS were quantified as described in our previous publications (Puttachary et al., 2015, 2016b; Sharma et al., 2018a).

2.7. Behavioral tests

The timeline for all behavioral tests is illustrated in Figure 10A.

Figure 10.

Impact of DFP and vehicle or 1400W on learning and memory in the MWM, motor function in the rotarod and horizontal bar test (HBT), and depression-like behavior in the forced swim test (FST). A, B: An overview of the experimental design for all four neurobehavioral tests (A) and a detailed plan for MWM (B) are illustrated. There were no differences in spatial learning and memory between groups in pre-DFP testing paradigm, at 10–15d (post-treatment 1) or at 28–33d (post-treatment 2) post-DFP exposure (C). Single probe trials conducted 24h after the 6 days of learning sessions did not reveal any changes in the short-term spatial memory between the groups in pre-DFP (Di), at 16d (Diii) or at 34d post-DFP (Dv). We observed a similar group results for the probe trials conducted at 9d (Dii) and 27d (Div) post-DFP (i.e., 11 days after the previous probe trial). However, there was a significant reduction in the time spent in the target quadrant in the vehicle group at 9d post-DFP when compared to the probe trial conducted prior to DFP exposure (Ei). This was not significant in control or the 1400W group (Ei). A similar comparison between the two probe trials conducted at 16d and 27d post-DFP (Diii and Div) did not reveal significant differences in the vehicle or 1400W group but the control group spent significantly less time in the opposite quadrant at 27d trial than the previous trial (Eii). No significant differences were observed between the groups in the rotarod (F), HBT (G) or FST (H). The animals in control group stayed on the rotarod significantly longer on the test day compared to the training day in their first round of training (pre-DFP) (F). In the HBT, the control and the vehicle groups showed a significant reduction in the time taken to reach the target in both pre- and post-exposure. The 1400W group showed a similar trend but the differences were not significant (G). In the FST, there were no significant differences although the vehicle and 1400W groups were immobile for longer time than the control group in post-DFP test (H). *p<0.05, **p<0.01, ***p<0.001 with respect to the first training day. n=6 control, 12 each in the vehicle and 1400W groups. NE, Northeast; NW, Northwest; SW, Southwest. D= day; Pre-Treatment = Pre-DFP. Also, see Videos.1S for the MWM videos.

2.7.1. Morris water maze (MWM)

The experimental design for MWM is illustrated in figure 10A and 10B. A PowerPoint slide containing MWM videos is included in the supplementary material (Video. 1S). We used a large aluminum water tank of an appropriate size (diameter, 180 cm; height, 60 cm) suitable for rats. The internal surface of the tank was coated with black paint to track and acquire the movement of rat by the camera operated by ANY-maze software (Stoeling Co., USA). The temperature of the tank water was maintained at 21–23°C. The tank was divided into four quadrants by imaginary (not based on compass) North (N)-South (S) and East (E)-West (W) longitudinal axes across the tank, and the quadrants were accordingly named as Northeast (NE), Northwest (NW), Southeast (SE), and Southwest (SW). The learning task was to find a camouflaged submerged platform (diameter, 12 cm; height, 32 cm; placed at the middle of one of the four quadrants) during the training (spatial learning) period followed by a single probe trial following 24h after the last training day to test the spatial memory. Black tempera paint was added to the water to camouflage the transparent platform. The rationale is that animals use distal cues to navigate to find the hidden platform. Decreased time or distance traveled to reach the platform when started at any point at the periphery of the tank was considered evidence of spatial learning. The amount of time taken to reach the target (hidden platform during training period) or the time spent at the target quadrant (platform removed in probe trial) was quantified. Four trials per day every 20–25 minutes for six days followed by a single probe trial on the 7th day were performed. The animals were handled for a while before gently placed in the water close to the periphery of the tank. The animals were dried after each test and left in the home cage until the next trial. All the tests were conducted at the same time of the day (8–11 am) in the same order by the same group of experimenters. The rats were trained with the platform visible above the water on the first day of training followed by 5 days of learning with a submerged platform. To determine whether the animals used in the experiment are normal with respect to spatial learning and memory, we acquired the baseline data with the target in NE quadrant prior to DFP exposure. At 9 days post-DFP, a second probe trial was conducted with the same target (i.e., NE) to test the impact of DFP and 1400W or vehicle on the long-term memory (i.e. 11 days after the first probe trial). Traditionally, a single probe trial 24h after 5 days of training is considered to be long-term memory (Vorhees and Williams, 2006). In our study, 11 days was defined as an evaluation of long-term memory while 24 hours after the training was defined as an evaluation of short-term memory. Between 10–15d post-DFP, the rats were subjected to a new learning task for six days by placing the platform in the SW quadrant, and the tank surrounded by a new set of cues. A single probe trial was conducted at 24h after the 6th day of spatial learning (i.e., 16d post-DFP) to test the impact of DFP and 1400W/vehicle on the spatial memory. Eleven days later (27d post-DFP, i.e., 11 days after the previous probe trial), a second single probe trial was conducted to test the long-term memory. On day 28, reversal learning task (for 6 days) was introduced with a target in the NW quadrant with new distal cues, including a hanging ball over the platform, followed by a single probe trial after 24h (i.e., 34d post-DFP) to investigate the impact of DFP and 1400W/vehicle at an even later time-point. Of the several parameters, we calculated the time to reach the hidden platform for learning (training) and the time spent in the target quadrant for probe trial (memory) in 60s. Group ±SEM for each trial and all 4 trials for each day were calculated and plotted graphs. For the training periods, we used a two-way ANOVA followed by a Tukey post-hoc analysis to compare between the groups at each training day. For the probe trials, we also used a two-way ANOVA in each quadrant in order to compare between the treatment groups as well as pre and post DFP within each treatment group (see, Table 4).

Table 4.

Details of statistical analyses applied for each experiment in the figures. P values and n numbers are provided in the figure legend.

Fig. #	Panel	Statistical Test
1	E	RM-two-way ANOVA: F(1, 14)=2.145; α =0.05; Sidak’s post-hoc
	F	Unpaired t-test, two-tailed: t=0.9137; df=14; α =0.05
2	E	Unpaired t-test, two-tailed df=6, α =0.05: iNOS 48h (t=4.216); iNOS 7d (t=4.661); 3NT 48h (t=3.36); 3NT 7d (t=6.868)
	F	Unpaired t-test, two-tailed df=6, α =0.05: iNOS 48h (t=4.626); iNOS 7d (t=8.445) 3NT 48h (t=3.481); 3NT 7d (t=3.007)
3	C	RM-two-way ANOVA: F(2, 19)=14.06; α =0.05; Tukey’s post-hoc
	D	RM-two-way ANOVA: F(2, 9)=10.7; α =0.05; Tukey’s post-hoc
4	C	One-way ANOVA; df=2,10; α =0.05; Tukey’s post-hoc; iNOS (F=12.42); 3NT (F=16.13)
	D	One-way ANOVA: df=2,10; α =0.05; Tukey’s post-hoc; iNOS (F=15.07) 3NT (F=11.86)
	E	one-way ANOVA: F(2, 16)=13.76; α =0.05; Tukey’s post-hoc
5	B	one-way ANOVA: df=2,14; α =0.05; Tukey’s post-hoc; IBA1 (F=91.65), GFAP (F=80.16); NEUN+FJB (F=128.1)
6	D	RM-two-way ANOVA: F(1, 12)=3.437; α =0.05; Tukey’s post-hoc
	E	(0–4 weeks) Mann-Whitney test, two-tailed, p-value=0.813 (5–8 weeks) Mann-Whitney test, two-tailed, p-value=0.9452 (9–12 weeks) Unpaired t-test, two-tailed: t=3.79; df=8; α =0.05
	G	RM-two-way ANOVA: F(1, 884)=10.16; α =0.05; Tukey’s post-hoc
	H	Exact Poisson Test; α =0.05
7	C	RM-two-way ANOVA: F(2, 19)=4.753; α =0.05; Tukey’s post-hoc
	D	RM-two-way ANOVA: F(2, 19)=6.679; α =0.05; Tukey’s post-hoc
	E	RM-two-way ANOVA: F(2, 19)=11.28; α =0.05; Tukey’s post-hoc
8	C	RM-two-way ANOVA: F(2, 18)=2.609; α =0.05; Tukey’s post-hoc
	D	RM-two-way ANOVA: F(2, 18)=7.211; α =0.05; Tukey’s post-hoc
9	B	RM-two-way ANOVA: F(2, 17)=16.26; α =0.05; Tukey’s post-hoc
10	C	RM-two-way ANOVA: df=2,23; α =0.05; Tukey’s post-hoc; Pre-Treatment (F=0.5818); Post-Treatment 1 (F=0.2469); Post-Treatment 2 (F=0.9319)
	D, E	RM-two-way ANOVA: df=2,23; α =0.05; Tukey’s post-hoc; (i)pre- vs post-DFP(9d) Opposite (F=0.4594); Target (F=0.9739): (ii) post-DFP(16) vs post-DFP(27d) Opposite (F=2.010); Target (F=0.5272)
	F	RM-two-way ANOVA: df=2,23; α =0.05; Sidak’s post-hoc; Pre (F=1.105); Post (F=0.625)
	G	RM-two-way ANOVA: df=2,23; α =0.05; Tukey’s post-hoc; Pre (F=0.6644); Post (F=0.5897)
	H	RM-two-way ANOVA: F(2, 23)=0.3695; α =0.05; Sidak’s post-hoc
11	A	one-way ANOVA; df=2,12; α =0.05; Tukey’s post-hoc. IL-1α (F=21.17), TNF-α (F=26.16), IL-1β (F=4.976), IL-2 (F=1.62), IL-6 (F=5.553), IL-10 (F=5.815), IL-12(F=11.4), IL-13(F=10.62), IL-17A(F=6.647), IL-18(F=2.128), IL-4 (F=3.837), IL-5 (F=0.335)
	B	one-way ANOVA; df=2,12; α =0.05; Tukey’s post-hoc. MCP-1 (F=16.89), LIX (F=12.09), Eotaxin (F=5.517), VEGF(F=3.997), EGF (F=0.9125), Leptin (F=10.83), MIP-1α (Kruskal-Wallis test, p=10.3563), Fractalkine (F=0.503), IP10 (F=2.425), RANTES (F=0.9415)
	C	one-way ANOVA: F(2, 23)=6.529; α =0.05; Tukey’s post-hoc

2.7.2. Accelerated rotarod Test

The MWM test alone does not distinguish between cognitive and motor deficits. To test motor function, we used accelerated rotarod and Horizontal Bar Tests (HBT). We used a motorized accelerating AccuRotor 4 channel (Omnitech Electronics Inc.) equipment to investigate the motor coordination during speed and acceleration. A grooved cylinder (70 mm diameter) facilitates the animals’ grip while rotating. The fall height was set at 36 cm. The acquisition software Fusion 6.2 was used to acquire the data in Excel Compatible Result File, which automatically stores the results of each trial such as, the speed of the rod, duration of the trial, and trial/batch/subject/profile names. The test was started at 18d post-DFP (i.e., 37 days after the start of the baseline test). We conducted three trials on both training and testing days. The latency to fall on an accelerating rod at 5 rpm to 60 rpm during a 2 minutes trial was measured. Two-way ANOVA with appropriate post-hoc analysis was used to compare treatment groups as well as differences between the training and test days in order to determine if each group could perform better following training (see, Table 4).

2.7.3. Horizontal Bar Test

Horizontal bar test determines static coordination in forelimbs. The test was started at 23d post-DFP (i.e. 36 days after the baseline test). Apparatus consisted of a 100 cm bar suspended about 58 cm from the ground. On one end the rat was placed on the bar, on the other was a secure black box which represented safety for the rats. After two days of training on the horizontal bar, 24h rest was given prior to the testing day. The time taken to reach the target at the end of the bar in 90s was scored. We conducted four trials per day for two days on a 2.5 cm beam (80 cm length) and a single trial on the testing day on 2.5 cm and 2.0 cm beams. Manual timing supplemented by video was used to determine the time taken to reach the target. Two-way ANOVAs were run with appropriate post hoc in order to compare differences between the treatment groups as well as between each day of training/testing (see, Table 4).

2.7.3. The Forced Swim Test (FST)

We utilized the FST to detect depression-like behavior in DFP treated rats and its mitigation by 1400W. The test was conducted in a plastic cylinder (35 cm height × 30 cm diameter) filled with tap water up to 2/3rd of the container. The temperature of the water was maintained at 21–23°C. The test was conducted after a day of rest after the rotarod. In the FST, the rats were allowed to acclimatize in the water for 2 min, and after that, the duration of immobility during a further 4 min swim was measured. We used the Any-maze Video Tracking and data acquisition system (Stoelting Co. USA) to determine the duration of immobility. Two-way ANOVA with appropriate post-hoc was used to determine significant differences between treatment groups and between pre- versus post- DFP to understand the long-term impact of DFP-induced dysfunction and the mitigation by 1400W (see, Table 4).

2.8. Tissue collection and processing, immunohistochemistry, imaging, and cell quantification

The blood and CSF were collected under terminal anesthesia. The blood was processed to separate the serum using the standard protocol from Thermo Fischer Scientific. The serum samples were used for the nitrite assay. Both serum and CSF were used for the multiplex assay. The brains were isolated and processed for WB and multiplex assays or IHC.

For IHC, the rats were perfused with 4% PFA in 0.1M PBS under terminal anesthesia with pentobarbital sodium. The brains were collected and post-fixed in the same solution for 24h at 4°C and cryopreserved in 25% sucrose solution for 3–4 days at 4°C (Cosgrave et al., 2008; Sharma et al., 2018a). The tissues were then embedded in gelatin (15% type A gelatin, 7.5% sucrose, and 0.1% sodium azide in PBS, Sigma, MO, USA), wrapped in cling film, and stored overnight at 4°C. The gelatin-embedded tissue blocks were snap-frozen in liquid nitrogen cooled iso-pentane and stored in −80°C as described previously (Beamer et al., 2012; Cosgrave et al., 2008). The coronal brain sections at 16 μm thickness were cut using a CryoStar NX70 cryostat (specimen head temperature −20°C; blade temperature −16°C, Thermo Scientific, MA, USA). The sections were collected on chrome alum gelatin (Pfaltz and Bauer, CT, USA) coated slides. The details of brain section sampling to represent different regions of the brain (rostral to caudal) on a slide has been described in our previous publication (Puttachary et al., 2016a). The sections were then either processed for IHC immediately or stored at −20°C for later use.

Prior to IHC, antigen retrieval was performed on the brain sections using citrate buffer (10 mM citric acid, 0.05% tween 20, pH 6.0) for 20 min at 95°C. The sections were washed with 0.1M PBS for an hour at room temperature (RT) followed by incubation with 10% donkey serum in PBS. The sections were then incubated with primary antibodies of interest (GFAP/NeuN/iNOS/3-NT/IBA1/CD68/S100β) overnight at 4°C. In addition to neutralizing IgG antibodies against primaries, primary antibody omission step was run in parallel as negative controls. The next day, the sections were washed in PBS for an hour at RT and probed with appropriate secondary antibodies (FITC or CY3 conjugated, or biotinylated) for an hour at RT followed by further washings in PBS and then treated with streptavidin CY3 (biotinylated sections) for an hour. The sections were washed in PBS, and finally with water to remove salt crystals and cover-slipped with Vectashield (Vector Laboratories Inc., CA, USA) containing 4’,6-diamidino-2-phenylindole (DAPI) to stain nuclei. To observe the extent of neurodegeneration in various parts of the brain, we did FJB-NeuN double staining. The brain sections were first stained with NeuN followed by FJB staining as described earlier (Puttachary et al., 2016b; Rao et al., 2006; Todorovic et al., 2012). For FJB staining, the sections were incubated in 0.006% potassium permanganate solution for 5 min with slow shaking. The sections were thoroughly washed twice with distilled water for a minute. The slides were submerged in 0.0003% FJB-0.1% acetic acid solution for 10 min in the dark followed by three washes for one minute each. The slides were air-dried in the dark at RT, cleared in xylene and then mounted in Surgipath acrytol (Surgipath, Leica Biosystems, IL).

The Axiovert 200 Zeiss inverted fluorescence microscope equipped with a Hamamatsu camera (Zeiss, Deutschland, Germany) was used for imaging the immunostained brain sections. As described previously, the systematic sampling of brain sections on slides in sequence provide a reliable representative region of interest from each slide (Puttachary et al., 2016a). The brain regions of interest (for example, CA1, CA3, amygdala, piriform cortex) were first identified using the bright field by referring to the rat brain atlas and verified for cell density using blue (DAPI) channel. Images were captured using HCImage Live 4 software (Hamamatsu Corporation, Sewickley, PA). All the images were taken at 20x magnification. Image J software was used to quantify the cells from a known area (0.14mm²). The area of cell counting for all sections on a slide in all groups was kept constant. The cell counts were done from a minimum of 3 sections per animal distributed across the brain as described in our previous publications (Beamer et al., 2012; Cosgrave et al., 2010a; Cosgrave et al., 2010b; Cosgrave et al., 2008; Puttachary et al., 2016a, 2016b). The NeuN positive cells with FJB staining were counted to determine neurodegeneration. IBA1 positive cells with CD68 were quantified to determine microgliosis. DAPI staining marked nuclei and was used to verify cell number. Since there is no consensus on reliable and reproducible markers for reactive microglia, we considered the area of cell body, the number of branches, and the junctions to distinguish reactive microglia (also referred by some as M1-type microglia) from the resting or alternative type microglia (often referred by some as M2-type) (Andersson et al., 1992; Block, 2014; Torres-Platas et al., 2014; Walker and Lue, 2015). Initially, we derived these parameters from the resting microglia in the control brain section from a known area (e.g., CA1 region) to compare with reactive microglia derived from a similar area in other groups at various time-points. Likewise, we determined the reactive astrocytes based on the large cell body, thick cytoplasmic processes, and intense GFAP staining in contrast to thin processes and moderate GFAP content in resting-type astrocytes. The percentage of reactive microglia or astrocytes were calculated from the total number of IBA1 or GFAP positive cells. We further calculated the percentage of IBA1+CD68 positive microglia from the total immuno-positive cells. The data collected from multiple brain regions were analyzed using a linear repeated mixed-effects regression model (also referred to, in this study, a two-way repeated measures ANOVA) to examine the overall effects of DFP and vehicle or 1400W on gliosis and neurodegeneration across brain within the same set of animals in each brain region. This allowed us to compare the effect of treatment for an average of all brain regions as well as to determine the simple effects of treatment within each region. A similar approach has been reported in the literature (Pessah et al., 2016; Wu et al., 2018).

2.9. Western blotting

The hippocampal and piriform cortex tissues were dissected immediately after euthanasia and snap frozen in liquid nitrogen. The tissues were homogenized and lysed in RIPA buffer containing 1% protease and phosphatase inhibitors (Thermo-Scientific, USA). We used the kit and the standard protocol from Thermo Scientific. The samples were then centrifuged at 16,000 × g for 10 min and the supernatant was collected. The protein concentration from the supernatant was determined using the Bradford assay kit (Biorad, USA). Equal amounts of protein (30–60 ug) were loaded in the wells of SDS-PAGE gels (8–10%) along with a molecular weight marker. The gels were run at 110V for 1–2h at 4° C. The proteins were transferred onto a nitrocellulose membrane and the transfer sandwich was placed into a mini transfer blot unit (Biorad, USA) at 4°C overnight at 25V for 14–16h according to the manufacturer’s instructions. Next day, the membrane was washed with PBS for few minutes and incubated with Fluorescent Western Blot blocking buffer (Rockland Immunochemicals, PA, USA) for an hour RT. After the blocking, the membranes were incubated overnight at 4°C with the primary antibodies. The following day, the membranes were washed with a mixture of PBS and 0.1% Tween® 20 (PBS-T) and then incubated with IR-680 or IR-800 dyes followed by further washes with PBS-T as described earlier. β-actin was used as a loading control and a molecular weight marker was used to determine appropriate band for the protein of our interest. Fluorescent Western Blot blocking buffer was used as a diluent for both primary and secondary antibodies. The bands were visualized on the Odyssey IR imaging system based on the molecular weight. The Region of Interest was determined by creating a rectangle frame around the largest bands of the row of the protein of interest using Image J software. The intensity of each band was quantified and analyzed by normalizing with the β-actin using Image J software.

2.10. Multiplex cytokine, chemokine, and growth factor immunoassays

In this study, we used Luminex xMAP technology for multiplexed quantification of 27 Rat cytokines, chemokines, and growth factors. The multiplexing analysis was performed using the Luminex™ 100 system (Luminex, Austin, TX, USA) by Eve Technologies Corporation (Calgary, Alberta). Twenty-seven markers were simultaneously measured in the samples using a MILLIPLEX Rat Cytokine/Chemokine 27-plex kit (Millipore, St. Charles, MO, USA) according to the manufacturer’s protocol. The hippocampal lysates were prepared and protein concentrations were estimated, as described in section 2.9, and used at 500 μg/mL in the assay. Serum was diluted at 1:2 in PBS at pH 7.2 while CSF was undiluted. The 27-plex consisted of: Eotaxin, EGF, Fractalkine, IFN-gamma,IL-1alpha, IL-1beta, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12(p70), IL-13, IL-17A, IL-18, IP-10, GRO/KC, TNF-alpha, G-CSF, GM-CSF, MCP-1, Leptin, LIX, MIP-1alpha, MIP-2, RANTES, and VEGF. For the 27-plex assay, the sensitivities of these markers range from 0.1–15.7 pg/mL. All assays were done in duplicates and previously known positive and negative controls were used simultaneously with the test samples. A standard curve of all the cytokines and chemokines were prepared using the standard reagents (Eve technologies, Canada) which was used to determine the concentrations of cytokines, chemokines, and growth factors in the test samples and expressed as pg/mL. One way ANOVAs were used to analyze each analyte (Pirela et al., 2013).

2.11. Nitrite assay

The Griess assay was performed to determine the nitrite concentration in the serum and culture media samples. Briefly, 50,000 cells/well were plated in 96-well plates. Next day, the cells were co-treated with LPS (0.1μg/mL) and various concentrations of 1400W (10, 50, 100, and 150 μM) or sterile distilled water for 24h. Further details of culturing, cell line, reagents, and media used are described in the supplementary figure legend (Fig. 3S). For the nitrite assay, 50 μL of media or undiluted serum was transferred to a 96-well plate and mixed with an equal volume of Griess reagent (Sigma-Aldrich) and incubated for 10 min. The color changes in the samples were detected at 540 nm absorbance using a Synergy 2 multi-mode microplate reader (BioTek Instruments, USA). The results were determined by comparing the readout from the samples and sodium nitrite standard curve, and the averages were compared between groups using one-way ANOVA.

2.12. Experimental design, methodological rigor, and statistical analyses

We had consulted a Biostatistician, Dr. Wang, College of Veterinary Medicine, Iowa State University, for the experimental design and statistical analyses for this study. The experimental groups were blinded until the data analyses were completed. We followed the pre-determined criteria to exclude animals from the data analyses. The criteria set were: i) non-responders to DFP i.e. the animals that do not show any stages of SE; ii) animals died during the course of the experiment; and, iii) in the telemetry group if animals do not regain the body weight within 8–10 days post-surgery or post-DFP exposure. We applied the Grubbs’ test to remove outliers from the data. The normality of the data was evaluated with Shapiro-Wilk test. We had taken measures to minimize variables by; i) randomizing the animals based on predetermined weight (≥200 g rat) and age (8 weeks) before the start of each experiment; ii) seizure severity during the SE was quantified by both direct observation and offline video analysis by at least two independent observers; iii) we acquired ~48 h of baseline EEG data covering at least 2 day-night cycles to normalize post-exposure EEG from the same animal; iv) authenticated the identity of 1400W and DFP using GC-MS and further tested 1400W in vitro before it was used in the animals (Fig. 3S); and, v) determined the optimum concentration of the primary antibodies by serial dilution and validated their specificity using neutralizing antibodies appropriate to the primary antibodies. We compared the standard error means of the groups using the Student t-test, the Mann-Whitney test, Fischer’s Exact Test, One-way ANOVA with Tukey multiple comparison post-test or repeated measures two-way ANOVA followed by Tukey’s or Sidak’s correction where appropriate. Information on specific statistical test/s for each experimental analysis is described in Table 4. We used the GraphPad Prism 7.0 and R-Studio version 1.1.463 or all statistical analysis. The p-value <0.05 was considered statistically significant.

3. Results

3.1. DFP exposure induced a varying degree of SE severity and mortality in telemetered and non-telemetered rats, and 1400W treatment prevented DFP-induced mortality.

All rats were randomized and exposed to a single dose of DFP (4 mg/kg, s.c.) in non-telemetry rats and 3 or 4 mg/kg in transmitter implanted rats at 10 days post-surgery. The rats that did not receive DFP served as naïve control. All rats that were exposed to DFP developed typical cholinergic symptoms such as SLUD and seizures of varying degree. The severity of seizures or SE was scored on a scale of 1 to 5 based on the following criteria: stage 1- SLUD, mastication, chewing; stage 2- tremors, wet-dog shakes, head nodding, neck jerks, kyphosis and opisthotonus; stage 3- forelimb clonus, Straub tail, rearing and rigid extension of forelimbs; stage 4- rearing, forelimb clonus and loss of righting reflex; stage 5- abducted limbs clonus or repeated rearing and falling with generalized seizures. In telemetered animals, we observed a correlation between the stages of behavioral seizures and epileptiform spiking, with respect to the amplitude and frequency, and power spectrum (Fig. 1A–1D). The gamma power increased during the convulsive seizures while the theta and delta powers increased if there were high amplitude spikes on EEG (Fig. 1C).

Stages 1 and 2 were considered non-convulsive seizures (NCS) and stage 3 to 5 convulsive seizures (CS). In non-telemetry rats, 86% had CS and 14% only NCS and 12% of the DFP-exposed rats died during SE or in <24h post-diazepam treatment (Table 2). The total amount of time the rats spent in CS during the 2h period between DFP and diazepam treatments was used to classify the severity of SE as mild (10–15 min), moderate (15–30 min) and severe (>30 min). In the non-telemetry group, 16% had mild, 9% moderate, and 61% severe SE. Out of the 23 telemetered rats, 16 rats survived following 3 mg/kg DFP, and all the rats showed almost a similar SE profile as the non-telemetered rats (Table 2). Following diazepam administration, the rats were regrouped based on the severity of SE and coded for 1400W or vehicle treatment. Both the vehicle and 1400W groups had an almost equal duration of CS, and no significant difference was observed in the duration of SE severity between the groups (Fig. 1E, 1F). We further analyzed the mortality that occurred within 24h post-diazepam, between the vehicle and 1400W treated non-telemetry groups, in those rats that had convulsive seizures during SE for >90 min. None of the 14 rats died in 1400W group after DFP exposure, while 6 out of 20 rats died (33%) in the vehicle group suggesting the protective effects of 1400W against DFP-induced mortality (Table 2).

3.2. DFP exposure significantly upregulated iNOS and 3-NT in the hippocampus and piriform cortex at 24h, 48h, and 7d post-DFP

The experimental design for early time-points is illustrated in Figure 2A and the number of animals used in each experiment is given in Table 1. We investigated the time course upregulation of both iNOS and 3-NT in the hippocampus and piriform cortex, the most widely investigated parts of the brain in CNA-induced brain pathology (Aroniadou-Anderjaska et al., 2009; Figueiredo et al., 2018). The WB in hippocampal and piriform cortex lysates revealed a significant increase in both iNOS and 3-NT levels at 24h, 48h, and 7d post-DFP (Fig. 2C–2F, 4A–4D). Serum nitrite levels were significantly higher at 24h post-exposure (Fig. 4E). A persistent increase of both iNOS and 3-NT in the hippocampus and piriform cortex prompted the investigation of their cellular distribution. IHC of the brain sections for iNOS and 3-NT, co-labeled with IBA1 or NeuN, at 7d post-DFP, revealed a significant increase in the iNOS in microglia (IBA1 positive cells) and 3-NT immunopositive staining in neurons (NeuN positive cells) in both CA1 region of the hippocampus and the piriform cortex (Fig. 3A–3D). Interestingly, the pattern of such co-localization was also persisted in the chronic phase at 12 weeks post-exposure (Fig. 3A–3D) suggesting the role of iNOS-mediated NO in the long-term brain pathogenesis in response to DFP exposure. Therefore, we hypothesized that iNOS inhibition by 1400W soon after the DFP exposure would protect the brain.

Figure 2.

Experimental design and early expression levels of iNOS and 3NT. A, B: The experimental design for non-telemetry early time-points (A) and telemetry (B) studies are illustrated. The number of animals used in each experiment is given in Table 1. The animals used for early time-point experiments were euthanized for WB at 24h and 48h, and for WB and IHC at 7 days. The animals used in the telemetry experiments (B) were subjected to continuous video-EEG recording for 12 weeks. The brains from these animals were used for IHC evaluation (gliosis and neurodegeneration). C-F: The WB analysis of tissue lysates from 48h and 7d post-DFP groups showed a significant increase in both iNOS and 3-NT levels in the hippocampus (C, E) and piriform cortex (D, F). Data presented as a group mean ±SEM (n=4 per group). *p<0.05, **<0.01, ***p<0.001 denote significant differences between DFP and control groups. n=4 per group. DZP=diazepam, SE=status epilepticus, SRS= spontaneous convulsive seizure.

Figure 4.

1400W inhibits DFP-induced iNOS and 3-NT upregulation in the hippocampus and piriform cortex, and the serum nitrite levels at 24 post-exposure. The animals were treated with two doses of 1400W (20 mg/kg) at 12h intervals. The WB analysis showed a significant increase in iNOS and 3-NT in the vehicle group and a reduction in the 1400W group in both hippocampus and piriform cortex lysates (A-D). Serum nitrite levels were significantly increased in the DFP group compared to control; this was significantly suppressed by 1400W intervention (E). n=4–5 per group. *p<0.05, **p<0.01, ***p<0.001 with respect to control; ^#p<0.05, ^{# #}p<0.001 with respect to the DFP+ vehicle group.

Figure 3.

A-D: IHC of the brain sections for iNOS, IBA1, 3-NT and NeuN from 7d and 12 weeks post-exposure animals. Panel A represents the images from CA1 of the hippocampus and piriform cortex co-immunostained for IBA1 (microglia marker, red label) and iNOS (green label), and in panel B, 3-NT (green label) and NeuN (neuronal marker, red label) to visualize co-localization (appears yellow). In both A and B, nuclei are stained with DAPI (blue). The majority of the IBA1 positive microglia were co-labeled with iNOS, likewise, the majority of NeuN were co-labeled with 3-NT in the DFP exposed animals (yellow labeled cells). There was a significant increase in both iNOS and 3-NT, co-labeled with IBA1 and NeuN respectively, at 7d and at 12 weeks post-DFP suggesting the long-term impact of iNOS-mediated neurotoxicity (C, D). *p<0.05, **<0.01, denote significant differences between DFP and control groups. n=4 per group. Scale bar 50 μm.

3.3. 1400W suppressed DFP-induced iNOS and 3-NT levels in the hippocampus and piriform cortex, and the serum nitrite levels at 24h post-exposure

At 24h post-exposure in rats, 1400W treatment (20 mg/kg, two doses at 12h interval, the first dose was administered 2h post-diazepam) suppressed DFP-induced iNOS and 3-NT upregulation in both hippocampus and piriform cortex as revealed by WB (Fig. 4A–4D) suggesting the pharmacological inhibition of iNOS as well as the suppression of iNOS synthesis. DFP-induced serum nitrite levels were also significantly suppressed by 1400W at 24h post-exposure (Fig. 4E) suggesting the activation of iNOS in peripheral leukocytes by DFP and suppression by 1400W.

3.4. 1400W suppressed gliosis and neurodegeneration in the hippocampus at 7d post-exposure

Having observed the upregulation of iNOS and 3-NT at 7d post-exposure, we investigated the effects of 1400W on gliosis and neurodegeneration in the hippocampus. 1400W was administered for the first three days of post-exposure (at 12h intervals) since we found a persistent increase in both iNOS and 3-NT at 48h (Fig. 2C–2F). IHC of brain sections at 7d post-DFP revealed a significant reduction in microgliosis (IBA1 positive cells), astrogliosis (GFAP positive cells), and neurodegeneration (NeuN+FJB) in the 1400W-treated group in contrast to the vehicle-treated group (Fig. 5A, 5B) suggesting the protective effects of iNOS inhibition.

Figure 5.

1400W significantly suppressed DFP-induced gliosis and neurodegeneration at 7d post- exposure. A: IHC of the brain sections for gliosis (IBA1 in green in the top panels, GFAP in red in the middle panels) and neurodegeneration (NeuN in red and FJB in green in the bottom panels; co-labeled cells appear yellow). The microglia and astrocytes (in DFP+ vehicle panel) were concentrated more in the pyramidal cell layer of CA3 and obscured the DAPI stained nuclei. The FJB positive neurons (in DFP+ vehicle panel) were also concentrated at the same CA3 area. All three images in DFP+ vehicle panel are from parallel sections of the same animal. 1400W treatment significantly reduced microgliosis, astrogliosis, and neurodegeneration (B). Scale bar, 100 μm. n=5–6 per group, *** p<0.001 with respect to control, ^{# # #} p<0.001 with respect to DFP+ vehicle group.

3.5. Diazepam controlled behavioral seizures but did not suppress electrographic seizures and epileptiform spiking

Diazepam administration suppressed the behavioral seizures in both telemetry and non-telemetry groups. The telemetry implanted rats further revealed that despite behavioral seizures suppression by diazepam, the epileptiform spikes persisted and the seizure clusters reappeared by 6h post-diazepam (Fig. 6A) suggesting that diazepam alone would not be sufficient to control abnormal brain electrical activity induced by DFP. Atropine sulfate and 2-PAM also had no impact on the DFP-induced EEG parameters (Fig. 6A) even though both were administered within a minute of DFP exposure.

Figure 6.

The effects of MCMs (atropine sulfate (ATS), 2-PAM, and diazepam), DFP and vehicle or 1400W on EEG characteristics and SRS in the rat DFP model. MCM alone did not suppress electrographic seizures and epileptiform activity during the first 1h and 6h post-exposure (A), and in the long-term (B). ATS and 2-PAM were administered in <1min after DFP. The diazepam was administered at 2h after DFP exposure. Diazepam controlled only behavioral seizures but did not affect epileptiform spikes or clusters (A). Representative EEG traces from 1 to 12 weeks (B, C) and the epileptiform spike counts (D, E) from 0 to 12 weeks post-exposure from the vehicle and 1400W treated animals demonstrate the differences in epileptiform spikes, NCS, and spike clusters. NCS were characterized by high amplitude spikes and an increase in the delta power (C). A significant reduction in epileptiform spiking in 1400W treated animals was observed at various time points (D). The epileptiform counts during the first 4 weeks (0–4 weeks), 5–8 weeks, and 9–12 weeks were pooled and compared between the groups (E). An example of an SRS episode that occurred at 27d post-exposure is illustrated (F). An integrated video and EEG trace along with the corresponding increase in the gamma power during a convulsive seizure followed by a brief stage-2 like high amplitude spikes characterized by an increase in the delta and theta powers are illustrated (F). The activity counts represent the measure of their musculoskeletal activity during a seizure. We considered four parameters (video-based behavior, EEG, power spectrum, and activity counts) to confirm SRS, and only such episodes were quantified in the 12 weeks of continuous video-EEG study (G-I and Table 3). In figure I, each dot represents an SRS episode that occurred at various time points during the 12 weeks. In the vehicle group, 4 out of 8 rats had several convulsive SRS episodes. In contrast, only one rat had just 3 convulsive SRS in the 1400W group (I). n=8 per group, *p<0.05, **p<0.01, ***p<0.001 (vehicle vs. 1400W).

3.6. 1400W suppressed epileptiform spiking and SRS and prevented epileptogenesis in some animals

The experimental design for the telemetry experiments is illustrated in Figure 2B. The telemetered rats were exposed to DFP at 10-days post-surgery. The SE profiles are given in Table 2. The 16 out of 23 rats exposed to DFP were survived and showed a similar SE profile as the non-telemetered rats (Table 2) and divided into two groups (n=8) for 1400W or vehicle treatment. Both groups had an almost equal duration of CS during the SE. In the DFP-exposed vehicle-treated animals, the EEG traces showed more frequent electrographic NCS, epileptiform clusters, spike trains, and interictal spikes (Fig. 6B, 6C). The spikes from these patterns were quantified and represented as epileptiform spikes and compared between the vehicle and 1400W groups. 1400W significantly attenuated the epileptiform spikes at multiple time points across 12 weeks observation (Fig. 6D). There was a significant reduction in overall epileptiform spiking in the 1400W treated group between 9–12 weeks post exposure when compared to the vehicle group (Fig. 6D–6E). Continuous video-EEG recording for 12 weeks revealed the developed of epilepsy in all 8 rats in the vehicle-treated group (based on spontaneous NCS and CS onset). In contrast, only 3 out of 8 rats in the 1400W-treated group developed epilepsy having convulsive or non-convulsive SRS. A few examples of an electrographic NCS and a convulsive SRS are illustrated (Fig. 6C, 6F). NCS episodes as such were not quantified, however, the epileptiform spike counts include the spikes in NCS (Fig. 6D, 6E). All convulsive SRS were verified against the synchronized video and a corresponding increase in the gamma power, and an increase in activity counts (increased locomotor activity) (Fig. 6F). SRS episodes occurred during the 12 weeks period are pooled and shown in Figure 6G–H. In the vehicle-treated group, 4 out of 8 rats had several convulsive SRS while in 1400W group only one rat had convulsive SRS (just 3 episodes) (Fig. 6I; Table 3). In the 1400W group, two other rats had a few electrographic NCS and occasional interictal spikes similar to the EEG pattern shown in Figure 6C. Further analysis of the severity of convulsive SRS, based on stage 3, 4, and 5, is summarized in Table 3.

Table 3.

Types of SRS during the 12 weeks of continuous video-EEG monitoring in the vehicle or 1400W treated groups.

Animal # → SRS stage↓	RT3	RT7	RT27	RT6	RT21
	DFP+ vehicle				DFP+ 1400W
Stage 3	28	57	4	13	3
Stage 4	17	20	1	6	0
Stage 5	14	26	0	7	0
Total SRS	59	103	5	26	3

3.6. 1400W suppressed DFP-induced microgliosis at 12 weeks post-exposure

Since there was a persistent upregulation of both iNOS and 3-NT at 12 weeks, and 1400W suppressed SRS and epileptiform spikes during the 12 weeks, we further examined the brains from the same animals that were used for video-EEG studies. In this study, we did not use sham surgery animals as control since our previous studies in the rats implanted with electrodes on the surface of the dura mater (not depth electrodes) did not show any significant differences in brain pathology with respect to gliosis and neurodegeneration at 12 weeks post-surgery (Fig. 4S). Off-note, however, we did not test these parameters at 10 days post-surgery, the time-point chose for the administration of DFP in telemetry groups.

We utilized IBA1 to mark the microglia (Fig. 7A). Main effects of treatment showed an increase in microglia in the vehicle group and a significant decrease by 1400W (Fig 7Ci). Following further region-wise post-hoc analysis, the vehicle group showed a significant upregulation of microglia in the amygdala and piriform cortex and significant decreases by 1400W (Fig. 7Cii). We further quantified reactive microglia which were identified based on a large cell body and thickened cytoplasmic processes with a few numbers of branching patterns when compared to the resting/alternative-type microglia (see, for example, Fig. 7B). The percentage of reactive (M1-like) microglia from the average number of IBA1 positive cells were calculated and compared between the groups. Main effects following two-way repeated measures ANOVA showed a significant increase in reactive microgliosis for the vehicle group and a significant decrease in the 1400W group (Fig. 7Di). Further region-wise post-hoc analysis showed a significant increase in reactive microgliosis in the vehicle group in all regions except the thalamus, as compared to controls, and a significant decrease by 1400W in the amygdala (Fig. 7Dii). Further, we co-stained IBA1 with CD68, a phagocytic marker, to identify whether co-labeled cells were also altered in response to DFP and 1400W. Main effects analysis following two-way repeated measure ANOVA, showed a significant increase in the percentage of CD68 containing microglia cells in the vehicle and 1400W groups as compared to the control (Fig. 7Ei). Post-hoc analysis of the same parameter showed a significant increase in all brain regions and no mitigation by 1400W (Fig. 7Eii). However, qualitatively, in the 1400W group, IBA1+CD68 positive cells were less frequently observed in the pyramidal cell layer of the CA1 region of the hippocampus when compared with the vehicle group (Fig. 7A).

Figure 7.

The effects of DFP and vehicle or 1400W treatment on microgliosis, reactive microgliosis, and CD68 containing IBA1 cells in various parts of the brain at 12 weeks post-exposure. A: IHC of brain sections for IBA1 (red label) and CD68 (green label, but co-localized cells appear yellow), representative images are from CA1 and amygdala. Reactive (M1-like) microglia were characterized by large cell body and thick cytoplasmic process in contrast to thin processes and small cell body of resting type microglia (B). CD68 positive reactive-type microglia were more frequently found at the vicinity of neurons in the pyramidal CA1 hippocampal region (A). Main effects showed a significant increase in the number of microglia averaged across brain regions in the DFP+ vehicle group; this was significantly suppressed by 1400W (Ci). Following further region-wise post-hoc analysis, the significant differences were found only in the amygdala and piriform cortex (Cii). Furthermore, main effects revealed a significant increase in the percentage of IBA1 positive cells with M1-like morphology; this was significantly reduced by 1400W (Di). Simple effects, using the same model, showed significant upregulation in all regions except the thalamus in the DFP+ vehicle group and significant suppression by 1400W in the amygdala (Dii). Since the majority of microglia were co-labeled with CD68, the main effects showed was an overall significant increase of the percent IBA1 positive cells with CD68 localization in the vehicle group when compared to the control, and the general suppression by 1400W but this was not significant (Ei). When region-wise simple effects were examined, significant upregulation of IBA1 cells with CD68 in the vehicle group was found in all regions (Eii). In panel Ci, Di, and Ei, the data are presented as a group median (horizontal black bar) and 95% CI. In panel Cii, Dii, and Eii, the data are presented as a group mean ±SEM. *p<0.05, **p<0.01, ***p<0.001.with respect to controls; #p<0.05, ##p<0.01, ###p<0.001 with respect to DFP + vehicle group. n=6–7 per group. Scale, 100 μm. CA, cornu ammonis; DG, dentate gyrus; AMY, amygdala; PC, piriform cortex; ENT, entorhinal cortex, TH, thalamus.

3.7. 1400W suppressed DFP-induced reactive astrogliosis at 12 weeks post-exposure

A similar analysis as microgliosis was done to evaluate the effects of DFP and 1400W on astrogliosis. Main effects did not show significant differences in the GFAP positive astroglia cell numbers between treatment groups (Fig. 8Ci). It should be noted that in the thalamus there was a significant decrease in GFAP positive cells in the 1400W group as compared to the vehicle group following post-hoc analysis (Fig. 8Cii). We further quantified reactive-type astrocytes which were identified based on thick cytoplasmic processes and intense GFAP staining (Fig. 8B). The vast majority of GFAP positive astrocytes were also co-localized with S100β in all treatment groups. Like the IBA1+CD68 reactive microglia, the reactive-type astrocytes with intense S100β were also frequently found in the vicinity of the pyramidal cell layer of the CA1 region of the hippocampus (Fig. 8A). Following main effects analysis, in the vehicle group, reactive astrocytes were significantly increased when compared to the control and significantly decreased in the 1400W group (Fig. 8Di). Further post-hoc analysis showed a significant increase in reactive astrogliosis in all regions except the thalamus. 1400W significantly suppressed reactive gliosis in CA1 and thalamus while in the other regions there was reduction but those were not statistically significant (Fig. 8Dii).

Figure 8.

The effects of DFP and vehicle or 1400W treatment on astrogliosis and reactive astrogliosis in various parts of the brain at 12 weeks post-exposure. A: IHC of brain sections for GFAP (green labeled cells) and S-100β (red label, but co-localized cells appear yellow) and representative images are from CA1 and amygdala. The majority of astrocytes were co-labeled with S-100β and the reactive type astrocytes were characterized by thick cytoplasmic process and large cell body in contrast to thin processes and small cell body (B). Like reactive microglia, the reactive-type astrocytes (with intense S-100β staining) were also more frequently found at the vicinity of neurons in the pyramidal CA1 hippocampal region in the vehicle group which were reduced in the 1400W (A). Main effects for the number of GFAP positive cells as such did not reveal significant differences between control, the vehicle, and 1400W groups (C-i). Further post-hoc analysis for region-wise simple effects of treatment also showed a similar results in all regions, except the thalamus where it was significantly suppressed in 1400W group (C-ii). Main effects for the percentage of GFAP positive cells with reactive-type morphology demonstrated a significant increase in reactive astrogliosis in the vehicle in contrast to the control with significant suppression by 1400W (Di). Tukey’s post-hoc for region-wise simple effects also showed a similar significant increase of reactive astrocytes in the vehicle group in all regions, except thalamus, however, their suppression by 1400W was only observed in CA1 and thalamic regions as compared to the vehicle-treated group (Dii). In panel Ci and Di, the data are presented as a group median (horizontal black bar) and 95% CI. In panel Cii and Dii, the data are presented as a group mean ±SEM. *p<0.05, **p<0.01, ***p<0.001 with respect to controls; #p<0.05, ##p<0.01, ###p<0.001 with respect to DFP + vehicle group. n=6–7 per group. Scale, 100 μm. CA, cornu ammonis; DG, dentate gyrus; AMY, amygdala; PC, piriform cortex; ENT, entorhinal cortex, TH, thalamus.

3.8. 1400W mitigates DFP-induced neurodegeneration at 12 weeks post-exposure

Since there was a persistent increase in iNOS and 3-NT at 12 weeks post-exposure (Fig. 2, 3), and 1400W suppressed DFP-induced neurodegeneration at 7d post-exposure (Fig. 5), we further investigated its long-term impact on neurodegeneration at 12 weeks. DFP exposed animals showed a significant increase in NeuN+FJB positive cells when compared to the control and significant decrease by 1400W, as revealed by the main effects analysis (Fig. 9Bi). This significance was found in all regions following further post-hoc analysis except the piriform and entorhinal cortex where 1400W treated animals showed a reduction that was not statistically significant (Fig. 9Bii). Representative IHC images from CA1 and amygdala are shown in Figure 9A.

Figure 9.

The effects of DFP and vehicle or 1400W treatment on neurodegeneration (FJB+NeuN positive cells) in various parts of the brain at 12 weeks post-exposure. A: IHC of NeuN (red labeled cells) coupled with FJB staining (green label, but co-localized cells appear yellow) of brain sections and representative images from CA1 and amygdala. The linear mixed-effects regression analysis of main effects revealed an overall significant increase in neurodegeneration in the vehicle when compared to the control and 1400W treatment significantly reduced DFP-induced neurodegeneration (B-i). Further, region-wise analysis also revealed significant increases in FJB+NeuN positive cells in the vehicle when compared to the control in all the regions with significant suppression by 1400W in all regions except the piriform cortex and entorhinal cortex (B-ii). In panel Bi, the data are presented as a group median (horizontal black bar) and 95% CI. In panel Bii, the data are presented as a group mean ±SEM. **p<0.01, ***p<0.001 with respect to controls; #p<0.05, ##p<0.01, ###p<0.001 with respect to DFP + vehicle group. n=6–7 per group. Scale, 100 μm. CA, cornu ammonis; DG, dentate gyrus; AMY, amygdala; PC, piriform cortex; ENT, entorhinal cortex, TH, thalamus.

3.9. Impact of DFP and vehicle or 1400W on cognition, motor function, and depression-like behavior

The experimental design for all behavioral tests is illustrated in Figure 10A and 10B. The basal neurobehavioral tests, conducted prior to the DFP exposure, did not show significant differences in spatial learning and memory between the groups in the MWM (Fig. 10C, 10Di) suggesting the animals used in the experiment were normal with respect to cognition. It is important to note that in this study, the short-term memory is defined as 24h after the last learning in the MWM and the long-term memory is defined as 11 days between the two probe trials with the target being in the same quadrant in both probe trials (Fig. 10B). This is in contrast to the commonly accepted 24h gap, after the last learning session, as the long-term memory in the MWM (for example, Vohrees and Williams, 2010).

In the MWM, there was a reduction in the time to reach the platform as the training period progressed in all three groups prior to DFP exposure (Fig. 10C). At 9d post-DFP (i.e. 11 days after the first single probe trial with the target in NE quadrant), a second probe trial with the same target (NE) was conducted to test whether the animals remembered the previously learned task. The vehicle group spent significantly less time at the target quadrant when compared to their own performance prior to the exposure to DFP (Fig. 10Ei) suggesting the early impact of DFP on the long-term memory. In contrast, there were no significant differences in the 1400W and control groups (Fig. 10Ei) which may suggest that 1400W might had mitigated the DFP-induced long-term memory impairment at the time-point tested. When the new learning was introduced at 10d post-DFP for 6 days with the platform in a new quadrant (SW), no significant differences were observed between groups in spatial learning and short-term memory (Fig. 10C, 10Diii). When the probe trial was repeated after 11 days (27d post-DFP) with the target in the same quadrant (SW), the vehicle group though spent less time at the target, a similar trend observed at 9d post-DFP, but the difference was not significant between the two probe trials (Fig. 10Eii). There were also no significant differences in the control and 1400W-treated groups in the long-term memory, considering the time spent in the target quadrant as a parameter (Fig. 10Eii). Likewise, there were also no differences between the groups in reversal learning and short-term memory conducted between 28–34d post-DFP (Fig. 10C, 10Dv).

The motor function tests did not reveal significant differences between groups in the latency to fall in the rotarod or the time taken to reach the target in HBT (Fig. 10F, 10G). In the accelerated rotarod test, prior to DFP exposure, the control group showed a significant increase in latency to fall when compared between the training and the test. Surprisingly, however, the vehicle and 1400W groups did not show a similar significant difference (Fig. 10F). In post-exposure, in both control and 1400W, there was a positive trend in the test results after the trial, and no difference in the vehicle group, but the differences were not significant (Fig. 10F). In HBT, prior to DFP exposure, all three groups showed a learning trend to reach the target quicker than the first training but the difference was not significant in the 1400W group. A similar pattern was observed in the post-exposure (Fig. 10G). The forced swim test was used to determine depression-like behavioral dysfunction in response to DFP exposure and mitigation by 1400W. In pre-exposure test, immobility duration during a 4-minute swim revealed no differences between groups. In post-exposure, animals in both vehicle and 1400W groups were immobile for a longer time than the control but the differences were not significant (Fig. 10H).

3.10. 1400W modulated DFP-induced key proinflammatory cytokines and chemokines, and serum nitrite levels at 6 weeks post-exposure

We used the Rat Cytokine 27 Plex Discovery assay to determine the levels of cytokines and chemokines from the hippocampal lysates from the rats that were used for the behavioral tests and euthanized at 6 weeks post-exposure. The key proinflammatory cytokines such as IL-1α, TNFα, IL-1β, IL-2, IL-6, IL-10, IL-12, and IL-17A were significantly upregulated in the DFP exposed animals that were treated with the vehicle when compared to the control (Fig. 11A). In the 1400W group, IL-1α, TNFα, IL-2, IL-12, and IL-13 were significantly suppressed in comparison to vehicle-treated animals, and IL-4, an anti-inflammatory cytokine was upregulated when compared to the control (Fig. 11A). IL-5 and IL-18 levels were unaltered. The key proinflammatory chemokines such as MCP-1, LIX and Eotaxin, and growth/trophic factors VEGF and leptin were upregulated in the vehicle group when compared to the control. Importantly, 1400W significantly reduced MCP-1 levels when compared to the vehicle (Fig. 11B). Though LIX, Eotaxin, and VEGF levels were reduced by 1400W, the differences were not significant (Fig. 11B). The fractalkine levels did not change in any group (Fig. 11B). It is of note that LIX and Leptin levels in the 1400W group were also significantly upregulated as compared to the control group (Fig. 11B). Other chemokines such as EGF, MIP-1α, and RANTES were increased in the vehicle and reduced in the 1400W group but the differences were not significant. We also observed an increase in IP-10 (CXCL10) levels in the 1400W group but it was not significant (Fig. 11B). Interestingly, we did not observe significant differences in any of the cytokines/chemokines in the serum or CSF at 6 weeks post-exposure (data not shown). However, there was a significant increase in serum nitrite levels in the vehicle group which was reduced in the 1400W group (Fig. 11C).

Figure 11.

The effects of DFP and vehicle or 1400W treatment on cytokines, chemokines, and growth factors in hippocampal lysates (A, B), and on the serum nitrite levels (C) at 6 week post-exposure. The key proinflammatory cytokines IL-1α, TNFα, IL-1β, IL-2, IL-6, IL-10, IL-12, and IL-17A were significantly upregulated in the vehicle when compared to the control (A). In 1400W treated animals, IL-1α, TNFα, IL-2, IL-12, and IL-13 were significantly suppressed, and IL-4, an anti-inflammatory cytokine was upregulated in contrast to the control group (A). The chemokines such as MCP-1, LIX and Eotaxin, and growth/trophic factors VGEF and leptin were upregulated in vehicle when compared to the control. MCP-1 was significantly suppressed in the 1400W group (B). The serum nitrite levels were significantly higher in the vehicle group and suppressed in the 1400W group (C). *p<0.05, **p<0.01, ****<p<0.001, ****p<0.0001 with respect to control; ^#p<0.05, ^##p<0.01, ^###p<0.001 with respect to the vehicle. n=4–6 per group.

Discussion

Our previous in vitro (brain slice multi-electrode array) and in vivo studies in the mouse and rat models, respectively, demonstrated the efficacy of 1400W in suppressing the KA-induced epileptiform spiking and the disease-modifying effects (Puttachary et al., 2016b). In the present study, we demonstrate similar neuroprotective effects of 1400W in the acute DFP exposure model both in the short- and long- term. DFP is a CNA and mimics nerve agents that have been used in chemical warfare such as soman and sarin with respect to AChE target engagement and the associated acute clinical signs (El Sayed et al., 2014; Lynch et al., 1986; McCarren and McDonough, 2016; Tripathi and Dewey, 1989). The short- and long- term consequences of DFP exposure on brain morphology and function in experimental models are also very much similar to soman and/or sarin (de Araujo Furtado et al., 2010; Kadar et al., 1992; Marrero-Rosado et al., 2018; Sisó et al., 2017). DFP-induced EEG changes, reactive gliosis, neurodegeneration, and their long-term consequences on functional outcomes are well known (McDonough et al., 1998; Pessah et al., 2016; Todorovic et al., 2012). It is also now well known that MCMs do not prevent the long-term consequences of acute CNA exposure in experimental models if they are not administered immediately after the exposure (Kadriu et al., 2011; Kuruba et al., 2018; McDonough et al., 1995; Todorovic et al., 2012; Wu et al., 2018). Similarly, in our DFP rat model, administering the diazepam 2h post-exposure did not prevent epileptiform spiking or brain pathology although it controlled behavioral seizures in the vast majority of animals. Based on evidence for 1400W as a potential disease-modifying agent in the rat KA model of chronic epilepsy (Puttachary et al., 2016b), we investigated whether 1400W can also mitigate the DFP-induced long-term neurotoxicity when it was administered for the first three days starting at 4h (i.e., 2h post-diazepam) after the acute exposure.

The treatment regimen to intervene CNA-induced brain pathology in our study differs from the published literature (for example, (Liang et al., 2017; Miller et al., 2017; Wu et al., 2018) in three aspects: i) No pyridostigmine (PYR) pre-treatment, ii) delayed AED treatment i.e., 2h post-exposure, and iii) a further delay in neuroprotectant treatment (2h post-AED). This strategy is expected to mimic an ‘after-field evacuation and in-hospital’ scenario. The rat DFP model recommended by the NIH NIAID Chemical Countermeasures Research Program was modified in this study by excluding PYR to mimic civilian population scenario with respect to the pre-medication procedure since it is impractical to pre-medicate the civilian population which could also be exposed to OP CNAs. A similar modification was made in a previous study (Pessah et al., 2016). ATS and 2-PAM are essential to minimize mortality although the timing of administration of these MCMs do not mimic real-world scenario. Likewise, the timing and the dose of diazepam treatment 2h post-exposure in this study may not be practical in some scenarios. It is reasonable to argue that there are no perfect animal model or protocols to mimic all aspects of human condition or scenarios. However, in this study, delayed AED treatment was expected to limit mortality and the confounding neuroprotective effects of AED if it were to be administered at an early time-point (Kuruba et al., 2018; Wu et al., 2018; Zhang et al., 2017). Our results confirm that delayed AED treatment causes sufficient SE-induced brain pathology which is useful to measure the real effects of neuroprotectants.

The DFP-induced mortality in this study was 12% in the non-telemetry rats while it was 30% in the telemetered rats. Our recent study in the rat KA model also suggested that the surgical procedure for electrode implantation reduces the threshold for seizurogenic agents at about 8–10 days post-surgery (Sharma et al., 2018a). However, we did not test seizurogenic agents at later time-points post-surgery. The dose, temperature, pH, vehicle, and the route of administration of DFP impact the severity of SE and mortality (Deshpande et al., 2010). We have kept these variables constant for all the rats tested in this study. The dose of DFP used in this study was sufficient to induce SE of varying degree to cause a progressive brain pathology as demonstrated in our results and the literature (Flannery et al., 2016; Sisó et al., 2017). To group the animals for vehicle or 1400W treatment, we considered the group average of the duration of SE to balance the number of animals of varying SE severity between the groups. There were no significant differences between the vehicle and 1400W treated groups in the initial seizure severity. Importantly, 1400W treatment also prevented acute mortality induced by DFP.

Chemoconvulsant-induced SE causes a persistent increase in iNOS transcription and/or production and nitro-oxidative stress in rodent models (Cosgrave et al., 2008; De Simoni et al., 2000; Gupta et al., 2001; Ryan et al., 2014; Zaja-Milatovic et al., 2009). NO nitrosylates certain proteins on tyrosine residue (3-NT) which can be used to measure NO bioactivity in addition to the other end products such as nitrite or citrulline (Higashi et al., 2017; Kim et al., 2018). We found that DFP-induced SE caused a significant increase in iNOS and 3-NT levels in the hippocampus and piriform cortex at 24h, 48h, 7d and persisted even at 12 weeks post-exposure (Fig. 2, 3). Previous studies in the rat DFP and KA models have shown that NO production occurs as early as 1h post-exposure and depletes high-energy phosphates (ATP) in the hippocampus, piriform cortex, and amygdala (Gupta et al., 2001; Milatovic et al., 2002). IHC of brain sections in our DFP model at both 7d and 12 weeks post-exposure revealed that iNOS and 3-NT were predominantly co-localized in IBA1 positive microglia and NeuN positive neurons, respectively (Fig. 3). We have previously shown that the SE-induced activation of glia in the hippocampus produces iNOS in the rat KA model (Cosgrave et al., 2008; Sharma et al., 2018a). In another rat KA study, increased NO levels and 3-NT in neurons in the hippocampus were shown to persist for up to 6 weeks post-SE (Ryan et al., 2014). The increased NO production in activated glia causes inflammatory neurodegeneration by inhibiting neuronal respiration, glutamate release, and excitotoxicity (Bal-Price and Brown, 2001). We also observed a significant increase in reactive gliosis and neurodegeneration (FJB+NeuN positive cells) in several brain regions at 12 weeks post-exposure (Fig. 7–9) suggesting the impact of DFP-induced iNOS in the long-term brain pathology. Previous studies in the rat DFP model have also demonstrated persistent neurodegeneration and reactive gliosis for up to two months post-exposure (Sisó et al., 2017). Positron emission tomography (PET) imaging of [11C]-(R)-PK11195, a ligand for the 18-kDa mitochondrial membrane translocator protein (TSPO) of the rats that were acutely exposed to DFP confirmed brain inflammation at 21 days post-exposure (Flannery et al., 2016). These past studies and our current observations indicate that acute DFP intoxication activates glia (neuroinflammation), produces RNS, and causes neurodegeneration. Therefore, our hypothesis was that dampening the iNOS-mediated RNS production at an early stage of DFP exposure could have a long-term impact on neuroinflammation and neurodegeneration.

We utilized 1400W dihydrochloride, a highly selective iNOS inhibitor, which is >5000- and 200-folds more selective against the purified human iNOS than the eNOS and nNOS, respectively (Garvey et al., 1997). 1400W suppressed DFP-induced iNOS and 3-NT levels in both hippocampus and piriform cortex confirming the iNOS inhibition and its impact on the synthesis or transcription of iNOS itself. IHC of brain sections revealed iNOS co-localization primarily in IBA1 positive microglia cells (Fig. 3). Seizures induce NFkB in microglia which regulate the transcription of several key pro-inflammatory cytokines and chemokines, and iNOS (Lubin et al., 2007; Vezzani et al., 2011b, 2011a, 2013; Vezzani, 2014). Interestingly, iNOS-mediated NO activates NFkB suggesting that iNOS indirectly mediates the production of key proinflammatory cytokines and chemokines, and promotes neurodegeneration (Chang et al., 2014; Vezzani et al., 2011b). Thus, iNOS is a key regulator of neuroinflammation and nitro-oxidative stress, and a potential therapeutic target to mitigate these deleterious effects. 1400W also suppressed DFP-induced serum nitrite levels, at 24h post-exposure, which can be a useful peripheral biomarker for determining RNS-mediated pathogenesis (Demchenko et al., 2003; Schopfer et al., 2003; Tsikas et al., 2006). The circulating leucocytes produce iNOS, but not nNOS, and OP toxicity has been known to increase serum nitrite levels (Hickey et al., 2001; Niess et al., 2000; Yurumez et al., 2007), therefore altered nitrite levels in the serum can be attributed to iNOS and 1400W. The serum nitrite levels were also increased at 6 weeks post-exposure suggesting that the persistent upregulation of brain iNOS was catalyzing the NO production to cause neurotoxicity and the suppression in 1400W treated animals suggest its long-term washout effects.

The reactive glial cells and a concomitant increase in brain iNOS promote the production of proinflammatory cytokines and chemokines in experimental models of epilepsy (De Simoni et al., 2000; Devinsky et al., 2013; Jankowsky and Patterson, 2001; Vezzani et al., 2011a; Wilms et al., 2010). The multiplex assay of the hippocampal lysates at 6 weeks post-exposure in our study showed a significant increase in the key proinflammatory cytokines and chemokines suggesting the existence of neuroinflammation in the chronic phase after acute exposure to DFP. Amongst the increased proinflammatory cytokines/chemokines in response to DFP exposure, IL-1α, TNFα, IL-2, IL-12, and MCP-1 were significantly suppressed by 1400W suggesting its long-term protective effects. Interestingly, 1400W did not significantly suppress DFP-induced IL-1β, IL-6, and IL-17A levels. Several experimental studies have implicated MCP1 in seizure onset and neurodegeneration (Arisi et al., 2015; Cerri et al., 2016). MCP1, through its receptor CCR2, induces neurodegeneration via STAT3 activation and IL-1β production after SE in a mouse model (Tian et al., 2017). Suppression of MCP1 levels and several other proinflammatory cytokines in our study further support the neuroprotective effects of 1400W in the DFP model. It is worth noting that we did not observe changes in any of the cytokines/chemokines in serum or CSF at 6 weeks post-exposure, however, we did not test at an early time point. In contrary, we did observe significant differences in serum nitrite levels at both early and late time-points suggesting that the nitrite can serve as a peripheral biomarker for nitro-oxidative stress induced by OP CNA exposure (Hickey et al., 2001; Niess et al., 2000; Yurumez et al., 2007).

Previous studies from the rat DFP model have demonstrated the gliosis and neurodegeneration in both hippocampus and piriform cortex at 7d post-exposure (Flannery et al., 2016; Liu et al., 2012). Our studies at 7d and 12 weeks post-exposure showed a significant increase in iNOS and 3-NT, and also a significant increase in gliosis and neurodegeneration in the hippocampus, and mitigation by 1400W (Figs. 3, 5, 7–9). Reactive gliosis and iNOS mediate the production of proinflammatory cytokines, sensitize neurons to cause hyperexcitability (epileptiform spiking) and reduce seizure threshold by decreasing GABA_A currents, the pathognomonic feature of epileptogenesis (Puttachary et al., 2016b; Roseti et al., 2015; Steinborn et al., 2014; Vezzani et al., 2011a, 2013). Continuous integrated video-EEG monitoring for 12 weeks confirmed the development of epilepsy (spontaneous onset of both CS and NCS) and increased epileptiform spike rate in all 8 animals that were treated with the vehicle and MCM in this study. In contrast, in the 1400W-treated group, only 3 out of 8 animals developed epilepsy and there was a significant reduction in epileptiform spiking (Fig. 6D–I). Previous studies also demonstrated the development of epilepsy in the rat DFP and the other OP CNA model despite treating with MCM (de Araujo Furtado et al., 2012, 2010; Rojas et al., 2018). Other rat DFP/CNA studies also confirmed that the conventional therapeutic drugs have little or no long-term beneficial effects (Apland et al., 2014; Kuruba et al., 2018; Rojas et al., 2018; Wu et al., 2018) implying the need for the investigation of a novel follow-on therapeutic approach to target long-term neurotoxicity.

There are reports of cognitive dysfunction and anxiety or depression-like behavior in CNA survivors (Figueiredo et al., 2018; Nishiwaki et al., 2001) and in experimental models (Deshpande et al., 2014; Flannery et al., 2016; Phillips and Deshpande, 2016). Increased epileptiform activity impacts learning and memory (Bermeo-Ovalle, 2018; Ibrahim et al., 2014; Kleen et al., 2010). S100β upregulation in astrocytes and neurodegeneration have been known to alter synaptic plasticity and impact spatial learning and memory (Gerlai et al., 1995; Nishiyama et al., 2002; Plata et al., 2018). We investigated whether DFP exposure had any impact on cognition, motor function, and depression-like behavior. In this study, even though we observed increased epileptiform activity, intensely stained S-100β in reactive astrocytes, proinflammatory cytokines, and hippocampal neurodegeneration, we did not observe significant differences in learning and short-term memory in the MWM when the animals were tested between 10–16 and 28–34 days post- exposure. Previous studies reported the impact of acute exposure of DFP on cognition in the MWM only after 4 or 12 weeks of exposure (Brewer et al., 2015, 2013). In a chronic exposure DFP study, 35 days after the first DFP exposure affected spatial learning (Prendergast et al., 1997). None of these MWM studies did pre-exposure baseline tests as we did in this study. It is plausible that the prior learning may have impacted the learning and memory at 10–16 and 28–34 days post-exposure in this study in contrast to the published literature (Brewer et al., 2015, 2013; Prendergast et al., 1997). On an off note, the animals in our study were also subjected to other behavioral tests which may have interfered with MWM performance or vice versa. We should acknowledge, however, that the vehicle group spent significantly less time in the target quadrant at 9 days post-exposure when compared to the same probe trial prior to DFP exposure (Fig. 10Ei). The control and 1400W groups did not show this difference, which may suggest that DFP may have impacted the long-term memory at 9 days and 1400W may have mitigated the effect.

Delayed learning and memory deficits are also reported in an acute DFP exposure rat model. In contextual fear conditioning (CFC) test, the rats showed a significant deficit only at 36 days but not at 7 or 14 days post-exposure when the animals were tested at 24h after a three-shock conditioning paradigm with foot shocks at 0.7 mA every 30 seconds (Flannery et al., 2016). However, in the same study but in a different set of rats, a single paired shock conditioning paradigm with foot shocks at 0.5 mA showed significant deficits in contextual learning and memory (Flannery et al., 2016). In our MWM study too, we did not observe learning and memory deficits in less than 33 days of post-exposure which could be due to the effect of prior learning. Though both CFC and MWM tests are commonly interpreted as hippocampal-dependent learning and memory tests, the former also tests the function of the amygdala, which is useful. It is unclear why learning was unaffected in our study even though there was neurodegeneration in the hippocampus and the other parts of the brain which warrants further investigation. It may also likely that the variation in the initial severity of DFP-induced SE impacts the disease progression and comorbidity (Kanner et al., 2014). As reported by Flannery et al., 2016, in our study also DFP had no impact on the motor function as measured by rotarod and horizontal bar tests at the time-points tested. Likewise, we did not observe significant differences in depression-like behavior at 21 days post-exposure, which contradicts the previous study (Wright et al., 2010). The plausible reason being that we had tested only once without a pre-test session and we had also tested the animals prior to the DFP exposure which may have confounded the post-exposure test results.

In summary, an acute DFP exposure caused a significant increase of iNOS and 3-NT in the hippocampus and piriform cortex during both early and late stages of post-exposure suggesting the role of RNS in OP-induced pathogenesis. IHC revealed iNOS expression in microglia and 3-NT in neurons implying glial-mediated nitro-oxidative stress and treating with 1400W suppressed DFP-induced brain iNOS and 3-NT and serum nitrite levels at 24h post-exposure. Importantly, 1400W prevented DFP-induced mortality in <24h exposure. The serum nitrite and multiplex assays for cytokines and chemokines from hippocampal lysates at 6 weeks, and IHC of brain sections for gliosis and neurodegeneration at 7d and 12 weeks post-DFP confirmed the role of both nitro-oxidative stress and neuroinflammation in DFP-induced long-term neurotoxicity. Continuous long-term video-EEG monitoring for 12 weeks revealed increased epileptiform spiking and SRS suggesting hyperexcitability of neurons and the onset of epilepsy, and 1400W treatment significantly reduced the effect. There were no significant differences in neurobehavioral tests between the groups, which could be due to the effects of pre-training or testing at an early time-point, and perhaps conducting several tests in the same animals (although the animals were rested between the tests). Overall, 1400W treatment mitigated the majority of the DFP-induced neurotoxicity effects when it was administered as a follow-on therapy. Further testing in real nerve agent models such as soman or sarin and VX is required to advance 1400W as a therapeutic agent for CNA-induced long-term neurotoxicity.

HIGHLIGHTS.

• 1400W, a selective iNOS inhibitor, suppresses DFP-induced brain iNOS and 3-NT at 24h

• 1400W prevents DFP-induced mortality in <24h post-exposure

• 1400W suppresses DFP-induced SRS and epileptiform spiking during 12 weeks

• 1400W suppresses DFP-induced gliosis and neurodegeneration at 7d and 12 weeks

• 1400W reduces DFP-induced key proinflammatory cytokines/chemokine and serum nitrite

Glossary

2-PAM

2-pyridine aldoxime methyl chloride

3-NT

3-Nitrotyrosine

AChE

Acetylcholinesterase

AED

Anti-epileptic drug

ATS

Atropine sulfate

CNA

Chemical nerve agents

CS

Convulsive seizure

DFP

Diisopropyl fluorophosphate

DZP

Diazepam

EEG

Electroencephalography

FST

Forced swim test

HBT

Horizontal bar test

IHC

Immunohistochemistry

iNOS

Inducible nitric oxide synthase

KA

Kainic acid

MCM

Medical counter measures

MWM

Morris water maze

NCS

Nonconvulsive seizure

nNOS

Neuronal nitric oxide synthase

OP

Organophosphate

ROS/RNS

Reactive oxygen/nitrogen species

SE

Status epilepticus

SLUD

Salivation, lacrimation, urination, deification

SRS

Spontaneous recurrent seizures

WB

Western blot