Drugs and chemicals were obtained from the following sources: chlorpyrifos oxon (CPO; Chem Service, Inc., West Chester, PA, USA), DFP (Sigma‐Aldrich Co., St. Louis, MO, USA), PB (Sigma‐Aldrich Co.), physostigmine (PHY; Sigma‐Aldrich Co.), ethanol (Sigma‐Aldrich Co.), CORT (Steraloids Inc., Newport, RI, USA), 5,5‐dithio‐bis‐(2‐nitrobenzoic acid) (Sigma‐Aldrich Co.), tetraisopropyl pyrophosphoramide (Sigma‐Aldrich Co.), and acetylthiocholine iodide (Sigma‐Aldrich Co.). Rabbit Anti‐phospho STAT3^tyr705 antibodies were obtained from Cell Signaling, Inc. (RRID: AB_621843; Beverly, MA, USA). The materials used in glial fibrillary acidic protein (GFAP) ELISA previously have been described in detail (O'Callaghan et al. 1991; O'Callaghan 2002). Material used for additional tissue analyses were of at least analytical grade and purchased from various commercial sources.

Adult male (8–12 weeks of age; weighing approximately 22 g) C57BL/6J mice were purchased from Jackson Labs (RRID: IMSR_JAX:000664; Bar Harbor, ME, USA). Upon arrival, mice were individually housed in a temperature‐ (21 ± 1°C) and humidity‐controlled (50 ± 10%) colony room that was maintained under filtered positive‐pressure ventilation on a 12 h light (0600 EDT)/12 h dark cycle (1800 EDT). The plastic cages were 46 cm in length by 25 cm in width by 15 cm in height; cage bedding consisted of heat‐treated pine shavings spread at a depth of 4 cm. Mice were single housed and given ad libitum access to food (Harlan 7913 irradiated NIH‐31 modified 6% rodent chow) and water. All mouse procedures were performed according to protocols approved by the Institutional Animal Care and Use Committee of the Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, and the animal colony was certified by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC).

Mice (n = at least four/group, animals were arbitrarily assigned to groups by the experimenter) were given intraperitoneal (i.p.) injections of saline (0.9%), peanut oil (CPO vehicle), CPO (8 mg/kg,) DFP (4 mg/kg), PB (3 mg/kg) or PHY (0.5 mg/kg) in the morning and returned to their home cage (experimenter was not blinded). The doses of AChE inhibitors were selected based on their ability to produce the symptoms of cholinergic crisis [e.g., SLUD (salivation, lacrimation, urination, and defecation) and seizures] while displaying mortality below LD25. CORT was given in the drinking water (400 mg/L in 1.2% EtOH) for 4 days prior to AChE inhibitor or vehicle exposure. This regimen of CORT was chosen because of its ability to achieve high circulating levels of this hormone, similar to those achieved with repeated stress (Ganon and McEwen 1990), and because it was capable of producing significant immunosuppression as evidenced by involution of the thymus (O'Callaghan et al. 1991). Here, average thymus weights were significantly reduced from 45.4 ± 2.7 mg in vehicle‐treated to 12.8 ± 1.8 mg in CORT‐treated mice with average body weight of 21.5 ± 0.4 g. This study used twice the CORT concentration in the drinking water, but employed a 4‐day instead of a 7‐day CORT exposure regimen in comparison to the CORT regimen used in our original mouse model of GWI (O'Callaghan et al. 2015). Mice were killed by decapitation at 30 min (for AChE activity assay) or 6 h (for all other analyses) post‐AChE or vehicle injection. No differences were seen for the endpoints evaluated between mice exposed to peanut oil and saline (data not shown); therefore, peanut oil alone groups were excluded from analyses.

Immediately after decapitation, whole brains were removed from the skull and the hippocampus and cortex were dissected free‐hand on a thermoelectric cold plate (Model TCP‐2; Aldrich Chemical Co., Milwaukee, WI, USA) using fine curved forceps (Roboz, Washington, DC, USA). Brain regions from one side of the brain were frozen and stored at −80°C until subsequent isolation of total RNA. Brain regions from the other side of the brain were weighed and then homogenized with the aid of a sonic probe (model XL‐2005; Heat Systems, Farmingdale, NY, USA) in 10 volumes of hot (90–95°C) 1% sodium dodecyl sulfate (SDS). This tissue was then stored at −80°C until total protein assay and immunoassays of GFAP and immunoblots of pSTAT3^tyr705 were conducted. For the AChE activity assay, brains were bifurcated and one half of the brain was frozen and stored at −80°C until analysis.

The total RNA from the hippocampus and cortex were isolated using Trizol^® reagent (Thermo Fisher Scientific, Waltham, MA, USA) and Phase‐lock heavy gel (Eppendorf, AG Hamburg, Germany), and purified using RNeasy mini‐spin columns (Qiagen, Valencia, CA, USA). Total RNA (1 μg) was reverse transcribed to cDNA using Superscript III and oligo (dT)_12–18 primers (Thermo Fisher Scientific) in a 20 μL reaction. Real‐time PCR analysis of glyceraldehyde‐3‐phosphate dehydrogenase (endogenous control), tumor necrosis factor‐alpha (TNFα), C–C chemokine ligand 2, leukemia inhibitor factor, interleukin 6 (IL‐6), interleukin 1beta (IL‐1β), oncostatin M and GFAP was performed using an Applied Biosystems 7500 real‐time PCR system (Thermo Fisher Scientific) in combination with TaqMan^® chemistry. All PCR amplifications (40 cycles) were performed in a total volume of 50 μL, containing 1 μL cDNA, 2.5 μL of the specific Assay of Demand primer/probe mix (Thermo Fisher Scientific), and 25 μL of Taqman^® Universal master mix (Thermo Fisher Scientific). Sequence detection software (version 1.7; Applied Biosystems/Thermo Fisher Scientific) results were exported into Excel for further analysis. Relative quantification of gene expression was performed using the comparative threshold (ΔΔC_T) method. Changes in mRNA expression levels were calculated after normalization to glyceraldehyde‐3‐phosphate dehydrogenase. The ratios obtained after normalization are expressed as fold changes over corresponding controls.

Activation of the Janus kinase (JAK)‐STAT3 neuroinflammation effector pathway (O'Callaghan et al. 2014) was assessed by quantifying pSTAT3^tyr705 from immunoblots of tissue homogenates, with detection of fluorescent signals using an infrared fluorescence scanner (Licor Biosciences; Lincoln, NE, USA) as previously described (Sriram et al. 2004; Dinapoli et al. 2010; O'Callaghan et al. 2014). Briefly, following incubation with primary antibodies (rabbit anti‐phospho‐STAT3^tyr705[1 : 500]; RRID: AB_331586; Cell Signaling, Danvers, MA, USA), blots were washed with phosphate buffered saline with 0.1% Tween‐20 and incubated with anti‐rabbit fluorescent‐labeled secondary antibody (1 : 2500; RRID: AB_621843) for 1 h prior to scanning by Licor. We note the general requirement for using focused microwave irradiation sacrifice to preserve steady‐state in vivo phosphorylation does not apply in the case of pSTAT3^tyr705 (O'Callaghan and Sriram 2004).

GFAP was assayed in accordance with a previously described procedure (O'Callaghan 1991, 2002). In brief, a rabbit polyclonal antibody to GFAP (1 : 400; RRID: AB_10013382; DAKO, Carpenteria, CA, USA) was coated on the wells of Immulon‐2 microtiter plates (Thermo Labsystems, Franklin, MA, USA). The SDS homogenates and standards were diluted in phosphate‐buffered saline (pH 7.4) containing 0.5% Triton X‐100. Standards consisted of SDS homogenates of hippocampus with known concentration of GFAP and were prepared the same way as the samples. After blocking non‐specific binding with 5% non‐fat dairy milk, aliquots of the homogenate and standards were added to the wells and incubated. Following washes, a mouse monoclonal antibody to GFAP (1 : 250; RRID: AB_477010; Sigma‐Aldrich Co.) was added to ‘sandwich’ the GFAP between the two antibodies. An alkaline phosphatase‐conjugated antibody directed against mouse IgG (1 : 2000; RRID: AB_2340075; Jackson ImmunoResearch Labs, West Grove, PA, USA) was then added and a colored reaction product was obtained by subsequent addition of the enzyme substrate, p‐nitrophenol. Quantification was achieved by spectrophotometry of the colored reaction product at 405 nm in a microplate reader, Spectra Max Plus, and analyzed using Soft Max Pro Plus software (Molecular Devices, Sunnyvale, CA, USA). The amount of GFAP in the samples was calculated as micrograms of GFAP per milligram total protein.

Acetylcholinesterase activity was assessed via a protocol adapted from the Ellman method (Ellman et al. 1961; Lein and Fryer 2005). Briefly, one frozen cerebral hemisphere was homogenized with a sonic probe (mode XL‐2005; Heat Systems) in 10 volumes of sodium phosphate buffer (0.1 M, pH 8.0) with 1% Triton X‐100. Immediately following homogenization, the brains were centrifuged at 13 400 × g and the supernatant was removed and diluted 1 : 10 with sodium phosphate buffer prior to analysis. Following addition of a 5,5‐dithio‐bis‐(2‐nitrobenzoic acid)/tetraisopropyl pyrophosphoramide solution, samples were incubated for 5 min and reaction began when acetylthiocholine iodide was added. Quantification was achieved by spectrophotometry of the colored reaction product at 405 nm over a 10 min kinetic assay (16 cycles) in a Spectra Max Plus microplate reader and analyzed using Soft Max Pro Plus software (Molecular Devices). Acetylcholinesterase activity (μM substrate formed/min/mg total protein) was calculated based on the amount of total protein determined using the Pierce™ BCA Protein Assay kit (Thermo Fisher Scientific), per manufacturer's instructions. Acetylcholinesterase activity was normalized to saline control and is reported as a percentage.

For calculation of sample size, anova power analysis was performed using SigmaPlot (Systat Software, Inc., San Jose, CA, USA; RRID:SCR_003210; v12.5) using previously obtained mean differences and standard deviations between treated and control tissue with a power of 0.8 and α = 0.05; the sample size was estimated at four mice per group. Larger sample sizes of 5–7 were utilized to control for AChE inhibition induced mortality and endpoint variability [removal of outliers via Grubbs’ test (α = 0.05)] to achieve final sample sizes of at least n = 4 per group. All statistical analyses were performed using SigmaPlot. Two‐way anovas were conducted on log transformed values with Fisher least significant difference post‐hoc tests. Statistical significance was set at α = 0.05 (p < 0.05), and all graphs show mean ± SEM of raw values, unless otherwise stated.

Administration of the irreversible AChE inhibitor, DFP (used as a sarin surrogate), resulted in an increased expression of CCL2 and TNFα in cortex and or hippocampus (Fig. 1). Prior treatment with CORT in the drinking water for 4 days resulted in significant increases in all six cytokines/chemokines in both brain regions (Fig. 1) (with the exception of IL‐6 in hippocampus). Small increases (TNFα) or decreases (IL‐6) also were observed for the CORT alone condition in cortex. The findings for DFP were extended to another irreversible inhibitor of AChE, CPO (the oxon metabolite of the insecticide CPF). Like DFP, administration of CPO alone produced a significant increase in the expression of some proinflammatory mediators in cortex and hippocampus (Fig. 2) (and a small decrease in expression of IL‐6 in hippocampus). Also consistent with the data for DFP, CPO‐induced neuroinflammation was markedly increased by prior treatment with CORT, with the exception of mRNA for IL‐6 (Fig. 2). These results act to expand our DFP‐based GWI model (O'Callaghan et al. 2015) to include another GW‐relevant exposure, chlorpyrifos.

Epidemiologic studies have associated the reversible AChE inhibitor, PB, a drug given as a prophylactic measure against nerve agent exposure during the Gulf War, with the symptomology of GWI (Steele et al. 2012; White et al. 2016). PB does not readily cross the blood‐brain barrier (BBB) (Rice et al. 1997; Tuovinen et al. 1999; Song et al. 2002; Amourette et al. 2009); therefore, to assess the potential for reversible AChE inhibition in the central nervous system (CNS) to produce neuroinflammation, mice were exposed to PB or the CNS‐penetrant reversible AChE inhibitor, PHY, with or without CORT pretreatment. In general, neither agent produced significant neuroinflammation alone or with prior CORT treatment, except for a minor effect on TNFα (Figs 3 and 4). In addition, reversible AChE inhibitors tend to be anti‐inflammatory, reducing neuroinflammation significantly below control levels in some cases (Figs 3 and 4).

Proinflammatory cytokine signaling can activate the JAK/STAT3 pathway, as evidenced by the pSTAT3^tyr705 (O'Callaghan et al. 2014). Here, we observed an increase in pSTAT3^Tyr705 in the cortex and hippocampus of CORT+DFP and CORT+CPO exposed mice, an effect not seen following CORT+PB or CORT+PHY exposure (Fig. 5). This observation is consistent with the enhanced neuroinflammation induced by DFP and CPO following CORT pretreatment (Figs 1 and 2). Activation of the STAT3 pathway in the absence of astrogliosis, as is the case for exposure to DFP and CPO in our model (see below and O'Callaghan et al. 2015), is suggestive of the activation of microglia (O'Callaghan et al. 2014).

One hypothesis regarding AChE inhibitors and the development of GWI is that cholinergic effects of these compounds have made lasting physiological impacts that may contribute to the underlying cause of GWI (Golomb 2008). To evaluate this theory, the AChE enzyme activity was measured in mice treated with both irreversible and reversible AChE inhibitors with or without prior CORT treatment. As expected, DFP, CPO, and PHY exposure resulted in significant inhibition of enzyme activity 30 min after treatment and PB had no effect on AChE activity in the brain (Fig. 6). Interestingly, pretreatment with CORT significantly ‘recovered’ some of the AChE activity inhibited by DFP and CPO (Fig. 6). However, CORT pretreatment did not ameliorate PHY‐induced AChE inhibition (Fig. 6). Furthermore, despite claims that stressors or stress hormone may increase BBB permeability to PB (Friedman et al. 1996; Hanin 1996; Shen 1998), there was no effect of PB on AChE activity in the brain with or without prior CORT exposure (Fig. 6). The recovery of AChE activity with DFP and CPO following CORT pretreatment suggests that the enhanced neuroinflammation seen with these conditions is not dependent on a particular degree of AChE inhibition. Moreover, CORT treatment prior to DFP or CPO exposure brings AChE activity back to a level comparable with PHY exposure, which we have shown does not instigate neuroinflammation (Fig. 4).

Damage to the CNS by any type of insult, including neurotoxicants, results in hypertrophy of astrocytes at sites of injury (O'Callaghan and Sriram 2005; O'Callaghan et al. 2008). Injury‐induced activation of astrocytes is associated with an accumulation of the astrocyte intermediate filament protein, GFAP (O'Callaghan and Sriram 2005; O'Callaghan et al. 2014). Thus, GFAP expression and levels can be used as a biomarker of underlying CNS damage. With minor exceptions, no changes in GFAP or GFAP mRNA were seen with any exposure condition (data not shown), results consistent with our prior findings for DFP with and without CORT (O'Callaghan et al. 2015), and findings suggestive of a lack of underlying damage as a result of any treatment.