Abstract
Repeated stress has been reported to cause reversible impairment in the central nervous system (CNS). It was proposed that alterations in glutamatergic, cholinergic, and monoamine neurotransmitter systems after exposure to stress are initial CNS events contributing to this impairment and that exacerbation could occur with concurrent exposure to cholinesterase inhibitors. Effects of concurrent exposure to repeated stress and chlorpyrifos on activities of acetylcholinesterase (AChE), carboxylesterase, and choline acetyltransferase (ChAT); concentrations of excitatory amino acids, monoamines, and their metabolites; and maximum binding densities (B
max) and equilibrium dissociation rate constants (K
d) of glutamatergic N-methyl-
Higher organisms respond to stress through the hypothalamic-pituitary-adrenal axis (HPA) feedback system (De Kloet et al. 1998). In addition to the structures and chemical messengers directly associated with this axis, this stress system interacts with other neurotransmitter systems and structures of the central nervous system (CNS), influencing the integration of information, initiation of specific actions, and control of emotion (Konstandi et al. 2000). Even though the stress response tends to protect the organism from threats, long-term exposure to stressors can decrease the effectiveness of physiological and behavioral defense mechanisms against a variety of threats to health, including neurotoxic substances. Stress-induced compromise of defense mechanisms can happen to military personnel in combat conditions. For example, the consequences of concurrent exposure to repeated stress and neurotoxicants have been a concern because veterans of the 1991 Gulf War were exposed to a combination of many neurotoxic agents, including cholinesterase-inhibiting nerve agents and insecticides, and stress was proposed to be a contributing factor to symptoms noted in veterans after the war was over (Institute of Medicine [IOM] 2003; Marshall, Davis, and Sherbourne 2000; Sapolsky 1998). Individuals not in the military who are under stress may be exposed to cholinesterase-inhibiting insecticides as well, and long-term psychological effects from exposure to these insecticides have been reported in the literature (IOM 2003; Jamal 1997). In this study, an animal model was used to investigate interactions of stress and a cholinesterase-inhibiting insecticide, factors that have been suggested to contribute to illness in veterans of the 1991 Gulf War (IOM 2003).
Stress is able to affect neurotransmitter systems not directly associated with the hypothalamic-pituitary-adrenal axis. For example, stress elevates glucocorticoid hormones (Mizuguchi et al. 2001; Sapolsky, Krey, and McEwen 1986) and glucocorticoid hormones subsequently modulate excitatory amino acids and their N-methyl-
The cholinergic system has also been reported to be altered when people and animals are under stress (Kaufer et al. 1998; Sunanda, Rao, and Raju 2000). For example, repeated restraint stress was reported to reduce acetylcholinesterase (AChE) activity and transcription of genes coding for acetylcholine (Kaufer et al. 1998). In addition, chronic stress has been reported to alter concentrations of the monoamines, norepinephrine, dopamine, serotonin, and their metabolites in the brain (Konstandi et al. 2000; Liu et al. 2003; Sudo 1983; Sunanda, Rao, and Raju 2000).
The esterase-inhibiting organophosphorus (OP) insecticide, chlorpyrifos, was used for pest control during the 1991 Gulf War (IOM 2003). Cholinesterase inhibitors have also been reported to down-regulate muscarinic and nicotinic acetylcholine receptors (Ecobichon 2001), modulate the glutamatergic system by increasing extracellular glutamate concentrations and glutamate-mediated neurotransmission (Lallement et al. 1992; Pavlovsky, Rowne, and Friedman 2003), and alter concentrations of monoamines (Bloomquist et al. 2002; Dam, Seidler, and Slotkin 1999; Lewis 2003). This includes changes in dopamine uptake and turnover reported following exposures to chlorpyrifos (Bloomquist et al. 2002; Karen et al. 2001).
Glutamatergic, cholinergic, and monoaminergic pathways innervate and interconnect in the hippocampus and cerebral cortex (Cooper, Bloom, and Roth 1996; Webster 2001b). These pathways can be affected by stress and by cholinesterase inhibitors (Mangiavacchi et al. 2001; Mizoguchi et al. 2000). Therefore, we hypothesized that concurrent exposure to repeated stress and cholinesterase inhibitors could lead to effects on the glutamatergic, cholinergic, and monoamine systems not seen under conditions of exposure to each alone. Another region of interest for such potential interactions is the hypothalamus, a crucial region of the HPA axis under control of many neurotransmitters, including monoamines (Lowry et al. 2003).
The first major objective of this study was to assess the effect of repeated stress models on plasma levels of corticosterone, norepinephrine, and epinephrine, as they all have been used as physiological indicators of stress (Axelrod and Reisine 1984; Goldstein and Pacak 2001). Restraint, swim, and restraint with occasional swim were selected stress models for our study in rats. According to the literature, restraint stress relates to psychological stress, and was included due to reported effects on the glutamatergic system (McEwen 2000; Sunanda, Rao, and Raju 2000). Swim causes both physical and psychological stress and has been reported to affect the cholinergic system (Kaufer, Friedman, and Soreq 1999). Restraint with occasional swim was employed with the intention of creating a model of unpredictable physical and psychological stress for our study. Included in this portion of the study was evaluation of the effects of chlorpyrifos on the indices of stress noted above.
The second major objective of the study described below was to examine the effects of repeated stress, chlorpyrifos, and their interactions on brain glutamatergic, cholinergic, and monoaminergic systems. This was done by measuring concentrations of monoamines, excitatory amino acids and NMDA glutamatergic receptors, and total muscarinic receptors and activities of the enzymes acetylcholinesterase, carboxylesterase, and choline acetyltransferase in the hippocampus and cerebral cortex. Measurements of these end points in the hypothalamus were also included, where feasible, as that region of the brain is an important contributor to the stress response.
MATERIALS AND METHODS
Animals, Stress Equipment, and Chemicals
Adult male Long-Evans rats obtained from Harlan Sprague-Dawley (Indianapolis, IN) were used for these studies. For studies on glutamate, cholinergic enzymes, and monomines, the 56 rats were 100 to 120 days of age and weighed 230 to 265 g; for studies on muscarinic and glutamatergic receptors, the 48 rats were 54 to 82 days of age and weighed 246 to 322 g. Rats were acclimated to their surroundings for 7 days. They were housed individually with food and water provided ad libitum. Room temperature was kept at 21°C to 23°C with a light cycle of 7
Restraint stress was performed using a plexiglass cylinder with air holes, 6.5 cm in diameter and 22 cm in length that could be closed at both ends. Swim stress was performed in a 30-gallon fish tank held at 21°C to 23°C that was divided equally into four sections so four rats could be tested in the same 30-min period, decreasing the time needed to stress a relatively large number of rats.
Chlorpyrifos (O,O′-diethyl-3,5,6-trichloro-2-pyridinyl-phosphorothioate, 99.5% pure) was obtained from Chem Service (West Chester, PA). Corticosterone radioimmunoassay kits were purchased from ICN Biomedicals, (Costa Mesa, CA). [3H]MK-801 (5H-dibenzo[a,d]cyclohepten-5,10-imine-10,11-dihydro-5-methyl-(5Sm, Z)-2-butenedioate(1:1), dizocilpinemaleate) is the NMDA antagonist. It had specific activity of 25 Ci/mmol and concentration of 1 mCi/ml (American Radiolabeled Chemicals, St. Louis, MO). 1-Quinuclidinyl[phenyl-4-3H]benzylate ([3H]QNB) is a nonselective muscarinic receptor antagonist. It had specific activity of 42 Ci/mmol and concentration of 1.0 mCi/ml (Amersham Biosciences, Piscataway, NJ). [Acetyl-1-14C]-Coenzyme A with a specific activity of 50 mCi/mmol and a concentration of 100 μCi/ml (ICN Biomedicals) was used as a radioactive substrate for determining choline acetyltransferase (ChAT) activity. Experiments with radioactive material were performed in accordance with Virginia Tech guidelines. Monoamine reagents (norepinephrine bitartrate, epinephrine hydrochloride, dopamine hydrochloride, 3,4-dihydroxyphenylacetic acid [DOPAC], 5-hydroxytryptamine oxalate [serotonin], 5-hydroxyindole-3-acetic acid [5-HIAA], and isoproterenol hydrochloride [an internal standard of monoamines in the high-performance liquid chromatography assay]) were purchased from Sigma Aldrich (St. Louis, MO). Other chemicals were high performance liquid chromatography (HPLC) or analytical grade. They were purchased from Sigma Aldrich, VWR (Suwanee, GA), and Fisher (Suwanee, GA).
Experimental Design and Sample Collection
The experimental design was a generalized, randomized, complete block design. Treatment structure was 4 × 2, which consisted of four types of repeated stress (handling as a control, restraint, swim, and restraint with occasional swim) and subcutaneous injection either with corn oil or chlorpyrifos. Seven to eight rats were randomly allocated into each cell of the 4 × 2 treatment design. Treatment and sacrifice were performed in two experimental blocks separated by 1 day (enzymatic and neurotransmitter experiments) or 8 days (receptor experiments). Body weight, plasma corticosterone, and serum catecholamines were analyzed as repeated measurements on various days. Concentrations of excitatory amines (glutamate, aspartate), NMDA and total muscarinic receptors, cholinergic enzymes, and monoamines and their metabolites in brain were assessed with a single measurement at sacrifice.
Control rats were handled 5 days/week over the 28-day testing period. Rats were stressed between 9
Concentrations of Plasma Corticosterone and Catecholamines
Corticosterone concentrations were determined using blood samples collected from the orbital sinus in heparinized micro-centrifuge tubes and spun on a Beckman microfuge Lite centrifuge at 3500 × g for 3 min. Plasma samples were kept at –70°C until assayed. Corticosterone concentrations were assayed using a 125I radioimmunoassay kit for rats and mice and autogamma detection (ICN Biomedicals) following the manufacturer’s directions. A CROBA II auto-gamma counter (Packard Bioscience Company, Meriden, CT) was used for gamma counting of the precipitate. Detection limit was in the range of ng/ml of plasma.
The catecholamines, norepinephrine and epinephrine, were analyzed in plasma by reverse-phase HPLC coupled with electrochemical detection (Jussofie, Lojewski, and Hiemke 1993; Wang, Fice, and Yeung 1999), with sensitivity in a range of ng/ml plasma. Norepinephrine and epinephrine were extracted from plasma by acid-washed alumina in Tris buffer pH 8.6. Isoproterenol, an internal standard (final concentration of 0.5 μM), was added to the plasma before the extraction. Acid-washed alumina was washed twice with distilled water and desorbed with a mixture of 0.04 M phosphoric acid–0.2 M acetic acid (20:80, v/v, pH 1.5 to 2.0). Then, the supernatant was filtered through an Acrodisc LC 13-mm syringe filter with a 0.2-μm polyvinylidene (PVDF) membrane into HPLC vials. The HPLC system consisted of a Hewlett Packard Series 1100 Quaternary Pump, degasser, autosampler, and electrochemical detector (Agilent Technologies, Wilmington, DE). Separation was performed on a reverse-phase C18 analytical column (Nucleosil 100 3 μ, 250 × 4 mm; Macherey-Nagel, Düren, Germany) preceded by a guard column (Nucleosil 100 C18 3 μ, 8 × 4 mm). Electrochemical detection was used in the oxidation mode set at +0.35 volt. Isocratic elution was performed at a flow rate of 0.6 ml/min. The mobile phase consisted of 0.1 M sodium acetate, 25 mM citric acid, 134 μM ethylenediaminetetraacetic acid disodium dihydrate salt (EDTA), 230 μM octanesulfonic acid, and 6% methanol at pH 4.7.
Concentrations of Excitatory Amino Acid and NMDA Receptors
Glutamate concentrations in the hippocampus and cerebral cortex were determined using reverse-phase HPLC with fluorescence detection (Phillips and Cox 1997; Piepponen and Skujins 2001). Brain tissues were homogenized in ice-cold Ringer’s solution (147 mM NaCl, 1.2 mM CaCl2, 2.7 mM KCl, 1.0 mM MgCl2, 0.04 mM ascorbic acid) along with an internal standard (homoserine at a final concentration of 2 μM) using a cell sonicator (Heat System-Ultrasonic, Farmingdale, NY) before centrifugation at 13,000 ×g, 4°C for 5 min. Supernatants were filtered through Acrodisc LC 13-mm syringe filters and diluted 20,000- to 24,000-fold with Ringer’s solution prior to amino acid analysis. The working derivatization reagent was freshly prepared using 5 mg of O-phthaladehyde in 5 ml of 0.10 M sodium tetraborate and 15 μl of 14.3 mol/L of β-mercaptoethanol. The filtrate was mixed manually with the derivatization reagent (1:2 by volume) 2 to 4 min prior to sample injection on the analytical column. The external standard series (glutamate, aspartate, glycine, and homoserine) was prepared in 1:1 water-methanol and run for quantitative analysis. A Hewlett Packard Series 1100 Quaternary Pump with a degasser and au-tosampler was used for HPLC (Agilent Technologies). The separation was performed on a C18 reversed-phase analytical column (Luna C18 (2) 5 μ, 150 × 4.6 mm, Phenomenex, Torrance, CA). The mobile phase was 88% 0.05 M disodium phosphate pH 6.1 and 12% acetonitrile. Flow rate of the mobile phase was 1.2 ml/min until glycine (the last analyte) appeared. Then, the column was washed with deionized water, 30% acetonitrile, and 100% acetonitrile to wash out other amino acids before equilbration with the mobile phase, for 14 min, in preparation for the next sample. Fluorescence detection was used with an excitation wavelength of 330 nm and emission wavelength of 425 nm. Detection limit was in the range of 0.05μM per 20 μl of injection and the tissue sample needed was 0.02 to 0.05 g.
The maximal number of receptors (B max) and equilibrium dissociation constant (K d) of NMDA glutamate receptors were determined by measuring binding at various concentrations of the radiolabeled NMDA antagonist [3H]MK-801. The radioligand was incubated with membrane suspensions to determine total binding. Specific binding was determined as total binding minus nonspecific binding (in the presence of 50 μM unlabeled MK-801) at each concentration (Bellinger et al. 2002; Betancourt and Carr 2004; Motulsky and Neubig 1997).
Brain tissues assayed for receptor studies were thawed in 10 volumes of 0.32 M sucrose. The tissue was homogenized followed by centrifugation at 1000 ×g, 4°C, for 10 min. The supernatant was divided into two parts; one for the NMDA receptor binding assays, the other for the muscarinic binding assay (see below).
For the NMDA receptor assay, brain supernatants were recentrifuged at 40,000 × g, 4°C, for 10 min. The pellet was washed with 5 mM Tris-HCl pH 7.4 by resuspending and centrifuging at 40,000 × g, 4°C, for 30 min. The pellet was washed with deionized water and this pellet was resuspended in less than 1 ml of 5 mM Tris-HCl and kept at –70°C until time for the assay. On the assay day, the pellet was thawed and washed twice with 5 mM Tris-HCl pH 7.4 by resuspending and centrifuging. Then it was resuspended in sufficient 5mM Tris-HCl to provide 0.1 to 0.4 mg of membrane protein/ml (Bio-Rad Laboratories, Richmond, CA). A membrane suspension volume of 25 μl was added to microplate wells containing 125 μl of 5 mM Tris-HCl pH 7.4. Glutamate, glycine, and spermidine (all at final concentration of 10 μM) were present in the buffer for maximum activation of NMDA receptors. [3H]MK-801 in a volume of 25 μl at six different final concentrations between 0.1 and 15 nM was added to the microplate. These membrane suspensions were incubated at 25°C for 120 min either with 50 μM of unlabeled MK-801 (to determine nonspecific binding) or without unlabeled MK-801 (to determine total binding). The reaction was terminated by rapid filtration over FilterMat (1 μm; Molecular Devices, Sunnyvale, CA) using a cell harvester (Skatron Instruments; available through Molecular Devices, Sunnyvale, CA). The filters were dried in an oven at 50°C for 1 h before they were punched into scintillation vials. Liquid scintillation cocktail (Ultima Gold F; PerkinElmer Life and Analytical Sciences, Boston, MA) was added at a volume of 3 ml. Radioactivity was measured using a liquid scintillation counter (LS6500; Beckman Coulter, Fullerton, CA). Counts per minute of each sample were converted to fmol/mg protein. B max and K d were calculated by GraphPad Prism Software package version 4 (Motulsky and Christopoulos 2003).
Concentrations of Cholinergic Muscarinic Receptors and Enzymes for Biotransformation and Synthesis of Acetylcholine
For determination of muscarinic receptor (mAChR) numbers, the original brain supernatant, prepared as described above, was centrifuged at 48,000 × g, at 4°C for 10 min in a Beckman L8-80M ultracentrifuge (Palo Alto, CA). The pellet was washed twice with cold 50 mM Tris-HCl buffer (containing 120 mM NaCl; 5 mM KCl; 2 mM CaCl2; 1 mM MgCl2) by resuspending the pellet in fresh cold buffer and centrifuging. The final pellet was resuspended in cold buffer to provide 0.1 to 0.4 mg of membrane protein/ml (Bio-Rad Laboratories). The mAChR binding was determined using six concentrations between 0.1 and 10 nM of the muscarinic antagonist [3H]QNB in a volume of 25 μl. The assay was performed in duplicate in a 96-well microplate. Atropine sulfate 1 μM (25 μl) was used as an unlabeled antagonist. Membrane suspension (100 μl) was incubated without atropine (total binding) or with atropine (nonspecific binding) at 37°C for 60 min. The incubation was terminated and the radioactivity measured as described above for NMDA receptors.
AChE activity in blood and brain regions (hippocampus and cerebral cortex) was analyzed by microplate spectrophotometry at 412 nm over a 30-min incubation period at room temperature using acetylthiocholine iodide as a substrate (Correll and Ehrich 1991). The whole blood was diluted to 1:1000 with phosphate buffer pH 8.0. Iso-OMPA (N,N′,N″,N′″-tetraisopropylpyrophosphamide) was added to the diluted blood at a concentration of 10–8 M to inhibit pseudocholinesterase activity. The diluted blood or diluted brain homogenate (0.4 mg in 50 μl) was added to microplate wells along with 150 μl of 0.1 M phosphate buffer at pH 8.0, 50 μl of 6 mM of 5,5′-dithio-bis-2-nitrobenzoic acid, and 50 μl of 4.5 mM of acetylthiocholine in triplicate. Results were expressed as nmol/min/mg protein, with protein determined using the Bio-Rad assay kit (Bio-Rad Laboratories).
Carboxylesterase activity in brain was determined by measuring hydrolysis of phenyl valerate (an ester substrate) to phenol, which then undergoes a redox reaction with potassium ferri-cyanide to form a product with absorbance at 510 nm (Correll and Ehrich 1991). For this assay, brain tissues were prepared as for the AChE assay then further diluted 1:8, with 25 μl of diluted sample, 125 μl buffer, and 50 μl of phenyl valerate (1.1 mg/ml in 0.03% Triton X-100), and incubated at 37°C for 15 min in triplicate. Sodium lauryl sulfate 5% was added to stop the reaction and 0.4% potassium ferricyanide was added to initiate the redox reaction. Absorbance at 510 nm was read 15 min later. Carboxylesterase activity was calculated using a standard curve created by reaction of phenol and potassium ferricyanide. Carboxylesterase activity was reported as nmole phenyl valerate hydrolyzed/min/mg protein.
ChAT was determined using our laboratory’s modification of a radiochemical microassay (Chambers and Chambers 1989; Flory and Correll, unpublished). Brain samples were homogenized with 50 mM Tris-HCl-5 mM EDTA buffer, pH 7.5, then further diluted with 2.5% Triton X-100 in 10 mM EDTA at a ratio of 4:1. The incubation mixture in a total volume of 7 μl consisted of the following reagents in millimolar concentrations: sodium phosphate buffer, pH 7.4, 50; sodium chloride, 300; choline bromide, 8; disodium EDTA, 20; eserine sulfate, 0.1; [14C]acetyl-CoA, 0.2; and homogenate containing 8 to 14 μg of protein. The reaction was incubated at 37°C for 15 min and stopped by adding 0.4 ml of 70 mM cold phosphate buffer, pH 7.4. Each sample was then washed with 1.6 ml of phosphate buffer. The washing solution was transferred to a scintillation vial that contained 2 ml of 0.5% sodium tetraphenylborate in butyronitrile and 10 ml of Scintilene (scintilation cocktail). Gentle shaking extracted the [14C]acetylcholine product into the cocktail, leaving the [14C]acetyl-CoA in the aqueous phase before beta radiation was counted by liquid scintillation (Beckman LS6500; Beckman Coulter, Fullerton, CA). Detection limit of ChAT activity was in the range of pmole/min/mg protein.
Concentrations of Brain Monoamines and Their Metabolites
Concentrations of monoamines and their metabolites in the hippocampus, cerebral cortex, and hypothalamus were determined by using HPLC combined with electrochemical detection (HPLC-ECD) (Jussofie, Lojewski, and Hiemke 1993). Brain samples (0.02 to 0.05 g) were homogenized in 250 μl of mobile phase pH 4.7 and an internal standard (1 μM isoproterenol hydrochloride; Sigma, St. Louis, MO) was added. Brain tissues were homogenized by using a cell sonicator (Heat System-Ultrasonic, Farmingdale, NY) with speed set at 6 cycles for 30 s. Homogenates were centrifuged by a Beckman Microfuge Lite at 10,000 × g for 5 min and supernatants were transferred and kept in HPLC light-protective vials at –20°C until assayed. On day of assay, supernatants were thawed on ice, filtered, and handled as described above for plasma norepinephrine. External standard solutions of norepinephrine bitartrate, epinephrine hydrochloride, dopamine hydrochloride, 3,4-dihydroxyphenylacetic acid (DOPAC), 5-hydroxytryptamine oxalate (serotonin), 5-hydroxyindole-3-acetic acid (5-HIAA), and isoproterenol hydrochloride were used for quantitative analysis. Range of concentrations used was 5 × 10–8 M to 1 × 10–5 M.
Statistical Analysis
Data were analyzed by two-way analysis of variance (ANOVA) using the SAS program (version 8.2; SAS Institute, Cary, NC). To test the effects of repeated stress on body weight and concentrations of plasma corticosterone before injection of chlorpyrifos from day 0 to day 24, the change of body weight and concentrations of plasma corticosterone were analyzed by mixed model ANOVA with post hoc comparisons as tests of simple main effect of repeated stress on separate days. To hold the overall alpha value to 0.05, the comparisonwise alpha level was adjusted by Bonferroni correction.
The main effects of repeated stress and chlorpyrifos on plasma corticosterone on day 24 were analyzed by two-way ANOVA using the SAS program (version 8.2; SAS Institute). Post hoc comparisons of stress effect were tested by Tukey’s test. The interaction of repeated stress with chlorpyrifos was tested at time points after the chlorpyrifos dosing. The main effects of repeated stress and chlorpyrifos, and the interaction of repeated stress and chlorpyrifos at day 28 on plasma concentrations of corticosterone, norepinephrine, and epinephrine, concentrations of brain glutamate and aspartate, B max and K d of NMDA and total muscarinic receptors, activities of blood AChE, brain AChE, brain carboxylesterase, choline acetyltransferase, and monoamines and their metabolites in the hippocampus, cerebral cortex, and hypothalamus were analyzed by two-way ANOVA. Post hoc comparison was tested using Tukey’s test for stress effects and by simple main effect for the interaction of stress and chlorpyrifos within stress treatment. A p < .05 was considered a significant difference and a p ≤ .08 was arbitrarily considered a trend toward statistical significance.
RESULTS
Although rats gained weight during the study, there was no main effect of stress prior to administration of chlorpyrifos on day 24. When measured 2 days after dosing, chlorpyrifos was found to significantly decrease change of body weight from day 0 to day 26 (df = 1, 55, p = .018) (chlorpyrifos-treated rats gained 62 ± 4 g over that time period; controls gained 75 ± 4 g). However, the overall gain in body weight from day 0 to day 28 between rats treated with chlorpyrifos and corn oil was not significant (df = 1, 55, p = .099; group average weight gains were 69 and 78 g, respectively).
Concentrations of Plasma Corticosterone and Catecholamines
Blood samples taken 10 to 40 min after stress on days 8 and 17 demonstrated increased concentrations of plasma corticosterone for swim stress and restraint with swim stress compared to samples from control rats or rats exposed to restraint alone (df = 1, 96, p < .0001, for both days and for the comparisons of swim versus control, swim versus restraint, restraint with swim versus control, restraint with swim versus restraint, Figure 1). Concentrations of plasma corticosterone in restrained rats were not different from those in control rats on days 8 and 17.
On day 24, there were no main effects of repeated stress or chlorpyrifos treatment, nor was there an interaction of repeated stress and chlorpyrifos treatment on concentrations of plasma corticosterone (df = 3, 99, p = .09; df = 1, 99, p = .23; df = 3, 99, p = .20, respectively). These blood samples were taken 7 h after all groups except the controls experienced restraint stress and 3 h after chlorpyrifos injection. Concentrations of plasma corticosterone on day 28, collected 10 to 30 min after stress were 1196 ± 59 and 1136 ± 63 ng/ml in the swim group and the restraint with swim group, respectively, compared to 462 ± 59 and 465 ± 59 ng/ml in the handling control group and the restraint group (Table 1).
Blood samples taken on day 28 indicated that swim and restraint with swim decreased plasma norepinephrine and epinephrine to levels below those of control (norepinephrine, swim versus control, df = 1, 47, p = .03; norepinephrine, restraint with swim versus control, df = 1, 47, p = .003; epinephrine, swim versus control, df = 1, 47, p = .01; epinephrine, restraint with swim versus control, df = 1, 47, p = .002; Table 1). Although our stressors that included swim had an effect on all three plasma indices of stress, chlorpyrifos administration did not affect any of these measures in the samples taken 4 days after it was administered (data not shown). There were no interactions of repeated stress and chlorpyrifos on plasma corticosterone, norepinephrine, or epinephrine.
Although there was no obvious behavioral evidence of cholinergic toxicity as rats were observed, chlorpyrifos adminstration produced a statistical trend toward decreased swimming time when this stress was experienced 4 days after dosing (on day 28 df = 1, 23, p = .08). Swimming times were 27 ± 1.9 min for chlorpyrifos-treated rats, whereas rats not given chlorpyrifos swam between 29.4 and 30.0 min. Chlorpyrifos-treated rats appeared less able to tolerate being in the water than rats that were given corn oil, as they had to be rescued from drowning before the end of the 30 min swimming period.
The Glutamatergic System: Concentrations of Excitatory Amino Acids and NMDA Receptors
Restraint with swim stress resulted in the highest concentrations of glutamate in the hippocampus (11.139 ± 0.315 μmole/g tissue) when determined at time of sacrifice (day 28). This concentration represented a trend toward being greater than in rats experiencing swim stress alone (10.018 ± 0.304 μmole/g tissue, df = 1, 45, p = .06). The latter values did not differ from hippocampal glutamate concentrations in handled (control) or restrained rats. Chlorpyrifos did not affect concentrations of hippocampal glutamate. In the cerebral cortex, no main effects of repeated stress, or chlorpyrifos, were noted nor was there an interaction of stress and chlorpyrifos on concentrations of glutamate.
Swim and restraint with swim increased hippocampal aspartate when compared with levels in control (handled) rats (df = 1, 45, p = .0035; df = 1, 45, p = .0003). When compared to control (handling), restraint stress did not affect hippocampal aspartate levels. Chlorpyrifos reduced the increase of hippocampal aspartate of rats that were restrained and swum (df = 1, 45, p = .006; Figure 2) but not of swum rats (df = 1, 45, p = .19). Although effects of stress were noted in the hippocampus, repeated stress, chlorpyrifos, or exposure to both repeated stress and chlorpyrifos did not affect concentrations of aspartate in the cerebral cortex.
B max of NMDA receptors in the hippocampus was not affected by repeated stress; however, K d of samples from restrained rats (1.84 ± 0.14 nM) was higher than samples from control (1.38 ± 0.15 nM; df = 1, 52, p = .03) and samples from restrained with swim rats (1.33 ± 0.14 nM; df = 1, 52, p = .01). Chlorpyrifos or interactions of chlorpyrifos and repeated stress did not alter B max and K d of NMDA receptors in the hippocampus. Furthermore, repeated stress, chlorpyrifos, or interactions of repeated stress and chlorpyrifos did not affect B max or K d of NMDA receptors in the cerebral cortex (Table 2) and hypothalamus (data not shown).
The Cholinergic System: Concentrations of Muscarinic Receptors and Enzymes of Acetylcholine Biotransformation and Synthesis
In contrast to results with NMDA receptors, ANOVA demonstrated interactions of repeated stress and chlorpyrifos on B max and K d of total muscarinic receptors in the cerebral cortex (df = 3, 54, p = 002; df = 3, 55, p = .01). Further analysis (Table 2) demonstrated that chlorpyrifos decreased B max of total muscarinic receptors in the cerebral cortex in the restrained rats (df = 1, 54, p = .009) and the swum rats (df = 1, 54, p = .001) when compared to their corn oil-treated counterparts. Table 2 also indicates that chlorpyrifos decreased K d of total muscarinic receptors in the cerebral cortex in the swum rats (df = 1, 55, p = .007). Neither B max or K d of total muscarinic receptors in the hippocampus, however, was affected by stress or by chlorpyrifos.
Blood AChE activity was not affected by repeated stress, but was markedly inhibited by treatment with chlorpyrifos, even when samples were taken 4 days later (activity 29% ± 4% of control at that time). Repeated stress did not affect activities of brain AChE and carboxylesterase in the hippocampus (data not shown), whereas chlorpyrifos inhibited activities of these enzymes in the hippocampus by 80% ± 3% and 46% ± 2%, respectively (df = 1, 45, p < .0001 for these comparisons). There were no interactions of repeated stress and chlorpyrifos on activities of these esterases in the hippocampus. Esterase activities were also measured in the cerebral cortex, where it was also noted that repeated stress did not affect the activity of AChE. However, in this brain region, restraint with swim stress decreased activities of carboxylesterase when compared to groups of rats that were handled, restrained, or swum (df = 1, 45, p < .0001; df = 1, 45, p < .0001; df = 1, 45, p = .0004, respectively), as shown in Figure 3. Chlorpyrifos (CPF) inhibited activities of AChE (by 85% ± 5%; df = 1, 45, p < .0001) and carboxylesterase (Figure 3, df = 1, 45, p < .0001) in the cerebral cortex. Even though restraint with swim stress and chlorpyrifos both inhibited carboxylesterase activity in the this brain region, there was no interaction of this repeated stress and chlorpyrifos (df = 1, 45, p = .23).
No effects of repeated stress or chlorpyrifos were noted on ChAT either in the hippocampus or cerebral cortex (data not shown). Enzyme concentrations in brain tissues from all groups of rats ranged from 1.42 to 1.58 pmol/min/mg protein.
Concentrations of Brain Monoamines and Their Metabolites
Repeated stress in our study did not alter concentrations of dopamine, serotonin, DOPAC, or 5-HIAA in the hippocampus and cerebral cortex, but swim stress decreased concentrations of norepinephrine in the hippocampus (df = 1, 45, p = .003; Table 3).
In the hypothalamus, repeated stress (swim and restraint with swim) affected concentrations of norepinephrine, dopamine, DOPAC and serotonin (p < .01), but not the concentration of 5-HIAA. The effects of swim and restraint with swim stress on norepinephrine were significant when compared to control (df = 1, 44, p < .0001) or restraint stressed rats (df = 1, 44, p ≤ .002; Table 3). Swim stress increased concentrations of dopamine over that of control (df = 1, 44, p = .05) and restraint stressed rats (df = 1, 44, p = .01). Swim and restraint with swim increased concentrations of DOPAC over that of control (df = 1, 44, p ≤ .001) and restraint rats (df = 1, 44, p = .002). Restraint stress decreased concentrations of serotonin in the hypothalamus more than swim (df = 1, 44, p ≤ .02) and restraint with swim (df = 1, 44, p = .01; Table 3).
Chlorpyrifos administered 4 days before sample collection decreased concentrations of norepinephrine in the hippocampus when compared to corn oil treatment (df = 1, 45, p = .04; Table 4), but exposure to this insecticide was without effects on concentrations of dopamine, DOPAC, or 5-HIAA in this brain region, as well as in the cerebral cortex and hypothalamus (Table 4). However, chlorpyrifos decreased concentrations of serotonin in the hippocampus and hypothalamus (df = 1, 45, p = .04; df = 1, 44, p = .02, respectively), with effects noted in the restrained rats (df = 1, 44, p = .03) and restrained with swim rats (df = 1, 44, p = .02), but not in the control or swum rats (Figure 4). In contrast to chlorpyrifos-induced decreases in serotonin in the hippocampus and hypothalamus, this insecticide increased concentrations of serotonin in the cerebral cortex (df = 1, 45, p = .02, Table 4), with statistical analysis suggesting a trend toward interactions between concurrent exposure to repeated stress and chlorpyrifos in the cerebral cortex (df = 3, 45, p = .06) and hypothalamus (df = 3, 44, p = .08). The chlorpyrifos-induced increased concentrations of serotonin in the cerebral cortex were evident in the control rats (df = 1, 44, p = .01) and restrained rats (df = 1, 44, p = .02), but not in the swum and restrained with swim rats (Figure 5).
DISCUSSION
The present study examined effects of repeated stress with and without concurrent exposure to the insecticide chlorpyrifos on glutamatergic, cholinergic, and monoaminergic systems of the hippocampus, cerebral cortex, and hypothalamus. Results identified several effects attributed to exposure of rats to repeated stress. These included elevated plasma corticosterone and decreased plasma norepinephrine and epinephrine in swim and restraint with swim rats, elevated aspartate in the hippocampus of swim and restraint with swim rats, decreased carboxylesterase activity in the cerebral cortex of restraint with swim rats, and decreased norepinephrine in the hippocampus and hypothalamus of rats that were swum (with and without restraint). Results also identified effects attributed to administration of the cholinesterase-inhibiting organophosphorus compound, chlorpyrifos. These included capability of chlorpyrifos to decrease elevated aspartate in the hippocampus of restraint with swim rats, decreased esterase activities in plasma and brain of all rats, and decreased B max of total muscarinic receptors in the cerebral cortex of restraint with swim rats. There was an interaction of chlorpyrifos and stress (swim) on B max of these receptors. Chlorpyrifos administration also affected serotonin levels, with an interaction of stress and chlorpyrifos in the cerebral cortex. Although we hypothesized that repeated stress and chlorpyrifos would have effects on neurotransmitter systems greater than either alone, results of the extensive investigations presented here indicate that interactions were few. Figure 6 diagrams relationships of our proposed model for the concurrent exposure to repeated stress and chlorpyrifos on the glutamatergic and cholinergic systems that have been elucidated based on our results, with pathways still to be investigated indicated by ?. Although the literature suggested numerous interconnections among stress, cholinesterase inhibition, and hippocampal effects, our experiments did not establish a direct relationship between acetylcholine increases that follow chlorpyrifos-induced AChE inhibition and effects on the glutamatergic system. Our studies, however, did note expected effects of chlorpyrifos on esterase activities, and confirmed that stress can have effects on the glutamatergic system.
Indices of Repeated Stress: Effects on Body Weight and Plasma Corticosterone, Norepinephrine, and Epinephrine Levels
Loss of body weight is a common response to many stressors. However, in this study, repeated stress did not have a notable effect on body weight. The present work supports studies of Hancock et al. (2003), which also reported that rats that underwent restraint and restraint with swim stress did not have body weights different from rats that were only handled. However, the swum rats in Hancock’s study had lower body weight than restrained rats. The discrepancy could be due to the amount of swimming. Rats in this study were swum one day per week, whereas in Hancock’s experiment the rats were swum 5 days per week.
Other groups have reported weight loss following restraint of rats for 0.5 to 6 h/day or swimming for 2 h/day for 21 to 28 days (Konarska, Stewart, and McCarty 1989 Konarska, Stewart, and McCarty 1990; Magarinos and McEwen 1995; Mizoguchi et al. 2001). In the study of Margarinos and McEwen (1995), concentrations of plasma corticosterone were very high on days 1, 7, and 14, but were decreased on day 21, suggesting rats could acclimate to restraint. Restraint rats in our study did not have high plasma corticosterone levels. Acclimation to restraint stress was noted in the present study by the increased ease with which rats went into the plexiglass tubes. Therefore, although noted by other investigators, our stress models were not sufficient to cause body weight change.
A chlorpyrifos dose of 160 mg/kg did not result in obvious clinical signs and only caused a transient effect on body weight. Lack of clinical evidence of cholinergic poisoning in the present study is supported by the work of Karanth and Pope (2003), who saw only a few overt signs of cholinergic toxicosis in rats given chlorpyrifos at the maximal tolerated dose (279 mg/kg). This higher dose did, however, decrease body weight of adult rats; whereas 60% of the maximal tolerated dose used in the present study did not.
Swim stress and restraint with swim stress increased concentrations of plasma corticosterone more than control and restraint alone but only when blood samples were drawn 10 to 40 min after swim stress. Plasma samples collected for corticosterone determinations 7 h after restraint stress or 3 h after chlorpyrifos were not different from corticosterone levels in rats that were only handled (the controls), indicating that effects of swim stress on corticosterone levels, although notable, were transient. Our results were similar to previous studies (Hancock et al. 2003; Konstandi et al. 2000; Magarinos and McEwen 1995; Mizoguchi et al. 2001), which also demonstrated increases in plasma corticosterone shortly after restraint and/or swim stress. The study of Mizoguchi et al. (2001) suggested that concentrations of plasma corticosterone remained high for 5 h. The elevated corticosterone levels suggest that swimming in our study was an effective stressor for at least 40 min.
The elevation of plasma corticosterone suggested that the hypothalamic-pituitary-adrenal axis was activated. The plasma concentrations of norepinephrine and epinephrine were anticipated to be elevated because the sympathetic adrenal medulla was supposed to be activated (Axelrod and Reisine 1984; Toates 1995). The reduction in plasma levels of these hormones seen in our experiments may be because repeated swim and restraint with swim caused exhaustion in rats and modified the hypothalamic-pituitary adrenal axis. Acclimation may also play a role. Glucocorticoids influence synthesis and modify release of norepinephrine and epinephrine from the adrenal medulla into the circulating blood (Toates 1995).
Indices of Repeated Stress and Chlorpyrifos on Excitatory Amino Acids
In the present study, swim and restraint with swim stress increased concentrations of aspartate (with a trend toward increased glutamate) in the hippocampus when compared to handling. Other studies have noted increases in glutamate or aspartate with stress. For example, restraint for 20 min or for 6 h daily for 21 days were both found to increase concentrations of one or both of these excitatory neurotransmitters in various brain regions of rats, including the hippocampus (Moghaddam 1993; Moghaddam et al. 1994; Sunanda, Rao, and Raju 2000). Swim stress has also been reported to elevate glutamate and aspartate (Moghaddam 1993). Although our results with glutamate after repeated stress were not as dramatic, our experiments differed from those cited above in length of exposure to stress, especially swim stress. Moreover, it appeared that our rats restrained for 28 days became acclimated to restraint because they easily entered the plexiglass tubes.
Chlorpyrifos did not affect hippocampal glutamate concentrations in our studies. These results contrast with studies conducted with another organophosphate compound, soman, which transiently elevated the extracellular glutamate concentrations in the hippocampus within 30 minutes of a seizure-inducing dose (Lallement et al. 1991). The dose of chlorpyrifos used in the present study did not cause seizures or any signs of cholinergic poisoning. Lack of convulsions may explain the lack of release of glutamate in our study. However, other cholinesterase-inhibiting compounds have been reported to elevate glutamate without causing seizures (Dijk et al. 1995). Although not contributing to changes in glutamate, chlorpyrifos decreased concentrations of aspartate in the hippocampus in our restrained with swim rats, suggesting that it can have an effect on excitatory amino acid neurotransmitters, an effect that has been suggested to modulate striatal and cortical cholinergic transmission (Morari et al. 1998). The interrelationship of glutamatergic and cholinergic systems may contribute to memory and learning (Cooper, Bloom, and Roth 1996).
Effects of repeated stress and chlorpyrifos on NMDA receptors in the present study suggested that our stress treatments were insufficient to affect maximum number of NMDA receptors or binding affinity to NMDA receptors across all types of stress and both brain regions examined. NMDA receptors have, however, been reported by others to be involved in the stress response (Magarinos and McEwen 1995). Even swimming, the apparently most potent stressor based on corticosterone changes, appeared to have a transient effect, as plasma corticosterone levels did not stay elevated. This may have contributed to the lack of notable effects of stress on NMDA receptors in the present study.
Indices of Effects of Repeated Stress and Chlorpyrifos on the Cholinergic System
Cholinesterase inhibitors not only inhibit AChE activity but also down-regulate muscarinic receptors (Ecobichon 2001). This was noted in the present study, as chlorpyrifos-treated rats concurrently exposed to restraint or repeated swim had reduced B max, indicating fewer total muscarinic receptors in the cerebral cortex. Similar results were seen with chlorpyrifos alone in a previous study (Pope et al. 1992). Repeated stress, chlorpyrifos, or the interaction of repeated stress and chlorpyrifos did not affect total muscarinic receptors in the hippocampus. Our results contrast with a previous study in which repeated immobilization of rats was reported to increase B max of muscarinic receptors in several brain regions (Gonzalez and Pazos 1992) when audioradiography was used for determination of this endpoint. As noted above, rats appeared to become acclimated to restraint in our studies. They did not, however, become acclimated to swimming, and exposure of rats to chlorpyrifos, which inhibits AChE in the brain and at neuromuscular junctions, resulted in earlier fatigue. Fatigue has been reported in applicators of organophosphate insecticides (Institute of Medicine 2003). Muscarinic receptors and neurotransmitters of the cholinergic system are suggested to be involved in learning and memory, and 1991 Gulf War veterans have self-reported alterations of learning and memory (IOM 2003). However, learning and memory are also modulated by other neuronal systems including the glutamatergic system (Cooper, Bloom, and Roth 1996; Lewis 2003).
Restraint or swim stress have been reported to modulate the cholinergic system. This, however, is a controversial issue. Some studies with carbamate cholinesterase inhibitors have suggested enhanced AChE inhibition and modulation of genes regulating acetylcholine availability following only 4 min of swim or 1 h of restraint (Beck et al. 2003; Friedman et al. 1996; Kaufer et al. 1998). Other studies did not see changes in AChE inhibition following 10 to 90 minutes of stress (Grauer et al. 2000; Song et al. 2002; Tian et al. 2002). Our results demonstrated that repeated restraint, swim, or restraint with occasional swim did not affect activities of blood AChE, or activities of AChE, carboxylesterase, and ChAT in the hippocampus.
Chlorpyrifos 160 mg/kg sc in this study inhibited activity of AChE in blood and in the hippocampus and cerebral cortex 4 days after injection. This corresponds to previous studies (Chanda et al. 1997; Karanth and Pope 2003; Pope et al. 1992), especially those of Karanth and Pope (2003), who showed that chlorpyrifos at the maximal tolerated dose of 279 mg/kg inhibited cholinesterase activity by 85 to 95%. The maximal cholinesterase inhibition in their adult rats was observed 96 h after dosing. In our studies, chlorpyrifos did not affect the activity of ChAT. Therefore, chlorpyrifos at 160 mg/kg sc decreased activities of degradatory enzymes of acetylcholine but did not affect the synthetic enzyme. Karanth and Pope (2003) also reported that chlorpyrifos had little effect on acetylcholine synthesis.
Restraint with swim stress decreased activity of carboxylesterase in the cerebral cortex more than other stressors. Activity of this esterase in brain was not determined during the stress treatments done in other laboratories. Carboxylesterase detoxifies foreign compounds and protects AChE in organophosphate poisoning (Chanda et al. 1997). The reduction of brain carboxylesterase activity in rats undergoing restraint with swim could enhance the toxicity of chlorpyrifos because chlorpyrifos would be less likely to be hydrolyzed and would be free to distribute to cholinergic synapses in this region of the brain. However, it is possible that enhancement of AChE inhibition was not observed because the inhibition of AChE was great enough with chlorpyrifos alone that no greater effect was noted in stressed than in unstressed rats.
Overall, our studies demonstrated results similar to Grauer et al. (2000), Hancock et al. (2003), Song et al. (2002), and Tian et al. (2002), who reported no effects of stress on AChE activity, as opposed to the studies of Beck et al. (2003), Friedman et al. (1996), and Sunanda, Rao, and Raju (2000), who did. In our studies, restraint, swim, and restraint with swim stress were not sufficient to inhibit AChE in blood or AChE, carboxylesterase, and ChAT in the hippocampus, nor did they further enhance the AChE inhibition caused by chlorpyrifos. However, restraint with swim and chlorpyrifos individually inhibited carboxylesterase activity in the cerebral cortex.
Indices of Effects of Stress and Chlorpyrifos on Monamine Neurotransmitters
The effects of stress on monoamines in the central nervous system have been investigated by others with mixed results (Chaouloff 2000; Goldstein and Pacak 2001; Konstandi et al. 2000). Norepinephrine, dopamine, serotonin, and their metabolites play an important role in adaptive responses to stressful situations.
Our results demonstrated that swim stress decreased concentration of norepinephrine in the hippocampus. Although acute stress may be expected to increase concentrations of norepinephrine (Axelrod and Reisine 1984; Toates 1995), repeated stress of rats using restraint or swim has previously been shown to either have no effect (Campamany, Pol, and Armario 1996; Hellriegel and D’Mello 1997) or to lower levels of norepinephrine in the brain regions such as the hippocampus, cerebral cortex, and hypothalamus (Konstandi et al. 2000; Sudo 1983; Sunanda, Rao, and Raju 2000). Restraint was 2 to 6 h per day when decreased norepinephrine levels were reported (Konstandi et al. 2000; Sunanda, Rao, and Raju 2000) and 1 h per day when it failed to change (Campamany, Pol, and Armario 1996; Hellriegel and D’Mello 1997). Administration of chlorpyrifos to rats undergoing repeated stress in our study decreased norepinephrine in the hippocampus when compared to rats not given chlorpyrifos. Because cholinergic pathways as well as monoamine pathways have projections to the hippocampus (Cooper, Bloom, and Roth 1996; Webster 2001a 2002b), inhibition of AChE and the decrease of norepinephrine concentrations by chlorpyrifos may be interrelated.
Although swim and restraint with swim decreased hippocampal norepinephrine, these stressors increased dopamine and DOPAC in the hypothalamus, supporting studies of Lowry et al. (2003). The high levels of dopamine and DOPAC in the swum and restrained with swim rats suggested that the dopaminergic system was activated, with dopamine being transformed to DOPAC rather than norepinephrine. However, stress and chlorpyrifos at 160 mg/kg sc did not alter concentrations of dopamine and DOPAC in the brain regions examined in our study, suggesting that dopamine in the hippocampus and cerebral cortex was not involved in the response to stress and chlorpyrifos in our study.
Chlorpyrifos increased concentrations of serotonin in the cerebral cortex of rats that were handled and restrained and decreased serotonin in the hypothalamus of rats that were restrained with swim. Both cholinergic pathways and serotonergic pathways innervate the cerebral cortex and hypothalamus, but the excess acetylcholine available after chlorpyrifos-induced AChE inhibition may have different modulating effects on serotonin in these brain regions (Azmitia 1999).
The serotonergic system is involved in the stress response (Reuter and Jacobs 1996). Restraint or forced swim have been reported to decrease (Adell, Casanoves, and Artiges 1997; Liu et al. 2003), increase (Reuter and Jacobs 1996), or have no affect (Kirby et al. 1997) on serotonin in the hippocampus.
Our findings demonstrated that the hypothalamus was vulnerable to the effects of concurrent exposure to stress and chlorpyrifos as indicated by changes in monamine concentrations. The hypothalamus relays information to neural pathways mediating neuroendocrine, autonomic, and behavioral responses to stress (Lowry et al. 2003) Stressors have been reported to alter hypothalamic concentrations of norepinephrine, dopamine, and serotonin (Lowry et al. 2003) and contribute to modulation of glutamate activity (DiMicco et al. 2002). It is possible that these changes may contribute to psychological and behavioral alterations reported in people under stress (IOM 2003; Konstandi et al. 2000; Marshall, Davis, and Sherbourne 2000).
CONCLUSIONS
In conclusion, repeated stress affected the glutamatergic, cholinergic, and monoaminergic systems with response dependent on the type of stress. Chlorpyrifos inhibited the neurotransmitter degrading enzyme of the cholinergic system (AChE) but did not affect glutamate. There was an interaction of stress and chlorpyrifos on muscarinic receptor levels. Restraint with swim tended to increase glutamate and aspartate and chlorpyrifos attenuated the elevated aspartate concentrations of restrained with swim rats, suggesting that excitatory amino acids were affected. Repeated stress and chlorpyrifos have effects on two interrelated systems (glutamatergic and cholinergic), but the interactions of stress and chlorpyrifos were few. Effects of stress and chlorpyrifos were present in all brain parts examined (cerebral cortex, hippocampus, and hypothalamus). Elucidation of the neurochemical effects that could lead to behavioral and/or pathological changes would require further studies.
Footnotes
Figures and Tables
Acknowledgements
The authors would like to acknowledge the assistance of Kristel Fuhrman, Geraldine Magnin-Bissel, Kristin Knight, Jonathan Hinckley, Sandy Hancock, Linda Correll, Delbert Jones, Barbara Wise, Ellen Uri, Nikola Agatic, David Moore, and Daniel Ward. Appreciations are to the Faculty of Liberal Arts and Science, Kasetsart University, and the Thai government for T. Pung’s scholarship. This work was supported by DAMD 17-99-1-9489 from the U.S. Army Medical Research and Materiel Command and by Virginia-Maryland Regional College of Veterinary Medicine.