This item is only available as the following downloads:
GULF WAR ILLNESS: AN EVALUATION OF NEUROGLIAL CHANGES AND LEARNING, MEMORY, AND ANXIETY IN GWI MOUSE MODELS BY ARIEL DUNAMIS GONZALEZ MENDOZA A Thesis Submitted to the Division of Natural Sciences New College of Florida in partial fulfillmen t of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr. Alfred Beulig Sarasota, Florida May, 2013
ii I would like to thank my thesis advisor, Professor Alfred Beulig, for all the mentorship and instruction he has given me t hroughout my four years at New College. I would also like to thank my thesis committee, Professors Gordon Bauer and Shelley Batts for dedicating their time to this endeavor. Thanks to my mentors at the Roskamp Institute for allowing me to conduct my thesi s experiments and guiding me through the treacherous shores. Laila Abdullah, Fiona Crawford, James Evans, Bunmi Ojo and Ghania Ait Ghezala without whom I would not be writing this dedication. This work is dedicated to my entire family, who have been with me through thick and thin. This thesis is one of the many representations of our mutual support and success. Hoy, Hoy ha de ser el da de las Rosas. Entre la lluvia celestial del sol vi a l a reina de las musas, la ms hermosa que mi corazn de pjaro a tocado no tenia anatoma de ningn total, era un regalo plural seda y jazmn que derrite el alma de cualquier poeta. Su cabellera bailaba y por siempre bailara en mis latidos con todos los sue os, que como este romance de luz solar son derretidos. Esa sonrisa que le pertenece a las lumbreras por siempre me cantara, y sin decir nada me hablara.
iii GWI Gulf War Illness GW Gulf War PB Pyridostigmine Bromide PER Permethrin CPF Chlorpyr ifos CNS Central Nervous System HPA Hypothalamic Pituitary Adrenal Axis PTSD Post Traumatic Stress Disorder ACh Acetylcholine AChE Acetylcholinesterase DoD Department of Defense OP Organophosphate DEET N ,N diethyl m eta toluamide BBB Blood Brain Barrier GFAP Glial Fibrillary Acidic Protein IBA1 Ionized Calcium Binding Adapter Molecule 1 BM Barnes Maze EPM Elevated Plus Maze FS Foot Shock PBS Phosphate Buffer Saline BLA Basolateral Amygdala CC Corpus Callosum
iv Ded ...ii Acknowledgement ... Table of C ontents ... 1. 1.1. 1.2. 1.3. 1.3.1. 1.3.2. 1.3.3. Pe 220.127.116.11. 18.104.22.168. 1.4. 1.5. 1.5.1. Cholinergi 1.5.2. 1.5.3. 1.5.4. 1.5.5. 1.6. 1.7. 2. 2.1. .. 2.1.1. 2.1.2. 2.2. 30 2.2.1. 2.2.2. 2.3. Foot 34 2.3.1. 2.3.2. 2.4. .. 3 9 2.4.1. 2.4.2. 2.5. 2.5.1. 2 2.5.2. Ionized Calcium Binding 3 2.5.3. 4 2.5.4. 6
v 3. 7 3.1. 7 3.1.1. Barnes Maze Immediate Post Exposure Testing Time Point ... 7 3.1.2. Barnes Maze Delay ed Post Exposure Testing Time Point PB + 9 3.1.3. Barnes Maze Delayed Post Exposure Time Point CPF + PB + 1 3.1.4. Foot Shock Delayed Post Exposure Time Point CPF + PB + 1 3.1.5. Elev ated Plus Maze Delayed Post Exposure Time Point CPF + 3.1.6. 2 3.2. 3 3.2.1. GFAP Staining of Immediate Post Exposure Time Point PB + PER + CPF 3 3.2.2. IBA1 Staining o f Immediate Post Exposure Time Point PB + 3 3.2.3. GFAP Delayed Post 8 3.2.4. IBA1 Delayed Post .5 9 4. Discussion and Co 60 5. Refere 1
vi GULF WAR ILLNESS: AN EVALUATION OF NEUROGLIAL CHANGES A ND LEARNING, MEMORY, AND ANXIET Y IN GWI MOUSE MODELS Ariel Dunamis Gonzalez Mendoza New College of Florida, 2013 ABSTRACT Gulf War Illness (GWI) is a complex set of symp toms associated with specific exposures during the 1990 Gulf War (GW). These exposures include organphosphate and pyrethroid pesticides, and the prophylactic use of pyridostigmine bromide (PB). All these agents act on the cholinergic system, and it is ther efore thought that cholinergic system alterations are a central theme in the GWI pathophysiology. In this thesis, memory, anxiety, and neuroglial activation were examined in three different GWI mouse models using behavioral and immunohistochemical data. Be cause all these actions utilize the cholinergic system, it was hypothesized that GW agent exposure would impair memory function and increase anxiety and neuroglia activation. Mouse models exposed to PB and the pyrethroid pesticide permethrin (PER) at an im mediate and delayed post exposure time points, and models exposed to the organophosphate pesticide chlorpyrifos (CPF), PB, and PER at the delayed time point were tested for learning and memory impairments using the Barnes maze. It was observed that both th e immediate and delayed PB + PER mouse models performed significantly worse than the control groups with a positive post exposure time correlation. The CPF + PB + PER delayed post exposure model performed better than the control mice. Anxiety was examined using the foot shock and elevated plus maze paradigms. During the training days, the CPF + PB + PER mice spent significantly more time frozen as compared to the controls, indicating heightened anxiety.
vii In the elevated plus maze however, this same model did not differ from the controls. The mouse models exposed to CPF + PB + PER and CPF alone at an immediate post exposure time point were used for immunohistochemical experiments examining for activated neuoglia. It was observed that both treated groups had si gnificantly increased astroglial activation in different brain regions, including the amygdala, hippocampus and several cortex regions. Microglia activation was not significantly different between the treated and control mice. At the delayed post exposure time point, PB + PER exposed mice were used to evaluate astroglial activation, and a significant increase in activated astroglia was seen in the hippocampus and amygdala. These data indicate that there are central nervous system (CNS) alterations that are related to GW agent exposure, namely memory impairments, anxiety, and astrogliosis. Dr. Alfred Beulig Division of Natural Sciences
1 INTRODUCTION 1.1. Background Gulf War Illness (GWI) is a complex multi symptomatic condition, which gets its name from the war thought to be its cause, namely the first Gulf War. GWI is characterized by a diverse spectrum of abnormalities including fatigue, hea dache, memory deficits, musculoskeletal pain and inflammation, and gastrointestinal, respiratory, and skin conditions. These symptoms, which comprise GWI, afflict about 25 32% of the 700,000 veterans that served in the 1990 1992 Gulf War (GW) (Barlow et al ., 2008). To date, GWI is the most prevalent disorder among these veterans (Barlow et al., 2008). A longitudinal study conducted over 10 years reported that deployed veterans, compared to their non deployed counterparts, continue to show persistence of thi s multi symptom illness (Li et al., 2011a). The complex symptomatology and the overwhelming number of affected veterans make this condition a significant military health issue. debi litating, makes it difficult to understand GWI etiology and to treat this illness effectively. To tackle such an issue, it is necessary to understand what the cause/s are and why each symptom is present. If the causes of the symptoms are traced back to the ir origin, then it may be plausible to treat the symptoms all together or even treat each symptom separately in a systematic way that is most efficient. Attempts at treating fatigue, pain, distress, cognitive problems, and mental health functioning as symp toms of GWI have been largely unsuccessful. To date there are 6 federally funded clinical trials for the treatment of GWI that have reached completion, and 15 others are currently
2 ongoing or have not yet started (National Institutes of Health, 2013). Two of these completed studies are major randomized trials, which have taken place in several different regions of the United States and its territories. The first was a doxycycline intervention trial, which was based on the hypothesis that GWI was caused by a systemic bacterial infection ( Donta et al., 2004 ) The other trial used cognitive behavioral therapy and exercise to improve some of the musculoskeletal symptoms pre sent in GWI (Donta et al., 2003 ). Unfortunately, neither trial significantly improved the symptoms of GWI, and so to this day, GWI remains an untreatable condition (Donta et al., 2003; Donta et al., 2004) Over twenty years have elapsed since the original war, and veterans with this condition continue to suffer from debilitating symptoms that a re not completely understood and therefore very difficult to treat. One of the initial steps into effectively tackling this problem is to characterize the sequelae of this illness in order to understand the pathology. There is a lack of focused pre clinica l research on the central nervous system symptoms of GWI, and in particular the cognitive dysfunctionality. For other traumatic stress disorder, and depression, pre clinical a nimal models have been extremely valuable, and GWI should be no different. Research and development of mouse models at the Roskamp Institute (2040 Whitfield Ave, Sarasota, FL 34234) have been successful in recapitulating the central nervous system related symptoms associated with GWI, namely memory impairment, anxiety, widespread pain, and motor problems (Abdullah et al., 2011; Abdullah et al., 2012). The creation of these GWI mouse models has been helpful as an initial step for the characterization of the GWI pathology, and may eventually lead
3 to the translation of therapies for the treatment of GWI (Abdullah et al., 2011; Abdullah et al., 2012). There are three mouse models of interest, differentiated by Gulf War agent exposure regimen. The first model wa s exposed only to chlorpyrifos (CPF); the second to pyridostigmine bromide (PB) and permethrin (PER); and the third to chlorpyrifos, pyridostigmine bromide, and permethrin. The main objective of this thesis was to use these three different GWI mouse models developed by Dr. Laila Abdullah to explore two cognitive aspects of GWI, memory impairment and anxiety. Attention was also given to glial changes in the central nervous system after GW agent exposure, as gliosis has been associated with the symptom compl ex ( Abdel Rahman et al., 2002 ) 1.2. GWI and the Central Nervous System The idea of a central nervous system component for GWI first drew attention after neurobehavioral data showed that some GW veterans were displaying memory and behavioral alterations In an initial survey of 249 members of a naval battalion, cognitive dysfunction and other neuronal abnormalities were observed as one of three major syndromes (Kurt, 1998; Haley et al., 1997). Lea Steele communicated that deployed GW veterans reported me mory problems more often than their non deployed counterparts (Steele, 2000). A study performed by Jeffrey Lange and colleagues demonstrated that compared to healthy GW veterans, those manifesting GWI symptoms had significant difficulty with information pr ocessing and the use of abstract concepts, particularly within the cognitive domains that require attention and concentration (Lange et al., 2001). Other groups showed that memory and cognitive alterations were more readily observed in deployed Gulf War ve terans compared to military personnel deployed elsewhere or not
4 deployed at all (David et al., 2002; Toomey et al., 2009; White et al., 2001). Furthermore, reports of 1990 1991 United Kingdom GW veterans displayed similar outcomes as the US veterans, where subjective memory failure and general cognitive and constructional impairment was significant even after adjusting for depression (David et al., 2002). These data show that the central nervous system component manifested in GW veterans is real and non dis criminate between US veterans or their allies. The next step in elucidating the central nervous system anomaly came from brain imaging studies performed on GW veterans and healthy civilians. One of the most interesting findings in these studies was a decr ease in hippocampal volume in deployed and reservist GW veterans compared to their healthy civilian counterparts ( Vythilingam et al., 2005 ) Furthermore, when the neural metabolic activity of GW veterans was studied, it was noted that hypo metabolism occur red in the hippocampuses of veterans with the GWI symptom complex, as compared to asymptomatic veterans who serve d in the GW (Menon et al., 2004 ). The connectivity passing through the hippocampus is extremely important for a variety of mental processes, in cluding but not limited to memory formation and memory retrieval (Schmolck et al., 2002). If this important structure were to be compromised, then a series of cognitive deregulations would be apparent, as well as structural changes in the hippocampus and a ssociated areas. Taken together, structural and metabolic abnormalities in the hippocampus can definitely account for memory alterations displayed in the neurobehavioral studies. In a recent report, Rosemary Toomey and colleagues communicated that deployed GW veterans continue to experience deficits in working memory almost two decades after the initial
5 exposure (Toomey et al., 2009). It is therefore apparent that GWI affects working memory and memory consolidating structures at both immediate and delayed t ime points. Although memory impairment remains one of the top four complaints among veterans with GWI, there are other neuropsychological problems also experienced by these veterans, such as anxiety, attention deficits, and impairment in motor speed (Toom ey et al., 2009; Gray et al., 1999). A series of studies performed by Dr. Julia Golier explore the impact the GW had on the stress response in these veterans. In 2006, Julia Golier, Julianna Legge, and Rachel Yehuda compared the hypothalamic pituitary adre nal axis (HPA) response to dexamethasone, a glucocorticoid agonist. The group examined the blood levels of adrenocorticotropic hormone and cortisol before and after dexamethasone treatment in Gulf War veterans with post traumatic stress disorder (PTSD), no n PTSD deployed veterans, and non deployed veterans. It was observed that adrenocorticotropic hormone and cortisol levels were not significantly different between the PTSD and non PTSD deployed veterans, but both were significantly different when compared to their non deployed counterparts. Furthermore, when controlling for PTSD, a positive correlation was observed between HPA axis feedback inhibition and both deployment status and the symptom complex associated with GWI (Golier et al., 2006a; Golier et al. 2006b). In another study, Julia Golier and colleagues found that since symptom clusters are similar in GW veterans with PTSD and GW veterans without PTSD, it appears that GWI is biologically distinct from PTSD alone. In fact, the data supported a positiv e correlation between HPA axis deregulation and exposures distinct to the GW, such as pyridostigmine bromide (PB) and pesticides (Golier et al., 2007). Furthermore, these changes in HPA axis are due to an increase in pituitary sensitivity, as
6 was seen when the pituitary response due to intravenous cortcotropin releasing factor in GW veterans was compared to other veterans (Golier et al., 2012). 1.3. GWI Etiology In the initial stages of GWI research, the causes of GWI were highly debated. This debate was d ifficult because of the complexity of the illness, and it was easy to be misled when each symptom could be due to a vast array of causes. This task was not made any easier when presented with the various unique circumstances GW veterans were exposed to. Ho wever, in 2008 when the Research Advisory Committee on Gulf War pyrethroid and organophosphate pesticide exposure and prophylactic use of PB against neurotoxins are likely th e main contributors for the development of GWI (Barlow et al., 2008). Data in favor of this conclusion continues to be released. In 2012, Lea Steele and colleagues reported that PB consumption elevates the risk for developing GWI 3 fold, and pesticide expo sure increases the risk by about 13 fold (Steele et al., 2012). 1.3.1. Oil Well Fires, Sarin Gas, and Depleted Uranium GW veterans, as compared to veterans from other wars, were highly exposed to depleted uranium, oil well fires, and accidental release of sarin gas, but the evidence supporting the notion that these agents are the cause for GWI is currently weak. There are well documented cases suggesting that the Kuwait oil well fires were unique to the deployed troops of the 1990 1991 GW, and therefore ma ke this exposure regimen suspect of GWI causation. However, this exposure is unique only to the ground troops stationed
7 in Kuwait, and does not account for the GWI cases seen in the troops stationed in support areas and ships during deployment (Barlow et a l., 2008). Other epidemiological studies evaluating the risk of oil well fires and the development of GWI symptoms have also indicated the lack of association between exposure to oil well fires and GWI symptoms (Smith et al., 2002; Cowan et al., 2002; Lang e et al., 2002). Similarly, there is currently a lack of clinical evidence suggesting any association between exposure to sarin gas or depleted uranium and the development of GWI (Barlow et al., 2008). Therefore, due to the lack of evidence and association it is not currently feasible to conclude that oil well fires, sarin gas, or depleted uranium cause the onset of GWI pathophysiology. 1.3.2. Pyridostigmine Bromide (PB) Exposure Pyridostigmine bromide (PB) is a reversible carbamate acetylcholinesterase (A ChE) inhibitor, and was used as a prophylactic measure against nerve gas in the GW (Gunderson et al., 1992). The idea of using PB to protect against nerve gas seems contradictory, since both PB and nerve gas inhibit AChE. However, the rationale for this tr eatment lies in the fact that PB is a reversible inhibitor, while nerve gas is an irreversible inhibitor. Therefore, if PB inhibits some portion of the total AChE when exposure to nerve gas occurs, then those same molecules would be protected against irrev ersible inhibition. Because nerve gas has a higher reactivity than PB, by the time the enzymes recover from PB inhibition, there would be very few organophosphate molecules left in circulation for AChE re inhibition (Pope et al., 2005). Although PB was use d to protect the troops, this exposure regimen was unique to the GW, thereby raising suspicions for a role in GWI development.
8 Early research supported PB as a possible correlate for GWI. When the neurobehavioral symptoms of GWI were first being evaluate d and distinguished from other conditions, six symptom clusters were associated with the illness, and from these three major syndromes were identified (Kurt, 1998; Haley et al., 1997). All of the participants in this study reported taking the anti nerve ga s treatment, pyridostigmine bromide, therefore exposure to PB was correlated with all three syndromes. Further data in support of an association between PB and GWI became available later on. A retrospective study showed that the veterans who took larger qu antities (>22 pills of 30mg) of PB as a prophylactic measure had a higher risk and more severe symptoms of GWI than those taking fewer pills (Haley et al., 1997; Golomb et al., 2008). Current evidence therefore strongly supports an association between PB a nd GWI pathophysiology. 1.3.3. Pesticide Exposure Pesticides were widely used during the Gulf War to combat desert pests. About 62% of the ground troops reported using pesticides, and of those about 44% reported using pesticide sprays containing permethri n (PER) or other pyrethroids. As mentioned above, Robert Haley and Thomas Kurt conducted surveys in an attempt at correlating syndromes observed readily in GW veterans with exposure regimens. Pesticide application was heavily correlated with cognitive impa irment. Also, confusion ataxia was correlated with the reports of chemical weapons attacks, some including nerve gas agent with an organophosphate based composition (Kurt, 1998; Haley et al., 1997). Furthermore, the Department of Defense (DoD) reported tha t the US service members
9 from the 1990 1991 GW may have been exposed to at least 64 different pesticides and insect repellents containing 37 active ingredients. Of these compounds, fifteen raised significant concern based on their known toxic effects. Seve n organophsphates and two pyrethroids are included among these fifteen (Department of Defense, 2003) Veterans with high pesticide use were at a greater risk of GWI than those with limited use, suggesting a dose response relationship (Hilborne et al., 2005) 22.214.171.124. Permethrin (PER) Permethrin (PER) is a type 1 synthetic pyrethroid and works by prolongin g the activation of the voltage gated sodium channels. Due to its high lipophilicity, PER is able to slide over to the lipids adjacent to the cha nnels and r ebind to the voltage gated sodium channels at a later time point. Therefore, the overuse of PER is a potential cause for concern due to its toxicological factors and excessive exposure regimens (Ray and Fry, 2006). During the GW, military personnel were is sued PER imbedded uniforms as well as 0.5% PER containing spray. Surveys conducted by the DoD report that PER was sprayed an average of 30 times per month despite the warning to only apply a light coat of PER on the uniforms every 4 5 days. The PER manufac suggested applying PER on the uniforms even less frequently once every six weeks, or after being washed six times (Barlow et al., 2008) The overuse of PER supports its association with GWI development.
10 126.96.36.199. Organophosphates (OP) As discussed above, GWI pathophysiology is likely heavily influenced by pesticide exposure, and there were at least seven organophosphates (OP) used that are of potential concern for GWI, including chlorpyrifos (CPF) (Department of Defense, 2003). Th ese exposures were due to pesticide control performed by the armed forces and services provided by the host nations (Department of Defense, 2002). Although it is clear that the GW veterans were exposed to these pesticides at elevated rates, it is difficult to determine the complete exposure report for these veterans, as it is well beyond their knowledge. However, supporting toxicological evidence suggests that OPs may be heavily associated with GWI. These agents act on acetylcholinesterase (AChE) in a simil ar way to pyridostigmine bromide (PB), but are more toxic due to the irreversible nature of their inhibition. When there is intoxication by an organophosphate a syndrome p rofound bronchial secretion, bronchoconstriction, miosis, increased gastrointestinal motility, diarrhea, tremors, muscular twitching, and various central nervous system effects including cognitive dysfunction (Eaton et al., 2008). The symptoms associated w ith OP intoxication are similar to the overall GWI complex, and the mode of action is similar to PB, another suspected GWI agent Therefore, overexposure to OPs may be associated with this multisymptom illness. 1.4. Combined Exposure to PB, PER, and OPs a s a Causative Factor in GWI The effects of a combined GW agent exposure regimen we re highlighted in a RAND (RAND C orp. for O bjective A nalysis and S olutions) survey showing that high
11 exposure to pesticides was associated with high PB use. The survey report ed that nearly one in four GW veterans were exposed to several pesticides on a monthly basis, ranging from 51 120 times, all the while taking 15 19 PB pills (30 mg each) (Hilborne et al., 2005). Although there are central nervous system structures that a re clearly affected, including those that regulate peripheral responses like the hypothalamic pituitary adrenal (HPA) axis and the acetylcholine anti inflammatory pathway, much of the work done on animal models suggests that these central nervous system al terations are due to adverse effects in the periphery or a possible compromise of the blood brain barrier (BBB). It has been shown that when stress is combined with PB, PER, and N,N Diethyl meta toluamide (DEET) for up to 28 days of exposure, the BBB may b e distorted in specific brain regions, including in the cortices and certain areas of the hippocampuses of exposed rats as compared to controls (Abdel Rahman et al., 2002). These results might be due to the vascularity found in these areas and susceptibili ty to stress hormones. Alon Friedman and colleagues showed that at low dosages, PB alone (0.1 0.5 mg/kg) had no effect on brain AChE activity. However, at higher dosages (2 mg/kg) or when combined with stress, PB inhibit s brain AChE activity and induce s ch anges in expression of genes, such as those encoding for brain derived neurotrophic factor, tyrosine related kinase B, and calcium calmodulin dependent protein kinase II, which are involved in learning and memory formation. The authors thereby suggested th at under stress ful conditions and at high concentrations, PB might be able to cross the BBB (Friedman et a., 1996; Barbier et al., 2009).
12 Other studies suggest that PB is not able to cross the BBB under any circumstances. Using radiolabeled PB, Christine A mourette and colleagues revealed that even when combined with stress, PB was unable to penetrate the BBB (Amourette et al., 2009). Despite these results, biochemical analysis continue to show that brain AChE activity is inhibited and neurobehavioral data a lso demonstrate a delayed presentation of cognitive impairment in rats exposed to PB and stress (Barbier et al., 2009; Lamproglou et al., 2009; Amourette et al., 2009). Furthermore, it was observed that nearly 50% of rats displayed neuropathological abnorm alities, such as neuronal death and an increase in astrogliosis at an immediate post exposure time point (Abdel Rahman et al., 2002). It has also been shown that the neurobehavioral consequences of combining these agents include an increase in anxiety and sensorimotor deficits in exposed rodents compared to their control counterparts (Abou Donia et al., 2004; Hoy et al., 2000). 1.5. Biological Mechanisms of PB, PER, and OPs in GWI Pyridostigmine bromide (PB), permethrin (PER), and the organophosphate chlo rpyrifos (CPF) all act on the peripheral and central cholinergic systems t hereby affecting cholinergic mediated inflammation and cognition. In order to have a complete mechanistic and biological rationale for the role of these compounds in GWI, data must be presented from both immediate and delayed post exposure time points. However, understanding the immediate biological effects of PB, PER, and OPs can contribute to understanding the biological and behavioral changes at delayed post exposure time points.
13 1.5.1. Cholinergic Transmission in GWI It is necessary to have an understanding of the functionality of the cholinergic system before any related abnormality is presented (See Figure 1.1). Under normal conditions, ACh is synthesized in the cholinergic n eurons by choline acetyltranferase, using choline and acetyl coenzyme A as substrates. This neurotransmitter utilizes muscarinic and nicotinic acetylcholine receptors in both the sympathetic and parasympathetic ne rvous systems. In the peripheral nervous sy stem, ACh is used by somatic motor neurons to communicate with muscle fibers and induce contractility, and by the autonomic nervous system to communicate with cells located in the spleen for the initiation of the inflammatory response. ACh is also found in the central nervous system, and it plays a crucial role in memory, learning, and cognition (Micheau and Marighetto, 2011; Davies, 1985). After synthesis, ACh is packed into synaptic vesicles, and upon terminal depolarization from an action potential the v esicle fuses with the plasma membrane and releases ACh into the synapse (Pope et al., 2005). In the synaptic cleft, ACh may bind to nicotinic post synaptic cholinergic receptors and activate numerous signaling pathways or be rapidly hydrolyzed by AChE. Cho linergic neurons also have a negative feedback mechanism via activation of muscarinic acetylcholine receptors. If activated, these receptors are able to inhibit further production of ACh (Pope et al., 2005). This mechanism may be altered either upstream of ACh release or downstream. Upstream deregulation involves hampering either the action potential via voltage gated sodium channels or the production of ACh via choline acetyltransferase. Downstream deregulation involves alterations in the clearance of ACh from the synapse, by inhibiting AChE, or improper receptor binding. As will be observed below, suspected GW agents
14 FIGURE 1.1 displays cholinergic activity bet ween two cholinergic neurons. On the left side, the pre synaptic neuron releases acetylcholin e into the synaptic cleft, where the compound either binds to post synaptic nicotinic receptors, gets hydrolyzed by acetylcholinesterase, or binds to pre synaptic muscarinic receptors. Binding to nicotinic receptors elicits an action potential in the post synaptic neuron and may lead to downstream signal transduction pathways. Whereas, binding to muscarinic receptor s leads to inhibition of further acetylcholine release.
15 alter the cholinergic system by altering the levels of ACh. Therefore, it is assumed th at GW agents act somewhere in this mechanism and disrupt proper cholinergic function. Data show that the improper use of PB, PER, or CPF alone may alter the functionality of the cholinergic system in ways that are similar to the GWI complex, including cogn itive dysfunction, motor abnormalities, bowel issues, and musculoskeletal pain and inflammation (Eaton et al., 2008). The use of AChE inhibitors, such as PB and CPF, causes prolonged accumulation of ACh in the synapse and thereby alters the cholinergic rec eptor mediated signaling pathways (see Figure 1.2). The use of PER stimulates the voltage gated sodium channels, which further activates the release of ACh. It makes intuitive sense that at least the combination of PER, PB, and CPF would have synergistic e ffects since each compound affects the cholinergic system via different pathways. Animal studies using a combined exposure regimen to a pyrethroid and a carbamate AChE inhibitor have shown that a synergistic increase in membrane depolarization is accompani ed by an increase in ACh release into the synapse (Corbel et al., 2006). On the other hand, it has been shown that excessive release of ACh into the synapse can itself form a negative feedback loop by binding to pre synaptic muscarinic receptors and haltin g the release of ACh (Zhang et al., 2002). It is conceivable that the combined exposure to PB, a carbamate AChE inhibitor; PER, a voltage gated sodium channel activator; and CPF, an OP irreversible AChE inhibitor may synergistically enhance ACh levels in t he short term, but may result in depletion of synaptic ACh levels in the long term.
16 FIGURE 1.2 displays the same cholinergic neurons seen in figure 1.1. However, above are also GW agents, chlorpyrifos (CPF), p ermethrin (PER), and pyridostigmine bromide (PB). The roles these agents play in the transmission of acetylcholine is shown above. CPF and PB both bind and inhibit acetylcholinesterase (inhibition shown by sign), and as a result cause an increase in the available synaptic acetylcholine (increase s hown by + sign). This increase in acetylcholine causes a rise in post synaptic neuron excitability, and alteration in downstream signal transduction pathways. Also, an increase in pre synaptic muscarinic receptor binding is shown, which halts the release o f acetylcholine into the synaptic cleft. Another character is PER, which activates voltage gated sodium channels, and causes membrane depolarization eventually leading to an increase in acetylcholine release. These G W agents act in a synergistic manner and cause deregulation in the cholinergic neurons.
17 Animal models have also been used to study acetylcholine receptor properties affected by GW agent exposure. One example is a study where enhanced ligand binding of the muscarinic 2 ACh receptor subtype in t he cortex was observed in a GWI rat model treated either with PB alone or a combination with PER and DEET (Abou Donia et al., 2004). Combining exposure to these GW agents also increased the ligand binding of nicotinic acetylcholine receptors in the cortex of rodents. However, combined exposure to these chemicals and stress resulted in a decrease in muscarinic receptor ligand binding in the midbrain and the cerebellum, although cortical muscarinic receptor binding was not examined (Abdel Rahman et al., 2004) A long term consequence of such an enhanced cholinergic transmission may include a compensatory increase in AChE production. This idea of compensating AChE levels after inhibition has been shown with CPF exposure, since CPF, alone or in combination with other agents, leads to alterations in ACh levels. CPF acts on AChE by phosphorylating a serine residue in the active site. Since AChE is a serine protease that utilizes the serine residue as a nucleophile during hydrolysis, the phosphorylation of this resi due will ultimately impede its action on the physiological substrate. AChE is hydrolyzed by water, however, this recovery process is very slow and may take days. By this point, most of the enzyme inhibitor complexes phosphoryla tion of the seri ne residue is irreversible, making this enzyme unusable. When this happens, the only way of replacing the enzymatic activity is by the synthesis of new AChE as a compensatory mechanism. This has been suggested by studies performed by Iti Bansal and colleag ues showing that after 30 days of AChE inhibitor exposure, there is an increase in AChE mRNA levels in the cortex of mice (Bansal et al., 2009).
18 A number of human studies provide further support for cholinergic deregulation in veterans with GWI. Recent fM RI imaging studies, for example, have shown an abnormal cholinergic response in veterans with GWI as compared to controls (Liu et al., 2011). Using atrial spin labeling Magnetic Resonance Imaging, Li and colleagues recently demonstrated that upon cholinerg ic dysregulation caused by an intravenous infusion with 0.3 mg of physostigmine, veterans with GWI exhibit an abnormal increase of regional cerebral blood flow in the hippocampus compared to control veterans (Li et al., 2011b). This work is supported by a previous experiment, which used Single Photon Emission Computed Tomography to show an increase in cerebral blood flow in response to cholinergic disturbance caused by physostigmine among veterans with GWI (Haley et al., 2009). These studies suggest choline rgic dysfunctionality in patients with GWI. 1.5.2. Cholinergic System and Learning and Memory choliner gic drugs. However, the specific mechanism of action by which the cholinergic system influences memory is unknown. What is known is that there is a role for this system in learning, memory formation, and memory retrieval. The classic studies of David Drach man demonstrated that anti cholinergics, such as scopolamine, cause robust deficits in memory functionality in young human subjects (Drachman and Leavitt, 1974; Drachman and Sahakinan, 1980). These memory alterations were not reversed with the use of amphe tamines, but were ameliorated with the use of physostigmine, a carbamate AChE inhibitor and a related PB compound. One possible place for cholinergic alteration
19 after anti cholinergic delivery is the nucleus basilis of Maynert and the diagonal band of chol inergic neurons that run through the basal forebrain, neocortex, and the hippocampus (Johnson et al., 1979; Lehmann et al., 1980). It is logical for anti cholinergics to act upon these sites, since these sites are heavily involved in the formation and retr ieval of disease patients that these cholinergic neurons are involved in the formation and retrieval of memories, since loss of these cells results in memory and cognitive impairments (Schliebs and Arendt, 2006). These studies led to the prop osition of the cholinergic hypothesis of geriatric memory dysfunction and treatment strategies consisting of cholinesterase inhibitors (Bartus et al., 1992). To date, 4 out of the 5 FDA approved o n the Alz.org website. Therefore, if these data are accompanied with the above mentioned mechanisms of action for PB, PER, and CPF, then it is conceivable that the mechanisms for learning and memory may in fact be altered via cholinergic deregulation. Although initial levels of AChE are lowered, a compensatory increase of AChE in the cortex has been shown after AChE inhibition (Bansal et al., 2009). Therefor e, the memory impai rments seen i n GWI patients might be due to a compromised central cholinergic system. 1.5.3. Cholinergic System and Stress It has been noted that in a healthy functioning brain, the application of AChE inhibitors and acute psychological stress both result in a temporary increase in ACh release, increased levels of ACh at the synapse, and a phase of enhanced neuronal
20 excitability. Furthermore, administration of AChE inhibitors can mimic the effects of acute stress on AChE gene expression. Both AChE inhibito rs and stress cause an increase in AChE mRNA levels and a simultaneous decrease in choline acetyltransferase mRNA levels (Kaufer et al., 1998; Kaufer et al., 1999). These changes go from increased levels of synaptic ACh to lower levels due to reduced produ ction and increased breakdown. AChE has other roles other than hydrolyzing ACh, including cell proliferation, differentiation, and post transcriptional responses to insults like stress. These responses will vary, at least in part, by the alternative spli cing of the AChE mRNA C terminal. Under normal conditions, the activation of AChE mRNA primarily produces the in muscle and brain (Grisaru et al., 1999; Salmon et al., 2005). However, under stressful conditions, whether chemical, psychological, or physical, a transcript splicing shift The latter is expressed in embryonic and tumor cel ls and has been shown to accumulate in the mammalian brain under acute stress. This process has been positively correlated with increased levels of glucocorticoids (Grisaru et al., 1999; Salmon et al ., 2005). When the effects of GW agent exposure and acute stress are evaluated, it is observed that these events are associated with neurodegeneration. However, excess amounts of produced after chemical or physical stress, result in neuroprotection. This is in part accomplished b during transient acute stress, which is a modulation that prevents the shift to progressive disease, namely chronic stress (Sternfield et al., 2000). On the other hand, chronic excess esult of failure to shift splicing to the
21 neuromuscular alterations, and increasing amounts of stress markers (Salmon et al., 2005). Therefore, neurodegeneration in co nditions like GWI and stress disorders like PTSD might have an underlying mechanism involving the compromise of the cholinergic system. Such a mechanism might involve the compensatory increase of AChE, due to stress or chemical inhibition, where the shift 1.5.4. Cholinergic S ystem and Gliosis Under normal conditions, astrocytes are involved in cell cell communication and in the maintenance of neuronal tissue and the blood brain barrier (Liedtke et al., 1996; Weintein et al., 1991). However, after acute injury or disease astr ocytes undergo cellular changes in response to the compromise of neuronal tissue. These changes ultimately result in astrocyte activation, including hypertrophy (astrogliosis), proliferation (astrocytosis), and increased glial fibrillary acidic protein (GF AP) expression (Reichenbach et al., 2010). In many neurodegenerative disorders the activation of astrocytes is considered to be neuroprotective. However, if the activation status is prolonged, then astrocytes may switch to play a neurotoxic role by releas ing inflammatory cytokines and cytotoxic factors. Furthermore, their normal physiological roles in communication and maintenance are surrendered, which may result in adverse effects in neuronal and cerebrovascular function (Reichenbach et al., 2010; Achour and
22 Pascual, 2010). Astrocytes contain muscarinic 7 and nicotinic receptors, and their activation and/or upregulation has been shown to be positively correlated with astrocytic activation (Van Der Zee et al., 1993; Liu et al., 2012). Sara Mense showed th at astrocytes isolated from human fetal brains showed alterations in gene expression after being treated with CPF. T hese alterations included increased levels of interleukin 6, a key inflammatory mediator, and glial fibrillary acidic protein, a marker of a strocyte activation (Mense et al., 2006). Blood brain barrier distortion in the cortices and the hippocampuses has been observed in rats treated with PB, PER, and DEET. These same studies also showed an increased amount of activated astrocytes and neuron al death (Abdel Rahman et al., 2002). These alterations seen in GWI models may be due to prolonged astrocyte activation that may be regulated by impaired cholinergic activity. Therefore, inflammatory activation of astrocytes might be an important mechanism underlying GWI pathobiology, as GW agents could potentially exert their effects via acetylcholine receptors on astrocytes in the brain directly or indirectly through the activation of the cholinergic inflammatory pathway (Borovikova et al., 2000; Tracey e t al., 2001). 1.5.5. Concluding Mechanistic Remarks These studies support the current view that combined exposure to PB and pesticides is one of the highlighting causes in GWI pathogenesis, since these agents use different pathways to affect the same syste m, namely the cholinergic system. Alterations in cholinergic transmission are closely associated with stress and gliosis. Gliosis in GWI might occur as a neuroprotective compensation for the loss of neurons, and may be
23 secondary to either AChE splicing va riations, or to direct neurotoxicity. Prolonged glial activation, chemical stress, and AChE alterations all affect the general functionality of the central cholinergic system, which is essential for memory formation and memory retrieval. However, due to th e complexity of these symptoms and related mechanisms, animal models are invaluable and will provide many clues to the mysteries surrounding GWI. 1.6. GWI Mouse Models There is very little information in regards to the short and long term adverse health ef fects of combined exposure to the GW neurotoxic chemicals. This problem was approached by Dr. Laila Abdullah at the Roskamp Institute with mouse models that focused on combinations of GW agents, namely PB, PER, and CPF; PB or PER; and CPF alone. In her the sis Abdullah used these mouse models to explore behavioral alterations, brain pathology, and molecular abnormalities. These combinations produced features closely following the symptoms observed in veterans wi studies have shown that alt erations in glial cells, memory impairments, motor deficits, and signs of anxiety are closely related to the clinical symptomatology ( Abdullah et al., 2011; Golier et al., 2006a; Golier et al., 2006b; Golier et al., 2007; Golier et al., 2012). Dr. Julia G olier has examined the HPA axis in GW veterans and has used adrenocorticotropic hormone, cortisol, and cortictrophin releasing factor as biomarkers for the stress response. Golier and colleagues indicated that the HPA axis is most likely deregulated at the pituitary in GWI (Golier et al., 2012). However, it has not been determined if this deregulation is due to alterations in neuronal centers upstream of the
24 pituitary and hypothalamus, or if it is due to a downstream feedback response. It is likely that GW agents may exert their effects on the peripheral nervous system before the onset of central nervous system symptoms via other mechanisms, and thereby influence the HPA axis. Such a mechanism would be the cholinergic inflamatory pathway, which involves comm unication between the brain and the immune system via the efferent and afferent cholinergic neurons of the vagus nerve. These neurons connect to cells in the spleen capable of initiating an immune response that may alter HPA axis functionality (Borovikova et al., 2000; Tracey et al., 2001). Although it has been shown that PB does not cross the blood brain barrier, AChE inhibition still occurs in the brain after PB exposure and it has also been shown the AChE splice variants may be associated with neurodegen eration and improper memory formation. Furthermore, when combining exposure regimens, including stress, there is an increase in anxiety and sensory alterations, which are regulated by central nervous system structures upstream of the HPA axis (Abou Donia e t al., 2004; Hoy et al., 2000). Another largely unanswered question is what is causing the anxiety related behaviors? When it comes to any anxiety related symptom, it is usually important to look at the center for emotional and stress regulation, namely th e amygdala and related limbic structures. Whether CNS deregulation or peripheral deregulation occurs first in GWI, the amygdala should theoretically be involved in the anxiety/stress response. This is because the amygdala is responsible for the integration of aversive stimuli and sensory stimuli, which would also apply to cholinergic alterations, which lead to inflammatory changes. It is also capable of communicating this integrated message to the hypothalamic nuclei and thereby able to activate the hypotha lamic pituitary adrenal (HPA) axis, which is a major
25 component in the physiological stress response and one that has been evaluated in GWI. Therefore, it is likely that the amygdala is involved in the anxiety states observed in GWI, since it is capable of changing and produce changes in other areas after altered homeostasis is detected, whether the initial response is in the CNS or in the periphery. Therefore, the generation of animal models to study the effects of synergistic processes and symptoms associ ated with GW agents is necessary. At this point it is important to compare immediate and delayed post exposure measurements in animal models to have a holistic idea of what GWI pathophysiology entails. In this thesis, mouse models exposed to 3 different co mbinations of GW agents (PB + PER + CPF, PB + PER, and CPF alone), and tested at immediate and delayed post exposure time points were used to explore memory and learning impairments, gliosis, and anxiety. 1.7. Objectives and Hypotheses The literature pre sented thus far shows that the cholinergic system is affected by GW agents, and this might affect cognition, anxiety states, and neuroprotection via gliosis. Furthermore, it has not yet been determined which brain areas are responsible for the anxiety irre gularities seen in GWI patients and mice. Therefore, the criteria described below w ere used to evaluate the effects of different combinations of GW agents at immediate and delayed post exposure measurement time points. 1. Evaluation of memory formation and me mory recollection with GWI mice using the Barnes Maze. 2. Expansion of the anxiety components using fear conditioning and the
26 elevated plus maze. 3. Evaluation of the cortices, hippocampi, and amygdalae pathology using glial immunohistochemistry against GFAP a nd Ionized Calcium Binding Adaptor 1 The data show that there is impairment of the cholinergic system in the brain, and this system is essential for neuroprotection and neurodegeneration, which includes gliosis. Also, the cholinergic system is heavily in volved in cognitive functionality, including memory, learning, and anxiety. Impairments in these areas has been observed in clinical and pre clinical GWI evaluations (Abdullah et al., 2011; Abdullah et al., 2012; Lange et al., 2001). Furthermore, it has be en shown that brain areas, including the hippocampuses, are structurally affected in GWI patients (Menon et al., 2004; Vythilingam et al., 2005). Hence, the hypotheses: 1. Mice exposed to GWI agents, at both immediate and delayed post exposure time points, wi ll display more spatial mem ory and learning impairments than non exposed mice. 2. Mice exposed to GWI agents will display heightened fear/anxiety than non exposed mice. 3. Mice exposed to GWI agents, at both immediate and delayed post exposure time points, wil l have more glial activity in the cortices, amygdalae, and hippocampuses than non exposed mice.
27 METHODS 2.1. Experimental Design Different GWI mouse models were used to observe the effects of several GW agents, including chlorpyrifos (CPF), pyridostigmin e bromide (PB), and permethrin (PER). The data presented includes immediate and delayed post exposure testing, as well as different combinations of GW agents. Table 2.1 breaks down the models into mouse strain and number of mice used, exposure regimens, po st exposure measurement time points, and tests administered. Below, the animals, groups, and chemical exposures are described, followed by a description of the tests administered, methods, and statistical analysis used. For each test described, the mouse m exposure time point at day of testing are mentioned, and table 2.1 may also be used as a reference. TABLE 2. 1. Table acronyms: GFAP Glial Fibrillary Acidic Protein, IBA1 Ionized Calcium Binding Adapter Molecule 1, BM Barnes Maze, FS Foot Shock, EMP Elevated Plus Maze MOUSE MODELS: EXPOSU RE REGIMENS AND TEST ING Mouse type and quantity Exposure Post Exposure Time Point Tests Histology Behavior C57Bl6 N = 4 CPF, PB, PER Immediate GFAP + IBA1 N/A C57Bl6 N = 4 CPF Immediate GFAP + IBA1 N/A C57Bl6 N = 5 PB, PER Immediate N/A BM C57Bl6 N = 12 CPF, PB, PER Delayed N/A BM + FS + EPM C57Bl6 N = 10 PB, PER Delayed GFAP + IBA1 BM
28 2.1.1. Animal Groups Wild type (male) C57BL6 mice were purchased from Jackson Laboratories, Maine, and were allowed to acclimatize to the new environment in accordance with American Association for Laboratory Animals Science guidelines. All mice were 9 weeks old at the time of agent exposure. The mice used for the immediate measure ment studies were 10 weeks old and the mice used for the delayed measurement studies were either 17 or 18 months old at the time testing was initiated (see Figures 2.2, 2.3, 2.5, and 2.7 for timelines). These mice were all housed in standard cages under a 12 hr light/dark cycle, at ambient temperature controlled between 22C 23C under specific pathogen free conditions. Animals were given food and water ad libitum and maintained under veterinary supervision throughout the study. For the immediate time poin t immunohistochemical studies, performed by Dr. Bunmi Ojo, twelve mice were randomly assigned to each of three different exposure regimens, and the experimental groups were composed of n=4 animals per group. As seen in table 2.2 the groups consisted of exp osures to CPF only, CPF + PB + PER, and the control group exposed to no GW agent. For the immediate time point behavioral studies, performed by Dr. Laila Abdullah, ten mice were randomly assigned to each of two different exposure regimens, and the experime ntal groups consisted of n=5 animals per group. As seen in table 2.2 the groups consisted of exposures to PB + PER and the control group exposed to no GW agent.
29 TABLE 2.2. IMMEDIATE POST EXPOSURE Immunohistochemistry Behavior CPF CPF + PB + PER Contr ol PB + PER Control 4 mice 4 mice 4 mice 5 mice 5 mice In the triple exposure delayed post exposure study, 24 mice were randomly assigned to each of two different exposure regimens. For the exposed group, 12 randomly selected mice were exposed to CPF + P B + PER. The control group consisted of 12 randomly select ed mice exposed to no GW agent (T able 2.3 ) In the double exposure delayed post exposure study, 20 mice were randomly assigned to two different exposure regimens. One group consisted of 10 mice expo sed to PB + PER. The control group consisted of 10 mice exposed to no GW agent (T able 2.3 ) TABLE 2.3. DELAYED POST EXPOSURE Double Exposure Study Triple Exposure Study PB + PER Control CPF + PB + PER Control 10 mice 10 mice 12 mice 12 mice These expe riments were performed in accordance with OLAW and NIH guidelines under a protocol approved by the Roskamp Institute IACUC. 2.1.2. Chemicals All GW agent treated mice were administered either (i) 5mg/kg CPF, (ii) combined 0.7mg/kg PB and 200mg/kg PER, or (iii) combined 5mg/kg CPF, 0.7mg/kg PB and 200mg/kg PER for 10 consecutive days. For combined exposure regimens (ii and iii), exposures were delivered in a single intraperitoneal injection volume in 50L dimethyl sulfoxide (DMSO). The controls were injecte d with a 50L volume of only 100% DMSO for 10 consecutive days. CPF and PB (99.4%) were purchased from Fisher Scientific,
30 Hanover Park, IL; PER (98.3% mixture of 27.2% cis and 71.1% trans isomers) was purchased from Sigma Aldrich, St. Louis, MO. Individual mice were weighed and monitored in their home cages on a daily basis for visible cholinergic behavioral symptoms (e.g., diarrhea, urination, excessive lacrimation, respiratory difficulties, muscle fasciculations) or other signs of pain or distress. No ove rt cholinergic toxicity was noticeable in any of the mice, which is consistent with the sub threshold level of exposures used in this study. The doses of GW agents were either chosen based in accordance with the relative level of exposure experienced by GW I veterans or previously established doses from in vivo animal studies. Moreover, these chosen dose regimes have been extensively reported to recapitulate detrimental effects on long term memory, as well as biochemical and neuropathological outcomes in pre vious rodent models (Abdullah et al., 2011; Abdullah et al., 2012; Dodd and Klein, 2009; Terry et al., 2003). Detailed rationales for the chosen doses for exposure have previously been provided (Speed et al., 2012; Steele et al., 2012). 2.2. Barnes Maze T he Barnes Maze has been widely used to measure the spatial and contextual memory of rodents and thereby the functionality of the hippocampus and associated areas recall spa tial cues and thereby evaluate the functionality of the hippocampus (see Figure 2.1). Together with the foot shock paradigm and the elevated plus maze, used to compare anxiety states (see Figure 2.5), the Barnes maze makes a good tool to evaluate cognition Training trials were administered at the immediate testing time point to the PB + PER
31 mouse model at one week post exposure, and the delayed testing time point PB + PER and CPF + PB + PER mouse models at 15 months post exposure to determine spatial learn ing (Figure 2.2 for Barnes maze timeline) (Tables 2.1, 2.2, and 2.3). The same procedures were used for all models. animal in the middle of the maze an d allowing it to find the escape hole, under which a dark chamber was placed. The surface of the maze was made aversive, by using bright light and a spinning fan. This increased the likelihood that the mouse entered the escape hole. During the training tri als, if the mouse did not find the hole the experimenter guided it to the hole. During testing trials the box was removed and the mouse looked for a virtual escape hole.
32 A) Immediate Testing Time Point PB + PER Barnes Maze Timeline B) Delayed Testing Time Point PB + PER Barnes Maze Timeline C) Delayed Testing Time Point CPF + PB + PER Barnes Maze Timeline FIGURE 2.2 shows the timeline for the Barnes maze protocol with the age of the mice and experimental timeframe for A) the imm ediate testing time point PB + PER model, B) the delayed testing time point PB + PER model, and C) the delayed testing time point CPF + PB + PER model, and it also shows the time difference between foot shock, Barnes maze, and elevated plus maze training periods.
33 2.2.1. Pr otocol Four training trials per mouse were conducted on each day over the course of 4 days, and each trial lasted for 180 seconds with an inter trial interval of 15 minutes. Each mouse was placed in the middle of the maze and allowed to explore, and the tr ial ended when a mouse entered the escape hole or after 180 seconds had elapsed. Throughout the trials, a rotating fan was placed near the maze while mice explored, but was turned off when a mouse escaped into the box through the target hole. Once a mouse was in the escape box, it was allowed to stay for 1 minute. If a mouse did not reach the escape hole within 180 seconds, the experimenter gently guided it to the escape hole, where it stayed for 1 minute (Figure 2.1). Twenty four hours after the last trai ning day, a testing trial was conducted for each mouse, during which the escape hole was removed and mice had 90 seconds to explore the maze and find a virtual hole. This served as a means to evaluate spatial memory and cognitive function. The number of po kes in each hole and the latency and distance traveled to reach the virtual target hole location were measured as outcome factors. To assess long term retention, an additional testing trial was administered approximately 1 month following the first testing trial. 2.2.2. Statistical Analysis For the Barnes Maze performed on PB + PER at the immediate post exposure time point, statistical analyses were conducted using a generalized linear model to accommodate non normally distributed dependent variables. For t he Barnes Maze
34 conducted at 15 months (delayed) post exposure to PB + PER, generalized linear model was utilized for the training trial but mixed linear model regression was used for the testing trials. For the Barnes Maze study conducted on the CPF + PB + PER GW agent mouse model at the 15 months (delayed) post exposure time point, mixed linear model regression was used, as data were normally distributed. All analyses were performed with SPSS 13.0 (IBM corp. Armonk, NY) and the statistical significance wa s set at the alpha 0.05 level. 2.3. Foot Shock This training paradigm was chosen based on previous work showing morphological as well as physiological changes in the amygdala following fear conditioning (Heinrichs et al., 2013). These changes suggest tha t the amygdala is highly involved in the formation of a fear memory as well as the integration of the conditioned and unconditioned stimulus and the generation of the conditioned response. Furthermore, this training paradigm also measures stress levels gen erated by the fear memories created during the training cycles. Therefore, with the fear conditioning paradigm it may be possible to discern whether or not there are pathological stress/anxiety differences between GWI mice and control mice, and whether the re is a difference in learning the fear conditioning. The foot shock paradigm was administered to 23 CPF + PB + PER exposed and control mice at the 15 months post exposure time point (see Figure 2.3 for foot shock timeline) (Table 2.1 and 2.3). Of the 12 C PF + PB + PER exposed mice, 1 died during the 15 month period.
35 Delayed Testing Time Point CPF + PB + PER Foot Shock Timeline FIGURE 2.3 shows the foot shock timeline with the age of the mice and experimental timeframe for the delayed post exposure C PF + PB + PER model, and it also shows the time difference between foot shock, Barnes maze, and elevated plus maze training periods.
36 2.3.1. Protocol The 43.8 cm17.1 cm chamber used for this behavior paradigm was a shuttle box obtained from Med Associates Incorporated (ENV 010MC). The chamber, except for the metal grid floor, is made of Plexiglas to prevent the mice from climbing or holding on to the surfaces while the shock is administered. The foot shock apparatus was controlled by an 8 channel I/R contr oller obtained from Med Associates Incorporated (ENV 253C). In order to record and analyze freezing response, a camera was placed over the shuttle box and mice were tracked using the EthoVision tracking system from Noldus Information Technology I nc. ( Leesb urg, VA 20176 ) The paradigm for this study involved 3 training sessions and a final testing trial 9 days after the last training session. Each individual training session and the testing trial lasted a total of 180 seconds. Tracking initiated 3 second s after mice were placed inside the time spent in the shuttlebox (180 secs) was divided into four parts to evaluate mouse movement for freezing/anxiety or lack there of. (I) The exploration segment consisted of an 88 second period where mice either explored or showed freezing behavior. (II) After this 88 second period, mice were exposed to 30 seconds of a 2KHz pure tone sound cue, where once again their movement was ev aluated. (III) Immediately following this 30 second sound cue, a 1mA shock was administered for 2 seconds followed by (IV) 60 seconds of recovery time during which their movements were evaluated (Figure 2.4A). Nine days after the third fear conditioning t raining session, each mouse was tested for the retention of contextual fear and cued fear, by evaluating their ability to
37 predict/recall an aversive event. This evaluation was based on freezing, a behavior rodents use to display fear/anxiety. The 9 day sep aration was chosen based on findings that neuroplasticity based remodeling occurs in the basolateral amygdala in relation to increased fear and anxiety related behaviors (Mitra et al., 2005). First, the animals were placed into the foot shock chamber for 3 minutes in order to evaluate their ability to integrate the noxious stimuli and the context memory without the sound cue (Figure 2.4B). About two hours after context memory was tested, the animals were placed in a novel box where the pure 2KHz tone was pr esented after 120 seconds for 30 seconds and the integration of the sound cue and the shock was evaluated over the remaining 60 seconds (Figure 2.4C). Freezing duration was recorded on both occasions. In this study our interest was on the percent time spe nt freezing, based on the absence of body movement except for respiration.
38 FOOT SHOCK TRAINING AND TESTING A) Box A: Training session B) Box A: Context Shock Association C) Box B: Cue Shock Association FIGURE 2.4: The images displayed above show the different stages of the foot shock paradigm. Figure 2.1A shows the training session, where the mice are placed in box A and conditioned. Figure 2.1B shows the testing session, where the mice are tested for th e integration of the context and the shock in box A. Figure 2.1C shows a mouse in novel box B, where testing for the integration of the sound cue and the shock occurred. 2.3.2. Statistical Analysis For the foot shock study conducted on the CPF + PB + PER GW agent mouse model at the15 months (delayed) post exposure time point, mixed linear model regression was used, as data were normally distributed. All analyses were performed with SPSS 13.0 (IBM corp. Armonk, NY) and the statistical significance was set at the alpha 0.05 level.
39 2.4. Elevated Plus Maze al., 1985; Rodgers and Dalvi, 1997). This test has been widely used to test anxiety levels in genetically modified mice an d for drug discovery (Hogg et al. 1996). The elevated plus maze has been reliable in the screening of anxiolytic and anxiogenic drugs (Rodgers and Dalvi, 1997; Mechiel Korte and De Boer, 2003; Crawley, 2007). Therefore, this test was chosen to evaluate the role GW agents play in anxiety (see Figure 2.6). The elevated plus maze paradigm was administered to 23 CPF + PB + PER exposed and control mice at the 17 months post exposure time point (see Figure 2.5 for timeline) (see Table 2.1 and 2.3). Delayed Testin g Time Point CPF + PB + PER Elevated Plus Maze Timeline FIGURE 2.5 shows the elevated plus maze timeline with the age of the mice and experimental timeframe in respect to the foot shock and Barnes maze training periods for the CPF + PB + PER model.
40 FIGURE 2.5 displays the elevated plus maze. This maze was used to examine the anxiety differences between the treated group and the untreated group. The mice were placed in the middle of the maze and allowed to move freely. The nature of the closed arms i s to create a sense of security, and the opened arms to create a sense of vulnerability since mice in general prefer confined spaces and the maze is elevated. Therefore, if a treated mouse spent more time in the closed arms as compared to the controls, it is regarded as more anxious. 2.4.1. Protocol The apparatus used for the elevated plus maze test is made from plastic and designed in a plus sign configuration. The maze is placed 50 cm above the floor, and is comprises 25 x 5 cm opened arms, placed acros s from each other and perpendicular to two 25 x 5 x 16 cm closed arms. There is also a platform center, measuring 5 x 5 cm, which divides the arms. The open arms are platforms without any walls, whereas the closed arms have 16 cm high sidewalls enclosing t he arms (Figure 2.5). The behavior testing room illumination level is kept relatively dim. A mouse is placed in the center area of the maze with its head directed toward a closed arm. The elevated plus maze test is recorded using a video camera attached to a computer. The and the time spent in the open arms are recorded and these measurements se rve as an
41 indication of anxiety related behavior. Anxious mice tend to spend mo re time in the closed arms. Mice are allowed to move freely about the maze for 5 minutes, while being recorded and tracked. Each mouse received one test trial. The application used for acquiring and analyzing the behavioral data was the EthoVision trackin g system from Noldus. The distance traveled, the number of entries into each arm, the time spent in each arm, and the percent of entries into the open arms are calculated by the Etho Vision system. After each trial, all arms and the center area are cleaned with Quatracide PV 15 amonium chloride solution, which is an efficient odor removal agent used to prevent a bias based on olfactory cues. 2.4.2. Statistical Analysis For the EPM study conducted on CPF + PB + PER GW agent mouse model at 15 months (delayed) post exposure time point, mixed lineal model regression was used because data were normally distributed. All analyses were performed with SPSS 13.0 (IBM corp. Armonk, NY) and the statistical significance was set at the alpha 0.05 level. 2.5. Immunohistoc hemical Staining Inflammation, typified by astrogliosis and microgliosis, is common in most brain related illnesses (Reichenbach et al., 2010; Zhang et al., 2010). Therefore, the activation of astroglia and microglia was evaluated in different brain region s. These regions include the cortex, hippocampus, and the amygdala. See figure 2.7A for the immediate testing
42 time point CPF + PB + PER and CPF alone models timeline, and see figure 2.7B for the delayed testing time point PB + PER model timeline. 2.5.1. G lial Fibrillary Acidic Protein (GFAP) Glial fibrillary acidic protein (GFAP ) is expressed in the CNS in astrocyte cells. GFAP is known to be involved in cell cell communication, including astrocyte astrocyte and astrocyte neuron communication, and is invol ved in the preservation of the blood brain barrier (Liedtke et al., 1996; Weinstein et al., 1991). It has been demonstrated that astroglial response to noxious events consists of hypertrophy (astrogliosis), proliferation (astrocystosis), and increased GFAP expression (Reichenbach et al., 2010). Once astroglia become activated, a neuroprotective role is assumed. If the activation status is prolonged, astrocytes may switch to a neurotoxic phenotype by releasing cytokines and inflammatory factors. Furthermore, this astrocytic switch compromises the normal physiological role, which may result in alterations in cerebrovascular function, cell cell communication, and signaling pathways (Reichenbach et al., 2010; Achour and Pascual, 2010; Abdullah et al., 2011). The refore, immunohistochemical staining against GFAP is a good way to observe astroglia activity in different areas of the brain. This approach was used, and astroglia activity was evaluated in the cortical, hippocampal, and amygdalar regions. The mice used f or GFAP staining included 12 mice from the immediate post exposure CPF + PB + PER model and the immediate post exposure CPF model, and 10 mice from the delayed post exposure PB + PER model (see Table 2.1, 2.2, and 2.3). The same procedures were used for al l models.
43 2.5.2. Ionized Calcium Binding Adapter Molecule 1(IBA1) AIF1 is a gene that is localized in a segment of the major histocompatibility complex class III region. This gene is highly expressed in the testes and the spleen, but under normal conditio ns it is not readily expressed in brain tissue. However, activated microglia cells, like macrophages, do express this gene when there is brain inflammation due to neuronal injury or when disease states compromise the integrity of the tissue (Ohsawa et al., 2004). The expression of this gene leads to the increased production of debris by activated microglia (Ohsawa et al., 2004). Therefore, immunohistochemical staining agains t IBA1 was used to evaluate the correlation, if any, of microglial activation in the cortex, hippocampus, and the amygdala. The mice used for IBA1 staining included 12 mice from the immediate post exposure CPF + PB + PER model and the immediate post exposu re CPF model, and 10 mice from the delayed post exposure PB + PER model (see Table 2.1, 2.2, and 2.3). The same procedures were used for all models.
44 A) Immediate Testing Time Point CPF + PB + PER and CPF alone Immunohistochemistry Timeline B) Delayed Test ing Time Point PB + PER Immunohistochemistry Timeline FIGURE 2.7 shows the timeline with the age of the mice and experimental timeframe for the i mmunohistochemistry experiments. A) displays the immediate testing time point CPF + PB + PER and CPF alone mo dels, and B) displays the PB + PER model. 2.5.3. Protocol All animals were deeply anesthetized with isoflurane before being intracardially perfused by gravity drip with a heparinized PBS solution (pH 7.4 ) After the perfusion, the brain was collected and post fixed overnight with 4% paraformaldehyde and paraffin embedd ed Using the known bregma coordinates, separate series of 5 6 m thick sagittal sections were cut using a microtome (2030 Biocut, Reichert/Leica, Grove, IL). These sections included the c ortex, hippocampus, and amygdala. The cut sections were mounted onto positively charged glass slides obtained from Fisher (Superfrost Plus, Pittsburg, PA). The sections were deparaffinized in xylene and then exposed to a rehydrating ethanol series in prepa ration for the immunohistochemical procedure. Sections were then
45 rinsed in water and subsequently incubated at room temperature in a solution of endogenous peroxidase blocking solution, containing 0.3% hydrogen peroxide diluted in phosphate buffer solution (PBS: 0.1M, pH 7.4) for 30 minutes. Sections were furthe r incubated with protein block serum free solution (Dako, Carpentaria, CA) for a period of 1hr in a humid chamber at room temperature. Series of sections were stained in batches with primary antibodi es made up in antibody diluent background reducing agent, raised against: (I) GFAP (rabbit anti GFAP, 1:10,000, Dako, Carpentaria, CA) for astrocyte activation, (II) and IBA1 (goat polyclonal anti Iba1, 1:1000, Abcam, Cambridge, MA) for microglia activatio n. After overnight incubation with the relevant primary antibodies, sections were rinsed with PBS, transferred to a solution containing the appropriate secondary antibody (from the Vecatastain Elite ABC Kit, Vector Lab, Burlingame, CA) for 1hr, and then in cubated with avidin biotin horseradish peroxidase solution (Vectastain Elite ABC kit; Vector Lab, Burlingame, CA) for a further hour. diaminobenzidine (DAB) chromogen, and hydrogen peroxide. Development with the chromogen was timed and applied as a constant across batches to limit technical variability (in immunodetection) before progressing to quantitative image analysis. The reaction was terminated by rinsing sections in distilled water. Finally, mounted section s were progressed through a dehydration gradient series of alcohols, cleared in xylene and coverslipped with permanent mounting medium. Immunoreacted sections were viewed using an Olympus (BX60) light microscope and photos taken using an Olympus MagnaFire SP camera.
46 2.5.4. Statistical Analysis The relationships between staining/cell counts in each brain area and treatment group were examined by one way analysis of variance (ANOVA) using a criterion of p<0.05 to assess group differences and post hoc (Tukey diff erences test. A non parametric Mann Whitney U test was used where the sample size did not fit the normal Gaussian distribution. All analyses were performed with SPSS 17.0 (IBM corp. Armonk, NY).
47 RESULTS 3.1. B ehavior The results below will display foot shock data for the delayed testing time point PB + PER + CPF exposure regimen, Barnes maze data for the immediate testing time point PB + PER and the delayed testing time point PB + PER and PB + PER + CPF exposur e regimens, and elevated plus maze for the delayed testing time point PB + PER + CPF exposure regimen. 3.1.1. Barnes Maze Immediate Post Exposure Testing Time Point PB + PER For the immediate testing time point PB + PER post exposure studies, training tr ials were conducted daily on post exposure days 10 13 to train mice to locate and enter the escape hole. Figure 3.1A shows that exposed mice learned this task better than control mice, as evidenced by a shorter duration of time spent in the arena over the 4 day training period (Wald = 12.08, p < 0.01). For testing trials, on post exposure day 14, exposed mice more frequently visited the escape hole location than control mice. On post exposure day 47, there were no differences between the two groups. However on post exposure day 77, exposed mice less frequently visited the escape hole than control mice and these differences were statistically significant (Wald = 5.93, p = 0.05, Figure 3.1B). Examination of primary errors (represented by a comparison of perce ntage frequency of nose poke in escape hole vs. other holes not including two holes adjacent to the target hole) provided supporting data that on post exposure day 77, exposed mice had less
48 accuracy than control mice in identifying the escape hole (Wald = 3.03, p = 0.08, Figure 3.1C). Barnes Maze: PB + PER Immediate Post Exposure Tests A) Duration of Time in Arena B) Frequency of Nose Pokes in During Training Trials Escape Hole During Testing Trials C) Correct Nose Pokes vs. Incorrect Pokes During Testing Trials FIGURE 3.1: Barnes Maze Mean SEM, n = 10 within each of the exposed group and the control group. (A) For the training trials, duration of time in arena was lower for P B +PER exposed mice than controls and a significant interaction between training days and exposure was observed, *denotes p < 0.01 for a main exposure effect (based on generalized linear model where training days and exposure were incorporated as fixed mai n factors). (B) Data on the frequency of nose pokes in the escape hole during the testing t rials showed a significant interaction between exposure and post exposure days on this outcome,
49 where exposed mice performed better than control mice on day 14, but this pattern reversed on day 77, p < 0.05 (based on generalized linear model with days and exposure were incorporated as fixed main factors). (C) Examination of primary error % (as measured by % frequency of correct nose poke vs. those in other holes outside of target zone) during testing trials showed that the error rate was numerically increas ed in PB+PER exposed mice compared to control mice on post exposure day 77, p > 0.05. 3.1.2. Barnes Maze Delayed Post Exposure Testing Time Point PB + PER The Barnes m aze protocol was performed at the 15 month post exposure time point and it was noted that during training while control mice showed proper learning following each training session, exposed mice performed poorly and their performance was highly variable as evidence by the total distance traveled by each group (Wald = 25.7, df = 12, p = 0.012 for an interaction between trials and exposure) (Figure 3.2A) T he total time spent in the arena was not significant ( data is not shown ) and the nose poke difference was also not significant However, during the testing trial immediately follow ing the training, a statistically significant difference was noted between exposed and control mice for distance to target hole and a similar pattern was also observed during the testing trial performed 30 days later (F = 13.1, p = 0.001) (Figure 3.2B)
50 Barnes Maze: PB + PER Delayed Post Exposure Tests A) Total Distance Traveled During Training Trials B) Total Distance Traveled During Testing Trials FIGURE 3. 2 : Barnes Maze Mean SEM, n = 10 within each of the exposed group and the cont rol group. A) Depicts the 4 training sessions and the total distance traveled during the training trials, as is seen this is lower for the controls as compared to the PB +PER exposed mice. P< 0.05 B) This graph depicts the variable distance traveled to th e target hole during the testing trial. A statistically significance effect of exposure was observed, where control mice traveled less distance before getting to the target hole compared to exposed mice, p < 0.05).
51 3.1.3. Barnes Maze Delayed Post Exposu re Testing Time Point CPF + PB + PE R For delayed post exposure studies, there were no differences between the exposure and the control group on learning during the training trials (p > 0.05, data not shown). Testing trials were conducted on the 15, 16, and 17 month post exposure time points, and as seen in f igure 3. 3 exposed mice learned this task better than control mice, as evidenced by a shorter distance to target hole in every testing trial (Wald = 4.6, p = 0.04) Barnes Maze: Delayed Post Exposure P F + PB + PER FIGURE 3. 3 : Barnes Maze Mean SEM, n = 12 within each of the exposed group and the control group Distance to target hole was lower for CPF+ PB+PER exposed mice than controls during testing p < 0.05. 3. 1 4 Foot Shock Delayed Post E xposur e Tim e Point CPF + PB + PE R The foot shock paradigm was performed at the 15 month post exposure time point. As can be seen in figure 3. 4 the exposed mice showed significantly higher
52 immobility than the control group on the third training day (t t est = 2.17, df = 21, p = 0.04) However, during the testing trial there were no significant differences between the two groups (data not shown). Foot Shock: Delayed Post Exposure CPF + PB + PER FIGURE 3. 4 : The results for the foot shock paradigm SEM are presented as the duration each group spent immobile during the three training days. On day 1 the exposed group spent more time frozen, but it was only day 3 that a statistical significance was observed. The testing trial data is not presented as it was not significant. 3. 1 5 Elevated Plus Maze Delayed Post Exposure Time Point CPF + PB + PER For the delayed post exposure CPF + PB + PER elevated plus maze, there were no significant differences between the exposed and control mice. These d ata are not displayed.
53 3.2. Immun o histochemistry The data below display s GFAP and IBA1 immunohistochemistry for the immediate post exposure testing time point PB + PER + CPF, CPF alone, and PB + PER models, as well as the delayed post exposure testing time point PB + PER and PB + PER + CPF models. 3.2.1 GFAP Staining of Immediate Post Exposure Time Point PB + PER + CPF and CPF Alone GW agent exposure upregulates astrocytic activation. Figure 3.2 shows the effects of GW agents on astrocytic activation as determined by GFAP immunostaining. Exposure to CPF + PB + PER treatment significantly increased GFAP immunoreactivity in the piriform cortex and the basolateral amygdala both qualitatively and quantitatively ( Figures 3.5 and 3.7G and I, M and O ). Exposure to CPF alone increased GFAP cell immunostaining in the motor cortex and CA3 ( Figure 3.7D, E, J, K ). However this change was only significant in the motor cortex ( Figure 3.5 ). Activa ted GFAP cells were hypertrophic with a prominent cell body and thick intermediate filaments ( Figure 3.7K ). 3.2.2. IBA1 Staining of Immediate Post Exposure Time Point PB + PER + CPF and CPF Alon e Post Exposure s Subtle effect s of G W agents on microglial cells are seen in figure 3.4. Analysis of Iba1 immunostaining after GW agent exposure showed minimal changes in all regions examined, following both qualitative and quantitative analysis ( Figure 3.6 and 3.8 ). A slight but noticeable increase was seen in the motor and piriform cortices following CPF
54 treatment, compared to controls, however this was not statistically significant (P<0.05). The morphological characteristics of (Iba1+) cells were relatively similar in each region, f or all groups examined. Positive cells resembled resting/quiescent microglia cells ( see Figure 3.8 ). Therefore neither the single nor the combined GW agent exposures had any significant effect on microglia activation at this time point. GFAP: Immediate Pos t Exposure CPF + PB + PER, and CPF alone FIGURE 3. 5 : Immunohistochemical reaction to GFAP SEM shows a significant increase in GFAP immunoreactivity (by 1.5 2fold) in the piriform cortex and basolateral amygdala (BLA) following CPF+PB+PER treatment compared to control (and CFP treatment alone). Combined CPF+PB+PER treatment did not alter GFAP immunoreactivity compared to control (DMSO), in the corpus callosum (CC), motor cortex and hippocampus. GFAP immunoreactivity was significantly lower in both t he cortex and hippocampus of mice treated with all three agents, as compared to treatment with CPF alone. CPF treatment alone also significantly increased GFAP immunoreactivity in the motor cortex. Data in (A D) are plotted as mean values SEM. (*) p<0.05 (**) p<0.01.
55 IBA1: Immediate Post Exposure CPF + PB + PER, and CPF alone FIGURE 3. 6 : Immunohistochemical reaction to IBA SEM shows a minimal effect on Iba1 immunost aining in all regions examined, following CPF or CPF+PB+PER treatment. A slight increase was noticed in the motor and piriform cortex (by 1.5 fold) following CPF treatment, compared to controls, however this was not statistically significant (P<0.05). Data in (A D) are plotted as mean values SEM. (*) p<0.05, (**) p<0.01.
56 FIGURE 3.7 shows GFAP immunoreactivity detected (400X) within the corpus callosum (CC), motor cortex (M cortex), piriform cortex (P cortex), hippocampus (CA3 region) and basolateral amygdala (BLA) of 11 week old mice treated with DMSO (A, D, G, J, M), CPF (B, E, H, K, N) or CPF+PB+PER (C, F, I, L, O). Combined CPF+PB+PER treatment dramatically increases GFAP immunoreactivity in the piriform cortex and basolateral amygdala compared to controls (G I; M O). GFAP immunostaining was greater in the CPF+PB+PER animals compared to CPF animals, within these same regions. In the motor/piriform cortex, hippocampus (see CA3) and basolateral amygdala, CPF treatment alone increases GFAP
57 immunostaini ng (E, K, N). This response was particularly evident in the CA3 of the hippocampus (K). Morphologically, these activated cells demonstrated hypertrophic cell somata and thicker/numerous intermediate filaments (see inset K, three stars). Scale bar represent s 75 m in all images. FIGURE 3.8 shows Iba1 immunoreactivity detected (600X) within the motor cortex (M cortex), piriform cortex (Pcortex), hippocampus (CA3 region) and basolateral amygdala (BLA) of 11 week old mice treated with DMSO (A, D, G, J), CPF (B E, H, K) or CPF+PB+PER (C, F, I, L). Changes in Iba1 immunoreactivity were minimal, however some notable effects were seen in the motor and piriform cortex. CPF treatment slightly increased I ba1 immunostaining in these regions (B, E) compared to controls The morphology of these (Iba1+) cells were relatively similar in each region, there was no indication of a dramatic increase in microglia activation characterized by; an increase cell somata or thicker/bushy cellular processes. Scale bar represents 37.5 m in all images.
58 3.2.3. GFAP Staining of Delayed Post Exposure Time Point PB + PER Figure 3. 9 shows that PB + PER exposure can upregulate astrocytic activation at the chronic post exposure time point. Astrocyte activation was observable in the CA3 region of the hippocampus and the basolateral amygdal a both qualitatively and quantitatively ( see Figure 3.9 and 3.10 C and D, E and F ) The GFAP staining was statistically significant in the hippocampus and amygdala (F = 11.7, p < 0.01) GFAP: Delayed Post Exposure PB + PER FIGURE 3. 9 : Immunohistoch emical reaction to GFAP SEM shows a significant increase in GFAP immunoreactivity in the CA3 region of the hippocampus and the basolateral amygdala following CPF + PB + PER treatment compared to control. Combined PB + PER treatment did not a lter GFAP immunoreactivity compared to control in the cortex.
59 FIGURE 3.10 shows GFAP immunoreactivity detected (400X) within the motor cortex (M cortex), piriform cortex (P cortex), hippocampus (CA3 region) and basolateral amygdala (BLA) of 16 mont h old mice treated with DMSO (A, C E ), or PB+PER ( B, D F ). Combined PB+PER treatment increases GFAP immunoreactivity in the CA3 region of the hippocampus and basolateral amygdala compared to controls (G and I; E and F ). Scale bar represents 75 m in all i mages. 3.2. 4 IBA1 Staining of Delayed Post Exposure Time Point PB + PER For the delayed post exposure PB + PER IBA1 immunohistochemistry stains, there were no significant differences between the exposed and control mice. These data are not displayed.
60 DISCUSSION Behavior Barnes Maze In the PB + PER immediate post exposure Barnes maze experiments, all mice learned how to escape into the target box. However, it was evident that the PB + PER exposed mice learned the task more quickly than control mice du ring post exposure training days 10 13 (see Figure 3.1A). This is supported by the fact that the exposed mice spent less time in the arena during these training days than the control mice, and a significant difference was observed on the last day of traini ng trials. The same pattern was evident in the testing trial carried out on post exposure day 14, as the exposed mice had a significantly higher number of nose pokes in the target hole as compared to the control mice. There were no differences between the two groups in the second testing trial carried out on post exposure day 47. However, in the third testing trial carried out on post exposure day 77, a reverse pattern was seen where exposed mice performed significantly worse than control mice (Figure 3.1B) This evidence suggests that there is an association between performance and the post exposure period, where the performance of the control mice got better over time, but that of the exposed mice decreased. Furthermore, when it comes to the percentage of nose poke errors, it was seen that exposed mice had a higher percentage of nose poke errors in all three of the testing trials, but it was only significant in the third trial (Figure 3.1C). These data demonstrate that the cognitive impairment observed in t his Barnes maze experiment is not an artifact of changes in control mice, as both exposed and control mice had lowered frequency of
61 nose pokes in the escape hole and an associated increase in the frequency of incorrect nose pokes when comparing the first t esting trial on day 14 to the third testing trial on day 77. This is, therefore, further proof that at an immediate post exposure testing time and impaired performance is directly correlated with increased post exposure time. In the delayed post exposure Barnes maze studies, like the immediate post exposure studies, both control and PB + PER exposed mice learned the task during the training trials. However, unlike the immediate post exposure studies, this delayed post exposure model took longer to learn the task. This is seen in figure 3.2A, where the data show that the exposed mice had a significantly higher distance traveled before going into the target hole as oppose d to the control mice. This effect is also seen during the testing trials carried out on the 15 and 16 month post exposure time point, where the exposed mice had a significantly higher distance traveled as compared to the control group (see figure 3.2B). T hese data suggest that at the delayed testing time point, the PB + PER model continue to perform worse than their control counterparts in a time dependent manner. It is apparent that at the immediate testing time point, the exposed mice performed better t han the control and progressively perform worse. It was also observed that at the delayed time point the exposed mice performed worse from the beginning of the training trials and continued to perform worse than the control mice. Therefore, the delayed pos t exposure testing model also suggests that the crippled performance seen in the PB + PER exposed mice is time point related. Learning is not affected in the immediate post
62 exposure PB + PER mice in the beginning, as evidence d by their ability to learn the task efficiently. However, there seems to be a problem as time passes in memory recall or long term memory storage as is seen in their diminished performance in recalling the target hole during the testing trials. During the delayed post exposure time po int for the PB + PER mice, there were impairments both in learning and memory recall as these mice performed significantly worse than their control counterparts during the training and testing phases of the experiment. It was observed that mice exposed to CPF + PB + PER learned the task just as well as the control mice during the training trials at a delayed post exposure time point, as there were no differences in their performance (data not shown). In fact, as is seen in figure 3.3, the exposed mice had a significantly lower distance traveled before reaching the target hole in the first, second, and third testing trials, carried out at the 15, 16, and 17 month post exposure time point respectively. These data suggest that the exposed mice performed better in this task than the control group, which is interesting when compared to the PB + PER immediate and delayed post exposure mice. Since the mice with the triple exposure (CPF + PB + PER) performed better compared to their control counter parts than the mi ce with the double exposure (PB + PER), it is theorized that the combination of CPF and PB may ameliorates the behavioral symptoms. Although in these experiments the data suggest that CPF ameliorates PB + PER damage, this is contradictory to the expected e ffects of PB and CPF activation. It is necessary to have a delayed CPF exposure group and compare the effects to the CPF + PB + PER. Comparing these exposure regimens holds promise for further studies, which may help
63 models get closer to the cognitive impa irment seen in GWI clinical and pre clinical settings, and elucidate new modes of action for these chemicals. Foot Shock During the training sessions, the exposed mice spent more time frozen as compared to the controls. This was significant on the third training day (Figure 3.4), however during the testing trials there was no significant difference between the control and exposed group (data not shown). This lack of significance during the testing trial e were 16 months old when exposed to this paradigm. This same factor might also be present in the Barnes maze, since these mice were the same age in the delayed post exposure testing group. A good way to rule this possibility out would be to perform this f ear conditioning paradigm on an immediate post exposure CPF + PB + PER model, a PB + PER only and CPF only delayed post exposure testing models. The immediate time point triple exposure model would serve as a fresh point of comparison as young mice should not have issues in learning. Also, the PB + PER and CPF only delayed time point models would help to provide insights on the interaction between PB and CPF, as this may be a possible reason for the lack of difference between exposed and control mice during the testing trials. Another possible variable that might have prevented the treated mice from being significantly different during the testing trials was the length of the trial. Perhaps if more time was given to evaluate the freezing response, a signific ant difference would be achieved. Although it was not significant a trend towards increased anxiety was observed in the treated mice.
64 Elevated Plus Maze During the elevated plus maze paradigm there were no significant differences observed between the d elayed post exposure testing time point CPF + PB + PER mice and their control counterparts. Once again, it may be speculated that the triple exposure does not have as serious an effect as the double exposure in the anxiety and memory aspects of cognition d ue to PB protection. As stated above, anxiety behaviors seen in the open field test have been noted before with the PB + PER mice, however it would be interesting to also use these mice in the elevated plus maze paradigm. Immunohistochemistry It was hypo thesized that astroglia activation would be apparent in the GWI mouse models, since astroglia displaying acetylcholine receptors may be activated by increased levels of acetylcholine in the synapse, and this is a feature of acute exposure to GW agents. Thi s was verified in the immediate PB, PER, and CPF post exposure models. It was observed that systemic administration of CPF treatment (alone), significantly upregulated GFAP immunoreactivity and increased the hypertrophic astrocytic phenotypes in the motor cortex and hippocampus of exposed mice compared to controls. A similar effect was also shown with CPF in combination with PB + PER in the piriform cortex and basolateral amygdala. This apparent impact on different brain regions is consistent with the heter ogeneity of GWI symptoms such as olfactory dysfunction, cognitive deficits and anxiety, which relate to these different cholinergic enriched sites in the brain (Barlow et al., 2008). A greater effect on GFAP immunoreactivity was observed with CPF + PB + PE R treatment compared with CPF alone, in the piriform cortex and
65 basolateral amygdala. The reason behind this greater effect seen with the combined chemical dosing regimen compared to individual GW agent exposure has been linked to the potential interaction between these chemicals on their metabolic rate of detoxification (Ma and Chambers, 1994; Sultatos et al., 1984) Pyridostigmine bromide and PER are metabolized by both human plasma and liver microsomal enzymes, while the major site of CPF metabolism is th e liver, where it is rapidly bioactivated (desulfurated) by a P 450 dependent monooxygenase to CPF oxon (Ma and Chambers, 1994; Sultatos et al., 1984). The rate of detoxification of CPF oxon is rapid and therefore it is rare to find this compound in body f luid samples, except at very high exposures (Nolan et al., 1984). However, a combined exposure to all these chemicals may increase their neurotoxicity as it has previously been shown that combined exposure to these or similar compounds bility to eliminate them, due to competition for detoxifying enzymes (Choi et al., 2004; Usmani et al., 2002). This therefore raises the significance of the interaction between combined GW agent exposure, and further supports the retrospective reports that veterans exposed to a combination of low levels of organophosphate and pyrethroid based pesticides, and PB pills had a higher prevalence of GWI (compared to single GW agent exposure) (Barlow et al., 2008). In addition to the aforementioned findings on ast rogliosis, there were differential effects on specific brain regions with either treatment. For example, CPF alone increase d the immunoreactivity of GFAP in the motor cortex and hippocampus, however this effect was not present with the combination of CPF w ith PB+PER. In the delayed post exposure time point PB + PER models, it was observed that astrogliosis was significantly increased in the hippocampus and basolateral amygdala of the treated mice as compared to the controls. The reason behind these
66 regional brain specific effects is unknown, but may be attributed to differential modulation of cholinergic activity in astrocyte populations located in different sites of the brain. Brain related illnesses usually affect neuronal integrity, which in turn influenc es the activation of glial cells (Reichenbach et al., 2010). Reactive astrogliosis (and inflammation) is a common feature of most neurodegenerative disorders, and, in support of our data, has been consistently shown to occur both in vitro and in vivo in an imal models following (single and combined) chronic exposure to various GW agents, such as PB, CPF, PER and N,N diethyl meta toluamide (DEET) (Abdel Rahman et al., 2004; Abdullah et al., 2011) Protoplasmic and fibrous astrocytes undergo certain cellular al terations in response to acute injury, classified by hypertrophy (astrogliosis), proliferation (astrocytosis) and increased GFAP expression. Once activated, astrocytes could potentially be neuroprotective. However if their activation status is prolonged, a strocytes may switch to a neurotoxic phenotype by releasing inflammatory cytokines and cytotoxic factors, including surrendering their normal physiological roles that could adversely affect neuronal and cerebrovascular function (Achour and Pascual, 2010; R eichenbach et al., 2010). Mechanisms behind the upregulation of astroglial activation in the brain after GW agent exposure are currently under debate. Astrocytes are known to ion of these receptors is linked to astrocytic activation (Liu et al., 2012) GW agents could potentially exert their effects via ACh receptors on astrocytes in the brain directly, or alternatively in directly through the peripheral nervous system (Borovikov a et al., 2000).
67 Reactive microglia with Iba1 immunostaining were not observed after exposure to CPF alone or combination with PB and PER in the immediate post exposure time point. Reactive microglia were also not seen at the delayed PB + PER post exposur e time point. These findings are consistent with previous work done at the Roskamp Institute showing a complete lack of microglial activation in two different mouse models of GW agent exposure; PB + PER (8 days post exposure) and PB + PER + DEET + Stress ( 150 days post exposure) (Abdullah et al., 2011; Abdullah et al 2012). This observed difference between astroglia and microglia responses may indicate cell specific cholinergic effects in vivo which warrant further studies. Improvements and Suggestions T he work done with these models thus far has been interesting and productive, however a great deal of work is still necessary to fully elucidate the pathology of GWI. For instance it is necessary to directly look at the integrity of the cholinergic system. This includes AChE activity for all variants, and the activity of choline acetyltransferase which will give information regarding the production of acetylcholine and how it is altered (if at all) after GW agent exposure. This would have to be done at immed iate and delayed post exposure testing time points with different exposure models. Another helpful avenue to take would be to create models that incorporate PTSD. For example an experimental design incorporating a GWI/PTSD mouse model, GWI only, PTSD only, and control would help to pinpoint differences in PTSD and GWI pathology, particularly dealing with the HPA axis. Finally, it would be helpful to perform experiments on the same mice at an immediate and then at a delayed post exposure time point, and use
68 another type of biomarker to detect stress, such as cortisol, which may be collected from saliva and urine. Concluding Remarks Previous thought has categorized GWI as a primarily somatic disorder. However recent findings including behavioral and neuroima ging studies show a prominent CNS component to this illness. In fact it would appear that the CNS symptoms seen in GWI may be separated into different subtypes, depending on the cognitive aspects being evaluated. Sensory and motor impairments are also see n in these patients. Although there are a n umber of avenues being explored ( as may be seen in the clinicaltrials.g ov website when GWI is searched), currently there are no approved treatments available for GWI that can either provide symptomatic relief or t arget the underlying CNS pathology. An additional 50,000 GW veterans continue to suffer from unexplained illnesses, but have not received a formal diagnosis of GWI. These issues may be targeted with a better understanding of the pathobiology, and this may in turn be addressed using preclinical animal models, as these present a great deal of information regarding the effects of GW exposures. Current research is already changing the way that the government responds to GW related exposures. Although not nece ssarily related to GWI, in a Department of Defense report ( 1993 ) the Deputy Under Secretary of Defense introduced pest management initiatives which have resulted in improved standards and practices. These included reducing pesticide use in military instal lations by 50%. This was also accompanied by improved training of pesticide applicators and an increased awareness
69 among military personnel of pesticide use. For instance, a 20 33% DEET formulation is now provided to military personnel involved in Operatio n Iraqi Freedom compared to the 75% DEET that was issued during the first Gulf War. In the civilian population both here in the United States and other countries, the use of pesticides is of concern and has been implicated in a number of neurological illn esses. Findings reported from a longitudinal study in Cache county Utah indicate that the use of organophosphate AChE inhibitors can increase the risk of developing in the Bordeaux region of France followed agriculture workers over the course of 4 years. The results indicated that agriculture workers directly exposed to pesticides during midlife have an increased risk of early neurocognitive decline (Baldi et al., 2011) In the past decade, the Environmental Protection Agency has initiated a gradual elimination of organophosphate use in the urban household (2002). However, this has resulted in a switch to other easily available pesticides including PER and nearly 100 mil lion applications of PER are made each year in US homes (Power and Sudakin, 2007). This is accordance with data released by the American Association of Poison Control Centers Toxic Exposure Surveillance System, which show an increase in pyrethroid exposure related adverse events that positively correlate with a decline in organophosphate related incidences (Power & Sudakin 2007). Therefore, the exposure to these chemicals may increase the likelihood of developing neurological disorders. These disorders are not limited to GW veterans or the developed world. These exposures are readily seen in the developing world, where such environmental and health protecting organizations do not exist and education on the toxicity of these chemicals is lacking. As a result
70 understanding of the molecular mechanisms of such exposures may help to identify pesticide related neurocognitive dysfunction so that appropriate corrective measures may be taken to reduce the CNS complications associated with pesticide use. Although the current work is performed within the context of GWI, these findings provide novel insights into the neurotoxic mechanisms associated with these chemicals.
71 REFERENCES 1. Abdel Rahman A, Abou Donia SM, El Masry EM, Shetty AK, Abou Donia MB. Stre ss and combined exposure to low doses of pyridostigmine bromide, DEET, and permethrin produce neurochemical and neuropathological alterations in cerebral cortex, hippocampus, and cerebellum. Journal of Toxicology and Environmental Health, Part A 2004;67:1 63 192. 2. Abdel Rahman A, Shetty AK, Abou Donia MB. Disruption of the blood brain barrier and neuronal cell death in cingulate cortex, dentate gyrus, thalamus, and hypothalamus in a rat model of Gulf War syndrome. Neurobiol Dis 2002;10:306 326. 3. Abdulla h L, Crynen G, Reed J, et al. Proteomic CNS profile of delayed cognitive impairment in mice exposed to Gulf War agents. Neuromolecular medicine 2011;13:275 288. 4. Abdullah L, Evans JE, Bishop A, et al. Lipidomic profiling of phosphocholine containing bra in lipids in mice with sensorimotor deficits and anxiety like features after exposure to Gulf War agents. Neuromolecular medicine 2012;14:349 361. 5. Abou Donia MB, Dechkovskaia AM, Goldstein LB, Abdel Rahman A, Bullman SL, Khan WA. Co exposure to pyridos tigmine bromide, DEET, and/or permethrin causes sensorimotor deficit and alterations in brain acetylcholinesterase activity. Pharmacology Biochemistry and Behavior 2004;77:253 262. 6. Amourette C, Lamproglou I, Barbier L, et al. Gulf War illness: Effects of repeated stress and pyridostigmine treatment on blood brain barrier permeability and cholinesterase activity in rat brain. Behav Brain Res 2009;203:207 214. 7. Baldi I, Gruber A, Rondeau V, Lebailly P, Brochard P, Fabrigoule C. Neurobehavioral effects of long term exposure to pesticides: results from the 4 year follow up of the PHYTONER Study. Occup Environ Med 2011;68:108 115. 8. Bansal I, Waghmare C, Anand T, Gupta A, Bhattacharya B. Differential mRNA expression of acetylcholinesterase in the central nervous system of rats with acute and chronic exposure of sarin & physostigmine. Journal of Applied Toxicology 2009;29:386 394. 9. Barbier L, Diserbo M, Lamproglou I, Amourette C, Peinnequin A, Fauquette W. Repeated stress in combination with pyridostigmi ne: Part II: Changes in cerebral gene expression. Behav Brain Res 2009;197:292 300. 10. Barlow C, Clauw DJ, Meggs WJ, et al. Gulf War Illness and the Health of Gulf War Veterans. 2008.
72 11. Bartus RT, Dean RL, Beer B, Lippa AS. The cholinergic hypothesis of geriatric memory dysfunction. Science 1982;217:408 414. 12. Ben Achour S, Pascual O. Glia: the many ways to modulate synaptic plasticity. Neurochem Int 2010;57:440 445. 13. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000;405:458 462. 14. Choi J, Hodgson E, Rose RL. Inhibition of trans permethrin hydrolysis in human liver fractions by chloropyrifos oxon and carbaryl. Drug Metabol Drug Interact 2004;20:233 246. 1 5. Corbel V, Stankiewicz M, Bonnet J, Grolleau F, Hougard JM, Lapied B. Synergism between insecticides permethrin and propoxur occurs through activation of presynaptic muscarinic negative feedback of acetylcholine release in the insect central nervous syst em. Neurotoxicology 2006;27:508 519. 16. Cowan DN, Lange JL, Heller J, Kirkpatrick J, DeBakey S. A case control study of asthma among US Army Gulf War veterans and modeled exposure to oil well fire smoke. Mil Med 2002;167:777 782. 17. David A, Farrin L, Hull L, Unwin C, Wessely S, Wykes T. Cognitive functioning and disturbances of mood in UK veterans of the Persian Gulf War: a comparative study. Psychol Med 2002;32:1357 1370. 18. Davies P. A critical review of the role of the cholinergic system in human memory and cognitiona. Ann N Y Acad Sci 1985;444:212 217. 19. Department of Defense. U.S. Department of Defense, Office of the Special Assistant to the Undersecretary of Defense (Personnel and Readiness) for Gulf War Illnesses Medical Readiness and Milita ry Deployments. Environmental Exposure Report: Pesticides Final Report. Washington D.C.:April 17, 2003. Available from: http://www.dtic.mil/whs/directives 20. Department of Defense. U.S. Depar tment of Defense, Office of the Special Assistant to the Undersecretary of Defense (Personnel and Readiness) for Gulf War Illnesses Medical Readiness and Military Deployments. Technical Report: Modeling and Risk Characterization of U.S. Demolition Operatio ns at the Khamisiyah Pit. Washington, D.C.:April 16, 2002. 21. Dodd C, Klein B. Pyrethroid and organophosphate insecticide exposure in the 1 methyl 4 phenyl 1, 2, 3, 6 immunohistochemical analysis o f tyrosine hydroxylase and glial fibrillary acidic protein in dorsolateral striatum. Toxicol Ind Health 2009;25:25 39.
73 22. Doebbeling BN, Clarke WR, Watson D, et al. Is there a Persian Gulf War syndrome? Evidence from a large population based survey of ve terans and nondeployed controls. American Journal of Medicine, The 2000;108:695 704. 23. Donta ST, Clauw DJ, Engel Jr CC, et al. Cognitive behavioral therapy and aerobic exercise for Gulf War Veterans' illnesses. JAMA: the journal of the American Medical Association 2003;289:1396 1404. 24. Donta ST, Engel CC, Collins JF, et al. Benefits and Harms of Doxycycline Treatment for Gulf War Veterans' IllnessesA Randomized, Double Blind, Placebo Controlled Trial. Ann Intern Med 2004;141:85 94. 25. Drachman DA, L eavitt J. Human memory and the cholinergic system: a relationship to aging? Arch Neurol 1974;30:113. 26. Drachman DA, Sahakian B. Memory and cognitive function in the elderly: a preliminary trial of physostigmine. Arch Neurol 1980;37:674. 27. Eaton DL, D aroff RB, Autrup H, et al. Review of the toxicology of chlorpyrifos with an emphasis on human exposure and neurodevelopment. Crit Rev Toxicol 2008;38:1 125. 28. Friedman A, Kaufer D, Shemer J, Hendler I, Soreq H, Tur Kaspa I. Pyridostigmine brain penetrat ion under stress enhances neuronal excitability and induces early immediate transcriptional response. Nat Med 1996;2:1382 1385. 29. Golier JA, Caramanica K, Yehuda R. Neuroendocrine response to CRF stimulation in veterans with and without PTSD in consider ation of war zone era. Psychoneuroendocrinology 2012;37:350 357. 30. Golier JA, Legge J, Yehuda R. The ACTH response to dexamethasone in Persian Gulf War veterans. Ann N Y Acad Sci 2006b;1071:448 453. 31. Golier JA, Schmeidler J, Legge J, Yehuda R. Twent y four hour plasma cortisol and adrenocorticotropic hormone in Gulf War veterans: relationships to posttraumatic stress disorder and health symptoms. Biol Psychiatry 2007;62:1175 1178. 32. Golier JA, Schmeidler J, Legge J, Yehuda R. Enhanced cortisol supp ression to dexamethasone associated with Gulf War deployment. Psychoneuroendocrinology 2006a;31:1181 1189. 33. Golomb BA. Acetylcholinesterase inhibitors and Gulf War illnesses. Proceedings of the National Academy of Sciences 2008;105:4295 4300.
74 34. Gray GC, Kaiser KS, Hawksworth AW, Hall FW, Barrett Connor E. Increased postwar symptoms and psychological morbidity among US Navy Gulf War veterans. Am J Trop Med Hyg 1999;60:758 766. 35. Grisaru D, Sternfeld M, Eldor A, Glick D, Soreq H. Structural roles of acetylcholinesterase variants in biology and pathology. European Journal of Biochemistry 1999;264:672 686. 36. Gunderson CH, Lehmann CR, Sidell FR, Jabbari B. Nerve agents: a review. Neurology 1992;42:946 950. 37. Haley RW, Kurt TL, Hom J. Is there a Gu lf War syndrome. JAMA 1997;277:215 222. 38. Haley RW, Spence JS, Carmack PS, et al. Abnormal brain response to cholinergic challenge in chronic encephalopathy from the 1991 Gulf War. Psychiatry Research: Neuroimaging 2009;171:207 220. 39. Hayden K, Norto n M, Darcey D, et al. Occupational exposure to pesticides increases the risk of incident AD The Cache County Study. Neurology 2010;74:1524 1530. 40. Heinrichs SC, Leite Morris KA, Guy MD, Goldberg LR, Young AJ, Kaplan GB. Dendritic Structural Plasticity i n the Basolateral Amygdala after Fear Conditioning and its Extinction in Mice. Behav Brain Res 2013. 41. Hilborne, Golomb, Marshall, Davis & Sherbourne. Examining Possible Causes of Gulf War Illness: RAND Policy Investigation and Reviews of the Scientific Literature. 2005;RB 7544 OSD. 42. Hogg S. A review of the validity and variability of the elevated plus maze as an animal model of anxiety. Pharmacology Biochemistry and Behavior 1996;54:21 30. 43. Hoy J, Cornell J, Karlix J, Tebbett I, Haaren Fv. Repeat ed coadministrations of pyridostigmine bromide, DEET, and permethrin alter locomotor behavior of rats. Vet Hum Toxicol 2000;42:72 76. 44. Johnston MV, McKinney M, Coyle JT. Evidence for a cholinergic projection to neocortex from neurons in basal forebrain Proceedings of the National Academy of Sciences 1979;76:5392 5396. 45. Kaufer D, Friedman A, Seidman S, Soreq H. Anticholinesterases induce multigenic transcriptional feedback response suppressing cholinergic neurotransmission. Chem Biol Interact 1999; 119:349 360. 46. Kaufer D, Friedman A, Seidman S, Soreq H. Acute stress facilitates long lasting changes in cholinergic gene expression. Nature 1998;393:373 377.
75 47. Kurt TL. Epidemiological association in US veterans between Gulf War illness and exposure s to anticholinesterases. Toxicol Lett 1998;102:523 526. 48. Lamproglou I, Barbier L, Diserbo M, Fauvelle F, Fauquette W, Amourette C. Repeated stress in combination with pyridostigmine: Part I: Long term behavioural consequences. Behav Brain Res 2009;19 7:301 310. 49. Lange G, Tiersky L, DeLuca J, et al. Cognitive functioning in Gulf War illness. Journal of clinical and experimental neuropsychology 2001;23:240 249. 50. Lange JL, Schwartz DA, Doebbeling BN, Heller JM, Thorne PS. Exposures to the Kuwait oi l fires and their association with asthma and bronchitis among gulf war veterans. Environ Health Perspect 2002;110:1141. 51. Lehmann J, Nagy J, Atmadja S, Fibiger H. The nucleus basalis magnocellularis: the origin of a cholinergic projection to the neocor tex of the rat. Neuroscience 1980;5:1161 1174. 52. Li B, Mahan CM, Kang HK, Eisen SA, Engel CC. Longitudinal health study of US 1991 Gulf War veterans: Changes in health status at 10 year follow up. Am J Epidemiol 2011a;174:761 768. 53. Li X, Spence JS, Buhner DM, et al. Hippocampal dysfunction in Gulf War veterans: investigation with ASL perfusion MR imaging and physostigmine challenge. Radiology 2011b;261:218 225. 54. Liedtke W, Edelmann W, Bieri PL, et al. GFAP is necessary for the integrity of CNS wh ite matter architecture and long term maintenance of myelination. Neuron 1996;17:607 615. 55. Liu P, Aslan S, Li X, et al. Perfusion deficit to cholinergic challenge in veterans with Gulf War Illness. Neurotoxicology 2011;32:242 246. 56. Liu Y, Hu J, Wu mediated neuroprotection against dopaminergic neuron loss in an MPTP mouse model via inhibition of astrocyte activation. J Neuroinflammation 2012;9:2094 2099. 57. Ma T, Chambers J. Kinetic parameters of desulf uration and dearylation of parathion and chlorpyrifos by rat liver microsomes. Food and chemical toxicology 1994;32:763 767. 58. Menon PM, Nasrallah HA, Reeves RR, Ali JA. Hippocampal dysfunction in Gulf War Syndrome. A proton MR spectroscopy study. Brain Res 2004;1009:189 194.
76 59. Mense SM, Sengupta A, Lan C, et al. The common insecticides cyfluthrin and chlorpyrifos alter the expression of a subset of genes with diverse functions in primary human astrocytes. Toxicological Sciences 2006;93:125 135. 60. Micheau J, Marighetto A. Acetylcholine and memory: a long, complex and chaotic but still living relationship. Behav Brain Res 2011;221:424 429. 61. Mitra R, Jadhav S, McEwen BS, Vyas A, Chattarji S. Stress duration modulates the spatiotemporal patterns of spine formation in the basolateral amygdala. Proc Natl Acad Sci U S A 2005;102:9371 9376. 62. National Institutes of Health. A service of the U.S. National Institutes of Health: Gulf War Illness. Available at: Available at: http://clinicaltrials.gov/ct2/results?term=gulf+war+illness&Search=Search Accessed Accessed April 3, 2013, 2013. 63. Nolan R, Rick D, Freshour N, Saunders J. Chlorpyrifos: pharmacokin etics in human volunteers. Toxicol Appl Pharmacol 1984;73:8 15. 64. Ohsawa K, Imai Y, Sasaki Y, Kohsaka S. Microglia/macrophage specific protein Iba1 binds to fimbrin and enhances its actin bundling activity. J Neurochem 2004;88:844 856. 65. O'Leary TP, Brown RE. Optimization of apparatus design and behavioral measures for the assessment of visuo spatial learning and memory of mice on the Barnes maze. Learning & Memory 2013;20:85 96. 66. Pellow S, Chopin P, File SE, Briley M. Validation of open: closed a rm entries in an elevated plus maze as a measure of anxiety in the rat. J Neurosci Methods 1985;14:149 167. 67. Pope C, Karanth S, Liu J. Pharmacology and toxicology of cholinesterase inhibitors: uses and misuses of a common mechanism of action. Environ T oxicol Pharmacol 2005;19:433 446. 68. Power LE, Sudakin DL. Pyrethrin and pyrethroid exposures in the United States: a longitudinal analysis of incidents reported to poison centers. Journal of medical toxicology 2007;3:94 99. 69. Ray DE, Fry JR. A reasse ssment of the neurotoxicity of pyrethroid insecticides. Pharmacol Ther 2006;111:174 193. 70. Reichenbach A, Derouiche A, Kirchhoff F. Morphology and dynamics of perisynaptic glia. Brain Res Rev 2010;63:11 25.
77 71. Rodgers R, Dalvi A. Anxiety, defence and the elevated plus maze. Neurosci Biobehav Rev 1997. 72. Salmon A, Erb C, Meshorer E, et al. Muscarinic modulations of neuronal anticholinesterase responses. Chem Biol Interact 2005;157:105 113. 73. Schliebs R, Arendt T. The significance of the cholinergi c system in the brain during J Neural Transm 2006;113:1625 1644. 74. Schmolck H, Kensinger EA, Corkin S, Squire LR. Semantic knowledge in patient HM and other patients with bilateral medial and lateral temporal lobe lesio ns. Hippocampus 2002;12:520 533. 75. Smith TC, Heller JM, Hooper TI, Gackstetter GD, Gray GC. Are Gulf War veterans experiencing illness due to exposure to smoke from Kuwaiti oil well fires? Examination of Department of Defense hospitalization data. Am J Epidemiol 2002;155:908 917. 76. Speed HE, Blaiss CA, Kim A, et al. Delayed reduction of hippocampal synaptic transmission and spines following exposure to repeated subclinical doses of organophosphorus pesticide in adult mice. Toxicological Sciences 2012; 125:196 208. 77. Steele L. Prevalence and patterns of Gulf War illness in Kansas veterans: association of symptoms with characteristics of person, place, and time of military service. Am J Epidemiol 2000;152:992 1002. 78. Steele L, Sastre A, Gerkovich MM, Cook MR. Complex factors in the etiology of Gulf War Illness: Wartime exposures and risk factors in veteran subgroups. Environ Health Perspect 2012;120:112. attenuates correlates. Proceedings of the National Academy of Sciences 2000;97:8647 8652. 80. Sultatos LG, Shao M, Murphy S. The role of hepatic biotransformation in mediating the acute toxicity of the phosph orothionate insecticide chlorpyrifos. Toxicol Appl Pharmacol 1984;73:60 68. 81. Terry AV, Stone JD, Buccafusco JJ, Sickles DW, Sood A, Prendergast MA. Repeated exposures to subthreshold doses of chlorpyrifos in rats: hippocampal damage impaired axonal transport, and deficits in spatial learning. J Pharmacol Exp Ther 2003;305:375 384. 82. Toomey R, Alpern R, Vasterling JJ, et al. Neuropsychological functioning of US Gulf War veterans 10 years after the war. Journal of the International Neuropsychological Society 2009;15:717.
78 83. Tracey KJ, Czura CJ, Ivanova S. Mind over immunity. The FASEB Journal 2001;15:1575 1576. 84. Usmani KA, Rose RL, Goldstein JA, Taylor WG, Brimfield AA, Hodgson E. In vitro human metabolism and interactions of repellent N, N diethyl m toluamide. Drug Metab Disposition 2002;30:289 294. 85. Van der Zee E, De Jong G, Strosberg A, Luiten P. Muscarinic acetylcholine receptor expression in astrocytes in the cortex of young and aged rats. Glia 1993;8:42 50. 86. Vythi lingam M, Luckenbaugh DA, Lam T, et al. Smaller head of the hippocampus in Gulf War related posttraumatic stress disorder. Psychiatry Research Neuroimaging Section 2005;139:89 100. 87. Weinstein DE, Shelanski ML, Liem R. Suppression by antisense mRNA demo nstrates a requirement for the glial fibrillary acidic protein in the formation of stable astrocytic processes in response to neurons. J Cell Biol 1991;112:1205 1213. 88. White RF, Proctor SP, Heeren T, et al. Neuropsychological function in Gulf War veter ans: relationships to self reported toxicant exposures. Am J Ind Med 2001;40:42 54. there a role for microglia? Mol Neurobiol 2010;41:232 241. 90. Zhang W, Yamada M, G omeza J, Basile AS, Wess J. Multiple muscarinic acetylcholine receptor subtypes modulate striatal dopamine release, as studied with M1 M5 muscarinic receptor knock out mice. The Journal of neuroscience 2002;22:6347 6352.