Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Behavioral deficits without motor neuron loss in mice fed cycad : implications for ALS Hawkes, Erin Lynne 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2005-0468.pdf [ 3.28MB ]
JSON: 831-1.0092095.json
JSON-LD: 831-1.0092095-ld.json
RDF/XML (Pretty): 831-1.0092095-rdf.xml
RDF/JSON: 831-1.0092095-rdf.json
Turtle: 831-1.0092095-turtle.txt
N-Triples: 831-1.0092095-rdf-ntriples.txt
Original Record: 831-1.0092095-source.json
Full Text

Full Text

B E H A V I O R A L DEFICITS WITHOUT MOTOR N E U R O N LOSS IN MICE FED C Y C A D : IMPLICATIONS FOR ALS? by ERIN L Y N N E H A W K E S B.Sc, Mount Saint Vincent University and Dalhousie University, 2003 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Neuroscience) THE UNIVERSITY OF BRITISH C O L U M B I A October 2005 © Erin Lynne Hawkes, 2005 Abstract Amyotrophic lateral sclerosis-parkinsonism dementia complex (ALS-PDC) is a neurodegenerative disease characterized by features of A L S , parkinsonism, and Alzheimer's disease (AD). Occurring primarily in Guam, it has been linked epidemiologically to the consumption of cycad. Similarly, mice fed cycad develop a neurological disorder that mimics ALS-PDC both behaviorally and neuropathologically. In previous studies employing this model, mice were fed cycad for the entire duration of the experiment. However, many human cases of ALS-PDC arise after a period in which little or no cycad is consumed. We therefore hypothesize that feeding cycad to mice for a limited amount of time will lead to the persistence and even worsening of symptoms despite the cessation of cycad exposure. In contrast, we found that the mice may recover from significant behavioral deficits (Experiment 1). In Experiment 2, we sought to extend these findings to older animals that were fed cycad for a longer time. These mice displayed no symptoms through most of the experiment but developed a severe behavioral syndrome months after the cessation of cycad feeding. Both groups of mice had no significant loss of motor neurons in the lumbar spinal cord. These data suggest the possibility of a "window of opportunity" in which symptoms are detectable but neurons are yet rescuable. This highlights the importance of early detection and treatment of ALS. Table of Contents Abstract i i Abbreviations v Introduction 1 ALS-PDC , 1 Recognizing ALS-PDC: Historical perspectives 1 Clinical observations 2 Neuropathology 3 Etiology of ALS-PDC 6 Cycad 7 Cycad consumption and acute toxicity 8 The B M A A story 9 Sterol glucosides: Causal agents in the development of ALS-PDC? 10 A murine model of ALS-PDC 12 Behavior and neuropathology in a murine model of ALS-PDC 12 Validity and value of our model 13 Combining risk factors: Age, genetics, and environment 14 The present study 15 Time course 15 Effects of age 16 When cycad feeing is stopped 16 Methods 18 Cycad 18 Animals 19 Experiment 1 19 Experiment 2 19 Behavioral testing 20 Leg extension 21 Wire/grid hang 21 Gait length 22 Y-maze 22 Five-arm water maze 23 Animal sacrifice and tissue preparation 24 Motor neuron counts 24 Results 25 Experiment 1 25 Experiment 2 27 Discussion 30 Experiments 1 and 2 : 30 Comparing the genetic SOD-1 model of fALS 32 Effects of age 34 Time course 36 Motor neuron vulnerability and other neuronal populations 38 Parkinsonism, the nigrostriatal system, and extrapyramidal symptoms 38 Putative actions and properties of cycad toxins 40 IV Conclusions and future directions 41 References • 43 Figure captions 62 Figures 64 Figure 1: Genetics, environment, and age 64 Figure 2: Feeding schedule 65 Figure 3: Weights (Experiment 1) 66 Figure 4: Behavior (Experiment 1) 67 4a: Gait length 67 4b: Leg extension 67 4c: Wire hang 67 4d: Y-maze 67 Figure 5: Cresyl violet (MN counts Experiment 1) 68 Figure 6: Cresyl violet (MNs Experiment 1) 69 Figure 7: Weights (Experiment 2a) 70 Figure 8: Behavior (Experiment 2a) 71 8a: Leg extension 71 8b: Grid hang 71 8c: Water maze 71 Figure 9: Cresyl violet (MN counts Experiment 2) 72 Figure 10: Cresyl violet (MNs Experiment 2a) 73 Abbreviations A D Alzheimer's disease ALS amyotrophic lateral sclerosis ALS-PDC amyotrophic lateral sclerosis-parkinsonism dementia complex ApoE apolipoprotein E B M A A B-methylamino-L-alanine B O A A P-N-oxalylamino-L-alanine BSA bovine serum albumin BSSG P-sitosterol-P-D-glucoside CDK-2 cyclin-dependent kinase CNS central nervous system DA dopamine D A B diaminobenzidine D Aergic dopaminergic Erk-1 extracellular signal-regulated kinase 1 FALS familial amyotrophic lateral sclerosis GFAP glial fibrillary acidic protein GLT-1 glutamate transporter 1 HPLC high performance liquid chromatography Hsp 70 heat shock protein 70 LB Lewy body L D H lactate dehydrogenase M A M methyloxymethanol M A M A L methyloxymethanol aldehyde M D A malondialdehyde M N motor neuron M R M magnetic resonance microscopy mSOD mutant superoxide dismutase NFT neurofibrillary tangle NGS normal goat serum N M D A N-methyl-D-aspartate NNOS neuronal nitric oxide synthase 3-NT 3-nitrotyrosine PBS phosphate-buffered saline PD Parkinson's disease PDC parkinsonism-dementia complex PFA paraformaldehyde PKB-2 protein kinase B 2 PKC protein kinase C Rsk-1 ribosomal protein S6 kinase SG sterol glucoside SN substantia nigra SNpc substantia nigra pars compacta SOD-1 superoxide dismutase 1 TH tyrosine hydroxylase 1 Introduction ALS-PDC Amyotrophic lateral sclerosis-parkinsonism dementia complex (ALS-PDC) is a complex neurodegenerative disorder characterized by symptoms of motor neuron disease, parkinsonism, and dementia that closely resemble more classical forms of A L S , Parkinson's disease (PD), and Alzheimer's disease (AD). Occurring primarily on the island of Guam in the South Pacific, A L S -PDC has been epidemiologically linked to the consumption of cycad seeds, a traditional food of the indigenous people of Guam, the Chamorros. Recognizing ALS-PDC: Historical perspectives Guam was first recognized as having a remarkable concentration of neurodegenerative disorders in the early 1950s. Shortly after the Second World War, a US Navy pathologist serving on Guam presented the first formal report of a high incidence of A L S - later documented as 50 to 100 times as in the continental United States (Kurland and Mulder 1954) - among the Chamorro population (Zimmerman 1945). A similarly high incidence of PDC was soon identified (Kurland et al. 1961). Teams of neurologists, pathologists, and epidemiologists have since become interested in understanding Guamian neurodegeneration in the hopes that it would provide insight into understanding neurological disease worldwide. It was also imperative to understand ALS-PDC in its own right, given that between the early 1950's and 1980's, nearly 25% of adult deaths among the Chamorros were due to A L S and PDC (Kurland 1992). By 1980-89 the incidence had decreased to 7.5/100,000/year which is still significantly higher than that in the US (2.3/100,000/year) or Japan (0.6/100,000/year) but is less than that in earlier decades on 2 Guam, when it was 50-100 times that of the US (Okumura 2003). Clinical observations ALS is a fatal paralytic disease characterized by loss of lower and upper motor neurons (MNs) resulting in progressive muscle weakness and atrophy. From a clinical perspective, Guamian A L S and classical A L S are nearly identical disorders (Mukai et al. 1982; Garruto and Yanagihara 1991; Kurland and Mulder 1954; Rodgers-Johnson et al. 1986). As in classical A L S , the most common features at diagnosis are atrophy and muscle weakness. Hyperreflexia, fasiculations, and spasticity are also typically present. Disease onset is insidious; an increasing paralysis leads to death on average three to five years (classical ALS) or five to seven years (Guamian A L S ; Mourelatos et al. 1994) after diagnosis due to failure of the respiratory muscles. Age at onset likewise differs slightly between classical and Guamian A L S : 68 years for the former and 56 years for the latter (Okumura 2003). As noted, PDC is characterized by PD-like parkinsonism and a cognitive decline reminiscent of A D . Parkinsonian symptoms include tremor, rigidity, gait disturbances, and bradykinesia. PDC is also often associated with a disturbance of speech and gait apraxia (Lilenfield et al. 1994). Fine motor movements are impaired and there is marked facial masking with reptilian stare and infrequent blinking (Elizan et al. 1966; Rodgers-Johnson et al. 1986). Cognitively, virtually all patients present with dementia, and this is often the dominant clinical feature of PDC (Hirano et al. 1961a). This dementia may occur prior to of after the appearance of parkinsonism (Elizan et al. 1966). It is characterized by memory deficits, disorientation with regard to time, place and person, and difficulty with reasoning. In addition, personality changes, including apathy, irritability, and aggression have been noted in many patients. Olfactory deficits 3 similar to those observed in A D and PD are present in most PDC patients (Doty et al. 1991). The clinical distinctions between A L S and PDC are not clear given that patients often express features of both A L S and PDC and familial analysis often reveals a history of A L S for patients with PDC and of PDC for those with A L S (McGeer et al. 1997). Taken together, this strongly suggests that A L S and PDC on Guam are not distinct disease entities, and instead points towards a common etiology and pathology. Furthermore, while A L S , PD, and A D have traditionally been perceived as distinct neurological disorders arising from different etiologies and expressing with unique behavioral and neuropathological features, the co-occurrence of these diseases in ALS-PDC suggests that there are significant etiological and pathogenetic similarities. The clinical and neuropathological similarities between Guamian ALS-PDC and the classic forms of A L S , PD, and A D suggest that an understanding of the etiology and progression of Guamian neurodegeneration might shed light on neurodegenerative disease in general. Further support to this hypothesis is the increasing awareness that motor diseases such as ALS or PD may also have a significant cognitive deficit (Vaphiades et al. 2002; Aarsland et al. 2003; Peavy et al. 1992; Lopez et al. 1994), while patients with disorders such as A D , that are characterized primarily by cognitive impairments, may also have problems with motor function such as tremor (Yokoyama 2002; Golaz et al. 1992; Kobayashi et al. 2000). For example, gait abnormalities typical of PD have been shown to be a significant predictor of the risk of developing dementia in elderly persons (Verghese 2002). Neuropathology Just as Guamian A L S is similar to classical ALS from a clinical perspective, the neuropathology is comparable. There is a progressive loss of spinal and cortical MNs resulting in 4 muscle weakness and atrophy. The notable difference in Guamian A L S is an abundance of neurofibrilliary tangles (NFTs) composed of the microtubule-associated protein tau (Rodger-Johnson et al. 1986). Tau regulates the assembly and stability of microtubules in a manner dependent upon its level of phosphorylation. Abnormally high levels of hyperphosphorylated tau are associated with various neurological pathologies including A D . Further differences are present in other CNS regions of Guamian patients with A L S include a reduced uptake of 6-fluorodopa in the striatum, a sign of DAergic cell loss that is the hallmark pathology of PD (Snowetal. 1990). While these changes are not features of classical A L S , their presence could indicate preclinical, concomitant PDC that has not yet reached the level of clinical detection. This is in keeping with numerous studies of neurodegenerative disease demonstrating significant neuron loss occurs before clinical symptoms appear. For example, the symptoms of PD only become apparent when more than 50% of nigral dopamine neurons are lost (McGeer et al. 1988). The neuropathological changes in PDC are also similar to those observed in PD and A D , with a few exceptions. PD is characterized by a loss of DAergic neurons in the substantia nigra pars compacta (SNpc) and their terminals that project to the striatum. Similar changes have been documented in the nigrostriatal system in PDC. Not only is there a dramatic loss of DAergic cells in the SNpc (Hirano et al. 1961b), but Snow and colleagues (1990) have also shown there is a dramatically reduced uptake of 6-ffourodopa in the striatum of patients with PDC indicative of terminal loss. However, one of the hallmark pathologies of PD, intracellular inclusions known as Lewy bodies (LBs) that consist of aggregated synuclein protein, is present in far fewer PDC than PD patients (Forman et al 2002). With regard to neuropathological features of PDC dementia, they closely resemble those 5 of A D (Hyman et al. 1984). There is marked cortical atrophy and NFTs in numerous brain regions including the hippocampus, entorhinal cortex, basal forebrain, and the neocortex (Hirano et al. 1961b; Kurland 1992). The ultrastructure and immunohistochemical profile of the NFTs appears to be identical to that in A D (Kurland 1992), but there are differences in their pattern and distribution. In PDC, there is a strong predilection for tangles in cortical layer 3, whereas in A D they are mainly found in layer 5 (Hof et al. 1994). Additionally, NFTs are present to a greater degree subcortically in PDC than in A D (Hirano et al. 1961b). Of note is the fact that there are many cases in which Chamorros who died without any apparent neurological disease yet displayed significant numbers of NFTs at autopsy. As many as 95% of unsymptomatic Chamorros older than 60 have significantly high levels of NFTs (Chen et al. 1981). This could be interpreted as preclinical neuropathology or it may indicate that NFTs are merely a background feature unrelated to neurodegenerative disease in the Chamorro population. Similar to the overlap of symptoms in classic neurodegenerative disorders previously described, features of the diseases may also cross conventional boundaries. For example, NFTs, characteristic of A D , have been identified in some cases of PD and A L S (Hamilton and Bowser 2004; Braak and Braak 1990). Similarly, synuclein, the major component of LBs in PD, have been isolated from amyloid plaques in A D patients (Kruger et al. 2000). This has led some (e.g., Calne and Eisen 1989) to suggest that since ALS, PD, and A D are all characterized by the accumulation of cytoskeletal debris (NFTs, LBs, etc.) these neurodgenerative disorders may be best described as "cytoskeletal disorders." A common denominator at the level of neuropathology may thus link these three diseases. 6 Etiology of ALS-PDC During the initial investigations into the etiology of ALS-PDC, investigators had hoped that straightforward casual factors would be readily unearthed. For example, the Chamorro population was relatively homogenous in genetic background, facilitating genetic analysis and identification of genes implicated in the disorder (Plato et al. 2003). However, detailed genetic surveys and analysis failed to identify a genetic basis for the disease (Reed et al. 1975). Further evidence against a genetic etiology came from studies of a Chamorro population living on the nearby island of Saipan (80 miles north of Guam). They had virtually the same genetic background as Chamorros on Guam, yet there was no evidence of an increased incidence of neurodegenerative disease. This is strongly indicative of an environmental etiology. The apparent lack of a genetic basis of the disorder is in keeping with observations regarding the classical forms of A L S (Armani et al. 1987), PD (Tanner et al. 1999), and A D (Rocchi et al. 2003) in which less than 10% of cases show a familial pattern of inheritance. Investigators began screening hundreds of potential environmental factors, including the levels of various minerals and heavy metals in the soil and ground water (Yasui et al. 1997), native food products, industrial materials associated with military activity, and finally, radiation. These potential factors were discounted as being causal in the development of ALS-PDC. For example, one hypothesis that proposed ALS-PDC was triggered by nutritional deficiencies of calcium and magnesium, leading to secondary hyperparathyroidism that then facilitated the entry of calcium and toxic heavy metals into the brain (Yanagihara et al. 1984). However, this was later refuted by a study that showed that ALS-PDC patients have no indications of abnormalities in calcium metabolism, have normal parathyroid hormone levels, and have levels of heavy metals in blood and urine samples that are statistically similar compared to that of controls 7 (Ahlskog et al. 1995). In contrast, cycad consumption appeared to be etiologically relevant to ALS-PDC. (Kurland et al. 1992). The incidence of ALS-PDC has mirrored changes in cycad consumption among the Chamorros. A spike of increased ALS-PDC incidence occurred after the Japanese occupation during World War II when, due to food shortages, the Chamorros increased their consumption of cycad. Since then, as the Chamorros adopted a more Westernized diet that does not include cycad, the incidence of ALS-PDC has decreased dramatically (Haddock and Chen 2003; Zhang et al. 1990). Furthermore, the genetically similar Chamorros living on the nearby island of Saipan, among whom ALS-PDC does not occur, rarely consume cycad. Cycad As noted above, ALS-PDC has been linked epidemiologically to the consumption of cycad, although the causal toxin(s) has yet to be conclusively determined. A variety of species of this palm-like tree are found throughout the world. For as many as several thousand years, cycad seeds, stem, roots, and leaves have been exploited as a traditional source of dietary starch by various indigenous populations around the world. Since seeds are a more renewable source than stems, these are the most commonly ingested part. Traditional processing renders the seeds into flour, which is then made into such products as "fadang" tortillas. Cycad contains toxic bioactive compounds, such as the azoxyglycosides cycasin and macrozamins, and the non-protein amino acid, B-methylamino-L-alanine (BMAA) that must be removed before ingestion. Traditional processing techniques, including washing and heating, have therefore been developed to remove or destroy these substances. While this removes toxins that are water-soluble and/or altered by heat, water-insoluble thermotolerant toxins remain. Since the former are associated with more acute illness, it was assumed that, because washing and cooking cycad prevents such acute toxicity, it rendered cycad safe for ingestion. That is, water-insoluble and heat-resistent toxins whose clinical effects are riot seen until many years, or even decades, after exposure would not have been easily linked to cycad consumption. It is therefore not surprising that although traditionally-prepared cycad has had many of its toxins removed or destroyed, the ingestion of processed cycad products is still associated with the development of Guam's unique neurodegenerative disorder, ALS-PDC (Kurland 1988). Cycas micronesica, a cycad species endemic to Micronesia, the Marianas Group, and the Western Caroline Islands, including the island of Guam, has been exploited by the indigenous Chamorro population of Guam as a food source (Audhali and Stevenson 2003). In recent times, cycad is more often resorted to as a "famine food" than a dietary staple (Beardsley 1964). For example, the occupation of Guam by the Japanese during World War II forced the people of Guam to rely on cycad due to food shortages. Since then, the adoption of a more Westernized diet has caused the consumption of cycad as a food source among indigenous populations to decreased substantially. Cycad ConsumpUon and Acute Toxicity Unprocessed cycad contains appreciable amounts of acutely toxic compounds. Although indigenous populations, such as the Chamorros of Guam, had tamed cycad's toxicity by developing extensive processing methods, early European travelers had no such understanding. When Europeans began exploration of areas to which cycads were native they naively consumed unprocessed cycad, with deleterious consequences. Norstog and Nicholls (1997) describe how, for example, members of Captain James Cook's crew ate cycad seeds and became violently i l l 9 with vomiting and vertigo. During periods of wartime scarcity (e.g., the Boer War, American Civil War, and World War II) soldiers ate cycad seeds with similar results (Norstog and Nicholls 1997). Such acute cycad toxicity is caused by azoxyglycosides present in the seed, leaf, and stem of cycad plants (Norstog and Nicholls 1997). These bioactive compounds can account for as much as 5 % of the plant's dry weight, and are composed of a sugar moiety (cycasin: glucose; macrozamins: glucose and xylose) that is joined via a P-glucoside linkage to methylaoxymethanol (MAM). This aliphatic azoxy side chain is highly toxic, but the sugar-M A M complex is itself innocuous. This allows for the safe storage of large amounts of azoxyglycosides in cycad tissues. Toxicity arises when these azoxyglycosides are cleaved by the relatively common enzyme, P-glucosidase. The freed M A M is itself toxic, but can also be converted to an aldehyde form ( M A M A L ) by the enzyme alcohol dehydrogenase (an enzyme found in the liver); M A M A L then decomposes, yielding destructive methylcarbonium ions. The BMAA story In addition to the presence of acutely toxic compounds, cycad also contains toxins that are instead associated with the development of more chronic, long-term illness. B M A A (p-methylamino-L-alanine) is one such cycad toxin. A non-protein but analogous amino acid, it is found in all cycad genera (highest concentrations are in the genus Cycas; Audhali and Stevenson 2003). B M A A is structurally very similar to glutamate (Audhali and Stevenson 2003). Agonism at glutamatergic N-methyl-D-aspartate (NMDA) receptors may cause extreme ionic flux across the cell membrane, resulting in cell death by excitotoxicity. However, this has only been shown in cell culture and not in vivo. 10 This profile of B M A A , coupled with negative findings regarding genetic, viral, and environmental (mineral deficiencies, cycad's azoxyglycosides, etc.) causes of ALS-PDC, led some researchers to propose that B M A A was the major causative agent in the development of ALS-PDC. Early work by Spencer and associates (1987) included exposing macaques to synthetic B M A A by gavage. These primates developed a transient motor disorder that did not persist without continual exposure to the toxin. Neuropathologically, there was no evidence of cell death but the morphology of cells was abnormal. This profile did not resemble human A L S -PDC. Furthermore, the amount of B M A A to which the primates had been exposed was not comparable to human consumption of cycad. The traditional washing of cycad before consumption removes, on average, 87% of the total B M A A content (Duncan et al. 1990). In fact, 50% of samples tested by Duncan and colleagues had virtually all - greater than 99% - of traces of B M A A removed, and others have reported B M A A levels in processed cycad flour to be in the range of 0.00 to 18.39 micrograms per gram of flour (Kisby et al. 1992). The experiments of Spencer and colleagues (1987) were therefore not comparable to the human situation. Furthermore, their use of much higher doses of B M A A than any human would ever encounter led not to the chronic formation of NFTs (as is found in human ALS-PDC cases) but instead to sub-acute chemical encephalitis (Spencer et al. 1987). Spencer and colleagues retracted their paper in 1990. The " B M A A theory" has since been superceded by other cycad-based etiological explanations of ALS-PDC. Sterol glucosides: Causal agents in the development of ALS-PDC? Since cycad processing effectively removes all water-soluble and/or heat-resistant compounds (i.e., cycasin, macrozamins, B M A A ) , the causal toxin(s) must be insoluble in water 11 and able to survive cooking temperatures. They must also be sufficiently lipophilic or able to recruit transport mechanisms in order to cross the blood brain barrier. Recent work in our laboratory has identified a group of toxins that fit this profile: sterol glucosides (SGs). Preliminary experiments using cortical wedge preparation and assays for lactate dehydrogenase (LDH) activity in cortical slices revealed biological activity and cell death, respectively. When the most active fractions were analyzed, it was found that these contained significant amounts of three SGs: P-sitosterol- P-D-glucoside (BSSG), campestrol- or dihydrobrassicasterol- P-D-glucoside, and stigmasterol- P-D-glucoside (Khabazian et al. 2002). Khabazian et al. (2002) went on to show that both isolated and synthesized SGs provoke similar cellular reactions. In a cortical wedge preparation, SGs give depolarizing responses that can be selectively blocked by the N M D A receptor antagonist D-AP5 and the noncompetitive antagonist MK-801. Rapid cell death is evidenced by L D H release. Although SG exposure leads to a significant release of glutamate, it does not compete with either glutamate or N M D A in competition binding assays. Exposure to N M D A or SG fractions changes the expression of some (e.g., CDK-2, PKC-beta) but not all (e.g., Erk-1, Rsk-1, Cot, PKB-2) protein kinases. Taken together, these results suggest that cycad's SG content is sufficiently neurotoxic so as to be considered a potential causal agent in the development of ALS-PDC. Interestingly, it is only in their glucosidated forms that these plant sterols are toxic (Khabazian et al. 2002). Aglycone sterols, such as P-sitosterol, may in fact have (neuro)protective effects (Bi et al. 2000). SGs are present in a number of plants besides cycad, some of which are also exploited nutrionally, medicinally, or recreationally. Rice bran, for example, contains the same three SGs as cycad (Fujino and Ohnishi 1979), as also does cocoa butter (cocoa beans Lome Tongo; Staphylakis and Gegiou 1985), and Iceburg lettuce (Lactuca sativa; Knapp et al 1968). Used in traditional herbal medicine, Juniper (Juniperus macropoda) and Chinese boxthorn (Lycium chinense) are among the plants that contain the highest amounts of BSSG (Duke 2003). Tobacco leaf and smoke likewise contains these compounds (Kosak and Swinehart 1960; Wright et al. 1962; Kallianos et al. 1963). A murine model of ALS-PDC Given our work showing the neurotoxicity of processed cycad seeds, along with the correlation between cycad consumption and ALS-PDC, we performed the critical in vivo experiments that would definitively link cycad consumption to ALS-PDC: feeding mice cycad processed in the traditional method. This would supersede experiments in which animals were exposed to water-soluble cycad toxins (e.g., azoxyglycosides; Sanger et al. 1972; B M A A ; Spencer 1987). Since these experiments neglected to consider the traditional preparation methods, our novel approach was designed to produce an accurate animal model human A L S -PDC. Adult male CD-I mice were fed washed cycad, with cycad constituting approximately 1/4 of their daily intake by weight. Control mice were fed pellets identical in weight and similar in nutritional content made of commercial grade processed white flour. To assess neurological effects of washed cycad consumption, we employed a battery of behavioral assays to monitor changes in motor, cognitive, and olfactory function. After animal sacrifice, the neuropathological consequences of cycad consumption were investigated using a variety of histological indices. Behavior and neuropathology in a murine model of ALS-PDC The consumption of washed cycad leads to both behavioral and neuropathological 13 outcomes in mice, which in many respects mirror features of ALS-PDC (Wilson et al. 2002). Behavioral analysis reveals a marked, progressive decline in both motor and cognitive function. With regards to the former, cycad-fed mice are impaired on the leg extension reflex and wire hang, indicating compromised motor neuron integrity. Reduced gait length, a feature of PD, is also present. Cognitively, the animals are impaired in their ability to perform the Morris Water Maze (spatial memory) and the radial arm maze (working and reference memory), suggesting hippocampal dysfunction. Histological analysis reveals neurodegeneration consistent with the observed behavioral deficits, as well as with regions of neurodegeneration found in ALS-PDC (Wilson et al. 2002). The ALS-like motor deficits are associated with a decrease the number of motor neurons (MNs) found in the lumbar spinal cord (Wilson et al. 2002) and the key neuropathological features of PD, decreased TH immunoreactivity in the striatum and DAergic cell loss in the SN, are also present (Schulz, personal communication). Cell loss in the cortex and hippocampus are assumed to underlie the cognitive deficits (Wilson et al. 2002). A downregulation of glutamate transporters may relate to excitotoxicity (Wilson et al. 2003) and astrogliosis, as measured by increased glial fibrillary acidic protein (GFAP) immunoreactivity, and activated microglia are also found. Validity and value of our model Standard criteria that any model must meet are etiological, construct, and predictive validity. Etiological validity is met, since our model mimics the human situation quite closely: daily consumption of processed cycad constituting a proportion of total dietary intake provokes the gradual emergence of the murine equivalent of ALS-PDC. That our model links specific 14 behavioural deficits with underlying neuronal dysfunction and death indicates that construct validity is also satisfied. With regards to predictive validity, we hope to construct a time course of the behavioral and neuropathological features so that we may predict outcomes by examining early markers or predict neuropathology by determining behavior and vice versa. Furthermore, to move from mouse model to the human experience of neurodegenerative disease we must transpose our data (Shaw et al. 2002; Shaw and Wilson 2003), which depends critically on the validity of the model. The time course construct adds significantly to the value of our model. Many animal models of neurodegenerative disease focus - at times exclusively - on the rapid attainment of end-stage. The slow emergence of disease in our model allows us to examine pre-clinical, symptomatic, and end-stage phases of ALS-PDC. We will also be able to distinguish between the disease's causal, co-incidental, and compensatory (successful or failed) components in this time course. This will facilitate effective intervention, moving from the largely palliative care suggested by most models to actually halting or even preventing neurological damage. Combining risk factors: age, genetics, and environment While cycad consumption appears to be the critical causal factor in the development of ALS-PDC, genotype may also contribute to susceptibility. Furthermore, ALS-PDC is an age-dependent disease. The interaction of these factors can be represented in a Venn diagram (Figure 1) such that the probability of disease increases with increased environmental toxin exposure, heightened genetic abnormality (or, alternatively, lessened protective mechanisms), and advanced age (open areas). As these predisposing factors combine (shaded areas), risk increases, culminating when all three are present (darkened area). 15 [FIGURE 1 N E A R HERE] Current work in our laboratory is examining the interaction of cycad toxicity (an environmental risk factor) with genetic factors (e.g., ApoE gentotype; Wilson et al. 2005). The present study begins to address the issue of age. The effect of gender is also under investigation (Wong, personal communication). This work extends our model of ALS-PDC, and furthers its etiological and predictive validity. The present study Time course As previously noted, disease progression in this cycad model of ALS-PDC is protracted, occurring over a course of weeks to months. This extended period of illness allows us to examine more closely and accurately the time course of the behavioral and neuropathological features. However, previous experiments in our laboratory have examined only end-stage mice. In the present experiments, we will attempt to construct a time line of the emergence, deterioration of function, and end-stage of the disorder by sacrificing animals at various time points. In the first of the experiments outlined below, we sacrifice mice at three time points: when symptoms begin to emerge, when animals are significantly impaired by their symptoms, and at a later time point. The second experiment was designed to have animals sacrificed at time points throughout the experiment. We predict that we will find both behavioral and neuropathological progression during disease development corresponding to cell vulnerability, dysfunction and compensation, 16 and, finally, death. We will begin this analysis by relating M N loss to behavioral deficits on motor tasks and will in future studies characterize the progression through cellular dysfunction to death. Effects of age Since ALS-PDC is an age-dependent disease, we begin to explore this issue in the experiments outlined below. Previous work with our cycad model has been done only with (three months old) mice, neglecting the crucial factor of age. In our first experiment, we replicate previous findings in three month-old mice and then use older (six months old) and younger (three weeks old, i.e., weanlings) mice in the second experiment. We predict that younger animals will have better resistance, compensatory mechanisms, and repair processes compared to older animals and will consequently be less susceptible to developing ALS-PDC, similar to the human situation where younger age is associated with better prognosis (Mukai et al. 1982). However, there may be a more U-shaped relationship, with both very young and very old animals being more prone to developing the disease due to immature (younger animals) or depleted (older animals) compensatory mechanisms. In addition to this increased susceptibility in younger and older, but not middle-aged, mice, we expect that the more vulnerable animals will also develop more severe and/or accelerated neurodegeneration. Our initial analysis in the present experiment will consist of comparing behavioral deficits with M N loss across the ages studied. When cycad feeding is stopped... A third feature of the cycad model of ALS-PDC that we explore is whether the disease 17 will continue to progress even if cycad feeding is ceased. After obtaining a time line of disease emergence and subsequent deterioration for each group of mice, we plan to cease cycad feeding but continue to monitor the animals behaviorally and to examine the neuropathology of their CNS at the beginning and end of this cycad-free period. This contrasts with previous work in our laboratory where mice were fed cycad more or less continually until end-stage. There are three possible outcomes to this: a) the mice will experience a worsening of symptoms, continuing to progress towards end-stage disease; b) symptoms will remain constant at the level achieved at the end of cycad feeding; or c) there will be some sort of recovery of function. Outcome will also depend on variables of dose and age. Previous studies in which mice were fed cycad continually until end-stage mimics the human experience for many ALS-PDC patients, who consumed cycad more or less continually throughout their lives. However, there are also cases in which patients consumed cycad during the early part of their life but for whom exposure subsided after moving outside of Guam or after the adoption of a more Westernized diet in Guam (Garruto et al. 1980). We therefore hypothesize that cycad exposure will induce neurological disease that sustains its progression even in the absence of further toxin exposure. We hope to show that disease progression, as defined by both behavioral deficits and M N loss, continues beyond cycad exposure. In Experiment 1, younger mice (three months) were used, and in Experiment 2 we sought to extend these findings to older (six months) that were exposed to cycad for a longer period. 18 Methods Cycad Raw cycad seeds were received fresh from Guam and processed according to the traditional method of the Chamorros. Seeds were extracted from their husks, thinly sliced, and soaked in approximately 2 L distilled water per batch of 25 seeds. The water was changed daily for seven days. The cycad chips were then left to air-dry overnight and the dry seeds ground into flour using a coffee bean grinder. A minimal amount of distilled water was mixed with the flour (0.5 g cycad/pellet) to make pellets. Pellets were flavored with of one of several liquid extract flavorings (e.g., vanilla, lemon, or coconut extract) had been added (1-2 drops per mL water) to prevent taste aversion. Regular commercial white flour of comparable caloric content (0.5 g) was similarly used to make control pellets. For the first experiment, pellets were left to air-dry overnight, while for the second experiment they were prepared immediately before feeding. To assess the neurotoxic content of the processed cycad flour, high performance liquid chromatography (HPLC) analysis was performed. Different batches of cycad arriving from Guam varied in their sterol glucoside content, with p-sitosterol glucoside concentrations ranging from 0.085 mg/g washed cycad to 0.412 mg/g. This variation may be due in part to differences in seed maturity, region of Guam from which they were harvested, season, amount of rainfall, and delay between sending from Guam and arrival in Canada. 19 Animals Experiment 1 32 CD-I male mice (3 months old at the beginning of the experiment) were used as subjects. Mice were housed individually and kept on a 12-hour light/dark cycle. Each morning, chow was removed from the cage and one pellet (cycad or control) introduced. Remaining portions of the pellets were weighed at the end of the day (approximately six hours after pellet was introduced) to determine how much cycad flour had been consumed. Chow was given ad libitum overnight. Overall, mice consumed approximately 95% of cycad introduced and 100% of control flour pellets. Mice were fed pellets for 100 days and sacrifices occurred on days 75 (n=8), 100 (n=T2) and 175 (i.e., 75 days after the end of feeding; n=10). Two mice were put down due to mite infestation. The behavioral data from these animals is included in analysis but excluded from neuropathological analysis. Experiment 2 40 CD-I male mice (six months old at the beginning of the experiment) and 39 CD-I male mice (three weeks old/weanlings) were used as subjects for the age-dependence study (Experiment 2a). For the time course study (Experiment 2b), 40 more of the six month-old male CD-I mice were used. Housing and feeding was the same as in Experiment 1, except that all mice were fed (i.e., fed cycad or control pellets) for 60 days, not fed for 30 days, fed for 90 days, and then not fed for 180 days. Due to inconsistent cycad availability and the fact that the mice did not consistently eat the cycad pellets, it is estimated that they consumed approximately 75% 20 of the total cycad offered. Ten mice (two from the first six-month-old group and four from each of the other groups) were put down due to mite infestation. The behavioral data from these animals is included in behavioral analysis but excluded from neuropathological analysis. While the six month old animals used in Experiment 2a are used as "older" animals, this may be misleading. These mice are indeed older than the weanlings or even the three month old animals used in Experiment 1, but at six months of age there are not considered to be "old." CD-I mice are instead considered "old" at 18 or even 24 months of age; six month olds are in fact simply "adult" animals. These younger animals were used as subjects because we hoped to examine disease progression from early, presymptomatic phases onward, phases which could well begin in adulthood. It would be valuable to repeat our studies using "old" CD-I mice. Behavioral testing Mice were subjected to a battery of behavioral tests weekly. These included several motor tasks (leg extension, wire or grid hang, and gait length) and one cognitive one (Y-maze or modified Morris Water Maze/radial arm maze). Testing was always performed in the morning (between 9:00am and 12:00 pm) and all mice were run through all of the tasks within this time in a standard order (leg extension, wire/grid hang, maze, gait length). Baseline was obtained prior to the commencement of cycad/control feeding. In the second experiment, there were three breaks (approximately day 50 to 105, day 250 to 295, day 310 to 340) during which no testing was performed due to the absence of the experimenter. 21 I. Leg extension. The leg extension reflex was used as a measure of alpha M N integrity (Barneoud and Curet 1999). Mice are held upside down by their tails and the position of their hind limbs immediately noted. They are scored on a non-continuous five-point scale: 0 - both (hind) limbs retracted; 1 - one limb extended, tremors or punching of this limb; 2 - one limb extended, no tremors or punching; 3 - both limbs extended, tremors or punching of one or both limbs; 4 -complete extension of both limbs/no retraction. That is, a score of 0 indicates a serious dysfunction of motor neurons, and a score of 4 indicates normal M N function. This differs slightly from Barneoud and Curet's three-point scale. This scale was adapted to increase its sensitivity. Baseline was obtained by performing the task five times over five consecutive days. The test was performed oriented in the testing room in the same way for all trials. II. Wire/grid hang. The wire/grid hang task also assesses M N integrity, as measured by muscle strength (Sango et al. 1996). Mice are either subtended by their forepaws on a string 30 cm above base (wire hang) or placed on a wire grid (8mm x 8mm squares) that is subsequently inverted 30 cm above base (grid hang). Mice grasp the grid with all four paws and the latency to fall is recorded for 3 trials (maximum per trial = 60 sec) and results averaged. The wire hang version was used for the first group of animals (Experiment 1) but the grid hang was deemed to be a better measure of muscle strength as the animals could not raise themselves on top of the wire/grid and balance instead of hanging. The grid hang was therefore used in Experiment 2 starting day 120 (wire hang used day 1-119). The apparatus is oriented in the testing room in the same way for all trials. 22 III. Gait length. Gait length, among other gait-related features, is related to severity of A L S (Goldfarb and Simon 1984) and the de Medinaceli pawprint test (de Medinaceli et al. 1982) is used to measure similar gait changes in mice. Mice are held by the tail, and a non-toxic paint applied to their hind paws. They are then trained to walk the length of a partially-enclosed (starting end open, finishing end enclosed) narrow (10 cm) runway (50 cm in length). The distance between consecutive ipsilateral pawprints (stride length) of both the right and left paws is measured using the middle toe tip as the reference point. One to three runs are performed per mouse for each session and between five and eight stride length measures taken depending on the quality of the prints. A run is unusable i f the mouse stops and/or turns around instead of progressing to the end of the runway. The runway is oriented in the testing room in the same way for all trials. IV. Y-maze. In the first experiment, a Y-maze was used to measure cognitive function. The three arms of the maze are of equal length and open to a central circular area. At the peripheral end (i.e., not the end of the arm that opens to the central circular area) of one of arm is a reward: a mouse house, covered to make it dark. This was used instead of the traditional reward pellets because the required food deprivation would interfere with the cycad/control feeding paradigm. At the peripheral end of the other arms, there is only the maze wall. Mice are placed at the peripheral end of one of these unrewarded arms, the starting arm (always the same arm). During baseline training as well as during testing, mice were placed in the starting arm and allowed to explore the maze until they found and entered the mouse house at the peripheral end of the reward arm. 23 Latency to enter the house (all four paws inside house) was recorded. The mice were left in the house for one minute and then (or when they came out of the house on their own, whichever came first) removed. The reward house is always in the same arm, and the entire maze oriented in the testing room in the same way for all trials. V . Five-arm water maze. For the second experiment, a different cognitive task was used. This task and its scoring reflected the need for a measure that is not dependent on latency, as ALS-PDC mice may display both cognitive and motor deficits. The maze was created by modifying and combining the traditional Morris Water Maze (Morris et al. 1984) and the radial arm maze (Olton and Samuelson 1976). A maze with four identical arms and one longer arm (starting arm) is placed in a large tub filled with approximately five centimeters of water. At the end of one of the four arms is a small stage onto which the mice can climb out of the water. The stage is always in the same arm (the arm at the extreme left of the longer starting arm), and the entire maze oriented in the testing room in the same way for all trials. Mice are placed at the peripheral end of the longer arm at the beginning of each run. During training, mice are allowed to explore freely until they find the stage. For testing, mice are placed at the peripheral end of the starting arm and must swim to the platform at the end of one of the arms. Mice that enter the correct arm are scored as successful and those that first enter another arm are scored as failing the task. "Entry" is defined by the placement of all four paws into the arm. When the mouse enters any of the non-staged arms, the arm is noted and the mouse is left to explore until it finds the stage. Three trials are performed each time and the results averaged across the trials. A score of 3 refers to the mouse 24 entering the reward arm on all three trials while a score of 0 refers to the mouse entering one of the non-reward arms first on all three trials. Animal sacrifice and tissue preparation Mice were deeply anesthetized by halothane inhalation in an isolation chamber. They were then perfused transcardially with approximately 20 mL ice-cold 0.1 M PBS, followed by approximately 3 0 - 4 0 mL ice-cold 4% paraformaldehyde (PFA). CNS tissues (brain and spinal cord) were extracted from the scull and vertebral column, post-fixed in PFA overnight, cryoprotected in 30% sucrose overnight, and flash-frozen in isopentane. Before flash-freezing, thoracic, lumbar, and sacral sections of the spinal cords were cut out and the intermediate portions discarded. These cuts were visually estimated using the thoracic and lumbar enlargements as reference points. Only lumbar sections were used in this study. 30 um coronal sections of lumbar spinal cord were cut on a cryostat and collected free-float in Millonig's storage buffer. Sections were collected in series of 12 wells such that each well contained a representative series of sections. 10 sections were collected in each well. Motor neuron counts The sections (10) from one well of lumbar spinal cord were mounted on Permafrost slides. On each slide, 10 sections from a control animal and 10 sections from a cycad-fed animal were mounted to ensure consistent staining across the groups. Mounted sections were left to ari-dry on the slides. These sections were then stained with cresyl violet to identify motor neurons. Briefly, slides were immersed in distilled water for two minutes and transferred to a 0.5 % cresyl violet solution for 25 minutes. Slides were then rinsed in distilled water for 30 seconds and 25 immersed for two minutes in each of 70%, 90%, and 100% ethanol, followed by immersion in xylenes. Slides were coverslipped using Entellan and left to air-dry. The occasional section fell off during the immersions but this never exceeded two sections per 10 mounted for each animal. This was assumed to be equally distributed across control and experimental groups and was not corrected for in analysis. For M N counts, digital images of the ventral horn of the spinal cord sections were captured using a Zeiss Axiovert microscope at lOOx magnification. A l l images for a given set were taken in one session to ensure light levels remained constant. Each image was opened in Adobe Photoshop 7.0 and a simple M N count was performed. A M N was identified by its size (large), morphology (see Figures 6 and 10), and staining (positive). The number of MNs was counted for both sides of the spinal cord and results averaged, resulting in 16 to 20 values per animal. Values for each M N count were used to calculate mean ± S.E.M. for each group. For Experiment 1, means were compared using one-way A N O V A while for Experiment 2a, means were compared using an unpaired, 2-tailed t-test (GraphPad Prism, San Diego, CA). p < 0.05 was used for significance. Means were normalized such that controls = 100%. Results Experiment 1 Over the course of the experiment, as the animals matured, weights increased. This was similar for both control and cycad-fed mice (Figure 3). Mice appeared healthy throughout the 26 experiment with respect to feeding, drinking, activity, and grooming, although these were neither specifically assessed nor quantified. [FIGURE 3 N E A R HERE] In cycad-fed mice, decreases in gait length (-26%; t = 5.529, df = 317, p < 0.0001; Figure 4a) emerged around day 50, leg extension deficits (-27%; t = 2.994, df = 4, p = 0.0402) by day 100 (Figure 4b), followed by difficulty with the wire hang task (-30% by day 100; t = 3.589, df = 18, p = 0.0021; Figure 4c). Deficits on the wire hang task remained for the course of the experiment while leg extension showed substantial recovery of function: cycad-fed mice returned to baseline (-0%; t = 0.0000, df - 8, p = 1.0000) around day 115 and were indistinguishable from controls for the remainder of the experiment. Differences between groups on the gait length task remained for most of the experiment, persisting even as both groups underwent increases in gait length due to the growth of the animals and a tendency to walk the length of the runway faster with more experience. There were no consistent significant differences between groups on the Y-maze test. [FIGURE 4 N E A R HERE] M N counts revealed no significant differences between cycad and control groups at any of the three time points: Day 75: t = 1.516, df = 4, p = 0.2040; Day 100: t = 0.3883, df = 7, p = 0.7093; Day 175: t = 0.4073, df = 6, p = 0.6979 (Figure 5), although there is a nonsignificant 27 trend (p > 0.05) towards an increase in M N number on day 75 in cycad-fed mice as compared to controls and as compared to all groups at the other time points. [FIGURE 5 N E A R HERE] Representative photomicrographs show cresyl violet staining of the lumber spinal cord (Figure 6). [FIGURE 6 N E A R HERE] Experiment 2 Over the course of the experiment, as the animals matured, weights increased. This was similar for both control and cycad-fed mice (Figure 7). Mice appeared healthy throughout the experiment with respect to feeding, drinking, activity, and grooming, although these were neither specifically assessed nor quantified. Analysis of the liver of revealed that 10% of control mice and 20% of cycad-fed mice had some liver pathology. On gross examination, affected lobes of livers were blackish brown in color and friable in consistency. On microscopic examination, lobular structure was deranged and no specific pigment deposition was found. Various areas of necrosis and hemorrhages were visible. The pattern of pathology was similar in both control and cycad-fed mice. The fact that similar pathology was present in control and cycad-fed mice and that colony mice are known for their tumor growth as they age suggests that in this experiment the pathology noted may not be a function of a carcinogen present in cycad. Liver pathology was not significantly associated with cycad feeding in mice (p > 0.05). 28 [FIGURE 7 N E A R HERE] With regards to behavior, there were no consistent, significant differences between controls and cycad-fed mice on the leg extension, grid hang, gait length, or water maze tests during most of the experiment for both age-dependent (Experiment 2a) and the time course (Experiment 2b) groups. Combined with variable cycad availability (due to typhoons in Guam) jand the fact that the mice were not consuming their cycad pellets regularly, this absence of behavioral changes led to the discontinuation of these projects; data is not shown. Among the older (six month-olds) group of mice in Experiment 2a, performance on the leg extension task was near perfect for both groups throughout the tasks, including through breaks in testing (Figure 8a). As a group, the cycad-fed mice generally showed greater variance but with an average indistinguishable from controls (F(14, 75) = 0.3130, p = 0.9907). There were no significant differences between groups on the gait length task (p > 0.05; data not shown). There was a nonsignificant trend (p > 0.05) for the cycad group to perform better on the grid hang task (Figure 8b). This was apparent even during the initial stages and therefore probably reflects group selection. Similarly, there are nonsignificant trends (p > 0.05) for the controls outperform the cycad-fed mice in the water maze task (Figure 8c). These trends generally appeared after a break from testing and likely reflect a slight impairment of the cycad-fed mice in long-term memory. The controls also performed more poorly after such breaks, but not to the same extent as the cycad-fed group. [FIGURE 8 N E A R HERE] 29 Late in the experiment, approximately 6 months after cycad/control feeding had ceased, several cycad-fed animals began to show severe neurological symptoms. The typical profile at this point was an inability to perform both the leg extension task (limbs remained fixed in the position they were in before the mouse was lifted up by the tail) and wire hang test. They were akinetic, and as such would not perform the gait length or water maze tasks. They displayed an intermittent tremor, with a head tremor that persisted longer than the whole-body tremor. Their limbs splayed out, unable to support the body due to paralysis. By this point, animals were not grooming adequately. Seven of the 18 experimental animals (39%) developed this syndrome and were sacrificed, along with matched controls, within four days of symptom onset due to severity of illness. The other 11 animals were not sacrificed at this time as it was expected that they would also develop this syndrome. However, they did not and were sacrificed at a later time. Tissues from these animals were not analyzed. There was no significant loss (t = 1.320, df = 7, p - 0.2284) of ventral horn MNs (Figure 9). There was, however, a non-significant trend (p > 0.05) for a loss of these neurons in these cycad-fed mice. [FIGURE 9 N E A R HERE] Representative photomicrographs show cresyl violet staining of the lumber spinal cord (Figure 10). 30 [FIGURE 10 N E A R HERE] Discussion Experiments 1& 2 The present experiments demonstrate that considerable motor symptoms may be present in the absence of M N loss in mice fed cycad. Based on previous studies in our laboratory, we had expected to find that the behavioral and neuropathological correlates of disease progression would worsen with time, even after cycad exposure had ceased. Instead, in Experiment 1, we found that there is a significant recovery of function. We therefore hypothesized that neuronal stress (e.g., oxidative stress) could be sufficient to cause performance on behavioral tasks to worsen. Removal of the stressor (cycad) could allow the cells to return to relative health, evidenced by behavioral recovery. We were able to extend the results of Experiment 1 to an older group of mice that had been exposed to cycad for a longer period of time (Experiment 2a). However, given that the emergence of behavioral deficits was sudden, severe, and preceded death by only a matter of days, subsequent animals were sacrificed within four days of symptom onset. This was approximately 180 days after cycad feeding had ceased. Since there were no detectible symptoms in mice at the cessation of feeding, none were sacrificed at that time. As a consequence, we only have one time point for the animals in Experiment 2a. This point can be compared to the third time point in Experiment 1 as both represent a state induced by previous, discontinued cycad exposure. 31 In Experiment 1, behavioral deficits emerged gradually. In contrast, in Experiment 2a, onset of symptoms was sudden. Despite this difference, the results of both experiments could be explained by either the actions of a slow-acting toxin or one that produces initial damage that itself leads to progressive deterioration. However, this toxicity had not yet caused a significant loss of MNs. Since liver pathology was not significantly associated with cycad feeding in Experiment 2a, it is unlikely that non-specific systemic changes were the cause of the observed behavioral dysfunction. It is possible that there is an initial stress response in the lumbar spinal cord to cycad exposure, accompanied by the recruitment of protective or compensatory measures such as the activation of astrocytes and the heat shock response. While this level of stress may be sufficient to impair performance at the behavioral level, this dysfunction may not overwhelm the system to the point of cell death. If examined at a later time point, it may be seen that the system moves beyond a state of stress and begins to undergo programmed cell, marked caspase 3 and other apoptosis-related immunoreactivity. Such stress and compensation were not examined in the present experiments but future studies could be conducted regarding these issues. The present data suggest that there is a certain "window of opportunity" in which behavioral deficits - analogous to clinical symptoms in human patients - are pronounced but M N loss has not yet begun to a significant degree. That is, we propose that M N dysfunction is sufficient to produce behavioral changes, and that these precede the cell death that characterizes end-state ALS both in humans and in cycad-fed mice (Wilson et al. 2002). If this were true, it would be a strong imperative to improve early detection and to focus on novel treatment regimes involving anti-oxidation and the boosting of compensatory mechanisms such as the heat shock response. 32 Comparing the genetic SOD-1 model of fALS The familial form of A L S (fALS) is observed in 5-10% of A L S patients and approximately 20%) of fALS patients have mutations in the gene coding for the antioxidant enzyme, superoxide dismutase (SOD) (Armani et al. 1987; Siddique and Deng 1996; Alexander et al. 2002; Robberecht 2000; Rosen 1993). Mutant SOD (mSOD) is expressed throughout life which is comparable to chronic exposure to cycad. Both murine models have onset of symptoms in adulthood despite this continued presence of mSOD or cycad toxins. The animal model most employed for the study of A L S in mice is the SOD-1 (Cu/Zn superoxide dismutase) transgenic. This model is characterized by rapid progression to death, as described by Chiu and colleagues (1995). The first sign of disease is tremor, which develops around day 90. Shortening of stride length appears by approximately day 125 and is the first sign of weakness. Severe paralysis and subsequent death occurs by approximately day 136. Postmortem analysis of end-stage spinal cord reveals significant loss of motor neurons and interneurons (Morrison et al. 1998). Neuropathologically, phosphorylated neurofilament inclusions are present (Morrison et al. 1998). Astrogliosis, as measured by increased GFAP immunoreactivity, appears by day 100 (Hall et al. 1998). This may correspond to an attempt to support oxidatively-stressed neurons; nNOS immunoreactivity is increased as of day 60 (Wengenack et al. 2004). Levels of 3-NT, indicative of protein nitrosylation, are increased in SOD-1 mice (Bruijn et al. 1997), as well as in human patients with either familial or sporadic ALS (Beal et al. 1997). Lipid peroxidation, as measured by M D A levels, is increased during days 30 to 100, but is decreased relative to wild-type controls by day 120 (Hall et al. 1998). 33 Caspase 3 expression is also increased in SOD-1 mice (Wengenack et al. 2004; Wootz et al. 2004). It would be reasonable to assume that a similar profile is present in cycad-fed animals. A striking similarity between the SOD-1 models and the findings of the present study is the lack of M N death in the ventral horn prior to symptom onset. Morrison and colleagues found that there is no reduction in the number of ChAT-immunopositive MNs in 30- or 80-day old SOD-1 mice (i.e., presymptomatic period) but numbers are decreased in 100-day old ones (i.e., period of symptom emergence, including decreased stride length; Chiu et al. 1995). Over the subsequent days to death, MNs are rapidly lost. By death, approximately 40% - 50% of MNs have degenerated in both the SOD-1 (Hamson et al. 2002) and cycad (Wilson et al. 2004) models. This pattern of no M N loss prior to the onset of symptoms holds across most of the various genetic models of A L S (i.e., different point mutations) and it is therefore valuable to demonstrate a similar pattern in an environmental model. This strongly suggests that the disease progresses similarly in response to very different etiologies. It appears that both the SOD-1 mutation and cycad induce some form of cell dysfunction that compromises the animals behaviorally. Using magnetic resonance microscopy (MRM), Wilson and colleagues (2004) have shown that there is a significant decrease (-21%) in total grey matter volume of the lumbar spinal cord in cycad-fed mice compared to controls when cycad feeding occurs continuously to (behavioral) end-state. This is similar to the 15% decrease found in mSOD mice as compared to wild-type G93 A (Hamson et al. 2002). Similarly, both mouse models showed decreases in ventral horn volume (cycad-fed: -20%; mSOD: -10%). This also suggests that both genetic mutation and environmental toxin exposure trigger similar processes. These similarities suggest that a single pathogenic pathway may be activated by either genetic mutation or exposure to environmental insult, resulting in ALS/ALS-PDC. If this is true, analysis of either model will provide possible therapeutic interventions. In contrast, i f different neurodegenerative pathways are activated by genetic versus environmental factors, then the situation becomes more complex and requires different approaches depending on the type of neurodegenerative pathway. The comparison of behavioural, morphological, and histological features of these different models is a first step in resolving this issue. Further comparisons between familial and sporadic A L S patients will help clarify this issue. Changes in the nigrostriatal system have also been documented in SOD-1 mice (Kostic et al. 1997). Decreased levels of D A occur during M N loss, and T H immunoreactivity is reduced by as much as 26%. Likewise, in the cycad model, the nigrostriatal DAergic system is affected and mice display signs of parkinsonism (Schulz, personal communication). Effects of age It has been estimated that the latency period for ALS-PDC is at least 18 years and may be as long as three decades (Garruto et al. 1980). This was calculated using cases of ALS-PDC in which the patients had lived on Guam and consumed cycad during their childhood and adolescence but who later migrated to other countries, such as the United States. Mortality rates for ALS-PDC among these migrants were greater than that of the United States population, but lesser than the Chamorros in Guam (Garruto et al. 1980). In order to study this latency period in our model, we designed our second experiment (2a) in which a latency period is introduced to cycad-fed mice before the development of any overt behavioral symptoms. Based on the human data described above, we predicted that the murine equivalent of ALS-PDC would develop in the 35 absence of continued cycad exposure. The latency was about 6 months, corresponding to approximately 20% of a typical lifespan for a laboratory mouse (average lifespan = 30 months). Similarly, the 18 year latency in humans compromises approximately 24% (average lifespan = 75 years). The complementary experiment is to not feed cycad during the first part of the mice's lives but then to introduce it later on. Although we conceive of the mice in Experiment 2a above as having received early cycad exposure, the cycad exposure could alternatively be thought of as beginning mid-life (six months). Analogous cases have been reported in humans. One study examined the mortality rate from ALS-PDC in Filipino migrants to Guam (Garruto et al. 1981). As with the patients who migrated out of Guam, the mortality rate for the migrants was greater than that of the United States but lower than that of Guam's Chamorro population, and there was a latency period as long as 24 to 29 years for the disease to develop in the Filipino migrants. This length of stay is important, as American construction workers working in Guam for one year or so were no more likely to develop ALS than their American counterparts working in the US (Brody et al. 1978). Based on this human data, we predicted that mice exposed to cycad for a significant part of the middle part of their life would develop ALS-PDC after a significant latency period. This is indeed what was found in Experiment 2. That the cycad exposure in this group of mice (Experiment 2a) can be categorized as either early or mid-life exposure highlights the crucial variable of age. Compared to 18-month-old mice, six-month olds are younger; compared to weanlings, they are older. We had hoped to address this issue of age dependence by varying the age at the beginning of cycad feeding and examining vulnerability to ALS-PDC. Since the weanlings did not at any point show signs of disease but the older group eventually did suggests that the younger mice are less vulnerable to 36 developing ALS-PDC. This may be explained by longer latency period for younger mice; had they lived longer, they may have eventually shown signs of disease. We can at least discard the hypothesis that vulnerability is U-shaped; the younger mice were not more susceptible than older ones. Time course Since the mice in the original time course study (Experiment 2b) showed no signs of illness, we have from them no data on the time course of ALS-PDC in mice. However, the mice in Experiment 1 were sacrificed at three time points and as such provide preliminary data regarding disease progression. Behaviorally, signs of parkinsonism (i.e., decreased gait length by day 55) emerge before motor deficits reminiscent of A L S (i.e., deficits on the leg extension and wire hang tasks by days 60 and 100, respectively). This seems to contrast with the fact that, in parkinsonism, there is substantial cell loss in the SN (approximately 60% loss as indicated by the loss of enzyme dopa-decarboxylase; Leenders et al. 1990) and, in A D , a 75% loss of neurons in the nucleus basalis of Meynert (Whitehouse et al. 1982) before clinical symptoms are detectable, while the present experiments indicate that symptoms of A L S can present before significant loss of motor neurons. This could be explained by a faster rate of degeneration in the SN compared to the lumbar spinal cord. However, research in our laboratory has revealed that there are no significant decreases in TH immunoreactivity until day 175, despite significant deficits on the gait length task on days 75 and 100 (Schulz, personal communication). A confounding factor that may be responsible for this unexpected display of parkinsonian behavioral deficits in the absence of SN cell loss is that fact that in human A L S patients, gait is disturbed. Stride length, as measured by 37 the gait length test in mice, decreases with the progression of A L S (Goldfarb and Simon 1984). It is therefore possible that the gait length deficits detailed in Experiment 1 may have resulted from ALS-linked changes in gait. This time course of symptoms before cell loss is at odds with some animal model and human data. One SOD-1 murine model is characterized by a protracted disease course (273 days to end-stage) and displays a biphasic progression of cell loss (Feeney et al. 2001). There is an initial presymptomatic loss, followed by stabilization and then a gradual loss that coincides with symptom onset. Like other models, there is approximately 50% loss of motor neurons by end-stage. However, this pattern is reported only in one model and one study. It may be specific to the protracted model used, although not as a characteristic of a protracted disease course in general since the present study demonstrates symptom onset before motor neuron loss in the protracted cycad model. For human cases of A L S , it has been suggested that there is a long presymptomatic phase during which cells are lost but for which there is compensation and therefore no detectable motor deficits (Swash and Ingram 1988). Compensatory mechanisms (e.g., motor unit enlargement by sprouting) are able to maintain normal muscle twitch even when as many as 70-80% of motor units are lost (Gordon et al. 2004). In contrast, Aggarwal and Nicholson (2005) showed that in human patients with A L S , in 17 out of 19 cases patients are symptomatic but without detectable loss of motor neurons, as determined by motor unit number estimation. Since at no point did the cycad-fed animals display statistically significant cognitive deficits as measured by the Y-maze or 4-arm water maze while manifesting significant motor ones, this suggests that motor symptoms precede cognitive deficits. This corresponds to human cases of ALS-PDC, in which symptoms of ALS precede those of PDC (Steele and Guzman 38 1987). However, we employed only one cognitive test (Y-maze) and the cognitive deficits in ALS-PDC mice may not correspond to those detected by this task. Previous work in our laboratory has utilized the Morris Water Maze for cognitive assessment and significant impairments are present in the cycad-fed mice (Wilson et al. 2002). As noted above, we did not use this task as it confounds motor and cognitive deficits in its latency-dependent scoring. The radial arm maze has been used in the past in our laboratory but cycad feeding interferes with its required food deprivation. Furthermore, the Y-maze task may have not been sensitive enough to detect subtle changes in cognition. MN vulnerability and other neuronal populations MNs are known to be particularly susceptible to stress and subsequent death. Their large size, with axons extending for as long as over a meter in length, places a high metabolic demand on these neurons. This in turn can easily result in a high state of oxidative stress and mitochondrial dysfunction. M N are also more vulnerable to excitotoxicity than other neuronal populations. Interneurons, for example, and largely inhibitory and as such are not significantly affected by a glutamate-rich environment. In contrast, MNs are highly responsive to excess glutamate. It has been shown that, in cycad-fed mice, there is a downregulation of the glutamate transporter, GLT-1, in MNs (Wilson et al. 2003). This can lead to the accumulation of toxic concentrations of glutamate in the synapse and subsequent M N death. Parkinsonsim, the nigrostriatal system, and extrapyramidal symptoms The presence of motor features without corresponding M N loss may instead be explained 39 as symptoms of parkinsonism. The extrapyramidal symptoms of parkinsonism include tremor, rigidity, temporary paralysis, and bradykinesia - features which were displayed in particular by the mice in Experiment 2. Extrapyramidal symptoms, such as defective pursuit eye movements (Jacobs et al. 1981) have been described in ALS and ALS-plus syndromes feature extrapyramidal signs such as cerebellar ataxia (Schimke et al. 2002), rigidity, tremor, and slowness of movement (Zoccolella et al. 2002). In both familial and sporadic A L S , degeneration of midbrain dopaminergic cells have been reported (Geracitano et al. 2003), which may explain the presence of extrapyramidal signs in some cases of A L S . Alternatively, some have explained this by a concomitant parkinsonism; this disorder and A L S coexist to a greater degree than expected by chance (Takahashi et al. 1993). Others have not found any significant extrapyramidal signs nor any striatal degeneration in ALS patients (Vogels et al. 2000), which may have resulted from the rarity of ALS-plus. It is difficult to determine whether the symptoms displayed by the mice in the present experiments were due to ALS-like pathology, a concomitant parkinsonism, Guamian PDC, or another ALS-plus syndrome. Of relevance to this question, other work in our laboratory includes the examination of the nigrostriatal system in the same mice as used for the present experiments. It is not until day 175 in the first group of animals that any significant loss of cells, as determined by TH immunoreactivity, occurs (Schulz, personal communication). Given that it has been well documented that clinical symptoms of parkinsonism do not appear until approximately 60% of nigral neurons are lost (Leenders et al. 1990) and that our mice yet displayed gait disturbances typical of parkinsonism, this suggests that the behavioral deficits may have in fact resulted from an ALS-like pathology (which does precede PDC; Steele and Guzman 1987). However, more research is needed to clarify this issue. 40 It should also be noted that in some genetic models of parkinsonism there is no loss of cells in the SN, despite considerable behavioral symptoms (Gomez-Isla et al. 2003). For example, parkin-deficient mice display behavioral signs of parkinsonism but show no evidence of nigrostriatal dysfunction, including normal catecholamine levels in the striatum (Perez and Palmiter 2005). On the other hand, others report both motor deficits and neuronal degeneration (i.e., loss of DAergic neurons in the SN) in genetic models of parkinsonsim (Hwang et al. 2005; Kirik et al. 2002; Masliah et al. 2000; van der Putten et al. 2000). It is not clear why some of these genetic models, like the SOD-1 genetic models of A L S , feature behavioral deficits before cell loss, but it is of note that we have reproduced this behavioral dysfunction without neuron loss our cycad model, both with respect to spinal MNs (Experiments 1 and 2a) and TH-immunopositive striatal terminals (Experiment 1, days 75 and 100). This is the first such work to replicate these findings that were, until now, unique to genetic models, suggesting that there are significant similarities between genetic and environmentally-induced models of neurodegenerative disease. Putative actions and properties of cycad toxins It is not known whether cycad toxins remain in the CNS for prolonged periods, unleash secondary biochemical cascades that lead to neurodegeneration, or both. In any case, age dependence suggests that the brains of older, more vulnerable animals differ from those of younger mice, explained by Morrison and colleagues (1998) with three general theories. The first hypothesizes an age-related accumulation of toxicity that must reach a threshold level before neurons are destroyed. Mutant SOD-lor continued cycad exposure could gradually move the system towards this threshold. In this scenario, oxidative damage to proteins by free radicals and 41 oxidants occurs. A similar nitrosylation of proteins is indicated by increased levels of free 3-nitrotyrosine (Bruijn et al. 1997), as was also found in cycad-fed mice in Experiment 1 on days 75 and 100. A second hypothesis is that an excitotoxic environment accumulates around the neurons with age (Morrison et al. 1998). In ALS patients (Rothstein et al. 1995), SOD-1 mice (Bruijn et al. 1997), and previous work in our laboratory with ALS-PDC mice (Wilson et al. 2002), there are decreases in the glial glutamate transporter, G L T - l . This transporter normally removes glutamate from the synapse and its dysfunction could allow subsequent elevations in extracellular glutamate to overwhelm the system, leading to neurodegeneration. Thirdly, as neurons age, their protective mechanisms begin to fail (Morrison et al. 1998). To combine the three theories, younger mice are able to compensate for oxidative or excitotoxic stress, but older ones may be unable to do so. It has been often suggested that MNs are particularly vulnerable to oxidative stress, excitotoxicity, mitochondrial dysfunction, and protein aggregation, rendering them more susceptible to degeneration than other cell populations. The cells involved in parkinsonian degeneration are similarly highly susceptible to degeneration which may explain the emergence of motor symptoms before those of dementia in ALS-PDC. Conclusions and Future Directions The present data suggest that there is a certain "window of opportunity" in which behavioural deficits - analogous to clinical symptoms in human patients - are pronounced but M N loss has not yet begun to a significant degree. We propose that M N dysfunction is sufficient to produce behavioral changes, and that these precede the cell death that characterizes end-state ALS both in humans and in cycad-fed mice (Wilson et al. 2002). A similar time line occurs in 42 SOD-1 mice, with the emergence of behavioral deficits that precede significant cell loss (Morrison et al. 1998). Combined with the results of the present study, this suggests that a similar course of neuronal dysfunction and degeneration occurs in both genetic and environmental models of ALS/ALS-PDC. The present experiments increase our understanding of the age dependence and time course of our cycad model and raise questions about ALS-linked deficits versus parkinsonian extrapyramidal symptoms. Together, the data suggest that there is a strong imperative to improve early detection and treatment. Future work includes delineating the frames of the "window of opportunity" for recovery both in animal models and in human patients. The cellular phases of M N stress, dysfunction, and death could be examined with such measures as 3-nitrotyrosine for protein nitrosylation, malondialdehyde for lipid peroxidation, 8-hydroxy-2-deoxyguanosine for oxidative damage to DNA, GFAP for activated astrocytes, and Hsp 70 for the heat shock response. The age effects could be examined in truly "old" mice (18-24 months old) and the time course study could be extended to both younger and older animals. Finally, it is important to attempt the time course study again in hopes that cycad availability and consumption would be improved, allowing the analysis of the disease progression beyond initial behavioral deficits and M N loss. 43 References Aarsland, D., Andersen, K. , Larson, J.P., Lolk, A . , and Kragh-Sorensen, P. (2003) Prevalence and characteristic of dementia in Parkinson's Disease: an 8-year prospective study. Archives of Neurology, 60: 387-92. Aggarwal, A . , and Nicholson, G. (2002) Detection of preclinical motor neuron loss in SOD1 mutation carriers using motor unit number estimation. Journal of Neurology, Neurosurgery, and Psychiatry, 73: 199-201. Ahlskog, J.E., Waring, S.C., Kurland, L.T., Petersen, R.C., Moyer, T.P., Harmsen, W.S., Maraganore, P.C., O'Brien, P.C., Esteban-Santillan, C., and Bush, V . (1995) Guamian neurodegenerative disease: investigation of the calcium metabolism/heavy metal hypothesis. Neurology, 45: 1340-4. Alexander, M.D. , Traynor, B.J., Miller, N . , Corr, B., Frost, E., McQuaid, S., Brett, F .M. , Green, A. , and Hardiman, O. (2002) "True" sporadic A L S associated with a novel SOD-1 mutation. Annals of Neurology, 52: 680-3. Armani, M . , Pierobon-Bormioli, S., Mostacciuolo, M L . , Cacciavillani, M . , Cassol, M.A. , Candeago, R .M. , and Angelini, C. (1987) Familial ALS: clinical, genetic and morphological features. Advances in Experimental Medical Biology, 209: 109-10. 44 Audhali, N . and Stevenson, D. (2003) The Cycad Pages, Available HTTP:> (accessed 2 June 2003). Barneoud, P. and Curtet, O. (1999) Beneficial effects of lysine acetylsalicylate, a soluble salt of aspirin, on motor performance in a transgenic model of amyotrophic lateral sclerosis. Experimental Neurology, 255: 243-51. Beal, M.F., Ferrante, R.J., Browne, S.E., Matthews, R.T., Kowall, N.W., and Brown, R.H. Jr. (1997) Increased 3-nitrotyrosine in both sporadic and familial amyotrophic lateral sclerosis. Annals of Neurology, 42: 644-54. B i , R., Broutman, G., Foy, M.R., Thompson, R.F., and Baudry, M . (2000) The tyrosine kinase and mitogen-activated protein kinase pathways mediate multiple effects of estrogen in hippocampus. Proceedings of the National Academy of Science of the United States of America, 97: 3602-7. Braak, H. , and Braak, E. (1990) Alzheimer's disease: striatal amyloid deposits and neurofibrillary changes. Journal of Neuropathology and Experimental Neurology, 49: 215-24. Brody, J.A., Edgar, A . H . , and Gillespie, M . M . (1978) Amyotrophic lateral sclerosis: no increase among US construction workers in Guam. Journal of the American Medical Association, 240: 551-60. 45 Bruijn, L.I., Becher, M.W., Lee, M.K. , Anderson, K.L . , Jenkins, N.A. , Copeland, N.G. , Sisodia, S.S., Rothstein, J.D., Borchelt, D.R., Price, D.L., and Cleveland, D.W. (1997) ALS-linked SOD-1 mutant G85R mediates damage to astrocytes and promotes rapidly progressive disease with SOD-1 containing inclusions. Neuron, 18: 327-38. Calne, D.B. and Eisen, A . (1989) The relationship between Alzheimer's disease, Parkinson's disease and motor neuron disease. Canadian Journal of Neurological Sciences, 16: 547-50. Chen, L. (1981) Neurofibrillary change on Guam. Archives of Neurology, 38: 16-8. Chiu, A . Y . , Zhai, P., Dal Canto, M.C. , Peters T.M. , Kwon, Y.W. , Prattis, S.M., and Gurney, M.E. (1995) Age-dependent penetrance of disease in a transgenic mouse model of familial amyotrophic lateral sclerosis. Molecular and Cellular Neuroscience, 6: 349-62. de Medinaceli, L. , Freed, W.J., and Wyatt, R.J. (1982) A n index of the functional condition of rat sciatic nerve based on measurements made from walking tracks. Experimental Neurology, 11: 634-43. Doty, R.L., Perl, D.P., and Steele, J.C. (1991) Odor identification deficit of the parkinsonism-dementia complex of Guam: equivalence to that of Alzheimer's and idiopathic Parkinson's disease. Neurology, 41: 77-80. 46 Duke, J. A . (2003) Dr. Duke's Phytochemical and Ethnobotanical Databases: beta-sitosterol-beta-D-glucoside. Agriculture Research Services. Online. Available HTTP: ml> (accessed July 13 2003). Duncan, M.W., Steele, J .C , Kopin, I.J., and Markey, S.P. (1990) 2-Amino-3-(mefhylamino)-propanoic acid (BMAA) in cycad flour: an unlikely cause of amyotrophic lateral sclerosis and parkinsonism-dementia of Guam. Neurology, 40: 767-72. Elizan, T.S., Hirano, A. , Abrams, B . M . , Need, R.L., Van Nuis, C , and Kurland, L.T. (1966) Amyotrophic lateral sclerosis and parkinsonism-dementia of Guam: neurological reevaluation. Archives of Neurology, 14:356-68. Feeney, S.J., McKelvie, P.A., Austin, L., Jean-Frankcois, M.J.B., Kapsa, R., Tombs, S.M., and Byrne, E. (2001) Presymptomatic motor neurone loss and reactive astrocytosis in the SOD-1 mouse model of amyotrophic lateral sclerosis. Muscle and Nerve, 24: 1510-9. Forman, M.S., Schmidt, M.L . , Kasturi, S., Perl, D.O., Lee, V . M . , and Trojanowski, J.Q. (2002) Tau and alpha-synuclein pathology in amygdala for parkinsonism-dementia complex of Guam. American Journal of Pathology, 160: 1725-31. Fujino, Y . and Ohnishi, M . (1979) Isolation and structure of diglycosylsterols and triglycosylsterols in rice bran. Biochimica et Biophysica Acta, 574: 94-102. 47 Garruto, R .M. , Gajdusek, D.C., and Chen, K . M . (1980) Amyotrophic lateral sclerosis among Chamorro migrants from Guam. Annals of Neurology, 8: 612-9. Garruto, R .M. , Gajdusek, D.C., and Chen, K . M . (1981) Amyotrophic lateral sclerosis and parkinsonism-dementia among Filipino migrants to Guam. Annals of Neurology, 10: 341-50. Garruto, R . M . and Yanagihara, R. (1991) Amyotrophic lateral sclerosis in the Mariana Islands. In de Jong, J .M.B.V. (ed): Diseases of the Motor System: Handbook of Clinical Neurology, vol. 15. Amsterdam: Elsevier Science, pp.253-71. Geracitano, R., Paolucci, E., Prisco, S., Guatteo, E., Zona, C , Longone, O., Ammassari-Teule, M . , Bernardi, G., Berreta, N . , and Mercuri, N.B. (2003) Altered long-term corticostriatal synaptic plasticity in transgenic mice overexpressing human CU/ZN superoxide dismutase (GLY(93)^ A L A ) mutation. Neuroscience, 118: 399-408. Golaz, J., Bouras, C , and Hof, P.R. (1992) Motor cortex involvement in presenile dementia: report of a case. Journal of Geriatric Psychiatry and Neurology, 5: 85-92. Goldfarb, B.J., and Simon, S.R. (1984) Gait patterns in patients with amyotrophic lateral sclerosis. Archives of Physical and Medical Rehabilitation, 65: 61-5. 48 Gomez-Isla, T., Irizarry, M . C , Mariash, A. , Cheung, B., Soto, 0., Schrump, S., Sondel, J., Kotilinek, L., Day, J., Scharzschild, M.A. , Cha, J.H., Newell, K. , Miller, D.W., Ueda, K. , Young, A.B. , Hyman, B.T., and Ashe, K . H . (2003) Motor dysfunction and gliosis with preserved dopaminergic markers in human alpha-synuclein A30P transgenic mice. Neurobiology of Aging, 24: 245-58. Gordon, T., Hegedus, J., and Tarn, S.L. (2004) Adaptive and maladaptive motor axonal sprouting in aging and motoneuron disease. Neurology Research, 26: 174-85. Haddock, R.L. and Chen, K . M . (2003) Amyotrophic lateral sclerosis and diabetes on Guam: Changing patterns of chronic disease in an island community. Southeast Asian Journal of Tropical Medicine and Public Health, 34: 659-61. Hall, E.D., Oostveen, J.A., and Gurney, M.E. (1998) Relationship of microglial and astrocytic activation to disease onset and progression in a transgenic model of familial A L S . Glia, 23: 249-56. Hamilton, R.L., and Bowser, R. (2004) Alzheimer disease pathology in amyotrophic lateral sclerosis. Acta Neuropathologica, 107: 515-22. Hamson, D.K., Hu, J.H., Krieger, C , and Watson, N . V . (2002) Lumbar motoneuron fate in a mouse model of amyotrophic lateral sclerosis. Neuroreport, 13: 2291-4. 49 Hirano, A. , Kurland, L.T., Krooth, R.S., and Lessel, S. (1961a) Parkinsonism-dementia complex, an endemic disease on the island of Guam. I. Clinical features. Brain, 84: 642-61. Hirano, A. , Malamud, N . , and Kurland, L.T. (1961b) Parkinsonism-dementia complex, an endemic disease on the island of Guam. II. Pathological features. Brain, 84: 662-79. Hof, P.R., Perl, D.P., Loerzel, A.J . , Steele, J.C., and Morrison, J.H. (1994) Amyotrophic lateral sclerosis and parkinsonism-dementia from Guam: differences in neurofibrillary tangle distribution and density in the hippocampal formation and neocortex. Brain Research, 650: 107-16. Hwang, D.Y., Fleming, S.M., Ardayfio, P., Moran-Gates, T., Kim, H. , Tarazi, F.I., Chesselet, M.F., Kim, K.S. (2005) 3,4-dihydroxyphenylalanine reverses the motor deficits in Pitx3-deficient aphakia mice: behavioral characterization of a novel genetic model of Parkinson's disease. Journal of Neuroscience, 25: 2132-7. Hyman, B.T., Van Hoesen, G.W., Damasio, A.R., and Barnes, C L . (1984) Alzheimer's disease: cell-specific pathology isolates the hippocampal formation. Science, 225: 1168-70. Jacobs, L., Bozian, D., Heffner, R.R. Jr., and Barron, S.A. (1981) An eye movement disorder in amyotrophic lateral sclerosis. Neurology, 31: 1282-7. Jones, D.L. (2002) Cycads of the World, 2 n d edn, Washington: Smithsonian Institution Press. 50 Kallianos, A .G. , Shelburne, F.A., Means, R.E., Stevens, R.K., Lax, R.E., and Mold, J.D. (1963) Identification of the D-glucosides of stigmasterol, sitosterol and campestrerol in tobacco and cigarette smoke. Biochemical Journal, 87: 596-600. Khabazian, I., Bains, J.S., Williams, D.E., Cheung, J., Wilson, J .M., Pasqualotto, B.A. , Pelech, S.L., Andersen, R.J., Wang, Y.T., Liu, L., Nagai, A. , Kim, S.U., Craig, U.K. , and Shaw, C A . (2002) Isolation of various forms of sterol P-D-glucoside from the seed of Cycas circinalis: neurotoxicity and implications for ALS-parkinsonism dementia complex. Journal of Neurochemistry, 82: 516-28. Kirik, D., Rosenblad, C , Burger, C , Lundberg, C , Johansen, T.E., Muzyczka, N . , Mandel, R.J., Bjorklund, A. (2002) Parkinson-like neurodegeneration induced by targeted overexpression of alpha-synuclein in the nigrostriatal system. Journal of Neuroscience, 22: 2780-91. Kisby, G.E., Ellison, M . , and Spencer, P.S. (1992) Content of the neurotoxins cycasin (methylazoxymethanol p-D-glucoside) and B M A A (p-N-methylamino-L-alanine) in cycad flour prepared by Guam Chamorros. Neurology, 42: 1336-40. Knapp, F., Aexel, R., and Nicholas, H.J. (1968) The non-saponifiable constituents of lettuce. Journal of Food Science, 33: 159-62. 51 Kobayashi, K. , Shimoda, K. , Higashima, M . , Nakano, H. , Miyazu, K. , Hayashi, M . , Tabata, O., and Koshino, Y . (2000) Report of three cases of Alzheimer's disease with focal motor symptoms: clinical correlates of neuroimaging findings. World Journal of Biological Psychiatry, 1: 164-9. Kosak, A.I. and Swinehart, J.S. (1960) Some neutral components of cigarette smoke. Journal of Organic Chemistry, 25: 222-5. Kostic, V . , Gurney, M.E. , Deng, H.X. , Siddique, T., Epstein, C.J., and Przedborski, S. (1997) Midbrain dopaminergic neuronal degeneration in a transgenic mouse model of familial amyotrophic lateral sclerosis. Annals of Neurology, 41: 497-504. Kruger, R., Muller, T., and Riess, O. (2000) Involvement of alpha-synuclein in Parkinson's disease and other neurodegenerative disorders. Journal of Neural Transmission, 107: 31-40. Kurland, L.T. (1988) Amyotrophic lateral sclerosis and Parkinson's disease complex on Guam linked to an environmental neurotoxin. Trends in the Neurosciences, 11: 51-4. Kurland, L.T. (1992) Amyotrophic lateral sclerosis-parkinsonism-dementia complex in Guam: epidemiologic and etiological perspectives, in A.C . Williams (ed.) Motor Neuron Disease. Williams, London: Chapman. 52 Kurland, L.T., Hirano, A. , Malamud, N . , and Lessell, S. (1961) Parkinsonism-dementia complex, an endemic disease on the island of Guma: Clinical, pathological, and epidemiological features. Transcripts of the American Neurological Association, 86: 115-20. Kurland, L.T. and Mulder, D.W. (1954) Epidemiological investigations on amyotrophic lateral sclerosis. I. Preliminary report on geographic distribution with special reference to the Mariana Islands including clinical and pathologic observations. Neurology, 4: 355-378, 438-448. Leenders, K .L . , Salmon, E.P., Tyrell, P., Brooks, D.J., Sager, I.I., Jones, T., Marsden, C D . , and Frackowiak, R.S. (1990) The nigrostriatal dopaminergic system assessed in vivo by positron emission tomography in healthy volunteer subjects and patients with Parkinson's disease. Archives of Neurology, 47: 1290-8. Lilienfeld, D.E., Perl, D.P., and Olanow, C W . (1994) Guam Neurodegeneration, in D.B. Calne, and W.B. Saunders (eds.) Neurodegenerative Diseases. Philadelphia: W.B. Saunders Co., pp. 895-908. Lopez, O.L., Becker, J.T., and DeKosky, S.T. (1994) Dementia accompanying motor neuron disease. Dementia, 5: 42-7. Masliah, E., Rockenstein, E., Veinbergs, I., Mallory, M . , Hashimoto, M . , Takeda, A. , Sagara, Y. , Sisk, A. , and Mucke, L. (2000) Dopaminergic loss and inclusion body formation in a-synuclein mice: Implications for neurodegenerative disorders. Science, 287: 1265-9. 53 McGeer, P.L., Itagaki, S., Akiyama, H., and McGeer, E.G. (1988) Rate of cell death in parkinsonism indicates active neuropathological process. Annals of Neurology, 24: 574-6. McGeer, P.L., Schwab, C., McGeer, E.G., Haddock, R.L., Steele, J.C. (1997) Familial nature and continuing morbidity of the amyotrophic lateral sclerosis-parkinsonism dementia complex of Guam. Neurology, 2: 400-9. Morris, R .G.M. (1984) Development of water-maze procedure for studying spatial learning in the rat. Journal of Neuroscience Methods, 11: 47-60. Morrison, B . M . , Janssen, W.G., Gordon, J.W., and Morrison, J.H. (1998) Time course of neuropathology in the spinal cord of G86R superoxide dismutase transgenic mice. The Journal of Comparative Neurology, 391: 64-77. Mourelatos, Z., Hirano, A. , Rosenquist, A.C. , and Gonatas, N . K . (1994) Fragmentation of the Golgi apparatus of motor neurons in amyotrophic lateral sclerosis (ALS). Clinical studies in A L S of Guam and experimental studies in deafferented neurons and in P, P'-iminodipropionitrile axonopathy. American Journal of Pathology, 144: 1288-300. Mukai, E., Sakakibara, T., Sobue, I., and Chen, K . M . (1982) Prognosis of A L S in Guam and Japan. Clinical Neurology, 22: 139-44. 54 Norstog, K.J . and Nicholls, T.J. (1997) The Biology of the Cycads. Ithaca: Cornell University Press. Okumura, H. (2003) Epidemiological and clinical patterns of western pacific amyotrophic lateral sclerosis (ALS) in Guam and sporadic ALS in Rochester, Minnesota, U.S.A. and Hokkaido, Japan: A comparative study. Hokkaido Igaku Zasshi, 78: 187-95. Olton, D.S., and Samuelson, R.J. (1976) Remembrance of places passed: spatial memory in rats. Journal of Experimental Psychology: Animal Behavioral Processes, 2: 97-116. Peavy, G.M. , Herzog, A .G. , Rubin, N.P., and Mesulam, M . M . (1992) Neuropsychological aspects of dementia of motor neuron disease: a report of two cases. Neurology, 42: 1004-8. Perez, F.A. and Palmiter, R.D. (2005) Parkin-deficient mice are not a robust model of parkinsonism. Proceedings of the National Academy of Science of the United States of America, 102: 2174-9. Plato, C C , Garruto, R .M. , Galasko, D., Craig, U.K. , Plato, M . , Gamst, A. , Torres, J.M., and Wiederholt, W. (2003) Amyotrophic lateral sclerosis and parkinsonism-dementia complex of Guam: changing incidence rates during the past 60 years. American Journal of Epidemiology, 157: 149-57. 55 Reed, W., Torres, J .M., and Brody, J.A. (1975) A L S and PD on Guam, 1945 - 1972. II. Familial and genetic studies. American Journal of Epidemiology, 101: 302-10. Robberecht, W. (2000) Genetics of amyotrophic lateral sclerosis. Journal of Neurology, 247: 2-6. Rocchi A. , Pellegrini, S., Siciliano, G., and Murri, L. (2003) Causative and susceptibility genes for Alzheimers's disease: a review. Brain Research Bulletin, 61: 1-24. Rodgers-Johnson, P., Garruto, R .M. , Yanagihara R., Chen, K . M . , Gajdusek, D.C., and Gibbs, C.J. Jr. (1986) Amyotrophic lateral sclerosis and parkinsonism dementia on Guam: a 30 year evaluation of clinical and neuropathological trends. Neurology 36: 7-13. Rosen, D.R. (1993) Mutation in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature, 364: 362. Rothstein, J.D., Van Kammen, M . , Levey, A.I., Martin, L.J. , and Kuncl, R.W. (1995) Selective loss of glial glutamate transporter GLT-1 in amyotrophic lateral sclerosis. Annals of Neurology, 38: 73-84. Sanger, V.L . , Yang, M . , and Mickelsen, O. (1972) Cycasin-induced central nervous system lesions in postnatal mice. Sixth International Cycad Conference. Bethesda: Federation of American Societies for Experimental Biology. 56 Sango, K., McDonald, M.P., Crawley, J.N., Mack, M.L. , Tifft, C.J., Skop, E., Starr, C M . , Hoffmann, A. , Sandhoff, K. , Suzuki, K., and Proia, R.L. (1996) Mice lacking both subunits of lysosomal P-hexosaminidase display gangliosidosis and mucopolysaccharidosis. Nature Genetics, 14: 348-352. Schimke, N . , Krampfl, K. , Petri, S., Dengler, R., and Buffer, J. (2002) Cerebellar symptoms in motor neuron disease: special form of amyotrophic lateral sclerosis plus syndrome. Nervenarzt, 73:751-3. Shaw, C A . , Wilson, J .M. (2003) Analysis of neurological disease in four dimensions: insight from ALS-PDC epidemiology and animal models. Neuroscience and Biobehavioral Reviews, 27: 493-505. Shaw, C A . , Wilson, J.M.B., and Khabazian, I. (2002) Reverse engineering neurological disease. Proceedings of International Conference on Complex Systems. Available HTTP: abstract.cgi?7759.html> (accessed 30 July 2003). Siddique, T., and Deng, H.X. (1996) Genetics of amyotrophic lateral sclerosis. Human Molecular Genetics, 5: 1465-70. 57 Snow, B.J., Peppard, R.F., Guttman, M . , Okada, J., Martin, W.R., Steele, J., Eisen, A. , Carr, G., Schoenberg, B., and Calne, D. (1990) Positron emission tomographic scanning demonstrates a presynaptic dopaminergic lesion in Lytico-Bodig. The amyotrophic lateral sclerosis-parkinsonism-dementia complex of Guam. Archives of Neurology, 47: 870-4. Spencer, P.S., Nunn, P., Hugon, J., Ludolph, A.C. , Ross, S.M., Roy, D.N., and Robertson, R.C. (1987) Guam ALS-PDC linked to a plant excitant neurotoxin. Science, 237: 517-22. Staphylakis, K . and Gegiou, D. (1985) Free, esterified and glucosidic sterols in cocoa butter. Lipids, 20: 723-8. Steele, J.C. and Guzman, T. (1987) Observations about amyotrophic lateral sclerosis and parkinsonism-dementia complex of Guam with regard to epidemiology and etiology. Canadian Journal of Neurological Sciences, 14: 358-62. Swash, M . , and Ingram, D. (1988) Preclinical and subclinical events in motor neuron disease. Journal of Neurology, Neurosurgery, and Psychiatry, 51: 165-8. Takahashi, H. , Snow, B.J., Bhatt, M.H. , Peppard, R., Eisen, A. , and Calne, C.B. (1993) Evidence for a dopaminergic deficit in sporadic amyotrophic lateral sclerosis on positron emission scanning. Lancet, 342: 1016-8. 58 Tanner, C M . , Ottman, R., Goldman, S.M., Ellenburg, J., Chan, P., Mayeux, R., and Langston, J.W. (1999) Parkinson's disease in twins: an etiologic study. Journal of the American Medical Association, 281: 341-6. van der Putten, H. , Wiederhold, K . H . , Probst, A. , Barbieri, S., Mistl, C , Danner, S., Kauffmann, S., Hofele, K., Spooren, W.P., Ruegg, M.A. , Lin, S., Caroni, P., Sommer, B., Tolnay, M . , and Bilbe, G. (2000) Neuropathology in mice expressing human alpha-synuclein. Journal of Neuroscience, 20: 6021-29. Vaphiades, M.S., Husain, M . , Juhasz, K., and Schmidley, J.W. (2002) Motor neuron disease presenting with slow saccades and dementia. Amyotrophic Lateral Sclerosis and Other Motor Neuron Disorders, 3: 159-62. Verghese, L., Lipton, R.B., Hall, C.B., Kuslansky, G., Katz, M.J. , and Buschke, H. (2002) Abnormality of gait as a predictor of non-Alzheimer's dementia. New England Journal of Medicine, 347: 1761-8. Vogels, O.J.M., Veltman, J., Oyen, W.J.G., and Horstink, M.W.I .M. (2000) Decreased striatal dopamine D2 receptor binding in amyotrophic lateral sclerorir (ALS) and multiple system atrophy (MSA): D2 receptor down-regulation versus striatal cell degeneration. Journal of the Neurological Sciences, 180:62-5. 59 Wengenack, T .M, Holasek, S.S., Montano, C M . , Gregor, D., Curran, G.L., and Poduslo, J.F. (2004) Activation of programmed cell death markers in ventral horn neurons during early presymptomatic stages of amyotrophic lateral sclerosis in a transgenic mouse model. Brain Research, 19: 73-86. Whitehouse, P.J., Price, D.L., Stuble, R.G., Clark, A.W., Coyle, J.T., and Delon, M.R. (1982) Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain. Science, 215: 1237-9. Wilson, J.M.B., Khabazian, I., Pow, D.V., Craig, U.-K. , and Shaw., C A . (2003) Decrease in glial glutamate transporter variants and excitatory amino acid receptor down-regulation in a murine model of ALS-PDC. Neuromolecular Medicine, 3: 41-53. Wilson, J.M.B., Khabazian, I., Wong, M . C , Seyedalikhani, A . , Bains, J.S., Pasqualotto, B.A., Williams, D.E., Andersen, R.J., Simpson, R.J., Smith, R., Craig, U.K. , Kurland, L.T., and Shaw, C A . (2002) Behavioral and neurological correlated of ALS-parkinsonism dementia complex in adult mice fed washed cycad flour. Neuromolecular Medicine, 1: 207-21. Wilson, J.M.B., Petrik, M.S., Grant, S.C, Blackband, S.J., Lai, J., and Shaw, C A . (2004) Quantitative measurement of neurodegeneration in an ALS-PDC model using M R microscopy. Neuroimage, 23: 336-43. 60 Wilson, J.M.B., Petrik, M.S., Moghadasian, M.H. , and Shaw, C A . (2005) Examining the interaction of ApoE and neurotoxicity on a murine model of ALS-PDC. Canadian Journal of Physiology and Phamacology, 83: 131-41. Wootz, H., Hansson, I., Korhonen, L., Napankangas, U. , and Lindholm, D. (2004) Caspase-12 cleavage and increased oxidative stress during motoneuron degeneration in transgenic mouse model of A L S . Biochemical and Biophysical Research Communications, 10: 281-6. Wright, H.E., Burton, W.W., and Berry, R.C. (1962) Identification of stigmasterol D-glucoside in aged hurley tobacco. Journal of Organic Chemistry, 27: 918-33. Yanagihara, R., Garruto, R .M. , Gajdusek, D . C , Tomita, A. , Uchikawa, T., Konagaya, Y. , Chen, K . M . , Sobue, I., Plato, C C , and Gibbs, C J . Jr. (1984) Calcium and vitamin D metabolism in Guamian Chamorros with amyotrophic lateral sclerosis and parkinsonism-dementia. Annals of Neurology, 15:42-8. Yasui, M . , Ota, K. , and Yoshida, M . (1997) Effects of low calcium and magnesium dietary intake on the central nervous system tissues of rats and calcium-magnesium related disorders in the amyotrophic lateral sclerosis focus in the K i i Peninsula of Japan. Magnes Research, 39-50. Yokoyama, K., Ikebe, S., Komatsuzaki, Y. , Takanashi, M . , Mori, H. , Mochizuki, H. , and Mizuno, Y . (2002) A 68-year-old woman with dementia and parkinsonism. No To Shinkei, 54: 175-84. 61 Zhang, Z.X. , Anderson, D.W., Lavine, L., and Mantel, N . (1990) Patterns of acquiring PDC on Guam: 1944 through 1985. Archives ofNeurology, 47: 1019-24. Zimmerman, H . M . (1945) Monthly Report to Medical Officer in Command. US Naval Medical Officer in Command, US Naval Medical Research Unit No.2. Zoccolella, S., Palagano, G., Graddosio, A. , Russo, I., Ferrannini, E., Serlenga, L. , Maggio, F., Lamberti, and Ilicet, G. (2002) ALS-plus: 5 cases of concomitant amyotrophic lateral sclerosis and parkinsonism. Neurological Sciences, 23: SI23-4. 62 Figure captions Figure 1. Venn diagram showing how the variables of genetics, environment, and age influence the risk of neurodegenerative disease. Vulnerability to developing A L S , PD, A D , or ALS-PDC is lower when only one of the three factors (deleterious genetic mutation, exposure to environmental toxin, or advanced age) is present and increases when two (shaded areas) or all three factors (darkened area) are present. Figure 2a-b. Feeding schedule for mice in Experiments 1 and 2. In Experiment 1, mice were fed cycad or control pellets ("fed;" filled box) for 100 days and then given normal chow ("not fed;" open box) for 75 days. The mice in Experiment 2 were fed for 60 days, not fed for 30 days, fed for 90 days, and not fed for 180 days. Figure 3. Weights of mice from Experiment 1. There are no significant differences between groups, p > 0.05 Figure 4 a-d. Results of behavioral tasks for Experiment 1. The gait length task (a) reveals some differences between groups during the second half of the experiment while leg extension (b) deficits are present from day 50 to 110 after which recovery to levels indistinguishable from control mice occurs. Differences in performance on the wire hang task (c) emerge around day 100 and continue for the remainder of the experiment. There are no significant differences between groups on the Y-maze test (d) although, on day 126, cycad-fed mice appeared to take longer to reach the reward, although this difference was nonsignificant (p > 0.05). 63 Figure 5. Results of cresyl violet motor neuron counts for Experiment 1. There are no significant differences between control and cycad-fed mice. F(5, 17) = 0.7510, p = 0.5967. Figure 6a-f. Photomicrographs of representative sections of cresyl violet-stained lumbar spinal cord (Experiment 1) for control (a) and cycad-fed (b) mice on day 75, control (c) and cycad-fed (d) mice on day 100, and control (e) and cycad-fed mice (f) on day 175. Note that there appears to be no significant differences between groups on any of the days (see text for details). Scale bar = 50 pm (a-f). Figure 7. Weights for animals in Experiment 2a. There are no significant differences between control and cycad-fed mice, p > 0.05. Figure 8a-c. Results of behavioral tasks for Experiment 2. Neither leg extension (a), wire hang (b), nor water maze (c) tests revealed any significant differences between groups, p > 0.05. Figure 9. Results of cresyl violet motor neuron counts for Experiment 2a. There are no significant differences between the groups, p > 0.05. Figure lOa-b. Photomicrographs of representative sections of cresyl violet-stained lumbar spinal cord (Experiment 2a) for control (a) and cycad-fed (b) mice. Note that there appears to be no significant differences between groups on any of the days (see text for details). Scale bar = 50 pm (a-b). 65 Experiment 1: H "feeding" I "no feeding" day 0 100 175 Experiment 2 (a & b): day 0 60 90 180 360 Day 67 c 150-(0 2 £ 100-is •3 o 0) 504 75 100 Day X r z z i Control i a Cycad 175 10-0H 1 1 1 1 1 1 1 1 1 0 50 100 150 2 0 0 2 5 0 3 0 0 3 5 0 4 0 0 4 5 0 Day Day 125-i £ 100-2 o •g O 75-| 2 Z © 50-ro © ~-S 25-Control I Cycad Contro l C y c a d 73 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items