Twenty-five years after 700,000 U.S. troops fought and won the first Gulf War with remarkably low casualties, research "clearly and consistently" shows that exposure to pesticides and other toxins caused Gulf War Illness, a complex and debilitating disorder that affects as many as 250,000 of those deployed, according to a new report led by a Boston University School of Public Health (BUSPH) researcher.
In a special issue of the journal Cortex that coincides with the 25th anniversary of the war, Roberta White, professor of environmental health at BUSPH, and colleagues from a dozen other institutions comprehensively review studies on Gulf War Illness (GWI), especially those since 2008. They conclude that exposure to pesticides and ingestion of pyridostigmine bromide (PB) - prophylactic pills intended to protect troops against the effects of possible nerve gas -- are "causally associated with GWI and the neurological dysfunction in Gulf War veterans."
The research team also cites multiple studies showing a link between veterans' neurological problems and exposure to the nerve-gas agents sarin and cyclosarin, as well as to oil well fire emissions.
These "toxic wounds" resulted in damage to veterans' nervous systems and immune systems, including neuroendocrine and immune dysregulation, autonomic nervous system irregularities, and reduced white and gray matter in veterans' brains, the review says.
White and colleagues have been studying the health of troops deployed in the 1991 Gulf War for more than 20 years to determine why so many of them suffer from a multi-system disorder characterized by fatigue, joint and muscle pain, headaches, concentration and memory problems, gastrointestinal distress, and skin rashes. They note that effective treatments for the illness have been elusive, but that a recent treatment research effort has begun to produce promising leads.
"Further research into the mechanisms and etiology of the health problems of (Gulf War) veterans is critical to developing biomarkers of exposure and illness, and preventing similar problems for military personnel in future deployments. This information is also critical for developing new treatments for GWI and related neurological dysfunction," they write.
In 2008, a Congressionally mandated panel directed by White -- the Research Advisory Committee on Gulf War Veterans' Illnesses -- issued a landmark report concluding that Gulf War Illness was a "real" disorder, distinct from stress-related syndromes, and urging a robust research effort into its causes and potential cures. Gulf War veterans have complained for years that the Department of Veterans Affairs (VA) has not taken the illness seriously.
In terms of toxic exposures, the authors note, six out of seven research studies have found "significant associations between self-reported pesticide exposure and GWI." Similarly, ingestion of PB pills dispensed by the military has been "consistently linked to ill health in GW veteran populations."
James Binns, a co-author of the report and former chairman of the Research Advisory Committee, equated the main causes of GWI to "friendly fire."
"We did it to ourselves," Binns said. "Pesticides, PB, nerve gas released by destroying Iraqi facilities -- all are cases of friendly fire. That may explain why government and military leaders have been so reluctant to acknowledge what happened, just as they tried to cover up Agent Orange after Vietnam. Certainly, the government should have been facing the problem honestly and doing research from the start to identify diagnostic tests and treatments."
In the report, the research team notes that in addition to veterans suffering from GWI, other deployed troops from the first Gulf War report a variety of neurological disorders, either in conjunction with GWI or as separate ailments. Studies have found that deployed troops suffer a higher incidence of stroke, brain cancer and ALS (amyotrophic lateral sclerosis), compared to non-deployed veterans. The VA's own study, published in 2009, found that deployed veterans were diagnosed with seizures, stroke and neuralgia at higher rates than non-deployed service members. Other studies have found excess rates of brain structure alterations and brain cancer deaths among veterans who had the greatest exposure to nerve agents or oil fire smoke.
The report makes clear that psychiatric problems "have been ruled out" as a cause of Gulf War Illness, noting that Gulf War veterans have lower rates of post-traumatic stress disorder (PTSD) and other psychiatric disorders than their counterparts who served in other wars.
The research team says that a number of studies using diagnostic imaging and EEG probes have identified "structural and electrical abnormalities" in the central nervous systems of deployed troops with GWI. Fourteen of 15 papers published since 2008 support that conclusion, the report says.
White and colleagues say that overall, the research to date supports the conclusion that veterans are suffering from a "persistent pathology due to chemical intoxication." They say further research into GWI could benefit other occupational groups, such as farmers and insecticide applicators, who have similar exposures. White said she is hopeful that new research efforts will lead to effective treatments, especially as veterans age and are at increased risk of neuro-degenerative diseases.
Co-authors on the study include Kimberly Sullivan, a research assistant professor in environmental health at BUSPH who is the principal investigator on a large, multi-site Gulf War Illness Consortium study; Lea Steele of Baylor University Institute of Biomedical Studies; James O'Callaghan of the Health Effects Laboratory Division, Centers for Disease Control and Prevention; Beatrice Golomb of the University of California, San Diego; Floyd Bloom of The Scripps Research Institute; James Bunker of the National Gulf War Resource Center; Fiona Crawford of the Roskamp Institute; Joel Graves, retired U.S. Army captain; Anthony Hardie of Veterans for Common Sense; Nancy Klimas of the Institute for Neuro-Immune Medicine, Nova Southeastern University; Marguerite Knox of McEntire Joint National Guard Base; William Meggs of the Brody School of Medicine, East Carolina University School of Medicine; Jack Melling of the U.S. Government Accountability Office; Martin Philbert of the School of Public Health, Ann Arbor; and Rachel Grashow of Northeastern University.
Boston University School of Public Health, founded in 1976, offers master's-and doctoral-level education in public health. The faculty in six departments (biostatistics; community health sciences; environmental health; epidemiology; global health; and health law, policy & management) conducts policy-changing public health research around the world, with the mission of improving the health of populations -- especially the disadvantaged, underserved, and vulnerable -- locally, nationally, and internationally.
NOTICE: The January 26, 2016 meeting of the Congressionally-chartered VA Research Advisory Committee on Gulf War Veterans' Illnesses (RAC) has been postponed due to the winter weather emergency in Washington, DC.
(91outcomes.com) - A new study published in the January 12, 2016 edition of the medical journal, "Frontiers in Integrative Neuroscience," shows that in an animal model of Gulf War Illness, exposures to Gulf War toxins led to persistent neuroinflammation and neuropathological changes. The researchers were also able to identify novel biological pathways, correlate the results with outcomes, and identify potential treatment targets. The study, led by Dr. Ghania Ait-Ghezala of the Roskamp Institute in Sarasota, Fla., and assisted by Dr. Fiona Crawford, Roskamp President and CEO, was funded by the treatment-focused Gulf War Illness Research Program,(GWIRP) and did what it set out to do. The GWIRP is part of the Department of Defense Congressionally Directed Medical Research Program (CDMRP) In the study, the laboratory animals were exposed to a combination of Pyridostigmine Bromide (PB) and Permethrin. During the 1991 Gulf War, many U.S. and some Coalition troops took the experimental Pyridostigmine Bromide (PB) pills as a protective measure in the event of nerve gas exposure. At the same time, many U.S. troops used copious amounts of pesticides during the war, often without being able to shower for long periods of time, including Permethrin -- a potent pyrethroid pesticide. Several animal studies have shown that PB, in combination with Gulf War pesticides or low-level chemical warfare agents, results in long-lasting symptoms that appear similar to those in Gulf War Illness. Another Roskamp researcher involved in the study, Dr. Laila Abdullah, is currently funded by the GWIRP in a related study aimed at restoring the brain's lipid homeostasis in order to treat neurogenic symptoms. Entitled, "Restoring the Brain's Lipid Homeostasis as a Therapeutic Avenue for Treating the CNS Symptoms of Gulf War Illness," the study is testing phospholipids, drugs that increase functional peroxisomes, and supplement lipid metabolites to enhance brain cell function. Dr. Crawford was also recently funded by the GWI CDMRP for another Gulf War Illness treatment study related to these newest findings. Entitled, "Double-Blind, Placebo-Controlled Crossover Study of the Anti-Inflammatory Compound Anatabine to Treat Pain in GWI Patients," the study will test an anti-inflammatory nutraceutical with the goal of reducing persistent neuroinflammation and improving debilitating, chronic Gulf War Illness symptoms. The small study is expected to be completed in 2018, paving the way for a broader treatment trial if successful. -Anthony Hardie, 91outcomes.com
SOURCE: PubMed (Front Integr Neurosci), Jan. 12, 2016, Zakirova et al
Gulf War Illness (GWI) is a chronic multisymptom illness with a central nervous system component that includes memory impairment as well as neurological and musculoskeletal deficits. Previous studies have shown that in the First Persian Gulf War conflict (1990-1991) exposure to Gulf War (GW) agents, such as pyridostigmine bromide (PB) and permethrin (PER), were key contributors to the etiology of GWI. For this study, we used our previously established mouse model of GW agent exposure (10 days PB+PER) and undertook an extensive lifelong neurobehavioral characterization of the mice from 11 days to 22.5 months post exposure in order to address the persistence and chronicity of effects suffered by the current GWI patient population, 24 years post-exposure. Mice were evaluated using a battery of neurobehavioral testing paradigms, including Open Field Test (OFT), Elevated Plus Maze (EPM), Three Chamber Testing, Radial Arm Water Maze (RAWM), and Barnes Maze (BM) Test. We also carried out neuropathological analyses at 22.5 months post exposure to GW agents after the final behavioral testing. Our results demonstrate that PB+PER exposed mice exhibit neurobehavioral deficits beginning at the 13 months post exposure time point and continuing trends through the 22.5 month post exposure time point. Furthermore, neuropathological changes, including an increase in GFAP staining in the cerebral cortices of exposed mice, were noted 22.5 months post exposure. Thus, the persistent neuroinflammation evident in our model presents a platform with which to identify novel biological pathways, correlating with emergent outcomes that may be amenable to therapeutic targeting. Furthermore, in this work we confirmed our previous findings that GW agent exposure causes neuropathological changes, and have presented novel data which demonstrate increased disinhibition, and lack of social preference in PB+PER exposed mice at 13 months after exposure. We also extended upon our previous work to cover the lifespan of the laboratory mouse using a battery of neurobehavioral techniques.
Original Funded Proposal: "Proteomic Immune Profiling for the Therapeutic Modulation of Cognitive Impairment in a Novel GWI Mouse Model", Dr. Ghania Ait-Ghezala, Principal Investigator, Roskamp Institute, Sarasota, Fla. (GW100076)
SOURCE: Gulf War Illness Research Program, Congressionally Directed Medical Research Program
Veterans of the Gulf War (1990-1991) have a higher rate of "chronic multi-symptom health problems" than either non-deployed military personnel or those deployed to other regions. The severity of illness can range from moderate effects (some problems, but ability to maintain a normal lifestyle) to highly debilitating, where it interferes with the activities of daily living. Nearly 20 years since the original conflict, most veterans with this illness continue to suffer from these chronic symptoms and only a few have either recovered or improved. Among the 37 persistent symptoms, the most commonly reported symptoms included fatigue, gastrointestinal problems, chronic and widespread pain, and neurological and musculoskeletal symptoms. Further investigations have revealed that symptoms relating to gastrointestinal, respiratory, and mood/cognition are the three key correlated symptoms reported to be present among both the UK and US Gulf War veterans. Additional support for significant cognitive component to Gulf War illnesses (GWI) comes largely from brain imaging studies showing damage to brain structures.
The cause of these central nervous system (CNS) abnormalities remains unknown. Interestingly, several clinical studies on GWI suggest that there is a significant impairment of immune and inflammatory responses in the periphery of veterans with GWI. Interestingly, the chemicals pyridostigmine bromide given to soldiers as prophylaxis against nerve gas attack and permethrin, a pesticide used to protect against pest-borne illnesses, have been shown to affect immune system and inflammation in animals. We suspect that these chemicals, in combination with each other, may have resulted in an increase of proinflammatory proteins, which could have subsequently entered the brain and caused clinical symptoms that we observe in veterans with GWI. Therefore, our work will focus on identifying biological functions in order to figure out the relationship between cognitive and the immune component of GWI. However, given the complexity of clinical presentation, which is probably obscured by the heterogeneity of the veteran population, due to hereditary and other environmental differences, the standard technical approaches may not be suitable for examining the underlying etiology of this illness. We have state-of-the- art technology available to address this problem as well as several decades of experience in researching brain inflammation associated with cognitive diseases. This "proteomic technology" is a method that enables us to identify and measure hundreds to thousands of proteins simultaneously even in a small amount of sample and therefore can provide a picture of several biological pathways at the same time. Therefore, we propose to use this technology to identify profiles that are specific to immune and inflammatory responses in a mouse model of GWI, thereby allowing us to examine the relationship between these GWI exposures and immune/inflammatory dysfunction in a more homogenous study sample. We will identify biological pathways that can be targeted using pharmacological approaches and other genetic animal modeling approaches. We will examine whether we can successfully target the immune/inflammatory component of GWI and whether by doing so we can resolve the brain inflammation and change neurobehavioral symptoms of cognitive impairment. We are confident that our approach here will potentially lead to several new therapeutic avenues for treatment of the brain inflammation component of GWI, ultimately leading to their clinical testing in veterans with this chronic and persistent sickness.
Background: Investigation of Gulf War illnesses (GWI) provides evidence that the chemicals used as pesticides or as prophylaxis for nerve agent exposure contributed to the symptoms observed in GWI. Among these agents, pyridostigmine bromide (PB) and permethrin (PER) have been implicated in GWI pathogenesis, and findings suggest that these GW agents may synergistically activate processes in the central nervous system (CNS) and periphery that are associated with an aberrant immune/inflammatory response observed in veterans with GWI. However, the mechanism by which they contribute to GWI is not well-defined. Further studies are necessary to better understand the association of these immune/inflammatory responses with cognitive impairment in GWI and to identify methods of treatment. Hypothesis: Our GWI mouse model recapitulates delayed cognitive impairment and CNS and peripheral immune/inflammatory response to exposure to PB and PER (GW agents). In exploring potential mechanisms through which CNS and peripheral dysfunction could result from such exposure, a plausible mechanism is suggested by studies that support a crosstalk through (1) a dysfunction of the "cholinergic anti-inflammatory pathway," which leads to (2) chronic reduction of vagal activity resulting in (3) reduced suppression of proinflammatory cytokine (Th1) production in the periphery and consequently (4) their entrance into the brain, which could be responsible for the multisymptom illness complex observed in veterans with GWI. We have much experience in applying proteomic approaches to the interrogation of complex disorders, and the use of our proteomic platform will allow us to examine the diverse biological functions and pathways that are associated with the complexity of GWI etiology, providing us with an opportunity to identify key regulators of impaired immune/inflammatory responses. Specific Aims: (1) To establish which proteomic profiles of CNS/peripheral immune/inflammatory response are associated with cognitive impairment/neuropathological abnormalities in a GWI animal model. (2) In order to determine the effects of modulation of immune/inflammatory response on the recovery of normal phenotypes of proteome profiles, cognition and neuropathology we will (2a) test the effects of the genetic manipulation of immune/inflammatory response on the recovery of normal phenotypes in a GWI animal model and (2b) test the effects of pharmacological modulators of the immune/inflammatory response on the recovery of normal phenotypes in a GWI animal model. Study Design: We will apply proteomics to evaluate brain and plasma samples from the GWI mouse model, quantify Th1/Th2 cytokines from these samples, and then interrogate the differentially expressed proteins using the Ingenuity Pathway Analysis (IPA) knowledge base to identify proteomic profiles specific to GW agent-induced immune/inflammatory response, which will then be correlated with cognitive impairment and neuropathology. We will target biofunctions identified by IPA to be significantly altered in response to GW agent exposure. We will use genetic and pharmacological modulators to manipulate the immune/inflammatory responses in a GWI mouse model. We have animal models with genetic modifications influencing cell-mediated and/or humoral immunity (e.g., mice "knocked out" for CD40), and if required, they will be available for the proposed studies. Based on our preliminary data, we have identified NFkB inhibitors as potential pharmacological modulators of the GW agent-induced inflammatory response and syk inhibitors as modulators of GW agent-induced B-cell and T-cell activation and the peripheral humoral response. As part of the proposed work, we will test these compounds as potential therapies in our GWI mouse model. From our IPA analyses of the proteomic data, we will identify additional biofunctions, which will be targeted using specific inhibitors of key regulatory proteins within those pathways. Innovation: There are no approved treatments for the cognitive impairment associated with GWI due to the complexity of this illness and a lack of knowledge about its etiology. The use of a proteomic approach will allow us to understand the functional and combinatorial protein changes that accompany immune/inflammatory response in GWI and to identify their neurobehavioral and neuropathological correlates. Impact: The work proposed will identify the immune/inflammatory response associated biofunctions with the cognition in GWI, which will then be targeted to identify potential therapies for treating this debilitating condition that affects, 20 years later, nearly 200,000 soldiers from the 1990-1991 Gulf War. Hence, understanding the molecular events that contribute to the pathogenesis, together with the identification of targets for pharmacological development, will potentially have an enormous impact on treatment and management of this illness.
Exposure to environmental toxins increases Alzheimer’s risk: study
January 20 2016 09:52 PM
Dr. Renee Richer
Studies about the Gulf War toxins by a former professor of Weill Cornell Medicine-Qatar (WCM-Q) have been highlighted in a new research breakthrough that prove environmental toxins may increase the risk of Alzheimer’s disease and Amyotrophic Lateral Sclerosis (ALS).
Former WCM-Q professor, Dr. Renee Richer, now of the University of Wisconsin-Marinette, analysed why one decade after returning from deployment to the First Gulf War, some US military personnel started coming down with the unusual paralytic symptoms of ALS at twice the incidence rate of those who received the same training but were not dispatched to the Gulf.
During her eight years as associate professor of Biology at WCM-Q, Dr Richer found that flat plains in the Gulf deserts are covered with dried cyanobacteria crusts waiting for winter rain to complete their life cycle. Cyanobacteria are the oldest living bacteria on earth, that normally live in water, but can thrive in a multitude of environments. If these soil crusts in the desert are disturbed by an off-road military vehicle or tank tread, the dust given off contains a neurotoxin, called BMAA.
A new study published this week by the Royal Society of London in the biological research journal, Proceedings of the Royal Society B, indicates that chronic exposure to BMAA may increase risk of neurodegenerative illness.
Brain tangles and amyloid deposits are the hallmarks of both Alzheimer’s disease and also of an unusual illness suffered by villagers on the Pacific Island of Guam. Pacific Islanders with this unusual condition suffer from dementia and symptoms similar to Alzheimer’s disease, ALS and Parkinson’s disease. The diet of the Chamorro people is contaminated by BMAA.
Scientists have long suspected a link between neurodegenerative disease and BMAA and also in the brains of people suffering from ALS and Alzheimer’s disease. But this week’s announcement provides a new level of proof.
“Our findings show that chronic exposure to BMAA can trigger Alzheimer’s-like brain tangles and amyloid deposits,” said Paul Alan Cox, an ethnobotanist at the Institute for EthnoMedicine and lead author of the study. “As far as we are aware, this is the first time researchers have been able to successfully replicate brain tangles and amyloid deposits in an animal model through exposure to an environmental toxin.”
When Dr. Richer first met Cox in the deserts of Qatar during her time at WCM-Q, she was intrigued with his hypothesis of inhaled BMAA-dust as an environmental trigger for ALS. Together with her WCM-Q postdoctoral student, Dr. Aspa Chatziefthimou, Dr. Richer began an extensive survey of toxins in desert crusts.
“We were astonished that up to 87% of the deserts of Qatar are covered with cyanobacterial crusts,” Dr. Richer said. “Even more concerning was our discovery that the toxins they produce accumulate in the desert soil beneath them.”
In the research findings announced this week, scientists conducted two separate experiments on vervet monkeys.In the first experiment, vervets were fed fruit that was dosed with BMAA for 140 days.A second experiment was conducted, which added a BMAA dose closer to the amount the Chamorro villagers would be exposed to over a lifetime.
“This study takes a leap forward in showing causality—that BMAA causes disease,” said Deborah Mash, director of the University of Miami Brain Endowment Bank and co-author of the study.
The discoveries are important because they have implications for populations in areas where cyanobacteria blooms are common, such as Qatar, the wider Gulf region and also Marinette, Wisconsin, which lies on the shore of Lake Michigan and close to Lake Winnebago.
Dr. Richer added: “These new results suggest that water quality may be an important issue to now address in terms of neurological health."
The parts of the research conducted by Dr. Richer in Qatar were made possible by a National Priorities Research Programme grant from the Qatar National Research Fund, a member of Qatar Foundation.
Veterans of the 1990-1991 Gulf War have been reported to have an increased incidence of amyotrophic lateral sclerosis (ALS) compared to personnel who were not deployed. An excess of ALS cases was diagnosed in Gulf War veterans younger than 45 years of age. Increased ALS among Gulf War veterans appears to be an outbreak time-limited to the decade following the Gulf War. Seeking to identify biologically plausible environmental exposures, we have focused on inhalation of cyanobacteria and cyanotoxins carried by dust in the Gulf region, particularly Qatar. Cyanobacterial crusts and mats are widespread in the deserts of Qatar, occupying up to 56% of the available area in some microhabitats. These cyanobacterial crusts, which help bind the desert sands, are dormant throughout most of the year, but during brief spring rains actively photosynthesize. When disturbed by vehicular traffic or other military activities, the dried crusts and mats can produce significant dust. Using HPLC/FD, an amino acid analyzer, UPLC/MS, and triple quadrupole LC/MS/MS we find that the dried crusts and mats contain neurotoxic cyanobacterial toxins, including beta-N-methylamino-L-alanine (BMAA) and 2,4 diaminobutyric acid (DAB). If dust containing cyanobacteria is inhaled, significant exposure to BMAA and other cyanotoxins may occur. We suggest that inhalation of BMAA, DAB, and other aerosolized cyanotoxins may constitute a significant risk factor for the development of ALS and other neurodegenerative diseases.
[PubMed - indexed for MEDLINE]
"Intriguingly, the identification of cyanobacteria in desert sand suggests that inhalation of BMAA may explain the higher incidence of ALS in United States troops returning from Iraq after serving in the First Gulf War of 1991 (Cox et al., 2009). BMAA has also been shown to accumulate in the brains of Alzheimer's disease (AD) sufferers in Canada, but not in patients with non-neurological "
"In this form, BMAA has been shown to cause complete cell death of corticol neurons as measured by LDH release 20 h after exposure to 3 mM BMAA for as little as 1 h (Weiss and Choi, 1988). However, given the long lag-time (10e15 years) from exposure to BMAA and the onset of ALS/PDC (Cox et al., 2009) it seems likely a chronic effect is also at play. This was also supported by a peak in cases of ALS in soldiers who had been deployed during the Gulf War some ten years after they returned to the US (Cox et al., 2009; Veterans, 2008). "
ABSTRACT: Since plants lack the ability to remove themselves from sites of predation, they have evolved alternative defences to keep predators at bay. Mechanical defences include camouflage, the addition of thorns and spikes as well as growing in locations not easily accessed by herbivores. Chemical defences include the synthesis of compounds that are toxic to predatory species and can also inhibit or retard the growth of other plant species in a process termed “allelopathy”. Amongst these chemicals is a reservoir of non-protein amino acids that mediate toxicity. These non-protein amino acids can be charged by tRNA synthetases and subsequently mis-incorporated into nascent polypeptides, resulting in aberrant, dysfunctional proteins that can be toxic to predators. Some species such as the bruchid beetle Bruchus rufimanus (Chrysomelidae), which feeds on the jack bean Canavalia ensiformis (Fabaceae) plant, have evolved advanced tRNA synthetases that are able to discriminate between protein and non-protein amino acids, thus remain unaffected. Other species including livestock and humans that do not possess such selectivity are susceptible to plant toxins. In this brief review we discuss the mechanisms of action and consequences of exposure to plant-derived non-protein amino acids with a focus on those derived from plants from arid environments.
Cyanobacteria and BMAA exposure from desert dust: a possible link to sporadic ALS among Gulf War veterans. Amyotroph Lateral Scler 10(S2):109-117. Available from: https://www.researchgate.net/publication/51439971_Cyanobacteria_and_BMAA_exposure_from_desert_dust_a_possible_link_to_sporadic_ALS_among_Gulf_War_veterans_Amyotroph_Lateral_Scler_10S2109-117 [accessed Jan 24, 2016].
Full-text · Article · Jan 2014 · Journal of Arid Environments
There have been few studies concerning cyanotoxins in desert environments, compared with the multitude of studies of cyanotoxins in aquatic environments. However, cyanobacteria are important primary producers in desert environments, where after seasonal rains they can grow rapidly both stabilising and fertilising arid habitats. Samples of cyanobacteria from wadis - dry, ephemeral river beds - and sabkha - supertidal salt flats - in Qatar were analysed for the presence of microcystins, nodularin, anatoxin-a, cylindrospermopsin and anatoxin-a(S). Microcystins were detected by HPLC-PDA and ELISA at concentrations between 1.5 and 53.7ngg(-1) dry wt of crust. PCR products for the mycD gene for microcystin biosynthesis were detected after amplification of DNA from desert crust samples at two out of three sample sites. The presence of anatoxin-a(S) was also indicated by acetylcholine esterase inhibition assay. As a function of area of desert crust, microcystin concentrations were between 3 and 56μgm(-2). Based on the concentration of microcystins detected in crust, with reference to the published inhalation NOAEL and LOAEL values via nasal spray inhalation of purified microcystin-LR in aqueous solution, and the amount of dust potentially inhaled by a person from these dried crusts, the dose of microcystins could exceed a calculated TDI value of 1-2ngkg(-1)day(-1) for an average adult. The presence of microcystins, and potentially of anatoxin-a(S), in desert crusts has important implications for human health. Further studies are required to monitor desert dust storms for the presence of cyanotoxins. An understanding of the risks of inhaling particles containing cyanotoxins is also warranted.
Neurofibrillary tangles (NFT) and β-amyloid plaques are the neurological hallmarks of both Alzheimer's disease and an unusual paralytic illness suffered by Chamorro villagers on the Pacific island of Guam. Many Chamorros with the disease suffer dementia, and in some villages one-quarter of the adults perished from the disease. Like Alzheimer's, the causal factors of Guamanian amyotrophic lateral sclerosis/parkinsonism dementia complex (ALS/PDC) are poorly understood. In replicated experiments, we found that chronic dietary exposure to a cyanobacterial toxin present in the traditional Chamorro diet, β-N-methylamino-L-alanine (BMAA), triggers the formation of both NFT and β-amyloid deposits similar in structure and density to those found in brain tissues of Chamorros who died with ALS/PDC. Vervets (Chlorocebus sabaeus) fed for 140 days with BMAA-dosed fruit developed NFT and sparse β-amyloid deposits in the brain. Co-administration of the dietary amino acid L-serine with L-BMAA significantly reduced the density of NFT. These findings indicate that while chronic exposure to the environmental toxin BMAA can trigger neurodegeneration in vulnerable individuals, increasing the amount of L-serine in the diet can reduce the risk.
(a) Toxins and neurodegenerative illness
The relationship between environmental toxins and neurological disease has been of interest since residents of Minamata Bay, Japan, were sickened by chronic dietary exposure to methyl-mercury-laden fish. Parkinson's disease (PD) has been linked to rotenone or paraquat exposures in agricultural workers . PD also was diagnosed in ‘frozen addicts’, users of a recreational drug contaminated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine . Exposures to pesticides, metals, solvents and certain types of volatile anaesthetics have additionally been linked to PD, while exposures to lead, mercury and pesticides have been suggested as risk factors for amyotrophic lateral sclerosis (ALS) . However, the role of naturally occurring environmental toxins in progressive neurodegenerative disease has not been extensively studied.
(b) A paralytic disease among Pacific Islanders
In the 1950s, US Army physicians described a puzzling ALS-like disease among the indigenous Chamorro villagers of Guam . In the 1960s, amyotrophic lateral sclerosis/Parkinsonism dementia complex (ALS/PDC) was described based on histopathology and clinical symptoms which resemble aspects of Alzheimer's disease (AD), ALS and PD. Neurofibrillary tangles (NFT) in the brains of individuals with ALS/PDC have similar immunohistology and structure as those found in the brains of AD patients but are biochemically and regionally more heterogeneous [4,5]. Many afflicted villagers suffered from dementia. Histopathology of Chamorros who died prior to manifesting clinical symptoms of ALS/PDC suggests that depositions in the brain pre-dated clinical onset . No clear pattern of inheritance for the disease has been ascertained . Since even outsiders who adopted a Chamorro lifestyle experienced an increased risk of illness , a common environmental exposure seemed likely. Determining the nature of the toxin, however, was difficult due to a significant delay between exposure and clinical symptoms, extending years or even decades .
(c) BMAA: a neurotoxic amino acid in cycad seeds
In the 1960s, consumption of flour made from the gametophyte of cycad seeds (Cycas micronesica Hill) was proposed as a cause of the disease. Interest increased when a novel neurotoxic amino acid, L-BMAA, was isolated from cycad seeds by Bell . In the 1980s, BMAA fed to macaques was found to cause acute neurological symptoms , a finding that was discounted when it was argued that an equivalent human dose would require the consumption of unreasonable amounts of cycad seed flour . BMAA was subsequently identified as a cyanobacterial product . The toxin is biomagnified in flying foxes, which are eaten by Chamorros . Equally important was the discovery that a majority of BMAA in cycad seeds binds to proteins and cannot be released by washing with water, but only on hydrolysis, suggesting that BMAA doses ingested by the Chamorros had been previously underestimated [14,15]. Evidence continued to build for the link between BMAA and neurodegenerative disease with respect to cyanobacterial exposure and epidemiology [15–20].
A key missing puzzle piece has been experimental evidence that chronic dietary exposure to BMAA triggers neuropathological changes consistent with ALS/PDC, which presumably should occur prior to the onset of clinical symptoms. It is now known that in vivo BMAA exposure generates fibril formation and cognitive deficits in rodents , although some earlier animal studies that focused on acute rather than chronic exposure were inconclusive . This finding suggests that chronic BMAA exposure more closely models early disease.
(d) BMAA in the Chamorro diet
BMAA is produced by symbiotic cyanobacteria of the genus Nostocharboured in specialized cycad roots emergent in the leaf litter above the soil. BMAA accumulates in the gametophytes of cycad seeds, which, after washing, are used by villagers to prepare tortilla flour, dumplings and to thicken soups and stews. Animals, including flying foxes, feral deer and pigs that feed on cycad seeds, which in turn are consumed by villagers, also accumulate BMAA in their tissues . BMAA is biomagnified up to 10 000-fold from its production by cyanobacteria to its concentration in volant mammals [12,13,15].
(e) BMAA exposures beyond Guam
Diverse taxa of cyanobacteria produce BMAA [23,24], which is biomagnified in some marine ecosystems, accumulating in sharks, bottom-dwelling fish and shellfish. BMAA also occurs in cyanobacterial soil crusts . BMAA exposure through inhalation of desert dust has been suggested as triggering the increased incidence of ALS a decade subsequent to the deployment of military personnel in Operation Desert Storm . Similarly, inhalation of aerosolized BMAA from wave break has been proposed to explain the increased risk of ALS in individuals who live near lakes with persistent cyanobacterial blooms [18,19]. Exposure through ingestion of drinking water has not been ruled out . Maternal exposures to BMAA may also increase the risk of ALS in neonates later in their life [20,21].
(f) Mechanisms of BMAA-induced neurodegeneration
Through activation of metabotropic glutamate receptors such as mGluR5  or ionotropic glutamate receptors including the N-methyl-D-aspartate receptor, kainate or the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor, BMAA selectively kills subpopulations of NADPH-diaphorase-positive motor neurons . It is toxic to glial cells  and causes motor neuron damage and astrogliosis in the ventral horn . BMAA potentiates different neurotoxic insults including methyl mercury , which co-occurs in some fish. BMAA rapidly crosses the blood–brain barrier, where it is captured by the central nervous system (CNS) in a time period consistent with protein misincorporation . BMAA can be mistaken by cellular machinery for L-serine and be misincorporated into proteins, leading to protein misfolding, aggregation and subsequent apoptosis . Misincorporation of BMAA into proteins has been proposed as a mechanism for bioaccumulation as well as a mechanism for slow release of BMAA within the CNS over years depending on rates of protein turnover . Misincorporation of even the 20 canonical amino acids at error rates as low as 1/10 000 can lead to neurodegeneration in laboratory animals . BMAA exposure results in hyperphosphorylated tau, possibly by decreasing activity of protein phosphatase 2A (PP2A) through activation of the mGluR5 receptor and subsequent dissociation of the catalytic subunit PP2Ac . In Chamorro ALS/PDC brains, PP2A activity is significantly decreased, resulting in a significant increase in hyperphosphorylated tau .
(g) Producing an animal model of BMAA-induced neuropathology
The occurrence of BMAA in post-mortem brain tissues of Chamorro ALS/PDC patients but generally not in non-Chamorro control patients suggests that chronic dietary exposure to BMAA is an environmental risk factor for ALS/PDC [12,15,37]. To satisfy Koch's postulates of disease causation , it is necessary to show that chronic exposure to BMAA causes healthy individuals to develop neurodegenerative disease and that BMAA can be re-isolated from those individuals in which neurodegeneration has been induced.
NFT and β-amyloid deposits have not both been produced in human neuronal cell culture, so in vivo experiments are necessary. However, no animal species other than humans is known to develop ALS/PDC or AD. Furthermore, NFT and β-amyloid deposits have not both previously been produced in any single animal model, with the exception of a triple transgenic mouse model  in which the structure and density of the NFT significantly differ from the human condition (K. Iqbal 2015, personal communication). Some non-human primates including squirrel monkeys, chimpanzees, gorillas and orangutans of great age as well as lemurs develop senile plaques that are immunopositive for β-amyloid, and a single 41-year-old chimpanzee was found to produce paired helical tau filaments . Vervets are known to accumulate vascular β-amyloid deposits with age, but not NFT and other tau inclusions . We therefore decided to chronically expose vervets to BMAA for an extended period and to examine their brain tissues for tau inclusions and amyloid deposition consistent with ALS/PDC pathology. Since matching the duration of chronic exposure to the years—even decades—required for Chamorros to develop ALS/PDC  is unfeasible, we shortened chronic exposure to BMAA to 140 days. Since L-serine has been found to prevent misincorporation of BMAA and apoptosis in human neuronal cell culture , we added a cohort of vervets which daily received equal amounts of BMAA and L-serine. Finally, to increase statistical rigour, we replicated the first experiment. We used an oral dose (210 mg kg−1 d−1) that previous investigators found using gavage could be tolerated by macaques  and in the second experiment added a cohort of vervets with a 10-fold dose reduction (21 mg kg−1 d−1) to produce a cumulative BMAA exposure closer to total lifetime Chamorro exposure.
2. Material and methods
(a) In vivo studies
The vervets studied in this report were housed in groups in large outdoor enclosures at the Behavioural Science Foundation (BSF) in St Kitts, West Indies. The BSF is a fully accredited biomedical research facility with approvals from the Canadian Council on Animal Care. The animal use protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of BSF and McGill University (Quebec, Canada). The vervet low-protein diet was supplemented with fruit dosed with L-BMAA or other test substances. Doses were prepared at the Institute for Ethnomedicine, Jackson Hole, using a Mettler Toledo balance with a Quantos automated powder dispensing module at a tolerance of ±0.1% of target dose. A small cavity was made in each piece of fruit, and the test substance was placed inside. In the first experiment, 16 juvenile vervets were presented daily with a dosed piece of fruit, approximating a 210 mg kg−1 d−1 dose based on average weight (3.1 kg) of the vervets. One cohort of four was fed daily for 140 days a piece of fruit containing 651 mg of L-BMAA, a second cohort was fed fruit with 651 mg of L-serine, a third cohort was fed fruit dosed with 651 mg of L-BMAA plus 651 mg of L-serine, and a fourth control cohort received a piece of fruit dosed with 651 mg of rice flour as a placebo.
For both the original experiment on the younger vervets and the replication experiment on the adult vervets, 14 regions of each vervet's brain were investigated for neuropathology with immunostaining for tau AT8 and β-amyloid 1–42. Three serial sections were studied on a 3 × 4 grid with 100× magnification. In the replication experiment, cohorts of eight adult vervets were fed dosed fruit for 140 days. These 7-year-old vervets, which were colony-born, were somewhat larger than the younger vervets in the first experiment, so to approximate a 210 mg kg−1 d−1 dose for these larger animals, the L-BMAA dose was increased from the first experiment in order to adjust for weight. A cohort of vervets was added in which the L-BMAA dose was reduced 10-fold to approximate 21 mg kg−1 d−1 to be closer to a lifetime Chamorro exposure. Thus in the replication experiment one cohort of eight vervets received daily 987 mg of L-BMAA, a second cohort received 98.7 mg of L-BMAA, a third cohort received 987 mg of L-BMAA and 987 mg of L-serine, and a fourth control cohort received 987 mg of rice flour. Periodic blood serum and cerebral spinal fluid (CSF) samples were taken under ketamine anaesthesia to confirm BMAA exposures in the vervets and absence of BMAA exposure in the controls.
In both experiments, one hemisphere of each vervet brain was frozen. The other hemisphere was immersion fixed in buffered formalin for histopathology. This hemisphere was freeze-sectioned at 40 µm and an adjacent series of coronal sections were processed with antibody stains using the MultiBrain Services of NeuroScience Associates (Tennessee, USA). In both experiments, adjacent sections were stained with AT8 immunohistochemistry (IHC) stain with a thionine Nissl counterstain for hyperphosphorylated tau and β-amyloid (1–42) IHC stain for β-amyloid deposits. In the first experiment, NFT (100× magnification) and β-amyloid deposits (10× magnification) were identified from blinded review of the stained sections and were quantified using manual counts in three sections in series from non-overlapping brain regions. In the second experiment, stained sections were examined using automated images prepared with a TissueScope LE (Huron Digital Pathology, Ontario, Canada). Stained serial sections were digitally scanned at 20× using a 350 µm2 grid for NFT and β-amyloid deposits. Thioflavine-S with a thionine Nissl counterstain was used to confirm the presence of NFT and plaques. The regions of interest (ROI) for each case were initially drawn on the Nissl section and the ROI was mapped to the immunostained slides. The ROI was marked with an array tool to identify regional boundaries of the amygdala, hippocampus, entorhinal, frontal, temporal, motor, occipital and cingulate cortices. Digital images were measured using NIH IMAGE J64 software (1.44) converted from RGB colour to 8-bit followed by applying a threshold to eliminate non-specific background staining. After threshold correction, the images were converted to binary allowing for quantification of pathological features detected above background. The high-contrast images were highly suited for digital quantification of pixel counts. Representative sections were examined in parallel to validate the digital measurements by comparison to manually derived β-amyloid deposits and NFT counts .
(c) Analytical chemistry
Blinded samples of brain tissue, blood serum and CSF were analysed for BMAA content using triple quadrupole tandem mass spectrometry (LC-MS/MS) with a precolumn 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) derivatization using the validated method determined by the Association of Analytical Communities AOAC International . Negative controls included matrix blanks from control vervets with no detectable BMAA, AQC-derivatized blanks, internal standards and solvent blanks (HCl, TCA). Product ion analysis of BMAA used m/z 459 as the precursor ion for collision-induced dissociation (CID) and two-step mass filtering was performed during selective reaction monitoring of BMAA after CID in the second quadrupole, monitoring the following transitions: m/z 459 to 119 CE 25 eV; 459 to 289 CE 23 eV; 459 to 171 CE 45 eV. The resultant product ions were detected after passing the third quadrupole and their relative abundances were quantified. BMAA was analytically distinguished from its isomers using m/z 459 to 188 CE 38 eV (2,4-diaminobutyric acid); 459 to 214 CE 35 eV (N-(2-aminoethyl)glycine); 459 to 258 CE 36 eV (BMAA) [44,45]. Double ionized AQC-derivatized BMAA was also monitored with a precursor ion of m/z 230 and a product ion of 171 CE 27 eV. Additionally, the following amino acids were monitored: single derivatized lysine m/z 317, double derivatized lysine m/z 487, leucine m/z 302, serine m/z 276 and an internal standard (β-N-methyl-d3-amino-DL-alanine-15N2) with a precursor ion of m/z 464, and product ions m/z 124 CE 25 eV, 171 CE 45 eV, 259 CE 36 eV and 294 CE 23 eV. BMAA tissue concentrations were determined relative to concentration curves run daily in spiked matrix samples from a control animal using β-N-methyl-d3-amino-DL-alanine-15N2as an internal standard. Sample preparation techniques and complete analytical methods appear in the electronic supplementary material.
(d) Statistical analysis
Because of the small sample size inherent in the experimental design, and to avoid assumptions of normal distribution of resultant data, we used non-parametric methods to compare medians. To determine if chronic dietary exposure to BMAA results in a greater density of NFT, the hypotheses H0 = there is no difference in median NFT density between treatment groups and H1 = there is a difference in median NFT density between treatment groups were evaluated with a Kruskal–Wallis H-test, a non-parametric analogue of an analysis of variance. A Jonckheere–Terpstra trend test was used to test the hypotheses H0 = median NFT density is independent of BMAA dose versus H1 = median NFT density increases with BMAA dose. Different hypotheses comparing median amounts of BMAA per cohort treatment group were evaluated with a Kruskal–Wallis H-test: H0 = there is no difference in median BMAA concentrations between treatment groups within plasma, brain or CSF, and H1 = there is a difference between median BMAA concentrations between treatment groups in plasma, brain or CSF. (In this case, medians of protein-bound BMAA were used for plasma and brain samples, while medians of total BMAA content were used for the CSF; see methods in electronic supplementary material.) Other hypotheses evaluated with a Kruskal–Wallis H-test were H0 = there is no difference in the median ratios of protein to total BMAA concentrations between treatment groups in plasma or brain, and H1 = there is a difference in the median ratios of protein to total BMAA concentrations between treatment groups in plasma or brain. To determine if chronic dietary exposure to BMAA is related to the presence of β-amyloid deposits, two alternative hypotheses were evaluated with a χ2 test: H0 = there is no difference between treatment types on the number of vervets that develop β-amyloid deposits, and H1 = there is a difference between treatment types on the number of vervets that develop β-amyloid deposits. Spearman's rank correlation coefficients were calculated to evaluate H0 = there is no relationship between protein-bound or protein-bound/total ratio of BMAA concentrations in the brain and NFT counts, and H1 = there is a relationship between protein-bound or protein-bound/total ratio of BMAA concentrations in the brain and NFT counts.
In the first experiment, AT8-positive tangles and neuronal processes, as well as sparse β-amyloid plaque-like deposits, were observed in brain tissues of the L-BMAA-dosed vervets. AT8-positive NFT were observed in the perirhinal and entorhinal cortices, amygdala (paralaminar nucleus), motor cortex, frontal cortex, temporopolar cortex and occipital cortex of the BMAA-fed animals (figures 1⇓–3). In contrast, no AT8 immunopositive inclusions were visualized in the hippocampus (CA1 or dentate gyrus). Sparse immunopositive β-amyloid deposits were observed primarily in the frontal, temporal and motor cortices. In the first experiment, the L-serine treated cohort and the control cohort of four vervets were generally negative for tau AT8 and β-amyloid 1–42 neuropathology, while there was an 80–90% reduction of NFT and plaques in the cohort fed equal amounts of L-BMAA and L-serine; these results will be published elsewhere.
Neuropathology of vervet brain tissue with chronic dietary BMAA exposures; a comparison of thioflavine-S and β-amyloid (1–42) immunoreactivity. (a) Thioflavine-S stained cells and neuropil threads in the motor cortex; scale bar, 150 µm. (b) Intraneuronal β-amyloid accumulation in neurons in motor cortex. (c) Vervet extracellular thioflavine-S deposits in the frontal cortex. (d) Localized β-amyloid immunostained neocortical deposits in vervet brains. (e) Thioflavine-S positive senile plaques and tangles in human AD temporal cortex. (f) β-amyloid senile plaques in human temporal cortex of AD patient (86-year-old male; 400× magnification). Human brain sections from AD patients were run as reference controls.
Microscopic pathology of chronic dietary L-BMAA exposures in vervets. Representative low-power images (5× magnification) of hyperphosphorylated tau (AT8) immunostained coronal hemisections from control (a,c) and L-BMAA-fed vervets (b,d). AT8 immunostaining is seen in the amygdala (Amy), entorhinal (EC), perirhinal (PrC), primary motor (M1) and temporal cortices of L-BMAA-fed vervets. Higher-power images show predominant tau AT8 staining in superficial cortical layers II and III with more robust staining over the entorhinal and perirhinal cortices (25× magnification) (d). Microscopic images (original magnification ×120) show NFT in vervets fed L-BMAA. Tangle-like tau aggregates are seen in the temporal gyrus (e,f). Dense intracellular tau immunolabelling (g–i) and extracellular deposits (j,k) were seen in the parahippocampal gyrus. Abundant neuropil threads, tangles and dystrophic neuronal processes are observed in layers II and III of the perirhinal cortex (I, high-power images shown in l,m) and the paralaminar nucleus of the amygdala (n). Tau plaques were seen in L-BMAA-fed vervets ranging from large and diffuse (o) to small dense aggregates (p). ac, anterior commissure; Amy, amygdala; Bmc, basal nucleus of the amygdala, magnocellular region; Bpc, basal nucleus of the amygdala, parvicellular subdivision; Cd, caudate; cgs, cingulate gyrus sulcus; EC, entorhinal cortex; L, lateral nucleus of the amygdala; LF, lateral fissure; M1, primary motor cortex; PrC, perirhinal cortex; PL, paralaminar nucleus; Pu, putamen; STS, superior temporal sulcus.
Median counts for density of AT8 IHC positive staining inclusions plus NFT per brain area by treatment type. Each horizontal surface represents the median of an eight-vervet cohort statistically significant for dose using the Jonckheere–Terpstra trend test. (a) Brain regions in which AT8 IHC positive density counts from the 210 mg kg−1 d−1 BMAA treatment are greatest compared with other treatment types. (b) Brain regions in which AT8 IHC positive density counts from the 210 mg kg−1 d−1 BMAA plus 210 mg kg−1d−1L-serine (high+SER) treatment is less than low-dose (low) BMAA (entorhinal cortex posterior) or in which low-dose NFT density is similar to controls (all other brain areas).
Median AT8 tau NFT quantification in brain tissues of vervets. Each treatment cohort consisted of eight vervets.
In the replication experiment, chronic L-BMAA exposures for 140 days again led to hyperphosphorylated tau deposits and NFT formation in all BMAA-fed vervets (figure 2). Median NFT density differed significantly between treatment groups (Kruskal–Wallis H statistic = 16.4, p < 0.001). Furthermore, there was a clear dose relationship between chronic dietary exposure to L-BMAA and density of NFT (Jonckheere–Terpstra trend test, Z = 4.4, p < 0.00001). NFT were abundant in vervets with chronic dietary exposure to BMAA in the superior frontal, temporopolar (dorsal and ventral), perirhinal, occipital and entorhinal (anterior and posterior) cortices, and in the amygdala (figure 3). In these brain areas, there was a highly significant dose relationship between increasing dietary exposure to L-BMAA and NFT density (Jonckheere–Terpstra trend test, Z-scores for the 14 brain regions range between 3.13 and 4.87, p < 0.001–0.00001; figure 3). The regional differences in NFT and β-amyloid deposit counts in the brain areas examined were profound. For example, in the occipital cortex, other than controls, vervets in the low-dose treatment had the lowest median count (65) NFT density, while the median density (136) of the high-dose BMAA cohort was more than twice the low-dose NFT density. Co-administration of L-serine with high-dose BMAA significantly reduced median NFT density (124). This reduction in NFT induced by L-serine occurred in all measured areas of the brain. Supplementing the diet with L-serine resulted in more than a 50% NFT reduction in median NFT densities within five brain regions: temporal (dorsal and ventral), primary motor, entorhinal (posterior) and insula cortices. In the perirhinal cortex, amygdala and anterior cingulate gyrus, L-serine reduced NFT densities more than 35%.
A highly significant (p < 0.00001) dose relationship between chronic dietary exposure to L-BMAA and NFT density was also found in other brain regions, but no profound differences in NFT densities were found between the low-dose and control cohorts in these regions which included the dentate gyrus, substantia nigra, caudate nucleus, anterior cingulate gyrus, primary motor cortex and the insula cortex (figure 3b). Finally, dose relationships by treatment group were also significant in the entorhinal cortex, but this brain area differed from the others in that co-administration of L-serine led to an NFT density not only lower than high-dose BMAA, but also lower than low-dose BMAA (figure 3b).
Chronic dietary exposure to BMAA significantly increased the likelihood of a vervet developing β-amyloid deposits (χ2 = 15, p < 0.01). One of the eight low-dose BMAA vervets, three of the eight high-dose BMAA vervets and two of the eight high-dose BMAA plus L-serine vervets had β-amyloid deposits. These β-amyloid deposits were diffuse and sparse in distribution (figure 1). β-amyloid deposits were not found in any of the control vervets.
The relationship between NFT counts and measured concentrations of BMAA in the occipital cortex was also of interest. BMAA could not be detected in control vervets or baseline samples using LC-MS/MS. Protein-bound BMAA occurred in brain tissues of individual L-BMAA-fed vervets at concentrations between 0.24 and 2.2 µg mg−1 (see electronic supplementary material), similar to Chamorro ALS/PDC brain tissues (median = 0.6 µg mg−1, range = 0.2–1.2 µg mg−1) , and was detected in blood plasma and CSF. Even within the low-dose cohorts, protein-bound BMAA within vervet brain tissues (0.24–0.78 µg mg−1) reached concentrations consistent with the Guam disease.
There was no significant difference in protein-bound BMAA concentrations in blood plasma between treatment groups, but there were significant differences for BMAA concentrations for brain and CSF samples (Kruskal–Wallis H statistics (corrected for ties) of 8.69 and 9.09 (p < 0.05). There was no significant difference in the ratio of protein to total BMAA concentrations in brain, but there was in blood plasma (H statistic = 13.24, p < 0.01). Finally, no significant relationship was found between protein-bound BMAA and NFT density as well as in the protein-bound/total BMAA ratio and NFT density in vervet brains as determined by Spearman's rank correlation coefficients.
(a) L-BMAA triggers neuropathology
Chronic dietary exposure to L-BMAA results in the formation of NFT and β-amyloid deposits in a clear dose relationship. Other protein inclusions similar to those found in brain tissues from Chamorros who died with ALS/PDC were also found. Chronic dietary exposure to L-BMAA triggered tauopathies in all BMAA-dosed vervets including those at the low-dose treatment but NFT densities varied between brain regions. Concentrations of BMAA in vervet brains fell within the range measured in post-mortem brain tissues of Chamorros who died with ALS/PDC confirming BMAA exposures in the vervets that are clinically relevant. Furthermore, the regional densities of NFT are similar in both the Chamorros and L-BMAA-fed vervets [47,48].
There was a significant dose relationship between BMAA and NFT density in all affected regions of the brains (figure 3). The distribution of NFT and their relationship to dose exposure in the temporal lobe is similar to Braak 1 early stage AD pathology  (figure 3). Consistent with the neuropathology of preclinical AD, no profound clinical symptoms were observed in any of these vervets in the two experiments. Specific immunological methods (AT8) permit evaluation of neuronal changes before the actual formation of NFT and neuropil threads (figure 2). In vervets with chronic dietary exposure to BMAA, we observed changes in the transentorhinal region of the temporal lobe, but none in Ammon's horn of the hippocampus. Extensively distributed NFT formations with gliosis characterize ALS/PDC. Dementia in these cases is attributable to tangles and neuronal dropout in the neocortex, resembling the pattern reported for AD, but far more widely distributed . NFT in ALS/PDC brain tissues stain positively with antibodies to hyperphosphorylated tau protein . Thus, the distribution of AT8-positive tangles following chronic dietary BMAA exposure in vervets is similar to the histopathology reported previously in ALS/PDC.
The paucity of clinical symptoms in the BMAA-fed vervets corresponds to the finding of NFT in 5/29 asymptomatic Chamorro patients who died without ALS or PD being recognized clinically . Furthermore, β-amyloid plaques have been detected in human ALS/PDC patients who remain cognitively intact . The fact that BMAA-dosed vervets produced NFT and rare β-amyloid deposits in both experiments supports the theory that BMAA in the traditional diet is a cause of the Chamorro disease.
(b) Neurofibrillary tangles and β-amyloid deposit formation
Normal microtubules which serve as the pathways for anterograde and retrograde transport within the neuron unravel as the soluble monomeric tau proteins become hyperphosphorylated and detach from the microtubule. These hyperphosphorylated tau proteins form paired helical filaments, and thence aggregates leading to NFT (figure 4a). In the β-amyloid plaque pathway (figure 4b), amyloid precursor protein (APP) is cleaved by β-secretase and γ-secretase into β-amyloid fragments. Although the initial confirmation of Aβ-42 is an α-helix, transformation to a β-pleated sheet conformation is a necessary step in plaque formation. When the β-pleated sheets oligomerize, they can eventually join into polymers which form plaques (figure 4b). Our data suggest that chronic dietary exposure to BMAA triggers both the NFT and β-amyloid pathways.
Theoretical pathways of development of ALS/PDC and AD neuropathology from chronic dietary BMAA exposure. (a) Tau proteins which bind microtubules become hyperphosphorylated, leading to dissociation of hyperphosphorylated tau fragments. These form paired helical filaments, leading to the formation of neurofibrillary tangles. (b) The APP is cleaved, producing β-amyloid (Aβ-42) fragments which are in an α-helix conformation. These change to a β-pleated sheet conformation, oligomerize, forming amyloid plaques.
(c) Neuroprotective mechanisms of L-serine
Larger BMAA doses resulted in increased protein-bound BMAA concentrations in the brain. However, although L-serine reduced NFT density, it did not alter the ratio of protein-bound to total BMAA, which may be physiologically invariant. Possible neuroprotective mechanisms of L-serine include prevention of BMAA misincorporation in specific proteins involved in NFT formation. Misincorporation at rates as low as 1/10 000 can result in neurodegeneration , but such levels may be below our ability to differentiate. There may also be additional neuroprotective mechanisms other than prevention of misincorporation.
(d) Implications of chronic BMAA exposure for neurodegenerative disease
BMAA-producing cyanobacteria occur globally, perhaps causing similar neuropathologies. Our finding that all of the low-dose vervets developed tauopathies with NFT has implications for human health. BMAA may serve as an environmental trigger for some forms of other neurodegenerative illnesses including sporadic ALS and AD. In human beings, increasing age is a risk factor for ALS, AD and PD. We have initiated experiments to determine if chronic dietary exposures of aged vervets to BMAA results in more profound histopathology.
We have sponsored FDA-approved human clinical trials (ClinicalTrials.gov Identifier NCT01835782) to determine if L-serine is a safe and efficacious treatment to reduce disease progression in ALS patients. We hope to initiate human clinical trials of L-serine for mild cognitive impairment and early onset AD in the near future.
In conclusion, Koch's postulates  have been satisfied with respect to establishing chronic dietary exposure to BMAA as a cause of a neurodegenerative illness: (i) BMAA has been identified in post-mortem brain tissue from ALS/PDC patients from Guam who consume a BMAA-rich diet but not in control patients who have not been exposed to the traditional Chamorro diet, (ii) vervets fed BMAA over 140 days developed NFT and β-amyloid deposits, and (iii) BMAA was isolated and identified in BMAA-fed vervets that had NFT and β-amyloid deposits in their brains. This study indicates that chronic exposure to BMAA can trigger neurodegenerative illness and that adding L-serine to the diet can reduce the risk of disease.
This work was carried out with permission from the Behavioural Sciences Foundation (BSF) in St. Kitts and all procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of BSF and McGill University (Quebec, Canada). Dosing and animal health was monitored by an on-site DVM veterinarian.
P.A.C., D.C.M. and S.A.B. participated in the conception and design of the experiment, D.C.M. and D.A.D. performed histopathological analyses, S.A.B. conducted the chemical analyses, and all authors participated in conducting the experiments, data analysis and writing the manuscript.
The Institute for Ethnomedicine has applied for patents for the use of L-serine to treat neurodegenerative illness (US 13/683,821) and for screening potential drug candidates using BMAA-induced neurodegeneration (US 14/229,624).
This study was supported by the Josephine P. and John J. Louis Foundation, the William Stamps Farish Fund, Douglas and Elizabeth Kinney, and Patrick and Heather Henry.
We thank Prof. Roberta Palmour, the late Prof. Frank R. Ervin, Dr Amy Beierschmitt, Mr James Powell and the staff of the Behavioural Science Foundation in St. Kitts for overseeing the care and dosing of the vervets, Dr Peter Wyatt of Queen Mary, University of London for technical advice, Ms Jane Cox and Mr James Powell for dose preparation, Mr W. Broc Glover for sample preparation, Dr Robert Switzer and his team at NeuroScience Associates in Tennessee for helpful discussions at the initiation of the study and choice of antibodies, Huron Digital Pathology for technical advice regarding automated digital scanning of the brain sections, Dr Walter Bradley at the Department of Neurology, Miller School of Medicine, Miami for clinical observations of the vervets, the staff of the Miami Brain Endowment Bank at the Miller School of Medicine for management of the biospecimen repository, Ms Marilyn Asay for helping to prepare the manuscript and graphs, and Mr Michael Rothman for the drawing.
We dedicate this study to the memory of Oliver Sacks, friend and mentor.
. 1968A review of the pathologic findings in amyotrophic lateral sclerosis. In Contemporary neurology symposium vol II: motor neuron diseases: research on amyotrophic lateral sclerosis and related disorders(eds FHNorris, LTKurland), pp. 51–60. New York, NY: Grune and Stratton.
. 1994Amyotrophic lateral sclerosis-Parkinsonism-dementia complex on Guam: epidemiologic and etiological perspectives. In Motor neuron disease (ed. ACWilliams), pp. 109–130. London, UK: Chapman and Hall Medical.
. 2013Is exposure to cyanobacteria an environmental risk factor for amyotrophic lateral sclerosis and other neurodegenerative diseases? Amyotroph. Lateral Scler. Frontotemp. Degener. 14, 325–333. (doi:10.3109/21678421.2012.750364)
. 2014Intracellular fibril formation, calcification, and enrichment of chaperones, cytoskeletal, and intermediate filament proteins in the adult hippocampus CA1 following neonatal exposure to the nonprotein amino acid BMAA. Arch. Toxicol. 89, 423–436. (doi:10.1007/s00204-014-1262-2)
. 2008Co-occurrence of β-N-methylamino-L-alanine, a neurotoxic amino acid with other cyanobacterial toxins in British waterbodies, 1990–2004. Environ. Microbiol. 10, 702–708. (doi:10.1111/j.1462-2920.2007.01492.x)
. 2015Determination of β-N-methylamino-L-alanine, N-(2-aminoethyl)glycine, and 2,4-diaminobutyric acid in food products containing cyanobacteria by ultra-performance liquid chromatography and tandem mass spectrometry: single-laboratory validation. J. AOAC Int. 98, 1559–1567. (doi:10.5740/jaoacint.15-084)
. 2012Cyanobacteria produce N-(2-aminoethyl)glycine, a backbone for peptide nucleic acids which may have been the first genetic molecules for life on earth. PLoS ONE 7, e49043. (doi:10.1371/journal.pone.0049043)
. 1994Amyotrophic lateral sclerosis and parkinsonism-dementia from Guam: differences in neurofibrillary tangle distribution and density in the hippocampal formation and neocortex. Brain Res. 650, 107–116. (doi:10.1016/0006-8993(94)90212-7)
. 1998Amyloid plaques in Guam amyotrophic lateral sclerosis/parkinsonism-dementia complex contain species of Aβ similar to those found in the amyloid plaques of Alzheimer's disease and pathological aging. Acta Neuropathol. 95, 117–122. (doi:10.1007/s004010050774)
For 28 years, Bill Gilmore lived in a New Hampshire beach town, where he surfed and kayaked. “I’ve been in water my whole life,” he said. “Before the ocean, it was lakes. I’ve been a water rat since I was four.”
Now Gilmore can no longer swim, fish or surf, let alone button a shirt or lift a fork to his mouth. Earlier this year, he was diagnosed with Amyotrophic lateral sclerosis (ALS), or Lou Gehrig’s disease.
In New England, medical researchers are now uncovering clues that appear to link some cases of the lethal neurological disease to people’s proximity to lakes and coastal waters.
About five years ago, doctors at a New Hampshire hospital noticed a pattern in their ALS patients—many of them, like Gilmore, lived near water. Since then, researchers at Dartmouth-Hitchcock Medical Center have identified several ALS hot spots in lake and coastal communities in New England, and they suspect that toxic blooms of blue-green algae—which are becoming more common worldwide—may play a role.
Now scientists are investigating whether breathing a neurotoxin produced by the algae may raise the risk of the disease. They have a long way to go, however: While the toxin does seem to kill nerve cells, no research, even in animals, has confirmed the link to ALS.
No known cause As with all ALS patients, no one knows what caused Bill Gilmore’s disease. He was a big, strong guy—a carpenter by profession. One morning in 2011, his arms felt weak. “I couldn’t pick up my tools. I thought I had injured myself,” said Gilmore, 59, who lived half his life in Hampton and now lives in Rochester, N.H.
Three years and many doctors’ appointments later, Gilmore received the news in June that the progressive weakening in his limbs was caused by ALS.
Neither Hampton nor Rochester is considered a hot spot for ALS. Gilmore is one of roughly 5,600 people in the United States diagnosed each year with the disease. The average patient lives two to five years from the time of diagnosis.
There is no cure, and for the majority of patients, no known cause. For 90 to 95 percent of people with ALS, there’s no known genetic mutation. Researchers assume that some unknown interaction between genes and the environment is responsible.
In recent years, some of this research has focused on blue-green algae, also known as cyanobacteria.
“There’s a growing awareness of the importance of gene/environment interactions with neurodegenerative diseases. There is more interest in examining environmental exposures, including exposures to cyanobacteria, as possible risk factors for sporadic ALS,” said Paul Alan Cox, director of the nonprofit Institute of Ethnomedicine in Wyoming, which focuses on treatments for ALS and other neurodegenerative diseases.
Cyanobacteria—some of the oldest organisms on the planet—can occur wherever there is moisture. Blooms are fed largely by nutrients in agricultural and urban runoff.
Some cyanobacteria produce toxic compounds that can sicken people. In August, hundreds of thousands of people in Toledo, Ohio, were left without tap water for days when toxins from an algal bloom in Lake Erie were found in the water supply.
While the cyanobacteria toxin that prompted the Toledo water crisis can cause diarrhea, intestinal pain and liver problems, other toxins produced by the blue-green algae can harm the nervous systems of humans and wildlife.
Scientists have long suspected that a cyanobacteria toxin could play a role in some forms of ALS. After World War II, U.S. military doctors in Guam found that many indigenous Chamorro suffered from a rapidly progressing neurological disease with symptoms similar to both ALS and dementia. Years later, scientists found the neurotoxin BMAA in the brains of Chamorro people who died from the disease. Cyanobacteria that grow on the roots and seeds of cycad trees produce the toxin.
Cox, a researcher in Guam in the 1990s, hypothesized that BMAA worked its way up the food chain from the cycad seeds to bats to the Chamorro who hunted them. But Cox and his colleagues also found BMAA in the brains of Canadian Alzheimer’s patients who had never dined on Guam’s fruit bats. In patients who had died from other causes, they found no traces of it. The source of the BMAA in the Canadians remains unknown.
Some researchers have suggested that fish and shellfish from waters contaminated with cyanobacteria blooms may be one way that people ingest BMAA. In southern France, researchers suspect ALS cases may be linked to consumption of mussels and oysters. Lobsters, collected off the Florida coast near blooms, also have been found with high levels of BMAA.
Scientists around the world are investigating how the neurotoxin gets into the body and whether it contributes to disease.
“We don’t really know what exposure routes are most important,” Cox said.
New England’s ALS hot spots In New Hampshire, Dartmouth neurologist Elijah Stommel noticed that several ALS patients came from the small town of Enfield in the central part of the state. When he mapped their addresses, he saw that nine of them lived near Lake Mascoma.
Around the lake, the incidence of sporadic ALS—cases for which genetics are not a likely cause—is approximately 10 to 25 times the expected rate for a town of that size.
“We had no idea why there appeared to be a cluster around the lake,” Stommel said.
Based on the link between ALS and the neurotoxin in other parts of the world, Stommel and his colleagues hypothesize that the lake’s cyanobacteria blooms could be a factor.
Across northern New England, the researchers have continued to identify ALS hot spots—a large one in Vermont near Lake Champlain and a smattering of smaller ones among coastal communities in New Hampshire and Maine.
Earlier this year, the researchers reported that poorer lake water quality increased the odds of living in a hot spot. Most strikingly, they discovered that living within 18 miles of a lake with high levels of dissolved nitrogen—a pollutant from fertilizer and sewage that feeds algae and cyanobacteria blooms—raised the odds of belonging to an ALS hot spot by 167 percent.
The findings, they wrote, “support the hypothesis that sporadic ALS can be triggered by environmental lake quality and lake conditions that promote harmful algal blooms and increases in cyanobacteria.”
How people in New England communities could be ingesting the neurotoxin remains largely a mystery. While fish in the lakes do contain it, not everyone in the Dartmouth studies eats fish.
“We’ve sent questionnaires to patients and there’s really no common thread in terms of diet or activities,” Stommel said. “The one common thing that everybody does is breathe.”
In other words, it’s possible that a boat, jet ski or even the wind could stir up tiny particles of cyanobacteria in the air, where people then breathe it in.
Testing the air for a neurotoxin Last August, at Lake Attitash, Jim Haney, a University of New Hampshire biologist, waded knee-deep into swirling green water. Cyanobacteria were blooming at the small lake in the northeastern corner of Massachusetts. Haney had rigged up three vacuum-like devices with pipes, plastic funnels and paper to suck up and filter air near the lake’s surface.
He took the filter papers back to his laboratory and measured the cyanobacteria cells, BMAA and other toxins stuck to them.
“We want to know what level lake residents may be exposed to through airborne particles,” said Haney, who is sampling the air at Massachusetts and New Hampshire lakes in collaboration with the Dartmouth team.
Stommel said,“it’s very compelling to look at the filter paper and see it just coated with cyanobacteria.”
At this point, Haney and graduate students are trying to understand under what conditions the toxins might be coming out of the lake and whether the airborne particles are an important route of exposure.
Preliminary findings suggest that BMAA and other cyanobacteria cells are being aerolized. “There is potentially a large quantity of cyanobacteria that could be inhaled,” Haney said. He noted, however, that the measurements were taken about eight inches above the water's surface, making it likely that concentrations would be much lower farther away.
While the toxins are likely to be most abundant in the air around lakes, they exist all over the planet, even in deserts.
In 2009, BMAA was even detected in the sands of Qatar. Crusts containing cyanobacteria may lie dormant in the soil for most of the year, but get kicked up during spring rainstorms. Cox and colleagues hypothesized that breathing in toxins from dust might be a trigger for a doubling of ALS incidence in military personnel after Operation Desert Storm.
Near Haney’s workstation at Lake Attitash, a child splashed in the shallow water off a dock. Haney scooped up a cupful of water. He peered at the tiny green particles in the cup that reflect the sunlight, making the mixture resemble a murky pea soup.
“We’ve developed this view of nature as idyllic, which is wonderful, but not everything in nature is benign,” he said. “Rattlesnakes are natural and you wouldn’t get too close to one of those.”
“Proximity does not equal causality” The hypothesis that exposure to BMAA may trigger the disease in some people remains controversial.
Researchers have evidence that people living close to lakes with blooms may be at increased risk for ALS. They’ve even found BMAA in the diseased brain tissue of people who have died of neurodegenerative diseases. Nevertheless, “proximity does not equal causality,” said Deborah Mash, a neuroscientist at the University of Miami in Florida.
The big, unanswered question is whether the toxin can actually cause the disease. So far, there’s little evidence to show how it could induce the type of brain changes seen in people with ALS.
Tests of human cells have found that BMAA kills the motor neurons—nerve cells that control muscles—implicated in ALS. Primates fed high levels of BMAA in the 1980s showed signs of neurological and muscular weakness. But the toxin did not kill their motor neurons.
“What is lacking at this point is a clear animal model that demonstrates that BMAA exposure results in ALS-like neuropathy,” Cox said.
So what is a possible mechanism for how the toxin may lead to the disease? The body may mistake BMAA for the amino acid L-serine, a naturally occurring component of proteins. When the toxin is mistakenly inserted into proteins, they become “misfolded,” meaning they no longer function properly and can damage cells.
Cox and colleagues soon will test two drugs in FDA-approved clinical trials. They’re about to enter second-phase testing with L-serine. The idea, explained Sandra Banack, a researcher at the Institute for Ethnomedicine, is that large doses of L-serine may be able to “outcompete” low levels of BMAA in the body, preventing it from becoming incorporated into proteins.
For ALS patients like Gilmore, the research can’t come soon enough. “If they can figure out a cause, then hopefully they can find a cure,” Gilmore said.
This article originally ran at Environmental Health News, a news source published by Environmental Health Sciences, a nonprofit media company.