Bmaa sigma

Biomagnification of cycad neurotoxins in flying foxes

Implications for ALS-PDC in Guam Sandra Anne Banack, PhD; and Paul Alan Cox, PhD

Abstract—!-Methylamino-L-alanine (BMAA) occurs in higher levels in museum specimens of the Guamanian flying fox than in the cycad seeds the flying foxes feed on, confirming the hypothesis that cycad neurotoxins are biomagnified within the Guam ecosystem. Consumption of a single flying fox may have resulted in an equivalent BMAA dose obtained from eating 174 to 1,014 kg of processed cycad flour. Traditional feasting on flying foxes may be related to the prevalence of neuropathologic disease in Guam. NEUROLOGY 2003;61:387–389

ALS–parkinsonism dementia complex (PDC), de- scribed in the Chamorro people of Guam, with as- pects similar to ALS, Parkinson’s disease (PD), and Alzheimer’s disease (AD), once occurred in Guam at 50 to 100 times the incidence of ALS elsewhere.1 Epidemiologic studies indicate that consumption of a Chamorro diet is the only variable significantly associ- ated with disease incidence.2 Seeds of cycad plants, Cycas micronesica Hill in the C. rumphii species com- plex, used by the Chamorro people as a source of torti- lla flour, have neurotoxins including !-methylamino-L- alanine (BMAA), a nonprotein amino acid that Spencer et al.3 and others have suggested may be associated with ALS-PDC. However, the concentrations of BMAA in processed cycad flour were believed to be too low to produce comparable disease states in animal models.4-6

Cox and Sacks7 suggested that high doses of cycad neurotoxins or their metabolites might be inadver- tently ingested by the Chamorros during their con- sumption of flying foxes during traditional feasts. The high rates and subsequent decline of ALS-PDC in Guam shadowed the rise and eventual decline in consumption of cycad-fed flying foxes by the Chamorro people.8

We analyzed three C. micronesica seeds obtained from Guam. Pteropus mariannus mariannus, an in- digenous flying fox of Guam, is now highly endan- gered. Therefore, we analyzed skin tissue from museum specimens of three flying foxes that were collected five decades ago in Guam, preserved as dried study skins, and deposited at the Museum of Vertebrate Zoology (MVZ) at the University of Cali-

fornia, Berkeley. We also analyzed three processed cycad flour specimens from Guam. We examined all samples for the presence of BMAA using high- performance liquid chromatography (HPLC). Results were confirmed with thin-layer chromatography (TLC) as well as gas chromatography–mass spectroscopy.

Methods. BMAA was quantified from free amino acid extracts of flying fox and cycad tissues. Samples were rehydrated for 30 minutes with water or trichloroacetic acid (mean tissue prepara- tion 80 " 32 [SD] mg/mL), macerated, and filtered. Extracts were derivatized with 6-aminoquinolyl-N-hydrozysuccinimidyl carbam- ate (ACQ) following standardized protocols.9 Free amino acids were separated by reverse-phase separation on a gradient HPLC system (Waters 717 Automated Injector, Waters 1525 Binary Sol- vent Delivery System, and Waters Nova-Pak C18 column, 300 # 3.9 mm; Waters, Milford, MA) at 37 °C. Individual compounds were eluted from the column with a gradient elution of 140 mM sodium acetate/5.6 mM triethylamine, pH 5.2 (mobile phase A) and 60% acetonitrile (mobile phase B) with a flow rate of 1.0 mL/min.9 Gradient conditions were as follows: initial $ 100% A, 2.0 minutes $ 90% A curve 11, 5.0 minutes $ 86% A curve 11, 10.0 minutes $ 86% A curve 6, 18.0 minutes $ 73% A curve 6, 30.0 minutes $ 60% A curve 10, 35.0 minutes $ 40% A curve 6, 39.0 minutes $ 10% A curve 6, followed by a wash with 100% B for 5 minutes and re-equilibration for 5 minutes at 100% A. BMAA peak identity was confirmed by comparison with a com- mercial standard (%94% pure; Sigma B-107, St. Louis, MO) and was reverified by modified gradient elution. The concentration of BMAA in samples was determined by fluorescence detection (Wa- ters 2487 Dual-l Fluorescence Detector) with excitation at 250 nm and emission at 395 nm. Detection of the ACQ-derivatized BMAA was dependent on concentration, and quantification was accom- plished with comparison of equal amounts of BMAA and a nor- leucine internal standard (representing a single midrange concentration), resulting in a mean response of 51.2%. These data express the average response of values for several experiments and depict the efficiency of the derivatization protocol and the