Development of an analytical procedure for quantifying the underivatized neurotoxin β-N-methylamino-l-alanine in brain tissues

Abstract

The cyanotoxin β-methylamino-L-alanine (BMAA) has received renewed attention as an environmental risk factor for sporadic cases of amyotrophic lateral sclerosis (ALS) (Nunn et al., Brain Res 410:375-379, 1987). The aim of the present study was to develop and to validate an analytical procedure that allows the quantification of native BMAA and of its natural isomer, 2,4 diaminobutyric acid (DAB), in brain tissues. An analytical procedure was previously reported by our group for the determination of underivatized BMAA in environmental samples. It included a step of sample clean-up by solid phase extraction (SPE) with a mixed-mode sorbent and the analyses were performed by LC/MS-MS using hydrophilic interaction chromatography and multiple reactions monitoring scan mode. As brain tissues have a higher lipid content, the crucial step of sample clean-up had been optimized by evaluating the efficiency of the addition of a liquid/liquid extraction step prior to the SPE procedure or alternatively, of washing steps to the SPE extraction procedure. The efficiency was checked by visualizing the complexity of the resulting chromatograms in LC/MS and their performance by using spiked brain samples. The optimized analytical procedure, including a washing step with cyclohexane to the SPE with a recovery yield close to 100%, was validated using the total error approach and allowed the quantification of BMAA in a concentration level ranging from 20 to 1,500 ng/g in brain samples. Finally, the feasibility of implementation of this procedure was verified in human brain samples from two patients who died of ALS.

Detection of BMAA in the human central nervous system

“…with a known direct genetic cause (Cox et al, 2003;Murch et al, 2004a,b;Pablo et al, 2009). However, as other studies analyzing similar human CNS based sample sets have failed to detect BMAA in either patients or matched controls (Montine et al, 2005;Kushnir and Bergquist, 2009;Snyder et al, 2009aSnyder et al, ,b, 2010Combes et al, 2014), the occurrence of BMAA in the human CNS is still under debate as well as the observed presence of BMAA in patients with neurodegenerative disorders.…”

“…Age ( Combes et al (2014) fractions of patient ALS1 contained 78.0 and 96.6 pg BMAA/ml CSF, respectively. The BMAA concentrations in the hydrolyzed protein fractions of the non-ALS1 and non-ALS2 samples were above the LOQ.…”

The neurotoxin BMAA in aquatic systems : analysis, occurrence and effects

“…We selected sample preparation methods that were published, validated and developed by the trainers of the workshop (see [93] for underivatised analysis of animal and cyanobacterial samples, [199] for underivatised analysis in brain and [197] for AQC derivatised analysis of all sample types). Where needed, the extraction methods were adapted to the available equipment.…”

“…The protocol used for brain samples followed by underivatised LC-MS/MS analysis was adapted from a method developed and validated for the underivatised analysis of free BMAA in brain tissue [199]. Main adaptations to this published method are that we included a method for total BMAA, that we started with freeze dried samples instead of with samples that were only frozen and thawed and that we therefore adapted the first extraction steps for free BMAA, and that the LC-MS/MS analysis was performed according to [195] on the same LC-MS/MS system as used for the other analyses in this experiment.…”

“…Chapter 3, [100,145]) and the performance of underivatised LC-MS/MS methods can be good as well (e.g. Chapters 3 and 7, [103,199]). If each method is properly applied, there seems to be no scientific support for the suggestion that AQC derivatised LC-MS/MS analysis would lead to false positives [77], but neither for the claims that underivatised LC-MS/MS methods will greatly underestimate BMAA concentrations [88] or that these methods simply are inferior to AQC derivatised LC-MS/MS methods [84].…”