Neurology -- Correspondence for Montine et al., 65 (5) 768-769

Correspondence published:

Lack of ß-methylamino-l-alanine in brain from controls, AD, or Chamorros with PDC: Paul Alan Cox, PhD, S Conradi; B Bergman; U Rasmussen; R Bidigare; GA Codd; JS Metcalf; H Johnson;R Speth; JH Weiss (22 December 2005)

Reply from the authors: Thomas J. Montine, Daniel Perl, Douglas Galasko (22 December 2005)

Lack of ß-methylamino-l-alanine in brain from controls, AD, or Chamorros with PDC 22 December 2005

Paul Alan Cox, PhD,
Jackson, WY
SA Banack; S Murch; A Bell; P Nunn; W Bradley; D Mash; S Papapetropoulos; J Pablo;L-O Ronnevi;,
S Conradi; B Bergman; U Rasmussen; R Bidigare; GA Codd; JS Metcalf; H Johnson;R Speth; JH Weiss
Send Correspondence to journal:
Re: Lack of ß-methylamino-l-alanine in brain from controls, AD, or Chamorros with PDC

wbradley{at}med.miami.edu Paul Alan Cox, PhD, et al.

We read with interest the article by Montine et al. [1] ß-Methylamino-L-alanine (BMAA) was discovered in free and protein fractions of blinded brain tissues of ALS/Parkinson–dementia complex (ALS/PDC) Chamorro patients from Guam. [2] Montine et al. did not detect BMAA, stating that “the reasons for our inability to confirm the results of others are not clear. All of these studies . . . use similar quantitative methods.”
The methods used by Montine et al. are not similar to the original methods, and therefore their failure to confirm does not refute the original findings. In contrast, an independent team found BMAA in brain tissues from patients with neurodegeneration, using the original methods (D. Mash, University of Miami, personal communication).
BMAA is detected using precolumn derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC), a reverse phase gradient HPLC separation, fluorescent detection, and LC/MS confirmation. [2] This method was validated by careful assessment of the precision, accuracy, limits of quantification (LOQ), limits of detection (LOD), linearity and range of the LOQ and LOD, and repeated determination of ruggedness and robustness. [3] AQC provided “dramatic improvements” [4] over the older FMOC method Montine et al. employed and is used for hospital amino acid quantification.
FMOC yields multiple derivatives, is “troublesome” for human tissues, and exhibits reduced derivatization efficiency under common conditions. [5,6] Insufficient details of tissue preparation, tissue amount, amino acid extraction, precolumn derivatization, columns, and HPLC instrumentation make it difficult to repeat Montine et al. Their chromatogram is not comparable to the sharp, highly resolved, numerous amino acid peaks produced by FMOC experts [5]; lacks sufficient time range to identify any other amino acid; and is deficient in detail.
The use of up-to-date validated methods is important for quantification of BMAA, an unusual, highly reactive amino acid, within a complex human physiologic matrix. Remarkably, Montine et al. did not repeat the newer method or test for protein-associated BMAA, which occurs at higher levels. [2] As they noted, the original discovery was from fixed, not frozen, tissues. Recently, matched frozen and fixed tissue samples of Chamorro patients were obtained from Guam Memorial Hospital. Nonparametric pairwise comparisons show that BMAA levels are significantly higher in frozen tissues (n = 24, T = 2.5, Z0 = 3.35, p < 0.01); these results will be published elsewhere.
Given the lack of basic information that would allow replication of Montine and colleagues’ results, their inability to detect BMAA in Chamorro ALS/PDC tissues may reflect absence of analytical rigor rather than absence of BMAA.
References References
1. Montine TJ, Li K, Perl DP, Galasko D. Lack of β-methylamino-L-alanine in brain from controls, AD, or Chamorros with PDC. Neurology 2005;65:768–769.
2. Murch SJ, Cox PA, Banack SA, Steele JC, Sacks OW. Occurrence of ß-methylamino- L-alanine (BMAA) in ALS/PDC patients from Guam. Acta Neurol Scand 2004;110:267–269.
3. Meyer VR. Practical high-performance liquid chromatography, 4th edition. Chichester, UK: John Wiley & Sons, 2004.
4. Cohen SA, De Antonis KM. Applications of amino acid derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate. J Chromatogr A 1994;661:25–34.
5. Fürst P, Pollack L, Graser TA, Godel H, Stehle P. Appraisal of four pre-column derivatization methods for the high-performance liquid chromatographic determination of free amino acids in biological materials. J Chromatogr 1990;499:557–569.
6. Cohen SA, Michaud DP. Synthesis of a fluorescent derivatizing reagent, 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate, and its application for the analysis of hydrolysate amino acids via high-performance liquid chromatography. Anal Biochem 1993;211:279–287.
The authors report no conflicts of interest.

Reply from the authors 22 December 2005

Thomas J. Montine,
University of Washington
Box 359791, Seattle, WA 98104,
Daniel Perl, Douglas Galasko
Send Correspondence to journal:
Re: Reply from the authors

tmontine{at}u.washington.edu Thomas J. Montine, et al.

We thank Cox et al for their interest in our work. In data published by Cox and collaborators, BMAA was detected in a free pool in formalin-fixed brain tissue from patients with AD and PDC. Although the free BMAA levels were much lower than the “protein-associated” levels, most evidence for BMAA supports the free molecule as the proximate neurotoxin perhaps acting as an excitotoxin.
We therefore attempted to replicate the finding of elevated tissue levels of the presumed biologically active agent, free BMAA. Our homogenization procedure of frozen brain tissue is quite standard. Our detection protocol has been widely used by many investigators, is described sufficiently in our article, and is detailed in the referenced manuscript.
Our data speak for themselves. We established a detection limit for authentic free BMAA spiked into human brain homogenates prepared from frozen tissue that was well below the free BMAA tissue level reported by Cox and colleagues in fixed tissue from affected Chamorros as well as patients with Alzheimer’s disease and controls from western Canada.
We found no detectable free BMAA in any sample from Chamorros with or without PDC, or from AD patients or control individuals from western USA. We would exchange homogenates or tissue samples with Cox et al if they wish to test their assertion of superior analytical rigor.
We cannot comment on the possibility of a large “protein-associated” pool of BMAA, as we did not attempt to detect this. Is this protein-bound BMAA or BMAA incorporated into protein during translation? Both appear possible given the techniques employed. We note that no specific proteins involved in this association have been identified by Cox et al. Proteins undergo turnover and replacement, most quite rapidly, so either means of protein association does not guarantee the formation of an “endogenous reservoir” that slowly releases toxin over years to account for slow neurotoxicity (or persistence of a toxin) many years after exposure to the putative agent. If BMAA is incorporated into proteins, leading to protein dysfunction or an immune reaction, this would be a remarkable and novel mechanism of toxicity that will require very rigorous identification of which proteins are involved to take this beyond conjecture.
The authors report no conflicts of interest.