Mass-spectrometry based survey of BMAAsources, distribution and transferNadezda Zguna

Academic dissertation for the Degree of Doctor of Philosophy in Analytical Chemistry atStockholm University to be publicly defended on Thursday 28 February 2019 at 10.00 inNordenskiöldsalen, Geovetenskapens hus, Svante Arrhenius väg 12.

Abstractβ-methylaminoalanine (BMAA) is a neurotoxic non-protein amino acid first isolated from cycad seeds in 1967. It is believed to be connected to neurodegenerative diseases such as Parkinson’s, Alzheimer’s and amyotrophic lateral sclerosis (ALS) and is a ubiquitous compound produced by cyanobacteria, diatoms and dinoflagellates. Consequently, elucidating natural production, distribution and routes for human exposure of BMAA are of particular importance. However, the natural function of BMAA and its mechanisms of toxicity have not been fully established yet. The contradictory results about BMAA presence in cyanobacterial cultures and food webs have been reported by different scientific groups, which required the development of more sensitive and reliable analytical methods.

This thesis is focused on the analytical chemistry dimension of BMAA research: covering both new method development and novel applications. New analytical methods for BMAA detection and quantification were developed, focusing on improving sensitivity, since BMAA is normally found in natural samples at low concentrations. In Paper I, a new derivatization technique was implemented, which increased sensitivity and selectivity in the analysis of BMAA and its isomers. Subsequently, this developed method was applied to determine the presence of BMAA in fat and oil matrices in Paper II, which is a step towards discovering BMAA forms other than the documented free and protein-bound BMAA species. In Paper III, a method for separation and quantification of L- and D-BMAA stereoisomers in complex biological matrix was developed and applied to determine the enantiomeric composition of BMAA in cycad seed. Studying environmental distribution of BMAA is important to evaluate potential exposure routes and health risks for humans. Part of the work was devoted to broaden assessment on environmental occurrence of BMAA by applying existing robust methodology to new samples, such as commercial seafood in Paper IV and Baltic Sea biota in Paper V. Some of the “overlooked” aspects in the existing BMAA literature (i.e., BMAA chiral analysis, possible BMAA presence in dietary oil supplements and defined food webs) were successfully addressed.

Overall, the thesis presents important analytical developments, which can help to further elucidate sources, distribution and transfer of BMAA.

Keywords: BMAA, LC-MS/MS, chiral analysis, seafood, toxins, Baltic Sea, derivatization.

Stockholm 2019http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-164088

ISBN 978-91-7797-544-1ISBN 978-91-7797-545-8

Department of Environmental Science and AnalyticalChemistry

Stockholm University, 106 91 Stockholm

MASS-SPECTROMETRY BASED SURVEY OF BMAA SOURCES,DISTRIBUTION AND TRANSFER 

Nadezda Zguna

Mass-spectrometry basedsurvey of BMAA sources,distribution and transfer 

Nadezda Zguna

©Nadezda Zguna, Stockholm University 2019 ISBN print 978-91-7797-544-1ISBN PDF 978-91-7797-545-8 Cover image author: Tanja Russita Printed in Sweden by Universitetsservice US-AB, Stockholm 2019

To my dear parents, Victor and Lyubov

Моим дорогим родителям, Виктору и Любови

Populärvetenskaplig sammanfattning

Aminosyror kallas ofta för livets byggstenar. Av de hundratals naturliga

aminosyrorna som man känner till är det dock endast 21 stycken som bygger

upp proteinerna i människokroppen. De övriga, så kallade icke-proteinogena

aminosyror, uppfyller andra viktiga biologiska funktioner i olika typer av

levande organismer.

På 1960-talet upptäcktes en hittills okänd icke-proteinogen aminosyra på ön

Guam och som misstänks ha ett samband med en förhöjd frekvens av

neurologiska sjukdomar, som amyotrofisk lateral skleros (ALS) och

Parkinsons sjukdom. Föreningen isolerades från fruktkärnorna från en

kottepalm och identifierades som β-metylamino-L-alanin (BMAA). Senare

studier har visat att BMAA också produceras av vanliga marina organismer,

som cyanobakterier. Dessutom har hypotesen om föreningens verkan som

nervgift fått ytterligare vetenskapligt stöd. Sammantaget har intresset för

BMAA ökat markant under de senaste åren.

I arbetet bakom avhandlingen har nya kemiska analysmetoder utvecklats för

att studera förekomst och spridning av BMAA i miljön, möjliga källor, som

mat och kosttillskott, samt eventuell transport inom näringskedjan.

Högupplösande masspektrometri har varit viktig för att kunna detektera

riktigt låga halter av BMAA. Alla analysmetoder som beskrivs i

avhandlingen har dessutom involverat kemisk omvandling, derivatisering,

med olika reagens innan analys, dels för att förbättra separationen från andra

ämnen, dels för en känsligare detektion. För BMAA har därför lägre

koncentrationer än tidigare kunnat uppmätas i exempelvis blåmusslor, fisk-

och algolja.

I ett av arbetena har två isomerer av BMAA, de optiskt aktiva D- och L-

formerna som är varandras molekylära spegelbilder, påvisats för första

gången i fruktkärnor från kottepalm. Dessa så kallade enantiomerer är

molekyler som har identiska fysikaliska och kemiska egenskaper och därför

inte kan åtskiljas genom vanlig kromatografi eller masspektrometri. Däremot

är det vanligt att enantiomerer har olika biologiska funktioner och det är

därför viktigt att kunna mäta dem separat. Genom att derivatisera

blandningen av BMAA-enantiomerer med en optiskt aktivt reagens erhålls

isomerer av BMAA-derivat som inte längre är spegelbilder och därför kan

separeras kromatografiskt och mätas var för sig.

Eftersom BMAA har kopplats till neurodegenerativa sjukdomar är det viktigt

att identifiera eventuella livsmedel som kan vara potentiella

exponeringskällor. Därför undersöktes förekomst av BMAA i fisk och

skaldjur från olika matbutiker i Stockholm. Låga halter observerades i

ostron, blåmusslor, räkor och rödspätta. Däremot kunde BMAA inte

detekteras i lax, torsk, abborre eller kräftor.

Möjliga transportvägar för BMAA inom näringskedjor från Östersjömiljö

undersöktes som modell för ett naturligt ekosystem. Endast växtplankton,

djurplankton och pungräkor visade detekterbara mängder av BMAA, medan

inga halter alls kunde uppmätas i sediment, bottenlevande ryggradslösa djur,

skarpsill eller skrubbskädda. Enbart prover tagna under sommarsäsongen

analyserades, men för en fullständigare bild skulle en längre, exempelvis

ettårig mätperiod behövas. Dessutom vore det intressant att undersöka fler

områden i Östersjön.

Det finns en risk för att personer som bor nära vattendrag skulle kunna

exponeras för BMAA genom inandning av naturligt bildade aerosoler.

Därför initierades en undersökning kring detta. Ytterligare studier krävs

dock för att fullt ut kunna bedöma riskens omfattning.

1

Table of Contents

List of publications ........................................................................................ 2

Abbreviations ................................................................................................ 5

Introduction ................................................................................................... 7

BMAA chemical properties ........................................................................ 7

BMAA natural production, distribution and human exposure .................... 8

Forms of BMAA in natural matrices ........................................................ 11

L-BMAA toxicity and neurodegenerative diseases ................................... 12

Toxicity of BMAA structural isomers and D-BMAA ............................... 16

Analytical strategies for BMAA ............................................................... 17

Aims of the thesis ........................................................................................ 21

Methods ........................................................................................................ 23

Sample pre-treatment and derivatization .................................................. 23

UHPLC separation .................................................................................... 24

MS/MS detection ...................................................................................... 25

Results and discussion ................................................................................ 29

Developing novel derivatization methods................................................. 29

Screening studies ...................................................................................... 37

Other analytical strategies and applications .............................................. 39

Conclusions .................................................................................................. 45

Further method development & application ............................................ 47

Acknowledgements ..................................................................................... 49

References .................................................................................................... 52

2

List of publications

I. Improved detection of β-N-methylamino-L-alanine using N-

hydroxysuccinimide ester of N-butylnicotinic acid for the

localization of BMAA in blue mussels (Mytilus edulis)

Rudolf Andrýs *, Javier Zurita

*, Nadezda Zguna, Leopold L. Ilag,

Klaas Verschueren and Wim M. De Borggraeve

Anal Bioanal Chem. 407(13) (2015)

II. Analysis of BMAA in fish oil and its vegan alternatives

Nadezda Zguna, Jan Holmback and Leopold L. Ilag

[Manuscript]

III. Chiral separation of β-Methylamino-L-alanine (BMAA)

enantiomers after (+)-1-(9-fluorenyl)-ethyl chloroformate (FLEC)

derivatization and LC-MS/MS

Javier Zurita*, Nadezda Zguna

*, Rudolf Andrýs, Anna Strzelczak,

Liying Jiang, Gunnar Thorsen and Leopold L. Ilag

Analytical Methods, 2019, DOI: 10.1039/C8AY02287A

IV. Quantification of neurotoxin BMAA (β-N-methylamino-L-

alanine) in seafood from Swedish markets

Liying Jiang, Nadezda Kiselova, Leopold L. Ilag and Johan Rosen

Scientific Reports, 4 (2014)

V. Is there a BMAA transfer in the pelagic and benthic food webs in

the Baltic Sea?

Nadezda Zguna, Agnes Karlson, Leopold L. Ilag, Andrius Garbaras

and Elena Gorokhova

[Submitted]

* authors contributed equally to the paper

3

The author’s contribution to the papers was as described below:

I. The author was responsible for part of the experimental work and

part of the data evaluation.

II. The author was responsible for designing and performing the

majority of the experiments, all data evaluation and writing the

paper.

III. The author was responsible for part of the experimental design;

significant parts of the experimental work, significant parts of data

evaluation and significant part of writing the paper.

IV. The author was responsible for all the experimental work, part of the

data evaluation and part of writing the paper.

V. The author was responsible for designing and performing the

analytical part of the work and associated sample preparation as well

as LC-MS/MS data evaluation and writing a significant part of the

paper.

4

Publications not included in this thesis

VI. Human Scalp Hair as an Indicator of Exposure to the

Environmental Toxin β-N-Methylamino-l-alanine

Simoné Downing, Laura Louise Scott, Nadezda Zguna and Timothy

Grant Downing

Toxins, Vol 10, Iss 1, p 14 (2017)

VII. Host cell-derived lactate functions as an effector molecule in

Neisseria meningitidis microcolony dispersal

Sara Sigurlásdóttir, Jakob Engman, Olaspers Sara Eriksson, Sunil D

Saro, Nadezda Zguna, Pilar Lloris-Garcerá, Leopold L Ilag and

Ann-Beth Jonsson

PLoS Pathogens, Vol 13, Iss 4, p e1006251 (2017)

VIII. A Collaborative Evaluation of LC-MS/MS Based Methods for

BMAA Analysis: Soluble Bound BMAA Found to Be an

Important Fraction

Elisabeth J. Faassen, Maria G. Antoniou, Wendy Beekman-

Lukassen, Lucie Blahova, Ekaterina Chernova, Christophoros

Christophoridis, Audrey Combes, Christine Edwards, Jutta Fastner,

Joop Harmsen, Anastasia Hiskia, Leopold L. Ilag, Triantafyllos

Kaloudis, Srdjan Lopicic, Miquel Lürling, Hanna Mazur-Marzec,

Jussi Meriluoto, Cristina Porojan, Yehudit Viner-Mozzini and

Nadezda Zguna

Marine Drugs, Vol 14, Iss 3, p 45 (2016)

IX. An automated mass spectrometry-based screening method for

analysis of sulfated glycosaminoglycans

Nadezda Kiselova, Tabea Dierker, Dorothe Spillmann and

Margareta Ramström

Biochemical and Biophysical Research

Communications 450(1):598-603 (2014)

5

Abbreviations

AEG N-(2-aminoethyl)glycine 1H-NMR Proton nuclear magnetic resonance spectroscopy

ALS Amyotrophic lateral sclerosis

AMPA α-amino-β-methylaminopropionic acid

AP7 2-amino-7-phosphonoheptanoic acid

AQC 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate

BAMA β-amino-N-methylalanine

BMAA β-methylaminoalanine

BOAA α-amino-β-oxalylaminopropionic acid

C4-NA-NHS N-hydroxysuccinimide ester of N-butylnicotinic acid

CE Capillary electrophoresis

DAB 2,4-diaminobutanoic acid

ELISA Enzyme linked immunosorbent assay

ESI Electrospray ionization

FLD Fluorescence detector

FLEC (+)-1-(9-fluorenyl)-ethyl chloroformate

GC Gas chromatography

HILIC Hydrophilic interaction liquid chromatography

(U)HPLC (Ultra) High performance liquid chromatography

LC Liquid chromatography

LLE Liquid-liquid extraction

LOD Limit of detection

LOQ Limit of quantification

LPI Low pressure impactor

MRM Multiple reactions monitoring

MS Mass spectrometry

MS/MS Tandem mass spectrometry

N-butyl-AQ-N-NHS N-butyl-aminoquinolyl-N-hydroxysuccinimide

N-methyl-AQ-N-NHS N-methyl-aminoquinolyl-N-hydroxysuccinimide

NFT Neurofibrillary tangles

NMDA N-methyl-D-aspartate

PD Parkinsonism-dementia

QC Quality control

SPE Solid phase extraction

TCA Trichloroacetic acid

UV Ultraviolet

6

7

Introduction

In the 1960s, the anthropologist Marjorie Whiting was working on Guam

Island, studying the indigenous Chamorro population lifestyle and, in

particular, their diet1. She was interested in the fact that the Chamorros

suffered from an unknown neurodegenerative disease similar to amyotrophic

lateral sclerosis (ALS)-parkinsonism-dementia (PD) complex at unusually

high occurrence (50-100 times higher than worldwide)2. It was suspected

that a plant that was widely used in Chamorros’ traditional cuisine (Cycas

micronesica), might contain a toxic amino acid, α-amino-β-

oxalylaminopropionic acid (BOAA), which had been previously linked to

lathyrism by Arthur E. Bell3, or an analogue to BOAA in terms of its toxic

properties.

A couple of years later, in 1966, Vega and Bell isolated an amino acid from

the seeds of Cycas micronesica. They proved the compound to be β-

methylaminoalanine (BMAA), though at that time it was named α-amino-β-

methylaminopropionic acid (AMPA). Vega and Bell also mentioned the

acute toxicity of BMAA to chicks and thus BMAA started its winding road

in science4.

BMAA chemical properties

BMAA is a non-protein amino acid with molecular mass of 118.1 Da. It is a

polar basic amino acid with pKa values of 2.1, 6.48 and 9.70 for carboxyl

group, primary and secondary amino group, respectively5. BMAA is a chiral

compound, which means it can be present as L-BMAA and D-BMAA. Thus,

BMAA has structural isomers and this fact should be taken into

consideration during development of analytical methods. Isomers which

were observed in natural matrices6 are 2,4-diaminobutanoic acid (DAB), N-

(2-aminoethyl)glycine (AEG) and β-amino-N-methylalanine (BAMA)

(Figure 1). BMAA is also known as a chelator of divalent metals, such as

zinc, copper and nickel7.

8

Figure 1. BMAA and its structural isomers. Chiral centers of the molecules are denoted with

asterisk (*).

BMAA natural production, distribution and human

exposure

Even though the biological role of BMAA is still not clear, it has been

detected in various ecosystems all around the globe. Guam Island4, Lake

Taihu in China8, Baltic Sea

9, British

10, Portuguese

11 and South Florida

12

water bodies, Dutch urban waters13

, and even deserts of Qatar14

are examples

of where BMAA has been discovered. BMAA is believed to be naturally

produced by ubiquitous organisms: cyanobacteria in marine, freshwater and

terrestrial environments (such as soil and limestone caves)15

, diatoms16,17

and

dinoflagellates11,17,18

as reported for Swedish coast and French ecosystems.

Generally, cyanobacteria are well-known producers of bioactive compounds,

including toxins, which may cause adverse health effects19

. Several lines of

evidence support the hypothesis of BMAA production by cyanobacteria. It

has been widely demonstrated that both field samples and axenic laboratory

cultures of cyanobacteria contain BMAA in detectable amounts, which

would be possible only in case of endogenous BMAA production15,20

.

Similar experiments have been performed with diatom cultures that were

also found to contain BMAA in the low µg∙g-1

range17

.

The biological function of BMAA in cyanobacteria has also been addressed.

In the experiment with cyanobacteria and the isotope-labelled 15

NH4+ that

was used as a sole nitrogen source, nitrogen starvation induced an increase in

free 15

N BMAA, indicating synthesis of BMAA. Moreover, the addition of

nitrate or ammonia to the culture medium after starvation reduced the

concentration of free BMAA and did not cause the increase in the protein-

9

associated fraction. That finding suggests that BMAA may have an

important role in nitrogen metabolism in cyanobacteria21

. Further

investigations by the same scientific group proposed a possible pathway for

BMAA metabolism in cyanobacteria, which presumes that a natural

precursor of BMAA is a keto acid undergoing reversible amination reaction

to produce BMAA22

.

Other researchers supported the idea of BMAA’s role in cyanobacterial

nitrogen metabolism. For instance, Berntzon and co-authors have

demonstrated that BMAA severely inhibits nitrogenase (enzyme

participating in nitrogen fixation) activity, reduces growth, induces

pigmentation change of cells from blue-green to yellow-green, and leads to

the massive accumulation of glycogen23

. However, publications reporting no

detectable levels of BMAA in cyanobacteria are also common24–26

. In most

cases considered nowadays, reliable analytical methods were used, so the

controversy of the findings could not be attributed to poor scientific

approach and, therefore, cannot be ignored.

Currently, no large-scale surveys comparing field and laboratory samples of

cyanobacteria are available. Indeed a systematic investigation is needed to

reveal which specific cyanobacteria produce BMAA and under which

conditions. Analysis of 27 publications reporting the presence of BMAA in

cyanobacteria has been made by Berntzon, and the conclusion was drawn

that BMAA production by cyanobacteria depends neither on their taxonomic

group nor on the geographic origin, growth stage, and nutritional status27

.

Naturally, the variability of BMAA occurrence and its possible connection

with neurodegenerative diseases raised concerns about human exposure to

the toxin through food consumption, inhalation, and other routes. Seafood

analysis from the Swedish supermarkets showed that BMAA might be

present in mollusks and crustaceans (in oyster, blue mussel and shrimp

BMAA was detected in 16 samples out of 16 tested) However, it is less

likely to be found in fish and crayfish (BMAA was detected in 5 samples out

of 27 tested)28

. The occurrence of BMAA in human scalp hair indicates that

consumption of shellfish, and, possibly, pork may be a dietary route of

exposure, whereas residential proximity to the water bodies and

10

consumption of vegetables irrigated with water from water reservoir with

algal blooms does not correlate with BMAA content in hair29

. Other human

exposure routes through food might not be as straightforward as seafood

consumption, but should also be explored. For instance, the demand for

algae-derived food supplements increases30

, which makes it more urgent to

investigate their toxicological safety. The process of algal oil (a vegan

alternative to fish oil) production includes extraction of lipid compounds

with supercritical CO231

, and the remaining algal debris can be dried and

further used as the food seasoning known as furikake32

, which should also be

analyzed for algal toxins, specifically BMAA. Furthermore, the fact that blue

mussels are used for feeding hens33

makes poultry an interesting subject to

analyze for the presence of BMAA.

Inhalation, which is another plausible exposure route, was studied using a rat

model, but no clinical signs of toxicity were observed in exposed rats both at

environmentally relevant levels of the aerosolized BMAA and higher

concentrations (0.9 µg – 1.4 g BMAA dose per rat corresponding to 65 days

– 267 years of chronic BMAA exposure through inhalation from

environmental sources)34

. Aerosols collected from a lake in an area with

elevated ALS frequency (25-fold the expected average) were positive for

BMAA35

. There are only a few studies on BMAA availability through

inhalation and its link to ALS, and we undoubtedly need stronger evidence

to make any conclusions on this exposure route relevance, so it is an

interesting path for further investigations.

Studies of developmental exposure to BMAA have suggested that mother’s

milk could be an important pathway of BMAA transfer from lactating

mothers to its offspring in mice36

. Moreover, a preferential enantiomer-

specific uptake of L-BMAA was found in a human mammary cell line. The

efficient absorption of BMAA into intestinal cells and its transport into

neuronal and glial cells suggest that in humans it might be transferred via the

milk from a mother to the brain of the nursed infant37

. BMAA

transplacentral transfer from exposed mothers to the brain of fetus has been

demonstrated in mice38

, suggesting that exposure to BMAA in utero may

influence further offspring development. Autoradiography analysis revealed

11

pronounced BMAA transfer and deposition in the eggs of exposed quails,

predominantly in the yolk39

. Thus, the developmental exposure to BMAA in

birds should also be studied in more detail, and the level of BMAA in eggs

consumed by humans should be evaluated.

Forms of BMAA in natural matrices

In natural matrices, BMAA can be present both as a free compound and/or in

bound form. Therefore the analytical method must be selected wisely to

determine the concentrations of those fractions separately or as total BMAA.

Initially, free BMAA was discovered in cycad seeds, being extracted with

50% ethanol4. Later, the bound form was discovered by the observation that

BMAA concentration in samples increased 10- to 240-fold after

hydrochloric acid hydrolysis (6M HCl, 110°C for 24 h)40

. This discovery led

to introducing strong acid hydrolysis as a routine step in sample preparation

and reporting BMAA in further studies as “total BMAA” or as two separate

fractions of “free BMAA” and “protein-bound BMAA”16

. Although strong

acid hydrolysis is an unquestionably efficient way to cleave peptide bonds, it

might be unsuitable in cases, where it is important to preserve the absolute

configuration of a chiral molecule, since it may cause racemization41,42

.

Overall, determination of total BMAA concentration in the samples does not

provide information regarding its original form, i.e. what specific chemical

species exist in the matrix before hydrolysis, but it gives a good estimation

of the amounts of BMAA one can be exposed to from certain samples.

It has been discovered that BMAA does not exist solely in the protein-bound

form, but also as soluble bound BMAA. That means that after hydrolyzing

trichloroacetic acid (TCA) extracts of free BMAA, the concentration of the

latter increased up to 40%43

, implying the existence of TCA-soluble BMAA-

containing compounds44,45

or what here is referred to as a conjugated form.

Similar observations were made in pioneering studies of BMAA many years

ago, but have not been studied extensively43,46

. Possible soluble-bound forms

12

of BMAA include metal complexes7, carbamate

47,48, imines

49 and low-

molecular weight peptides50

, such as galantine and γ-glutamyl peptide51

.

There also exist a number of lipid compounds conjugated to amino acids,

namely lipoaminoacids and lipopeptides and various forms of these

compounds were discovered in marine organisms52

. Moreover, serine (one of

20 common amino acids), is a part of the most abundant acidic phospholipid

in human cell membranes – phosphatidylserine. Phosphatidylserine also

constitutes up to 20% of the total phospholipids of human blood plasma and

plays a crucial role in the human brain functions by influencing the

metabolism of neurotransmitters53

. Lipid compounds, as possible conjugates

or natural precusors of BMAA, to the best of our knowledge, have not yet

been studied.

L-BMAA toxicity and neurodegenerative diseases

In the half-century of BMAA research, its toxicity has always been in sharp

focus. The idea of the toxicity of BMAA arose from the fact that cycad seeds

are toxic both to animals and humans1. It has been demonstrated that BMAA

solution in low micromolar (10 µM) concentrations can cause selective death

of motor neurons in cortical cultures and that the same damage to the

neurons can be caused by crude cycad extract54,55

. Cox et al. hypothesized

that BMAA in cycads may function as a toxin for herbivores56

. Early

experiments by Vega et al.57

with purified BMAA showed that BMAA-

administered chicks and rats exhibited acute weakness, convulsions and

troubled gait. Furthermore, BMAA exposure experiments with macaques,

showed that the animals developed parkinsonian features and behavioral

anomalies; also degenerative changes were registered in their motor neurons

in the cerebral cortex and spinal cord58

. In a study by Cox and co-workers,

vervet monkeys were fed with L-BMAA or L-BMAA plus L-serine-dosed

fruit for 140 days. Animals developed neurofibrillary tangles (NFT) and β-

amyloid deposits in the brain, however, the animals co-administered with L-

serine expressed a significantly reduced density of NFT59

. Rats and mice

have also been used as animal models for studying BMAA toxicity. In a

13

study by Karlsson et al.60

, neonatal rats were dosed with 14

C-L-BMAA and

the radiographic data was obtained followed by ultra-high performance

liquid chromatography tandem mass spectrometry (UHPLC-MS/MS)

analysis. A dose-dependent increase of protein-associated BMAA in

neonatal rat tissues was observed; and BMAA concentration in the rat liver

was 10-fold higher than in the brain. No BMAA was detected in the samples

from adult rats, which suggests that it had been metabolized and not re-used

in protein synthesis. In other studies by the same group, it has been shown

that the rats neonatally treated with BMAA expressed long-term changes,

such as decrease in learning ability61

, deficits in spatial learning and memory

impairments62

. Rats dosed neonatally with 600 mg/kg of BMAA also

expressed morphological changes in the brain, developing fibrils63

. Long-

term effects of treating rats with BMAA manifested on the molecular level

as the decrease in several proteins expression. Specifically, the expression of

proteins involved in energy metabolism and intracellular signaling has been

registered64

. Other research groups have noted that L-BMAA-treated rats

mimic symptoms presented by ALS patients such as muscular atrophy, brain

structural changes, motor neuronal death and other65,66

. Moreover,

Alzheimer’s and Parkinson’s disease-like symptoms, such as decreased

locomotor activity, body tremors, inability to identify a learned odor and

developmental delays, could simultaneously be observed in BMAA-exposed

rats67

. Finally, BMAA toxicity was demonstrated to be gender and age

dependent, affecting younger rats more than older ones and showing that

different neurodegenerative disruptions manifested in male and female rats;

for instance, male rats exhibited unilateral splay of hind limb while female

rats demonstrated whole body tremors67

.

Linking BMAA to neurodegenerative diseases calls for elucidating the

mechanisms behind its neurotoxicity. The first suggested mechanism of

BMAA toxicity was presented in 1987 and described a concentration-

dependent excitotoxicity through N-methyl-D-aspartate (NMDA) glutamate-

receptor overactivation, manifesting as increased frequency of postsynaptic

vacuoles and dark shrunken cells in mouse cortex explants exposed to

BMAA. The ability of BMAA to block glutamate receptors has been

attributed to structural similarity between glutamate and BMAA carbamate

14

adduct (Figure 2), which can readily form47

in the presence of bicarbonate,

which is considered a required co-factor for BMAA neurotoxicity5. It was

also noticed that the acute neuronal toxicity of BMAA was selectively

blocked by specific antagonists at separate glutamate receptor sites. For

instance, 2-amino-7-phosphonoheptanoic acid (AP7), NMDA antagonist,

was shown specifically to block the neuronal changes produced by BMAA68

.

Another possible mechanism has been discovered in an aquatic plant,

Ceratophyllum demersum, where BMAA inhibits the activity of several

oxidative stress defense enzymes, such as superoxide dismutase, catalase,

guaiacol peroxidase, glutathione peroxidase and glutathione reductase,

therefore allowing reactive oxygen species to damage cell walls, DNA and

proteins69

.

Figure 2. Glutamate and BMAA carbamate adduct are structurally similar70.

Dunlop and co-workers suggested that BMAA may misincorporate into

newly synthesized proteins in human cell lines. This could potentially cause

protein misfolding and lead to the formation of protein plaques in the brain,

characteristic of many neurodegenerative diseases. The hypothesis of

15

BMAA misincorporation was based on the concentration-dependent increase

of the protein-associated radiolabel when exposed to 3H-BMAA, as well as

observing toxicity of BMAA to neuronal cell culture; the latter could also be

attributed to other toxicity mechanisms, not necessarily misincorporation.

Moreover, since radioactivity measurements were used for detection, there

might exist a possibility of radio-label migration giving false positive

results71

. Subsequently, the concept of BMAA misincorporation into proteins

has been criticized. For instance, BMAA and canavanine toxicity and their

possible misincorporation into proteins in various bacterial species was

compared by van Onselen et al. The authors concluded that growth reduction

was not observed for any of the tested bacterial species treated with BMAA

in contrast to those incubated in canavanine-containing media. The

following liquid chromatography mass-spectrometry (LC-MS) analysis

showed no detectable BMAA in proteins produced in an Escherichia coli

expression system72

.

Two abovementioned potential mechanisms of toxicity were further

compared by van Onselen et al.73

using rat pheochromocytoma cell line

PC12. It was concluded that excitotoxicity through activation of mGluR1

glutamate receptors was the main and possibly the only, mechanism of

BMAA toxicity in PC12 cells. Moreover, BMAA capability to

misincorporate into proteins was studied on the human cell lines and no

signs of misincorporation into proteins (i.e., no concentration-dependent

apoptosis or necrosis and no BMAA in purified protein extracts of BMAA-

exposed cells) were observed in contrast with the known misincorporating

analogues L-4-fluorophenylalanin and m-tyrosine in the same cell lines74

.

Nunn and Ponnusamy proposed that BMAA reacts non-enzymatically with

pyridoxal-5’-phosphate, forming 2,3-diaminopropanoic acid and

methylamine. This was observed in model experiments at physiological pH

as well as in rat liver and kidney homogenates incubated with BMAA.

Methylamine release is toxic, because in the organism it converts into

formaldehyde, which in turn participates in protein cross-linking processes

in a variety of tissues49

.

16

Recently, it has been shown that BMAA can inhibit the activity of certain

enzymes, such as β-amylase, catalase and glutathione S-transferase, as well

as associate in vitro with commercial proteins (becoming impossible to

remove by protein precipitation or denaturation) and interfere with protein

folding in the absence of de novo protein synthesis75

. These findings suggest

an additional novel mechanism of BMAA toxicity that needs to be

elucidated to possibly clarify the role of BMAA in the development of

neurodegenerative diseases.

Toxicity of BMAA structural isomers and D-BMAA

The toxicity of BMAA structural isomers, such as BAMA, DAB and AEG

has not been studied extensively, however the neurotoxicity of DAB has

been reported in the early 70s, when it was shown that L-DAB administered

to rats in low mmol/kg body weight doses resulted in hyperirritability,

tremors and convulsions, while D-DAB at even much larger doses did not

cause toxic symptoms76

. A more recent study compared individual and

combined toxicity of BMAA, DAB and AEG in vitro on human

neuroblastoma cells and showed that both BMAA and 2,4-DAB treatment in

µM concentration range resulted in reduction of cell viability, while AEG at

the same concentration was not cytotoxic and its toxicity was observed at

concentrations above 1 mM77

. Moreover, combined treatment with BMAA

and DAB resulted in increased apoptosis compared to the individual

treatment with each compound.

A crucial aspect in studying BMAA toxicity is to distinguish between

stereoisomers and to understand the differences in toxicity between D and L-

BMAA. Although it is known that L-amino acids are predominantly used in

the organisms to build proteins, it is important to give due consideration to

the investigation of D-amino acids and their biological role. This was

summarized in the recent review by Genchi, that contrary to the opinion that

D-amino acids have a minor function in biological processes in comparison

with L-enantiomers, there is a growing body of evidences that D-amino acids

are present in relatively high concentrations in various microorganisms,

17

plants and mammals and fulfil specific biological functions78

. Examples

include marine invertebrates, the cellular fluids of which contain D-amino

acids as a main component; marine shellfish in which D-amino acids can

exceed 1%; amphibian whose skin secretion contains active peptides with

incorporated D-amino acids. Also higher plants, such as pea seedlings,

tobacco leaves, wild rice and lentils contain D-amino acids in free or bound

form. D-amino acids are routinely found in the peptidoglycan cell wall of

bacteria. It has been shown that D-amino acids have regulatory roles in

diverse bacteria. They regulate cell wall remodeling and cause biofilm

dispersal in aging bacterial communities79

.

It was believed that L-BMAA is neurotoxic, while D-BMAA had less or no

harmful effect on the organisms46,57

. In a later study by Andersson et al.37

it

was demonstrated that the uptake of L-BMAA in rats was much higher

compared to D-BMAA. However, a more recent study by Metcalf et al.

assessed in vitro toxicity of D-BMAA using mixed cortical cell cultures from

mice and found it to be comparable to the toxicity of L-BMAA80

. In the same

study it was also reported that L-BMAA used for dosing was converted into

D-BMAA in mouse brain and vervet monkey cerebrospinal fluid but

remained in the form of L-BMAA in the peripheral nervous system of mice.

Moreover, D-BMAA was found in mice liver, suggesting that racemization

might occur during digestion. Altogether, the most recent research calls for

toxicological studies of the D-enantiomer, as well as analytical methods

development that would take BMAA chirality into consideration.

Analytical strategies for BMAA

Historically, BMAA detection and quantification has been performed with

various analytical methods, including different separation techniques, such

as capillary electrophoresis81

(CE), gas chromatography82

(GC) and liquid

chromatography (LC), in combination with a selection of detectors

(fluorescence (FLD), ultraviolet81

(UV), mass spectrometry82

(MS) and

tandem mass spectrometry (MS/MS)). Additionally, proton nuclear magnetic

resonance spectroscopy (1H-NMR) was used in one study

83 and

18

commercially available enzyme linked immunosorbent assay (ELISA) for

BMAA determination in environmental samples was evaluated.

Unfortunately, the latter gave unacceptable false positive results and was not

found to be useful for BMAA analysis84

.

One of the commonly used methods for a long time for BMAA analysis has

been high performance liquid chromatography (HPLC)-FLD10,15,85,86

preceded by pre-column derivatization with 6-aminoquinolyl-N-

hydroxysuccinimidyl carbamate (AQC). However, FLD detection is less

selective than MS/MS and thus tends to produce false positive results. For

instance, two publications reporting unusually high BMAA content in

sharks87

and flying foxes88

(10-250 µg/g range compared to normally

reported 0.1-1 µg/g range for wild fish and animals) used HPLC-FLD for

quantification. That puts the accuracy of the reported results in question.

Overall, in comparison with MS/MS, FLD detection has less stringent

identification criteria for the analyte and cannot be the method of choice for

accurate and precise BMAA analysis86

.

Currently, the method of choice for BMAA analysis is HPLC or UHPLC

coupled to tandem mass spectrometry. This approach may involve analysis

of either derivatized or non-derivatized BMAA. For the derivatization,

normally a commercial AccQ∙Tag Kit with AQC reagent from Waters is

used prior to chromatography on C18 column. Other derivatization

techniques rely on using propyl chloroformate72

and dansyl chloride89,90

as

derivatization agents.

In this thesis, novel strategies of BMAA derivatization were studied. For

instance, the successful use of N-hydroxysuccinimide ester of N-

butylnicotinic acid led to the efficient separation of BMAA from its

structural isomers and higher sensitivity in detecting BMAA compared to the

AQC derivatization (Papers I and II). Specific analytical strategies are

required for the enantiomeric analysis of BMAA. In some methods,

derivatization serves the purpose to create diastereomers and make

enantiomer separation possible. One methodology is a pre-column

derivatization with o-phthaldialdehyde followed by the detection by

fluorimetry91

. However, this approach is not currently recognized as a

19

method with sufficient selectivity. Another approach using (+)-1-(9-

fluorenyl)-ethyl chloroformate (FLEC) derivatization has been proposed in

this thesis (Paper III). In yet another method described by Metcalf et al., the

underivatized enantiomers of BMAA are separated by HPLC on the chiral

chromatographic column and collected as fractions80

. The advantage of

derivatization is the added molecular mass and, in certain cases, increased

hydrophobicity, which leads to a better separation on the reversed phase

chromatographic column and can generate more detectable fragments in

tandem mass spectrometry. For underivatized BMAA analysis, hydrophilic

interaction liquid chromatography (HILIC) columns are commonly used.

However, HILIC-MS/MS methods do not resolve the issue of BMAA

structural isomers except DAB26,92

and lack diagnostic fragmentation for

AEG26

.

As mentioned before, BMAA may be present in samples in different forms:

free BMAA, protein-bound BMAA and in conjugated forms other than to

protein. This defines the choice of appropriate analytical method for sample

pretreatment depending on which fraction is the focus of a certain study44,45

.

20

21

Aims of the thesis

Investigating BMAA requires sensitive, reliable, and validated analytical

methods. New instrumental and sample preparation strategies are to be

explored to provide an adequate toolbox for confident BMAA detection and

quantification in various biological matrices. Of particular interest is chiral

BMAA analysis, which is not extensively covered in current literature. Apart

from the analytical method development, it is crucial to understand the yet

unknown natural functions of BMAA and its possible dietary routes for

humans. To facilitate further elucidation of BMAA identification,

distribution and toxicity, the following aims were set:

To develop and analytically evaluate a more sensitive method for BMAA

quantification than the current method of choice, i.e., LC-MS/MS using

AQC derivatization.

To develop a quantitative method for the chiral analysis of BMAA.

To apply existing and/or developed methods to detect and quantify

BMAA in oil matrices, such as food supplements (fish and algal oil).

To conduct screening of BMAA presence in seafood from Swedish

supermarkets.

To explore BMAA distribution in pelagic and benthic Baltic Sea food

webs.

22

23

Methods

Sample pre-treatment and derivatization

Samples for the projects were either bought from commercial sources like

supermarkets (seafood for Papers III & IV, olive oil for Paper II) and

pharmacies (fish oil for Paper II) or received from collaborators (algae oil

for Paper II, cycad seeds for Paper III and Baltic Sea fauna for Paper V).

Main steps of sample preparation included homogenization, strong acid or

enzymatic hydrolysis, clean-up (filtering through spin filters, optional liquid-

liquid extraction with chloroform (LLE) and solid phase extraction (SPE))

followed by derivatization.

Rough homogenization was performed by mortar and pestle in liquid

nitrogen and was followed by sample homogenization by ultrasonication in

MilliQ water. Conditions of hydrolysis and the necessity of an LLE step

varied depending on the sample material and the detailed description can be

found in the corresponding papers (Papers I-V). SPE was always performed

according to the same protocol (column subsequent conditioning with

methanol and 0.1% formic acid in water, sample application, washing with

0.1% formic acid in water and methanol and elution with 5% NH4OH in

methanol). However, the derivatization step varied.

Different types of derivatization were studied and applied to address various

aspects of BMAA research. Briefly, a well-established derivatization

technique with commercial AccQ∙Tag Kit (AQC) from Waters was used in

Paper IV and Paper V to quantify BMAA in seafood and Baltic Sea food

webs respectively. Another derivatization technique with specially

synthesized N-hydroxysuccinimide ester of N-butylnicotinic acid (C4-NA-

NHS) was developed in Paper I and implemented in Paper II to quantify

BMAA extracted from spiked oil matrices. In Paper III, derivatization was

used to separate BMAA stereoisomers by adding FLEC reagent, which has

not been reported to be used previously with BMAA. The BMAA molecule

derivatized with each of three derivatization agents, used in the current

thesis, is shown in Figure 3.

24

Figure 3. BMAA (blue color) derivatized with AQC, C4-NA-NHS and FLEC reagents.

MS/MS fragmentation patterns shown for each compound.

UHPLC separation

In the framework of the current thesis, all chromatography was performed on

HPLC/UHPLC instruments equipped with different columns and using

different mobile phases according to the task. The following instruments

were used: Agilent G6410B Triple Quad LC/MS system (Santa Clara, CA,

USA); another system consisting of an Accela pump, a degasser and an

Accela auto-sampler coupled with TSQ Vantage triple quadrupole mass

spectrometer (Thermo Fisher Scientific, San Jose, USA) and another system

consisting of Acquity UPLC coupled with Xevo TQ-S micro mass

spectrometer (Waters, Milford, MA, USA).

For the separation of C4-NA-NHS and FLEC derivatized analytes a Hypersil

Gold C18 column (100×2.1 mm, 3 μm particle size; Thermo Scientific,

USA) was used. Elution was performed according to the linear gradient

25

elution program with a binary mobile phase (solvent A: 0.3 % or 0.075%

acetic acid in water; solvent B: 0.3 % or 0.075 % acetic acid in acetonitrile

for C4-NA-NHS or FLEC derivatives, respectively) at a flow rate of

300 μl/min. The first minute of elution was diverted to waste to avoid

introducing buffer salts into the mass spectrometer.

The AQC derivatized standards and analytes were separated on an ACCQ-

TAG ULTRA C18 column (100×2.1 mm, 1.7 µm particle size; Waters,

Ireland). Gradient elution with a binary mobile phase (solvent A: 0.3 %

acetic acid and 5% acetonitrile in water; solvent B: 0.3 % acetic acid in

acetonitrile) was run at a flow rate of 200 µl/min for 10 min and then 400

µl/min for 6 min. The post-column addition of solvent B at the flow rate of

600 µL/min was introduced to improve the detection. For the first two

minutes elution was diverted to waste to avoid introducing buffer salts into

the mass spectrometer.

MS/MS detection

Triple quadrupole mass analyzers consist of three four-rod quadrupoles

connected in series. The first and the third quadrupole are mass

spectrometers and the center quadrupole is a collision cell, where a collision

gas can be introduced and entering ions can undergo one or several

collisions93

.

In all studies, ion source was electrospray operating in positive ionization

mode (ESI+). Detection was performed in multiple reactions monitoring

(MRM) mode, where both first and third quadrupole were set to determine

ions with defined m/z. During the method development stage, product ion

scan mode was used to determine distinctive fragmentation patterns for each

compound derivatized with different tags, which means that the first

quadrupole selected the defined precursor ion and the third quadrupole

scanned m/z of all transmitted ions. In both modes, collision gas (argon) was

introduced to the second quadrupole (collision cell). For the consistency of

the results, collision energies and instrument parameters were tuned

26

specifically for each instrument. Schematic depiction of the triple

quadrupole instrument and its working modes can be seen in Figure 4.

A

B

Figure 4. Schematic diagram of a triple quadrupole instrument. A) Q1 and Q3 are mass

spectrometers, the center quadrupole, q2, is a collision cell (quadrupole using radio-frequency

(RF) only); B) Different operation modes of a tandem mass-spectrometer93.

To ensure that BMAA and each of its isomers were detected unambiguously,

four criteria were used for the identification: 1) compound retention time; 2)

presence of precursor ion; 3) presence of characteristic product ions (both

general and diagnostic); and 4) constant peak area ratio between two chosen

product ions. The complete workflow is summarized in Figure 5.

27

Figure 5. Diagram of sample preparation and analysis workflow, used in all Papers. Some

steps were excluded depending on the nature of the samples.

28

29

Results and discussion

Papers presented in the current thesis illustrate different aspects of analytical

chemistry: sampling; method development of sample treatment (Papers II

and IV) and instrumental analysis (Papers I, II, III and IV); method

application to matrices not previously analyzed for BMAA (Paper II);

method evaluation (Paper II, III and IV) and screening studies to deepen the

knowledge regarding BMAA natural distribution (Papers II, IV and V).

Developing novel derivatization methods

Paper I. Improved detection of BMAA using C4-NA-NHS for the localization

of BMAA in blue mussels.

In Paper I, chromatographic separation of BMAA and its structural isomers

was achieved after derivatization with C4-NA-NHS. Singly labelled species

of AEG and BAMA as well as doubly labelled species of BMAA and DAB

were predominant. For each compound, a characteristic fragmentation

pattern was obtained by MS/MS and diagnostic fragments were identified.

C4-NA-BMAA has three diagnostic product ions, at m/z 137.3, 192.9 and

249.1 (Figure 6), which allows more reliable determination of BMAA

compared to using only one diagnostic product ion at m/z 258.0 when AQC

derivatization is used6. It also allows quantification based on diagnostic ions,

which decreases the probability of false positive results, which can be rarely

done with AQC-derivatized BMAA due to the low abundance of diagnostic

product ion at m/z 258.0 compared to general product ion at m/z 119.0.

Calibration curves were produced for both C4-NA-NHS and AQC

derivatized BMAA standards in the range 100-1000 µg/L. These showed

that the curve for C4-NA-BMAA has a three-fold higher slope and thus

higher sensitivity (Figure 7). Instrumental limit of detection (LOD, defined

as S/N=3) of 40 μg/L was reached for C4-NA-BMAA.

30

Figure 6. Product ion spectra of doubly charged C4-NA-BMAA (m/z 221.2, CE=12, panel (a)

and singly charged AQC-BMAA (m/z 459.3, CE=20, panel (b).

Figure 7. Calibration curves for C4- NA-BMAA and AQC-derivatized BMAA. Higher

sensitivity is observed for C4-NA-BMAA.

31

The presence of BMAA in blue mussel tissues was detected using C4-NA-

NHS tag and two diagnostic ions were present for C4-NA-BMAA. For AQC

derivatized BMAA the diagnostic ion at m/z 258.0 was absent. This suggests

that signal enhancement through matrix effect is more pronounced for C4-

NA-NHS than for AQC derivatization probably due to the permanent

positive charge of the quaternary ammonium. In the electrospray ionization,

permanently ionized nicotine tag possibly facilitates charge repulsion, thus

favoring Coulomb explosion and faster desolvation. For AQC-derivatized

BMAA signal enhancement through matrix effect of around 10% was

observed and reported previously94

.

BMAA localization in blue mussels was investigated. Fresh mussels were

dissected in visceral (digestion organs, dark color) and non-visceral (flesh,

light color) parts and analyzed. BMAA was detected in both parts, which

indicates that depuration (method used to get rid of toxins from shellfish) is

unsuitable for BMAA removal.

Paper II. Analysis of BMAA in fish oil and its vegan alternatives

Although the method developed in Paper I showed promising results for

increased sensitivity in BMAA detection, it was not a quantitative study

because the tag used was not yet so well characterized regarding its stability.

Thus, the method did not employ the use of internal standard, which is

crucial for BMAA quantification in native samples, since it compensates for

analyte loss during sample preparation. Internal standard (d3-BMAA) was

introduced during further method development. The mass transitions for

deuterated BMAA have been determined at m/z 222.73>195.90 and

222.73>140.11, and the internal standard was successfully incorporated in

the method, which increased its reliability for quantification purposes. The

developed quantitative method accuracy was determined from quality

control (QC) samples as 94-99% and precision was 0.9-1.8% RSD. Method

limit of quantification (LOQ) was determined from low concentration (1-10

µg/L) standards S/N ratio. Standards of 10 µg/L demonstrated stable S/N>10

and thus 10 µg/L was determined to be LOQ concentration for the method.

32

Method performance was tested on food supplement samples. Since there are

evidences that BMAA might be present in the natural matrices in conjugated

forms, which are not protein-bound BMAA44,45

, determination of the

possible forms is an interesting challenge. In the current paper the focus was

on the oil matrices, which have not been studied for possible BMAA

presence before. Fish oil, algal oil and olive oil samples were subjected to

hydrolysis, which required the development of a new approach compared to

previous studies on BMAA quantification to hydrolyze lipids. The new

hydrolysis method was successfully applied and its efficiency was proven by 1H-NMR analysis. BMAA presence was not detected in any of the analyzed

samples.

Cycad seed samples (n=3) were analysed as a positive control and BMAA

presence was quantified as 180±11 µg/g wet weight, which is in agreement

with another BMAA quantification in the same material by another method

(Paper III). To further evaluate method reliability, all analysed food

supplements and olive oil (n=3 for each sample) were spiked with standard

L-BMAA solution (1000 µg/L) and subjected to the same sample preparation

protocol as the native samples. Subsequently, BMAA concentration was

quantified in the spiked samples, and the following recoveries were

obtained: 87% for fish oil; 95% for olive oil; 83% for pigment free algal oil

and 91% for algal oil.

Although it was clear that BMAA was not detectable in all analyzed

samples, a peak was observed on the AQC-derivatized algal oil sample

chromatogram corresponding to the diagnostic mass transition for BMAA

(459>258) at the retention time of BMAA (Figure 8). No peak was observed

for the general MRM transition of AQC-derivatized BMAA (459>119) as

well as no peaks of BMAA were observed in C4-NA-NHS-derivatized

samples. However, the presence of the potentially confounding interfering

peak for 459>258 transition introduces the risk of data misinterpretation and

potential false negative or false positive results. If BMAA were present in

the sample, the interfering peak would skew the peak area ratio between

general and diagnostic transitions of BMAA by contributing to the peak area

of diagnostic BMAA transition. This would lead to not meeting one of the

33

peak identification criteria (constant peak area ratio between two chosen

product ions) and therefore to the erroneous interpretation of data. In turn,

C4-NA-NHS derivatization does not share the described issue and could be

recommended as a complementary parallel method to the routine AQC

derivatization of BMAA, serving to resolve cases of having false negatives

or false positives.

Figure 8. Chromatograms of transitions

characteristic for BMAA, DAB and d3-

BMAA (IS) monitored in AQC-derivatized

algal oil sample. A peak for the general

transition of BMAA (459>119) is absent,

while a peak for diagnostic transition

(459>258) is present. This may introduce

false negative or false positive results in cases

when the sample indeed contains BMAA by

skewing the peak area ratio.

Figure 9. Chromatograms of transitions

characteristic for BMAA, DAB and d3-

BMAA (IS) in C4-NA-NHS-derivatized

algal oil sample. An unidentified compound

gives rise to the peaks around 7-7.5 min with

m/z transitions 221>193 and 221>123.

34

Although BMAA and its isomers were not detected in the C4-NA-NHS-

derivatized samples based on peak-identification criteria, the intense peaks

for the transitions 221>193 and 221>123 similar to BMAA and DAB,

respectively, were observed at the retention time around 7.0-7.5 min (Figure

9). These are actually suggestive indicators of possible unknown isomers.

Speculating on their identities, one can hypothesize that it can be doubly-

labelled BAMA. In a previous study (Paper I) BAMA was detected only as

singly-labelled species, but theoretical fragmentation pattern of doubly-

labelled BAMA (Figure 10) shows that the precursor ion at m/z=221 can

potentially fragment into products with m/z=193 and 123. According to

elution order of singly-labelled compounds determined in Paper I, BAMA

elutes after BMAA close to DAB. If doubly-labelled compounds follow the

same elution order, observed retention time is consistent for BAMA.

However, further studies should be carried out to clarify the identity of the

observed peaks and the identity of potential yet-unidentified BMAA

isomers.

Figure 10. Suggested fragmentation pattern for C4-NA-NHS-derivatized doubly-labelled

BAMA species. Parent ion mass m/z=221; fragment masses m/z=193, 179 and 57. A fragment

with m/z=123 can be obtained from m/z=179 by loss of alkyl group with m/z=57.

35

Paper III. Chiral separation of BMAA enantiomers after FLEC

derivatization and LC-MS/MS

In Paper III, we focused on establishing a simple and accurate method for

BMAA enantiomer analysis, since the existing methodologies do not address

resolving enantiomers to determine accurately the true amount of either L- or

D-BMAA. A key aspect in this study was the recognition that racemization

of BMAA occurs during strong acid hydrolysis, which is a widely used step

for measuring bound BMAA from complex matrices. Thus, it might be

recommended to use enzymatic hydrolysis with Pronase to avoid

racemization, which was done in the current study (Figure 11).

Standards and samples were derivatized with FLEC prior to UHPLC-ESI-

MS/MS to produce chromatographically separable derivatives of D- and L-

BMAA. FLEC derivatization converts enantiomers into diastereomers†, thus

resulting in different physico-chemical properties and allowing separation on

the common achiral chromatography column. We established quick and

clean separation of BMAA from its structural isomers as well as of BMAA

enantiomers. Useful fragment ions for confident identification and

quantification were determined. D-BMAA presence in cycad seeds was

discovered, which is an important novel finding. The enantiomeric

composition of BMAA in cycad seed was 50.13 ± 0.05 and 4.08 ± 0.04 µg

for L- and D-BMAA/g wet weight, respectively (n=3).

Although it was claimed that enzymatic hydrolysis should be preferred for

enantiomer quantification, it must be noted that it is less efficient than strong

acid hydrolysis. We quantified total BMAA in the Cycad sample after HCl

hydrolysis to be 133 µg BMAA/g wet weight and after Pronase hydrolysis it

was 54 µg BMAA/g wet weight. Our findings are in accordance with reports

on albumin hydrolysis, where the evaluated hydrolysis efficiency was 95%

and 25% for 6M HCl and Pronase, respectively95

. These findings indicate the

need for further development of the hydrolysis protocol, which will yield

higher outcome while simultaneously preserving the absolute configuration

† Diastereomers of FLEC-derivatized BMAA have different configurations at one of

the equivalent stereocenters and, in contrast with enantiomers, are not mirror images

of each other.

36

of enantiomers for unambiguous determination of those. Probably, other

enzymatic cocktails may be tested to find a more efficient one. Another

hydrolysis improvement option may include protein denaturation followed

by internal disulfide bonds reduction and subsequent alkylation to expose

more sites to proteolytical digestion95

.

Moreover, an experiment with cycad seed samples, incubated with the same

buffer, but without Pronase has been performed. Surprisingly, BMAA was

detected in those samples as well and was quantified to be 50 µg BMAA/g

wet weight. That can be interpreted as the fact that BMAA in the cycad seed

occurs in conjugated form, which not necessarily protein-bound, but rather a

fraction called soluble-bound in the literature44,45

.

Figure 11. MRM chromatograms of L-and D standard solutions (1 mg/L). A) L-BMAA and D-

BMAA, no treatment; B) L-BMAA and D-BMAA after acid hydrolysis demonstrate that

racemization occurs; C) L-BMAA and D-BMAA after enzymatic hydrolysis do not racemize.

The method evaluation parameters were evaluated from QC samples (n=5)

and were determined as follows: accuracy 98-108%, intraday precision 1.0-

37

2.7 RSD % and interday precision 1.3-2.4 RSD %. LOQ was estimated to be

0.1 mg/L for both L- and D-BMAA.

The proposed methodology can be used to determine the enantiomeric

composition of BMAA in other biological matrices and improve the

understanding of the contribution of enantiomers to the biological role of

BMAA and its toxicity. Despite that it has been believed that D-BMAA has

less pronounced to no toxic effect on the organisms46,57

, later studies

demonstrated that in vitro toxicity of D-BMAA on cell cultures is

comparable to the toxicity of L-BMAA80

. The same study reported that L-

BMAA conversion into D-BMAA occurs in vivo in mouse brain and liver as

well as in vervet monkey cerebrospinal fluid. That requires more

investigation of possible in vivo BMAA racemization and its toxicological

assessment.

Screening studies

Papers IV and V present screening studies, where BMAA was determined in

seafood from Swedish supermarkets as well as in pelagic and benthic food

webs of the Northern Baltic Proper.

Paper IV. Quantification of BMAA in seafood from Swedish markets

Taking into consideration that yearly seafood consumption in Sweden has

grown from 8.2 kg per capita in 2011 to 11.0 kg per capita in 2015 and is

still growing96

, it is crucial to determine potential dietary exposure to BMAA

through seafood. In Paper IV we present data on the occurrence of BMAA in

seafood available to the general public. Samples were obtained from popular

supermarkets in Stockholm. We detected BMAA in all blue mussel, oyster,

shrimp and plaice samples with little variation in concentrations between the

samples of the same species. We also detected BMAA in one out of three

samples of Baltic herring and char. Levels of BMAA in the positive-tested

seafood varied between 0.01 and 0.90 mg/kg seafood. However, no BMAA

38

was detected in crayfish and all other fish (salmon, cod and perch) samples

using the validated quantitative method with LOQ<0.01 mg BMAA/g wet

tissue.

Comparing with other studies containing quantitative data on BMAA in

environmental samples, our results are in the agreement with the studies

using LC-MS/MS for quantification8,97

and show lower values than those

obtained by quantification with LC-FLD method12,98

. This discrepancy adds

evidence for the tendency of LC-FLD method to give overestimated values

for BMAA concentration in the samples due to the method’s lower

selectivity. Nevertheless, the reason for the significant difference of several

magnitudes in the reported values could also be attributed, to a certain

extent, to the diverse nature of the samples (species, lifestyle, environmental

conditions).

Paper V. Is there BMAA transfer in the pelagic and benthic food webs in the

Baltic Sea?

In addition to BMAA screening in the seafood, it is important to elucidate

BMAA distribution according to trophic pathways. In Paper V, we analyzed

BMAA in defined pelagic and benthic food webs in the Baltic Sea (Northern

Baltic Proper) applying the same analytical method as in Paper IV. To

confirm that all animals in our study occupied the trophic positions as

expected and belonged to the same food web, we applied stable isotope

analysis (SIA). SIA uses elemental analyzer coupled to isotope ratio mass

spectrometer to detect ratios 14

N:15

N and 12

C:13

C, thus providing information

about trophic level of the analyzed organism to clearly distinguish producers

from consumers of various levels and define a food web99

.

In our study, only pelagic (phytoplankton, zooplankton and mysid shrimps)

organisms tested positive for BMAA, whereas none of the benthic

invertebrates and fish species as well as the sediment were found to contain

BMAA. In the BMAA-positive samples, the concentration varied between

7.5 to 18.1 µg/g dry weight, which is about 100-fold higher than previously

39

reported concentrations for BMAA in the Baltic Sea plankton by Jonasson et

al9. The conflicting results might be explained by the fact that BMAA

quantification in the Jonasson et al. study was based on a semi-quantitative

method, with no internal standard and no reported LOD. Their method

involved a clean-up procedure, which can introduce a significant (up to

90%44

) loss of BMAA. The absence of the internal standard that is subjected

to the same sample preparation procedures as the analyte can lead to the

underestimated concentrations of BMAA. For our study we used the

evaluated method28

with the reported LOQ<0.01 mg BMAA/g wet tissue.

Results suggest that BMAA was not present in the benthic food web, and

levels in the pelagic food web are below the risk level considering the effect

concentrations reported for invertebrates100–102

. However, to see the full

picture of BMAA distribution in various ecosystem compartments, a

systematic sampling throughout a year and in different parts of the Baltic

Sea should be conducted.

Other analytical strategies and applications

In addition to the studies described above, there are several on-going

projects that are important for BMAA quantification in the environmental

samples and understanding its natural pathways.

Investigation of novel derivatization possibilities

One of the studies addressed applicability of BMAA derivatization tags that

have not been used in the other studies of the thesis. These tags were N-

methyl-aminoquinolyl-N-hydroxysuccinimide (N-methyl-AQ-N-NHS) and

N-butyl-aminoquinolyl-N-hydroxysuccinimide (N-butyl-AQ-N-NHS)

(Figure 12). The idea behind designing these tags was to have permanent

positive charge and simultaneously introduce additional aromatic rings to

increase hydrophobicity of the molecule, which would be favorable for

ionization of the analyte in ESI source.

40

O

N+

N

O

O

O

CH3

N-methyl-aminoquinolyl-NHS N-butyl-aminoquinolyl-NHS

Figure 12. Alternative tags for BMAA derivatization. N-butyl-aminoquinolyl-NHS has

longer side chain, which increases the hydrophobicity of the molecule.

Using N-methyl-AQ-N-NHS for BMAA derivatization, we were not able to

detect any signal from BMAA or its isomers. However, analysis of N-butyl-

AQ-N-NHS BMAA-derivatives was far more promising. In Full Scan

positive ion mode, the molecular ion at m/z=271.0 was detected, which

corresponds to the BMAA molecule with the tag molecule attached to both

amino groups (doubly labelled and doubly charged species). Furthermore,

the product ion scan mode was applied to detect the fragmentation pattern of

the derivatized species. Two characteristic product ions at m/z 199.0 and

m/z 174.1 were determined at sufficient intensities to be qualified for using

in MRM mode.

Standard solutions of BMAA in concentrations ranging from 1 to 10 ppm

were derivatized with AQC, C4-NA-NHS and N-butyl-AQ-N-NHS and

analyzed in MRM mode to compare the performances of these tags. As seen

in Figure 13, BMAA derivatized with N-butyl-AQ-N-NHS demonstrated the

highest response at m/z 271.0>174.1, which is also higher than the response

from BMAA derivatized with AQC and C4-NA-NHS tags. More

development and evaluation is required to establish usable and robust

quantitative methods employing N-butyl-AQ-N-NHS for derivatization, but

this study shows that there is a potential for development of methods based

on novel derivatization techniques for BMAA.

O

N+

N

O

O

O

CH3

41

Figure 13. Calibration curves for BMAA (0.1-10 ppm) derivatized with AQC, C4-NA-NHS

and N-butyl-AQ-NHS tags with N-butyl-AQ-NHS demonstrating the highest signal.

Although novel derivatization strategies look promising, it must be stressed

that C4-NA-NHS and N-butyl-AQ-N-NHS tags are not yet commercially

available, but were custom synthesized for us. Thus, the stability and shelf-

life of the reagents have not been evaluated. During our studies involving

C4-NA-NHS derivatization agent, it was noted that proper storage conditions

(tube or vial with tight lock, covered in parafilm and aluminum foil, kept in a

dessicator) were crucial to preserve the reagent as stable as possible

preventing it from hydrolysis. The shelf-life of the reagent under proper

storage conditions was about 8-9 months and it cannot be recommended for

use after that time. The appearance of tag gradually changes with age from

dark-yellow coarse powder to dark-brown sticky liquid due to hydrolysis by

water vapor in the atmosphere, so the visual inspection of the tag is

suggested before use. Further detailed investigation should be performed to

determine storage time more precisely.

0

500000

1000000

1500000

2000000

2500000

0 2 4 6 8 10

Pe

ak a

rea

BMAA concentration, µg/L

AQC 459>258

C4-NA-NHS 221>193

N-butyl-AQ-NHS 271>174

N-butyl-AQ-NHS 271>199

42

Sea spray aerosol as possible exposure route to BMAA.

Potential aerosolisation of BMAA was studied using the Baltic

cyanobacterium Nodularia spumigena under controlled conditions. More

specifically, we aimed to evaluate whether these cyanobacteria and their

metabolites (such as BMAA) could be aerosolized in laboratory conditions.

The Baltic Sea is known to have extensive summer cyanobacterial blooms103

,

which are a growing environmental concern due to being harmful to a wide

range of organisms. The risks of marine toxins to humans have been

evaluated based on seafood consumption as the major route of human

exposure. However, inhalation of sea spray containing cyanobacteria or their

metabolites may be an alternative exposure pathway which has not yet been

studied.

Preliminary experiments, using a sea spray chamber (Figure 14) were

conducted.

Figure 14. Schematic picture of the sea spray chamber used to study BMAA aerosolisation104.

43

Figure 15. BMAA content on filters from the sea spray chamber. BMAA was detected on low

pressure impactor (LPI) stages 2, 7 and 14 (n=5) in different amounts. The medians are

marked with red horizontal lines, the ends of the box are the upper and lower quartiles and the

whiskers extend to the highest and lowest observations.

Briefly, artificial seawater was spiked with BMAA standard and d3-BMAA

was added as an internal standard. Sample was aerosolized in the sea spray

chamber with low pressure impactor (LPI) containing a set of filters with

different pore sizes (16 nm to 10 µm) and arranged with 14 levels of

decreasing pore size. BMAA was collected from several levels of filters,

namely 16 nm, 250 nm and 2.46 µm, and analyzed with UHPLC-MS/MS on

a system consisting of an Accela pump, a degasser and an Accela auto-

sampler coupled to TSQ Vantage (Thermo Fisher Scientific, San Jose, USA)

triple quadrupole mass spectrometer.

The results showed that different amounts of BMAA could be found on the

filters with the certain diameters (Figure 15). Highest BMAA concentration

was detected on stage 7 with pore size 250 nm, while filters on stage 2 (pore

size 16 nm) and stage 12 (pore size 2.46 µm) showed values close to blank.

44

These results point to the fact that filter’s size for BMAA analysis in

aerosols should be selected very carefully and suggest that 250 nm pore size

is suitable for BMAA collection from aerosols. Thus, previous studies that

report poor BMAA aerosolisation14,34

may have underestimated BMAA

concentration due to unsuitable pore size of the filters used (100 µm).

Next steps to be taken in this project include use of cyanobacterial culture at

different stages of the bloom in the sea spray chamber to determine the

BMAA transfer to aerosols as a function of the cyanobacterial abundance

and bloom stage.

45

Conclusions

The current thesis covered analytical aspects of BMAA research, more

specifically the development of new, more sensitive analytical methods for

BMAA quantification in biological matrices as well as the application of

developed and existing methods to natural samples to address issues on

BMAA distribution and possible public exposure.

A new method for BMAA quantification using C4-NA-NHS derivatization

agent was developed. It showed better sensitivity in comparison with

routinely used AQC derivatization, due to the permanent charge, which

facilitates compound detection in MS. The developed method was tested on

blue mussel samples and oil samples. Repeatability and accuracy of the

method were evaluated. The developed method can be recommended as a

complementary technique to the routine AQC analysis of BMAA to avoid

false negative or false positive results.

Another quantitative method was developed for the chiral analysis of

BMAA. BMAA optical isomers were successfully separated on a C18

column after derivatization with FLEC reagent. The study provided a crucial

observation that, to avoid racemization, enzymatic hydrolysis should be used

instead of acidic hydrolysis when chirality of the molecule is important. The

method was applied to quantify BMAA in cycad seeds, and the

concentrations were 50.13 ± 0.05 and 4.08 ± 0.04 µg of L- and D-BMAA/g

wet weight, respectively.

Studies for BMAA detection and quantification in seafood from Swedish

supermarkets and Baltic Sea food webs were also carried out. It was

demonstrated that BMAA is present in bivalves (mussels and oysters) and

shrimps as well as in several fish species (plaice, herring and char) in low

µg/g wet weight concentrations. In the Baltic Sea food webs, BMAA was

detected in the pelagic food chain (phytoplankton, zooplankton and mysids)

at 0.75-2.0 µg/g wet weight concentrations, but not in the benthic system;

moreover, no detectable levels were found in fish.

46

47

Further method development & application

Undoubtedly, the BMAA research field is fast-developing and challenging in

many ways. There are still more questions than answers, and due to possible

links to neurodegenerative diseases and ubiquity of BMAA, it remains a

highly relevant topic. Here are some suggestions about next steps to be taken

in analytical method development, application and evaluation in BMAA

studies.

Analytical method development

Further investigations regarding the identity of potential yet-unidentified

BMAA isomers should be carried out based on the observations made in

algal oil samples.

For the chiral analysis of BMAA, it would be beneficial to find the

appropriate enzyme cocktail or to develop an additional treatment method

for efficient hydrolysis without racemization.

Novel derivatization tags should be explored and tested on natural

samples to further improve selectivity and sensitivity of the quantification

methods.

Novel derivatization tags should be explored as a possible route to more

sensitive and robust analysis. Deuterated tags105

, such as isotope-coded

affinity tag (ICAT) and isobaric tags for relative and absolute quantitation

(iTRAQ), might be useful for labeling BMAA from different samples,

which are then pooled and analyzed by LC-MS/MS. This approach will

allow comparing BMAA concentrations in different sources while also

reducing the time required for the analysis.

Method application

For the production of food supplements, BMAA should be monitored at

every production step, including by-products, to get the full picture of

possible exposure routes.

48

Aerosolisation of BMAA should be studied as a possible exposure route

through inhalation. It is important to choose correct filter sizes for the

aerosol collection to perform accurate evaluation.

Chiral composition of BMAA in natural samples should be studied more

extensively, thus elucidating BMAA’s natural role.

Method evaluation

Collaborative efforts and interlaboratory projects should be directed

towards evaluation and validation of existing and newly developed

methods for BMAA quantification. This will reduce controversy about

the results and allow mapping BMAA distribution in various ecosystems

and relevant pathways.

49

Acknowledgements

My more-then-five-year PhD studies journey is coming to its end, and

before arrival to the PhD defense destination I would like to take the

opportunity and thank everyone who has been there on the road together

with me: during all time, part of the journey or just for a couple of stations.

First of all, I would like to thank the Department of Environmental Science

and Analytical Chemistry for the opportunity to study here and grow both

scientifically and personally in a friendly and inspiring atmosphere. I am

especially grateful to the Head of the Unit of Analytical Chemistry Cynthia

de Wit for her understanding, support and kindness.

I am deeply thankful to my supervisor Leopold L. Ilag for his guidance,

genuine support and scientific enthusiasm. It was an honor and a pleasure to

work with you during these years.

I was lucky to have three co-supervisors during the PhD studies and I thank

Ulrika Nilsson, Ingrid Granelli and Elena Gorokhova for their

involvement in my work at different stages and their readiness to help

whenever I needed that.

The PhD studies gave me a gift of meeting and collaborating with wonderful

people and I am thankful for the opportunity to know all of you: Anders

Colmsjö, Conny Östman, Magnus Åberg, Leopold L. Ilag, Roger

Westerholm and Gunnar Thorsén (thank you both for sharing teaching

experience in the department courses), Ulrika Nilsson, Ingrid Granelli, Jan

Holmbäck (thank you for introducing me to lipid chemistry), Vibhu Rinwa,

Karin Englund, Carlo Crescenci, Lena Elfver and Jonas Rutberg (thank

you both for your invaluable help with papers and documents), Emilia

Eklund, Rudolf Andrýs, Isabella Karlsson, Anna Sroka-Bartnicka,

Cynthia de Wit, Margareta Törnqvist, Lillemor Asplund, Hitesh

Motwani, Jenny Aasa, Henrik Carlsson, Ioannis Athanassiadis, Elena

Gorokhova, Oskar Sandblom and Matthew Salter.

50

I thank Oskar Karlsson, Hitesh Motwani, Elena Gorokhova, Cynthia de

Wit and Javier Zurita for the valuable comments on this thesis, which

certainly made it better.

I thank Tanja Russita for creating a very special cover image for my thesis.

I also thank all PhD students for being friendly, open and ready to help.

Most of you are already doctors and it was a pleasure to share our paths and

to witness your achievements: Aljona (thank you for sharing your

experience with instruments and being very supportive at the beginning of

my studies); Liying (thank you for the inspiration and support you gave me

and for our collaboration); Javier (thank you for all the work we did

together and for being so kind); Petter, Giovanna, Erik and Nicolò (thank

you all for sharing the room and having nice chats); Silvia, Rozanna, Jonas,

Hwanmi (thank you for nice conversations and help), Mohammad, Farshid

(thank you for sharing your knowledge and time so generously and for being

so helpful), Sara, Aziza, Trifa, Ioannis and Ramzy. I wish all the best to

the continuing PhD students and thank you for the time spent together:

Alessandro (thank you for the collaborations and always being ready to

help), Pedro (thank you for sharing a room and nice Lego-chats), Francesco

(thank you for sharing experience and helping me with Xevo), Hatem,

Josefine (thank you for nice health-walks and for the Swedish translation of

my abstract), Nikola (thank you for help with aerosols), Nicklas (thank you

for help with the sample preparation), Johan and Lorena.

Being teaching assistant was an important part of my PhD studies, so I want

to thank all the students I met in the courses as well as the exchange students

from France. I am particularly thankful to Patrik, Emil, Petra, Nargiza,

Maïwenn, Lea and Anthony. I learned a lot while teaching you.

I want to thank the external collaborators for their significant input in our

shared projects: Rudolf Andrýs, Klaas Verschueren, Wim M. De

Borggraeve, Johan Rosén, Agnes Karlson, Andrius Garbaras, Simoné

Downing, Laura L. Scott, Timothy G. Downing, Sara Sigurlásdóttir and

Elisabeth J. Faassen as well as organizers and participants of Cyanocost

Training School in Wageningen (2015).

51

My special thanks are for Zhongping Yao from Hong Kong Polytechnic

University who was supervising me during three-month internship in Hong

Kong and to the members of his research group who I was collaborating

with: Betty, Ho Yi and Tommy.

I am sincerely grateful to my friends who were always there for me and gave

me so much love and support: Aleksandr, Juta, Esther, Artur and Julia.

My very special thanks are to my family who always believes in me. I thank

my parents Victor and Lyubov for supporting me and being truly

understanding. I always feel your love no matter the distance.

I deeply thank my husband Petr for his strength, care and support. I am so

happy to share my life and my love with you.

52

References

1. Whiting, M. G. Toxicity of cycads. Econ. Bot. 17, 270–302 (1963).

2. Holtcamp, W. The emerging science of BMAA: do cyanobacteria

contribute to neurodegenerative disease? Environmental health

perspectives 120, 110–116 (2012).

3. Bell, E. A. Relevance of biochemical taxonomy to the problem of

lathyrism. Nature 203, 378–380 (1964).

4. Vega, A. & Bell, E. A. α-Amino-β-methylaminopropionic acid, a

new amino acid from seeds of Cycas circinalis. Phytochemistry 6,

759–762 (1967).

5. Norden. Analysis, occurence, and toxicity of BMAA, A risk for the

consumer? (Norden, 2007). doi:10.15713/ins.mmj.3

6. Jiang, L., Aigret, B., De Borggraeve, W. M., Spacil, Z. & Ilag, L. L.

Selective LC-MS/MS method for the identification of BMAA from

its isomers in biological samples. Anal. Bioanal. Chem. 403, 1719–

1730 (2012).

7. Nunn, P. B., O’Brien, P., Pettit, L. D. & Pyburn, S. I. Complexes of

zinc, copper, and nickel with the nonprotein amino acid L-alpha-

amino-beta-methylaminopropionic acid: a naturally occurring

neurotoxin. J. Inorg. Biochem. 37, 175–183 (1989).

8. Jiao, Y. et al. Occurrence and transfer of a cyanobacterial neurotoxin

β-methylamino-l-alanine within the aquatic food webs of Gonghu

Bay (Lake Taihu, China) to evaluate the potential human health risk.

Sci. Total Environ. 468–469, 457–463 (2014).

9. Jonasson, S. et al. Transfer of a cyanobacterial neurotoxin within a

temperate aquatic ecosystem suggests pathways for human exposure.

Proc. Natl. Acad. Sci. 107, 9252–9257 (2010).

10. Metcalf, J. S. et al. Co-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 (2008).

11. Lage, S. et al. BMAA in shellfish from two Portuguese transitional

water bodies suggests the marine dinoflagellate Gymnodinium

catenatum as a potential BMAA source. Aquat. Toxicol. 152, 131–

138 (2014).

12. Brand, L. E., Pablo, J., Compton, A., Hammerschlag, N. & Mash, D.

C. Cyanobacterial blooms and the occurrence of the neurotoxin, beta-

N-methylamino-l-alanine (BMAA), in South Florida aquatic food

webs. Harmful Algae 9, 620–635 (2010).

53

13. Faassen, E. J., Gillissen, F., Zweers, H. A. J. & Lrling, M.

Determination of the neurotoxins BMAA (β-N-methylamino-L-

alanine) and DAB (α-,γ-diaminobutyric acid) by LC-MSMS in Dutch

urban waters with cyanobacterial blooms. Amyotroph. Lateral Scler.

10, 79–84 (2009).

14. Cox, P. A. et al. Cyanobacteria and BMAA exposure from desert

dust: A possible link to sporadic ALS among Gulf War veterans.

Amyotroph. Lateral Scler. 10, 109–117 (2009).

15. Cox, A. et al. Diverse taxa of cyanobacteria produce N-methylamino-

L-alanine, a neurotoxic amino acid. Proc. Natl. Acad. Sci. 102, 5074–

5078 (2005).

16. Jiang, L. et al. Diatoms: A novel source for the neurotoxin BMAA in

aquatic environments. PLoS One 9, 1–8 (2014).

17. Réveillon, D. et al. β-N-methylamino-l-alanine (BMAA) and

isomers: Distribution in different food web compartments of Thau

lagoon, French Mediterranean Sea. Mar. Environ. Res. 110, 8–18

(2015).

18. Jiang, L. & Ilag, L. L. Detection of endogenous BMAA in

dinoflagellate (Heterocapsa triquetra) hints at evolutionary

conservation and environmental concern. PubRaw Sci. 1, 1–8 (2014).

19. Van Apeldoorn, M. E., Van Egmond, H. P., Speijers, G. J. A. &

Bakker, G. J. I. Toxins of cyanobacteria. Mol. Nutr. Food Res. 51, 7–

60 (2007).

20. Li, A. et al. Detection of the neurotoxin BMAA within cyanobacteria

isolated from freshwater in China. Toxicon 55, 947–953 (2010).

21. Downing, S., Banack, S. a, Metcalf, J. S., Cox, P. a & Downing, T.

G. Nitrogen starvation of cyanobacteria results in the production of β-

N-methylamino-L-alanine. Toxicon 58, 187–94 (2011).

22. Downing, S. & Downing, T. G. The metabolism of the non-

proteinogenic amino acid β-N-methylamino-L-alanine (BMAA) in

the cyanobacterium Synechocystis PCC6803. Toxicon 115, 41–48

(2016).

23. Berntzon, L. et al. BMAA inhibits nitrogen fixation in the

cyanobacterium nostoc sp. PCC 7120. Mar. Drugs 11, 3091–3108

(2013).

24. Krüger, T., Mönch, B., Oppenhäuser, S. & Luckas, B. LC-MS/MS

determination of the isomeric neurotoxins BMAA (beta-N-

methylamino-L-alanine) and DAB (2,4-diaminobutyric acid) in

cyanobacteria and seeds of Cycas revoluta and Lathyrus latifolius.

Toxicon 55, 547–57 (2010).

54

25. Rosén, J. & Hellenäs, K. E. Determination of the neurotoxin BMAA

(β-N-methylamino-L-alanine) in cycad seed and cyanobacteria by

LC-MS/MS (liquid chromatography tandem mass spectrometry).

Analyst 133, 1785–1789 (2008).

26. McCarron, P., Logan, A. C., Giddings, S. D. & Quilliam, M. A.

Analysis of β-N-methylamino-L-alanine (BMAA) in spirulina-

containing supplements by liquid chromatography-tandem mass

spectrometry. Aquat. Biosyst. 10, 1–7 (2014).

27. Berntzon, L. Detection, transfer and role of environmentally spread

neurotoxin (BMAA) with focus on cyanobacteria and the Baltic Sea

region. (Holmbergs, 2015).

28. Jiang, L., Kiselova, N., Rosén, J. & Ilag, L. L. Quantification of

neurotoxin BMAA (β-N-methylamino-L-alanine) in seafood from

Swedish markets. Sci. Rep. 4, 1–7 (2014).

29. Downing, S., Scott, L. L., Zguna, N. & Downing, T. G. Human scalp

hair as an indicator of exposure to the environmental toxin β-N-

methylamino-L-alanine. Toxins (Basel). 10, (2018).

30. Grahl, S., Strack, M., Weinrich, R. & Mörlein, D. Consumer-

Oriented Product Development: The Conceptualization of Novel

Food Products Based on Spirulina (Arthrospira platensis) and

Resulting Consumer Expectations. J. Food Qual. 2018, (2018).

31. Patil, P. D., Dandamudi, K. P. R., Wang, J., Deng, Q. & Deng, S.

Extraction of bio-oils from algae with supercritical carbon dioxide

and co-solvents. J. Supercrit. Fluids 135, 60–68 (2018).

32. Japan Centre. Furikake Rice Seasoning. Available at:

https://www.japancentre.com/en/pages/40-furikake-rice-seasoning.

33. Jönsson, L. & Holm, L. Effects of toxic and non-toxic blue mussel

meal on health and product quality of laying hens. J. Anim. Physiol.

Anim. Nutr. (Berl). 94, 405–412 (2010).

34. Scott, L. L., Downing, S. & Downing, T. G. The Evaluation of

BMAA Inhalation as a Potential Exposure Route Using a rat Model.

Neurotox. Res. 1–9 (2017). doi:10.1007/s12640-017-9742-9

35. Banack, S. A. et al. Detection of cyanotoxins, β-N-methylamino-L-

alanine and microcystins, from a lake surrounded by cases of

amyotrophic lateral sclerosis. Toxins (Basel). 7, 322–336 (2015).

36. Andersson, M., Karlsson, O., Bergström, U., Brittebo, E. B. &

Brandt, I. Maternal Transfer of the Cyanobacterial Neurotoxin β-N-

Methylamino-L-Alanine (BMAA) via Milk to Suckling Offspring.

PLoS One 8, 1–11 (2013).

55

37. Andersson, M., Ersson, L., Brandt, I. & Bergstr�m, U. Potential

transfer of neurotoxic amino acid β-N-methylamino-alanine (BMAA)

from mother to infant during breast-feeding: Predictions from human

cell lines. Toxicol. Appl. Pharmacol. 320, 40–50 (2017).

38. Karlsson, O., Lindquist, N. G., Brittebo, E. B. & Roman, E. Selective

brain uptake and behavioral effects of the cyanobacterial toxin

BMAA (β-N-Methylamino-L-alanine) following neonatal

administration to rodents. Toxicol. Sci. 109, 286–295 (2009).

39. Andersson, M., Karlsson, O. & Brandt, I. The environmental

neurotoxin β-N-methylamino-L-alanine (L-BMAA) is deposited into

birds’ eggs. Ecotoxicol. Environ. Saf. 147, 720–724 (2018).

40. Murch, S. J., Cox, P. A. & Banack, S. A. A mechanism for slow

release of biomagnified cyanobacterial neurotoxins and

neurodegenerative disease in Guam. Proc. Natl. Acad. Sci. U. S. A.

101, 12228–31 (2004).

41. Bada, J. L. Racemization of amino acids. Chem. Biochem. Amin.

Acids 399–414 (1985).

42. Neuberger, A. Stereochemistry of Amino Acids. Adv. Protein Chem.

4, 297–383 (1948).

43. Dossaji, S. F. & Bell, E. A. Distribution of α-amino-β-

methylaminopropionic acid in Cycas. Phytochemistry 12, 143–144

(1973).

44. Faassen, E. J. et al. A collaborative evaluation of LC-MS/MS based

methods for BMAA analysis: Soluble bound BMAA found to be an

important fraction. Mar. Drugs 14, (2016).

45. Rosén, J., Westerberg, E., Schmiedt, S. & Hellenäs, K. E. BMAA

detected as neither free nor protein bound amino acid in blue mussels.

Toxicon 109, 45–50 (2016).

46. Polsky, F. I., Nunn, P. B. & Bell, E. A. Distribution and toxicity of

alpha-amino-beta-methylaminopropionic acid. Fed. Proc. 31, 1473—

1475 (1972).

47. Nunn, P. B. & O’Brien, P. The interaction of β-N-methylamino-L-

alanine with bicarbonate: an1H-NMR study. FEBS Lett. 251, 31–35

(1989).

48. Myers, T. G. & Nelson, S. D. Neuroactive carbamate adducts of β-N-

Methylamino-L-alanine and ethylenediamine. J. Biol. Chem. 265,

10193–10195 (1990).

49. Nunn, P. B. & Ponnusamy, M. Beta-N-methylaminoalanine

(BMAA): metabolism and metabolic effects in model systems and in

neural and other tissues of the rat in vitro. Toxicon 54, 85–94 (2009).

56

50. Cheng, R. & Banack, S. A. Previous studies underestimate BMAA

concentrations in cycad flour. Amyotroph. Lateral Scler. 10, 41–43

(2009).

51. Nunn, P. B. 50 Years of Research on Α-Amino-Β-

Methylaminopropionic Acid (Β-Methylaminoalanine).

Phytochemistry 144, 271–281 (2017).

52. AMINO COMPOUND-CONTAINING LIPIDS. Available at:

http://www.cyberlipid.org/simple/simp0005.htm.

53. Glade, M. J. & Smith, K. Phosphatidylserine and the human brain.

Nutrition 31, 781–786 (2015).

54. Lobner, D., Piana, P. M. T., Salous, A. K. & Peoples, R. W. β-N-

methylamino-l-alanine enhances neurotoxicity through multiple

mechanisms. Neurobiol. Dis. 25, 360–366 (2007).

55. Rao, S. D., Banack, S. A., Cox, P. A. & Weiss, J. H. BMAA

selectively injures motor neurons via AMPA/kainate receptor

activation. Exp. Neurol. 201, 244–252 (2006).

56. Cox, P. A., Banack, S. A. & Murch, S. J. Biomagnification of

cyanobacterial neurotoxins and neurodegenerative disease among the

Chamorro people of Guam. Proc. Natl. Acad. Sci. U. S. A. 100,

13380–3 (2003).

57. Vega, A., Bell, E. A. & Nunn, P. B. Preparation of L- and D-α-

amino-β-methylaminopropionic acids and the identification of the

compound isolated from Cycas circinalis as the L-isomer.

Phytochemistry 7, 1885–1887 (1968).

58. Spencer, P. S. et al. Guam Amyotrophic Lateral Sclerosis--

Parkinsonism--Dementia Linked to a Plant Excitant. Science (80-. ).

237, 517–522 (1987).

59. Cox, P. A. et al. Dietary exposure to the environmental toxin b-N-

methylamino-L-alanine triggers neurofibrillary tangles and amyloid

deposits in the brain. Rspb.Royalsocietypublishing.Org (2016).

doi:10.1098/rspb.2015.2397

60. Karlsson, O., Jiang, L., Andersson, M., Ilag, L. L. & Brittebo, E. B.

Protein association of the neurotoxin and non-protein amino acid

BMAA (β-N-methylamino-l-alanine) in the liver and brain following

neonatal administration in rats. Toxicol. Lett. 226, 1–5 (2014).

61. Karlsson, O., Roman, E. & Brittebo, E. B. Long-term cognitive

impairments in adult rats treated neonatally with β-N-methylamino-

L-Alanine. Toxicol. Sci. 112, 185–195 (2009).

57

62. Karlsson, O., Roman, E., Berg, A.-L. & Brittebo, E. B. Early

hippocampal cell death, and late learning and memory deficits in rats

exposed to the environmental toxin BMAA (β-N-methylamino-L-

alanine) during the neonatal period. Behav. Brain Res. 219, 310–20

(2011).

63. Karlsson, O. et al. Intracellular 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

(2015).

64. Karlsson, O. et al. Neonatal exposure to the cyanobacterial toxin

BMAA induces changes in protein expression and neurodegeneration

in adult hippocampus. Toxicol. Sci. 130, 391–404 (2012).

65. Tian, K. W., Jiang, H., Wang, B. B., Zhang, F. & Han, S. Intravenous

injection of l-BMAA induces a rat model with comprehensive

characteristics of amyotrophic lateral sclerosis/Parkinson-dementia

complex. Toxicol. Res. (Camb). 5, 79–96 (2015).

66. de Munck, E. et al. Morphometric and neurochemical alterations

found in l-BMAA treated rats. Environ. Toxicol. Pharmacol. 39,

1232–1245 (2015).

67. Scott, L. L. & Downing, T. G. B-N-methylamino-L-alanine (BMAA)

toxicity is gender and exposure-age dependent in rats. Toxins (Basel).

10, 18–21 (2018).

68. Ross, S. M., Seelig, M. & Spencer, P. S. Specific antagonism of

excitotoxic action of ‘uncommon’ amino acids assayed in

organotypic mouse cortical cultures. Brain Res. 425, 120–127 (1987).

69. Esterhuizen-Londt, M., Pflugmacher, S. & Downing, T. G. The effect

of β-N-methylamino-L-alanine (BMAA) on oxidative stress response

enzymes of the macrophyte Ceratophyllum demersum. Toxicon 57,

803–10 (2011).

70. Weiss, J. H., Christine, C. W. & Choi, D. W. Bicarbonate

dependence of glutamate receptor activation by β-N-methylamino-l-

alanine: Channel recording and study with related compounds.

Neuron 3, 321–326 (1989).

71. Dunlop, R. A., Cox, P. A., Banack, S. A. & Rodgers, K. J. The Non-

Protein Amino Acid BMAA Is Misincorporated into Human Proteins

in Place of l-Serine Causing Protein Misfolding and Aggregation.

PLoS One 8, (2013).

72. Van Onselen, R., Cook, N. A., Phelan, R. R. & Downing, T. G.

Bacteria do not incorporate β-N-methylamino-l-alanine into their

proteins. Toxicon 102, 55–61 (2015).

58

73. van Onselen, R., Venables, L., van de Venter, M. & Downing, T. G.

β-N-Methylamino-L-Alanine Toxicity in PC12: Excitotoxicity vs.

Misincorporation. Neurotox. Res. 1–9 (2017). doi:10.1007/s12640-

017-9743-8

74. van Onselen, R., Downing, S., Kemp, G. & Downing, T.

Investigating β-N-methylamino-L-alanine misincorporation in human

cell cultures: A comparative study with known amino acid analogues.

Toxins (Basel). 9, 17–27 (2017).

75. van Onselen, R. & Downing, T. G. BMAA-protein interactions: A

possible new mechanism of toxicity. Toxicon 143, 74–80 (2018).

76. Chen, C., Flory, W. & Koeppe, R. E. Variation of Neurotoxicity of L-

and D-2,4 Diaminobutyric Acid with Route of Administration.

Toxicol. Appl. Pharmacol. 23, 334–338 (1972).

77. Main, B. J. & Rodgers, K. J. Assessing the Combined Toxicity of

BMAA and Its Isomers 2,4-DAB and AEG In Vitro Using Human

Neuroblastoma Cells. Neurotox. Res. (2017). doi:10.1007/s12640-

017-9763-4

78. Genchi, G. An overview on d-amino acids. Amino Acids 49, 1521–

1533 (2017).

79. Cava, F., Lam, H., De Pedro, M. A. & Waldor, M. K. Emerging

knowledge of regulatory roles of d-amino acids in bacteria. Cell. Mol.

Life Sci. 68, 817–831 (2011).

80. Metcalf, J. S. et al. Analysis of BMAA enantiomers in cycads,

cyanobacteria, and mammals: in vivo formation and toxicity of d-

BMAA. Amino Acids 49, 1427–1439 (2017).

81. Baptista, M. S., Cianca, R. C. C., Lopes, V. R., Almeida, C. M. R. &

Vasconcelos, V. M. Determination of the non protein amino acid β-

N-methylamino-l-alanine in estuarine cyanobacteria by capillary

electrophoresis. Toxicon 58, 410–414 (2011).

82. Esterhuizen, M. & Downing, T. G. β-N-methylamino-l-alanine

(BMAA) in novel South African cyanobacterial isolates. Ecotoxicol.

Environ. Saf. 71, 309–313 (2008).

83. Moura, S., Ultramari, M. D. A., de Paula, D. M. L., Yonamine, M. &

Pinto, E. 1H NMR determination of beta-N-methylamino-L-alanine

(L-BMAA) in environmental and biological samples. Toxicon 53,

578–83 (2009).

84. Faassen, E. J., Beekman, W. & Lürling, M. Evaluation of a

Commercial Enzyme Linked Immunosorbent Assay (ELISA) for the

Determination of the Neurotoxin BMAA in Surface Waters. PLoS

One 8, 1–8 (2013).

59

85. Cervantes Cianca, R. C., Baptista, M. S., Lopes, V. R. &

Vasconcelos, V. M. The non-protein amino acid β-N-methylamino-l-

alanine in Portuguese cyanobacterial isolates. Amino Acids 42, 2473–

2479 (2012).

86. Faassen, E. J., Gillissen, F. & Lürling, M. A comparative study on

three analytical methods for the determination of the neurotoxin

BMAA in cyanobacteria. PLoS One 7, (2012).

87. Hammerschlag, N. et al. Cyanobacterial neurotoxin bmaa and

mercury in sharks. Toxins (Basel). 8, 1–14 (2016).

88. Banack, S. A., Murch, S. J. & Cox, P. A. Neurotoxic flying foxes as

dietary items for the Chamorro people, Marianas Islands. J.

Ethnopharmacol. 106, 97–104 (2006).

89. Lampinen Salomonsson, M., Hansson, A. & Bondesson, U.

Development and in-house validation of a method for quantification

of BMAA in mussels using dansyl chloride derivatization and ultra

performance liquid chromatography tandem mass spectrometry. Anal.

Methods 5, 4865–4874 (2013).

90. Roy-Lachapelle, A., Solliec, M. & Sauvé, S. Determination of

BMAA and three alkaloid cyanotoxins in lake water using dansyl

chloride derivatization and high-resolution mass spectrometry. Anal.

Bioanal. Chem. 407, 5487–5501 (2015).

91. Euerby, R., Partridge, Z. & Nunn, B. Resolution of neuroactive non-

protein amino acid enantiomers by high-performance liquid

chromatography utilising pre-column derivatisation with o-

phthaldialdehyde-chiral thiols : Application to 2-amino-ω-

phosphonoalkanoic acid homologues and α-amino-β-N-m. J.

Chromatogr. A 469, 412–419 (1989).

92. Combes, A. et al. Validation of the analytical procedure for the

determination of the neurotoxin β-N-methylamino-L-alanine in

complex environmental samples. Anal. Chim. Acta 771, 42–9 (2013).

93. Hoffman, E. de & Stroobant, V. Mass Spectrometry: principles and

applications. (2011).

94. Jiang, L., Johnston, E., Åberg, K. M., Nilsson, U. & Ilag, L. L.

Strategy for quantifying trace levels of BMAA in cyanobacteria by

LC/MS/MS. Anal. Bioanal. Chem. 405, 1283–1292 (2013).

95. Sabbioni, G. & Turesky, R. J. Biomonitoring human albumin

adducts: The past, the present, and the future. Chem. Res. Toxicol. 30,

332–366 (2017).

96. Ziegler, F. & Bergman, K. Svensk konsumtion av sjömat- en växande

mångfald. (2017).

60

97. Christensen, S. J., Hemscheidt, T. K., Trapido-Rosenthal, H., Laws,

E. A. & Bidigare, R. R. Detection and quantification of β-

methylamino-L-alanine in aquatic invertebrates. Limnol. Oceanogr.

Methods 10, 891–898 (2012).

98. Mondo, K. et al. Cyanobacterial neurotoxin β-N-methylamino-L-

alanine (BMAA) in shark fins. Mar. Drugs 10, 509–20 (2012).

99. Michener, R. & Lajtha, K. Stable Isotopes in Ecology and

Environmental Science. (2007).

100. Esterhuizen-Londt, M., Wiegand, C. & Downing, T. G. β-N-

methylamino-l-alanine (BMAA) uptake by the animal model,

Daphnia magna and subsequent oxidative stress. Toxicon 100, 20–26

(2015).

101. Lepoutre, A. et al. Genotoxic and cytotoxic effects on the immune

cells of the freshwater bivalve Dreissena polymorpha exposed to the

environmental neurotoxin BMAA. Toxins (Basel). 10, (2018).

102. Faassen, E. J., García-Altares, M., Mendes e Mello, M. & Lürling, M.

Trans generational effects of the neurotoxin BMAA on the aquatic

grazer Daphnia magna. Aquat. Toxicol. 168, 98–107 (2015).

103. Bianchi, T. S. et al. Cyanobacterial blooms in the Baltic Sea: Natural

or human-induced? Limnol. Oceanogr. 45, 716–726 (2000).

104. Salter, M. E. et al. Calcium enrichment in sea spray aerosol particles.

Geophys. Res. Lett. 43, 8277–8285 (2016).

105. DeSouza, L. V. & Siu, K. W. M. Mass spectrometry-based

quantification. Clin. Biochem. 46, 421–431 (2013).