Amyotrophic Lateral Sclerosis, 2009; (Supplement 2): 56–60

ISSN 1748-2968 print/ISSN 1471-180X online © 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)DOI: 10.3109/17482960903269062

Correspondence: D. Lobner, Department of Biomedical Sciences, Marquette University, 561 N. 15th Street, Rm 446, Milwaukee, Wisconsin 53233, USA. Fax: 414 288 6564. E-mail: Doug.Lobner@marquette.edu

(Received 27 April 2009; accepted 18 August 2009)

ORIGINAL ARTICLE

Mechanisms of b -N-methylamino-L-alanine induced neurotoxicity

DOUG LOBNER

Department of Biomedical Sciences, Marquette University, Milwaukee, Wisconsin, USA

AbstractSince the initial discovery that the amino acid -N-methylamino-L-alanine (BMAA) was a neurotoxin, a great deal has been learned about its mechanism of action. However, exactly how it causes death of motor neurons, and how its actions may interact with other neurotoxins or pathological conditions, is not well understood. The focus of study on the mecha-nism of BMAA toxicity has been on its action as a glutamate receptor agonist. There is evidence that BMAA has effects on all of the main types of glutamate receptors: NMDA, AMPA/kainate, and metabotropic receptors. However, recent results suggest that BMAA may also act through other mechanisms to induce neuronal death. One such action is on the cystine/glutamate antiporter (system xc -). Through its effect of system xc-, BMAA can induce oxidative stress and increase extracellular glutamate. This action of BMAA provides an attractive mechanism for the multiple neurological defi cits that BMAA has been implicated in inducing.

Key words: Excitotoxicity , BMAA, glutamate , cystine , system xc-

Introduction

Overactivation of glutamate receptors, particularly the NMDA receptor, has been implicated in neu-ronal death occurring during Alzheimer’s disease, Parkinson’s disease, and amyotrophic lateral sclero-sis (ALS) ( 1). When it was discovered that BMAA was a glutamate receptor agonist, and specifi cally an NMDA receptor agonist, this became an obvi-ous potential mechanism of action by which BMAA may induce neurodegeneration. Activation of NMDA receptors may be the main mechanism of action of BMAA in toxicity to motor neurons. However, the possibility of other toxic effects of BMAA also needs to be explored. This paper will review the evidence for BMAA action at each of the main types of glutamate receptors and their possi-ble role in inducing neurodegeneration, but will also explore two additional actions of BMAA. First, the ability of BMAA to act at the cystine/glutamate antiporter (system xc -) to induce oxidative stress and glutamate release will be described. Secondly, the possible interaction of BMAA with other neu-rotoxins will be explored.

BMAA as a glutamate receptor agonist

NMDA receptors

That BMAA can act on NMDA receptors and induce excitotoxicity has been known since 1987 ( 2) and has been an important component of the BMAA hypothesis. Because of the dominant role that NMDA receptor activation plays in excitotoxic neu-ronal death this action has typically been viewed as the main mechanism of BMAA induced toxicity. However, there are a number of questions involving the actions of BMAA at the NMDA receptor.

An interesting aspect of BMAA action is that for it to activate the NMDA receptor/channel, it requires the presence of bicarbonate. There appears to be an interaction between bicarbonate and the β-amino group of BMAA that causes the forma-tion of a molecule capable of activating NMDA receptors ( 3,4).

A question regarding the physiological relevance of BMAA activation of NMDA receptors is raised by the fi nding that signifi cant BMAA concentrations (1–3 μM) “produce acute neuronal swelling and

Am

yotr

oph

Lat

eral

Scl

er D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Ade

laid

e on

11/

13/1

4Fo

r pe

rson

al u

se o

nly.

Mechanisms of BMAA toxicity 57

calcium that is blocked by AMPA/kainate receptor antagonists ( 10).

Metabotropic glutamate receptors (mGluRs)

Evidence indicates that BMAA is an agonist of both group 1 and group 2 mGluRs. Group 1 recep-tors induce phosphatidylinositol hydrolysis with the formation of inositol trisphosphate (IP3) and subsequent calcium release from the endoplasmic reticulum. Activation of these receptors has been shown to induce neuronal death ( 13). BMAA has been shown to have low affi nity, but full potency, at mGluR1 receptors ( 14), a type of group 1 mGluR. The low affi nity raises questions about the phy-siological relevance of action at this receptor; however, we found that activation of mGluR5, another type of group 1 mGluR, was one of the mechanisms by which BMAA induced neuronal death in cortical culture ( 7). The mGluR mediated release of calcium from the endoplasmic reticulum may have an additive effect with the calcium infl ux through NMDA receptors or calcium permeable AMPA receptors.

Activation of group 2 mGluR receptors causes decreased cAMP levels that may be neuroprotective (15). The ability of BMAA to act at group 2 mGluRs was shown by its inhibition of forskolin induced cAMP formation. However, this effect of BMAA was found to have low potency ( 16).

BMAA action on the cystine/glutamate antiporter (system xc -)

Effects on cystine uptake

We have found that BMAA inhibits cystine uptake, causing depletion of cellular glutathione, increased cellular oxidative stress, and free radical mediated neuronal death in cortical cultures ( 17). Most of the cystine uptake in cortical cultures is mediated by system xc -, and it was this component of the uptake that was inhibited by BMAA. The system xc- mediated cystine uptake in cortical cultures, measured by the uptake of 14C-cystine, is mainly into astrocytes. We found that the system xc - dependent uptake of 14C-cystine over a 20-min period was 170,083� 16,443 CPM/mg protein in astrocytes and 34,720 � 8326 CPM/mg protein in neurons (n� 8). Therefore, about 83% of the uptake was into astrocytes.

Effects on glutamate release

BMAA not only blocks cystine uptake via system xc -,but it also appears to drive the antiporter to releaseglutamate. The evidence for this is that BMAA and cystine at the same concentration (3 μM) increase extracellular glutamate to the same level, and both induce mGluR5 mediated neuronal death, probably

substantial late neuronal degeneration” ( 3) and are required to induce neuronal death in cell culture in conditions in which the death is mediated by activa-tion of NMDA receptors ( 2,4–7). However, at much lower BMAA concentrations, “100 μM or less, it potentiated N-methyl-D-aspartate (NMDA)” (7). Rapid electrophysiological studies have estab-lished that the effects of BMAA on NMDA recep-tors are due to direct actions on the receptor, i.e. BMAA induces rapid currents (ms) when directly applied to neurons ( 4,7). The current is blocked by D-amino-5-phosphonovalerate (APV), a competitive antagonist of the glutamate binding site at the NMDA receptor, suggesting that this is the site at which BMAA acts. Also, while signifi cant BMAA concen-trations are required to cause acute toxicity in the absence of other neurotoxic insults such as NMDA (7), low-level NMDA receptor mediated currents can be induced at μM concentrations ( 7). Another important point is that while BMAA induced neu-ronal death is often largely mediated by NMDA receptor activation, the protection provided by NMDA receptor antagonists is generally not com-plete ( 2,4,7), suggesting that other mechanisms of BMAA toxicity may also occur.

AMPA receptors

Blocking of AMPA/kainate receptors was found to not be protective against BMAA toxicity in sagittal mouse brain slices ( 6) or in cortical cultures ( 4,7). However, BMAA was shown to inhibit the binding of both glutamate and AMPA in vivo ( 8). Also, the ability of BMAA to induce seizures was not blocked by an NMDA receptor antagonist, but was blocked by an AMPA/kainate receptor antagonist ( 9). In con-trast to the general toxicity in cortical cultures, selective toxicity of BMAA to specifi c neuronal popu lations has been shown to be mediated by AMPA/kainatereceptors. A small subset of neurons in cortical cultures, distinguished by the presence of NADPH-diaphorase, is highly sensitive to BMAA toxicity, and this toxicity is mediated by AMPA/kainate receptors ( 4). Motor neurons are also highly sensitive to BMAA toxicity, which is also mediated by AMPA/kainate receptors. Importantly, the toxicity of BMAA on motor neuronsoccurs at a concentration of 30 μM ( 10), a concentra-tion much lower than the μM levels required to induceNMDA receptor mediated neuronal death in cortical cultures in the absence of other insults. A possible explanation for these results is that motor neurons, and cortical NADPH-diaphorase neurons, are unusual in that they contain high numbers of calcium perme-able AMPA/kainate receptors ( 11,12).

The possibility exists that AMPA/kainate receptor activation is secondary to BMAA induced release of endogenous glutamate. However, at least in the case of motor neurons, there is compelling evidence that BMAA directly acts on AMPA/kainate receptors, i.e. BMAA induces a rapid increase in intracellular free

Am

yotr

oph

Lat

eral

Scl

er D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Ade

laid

e on

11/

13/1

4Fo

r pe

rson

al u

se o

nly.

58 D. Lobner

through inducing glutamate release ( 17). It is known that glutamate release from system xc - can cause tox-icity. Activated astrocytes can release glutamate via system xc - that can kill cortical neurons ( 18). Also, microglia can release glutamate via xc - that can kill cerebellar granule cells ( 19) and enhance amyloid- induced neuronal death in cortical cultures ( 20).

Effects of BMAA uptake via system xc-

The fi nding that BMAA not only inhibits cystine uptake mediated by system xc -, but also stimulates glutamate release, suggests that it is transported by system xc -. This transport provides a mechanism by which BMAA can accumulate in cells and possibly be incorporated into proteins ( 21). If such uptake and incorporation into proteins occurs, this could potentially play a role in the protein misfolding found in neurodegenerative diseases ( 22).

Model for mechanism of BMAA action

Figure 1 presents a proposed model for the actions of BMAA in cortical cultures. Data indicate that BMAA stimulates glutamate release through action on system xc -. One possible explanation for the mul-tiple effects of BMAA on glutamate receptors is that some of the effects are actually mediated glutamate release from system xc -. This effect would be unlikely to account for the rapid NMDA receptor mediated responses recorded electrophysiologically. Therefore, we propose that BMAA acts directly on the NMDA receptor. However, we propose that the actions of

BMAA on mGluR5 receptors may be, at least par-tially, mediated by glutamate release from system xc -.Since system xc - is primarily on astrocytes, the glu-tamate release and inhibition of cystine uptake would occur primarily on astrocytes. With regard to the BMAA induced glutathione depletion, it is proposed that the decreased cystine uptake into astrocytes leads to decreased astrocytic glutathione, and reduced release of glutathione from astrocytes, in this way limiting the availability of the main precursor for neuronal glutathione synthesis, cysteine. However, we cannot exclude the possibility that the small amount of system xc - mediated cystine uptake by neurons is critical for their production of glutathione.

Synergistic effects of BMAA and other neurotoxins

Initial studies in cortical culture found that μM concentrations of BMAA were required to induce neuronal death (2,4–6). We found similar results when BMAA was tested in isolation, but when combined with other neurotoxins it could enhance neuronal death at much lower concentrations ( 7). We found that BMAA as low as 10 μM could potentiate neuronal injury induced by exposure to amyloid- or 1-methyl-4-phenylpyridinium ion (MPP � ). These insults are of particular interest because exposure to amyloid- is considered to be a model of Alzheimer’s disease ( 23) and MPP � of Parkinson’s disease ( 24). The common point of mechanism of these two tox-ins is that they are known to induce oxidative stress (25,26).

Figure 1. Proposed model for the neurotoxic actions of BMAA. The model is consistent with data obtained using mixed neural and glial cortical cultures.

Am

yotr

oph

Lat

eral

Scl

er D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Ade

laid

e on

11/

13/1

4Fo

r pe

rson

al u

se o

nly.

Mechanisms of BMAA toxicity 59

BMAA in vitro provides at least some insight into long-term exposure to lower concentration in vivo. In the future, studies of the mechanism of BMAA neurotoxicity must be moved in vivo. Previous in vivo studies using BMAA have studied the effects of BMAA in isolation, in healthy, young, animals ( 27). More complex in vivo studies must be performed. These studies may include testing for synergistic effects of BMAA with other neurotoxins and testing whether BMAA hastens the onset of defi cits in animal models of neurodegenerative diseases.

Declaration of interest: The author reports no confl icts of interest. The author alone is responsible for the content and writing of this paper.

References

Lipton SA. Failures and successes of NMDA receptor antag-1. onists: molecular basis for the use of open-channel blockers such as memantine in the treatment of acute and chronic neurologic insults. NeuroRx. 2004;1:101–10.Ross SM, Seelig M, Spencer PS. Specifi c antagonism of 2. excitotoxic action of ‘uncommon’ amino acids assayed in organotypic mouse cortical cultures. Brain Res. 1987;425: 120–7.Weiss JH, Choi DW. Beta-N-methylamino-L-alanine neuro-3. toxicity: requirement for bicarbonate as a cofactor. Science. 1988;241:973–5.Weiss JH, Koh JY, Choi DW. Neurotoxicity of 4. β-N-methylamino-L-alanine (BMAA) and β-N-oxalylamino-L-alanine (BOAA) on cultured cortical neurons. Brain Res. 989;497:64–71.Zeevalk GD, Nicklas WJ. Acute excitotoxicity in chick retina 5. caused by the unusual amino acids BOAA and BMAA: effects of MK-801 and kynurenate. Neurosci Lett. 1989; 102:284–90.Pai KS, Shankar SK, Ravindranath V. Billion-fold difference 6. in the toxic potencies of two excitatory plant amino acids, L-BOAA and L-BMAA: biochemical and morphological studies using mouse brain slices. Neurosci Res. 1993;17: 241–8.Lobner D, Piana PM, Salous AK, Peoples RW. 7. β-N-methylamino-L-alanine enhances neurotoxicity through multiple mechanisms. Neurobiol Dis. 2007;25:360–6. Rakonczay Z, Matsuoka Y, Giacobini E. Effects of L-8. β-N-methylamino-L-alanine (L-BMAA) on the cortical choli-nergic and glutamatergic systems of the rat. J Neurosci Res. 1991;29:121–6.Smith SE, Meldrum BS. Receptor site specifi city for the 9. acute effects of β-N-methylamino-alanine in mice. Eur J Pharmacol. 1990;187:131–4.Rao SD, Banack SA, Cox PA, Weiss JH. BMAA selectively 10. injures motor neurons via AMPA/kainate receptor activa-tion. Exp Neurol. 2006;201:244–52.Weiss JH, Turetsky D, Wilke G, Choi DW. AMPA/kainate 11. receptor-mediated damage to NADPH-diaphorase-containingneurons is Ca 2+ dependent. Neurosci Lett. 1994;167:93–6.Carriedo SG, Yin HZ, Lamberta R, Weiss JH. In vitro kain-12. ate injury to large, SMI-32( � ) spinal neurons is Ca 2�

dependent. Neuroreport. 1995;19:945–8.Strasser U, Lobner D, Behrens MM, Canzoniero LM, 13. Choi DW. Antagonists for group I mGluRs attenuate exci-totoxic neuronal death in cortical cultures. Eur J Neurosci. 1998;10:2848–55.Manzoni OJ, Prezeau L, Bockaert J. 14. β-N-methylamino-L-alanine is a low-affi nity agonist of metabotropic gluta-mate receptors. Neuroreport. 1991;2:609–11.

The implication of these studies is that while low level BMAA may not be the sole cause of neuro-logical diseases, it may be able to combine with underlying pathologies to either induce expression of the disease or quicken its progression. Also, the cell culture studies represent effects of BMAA over only a brief period of time (24 h), in contrast to the long-term exposure that is probably involved in neurodegenerative diseases. It seems likely that even lower concentrations of BMAA over such long periods of exposure could enhance neuronal death. Previous studies in mice have indicated that admin-istration of BMAA does not induce neurological defi cits or neuronal death in vivo ( 27). These results are not inconsistent with the proposed mechanism of BMAA action. Our results suggest that BMAA alone is unlikely to cause neurological defi cits unless very high concentrations are consumed, but that the combination of low levels of BMAA consumption plus another insult may lead to the development of neurological diseases.

Conclusions

The importance of understanding the mechanism of BMAA toxicity is two-fold. First, understanding how BMAA acts will provide insight into whether it may be responsible for inducing neurodegenera-tive diseases. For BMAA to be a major cause of ALS/PD on Guam requires that its actions explain the long latency of the disease following exposure and the multiple defi cits associated with this disease. The action of BMAA at system xc - in particular pro-vides an attractive explanation for the observed pathologies. By inhibiting cystine uptake, BMAA could cause a long-term decrease in glutathione. There is evidence that glutathione depletion plays a role in Alzheimer’s disease, Parkinson’s disease, and ALS (28–30). This kind of low-level, long-term insult could be responsible for delayed neuronal death in many areas of the central nervous system, depending on the sensitivity of the cells. Secondly, understand-ing the mechanism of the toxic effects of BMAA could potentially lead to therapeutic approaches, i.e. if excitotoxicity or oxidative stress is a major factor in the neuronal death, therapeutic agents designed to address those insults may be effective treatments. Greater knowledge about the actions of BMAA will provide for greater specifi city of drug treatment. For example, if depletion of cellular glutathione is the key event, then strategies to specifi cally enhance glutathione levels should be targeted.

A concern regarding the studies involving the mechanism of BMAA neurotoxicity is that they have been performed in cell culture. The main limitation of this system is that effects can only be studied for brief periods of time, while the toxic actions of BMAA in vivo are believed to occur following pro-longed exposure. Therefore, the assumption must be made that brief exposure to high concentration

Am

yotr

oph

Lat

eral

Scl

er D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Ade

laid

e on

11/

13/1

4Fo

r pe

rson

al u

se o

nly.

60 D. Lobner

Gotz J, Schild A, Hoerndli F, Pennanen L. Amyloid-induced 23. neurofi brillary tangle formation in Alzheimer’s disease: insight from transgenic mouse and tissue-culture models. Int J Dev Neurosci. 2004;11:453–65.

24. Gerlach M, Riederer P. Animal models of Parkinson’s disease: an empirical comparison with the phenomeno-logy of the disease in man. J Neural Transm. 1996;103: 987–1041. Johannessen JN, Adams JD, Schuller HM, Bacon JP, 25. Markey SP. 1-methyl-4-phenylpyridine (MPP � ) induces oxidative stress in the rodent. Life Sci. 1986;38:743–9.Kwon YS, Koh JY, Song DK, Kim HC, Kwon MS, Choi YS, 26. Wie MB. Danthron inhibits the neurotoxicity induced by various compounds causing oxidative damages including beta-amyloid (25-35) in primary cortical cultures. BiolPharm Bull. 2004;27:723–6. Karamyan VT, Speth RC. Animal models of BMAA neuro-27. toxicity: a critical review. Life Sci. 2008;82:233–46.Bains JS, Shaw CA. Neurodegenerative disorders in humans: 28. the role of glutathione in oxidative stress-mediated neuronal death. Brain Res Brain Res Rev. 1997;25:335–58.Liu H, Wang H, Shenvi S, Hagen TM, Liu RM. Glutathione 29. metabolism during aging and in Alzheimer’s disease. Ann N Y Acad Sci. 2004;1019:346–9.Zeevalk GD, Razmpour R, Bernard LP. Glutathione and 30. Parkinson’s disease: is this the elephant in the room? BiomedPharmacother. 2000;62:236–49.

Buisson A, Choi DW. The inhibitory mGluR agonist, 15. S-4-carboxy-3-hydroxy-phenylglycine selectively attenuates NMDA neurotoxicity and oxygen-glucose deprivation-induced neuronal death. Neuropharmacology. 1995;34:1081–7.Genazzani AA, Casabona G, L’Episcopo MR, Condorelli DF, 16. Dell’ Albani P, Shinozaki H, Nicoletti F. Characterization of metabotropic glutamate receptors negatively linked to adenylyl cyclase in brain slices. Brain Res. 1993;622:132–8.Zapata J, Hjelmhaug JA, Lobner D. Role of the cystine-glutamate 17. antiporter in β -N-methylamino-L-alanine (BMAA) toxicity. Soc Neurosci Abs. 2007;604:9. Fogal B, Li J, Lobner D, McCullough LD, Hewett SJ. System18. X(c)- activity and astrocytes are necessary for interleukin-1beta-mediated hypoxic neuronal injury. J Neurosci. 2007;27:10094–105.Piani D, Fontana A. Involvement of the cystine transport 19. system xc- in the macrophage-induced glutamate-dependent cytotoxicity to neurons. J Immunol. 1994;152:3578–85.Qin S, Colin C, Hinners I, Gervais A, Cheret C, Mallat M. 20. System xc- and apolipoprotein E expressed by microglia have opposite effects on the neurotoxicity of amyloid-beta peptide 1-40. J Neurosci. 2006;26:3345–56.Murch SJ, Cox PA, Banack SA, Steele JC, Sacks OW. Occur-21. rence of β-methylamino-L-alanine (BMAA) in ALS/PDC patients from Guam. Acta Neurol Scand. 2004;110:267–9.Uversky VN. Alpha-synuclein misfolding and neurodegen-22. erative diseases. Curr Protein Pept Sci. 2008;9:507–40.

Am

yotr

oph

Lat

eral

Scl

er D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Ade

laid

e on

11/

13/1

4Fo

r pe

rson

al u

se o

nly.