EXPERIMENTALNEUROLOGY

108,141-150(1990)

Potential Environmental Neurotoxins Related to 1-Methyl-4phenylpyridinium: Selective Toxicity of 1-Methyl-4-(4’-acetamidophenyl)pyridinium and 1-Methyl-4-cyclohexylpyridinium for Dopaminergic Neurons in Culture PATRICK P. MICHEL, Andrus

Gerontology Center, University Miami, Florida 33101;

BHUVANESHARI

K. DANDAPANI,

Academic

Press,

AND

FRANZ

of Neurology, Minnesota

HEFTI University 55455

of Miami,

nylhydrazine; E&, effective concentration 50, GABA, y-aminobutyric acid, IEG+, 2-(3-indolyl)ethylguanidinium; Z.i, Zs2, indices of selective toxicity 1 and 2, respectively; MACBP+, 1-methyl-4(4’-acetamidobenzyl)pyridinium; MACPP+, 1-methyl-4-(4’-acetamidophenyl)pyridinium; MACPTP, 1-methyl-4-(4’-acetamidophenyl)-1,2,3,6-tetrahydropyridine; MACSP+, 1-methyl-4-(4’-acetamidostyryl)pyridinium; MBSP+, 1-methyl-4-(3’bromostyryl)pyridinium; MCP+, 1-methyl-4-cyclohexylpyridinium; MCTP, 1-methyl-4cyclohexyl1,2,3,6-tetrahydropyridine; MOMBP+, l-methyl-4-orthomethylbenzylpyridinium; MPMBP+, I-methyl-4-paramethylbenzylpyridinium; MPP+, 1-methyl-4-phenylpyridinium; MPTP, 1-methyl-4-phenyl1,2,3&tetrahydropyridine; NBOMP+, l-(3’-nitrobenzyloxymethyl) pyridinium; PAMPP+, p-amino-MPP+; PAMPTP, p-amino-MPTP; PBS, phosphate-buffered saline; PPTEA+, 3-phenyl-2-propenyltriethylammonium; TH, tyrosine hydroxylase; TH+, tyrosine hydroxylase positive.

Inc.

INTRODUCTION 1-Methyl-4-phenyl-1,2,3,6-tetmhydropyridine (MPTP)’ is an extremely potent neurotoxin, inducing lesions of ’ Abbreviations used: AZB, azobenzene; BMAA, L-alanine; CEB+, 2-chloro-3-ethylbenzoxazolium; DMHET+, 3,4-dimethyl-5-(2-hydroxyethyl)thiazolium;

EFANGE,* Department Minneapolis,

the nigrostriatal system and long-lasting parkinsonian symptoms in man (8, 22). Experimental studies in MPTP-treated monkeys revealed pronounced deficits of dopamine within the striatum, selective destruction of dopaminergic neurons of the substantia nigra, and parkinsonian impairment of motor functions (4,19,21,23). Hence it was speculated that Parkinson’s disease could be induced by ubiquitous environmental neurotoxins related to MPTP (5). The mechanism of action of MPTP is largely understood. First, MPTP reaches the brain by passing the blood-brain barrier via unspecific diffusion based on its high lipophilicity (33). In the brain, MPTP is transformed into a polar metabolite, 1-methyl-4-phenylpyridinium (MPP+), by initial oxidation catalyzed by monoamine oxidase B (MAOB) (6,15, 24). MPP+ is then released from these cells into the extracellular fluid and subsequently transported into dopaminergic neurons via the high affinity uptake carrier for dopamine (l&40). It is likely that MPP+ causes death of dopaminergic neurons by inhibiting the mitochondrial respiration system at the site I level (32,34,41,46). This high mitochondrial toxicity appears to be closely linked to the strong cationic delocalized charge present in MPP+ (36). Over the past few years we have developed a culture system of dissociated mesencephalic cells containing do-

Mesencephalic cells in culture were exposed to various compounds which we hypothesized to be selective toxins for dopaminergic neurons. The culture system was previously shown suitable for assessing selective dopaminergic neurotoxicity, since l-methyl-l-phenylpyridinium (MPP+), the active metabolite of l-methyl4-phenyl-1,2,3,6-tetrahydropyridinium, destroyed dopaminergic neurons without affecting other cells. Some compounds tested were selected to fulfill two criteria believed to underly the selective dopaminergic neurotoxicity of MPPS, i.e., to be a potential substrate for the uptake carrier for dopamine and to possess a strong delocalized positive charge to inhibit the mitochondrial respiratory system. Other compounds were chosen on the basis of clinical or anecdotal evidence linking them to Parkinson’s disease. Among the tested compounds two pyridinium analogs, 1-methyl-4-(4’acetamidophenyl)pyridinium (MACPP+) and l-methyl-4-cyclohexylpyridinium (MCP+) were found to be selectively toxic toward dopaminergic neurons. Incubation of cultures with both MACPP+ and MCP+ produced a dramatic reduction in the number of tyrosine hydroxylase-positive cells and the uptake of [3H]dopamine without reducing the number of cells visualized by phase-contrast microscopy or the uptake of [3H]aminobutyric acid. Besides MACPP+ and MCP+ none of the tested compounds exhibited any selective dopaminergic neurotoxicity. Together with earlier findings, these data suggest that the structural requirements are rather strict for a chemical to be a selective dopaminergic neurotoxin and make it unlikely that there is a wide spectrum of environmental dopaminergic toxins. 0 1990

SIMON M. N.

of Southern California, Los Angeles, California 90089; and *Department of Radiology, University of Minnesota,

fl-N-methylaminoDA, dopamine; DPH, diphe141

0014-4s86/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

142

MICHEL

paminergic and nondopaminergic neurons and shown that MPP+, the active metabolite of MPTP, selectively destroys dopaminergic neurons without affecting other cells (30,38,39). These cultures appear then suitable for searching for dopaminergic toxins that could occur in the environment. Several rationales were used to select compounds for testing in our culture system. First, on the basis of the described mechanism of action of MPP+, we hypothesized that other compounds fulfilling the structural requirements of the crucial metabolic steps leading to its toxicity (i.e., to be a substrate for the dopamine uptake system and exhibit mitochondrial toxicity) could induce selective toxic effects on dopaminergic cells. Second, we tested a group of alkaloids which are pyridiniums or close derivatives of them. These compounds are ubiquitous plant molecules and therefore good candidates for environmental neurotoxins. Third, @-N-methylamino-L-alanine (BMAA), a plant excitotoxin, was tested because of its possible link to the amyotrophic lateral sclerosis-Parkinson-dementia complex of Guam (44,45). Fourth, azobenzene (AZB) and diphenylhydrazine (DPH), which are by-products of aniline synthesis, were tested for dopaminergic toxicity because of anecdotal evidence that chemical workers engaged in large scale synthesis of aniline may have a higher probability of becoming victims of Parkinson’s disease. The present findings extend our earlier studies on cultures which showed that close structural analogs of MPP+ are selective dopaminergic neurotoxins, whereas selected guanidines, benzimidazoles, isoquinolines, and excitotoxins are ineffective or nonselectively toxic (31,39). MATERIAL

AND

METHODS

Chemicals MPMBP+, MOMBP+, MACBP+, MACSP+, MBSP+, PPTEA+, and IEG+ were synthesized. MPP+ and BMAA were purchased from RBI (Natick, MA). MCP+ was provided by Dr. Heikkila (Rutgers University, Piscataway, NJ). MACPP+ and PAMPP+ were provided by Dr. Markey (National Institute of Health, Bethesda, MD). NBOMP+, CEB+, DMHET+, AZB, and DPH were purchased from Aldrich (Milwaukee, WI). Berberine, anabasine, cotinine, and cytisine were purchased from Sigma (St. Louis, MO). [3H]DA (37 Ci/mmol) and [3H]GABA (37 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Cell Cultures Dopaminergic cultures were prepared from the ventral mesencephalon of rat embryos at E15. The area dissected contained the dopaminergic nuclei A8 and A9 (substantia nigra) and the A10 nucleus (ventral tegmental area) (11,27,43). The collected tissue did not include the A6 nucleus (locus coeruleus). After dissection the tis-

ET

AL.

sue was gently triturated with an Eppendorf blue pipet and the resulting pool of dissociated cells was plated in 35-mm dishes, on Lab-Tek double-chamber slides, or on 16-mm multiwell plates, precoated with polyethyleneimine (1 mg/ml) and containing 2, 1.5, and 0.5 ml of N5 growth medium, respectively, which was supplemented with 5% horse serum and 0.5% fetal calf serum. The density after 1 week in vivo ranged from 0.8 to 1.2 X lo5 cells/cm2. Of these cells 0.5 to 1% were TH+. For further experimental details see Refs. (31) and (39). MPP+

Treatment

of Cultures

Mesencephalic cultures were usually grown for 6-7 days and then exposed for 24 h to the various toxins. Toxins were dissolved in sterile distilled water and then added to the cultures to the final concentration. After exposure, cultures were washed three times with fresh medium and processed for staining or uptake studies. For recovery studies, cultures were left for 3 additional days in MPP+-free medium and then taken for uptake measurements. Tyrosine

Hydroxylase

(TH) Immunocytochemistry

Cultures were washed with phosphate-buffered saline (PBS), fixed with 4% freshly made formaldehyde for 20 min and washed again with PBS. They were then incubated for 24-36 h with a rabbit anti-TH serum (a gift from Dr. J. Reinhard, Burroughs Wellcome Co., Research Triangle Park, NC) diluted 1:400 in PBS containing 0.2% Triton X-100. After washing the cells were incubated for 2 h with a biotinylated anti-rabbit IgG fraction (Vector, Burlingame, CA) diluted 1:lOO in PBS containing 0.2% Triton X-100. They were washed again with PBS and incubated for 2 h with a preformed avidinbiotinylated horseradish peroxidase complex diluted 1: 50 in PBS. The peroxidase was revealed by incubation with a solution of 1 mg/ml diaminobenzidine in PBS containing 0.015% hydrogen peroxide. The number of surviving cells in a culture dish was counted by microscopic analysis. TH+ neurons were counted in 2 to 10% of the total area of the culture dish. Unstained cells were counted by phase-contrast microscopy in 0.1 to 1% of every dish. Dopamine and GABA Uptake as Parameters Selective and Nonselective Neurotoxicity

Reflecting

After termination of MPP+ treatment cultures were washed, kept in MPP+-free medium for 1 h, and then processed for uptake studies. The method used to measure [3H]DA uptake was a modification of that described by Prochiantz et al. (37). Briefly, cells were washed three times with incubation solution (5 mM glucose, 0.1 mA4 ascorbic acid in PBS, pH 7.4). Then cultures were preincubated for 5 min at 37°C with 1 ml of incubation solu-

ENVIRONMENTAL

DOPAMINERGIC

tion containing 1 mM pargyline (MAO inhibitor). The cultures were then incubated with 50 nM [3H]DA (37 Ci/mmol) for 15 min at 37°C. Blanks were obtained by incubating the cultures at 0°C. The uptake was stopped by removing the incubation mixture followed by three rapid washes with cold PBS. Cells were scraped twice with 1 ml PBS containing 1% Triton X-100 and 6% perchloric acid. The radioactivity was measured by liquid scintillation spectrometry after addition of 10 ml of Quantafluor (Mallinckrodt) to each vial. Measurement of GABA uptake was performed using essentially the same technique (31). The concentration of GABA was 50 nA4 (37 Ci/mmol) and the incubation time was 4 min. Aminooxyacetic acid (10 PM) was used as inhibitor of GABA catabolism. Blanks were made at 0°C. Quantification

of Selective and Nonselective

Toxicity

Cultures were analyzed for selective and nonselective toxicity toward dopaminergic neurons by either immunochemical or biochemical methods. For immunochemical studies 9 or 10 different concentrations (0.001-2000 @I) were tested for each compound and three culture dishes were used for each concentration, i.e., a total of 27-30 dishes was taken for each compound. Each experiment was triplicated. These cultures were sister cultures and handled in parallel. Selective toxicity for dopaminergic neurons was quantitated by calculating the effective concentration which induced the loss of 50% of the population of tyrosine hydroxylase-positive (TH+) neurons (EC&&. Nonselective toxicity was quantitated by calculating the concentration which induced the degeneration of 50% of all cultured cells as visualized by phase-contrast microscopy (EC,,,). The ratio EC&,,/ ECSOTH was defined as IsI, i.e., index 1 of selective toxicity. Selective toxicity of some compounds was defined on the basis of uptake studies. Six to eleven concentrations (0.001-2000 PM) were tested for each compound and three culture dishes were used for each concentration. Each experiment was triplicated. The ratio ECSOGABA/ ECBODA was defined as IS2, i.e., index 2 of selective toxicity. The calculations of the EC&‘s were performed with the RSl fitting curve program (BBN Software Products Co., Cambridge, MA) according to the mathematical model Y = A/(1 + (XJB)Iz), where A is the theoretical limit when the concentration of the test compound approaches 0, B is the EC&, and n > 0. Inhibition of Dopamine Experiments

Uptake in Competition

Sister cultures were preincubated as described before and coincubated for 15 min with 50 nM [3H]DA and different concentrations of cold inhibitors for 15 min. For each compound seven or eight concentrations (0.001-1000 PM) were assessed and two or three wells

143

NEUROTOXINS

TABLE Selective Neurotoxicity Compounds

ECLOTH(PM)

MPP+ MACPP+ MCP+ MPMBP+ PPTEA+

0.43 t0.13** 5.5 * o.l3** 15+3* 205f84 110234

1

for Dopaminergic EC,,

(PM)

62k4 455+29 78k7 170+79 174+39

Neurons I 81

144 83 5.2 0.8 1.6 I 82

MPP+ MACPP+ MCPf MPMBP+ PPTEA

0.063 * 0.007** 0.35 2 0.03** 0.74 + 0.15* 66.7 f 7.1 205+18

93 f 6.1 807f117 79flO 184+65 241235

1470 2300 106 2.8 1.2

Note. Comparison of the selective toxic effects of MPP+, MACPP+, MCP+, MPMBP+, and PPTEA+ in mesencephalic dissociated cultures. Cultures were exposed to the various toxins for 24 h and then taken for analysis. EC&n, EC,,, ECsOoA, and ECSOoAB~ correspond to the concentration of toxin inducing a 50% decrease in, respectively, the number of TH+ cells visualized by immunohistochemistry, the number of cells visualized by phase-contrast microscopy, [3H]DA uptake, and 13H]GABA uptake. Is1 and Is2 are defined as the ratios EC,,/EC,,u and ECSOoAsA/ECSOnA, respectively, and represent two indices of selective toxicity for dopaminergic cells. MPMBP+ is included as an example of a pyridinium inducing nonselective toxic effects only. PPTEA+ is included as an example of a nonpyridinium molecule inducing nonselective toxic effects. * ECSOTH or E&o,, significantly different from corresponding or ECSEABA respectively, P < 0.01. EC,, **P < 0.001.

were used per concentration. Ki values were calculated with the RSl fitting curve program according to the model Y = A/(1 + X/B), where A is the percentage of uptake of [3H]DA in the absence of inhibitor and B is the Ki. Each experiment was triplicated. RESULTS Selective Toxicity for Dopaminergic Neurons Pyridiniums. As reported earlier (30,31,38,39), immuno-chemical methods showed that MPP-t selectively destroys dopaminergic neurons in our cultures (EC50TH = 0.43 PM, IsI = 144) (Table 1). Selective toxic effects of MPP+ for dopaminergic neurons were seen at even lower concentrations when biochemical uptake parameters were used to assessthe function of dopaminergic cells (EC5,,nA = 0.063 PM, IS2 = 1470; Table 1). Besides MPP+, dopaminergic neurons were selectively destroyed by two other pyridinium derivatives, i.e., lmethyl-4-(4’-acetamidophenyl)-pyridinium (MACPP+) and 1-methyl-4-cyclohexyl-pyridinium (MCP+) (Table 1; Figs. 1 and 2). MACPP+ was intrinsically less potent than MPP+ for dopaminergic neurons (ECBOrn = 5.5 PM, ECBODA= 0.35 PM). However its very low toxicity

144

MICHEL

ET

AL.

ENVIRONMENTAL

DOPAMINERGIC

145

NEUROTOXINS

TABLE2 Compounds Lacking Selective Dopaminergic &OTH (PM)

0

0.01

0.1

1

10

OO 100 1000

100 80 60 40 20 0

m

0.01

0.1

1

10

100

1000

PM

FIG. 2. Dose-response curves of the toxic actions of MACPP+ (A, B) and MCP+ (C, D) in mesencephalic cultures. Number of cells and uptake are expressed as percentage of control values in the absence of the toxin. Three culture dishes were used for each concentration and the results are given as means + SEM. Open circles represent the percentage of TH+ cells in A and C and the percentage of DA uptake in B and D (selective toxicity). Filled circles represent the percentage of cells visualized by phase-contrast microscopy in A and C and the percentage of GABA uptake in B and D (nonselective toxicity).

for nondopaminergic cells (EC,,, = 455 pCLM;EC&oAsA = 807 PM) compared to that of MPP+ (EC,,, = 62 pCLM; = 93 PM) made this toxin as selective as EC 50GABA MPP+ for dopaminergic neurons (1,i = 83; IS2 = 2300). The intrinsic toxicity of MCP+ (ECSOru = 15 PM; EC 50nA = 0.74 &f) and its selectivity (1,i = 5.2; I,2 = 106) were lower than those of MACPP+ and MPP+. None of the other pyridiniums was found to be selectively toxic for dopaminergic neurons as judged by staining techniques. Methylparamethylbenzylpyridinium and methylorthomethylbenzylpyridinium (MPMBP+ and MOMBP+)showed some toxicity for dopaminergic and nondopaminergic cells (EC50’~ < 300 puM) but no dopaminergic selectivity (Table 3). Uptake studies confirmed the lack of selectivity of MPBP+ (Table 1). The other pyridiniums tested (MACBPS, MACSP+, MBSP+, and NBOMP+) failed to show toxic effects toward all cultured cells (Table 2). Alkaloids. In this class of compounds only berberine showed some toxic effects (ECSOTH = 48 PM). However this compound was not selectively toxic for dopaminergic neurons (I,, = 1.6). The other alkaloids tested exhibited little toxicity for all cultured cells (EC50’~ > 3000 PM). FIG. 1. MACPP+ visual field 3, 10, 100, toxin. The

Pyridines NOMBP+ MPMBP+ MACBP+ MACSP+ MBSP+ NBOMP+ Alkaloids Berberine Anabasine Cotidine Cytisine Other molecules PPTEA+ IEG+ CEB+ BMAA AZB DPH

Neurotoxicity

EC,, (t.M

I Sl

151 f 38 205 f 84 23000 >3000 >3000 >3000

217 f 10 170 * 79 >3000 >3000 >3000 >3000

1.4 0.8

48+ 15 >3000 >3000 >3000

78 f 19 >3000 >3000 >3000

1.6

110 t 34 91+ 14 >3000 >3000 >3000 >3000

174 f 39 129 f 12 >3000 >3000 >3000 >3000

1.6 1.9

Note. Cultures were exposed taken for immunocytochemistry.

to the test compounds

for 24 h and then

Compounds related to dopamine or serotonin. PPTEA+ which has some structural similarity to dopamine and possesses a cationic delocalized charge was found toxic (EC&ru = 110 PM, ECsooA = 205 PM) but lacked selective toxicity for dopaminergic neurons (1,i = 1.6; IS2 = 1.2; Table 1). IEG+, CEB+, and MHET+ are related to serotonin and show a strong delocalized positive charge. IEGS was found relatively potent (ECsorH = 91 PM) but none of these compounds was selectively toxic for dopaminergic neurons (Table 2). BMAA. BMAA a plant amino acid which has been linked to the amyotrophic lateral sclerosis-Parkinsondementia complex of Guam (44,45) was found not selectively toxic for cultured dopaminergic neurons (Table 2). Biphenyl derivatives. AZB and DPH are by-products of large-scale aniline synthesis. Even though they do not possess a cationic delocalized positive charge they present some analogy with the chemical structure of MPP+. These two compounds were found poorly toxic for all cultured cells (Table 2). Recovery

Studies

Ten micromolar MACPP+ or 10 PM MCP+ for 24 h induced dramatic deficits in [3H]dopamine uptake when

Selective dopaminergic toxicity of MACPP+ for mesencephalic cultures. Cultures were exposed to different concentrations of for 24 h and then taken for immunochemistry. Left panels show TH immunochemistry staining; right panels show the identical seen by phase-contrast microscopy. a/f correspond to a control culture. b/g, c/h, d/i, and e/j correspond to cultures incubated with and 1000 pM of MACPP+, respectively. Note the progressive disappearance of TH+ cells with increasing concentrations of the number of cells seen by phase-contrast microscopy is not affected until concentrations higher than 300 NM. (magnification, 47X).

146

MICHEL

TABLE Lack

of Recovery Treatment

Days after treatment 0 0 0 3 3 3

TABLE Neurons or MCP+

100.0 3.4 7.9 100.0 3.5 8.1

f f + * f +

Inhibition

after

DA uptake (% of control)

Control MACPP+ MCP+ Control MACPP+ MCP+

GABA uptake (% of control)

1.2 0.4* 2.9* 1.3 0.9* 1.7*

100.0 104.9 92.2 100.0 97.9 96.8

f 3.6 + 10.4 f 3.2 f 3.6 + 3.8 f 3.4

Note. Cultures were treated for 24 h with MACPP+ or MCPP+ (10 PM) and then processed immediately for uptake studies or kept in toxin-free medium for 3 additional days. Control values for DA uptake were 39.6 and 37.3 fmol/min/well at Day 0 and Day 3, respectively. Corresponding values for GABA uptake were 1018 and 1117 fmol/ min/well. * Significantly different from corresponding control P < 0.001, t test.

the cultures ment. Deficits

were processed immediately of similar magnitude were

cultures

were

toxin-free found to

medium be affected

grown

for (Table under

an

after found

when

3-day

period

uptake conditions.

was

additional

3). GABA the

same

treatthe in not

Inhibition of Dopamine Uptake in Competition Experiments Similar to DA and MPP+, MACPP+ and MCP+ were found to inhibit competively the uptake of [3H]DA (Fig. 3). Ki values for these compounds and are shown in Table 4.

DISCUSSION

Our culture system of dissociated mesencephalic cells which contains 0.5-l% dopaminergic cells is particu-

25

SO

? P 3

60

b

40

AL.

3

of Dopaminergic with MACPP+

Compound

ET

0 0 A A 0

INHIBITOR

DA MPP+ PAMPP+ MACPP+ tvicp+

pLM

FIG. 3. Potencies of various selective dopaminergic neurotoxins to inhibit the uptake of [3H]DA. For each compound sister cultures were processed for [3H)DA uptake and coincubated for 15 min with 50 nM [3H]DA and various concentrations of unlabeled DA or neurotoxins. Three culture dishes were taken for each concentration. The uptake of [3H]DA was expressed as the percentage of the control value in the absence of toxin.

of Dopamine Neurotoxins

4

Uptake by Selective Dopaminergic in Competition Experiments

Inhibitor

Ki + SEM

Dopamine MPP+ MACPP+ MCP+

0.28 2.1 22.2 27.5

(rM)

f 0.07 fO.l + 2.1 + 5.2

Note. Sister cultures were coincuhated for 15 min with 50 nM [3H]DA and various concentrations of unlabeled inhibitors. For each compound seven or eight concentrations (0.001-1000 PM) were assessed and two or three wells were used per concentration. Ki values are the means of three independent experiments.

larly suitable for searching for selective dopaminergic toxins, since it allows the investigator to assess,in the same culture dish, dopaminergic and nondopaminergic toxicity. We have previously shown that MPP+, the active metabolite of MPTP produced in uiuo, induces selective toxic effects for dopaminergic neurons in these cultures, whereas MPTP itself is not toxic, probably because of insufficient conversion to MPP+ (31, 38, 39). Accordingly, our culture system was used to search for MPP+-like neurotoxins (Fig. 4). Among the pyridinium derivatives tested MACPP+ and MCP+ were found selectively toxic for dopaminergic neurons. It. is likely that MACPP+ and MCP+ destroy dopaminergic cells by acting similarly to MPP+. Both compounds inhibit the uptake of [3H]dopamine in competition experiments, suggesting that they are transported into the dopaminergic cells. MCP+ has been shown to be equipotent to MPP+ in inhibiting the respiration of isolated mitochondria at site I (42), whereas there is no corresponding experimental data concerning MACPP+. However, the presence of a strong delocalized positive charge is likely to make this compound a strong mitochondrial toxin (36). Due to proteolytic activity within the dopaminergic cells the amide bond of MACPP+ might undergo hydrolysis to give two breakdown compounds, i.e., acetic acid and PAMMP+. Consequently, not only MACPP+, but also PAMMP+ might induce mitochondrial inhibition. Acetic acid, by disrupting cell homeostasis, might contribute to the death of intoxicated dopaminergic cells. For all toxins such as MPP+, MACPP+, and MCP+, the parameter for dopaminergic toxicity based on dopamine uptake (EC&& is 7-20 times lower than the parameter based on TH immunoreactivity (EC&&. This finding is compatible with the view that such toxins first impair uptake processes of dopamine at the nerve endings and then induce retrograde degeneration of the cell body (2). The two selectively toxic pyridiniums, MACPP+ and MCP+, were found to produce an irreversible decline in dopaminergic parameters. Similar irreversible toxicity

ENVIRONMENTAL

MPP+

MACPP+

ALKALOIDS

MCP+

DOPAMINERGIC

MOMBP

+

MPMBP

147

NEUROTOXINS

+

MACBP+

MACSP+

MBSP+

NBOMP

+

01

cH%sA

81

&f&

H

Anabasine OTHER

GH

MOLECULES

0 +

CH3/jj

HOOC

1

NH2

BMAA

AZB

FIG.

4.

Chemical

Et;

CH,

CH2-

R :

CH3CONH-

DPH

structures

of the compounds

was shown earlier for MPP+ itself (31). Irreversible degeneration and impairment of the function of the dopaminergic system is also seen in parkinsonian patients (7). Our findings, obtained in culture with MPP+ and its analogs are at variance with those obtained in some in viva studies where partial behavioral and biochemical recovery processes were observed (10,47). Other studies, however, suggest that repeated injection of small doses of MPTP induce complete and permanent depletion of DA within the striatum (17). l-Methyl-4-cyclohexyl-1,2,3,4-tetrahydropyridinium (MCTP), the prodrug of MCP+, has been found to be a potent dopaminergic neurotoxin in viuo (48, 49). However this correlation between toxicity of MPTP analogs in uiuo and toxicity of their corresponding pyridinium analogs in vitro does not hold true for MACPP+. Indeed, its corresponding reduced analog, 1-methyl-4-(4’-acetmido)-1,2,3,6-tetrahydropyridine (MACPT) was not found to be a dopaminergic toxin in vivo (Markey S. personal communication). Since MPTP itself crosses the blood-brain barrier (33), it is possible that the acetamido substituent, which is highly hydrophilic (14), reduces the initial lipophilic nature of MPTP so that the resulting compound, MACPTP, does not reach the brain compartment. A similar speculation can be made if we consider MACPTP and MPTP as methylpyridiniums

tested

in this study.

substituted in position 4 by phenyl or acetanilide moities (14). Alternatively, the lack of toxicity of MACPTP could reflect the inability of MAOB to transform this compound into its active pyridinium species within the brain. Strict structural requirements for dopaminergic neurotoxicity are suggested by the fact that the analog structurally most close to MACPP+, i.e., MACBP+, lacked selective toxicity. Similarly, no selective toxicity was found for MOMBP+, a close analog of 2’-methylMPP+, another toxin previously found to be selectively toxic for dopaminergic neurons (30). Since MPP+, but not its corresponding benzyl derivative, i.e., 1-methyl-4benzylpyridinium (30) is selectively toxic, it seems generally true that the insertion of a methyl group between the methylpyridinium and its aromatic substituent in position 4 leads to complete disappearance of selectivity. Alkaloids, which are ubiquitous compounds and therefore good candidates for environmental neurotoxins, are very often structurally related to pyridines and usually possess a quaternary ammonium group. Anabasine and cotinine both show structural similarity to MPP+ and are found in cigarette smoke (25). However, their positive charge is not delocalized. Similar to other tested unsubstituted pyridines (phenylpyridine, bipyridinium, nicotine) which lack strong positive delocalized charge,

148

MICHEL

anabasine and cotinine did not show selective dopaminergic toxicity (1,29,30,35). Berberine is a very complex molecule with a strong delocalized positive charge. Its failure to induce selective dopaminergic toxicity probably reflects the lack of uptake by the dopaminergic carrier. PPTEA+ which is structurally related to dopamine and possessesa delocalized positive charge combines the two theoretical features for an ideal selective dopaminergic neurotoxin, i.e., potential ability to be transported by the uptake carrier for dopamine and to inhibit the respiratory chain at site I. However, PPTEA+ failed to induce selective dopaminergic selectivity. Given the lack of precise knowledge about the structural requirements of the dopamine carrier system, it remains possible that this compound is not taken up by dopaminergic neurons. Since serotonin and tryptamine have been shown to be taken up by central catecholaminergic neurons (26), we hypothesized that compounds with a structural analogy to serotonin and a strong delocalized positive charge could show selective toxic effects for dopaminergic neurons. The analogy of IEG+ to serotonin is given by the indole moiety, and the guanidinium substituent brings the delocalized positive charge. However, similar to CEB+ and DMHET+, which show some analogy to serotonin and have a delocalized positive charge on their aromatic ring, IEG+ was not selectively toxic. BMAA, an excitotoxin found in a palm plant, has been linked to the amyotrophic lateral sclerosis-Parkinsondementia complex of Guam (44, 45). Recent reports, however, do not support this hypothesis (9). In our culture system BMAA lacked selective toxicity for dopaminergic neurons and was not toxic toward other cultured cells. Its close analog, P-oxalylaminoalanine, which induces human lathyrism, has not been found selectively toxic in the same system (30). Since these compounds do not present any chemical analogy to MPP+ or dopamine they are probably not taken up by dopaminergic neurons. If they are the cause of neurological disorders, their mechanism of action is probably very different from that of MPP+. The present study shows that two analogs of MPP+, MACPP+ and MCP+, display selective and irreversible toxicity for dopaminergic neurons in culture. Earlier studies demonstrated selective dopaminergic toxicity for l-ethyl-4-phenylpyridinium, 2’-methyl-MPP+, and pamino-MPP+ (PAMPP+) (30, 39). A large number of other pyridiniums and “MPP+-like compounds” tested failed to show such selective toxicity (this study, 30). According to these results and those of studies by other investigators (3,12,13,16,20,28,48,49,50,51) it appears that only restricted modifications of the original structure of MPP+ retain the selective neurotoxicity for dopaminergic neurons. Taken together these studies do not support the hypothesis of an environmentally in-

ET AL.

duced Parkinson’s disease, since this hypothesis and the fact that the geographical distribution of Parkinson’s disease is rather homogeneous predict that a wide spectrum of dopaminergic neurotoxins is to be found. However, the discovery of still untested neurotoxins, with other chemical features and different mechanisms of toxicity, could invalidate this tentative conclusion. ACKNOWLEDGMENT This work was supported by a grant from the National Parkinson Foundation, Miami, Florida. REFERENCES 1. BRADBURY, A. J., B. COSTALL, A. M. DOMENEY, B. TESTA, P. G. JENNER, C. D. MARDSEN, AND R. J. TAYLOR. 1985. The toxic actions of MPTP and its metabolite MPP+ are not mimicked by analogues of MPTP lacking an N-methyl moiety. Neurosci. Lett. 61: 121-126. 2. BRADBURY, A. J., B. COSTALL, P. G. JENNER, M. K. KELLY, C. D. MARDSEN, AND R. J. NAYLOR. 1986. MPP+ can disrupt the nigrostriatal dopamine system by acting in the terminal area. Neuropharmacology 25: 939-941. 3. BOOTH, R. G., N. CASTAGNOLI, JR., AND H. ROLLEMA. 1989. Intracerebral microdialysis neurotoxicity studies of quinolline and isoquinoline derivatives related to MPTP/MPP+. Neurosci. I&t. 100: 306-312. 4. BURNS, R. S., C. C. CHIUEH, S. P. MARKEY, M. H. EBERT, D. M. JACOBOWITZ, AND I. J. KOPIN. 1983. A primate model of parkinsonism: Selective destruction of dopaminergic neurons in the pars compacta of the substantia nigra by n-methyl-4-phenyl1,2,3,6-tetrahydropyridine. Proc. Natl. Acad. Sci. USA 80: 45464550. 5. CALNE, D. B., AND J. W. LANGSTON. 1983. Aetiology of parkinson’s disease. Lancet 2: 1457-1459. 6. CHIBA, K., A. J. TREVOR, AND N. CASTAGNOLI, JR. 1984. Metabolism of the neurotoxic tertiary amine, MPTP, by brain monoamine oxidase. Biochem. Biophys. Res. Commun. 120: 574-578. 7. COTI?, L. 1981. Basal ganglia: The extrapyramidal motor system and diseases of transmitter metabolism. In Principles of Neural Science (E. R. Kandel and J. H. Schwartz, Eds.), pp. 347-357. Elsevier/North-Holland, New York. 8. DAVIS, G. C., A. C. WILLIAM, S. P. MARKEY, M. H. EBERT, E. D. CAINE, C. M. REICHERT, AND I. J. KOPIN. 1979. Chronic parkinsonism secondary to intravenous injection of meperidine analogues. Psychiatry Res. 1: 249-254. 9. DUNCAN, M. W., R. M. GARRUTO, L. LAVINE, AND S. P. MARKEY. 1988. 2,2-Amino-3-methylaminopropionic acid in cycad derived foods is an unlikely cause of amyotrophic lateral-sclerosis/parkinsonism. Lancet 10: 631-632. 10. EIDELBERG, E., B. A. BROOKS, W. W. MORGAN, J. G. WALDEN, AND R. H. KOKEMOOR. 1986. Variability and functional recovery in the N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine model of parkinsonism in monkeys. Neuroscience 18: 817-822. 11. FALLON, J. M., AND S. E. LOUGHLIN. 1985. Substantia nigra. In The rat neruous system (G. Paxinos, Ed.), Vol. 1, pp. 353-374. Academic Press, Orlando, FL. 12. FINNEGAN, K. T., I. IRWIN, L. E. DELANNEY, G. A. RICAURTE, AND W. LANGSTON. 1987. 1,2,3,6-tetra-1-methyl-4-(methylpyrrol-2-yl)-pyridine: Studies on the mechanism of action of l-

ENVIRONMENTAL methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Ther. 242: 1144-1151.

J. Phnrmacol.

DOPAMINERGIC Exp.

13. FULLER, R. W., D. W. ROBERTSON, AND HEMRICK-LUECKE. 1987. Comparison of the effects of two MPTP analogs, l-methyl4-(2-thienyl)-1,2,3,6&trahydropyridineand I-methyl-4-(3-thienyl)-1,2,3,6-tetrahydropyridine, on monoamine oxidase in vitro and dopamine in mouse brain. J. Pharmacol. Exp. Ther. 240: 415-420. 14. GRANT, J. 1969. Hackh’s Chemical Dictionary, 4th ed. McGrawHill, New York. 15. HEIKKILA, R. E., L. MANZINO, F. S. CABBAT, AND R. C. DUVOISIN. 1985. Studies in the oxidation of the dopaminergic neurotoxin MPTP by monoamine oxidase B. J. Neurochem. 45: 10491054. 16. HEIKKILA, R. E., L. MANZINO, F. S. CABBAT, AND R. C. DUVOISIN. 1985. Effects of 1-methyl-4-phenyl-1,2,3,6%etrahydopyridine (MPTP) and several of its analogs on the dopaminergic nigrostriatal pathway in mice. Neurosci. L&t. 58: 133-137. 17. HORNYKIEWICZ, O., C. PILF, S. J. KISH, K. SHANNAK, AND G. SCHINGNITZ. 1989. Biochemical changes in idiopathic Parkinson’s disease, aging and MPTP parkinsonism similarities and differences. In Parkinsonism and Aging (D. B. Calne, D. Crippa, G. Comi, R. Hrowski, and M. Trabucchi, Eds.), pp. 57-65. Raven Press, New York. 18. JAVITCH, J. A., AND S. H. SNYDER. 1984. Uptake of MPP+ by dopaminergic neurons explains selectivity of parkinsonism inducing neurotoxin MPTP. Eur. J. Pharmacol. 106: 455-456. 19. JENNER, P., N. M. J. RUPNIAK, S. ROSE, E. KELLY, G. KILPATRICK, A. LEES, AND C. D. MARDSEN. 1984. 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced parkinsonism in the common marmoset. Neurosci. Lett. 50: 85-90. 20. JOHNSON, E. A., E. Y. WV, H. ROLLEMA, R. 0. BOOTH, A. J. TREVOR, AND N. CASTAGNOLI, JR. 1989.1-methyl-4-phenylpyridinium (MPP+) analogs: In vivo neurotoxicity and inhibition of striatal synaptosomal dopamine uptake. Eur. J. Pharmacol. 166:65-74. 21. KITT, C. A., L. C. CORK, F. EIDELBERG, T. H. JOH, AND D. L. PRICE. 1986. Injury of nigral neurons exposed to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine: A tyrosine hydroxylase immunocytochemical study in monkey. Neuroscience 17: 1089-1103. LANGSTON, J. W., P. BALLARD, J. W. TETRUD, AND I. IRWIN. 22. 1983. Chronic parkinsonism in humans due to a product of meperidine analog synthesis. Science 219: 979-980. 23. LANGSTON, W. J., L. S. FORNO, C. S. REBERT, AND I. IRWIN. 1984. Selective nigral toxicity after systemic administration of l-methyl-4-phenyl-1,2,5,6-tetrahydropyridine (MPTP) in the squirrel monkey. Brain Res. 292: 390-394. 24. LANGSTON, J. W., I. IRWIN, E. B. LANGSTON, AND L. S. FORNO. 1984. Pargyline prevents MPTP-induced parkinsonism in primates. Science 225: 1480-1482. 25. LARSON, P. S., AND H. SILVETTE. 1971. In Tobacco: Experimental and Clinical Studies (P. S. Larson and H. Silvette, Eds.), Suppl. 2, pp. 147-167. William & Wilkins, Baltimore. 26. LICHTENSTEIGER, W., U. MLJTZNER, AND H. LANGEMANN. 1967. Uptake of 5-hydroxytryptamine and 5-hydroxytryptophan by neurons of the central nervous system normally containing catecholamines. J. Neurochem. 14: 489-497. 27. LINDVALL, O., AND A. BJORKLUND. 1983. Dopamine and norepinephrine containing neuron systems: Their anatomy in the rat brain. In Chemical Neuroanatomy (P. C. Emson, Ed.), pp. 229255. Raven Press, New York. 28. MARKEY, S. P., C. J. MARKEY, J. M. SAVITT, J. P. BACON, AND J. N. JOHANNESSEN. 1986. Search for environmental neurotoxins related to MPTP: Preparation of antibodies to MPTP and

29.

NEUROTOXINS

149

MPP+. In (S. P. Markey, N. Castagnoli, Jr., A. J. Trevor, and I. J. Kopin, Eds.), pp. 487-493. MPTP: A Neurotoxin Producing a Parkinsonian Syndrome Academic Press, Orlando, FL. MICHEL, P. P. 1985. Application du mod& d’ungerstedt 6 l’t&de des effets neurotoxiques dopaminergiques yl-4-phenyl-1,2,3,6-tetrahydropyridine

chez le rat de la l-meth(MPTP). Thesis, School

of Pharmacy, University of Lyon I, France. 30. MICHEL, P. P., B. K. DANDAPANI, J. SANCHEZ-RAMOS, S. EFANGE, B. C. PRESSMAN, AND F. HEFTI. 1989. Toxic effects of environmental neurotoxins related to 1-methyl-4-phenylpyridinium on cultured rat dopaminergic neurons. J. Pharmacol. Exp.

Ther.

248:

842-850.

31. MICHEL, P. P., B. K. DANDAPANI, B. KNUSEL, J. SANCHESRAMOS, AND F. HEFTI. Toxicity of 1-methyl-4-phenylpyridinium (MPP+) for rat dopaminergic neurons in culture: Selectivity and irreversibility. J. Neurochem., in press. 32. MIZUNO, Y., N. SONE, AND T. SAITOH. 1987. Effects of l-methyl4-phenyl-1,2,3,6-tetrahydropyridine and 1-methyl-4-phenylpyridiniumion on activities of the enzymes in the electron transport system in mouse brain. J. Neurochem. 48: 1787-1793. 33. PARDRIDGE, W. M. 1988. Recent advances in blood brain barrier transport. Annu. Rev. Pharmacol. Toxicol. 28: 25-39. 34. NICKLAS, W. J., I. VYAS, AND R. E. HEIKKILA. 1985. Inhibition of NADH-linked oxidation in brain mitochondria by l-methyl4-phenylpyridinium, a metabolite of the neurotoxin l-methyl-4phenyl-1,2,3,6-tetrahydropyridine. Life Sci. 36: 2503-2508. 35. PERRY, T. L., K. JONES, S. HANSEN, AND R. A. WALL. 1987. 4Phenyl pyridine and three other analogues of I-methyl-4-phenyl-tetra-hydropyridine lack dopaminergic nigrostriatal neurotoxicity in mice and marmosets. Neurosci. Lett. 75: 65-70. 36. PRESSMAN, B. C. 1963. The effects of guanidine and alkyl-guanidines on the energy transfer reactions of bitochondria. J. Biol. Chem. 238: 401-409. 37. PROCHIANTZ, A. M., C. DAGUET, A. HERBET, AND J. GLOWINSKI. 1981. Specific stimulation of in vitro maturation of mesencephalic dopaminergic neurons by striatal membranes. Nature (London)

38.

39.

40.

41.

42.

43.

293:

570-572.

SANCHEZ-RAMOS, J. R., J. N. BARRETT, M. GOLDSTEIN, W. WEINER, AND F. HEFTI. 1986. MPP+ but not MPTP selectively destroys dopaminergic neurons in cultures of dissociated rat mesencephalic neurons. Neurosci. Lett. 72: 215-220. SANCHEZ-RAMOS, J., P. P. MICHEL, W. WEINER, AND F. HEFTI. 1988. Selective destruction of cultured dopaminergic neurons from fetal rat mesencephalon by the neurotoxin 1-methyl-4phenylpyridinium (MPP+): Cytochemical and morphological evidence. J. Neurochem. 50: 1934-1944. SCHINELLI, S., A. ZUDDAS, I. J. KOPIN, J. L. BARKER, AND H. DI PORZIO. 1988. 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine metabolism and 1-methyl-4-phenylpyridinium uptake in dissociated cell cultures from the embryonic mesencephalon. J. Neurothem. 50: 1900-1907. SINGER, T. P., N. CASTAGNOLI, JR., R. R. RAMSAY, AND A. J. TREVOR. 1987. Biochemical events in the development of parkinsonism induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. J. Neurochem. 49: 1-8. SINGER, T. P., R. R. RAMSAY, AND K. A. MCKEOWN. 1988. Biochemistry of the neurotoxic action of MPTP and what it may teach us about the etiology of idiopathic parkinsonism. In Recent Advances in Parkinson Research (F. Hefti and W. Weiner, Eds.), pp. 101-111. Plenum, New York. SPECHT, L. A., W. M. PICKEL, T. H. JOH, AND D. REIS. 1981. Light microscopic immunochemical localization of tyrosine hydroxylase in prenatal rat brain. I. Early ontogeny. J. Comp. Neurol. 199:233-253.

150 44.

45.

46.

47.

MICHEL SPENCER, P. S. 1989. Western pacific ALS-parkinsonism-dementia: A model of neuronal aging triggered by environmental toxins. In Parkinsonism and Aging (D. B. Calne, D. Crippa, G. Comi, R. Horowski, and M. Trabucchi, Eds.), pp. 133-144. Raven Press, New York. SPENCER, P. S., P. B. NUNN, J. HUGON, A. C. LUDOLPH, S. M. Ross, D. N. ROY, AND R. C. ROBERTSON. 1987. Guam amyotrophic lateral sclerosis parkinsonism-dementia linked to a plant excitant neurotoxin. Proc. Natl. Acad. Sci. USA 237: 517-522. VYAS, I., R. E. HEIKKILA, AND W. J. NICKLAS. 1986. Studies on the neurotoxicity of MPTP. Inhibition of NADH-linked substrate oxidation by its metabolite MPP+. J. Neurochem. 46: 1501-1507. WATERS, C. M., S. P. HUNT, P. JENNER, AND C. D. MARDSEN. 1987. An immunohistochemical study of the acute and long term effects of 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine in the marmoset. Neuroscience 23: 1025-1039.

ET 48.

49.

50.

51

: AL. YOUNGSTER, S. K., W. S. SAARI, AND R. E. HEIKKILA. 1987. lmethyl-4-cyclohexyl-1,2,3,6-tetrahydropyridine (MCTP); an alicyclic MPTP-like neurotoxin. Life Sci. 79: 151-156. YOUNGSTER, S. K., P. K. SONSALLA, AND R. E. HEIKKILA. 1987. Evaluation of the biological activity of several analogs of the dopaminergic neurotoxin 1-methyl-4-pheny-1,2,3,6-tetrahydropyridine. J. Neurochem. 48: 929-934. YOUNGSTER, S. K., P. K. SONSALLA, B-A. SIEBER, AND R. E. HEIKKILA. 1989. Structure-activity study of the mechanism of 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP)-induced neurotoxicity. I. evaluation of the biological activity of MPTP analogs. J. Pharmacol. Exp. Ther. 249: 820-828. YOUNGSTER, S. K., W. J. NICKLAS, AND R. E. HEIKKILA. 1989. Structure-activity study of the mechanism of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced neurotoxicity. II. Evaluation of the biological activity of the pyridinium metabolites formed from the monoamine oxidase-catalyzed oxidation of MPTP analogs. J. Pharmacol. Exp. Ther. 249: 829-835.