Journal vjilieurochemistry Raven Press, Ltd., New York 0 199 I International Society for Neurochemistry

Prolonged Alterations in Canine Striatal Dopamine Metabolism Following Subtoxic Doses of 1 -Methyl-4-Phenyl- 1,2,3,6-Tetrahydropyridine (MPTP) and 4’-Amino-MPTP Are Linked to the Persistence of Pyridinium Metabolites

*?Jan N. Johannessen, ?Thomas J. Sobotka, $Virginia K. Weise, and *Sanford P. Markey

*Laboratory of Clinical Science, National Institute of Mental Health, $Laboratory of Neural Regeneration, National Institute ofNeurologica1 Disorders and Stroke, Bethesda, Maryland; and tCenter for Food Safety and

Applied Nutrition, Food and Drug AdminisIration, Washington, D.C.. U S A .

Abstract: Single toxic doses of I -methyl-4-phenyl- I ,2,3,6- tetrahydropyridine (MPTP) * HCI (2.5 mg/kg i.v.) and 4’- amino-MPTP- 2HC1(22.5 mg/kg) induce loss of stnatal do- pamine (DA) and tyrosine hydroxylase (TH) activity and of nigral DA neurons in the dog. To examine the subacute neu- rochemical changes induced by low doses of MPTP and 4’- amino-MPTP, dose-response studies of these compounds were camed out in the dog, using 6- and 3-week survival times for these two compounds, respectively. Low single doses of MPTP ( 1 .O, 0.5, and 0.1 mg/kg i.v.) and 4’-amino-MPTP (1 5,7.5, and 3.75 mg/kg i.v.) did not cause depletion of canine striatal DA or TH or a loss of nigral neurons. However, levels of the DA metabolites 3,4-dihydroxyphenylacetic acid (DO- PAC) and homovanillic acid (HVA) were decreased in a dose- related fashion, with significant loss of DOPAC being evident 6 weeks after the lowest administered dose of MPTP and 3 weeks after 4’-amino-MPTP. This selective loss of DA me- tabolites following nontoxic doses of MPTP and 4‘-amino- MPTP led to a shift in the ratio of DA to DOPAC or HVA, which was characteristic for each compound. The measure-

ment of striatal 1-methyl-4-phenylpyridinium (MPP+) and 4‘-amino-MPP’ levels revealed that high concentrations (up to 150 p M ) persist in the striatum for weeks following ad- ministration of a single nontoxic dose of MPTP or 4‘-amino- MPTP. A causal relationship between the striatal concentra- tion of MPP’ or 4’-amino-MPP+ and the change in DA me- tabolism as reflected in the DA/DOPAC ratio is suggested by a significant correlation between these measures. It is sug- gested that presynaptic sequestration and retention of MPP’ and 4-amino-MPP’ by striatal DA terminals result in the inhibition of the monoamine oxidase contained within these terminals. Key Words: 1-Methyl-4-phenyl- 1,2,3,6-tetrahy- dropyndine- 1 -Methyl-4-phenylpyndinium-Parkinson- ism-Dopamine-Dog-Dose response. Johannessen J. N. et al. Prolonged alterations in canine striatal dopamine me- tabolism following subtoxic doses of I -methyI-4-phenyl- 1,2,3,6-tetrahydropyridine (MPTP) and 4’-amino-MPTP are linked to the persistence of pyridinium metabolites. J. Neu- rochem. 57,98 1-990 ( I 99 1).

1 - Methyl - 4 - phenyl - 1,2,3,6 - tetrahydropyridine (MPTP) and its pyridinium metabolite 1 -methyl-4- phenylpyridinium (MPPf) exhibit a range of biological activities, despite the simplicity of their chemical forms. MPTP is a good substrate for monoamine oxidase (MAO) [amine: oxygen oxidoreductase (deaminating) (flavin containing); EC 1.4.3.41 B (Chiba et al., 1984), whereas MPPf is a substrate for the catecholamine transport system (Javitch and Snyder, 1984), is con- centrated within mitochondria (Ramsay and Singer,

1986; Hoppel et al., 1987), and is a potent inhibitor of mitochondria1 respiration (Nicklas et al., 1985). This complex sequence of biological events necessary for the expression of MPTP toxicity-transport into the brain, metabolic activation, accumulation by dopa- minergic terminals, and mitochondrial toxicity-leads to the suspicion that MPTP is quite unique. This may, however, not be the case. Several MPTP analogs with substitutions on the phenyl ring have been synthesized and found to be toxic (Youngster et al., 1986, 1987,

Received October 18, 1990; revised manuscript received February 1 1, 1991; accepted February 14, 1991.

Address correspondence and reprint requests to Dr. J. N. Johan- nessen at Laboratory of Clinical Science, National Institute of Mental Health, Building 10, Room 3D-40, Bethesda, MD 20892, U.S.A.

Abbreviations used: DA, dopamine; DDC, DOPA decarboxylase; DOPA, 3,4-dihydroxyphenylalanine; DOPAC, 3,4-dihydroxyphen- ylacetic acid; HVA, homovanillic acid; MAO, monoamine oxidase; MPPt, 1-methyl-4-phenylpyridinium; MPTP, 1 -methyl-.l-phenyl- 1,2,3,6-tetrahydropyndine; TH, tyrosine hydroxylase.

981

982 J. N. JOHANNESSEN ET AL.

1989). This work has demonstrated that both forms of MA0 (A and B) can act as catalysts for the oxidation of certain MPTP analogs. In addition, there have been reports of nigrostriatal toxicity from analogs with more extensive structural dissimilarities (Wilkening et al., 1986). It is interesting that some of these MPTP analogs could have an endogenous origin (Niwa et al., 1987) or occur naturally in certain foods (Makino et al., 1988).

Selective toxins for catecholamine- and indolamine- containing neurons have been known for some time. What distinguishes MPTP-like toxins from previously described toxins, such as 6-hydroxydopamine and 5,7- dihydroxytryptamine, is their ability to cause lesions of specific brain areas after systemic administration. This fact, coupled with the extremely low doses re- quired in sensitive species, such as the primate and the dog, has led to speculation that endogenous or envi- ronmental molecules similar to MPTP may contribute to the etiology of parkinsonism in humans. Certain epidemiological studies (Barbeau et al., 1987), coupled with the observations that treatment with an MAO-B inhibitor delays the onset of symptoms (Tetrud and Langston, 1989) and increases the longevity of parkin- sonian patients when combined with ~-3P-dihydroxy- phenylalanine (L-DOPA) therapy (Birkmayer et al., 1985), have provided circumstantial evidence consis- tent with this hypothesis, but these findings are ame- nable to alternative explanations.

The onset of parkinsonism is gradual, in contrast to the relatively rapid onset of symptoms following toxic doses of MPTP in humans (Davis et al., 1979; Langston et al., 1983), primates (Burns et al., 1983), and dogs (Johannessen et al., 1989). At least two alternative ex- posure scenarios might explain this discrepancy. In the first, one could assume that chronic exposure to low doses of an MPTP-like chemical eventually leads to slow neural degeneration. Alternatively, it has been hypothesized that exposure over a short interval to doses that cause a partial lesion may hasten the onset of parkinsonism in later years. In either case, a more detailed understanding of the long-term consequences of low doses of MPTP in a sensitive species is central to understanding the mechanisms by which toxins might lead to parkinsonism. Moreover, discovery of biological markers for exposure to MPTP-like com- pounds would allow some etiological questions about parkinsonism to be addressed more directly. In an ini- tial effort to address these questions, we have under- taken a dose-response study of MPTP and its 4‘-amino analog in the dog and examined the subacute effects (3-6 weeks) of this treatment on markers for nigrostri- atal dopamine (DA) neurons.

Protocols for the two dose-response studies were de- veloped independently, accounting for the difference in survival times (3 vs. 6 weeks). Because the qualitative effects of MPTP and 4’-amino-MPTP were similar, we thought it important to present the results of these studies together, even though strict comparisons of the

magnitude of effects produced by the two compounds are not possible because of the differences in survival times, age, and sex. Throughout this report, “toxic” doses are defined as those that cause an overt loss of nigrostriatal DA neurons. Doses that cause subacute neurochemical changes without evidence of cell loss are termed “subtoxic.”

MATERIALS AND METHODS

Adult beagle dogs were used in all studies. In the MPTP dose-response study, all were male, 48-70 weeks old, bred and raised at the Food and Drug Administration Beltsville Research Facility and housed singly in kennels with both indoor and outdoor areas. Dogs for the 4‘-amino-MPTP dose- response study were from Ridglan Farms (Mt. Horeb, WI, U.S.A.) and were female, 3-4 years old, and housed in pairs in 1 3-m2 kennels. All dogs had free access to food and water. An outside run was adjacent to each kennel and was freely accessible through a swinging door.

Sterile solutions of MPTP. HCl (Aldrich) or q-amino- MPTP * 2HC1 [previously synthesized (Johannessen et al., 1987)] in saline were injected through a catheter placed in the cephalic vein. Sterile saline was used in control animals. To minimize the exposure of laboratory personnel to 4‘- amino-MPTP, injections of this compound were given through a sterile length of tubing with the animals in a cage equipped with a mesh floor, with a pan covered by absorbent paper underneath the cage. Treated animals remained in the cages for at least 48 h before being returned to their runs for the remainder of the study. The absorbent paper was bagged and incinerated as medical pathological waste, and the cages were washed, MPTP injections were given in a similar man- ner, but the dogs were returned immediately to their runs, which were thoroughly washed each day. The principal in- vestigator was responsible for handling any materials, in- cluding waste, that might have become contaminated with MPTP. During all procedures in which exposure to MPTP or 4’-amino-MPTP was possible, disposable gloves, mask, cap, and laboratory coats were worn and subsequently incinerated as medical pathologic waste. Details of the exact dosages, number of animals per group, and survival times are shown in Table 1.

At the end of the experiments, animals were killed with an overdose of barbiturate (Lethane), and the brains were

TABLE 1. Experimentul treutments

Survival No. of Dose Treatment (mg/kg) (weeks) dogs

Saline MPTP . HCI MPTP. HCI MPTP * HCI MPTP. HCI

Saline 4-Arnino-MPTP * 2HC1 4-Amino-MPTP. 2HCl 4‘-Amino-MPTP - 2HCI 4‘-Amino-MPTP * 2HC1

- 0.1 0.5 1 .o 2.5

- 3.15 7.5

15.0 22.5

6 6 6 6

16

3 3 3 3 6

5 4 4 4 1

3 3 3 3 4“

a Data from this group are taken from J. N. Johannessen et al. (manuscript in preparation).

J . Nrurochem.. Vol. 57, No. 3, 1991

MPTP AND 4-AMlNO-MPTP DOSE-RESPONSE STUDY 983

rapidly removed, blocked in the coronal plane, and frozen on dry ice. Half of the brainstem block containing the sub- stantia nigra was immersed in buffered formalin for histo- logical analysis. All injections and treatments were performed in accordance with approved protocols.

Striatal DA and its metabolites, 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA), were quantified by HPLC with electrochemical detection. In brief, small punches (5-10 mg) were taken from the central part of the head of the caudate and sonicated in 1 .O ml of Ice-cold 0.1 M perchloric acid. After removal of two 10-p1 aliquots for the quantification of protein (Lowry et al., 1951), the ho- mogenates were spun at 12,000 g for 10 min in a refrigerated centrifuge. Aliquots (20 pl) of the supernatant were injected onto a column (4.6 mm X 10 cm) packed with 3-pm C18 (Hypersil; Shdndon). The mobile phase consisted of 0.2 A4 citric acid, 6% acetonitrile, 100 mg/L of EDTA, 450 mg/L of octanesulfonic acid, and 5 ml/L of triethylamine (no pH adjustment). Levels of DA and metabolites were measured amperometrically with the potential set to 0.75 V, by com- parison with external standards. Data on striatal DA, DO- PAC, and HVA concentrations from the toxic dose of 4'- amino-MPTP (22.5 mg/kg) were taken from another study (J. N. Johannessen, manuscript in preparation).

The striatal concentrations of the pyridinium metabolites MPP+ and 4'-amino-MPP+ were also assayed in small punches. Tissues were sonicated for 10 s in 0.25 ml of ice- cold 0.1 M perchloric acid, and after removal of a 1 0-pl al- iquot for protein content determination, the homogenates were spun at 12,000 g for 10 min. For the estimation of MPP', 50-pI aliquots of the supernatant were injected onto a 4.6-mm X 15-cm, 5-pm C18 column with a mobile phase of 80% 0.1 M sodium acetate containing 0.1% triethylamine (pH 5.6) and 20% acetonitrile, which also contained 0.1% triethylamine. The flow rate was 1 .O ml/min, through a UV flow cell set at 290 nm. MPP' was quantified by comparison of peak heights with a standard curve generated with authentic MPP'. The assay was validated against a previously developed method using gas chromatography-mass spectrometry (Shih and Markey, 1986; S. P. Markey et al., unpublished data). A similar procedure was used to quantify 4'-amino-MPP+, ex- cept that the percentage of acetonitrile in the mobile phase was reduced to 10% and the monitoring wavelength was 380 nm.

Dissected pieces of caudate were used for the assessment of tyrosine hydroxylase (TH) [I>-tyrosine, tetrahydropterin: oxygen oxidoreductase (3-hydroxylating); EC 1.14.16.21 ac- tivity and DOPA decarboxylase (DDC) (aromatic L-amino acid carboxylase; EC 4.1.1.28) activity. TH activity was as- sayed according to the method of Waymire et al. (197 1) in- corporating the modifications of Lerner et al. ( 1978). DDC was assayed according to the procedure of Lloyd and Hor- nykiewicz (1 972).

Fixed blocks containing the substantia nigra were cut with a freezing microtome at 50 pm, and sections were mounted and stained with thionin. Estimates of cell density were made by counting the number cells within a 0.325-mm2 grid in matched sections. Three such grids were counted in two se- quential sections containing the substantia nigra pars com- pacts, and the six numbers were averaged to yield the mean cell density.

Group comparisons were made by using an analysis of variance followed by the posthoc Sheffk PLSD f test. Com- parisons of slopes generated by regression analyses were made using the general Linear models procedure.

RESULTS

MPTP dose-response study The effects of various doses of MPTP on striatal DA

and DA metabolites 6 weeks after injection are shown in Fig. 1A. The toxic dose of MPTP - HCl(2.5 mg/kg), which in contrast to the lower doses was given 16 weeks before the animals were killed, did cause a complete loss of striatal DA, TH activity, and massive cell loss

0 0.1 0.5 1 2.5

MPTP-HCI Dose (mg/kg)

FIG. 1. A: Effect of the indicated doses of MPTP on striatal con- centrations of DA, DOPAC, and HVA 6 weeks after injection (16 weeks for the one dog given 2.5 mg/kg). Subtoxic doses cause a selective fall in levels of the DA metabolites DOPAC and HVA. Control averages (mean k SEM) were as follows: DA, 11 6.8 k 4.4; DOPAC, 19.4 _t 2.2; and HVA, 254 i 26 pg/pg of protein. The number of dogs per group is given in Table 1. Statistically significant differences with respect to the control mean are indicated: ' p < 0.05. 8: Ratios of DA/DOPAC and DA/HVA calculated from the data shown in A. Ratios of DA to its metabolites increased at subtoxic doses of MPTP but decreased in the lesioned animal. Control ratios (mean k SEM) were as follows: DA/DOPAC, 6.06 ? 0.55; DA/HVA, 0.468 +- 0.003. Statistically significant differences with respect to the control mean are indicated: ' p < 0.05. C Striatal concentrations of MPP+ (in pg/pg of protein). D: Striatal TH activity (in nmol of C02/mg of protein/h). Only the standard toxic dose of MPTP (2.5 mg/kg) caused a depletion of TH activity. Statistically significant differences with respect to the control mean are indi- cated: *p i 0.05.

.I Neurochem.. Vol 57 , No. 3, 1991

984 J. N. JOHANNESSEN ET AL.

within the substantia nigra pars compacta. This is vir- tually identical to earlier results demonstrating a com- plete loss of the canine nigrostriatal system within 3 weeks of treatment with the same dose (Johannessen et al., 1989). Lower doses of MPTP (0.1, 0.5, and 1.0 mg/kg) decreased striatal DA content only slightly (1 .O mg/kg) or not at all (0.1 and 0.5 mg/kg). The DA me- tabolites, especially DOPAC, were affected in an en- tirely different way. The lowest dose given (0.1 mg/ kg), only 1/25th of the standard toxic dose, caused a significant decline in striatal DOPAC level that was still evident 6 weeks after injection. The effect of MPTP treatment on DA metabolites continued at the 0.5 and 1 .O mg/kg dose, with progressive declines being seen in both DOPAC and HVA levels with only a minimal decrease in the mean DA concentration. Whereas the mean DA concentration at the 1.0 mg/kg dose was only slightly decreased, two of the four animals in this dose group had striatal DA levels that were 60-70% below the control mean, with the others having values within the control range. The decrease in metabolite levels seen 6 weeks after treatment with subtoxic doses of MPTP, coupled with the slight tendency of the stria- tal DA content to increase at the 0.1 and 0.5 mg/kg doses, resulted in a marked increase in the DA/DOPAC and DA/HVA ratios, as shown in Fig. 1 B. The direction of change in the DA/metabolite ratio caused by low doses of MPTP was opposite that induced by the toxic dose, after both a 16-week survival or a 3-week survival (Johannessen et al., 1989).

The striatal concentrations of MPP’ were surpns- ingly high (Fig. lC), increasing from - 1 pg/pg of pro- tein after the 0.1 mg/kg dose to - 5 pg/pg of protein after the 0.5 mg/kg dose. Further increases in concen- tration did not occur at the 1.0 mg/kg dose. For pur- poses of comparison, the highest individual concen- tration measured (-9 pg/pg of protein) is roughly twice the normal striatal serotonin concentration and almost 10% of the normal striatal DA concentration. No MPP+ was detected in the one animal treated with MPTP at 2.5 mg/kg; however, the prolonged survival time of this animal makes direct comparison with the other subjects impossible. However, it should be noted that the two animals in the 1 .O mg/kg group that sustained a substantial striatal DA loss also had low striatal MPP+ concentrations (Table 2), a result suggesting a parallel loss of DA and MPP+.

Striatal TH activity generally paralleled striatal DA concentrations (Fig. ID). No changes in TH activity occurred at the 0.1 and 0.5 mg/kg doses. Following the 1.0 mg/kg dose, there was a slight decrease in the mean striatal TH activity. The two animals in this dose group that had decreased striatal DA content also had substantial losses of TH activity.

4‘-Amino dose-response study The 4’-amino analog of MPTP was by weight much

less potent than MPTP, but the toxic dose (22.5 mg/ kg) produced all the neurochemical, behavioral, and

TABLE 2. Substantia nigra cell density and striatal DA markers in dogs with a partial lesion

of the nixrostriatal system

Striatal MPTP Cell density

Dog no. (mg/kg) (mean 2 SE) DA MPP+ TH activity

59B 0 58.2 k 2.4 124 0 47.7 51B 1 .O 51.6 i= 3.2 103 9.56 45.4 87B 1 .o 47.7 f 3.5 104 5.34 39.6 58B 1 .o 45.2 f 4.8 38 2.32 22.1 73B 1 .o 45.7 f 2.5 44 3.23 29.3

Comparison of individual values for cell density (no. of cells/0.325 mm2), striatal DA (pg/pg of protein), MPP+ (pglpg of protein), and TH activity (nmol of C02/mg of protein/h) in a control dog (59B) and four dogs treated with the 1.0 mg/kg dose of MPTP (51B, 87B, 58B, and 73B). Although two of the dogs receiving 1.0 mg/kg had decreased values for striatal markers for dopaminergic terminals (58B and 73B), the cell density was comparable to that in dogs given the same dose whose values for striatal DA markers were normal ( 5 1 B and 87B). Note also that the highest concentrations of striatal MPP+ are not associated with decreased TH activity.

neuropathologic changes indicative of the complete loss of the nigrostriatal system 6 weeks after administration (J. N. Johannessen et al., manuscript in preparation). The changes seen 3 weeks after administration of sub- toxic doses of 4’-amino-MPTP were similar to those seen after subtoxic MPTP doses but were more pro- nounced (Fig. 2). Thus, after the lowest dose (3.75 mg/ kg), DA levels were nearly doubled, whereas DOPAC and HVA concentrations were approximately halved (Fig. 2A). The loss of striatal DOPAC and HVA was virtually complete with the 7.5 mg/kg dose, as there were no significant differences in striatal DOPAC or HVA levels between the 7.5 and 15 mg/kg doses. There was no significant decrease in the mean DA concen- tration at any of the subtoxic doses; however, it should be noted that one of the three animals at the 15 mg/ kg dose sustained a 90% loss of striatal DA. The in- creases in the DA/DOPAC and DA/HVA ratios seen after 4’-amino-MPTP were more pronounced than those seen after MPTP (Fig. 2B). It is not known whether this was due to differences in action between MPTP and 4‘-amino-MPTP or to the difference in sur- vival time between experiments (6 weeks for MPTP and 3 weeks for 4‘-amino-MPTP).

4’-Amino-MPP+ was also retained within the stria- tum 3 weeks after injection of 4‘-amino-MPTP (Fig. 2C). After the lowest dose of 4’-amino-MPTP, the striatal concentration of 4’-amino-MPP+ was 16 pg/pg of protein. The intermediate dose yielded a concentra- tion of 24 pg/pg of protein, with some individual values approaching 30 pg/pg of protein. As with MPTP, the striatal pyridinium concentration following injection of 4’-amino-MPTP seemed to reach a maximum be- yond which increasing the dosage of 4’-amino-MPTP did not yield an increased striatal concentration of 4- amino-MPP+. No 4’-amino-MPP+ was detected in dogs that received a toxic dose of 4’-amino-MPTP, but again

J. Neurochem , Vol. 57, No. 3. 1991

MPTP AND 4'-AMI"-MPTP DOSE-RESPONSE STUDY 985

lcll T T

I

1'5 22.5 4'Amino MPTP-2HCI Dose (mg/kg)

FIG. 2. A Striatal concentrations of DA, DOPAC, and HVA 3 weeks after indicated doses of 4'-amino-MPTP (the 22.5 mg/kg group was killed 6 weeks after treatment). The lowest dose caused an increase in DA level and a decrease in DOPAC and HVA levels. Progressive declines in DOPAC and HVA concentrations occurred as the dose increased. Control concentrations (mean f SEM) were as follows: DA, 137 2 23; DOPAC, 19.8 ? 7.8; and HVA, 284 t- 73 pg/pg of protein. Statistically significant differences with re- spect to the control mean are indicated: *p < 0.05. 8: Ratios of DA/DOPAC and DA/HVA calculated from the data in A. Subtoxic doses of 4'-amino-MPTP caused a progressive increase in these ratios, but the toxic dose caused a decline. Control ratios (mean % SEM) were as follows: DA/DOPAC, 9.81 * 3.80; DA/HVA, 0.522 % 0.077. Statistically significant differences with respect to the control mean are indicated: 'p < 0.05. C: Striatal concentrations of 4'-amino-MPP+ (in pg/pg of protein). D: Striatal activities of TH and DDC 3 weeks after indicated doses of 4'-amino-MPTP. Control activities (mean ? SEM) were as follows: TH, 24.3 f 3.3; DDC, 41.6 k 4.9 nmol of COdmg of protein/h. Statistically significant differences with respect to the control mean are indicated: *p c: 0.05.

the survival time for this dose was twice that of the lower doses, making direct comparisons impossible. A parallel loss of DA and MPP' was suggested by the fact that one animal in the 15 mg/kg group that sus- tained a 90% striatal DA loss had a striatal MPP+ con- centration that was only 10% of that of the other two subjects in the group.

The activities of the enzymes TH and DDC paral- leled the DA values (Fig. 2D). Other than a slight de- crease in the mean DDC activity after the 15 mg/kg dose, neither mean enzyme activity was significantly altered by the subtoxic doses of 4'-amino-MPTP, but the animal in the 15.0 mg/kg group that had a 90% loss of striatal DA also had a comparable loss of striatal TH activity.

Correlations The fundamental change in DA metabolism after

subtoxic doses of MPTP or 4'-amino-MPTP is mani- fested as an increase in the DA/DOPAC ratio (Figs. 1B and 2B). To assess whether changes in this ratio were related to striatal concentrations of the pyridinium metabolites, individual values of the striatal MPP+ or 4'-amino-MPPf concentration were compared with the corresponding DA/DOPAC ratios in those animals re- ceiving subtoxic doses of these toxins or saline. As seen in Fig. 3A, there was a very high degree of correlation between MPP' concentrations and the DA/DOPAC ratio. The correlation was even stronger between 4'- amino-MPP' concentrations and corresponding DA/ DOPAC ratios (Fig. 3B).

To examine the individual relationships between the degree of loss of TH activity and striatal DA concen- trations, a correlational analysis was undertaken to compare individual striatal TH activities with the cor- responding DA concentrations. TH activities and DA concentrations were highly correlated: for the MPTP dose-response study, R2 = 0.85, p < 0.001; for the 4'- amino-MPTP dose-response study, R2 = 0.73, p < 0.001. Thus, the animals that sustained partial le- sions as judged by decreased striatal TH activity had correspondingly decreased striatal DA concentrations.

Because of the opposite nature of the changes in levels of DA and its metabolites following subtoxic doses of MPTP and 4'-amino-MPTP (Figs. 1 and 2), one would predict a poor relationship between the DA metabolite concentrations and the DA concentrations. This was in fact the case. When individual DA con- centrations from all animals given a subtoxic dose of MPTP or 4'-amino-MPTP were compared with the

a

R2 =.71 p< ,001

40

R2 =.91 p< .001

0 2 4 6 8 10 0 10 20 30 40 MPP+ (pg/&g prot.) 4'NHp-MPP+ (pg/pg prot.)

FIG. 3. The tight relationship between the measured striatal con- centrations of (A) MPP' or (6) 4'-amino-MPP+ and an index of DA metabolism, the ratio of DA/DOPAC. After either MPTP treatment (A) or 4'-amino-MPTP treatment (B), increasing striatal concentra- tions of the pyridinium metabolites were clearly linked to increases in the DA/DOPAC ratio.

J. Neurochem., Vol. 57. No. 3, 1991

986 J. N. JOHANNESSEN ET AL.

corresponding DOPAC or HVA concentrations, no significant correlations were found: for DA versus DO- PAC, R2 = 0.03, p = 0.370; for DA versus HVA, R2 = 0.06, p = 0.223.

Quite a different picture was seen if the data were analyzed by grouping the animals according to whether they were controls (pooled from both dose-response experiments), MPTP-treated, or 4‘-amino-MPTP- treated. In this case (Fig. 4A), DA and DOPAC were highly correlated within a given group. Furthermore, each treatment group exhibited a unique slope that was significantly different from the control group. Similar highly significant correlations were seen be- tween DA and HVA for all groups (Fig. 4B), with each group (control, MPTP, or 4’-amino-MPTP) having a different slope. DOPAC and HVA also covaried within each group, but the slopes were not different from each other (Fig. 4C).

Histology Examination of the thionin-stained sections of the

substantia nigra pars compacta revealed that dogs in the lower two dose groups from either dose-response group (MPTP, 0.1 and 0.5 mg/kg; 4’-amino-MPTP, 3.75 and 7.5 mg/kg) showed no evidence of cell loss or of the loss of striatal DA or TH activity. In contrast, dogs in the highest dose groups (MPTP, 2.5 mg/kg; 4’- amino-MPTP, 22.5 mg/kg) all had extensive cell loss as well as striatal DA depletion. Although the survival times for the animals receiving toxic doses of MPTP were considerably longer than for animals receiving lower doses, previous work in the dog suggests that depletion of striatal markers of DA terminals and nigral cell loss occurs within 3-4 weeks of dosing (Johannes- sen et al., 1989). Following the dose of each compound just below the standard toxic dose (MPTP, 1 .O mg/kg; 4’-amino-MPTP, 15.0 mg/kg), the loss of striatal DA and TH activity ranged from none to >90%. In the 4’- amino-MPTP-treated animal that sustained >90% loss of striatal DA and TH activity, the cells within the nigra were still visible but appeared abnormal, with very weak Nissl staining, possibly reflecting ongoing degeneration when they were killed (3 weeks postin- jection). However, the two animals that had a 60-70% loss of striatal DA after receiving 1 .O mg/kg of MPTP had no evidence of cell loss in the substantia nigra pars compacta 6 weeks after dosing (Table 2). For example, Fig. 5 shows Nissl-stained sections from a control an- imal (Fig. 5A), a lesioned animal (Fig. 5B), and two animals at the 1 .O mg/kg MPTP dose, one with normal striatal DA content (Fig. 5C) and the other with partial DA depletion (Fig. 5D). The animal that sustained a 70% loss of striatal DA (Fig. 5D) did not have a com- parable loss of nigral cells (Table 2). Thus, losses in stnatal DA and TH activity were not always accom- panied by nigral cell loss, an observation borne out by the comparison of stnatal markers for DA terminals and cell densities within the substantia nigra pars com- pacts in partially lesioned animals (Table 2).

8 g 10 a - &A

0 0 100 200

Doparnine

Control A MPTP 0 4‘-Amino 1 MPTP 1

0 100 200 300 Doparnine

0 10 20 30 DOPAC

FIG. 4. Correlation of (A) DA with DOPAC, (6) DA with HVA, and (C) DOPAC with HVA for each treatment group. Controls (0) were pooled from both dose-response curves. Dogs treated with sub- toxic doses of MPTP (A) and dogs treated with subtoxic doses of 4’-amino-MPTP (0) were examined. Because the purpose was to gauge the influence of subtoxic doses of MPTP and 4‘-amino- MPTP on DA metabolism, data for fully lesioned animals were not included in the analysis. A: Within all treatment groups (saline, MPTP, and 4‘-amino-MPTP), the correlation between striatal DA and striatal DOPAC levels was highly significant. However, each treatment group had a different slope: control, p < 0.01, R2 = 0.75, slope = 0.281; MPTP, p < 0.005, R2 = 0.57, slope = 0.070; and 4’-amino-MPTP, p < 0.005, R2 = 0.76, slope = 0.035. B: Within all treatment groups (saline, MPTP, and 4’-amino-MPTP), the cor- relation between striatal DA and striatal HVA levels was highly significant. However, each treatment group exhibited a different slope: control, p < 0.002, R2 = 0.83, slope = 3.04; MPTP, p < 0.001. R2 = 0.74, slope = 1.38; and 4’-amino-MPTP,p < 0.001, R2 = 0.93, slope = 0.51. C: Striatal DOPAC and HVA levels were also highly correlated within all groups. Unlike correlations between DA and its metabolites, the slope of these regressions were not significantly different in any of the groups: control, p < 0.005, R2 = 0.76, slope = 9.00; MPTP,p < 0.001, R2 = 0.79, slope = 15.3; and 4’-amino-MPTP, p < 0.001, R2 = 0.91, slope = 12.8.

DISCUSSION

Our data demonstrate that low doses of MPTP or 4’-amino-MPTP, doses that represent a small fraction of a toxic dose, can have relatively long-lasting (>6 weeks) effects on striatal DA metabolism. Although the mechanism of this effect is unknown, a good can- didate is the selective inhibition by MPP’ or 4‘-amino- MPP’ of MA0 contained within DA terminals. Several observations support this explanation. MPP+ is a com-

J . Neurochem., Vol. 57, No 3, 1991

MPTP AND 4'-AMINO-MPTP DOSE-RESPONSE STUDY 98 7

FIG. 5. Photomicrographs of Nissl-stained sections through the substantia nigra pars compacta (SNc) of animals receiving (A) saline, or MPTP. HCI at (6) 2.5 or (C and D) 1 .O mg/kg. Although one of the dogs treated with the 1 .O mg/kg dose sustained a substantial loss (60%) of striatal DA (D), there was no comparable loss of cells. A list of values for striatal biochemical markers and nigral cell counts for all animals at the 1 .O mg/kg MPTP dose group is given in Table 2.

petitive inhibitor of both forms of MA0 in vitro (Ki- nemuchi et al., 1985), although MAO-A is more sen- sitive to inhibition from MPP' (Takamidoh et al., 1987). We have demonstrated the presence of MPP' in the striatum in concentrations that correlate well with the degree of apparent MA0 inhibition, as judged by increases in the DA/DOPAC ratio. In the rat, high doses of MPTP have been shown to inhibit selectively presynaptic MAO-A and -B ex vivo (Arai et al., 1986) but only for a short time after injection; this observation is in keeping with the short half-life of MPP+ in the brain of this species (Johannessen et al., 1985). If the concentration of MPP' in the rat striatum is held con- stant by introducing the compound through a dialysis cannula, the efflux of DOPAC is decreased selectively (Caliguri and Johannessen, 1991), an effect that is de- pendent on uptake of MPP' into DA terminals because

it is blocked if the entrance of MPP' into the terminals is prevented by the coinfusion of a DA uptake blocker. In PC12h cells subtoxic concentrations of MPP' can inhibit MAO-A (Naoi et al., 1987) without affecting TH activity or DA content, a result that parallels these in vivo results.

Although the preponderance of evidence suggests presynaptic MA0 inhibition by MPP+ as a likely mechanism for the neurochemical effects of low doses of MPTP, other mechanisms are not precluded. In- creased vesicular storage of DA, decreased TH activity, inhibition of cellular activity, or inhibition of DA reup- take could all contribute to the tendency for DOPAC and HVA levels to decline whereas the DA content increases.

The persistence of such high concentrations of MPP' and 4'-amino-MPP+ in striatum is quite remarkable.

J. Neurochem., Vol. 57, No. 3, 1991

988 J. N. JOHANNESSEN ET AL.

Because striatal MPP’ concentrations are dramatically reduced following MPTP injections in animals with prior lesions of the nigrostriatal system (Herkenham et al., 1991) and were also low in the present study in animals with partial lesions (low striatal DA concen- trations), the site of storage is apparently dopaminergic terminals [striatal serotonin concentrations were nor- mal after treatment and noradrenergic terminals are a minor constituent of striatum (J. N. Johannessen et al., unpublished data)]. If one assumes that 10% of tis- sue wet weight is protein and that dopaminergic ter- minals occupy 10% of the striatal neuropil, then the concentration of MPP+ within dopaminergic terminals 6 weeks after a 0.5 mg/kg dose approaches 30 pA4 [ 3 weeks after the same dose the concentration is roughly 80 pA4 (J. N. Johannessen, unpublished data)]. Based on these same assumptions, the concentration of 4’- amino-MPP+ within dopaminergic terminals 3 weeks after 7.5 mg/kg of 4’-amino-MPTP would exceed 150 pM.

Clearly, the persistent striatal pyridinium metabolites are not bound to neuromelanin, because that biopoly- mer is confined to the soma. This leaves two obvious candidates for the subcellular localization of the pyri- dinium metabolites: mitochondria and synaptic vesi- cles. Mitochondria have been suggested as the ultimate site of action of MPP’, because MPP’ blocks mito- chondrial NADH oxidation in vitro (Nicklas et al., 1985). In addition, MPP+ is accumulated within mi- tochondria against its concentration gradient (Ramsay and Singer, 1986), probably by a passive Nernstian transport mechanism (Hoppel et al., 1987). Although uptake of MPP+ into brain synaptic vesicles has not been described, Reinhard et al. ( 1989) have character- ized a reserpine-sensitive uptake in chromaffin granules that appears to utilize the catecholamine transporter. Similarly, MPP’ is taken up and stored within Shy- droxytryptamine organelles in platelets (Cesura et al., 1987). It is of interest that the K,,, for uptake into chro- maffin granules for MPP’ is similar to that of cate- cholamines. If this similarity holds true for synaptic vesicles, the K, for uptake of MPP’ would be 4 pM. Of the two possibilities, storage in dopaminergic syn- aptic vesicles seems more likely for several reasons. First, the striatal dopaminergic terminals have an enormous vesicular storage capacity relative to other areas and are well suited to store high concentrations. Second, the high concentrations of pyridinium metab- olites being stored are not exerting a toxic effect and are thus most likely being sequestered in an intracel- lular compartment, such as the vesicular compartment. In addition, the distribution of MPP’ in the primate brain closely parallels the distribution of tetrabenazine- binding sites (Herkenham et al., 199 1). Thus, whereas DA metabolism is affected, TH activity is unaffected, and cell numbers are not reduced in the two lowest dose groups for each compound. Such high concen- trations of MPP+ might be expected to have a more

deleterious effect if localized in mitochondria. This possibility is underscored by the finding of Reinhard et al. (1987) that reserpinized chomaffin cells are more sensitive to MPP+. Perhaps slow leakage of MPP+ from a vesicular compartment might make continual low concentrations of MPP+ available to exert effects on mitochondria1 MA0 activity.

The dose-response data from MPTP and 4‘-amino- MPTP have been combined into a working model of the subacute effects of MPTP-like compounds on striatal DA metabolism (Fig. 6). The model has five salient features. First, there is a dramatic loss of the DA metabolites DOPAC and HVA at low doses. The maximal decrease in the concentration of these me- tabolites occurs at a dose well below the one that causes DA loss or cell death. Second, the DA concentration tends to rise at the lower doses. Third, extensive cell death, which is typical in the MPTP model, does not

e 0 .- L

E L c a 0 c

8 % v) v)

i=

I 4 D o p a m i n e b TH Activitv

Death 0 MPTP Dose ___)

FIG. 6. Working model of the subacute biochemical effects of MPTP-like toxins. Significant concentrations of the pyridinium me- tabolites can persist within dopaminergic terminals for weeks and perhaps months after a low dose of an MPTP-like toxin. This re- sidual pyridinium metabolite exerts an effect on DA metabolism, possibly by blocking the actions of MA0 contained within dopa- minergic terminals. At low increasing doses, the striatal pyridinium concentrations increase in a dose-related fashion, and this increase is paralleled by a decrease in DOPAC and HVA levels and an in- crease in DA content. At intermediate doses the striatal pyridinium concentrations reach their maximum, as do the ratios of DA/ DOPAC. Although the metabolite concentrations continue to decline slightly beyond this dose, so do MPP’ concentrations and levels of markers for dopaminergic terminals, such as DA and TH activity. This suggests that a saturable intraneuronal compartment exists that protects neuronal mitochondria from MPP+-induced damage. As this ”safe” compartment becomes saturated, terminals begin to die, leading to the parallel loss of DA, TH, and MPP’. Saturations of this “safe” compartment may have occurred at the 1 .O mg/kg (MPTP) and 15 mg/kg (4’-amino-MPTP) groups. In each case the mean striatal concentrations of the pyridiniums did not increase over that of the next lowest dose. Furthermore, the striatal DA content in these animals was either normal or substantially depleted, a finding suggesting the saturation point had been exceeded and toxicity manifested. The work of Reinhard (1989) suggests that synaptic vesicles may constitute the “safe” MPP’ Compartment. In the upper dose ranges, there is a significant loss of striatal DA and TH activity that precedes significant nigral cell loss. At the standard toxic dose, all striatal markers for dopaminergic terminals disappear, as do the majority of cells within the substantia nigra pars compacta.

J. Neurochem.. Vol. 57, No. 3, 1991

MPTP AND 4’-AMINO-MPTP DOSE-RESPONSE STUDY 989

occur until >70-80% of the striatal DA has been de- pleted. Fourth, the metabolic changes are paralleled by, and therefore possibly linked to, striatal concen- trations of a pyridinium metabolite. Fifth, the phe- nomenon is relatively long lived (>6 weeks) but re- versible over a 4-6-month period (J. N. Johannessen et al., unpublished data). The model has implications, detailed below, for understanding striatal DA metab- olism, for the use of the MPTP model in transplant studies, and for the hypothesis of toxin-induced par- kinsonism.

First, the fact that low doses of MPTP and 4’-amino- MPTP cause such drastic decreases in DOPAC and HVA levels, coupled with the dialysis data suggesting that this effect is dependent on uptake of the pyridin- ium metabolites into DA terminals (Caliguri and Jo- hannessen, 199 1 ), indicates that deamination of stnatal DA in vivo is overwhelmingly a presynaptic event. If inhibition of presynaptic MA0 is responsible for the metabolic changes, then presynaptic MA0 plays the predominant role in the oxidative deamination of DA, despite the fact that it constitutes only a small fraction of the total striatal MA0 activity. These results are in agreement with dialysis data showing that the efflux of DOPAC and HVA is not dependent on reuptake of DA and therefore is primarily a product of unreleased, nonvesicular DA (Zetterstrom et al., 1988).

Second, despite the fact that loss of the nigrostriatal system is associated with a large decrease in the HVA concentration in CSF (Bums et al., 1983), this model implies that CSF HVA concentrations may not always be used as an accurate index of the degree of striatal DA loss in animals treated with MPTP. In the case of 4‘-amino MPTP, there was no significant difference in HVA or DOPAC concentrations between animals with a normal DA content (7.5 mg/kg dose) and those with virtually no DA (22.5 mg/kg group). It should be em- phasized that the complete time frame of this HVA loss has not been assessed, but clearly low HVA levels measured in CSF within 6 weeks of an MPTP injection are not by themselves an unambiguous index of striatal DA loss. Because of the popularity of the MPTP-le- sioned primate for surgical transplant studies, this cau- tion is particularly relevant. The persistence of MPP’ in primate brain necessitates 12- 16 weeks post-MPTP treatment to obtain baseline neurochemical/behavioral data.

The data presented in Fig. 5 and Table 2 demonstrate that substantial striatal DA and TH loss may not always be accompanied by an equivalent loss of nigral cell bodies. This observation suggests that the DA terminals in the striatum are more susceptible to the effects of MPTP than the cell bodies, in keeping with the ret- rograde course of degeneration seen in the dog (Wilson et al., 1987) and monkey (Burns et al., 1983; Herken- ham, et al., 1991) and with the sparing of nigral cell bodies often seen in less susceptible species, such as the mouse. Alternatively, the decreases in striatal DA

and TH activity could reflect a transient decrease in the production of the TH enzyme and not destruction of DA terminals. In either case, in the context of toxin- induced parkinsonism in humans, loss of function through a decrease in striatal DA content would be likely to precede cell loss within the substantia nigra pars compacta. It should be noted that other primate models of MPTP-induced parkinsonism have been de- veloped where nigral cell bodies die rapidly (within - 10 days), roughly paralleling the terminal degener- ation (Irwin et al., 1990).

The severe alteration of striatal DA metabolism and long-term retention of pyridinium metabolites after subtoxic doses of MPTP and 4’-amino-MPTP also have several implications for the chronic neurotoxin expo- sure model of human parkinsonism. The high storage capacity of the striatal DA terminals for MPP’ and 4’- amino-MPP’ suggests that exposure to an MPTP-like toxin over a prolonged period could lead to a gradual filling of this striatal compartment, whatever its nature. If the contents of that compartment were suddenly caused to be released or if the storage capacity was exceeded, then precipitation of a toxic reaction might ensue. Furthermore, at the early stages of exposure, while patients are asymptomatic, one would expect de- creased striatal DOPAC and HVA levels and elevated DA levels. Thus, clinical symptoms might be delayed by the pharmacologic elevation of intraneuronal DA content even though synthetic capacity is decreased. This suggests that if the continued presence of an MPP+-like toxin (one that causes prolonged reductions in levels of dopamine metabolites) underlies the slow death of nigral neurons, substantial decreases in CSF levels of HVA would act as a prelude to the clinical manifestations of parkinsonism.

In conclusion, the subacute changes in DA metab- olism induced by low doses of MPTP-like compounds offer a clue to the neuronal disposition of subtoxic concentrations of MPP+. By studying acute and chronic exposure to subtoxic doses of MPTP and its analogs, we should be able to define the neurochemical changes induced by such treatment and apply them to the question of the possible contribution of naturally oc- curring MPTP-like toxins to human parkinsonism, keeping in mind that a negative finding will not elim- inate the possibility that other dissimilar toxins may contribute to the etiology of parkinsonism.

REFERENCES

Arai Y., Toyoshima Y . , Kinemuchi H., Tadano T., and Kisara K. (1986) The ex vivo effect of l-methyl-4-phenyl-l,2,3,6-tetrahy- dropyridine (MPTP) on rat intra- and extraneuronal monoamine oxidase activity. Neurosci. Lett. 10, 266-27 1.

Barbeau A., Roy M., Bernier G., Campanella G., and Pans S . (1987) Ecogenetics of Parkinson’s disease: prevalence and environ- mental aspects in rural areas. Can. J . Neurol. Sci. 14, 36-41.

Birkmayer W., Knoll J., Riederer P., Youdim M. B., Hars V., and

J. Neurochem.. Vol. 57, No. 3, 1991

990 J. N. JOHANNESSEN ET AL.

Marton J. (1985) Increased life expectancy resulting from ad- dition of L-deprenyl to Madopar treatment in Parkinson’s disease: a longterm study. J . Netirul. Trunsm. 64, 1 13- 127.

Burns R. S., Chiueh C. C., Markey S. P., Ebert M. H., Jacobowitz D. M. and Kopin 1. J. (1983) A primate model ofparkinsonism: selective destruction of dopaminergic neurons in the pars com- pacts of the substantia nigra by N-methyl-4-phenyl- 1,2,3,6-tet- rahydropyridine. Proc. Natl. Acacl Sci. USA 80, 4546-4550.

Caliguri E. J. and Johannessen J. N. (1991) Selective decrease in extracellular DOPAC concentrations in rats striatum following in vivo dialysis with low concentrations of MPP+. Bruin Res. 548,94-99.

Cesura A. M., Ritter A., Picotti G. B.. and Da Prada M. (1987) Uptake. release, and subcellular localization of I -methyl-4- phenylpyridinium in blood platelets. J . Neurochem. 49, 138- 145.

Cbiba K., Trevor A.. and Castagnoli N. J. (1984) Metabolism ofthe neurotoxic tertiary amine, MPTP, by brain monoamine oxidase. Biochem . Biophys Res. C( )mm iin . 1 20, 5 74-5 7 8.

Davis G. C.. Williams A. C., Markey S. P.. Ebert M. H., Caine E. D., Reichert C. M., and Kopin 1. J. (1979) Chronic parkin- sonism secondary to intravenous injection of meperidine ana- logues. P.yjehiatrj, Rcs. l , 249-254.

Herkenham M., Little M. D., Bankiewicz K., Yang S.-C., Markey S . P., and Johannessen J. N. (199 I ) Selective retention of MPP+ within the monoaminergic systems ofthe primate brain following MPTP administration: an in vivo autoradiographic study. Neu- ro.scirvm> 40, 133-1 58.

Hoppel C. L., Greenblatt D., Kwok H.-C., Arora P. K., Singh M. P., and Sayre L. M. (1987) Inhibition of mitochondria1 respiration by analogs of 4-phenylpyridine and 1 -methyl-4-phenylpyridine cation (MPP’), the neurotoxic metabolite of MPTP. Bioc,hem. Eiophys. Rm. Commun. 148, 684-693.

Irwin I., DeLanney L. E., Forno L. S., Finnegan K. T., Di Monte D. A,, and Langston J. W. (1990) The evolution of nigrostriatal neurochemical changes in the MPTP-treated squirrel monkey. Erain Rc.F. 532, 242-252.

Javitch J. A. and Snyder S. H. (1984) Uptake of MPP+ by dopamine neurons explains selectivity of parkinsonism-inducing neuro- toxin, MPTP. Eur. J . Phurmucol. 106, 455-456.

Johannessen J. N., Chiueb C. C., Burns R. S., and Markey S . P. (1985) Differenccs in the metabolism of MPTP in the rodent and primate parallel differences in sensitivity to its neurotoxic effects. I& Sci 36, 2 19-224.

Johannessen J. N., Savitt J. M., Markey C. J., Bacon J. P., Weisz A,, Hanselman D. S., and Markey S. P. (1987) The development of amine substituted analogues of MPTP as unique tools for the study of MPTP toxicity and Parkinson’s disease. Lifi Sci. 40, 697-704.

Johannessen J. N., Chiueh C. C., Bacon J. P., Garrick N. A,, Burns R. S., Weise V. K., Kopin I. J., Parisi J. E., and Markey S . P. ( 1989) Effects of I -methyl-4-phenyl- 1,2,3,6-tetrahydropyridine in the dog: effect of pargyline pretreatment. J . h’ezirochetn. 53,

Kinemuchi H., Arai Y., and Toyoshima Y. (1985) Participation of brain monoamine oxidase B form in the neurotoxicity of I - methyl-4-phenyl-1,2,3.6-tetrahydropyridine: relationship be- tween the enzyme inhibition and the neurotoxicity. Neuro.~ci L m 58, 195-200.

Langston J. W., Ballard P., Tetrud J. W., and Irwin 1. (1983) Chronic parkinsonism in humans due to a product of meperidine-analog synthesis. Sciencc 219, 979-980.

Lerner P., Nose P.. Ames M. W., and Lovenberg W. (1978) Modi- fication of the tyrosine hydroxylase assay: increased enzyme ac- tivity in the presence of ascorbic acid. Ncitrochem. Res. 3, 64 I - 651.

Lloyd K. G. and Hornykiewicz 0. (1972) Occurence and distribution of aromatic L-amino acid (L-DOPA) decarboxylase in the human brain. J. Neurochem. 19, 1549- 1559.

Lowry 0. H., Rosebrough N. J., Farr A. L., and Randall R. J. (195 1)

582-589.

Protein measurement with the Fohn phenol reagent. J. Biol. Chem. 193, 265-275.

Makino Y., Ohta S., Tachikawa O., and Hirobe M. (1988) Presence of tetrahydroisoquinoline and 1 -methyl-tetrahydro-isoquinoline in foods: compounds related to Parkinson’s disease. Lijh Sci.

Naoi M., Suzuki H., Kiuchi K., Takahashi T., and Nagatsu T. ( 1 987) Effect of N-methyl-4-phenylpyridinium ion on monoamine ox- idase in a clonal rat pheochromocytoma cell line, PCI2h. J. Neurochem. 48, 19 12- I9 16.

Nicklas W. J., Vyas I., and Heikkila R. E. (1 985) Inhibition of NADH- linked oxidation in brain mitochondria by I-methyl-4-phenyl- pyridine, a metabolite of the neurotoxin, I-methyl-4-phenyl- 1,2,5,6-tetrahydropyridine. Life Sci. 36, 2503-2508.

Niwa T., Takeda N., Kaneda N., Hashizume Y., and Nagatsu T. (1987) Presence of tetrahydroisoquinoline and 2-methyl-tetra- hydroquinoline in parkinsonian and normal human brains. Biochetn. Biophys. Rev. Commun. 144, 1084- 1089.

Ramsay R. R. and Singer T. P. (1986) Energy-dependent uptake of N-methyl-4-phenylpyridinium, the neurotoxic metabolite of I - methyl-4-phenyl- I ,2,3,6-tetrahydropyridine, by mitochondria. J . Biol. Chem. 261, 7585-7587.

Reinhard J. J., Diliberto E. J., Viveros 0. H., and Daniels A. J. ( 1 987) Subcellular compartmentalization of I-methyl-4-phen- ylpyridinium with catecholamines in adrenal medullary chro- maffin vesicles may explain the lack of toxicity to adrenal chro- maffin cells. Proc. Natl. Acad. Sci. US,4 84, 8 160-8164,

Reinhard J. J., Diliberto E. J., and Daniels A. J. (1989) Character- ization of cellular transport, subcellular distribution, and secrc- tion of the neurotoxicant I-methyl-4-phenylpyridinium in bovine adrenomedullary cell cultures. J . Neurochem. 52, 1253-1259.

Shih M.-C. and Markey S. P. (1986) Quantification of I-methyl-4- phenyl- I ,2,3,6-tetrahydropyndine and I-methyl-4-phenylpyri- dinium ion in brain tissue by gas chromatography/mass spec- trometry. Biomed. Environ. Muss Spectrom. 13, 85-89.

Takamidoh H., Naoi M., and Nagatsu T. (1987) Inhibition of type

43,373-378.

A monoamine oxidase by I-methyl-4-phenylpyridine. Neurosci. L.ett. 73. 293-297.

Tetrud J. W. and Langston J. W. (1989) The effect of deprenyl (Se- legiline) on the natural history of Parkinson’s disease. Science 245,s 19-522.

Waymire J. C., Bjur R., and Weiner N. (1971) Assay of tyrosine hydroxylase by coupled decarboxylation of DOPA formed from I -I4C-~-tyrosine. Anal. Biochem. 43, 588-600.

Wilkening D., Vernier V. G., Arthaud L. E., Treacy G., Kenney J. P., Nickolson V. J., Clark R., Smith D. H., Smith C., and Boswell G. ( 1 986) A Parkinson-like neurologic deficit in primates is caused by a novel 4-substituted piperidine. Bruin Re.s. 368,

Wilson J. S., Turner B. H., Morrow G. D., and Hartman P. J. (1987) MPTP produces a mosaic-like pattern of terminal degeneration in the caudate nucleus of the dog. Bruin Res. 423, 329-332.

Youngster S. K., Duvoisin R. C., Hess A., Sonsalla P. K., Kindt M. V., and Heikkila R. E. (1986) I-Methyl-4-(2’-methylphenyl)- 1,2,3,6-tetrahydropyridine (2’-CH,-MPTP) is a more potent do- paminergic neurotoxin than MPTP in mice. Eur. J . Pharmacol. 122,283-287.

Youngster S. K., Sonsalla P. K., and Heikkila R. E. (1987) Evaluation of the biological activity of several analogs of the dopaminergic neurotoxin I -methyl-4-phenyl- 1,2,3,6-tetrahydropyridine. J. Nmrochem. 48, 929-934.

Youngster S. K., McKeown K. A,, Jin Y. Z., Ramsay R. R., Heikkila R. E., and Singer T. P. (1989) Oxidation of analogs of I-methyl- 4-phenyl- 1,2,3,6-tetrahydropyridine by monoamine oxidases A and Rand the inhibition ofmonoamine oxidases by the oxidation products. J . Nmrochem. 53, 1837- 1842.

Zetterstrom 7.. Sharp T., Collin A. K., and Ungerstedt U. ( 1 988) In vivo measurement of extracellular dopamine and DOPAC in rat striatum after various dopamine-releasing drugs; implications for the origin of extracellular DOPAC. E w . J . Phurmucol. 148, 327-334.

239-246.

J . Ncwochem.. Vol, 57. No. 3, 1991