Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Since it was first identified as a cause of parkinsonism in humans (Langston et al., 1983) and nonhuman primates (Burns et al., 1983; Langston et al., 1984a), much has been learned about the biological cascade of events that lead to neurotoxicity caused by MPTP. After systemic administration, MPTP readily crosses the blood-brain barrier, where it is converted to MPP⁺ by monoamine oxidase B (Chiba et al., 1984; Langston et al., 1984b). This process appears to occur in astrocytes (Di Monte et al., 1991). MPP⁺ is then selectively taken up by the catecholamine transporter into dopamine neurons (Chiba et al., 1985; Javitch et al., 1985). The cytotoxic effects of MPP⁺ are probably due to its accumulation in mitochondria (Ramsay and Singer, 1986) and inhibition of Complex I of the respiratory chain (Nicklas et al., 1985; Ramsay et al., 1991), thereby impairing ATP production (Di Monte et al., 1986; Chan et al., 1991). These biochemical changes may ultimately lead to dopamine depletion and neuronal death.

In rodents, these biochemical events occur within a few hours after MPTP exposure. Both MPTP and MPP⁺ are rapidly cleared from the brain (within 12 hr), and nigrostriatal ATP depletion is no longer evident 24 hr after administration (Irwin and Langston, 1985; Chan et al., 1991; 1993a). However, the process of actual neuronal degeneration appears to evolve for a much longer period of time (Ricaurte et al., 1986; Janson et al., 1988a; Chadi et al., 1993; Jackson-Lewis et al., 1995). These observations raise the possibility that MPTP/MPP⁺ trigger other events that are ultimately responsible for neurotoxicity.

There is good reason to believe that excitotoxicity represents one of these events, because an impairment of energy metabolism is thought to lead to the activation of NMDA receptors (Henneberry et al., 1989). However, the results of studies wherein NMDA antagonists were used to protect against MPTP/MPP⁺ neurotoxicity have been conflicting; some investigators have reported success (Turski et al., 1991; Zuddas et al., 1992; Storey et al., 1992; Brouillet and Beal, 1993), and others have failed to demonstrate protection (Sonsalla et al., 1989; 1992; Kupsch et al., 1992; Chan et al., 1993b). In a recent study, we found that (+)MK-801 delays, but does not prevent, MPTP-induced dopamine depletion in the mouse striatum (Chan et al., 1993b). The present studies were initiated in order to reconcile this observation with other reports that (+)MK-801 effectively protects nigral neurons against the effects of MPTP (Turski et al., 1991; Zuddas et al., 1992; Srivastava et al., 1993). More specifically, using an unbiased stereological method (Janson and Møller, 1993), we tested the hypothesis that despite its only temporary effect in the striatum, (+)MK-801 might protects MPTP-induced dopamine depletion and neuronal loss in the mouse substantia nigra. As part of these studies, we also characterized the action of (+)MK-801 on the pharmacokinetics of MPP⁺ in the VME, because (+)MK-801 has previously been shown to alter the elimination of MPP⁺ from the striatum (Chan et al., 1993b).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

All experiments were performed in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by our Animal Care and Use Committee.

Animals and treatments. These experiments were carried out using a previously reported lesioning regimen (Chan et al., 1993b). Briefly, male C57BL/6 mice 6 to 7 weeks old (Simonsen Laboratory, San Jose, CA) were divided into four groups: 1) saline, 2) (+)MK-801 alone, 3) MPTP alone and 4) (+)MK-801 and MPTP. Animals were given a single s.c. injection of 40 mg/kg MPTP (hydrochloride salt; Research Biochemicals, Natick, MA). This regimen has been shown typically to induce a reproducible 80% striatal dopamine depletion in C57BL/6 mice (Langston et al., 1990). An injection of (+)MK-801 (Research Biochemicals, Natick, MA) (1 mg/kg) was administered i.p. 30 min before MPTP exposure and followed by additional injections at 4 and 8 hr after the first injection. Four or more injections of (+)MK-801 caused death in more than 40% of the animals. Control animals received equal volumes of saline instead of MPTP and/or (+)MK-801.

Immunohistochemistry. After perfusion, the brain was rapidly removed and transected coronally through the median eminence. The caudal block was postfixed for 90 min in the fixative described above and cryoprotected in sucrose (10% for 24 hr followed by 30% for 1-2 weeks). The midbrain was cut into coronal sections 40 µm thick on a cryostat (Microm, HM 500 M, Walldorf, Germany). The sections were sampled systematically throughout the entire substantia nigra with a random start. Every third section in rostro-caudal order was processed for free-floating immunohistochemistry (Janson et al., 1988b, Janson and Møller, 1993).

View larger version (K): [in this window] [in a new window]   Fig. 1.   Microphotographs illustrating the procedure used for quantitative immunohistochemistry. A) Mouse substantia nigra showing brownish TH immunoreactive cell bodies and neuropil with a cresyl violet counterstain of the Nissl substance and nucleoli. The SNc is encircled at low magnification in each of the unbiasedly sampled coronal sections in rostro-caudal order. B and C) The stereological analysis is performed as described in "Materials and Methods" using a 100× objective with a high numerical aperture. The motorized X-Y stage systematically stops at regular intervals with a random start inside the encircled SNc. Neurons are sampled if their nucleoli (arrow) lie within the sampling frame without touching its red lines. Optically dissecting the section in 0.5-µm thin slices makes the sampling area a sampling volume. In panel B, the vertical distance z from the top of the section is 5.0 µm; in panel C, z = 9.0 µm. Neurons are sampled only if their nucleoli lie within an 8-µm-thick slice of the section excluding the top 3.5 µm closest to the coverslip. The estimated total number of TH immunoreactive neurons is independent of any dimensional changes in the tissue. Abbreviations: SNr = substantia nigra pars reticulata; SNlat = substantia nigra pars lateralis; ml = medial lemniscus. The bars represents 100 µm in panel A (blue) and 5 µm in panels B and C (yellow).

Stereological analysis. The stereological analysis (Gundersen et al., 1988) was performed using an Olympus BH2 microscope with a motorized specimen X-Y stage linked to a computer-assisted stereological system consisting of a color video camera (CCD-Iris, Sony), a PC with a high-resolution SVGA monitor and GRID software (Olympus, Albertslund, Denmark) and a microcator (VRZ 401, Heidenhain, Traunreut, Germany) for monitoring movements in the Z-axis with a vertical resolution of 0.5 µm.

Measurements of dopamine and MPP⁺. After dissection, tissues were immediately immersed in 1 ml of 0.4 N perchloric acid and sonicated (Soniprep 150; MSE Scientific Instruments, UK). Samples were centrifuged and dopamine and MPP⁺ were measured in the supernatants. The remaining pellets were dissolved in 0.5 N NaOH overnight for the determination of protein. Dopamine and MPP⁺ were analyzed as previously described (Finnegan et al., 1988; Di Monte et al., 1991). Proteins were assayed by the methods of Lowry et al., (1951).

Statistics. Statistical analysis was performed using the Newman-Keuls (for biochemical data) and Schiff's (for cell counts) post-hoc tests after a two-way analysis of variance (ANOVA).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of (+)MK-801 on MPTP-induced dopamine depletion in the striatum and VME. In the striatum, dopamine levels fell to 35% of control values at 8 hr and recovered to 52% of control values at 24 hr after MPTP administration (fig. 2). Dopamine concentrations fell again at 1 and 3 weeks to 20% to 25% of control values. A biphasic pattern of dopamine depletion was even more evident in the VME, an area that includes the substantia nigra (fig. 3). Eight hours after MPTP exposure, dopamine was decreased by approximately 60% in the VME. By 24 hr, however, nigral dopamine levels returned to values only slightly lower than controls. A second phase of dopamine loss occurred between 24 hr and 1 week. Dopamine values were approximately 75% of controls at this later time-point and did not significantly change at 3 weeks.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effects of (+)MK-801 on MPTP-induced striatal dopamine depletion. Animals (n = 10/group) were given an i.p. injection of either saline or (+)MK-801 (1 mg/kg) 30 min before a s.c. injection of MPTP (40 mg/kg) or equal volumes of saline. One (8-hr time-point) or two additional doses of (+)MK-801 were given at 4-hr intervals after the first injection. Animals were killed at 8 and 24 hr and 1 and 3 weeks after MPTP exposure, and dopamine levels were measured. Data (mean ± S.E.M.) for MPTP alone (open circle) and (+)MK-801/MPTP (open diamond) are expressed as a percentage of the respective control values at the corresponding time-points. Values for control animals not receiving MPTP did not differ statistically at any time-point, and their mean was 129.7 ± 3.42 ng/mg protein. * Statistically different (P < .0001) from the corresponding value for (+)MK-801/MPTP animals.  Statistically different (P < .01) from the values at 8 hr and 1 and 3 weeks for MPTP-alone animals.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effects of (+)MK-801 on MPTP-induced dopamine depletion in the VME. Animals (n = 10/group) were treated as described in figure 2. Data (mean ± S.E.M.) for MPTP-alone (open circle) and (+)MK-801/MPTP (open diamond) are expressed as a percentage of the respective control values at the corresponding time-points. Values for control animals not receiving MPTP did not differ statistically at any time-point, and their mean was 7.16 ± 0.76 ng/mg protein. * Statistically different (P < .01) from the corresponding value for (+)MK-801/MPTP animals.  Statistically different (P < .05) from the values at 8 hr and 1 and 3 weeks for MPTP-alone animals.

Effect of (+)MK-801 on MPTP-induced cell loss in the substantia nigra. The total number of TH-immunoreactive/Nissl-stained positive neurons was estimated in the entire SNc (fig. 4, A and B). The CE for each estimate was between 0.05 and 0.10. Control animals had 11,800 ± 1,400 (mean ± S.D.) nigral cells bilaterally. This number is similar to other stereological estimates reported in normal mice (Chadi et al., 1993).

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Effects of (+)MK-801 on MPTP-induced loss of TH immunoreactive neurons counterstained with cresyl violet in the entire SNc on both sides. Treatment to the animals (n = 5-7/group) was the same as described in the legend to figure 2. Animals were killed at 1 week (panel 1A) and 3 weeks (panel B) after MPTP exposure. The unbiased stereological estimates of the total number of TH-immunoreactive and Nissl-stained neurons were performed as described in "Materials and Methods." Data are the mean ± S.E.M. of the total number of TH-immunoreactive and Nissl-stained neurons in the entire SNc. * Statistically different (P < .01) from the saline control and (+)MK-801 groups, respectively.

Effect of (+)MK-801 on MPP⁺ biodisposition in the VME. MPP⁺ levels were measured at multiple time-points (1.5, 2, 4, 8, 12, 16 and 24 hr) as described in our previous study (Chan et al., 1993b). In saline/MPTP-treated animals, MPP⁺ concentrations in the VME reached peak values at the 1.5-hr time-point (table 1). They then declined rapidly by 8 hr and were not detected at 12 hr. In the (+)MK-801/MPTP group, MPP⁺ concentration reached the peak level at 1.5 hr, a result that was similar to the saline/MPTP group. At 8 hr, MPP⁺ levels had declined less in the (+)MK-801/MPTP group than in animals receiving MPTP alone, but as in the saline/MPTP group, clearance was completed by 12 hr.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The role of excitotoxicity in the cascade of events leading to MPTP neurotoxicity has been surprisingly difficult to elucidate, particularly in view of the often-conflicting reports on the protective effects of NMDA antagonists. These different results may be related to wide variations in experimental design between studies, including differences in species, routes of injection, regimens of MPTP and (+)MK-801, methods for assessing toxicity and even the anatomic locations examined (i.e., the striatum vs. the substantia nigra). For example, at least two studies in rats (Turski et al., 1991; Srivastava et al., 1993) have shown that NMDA antagonists prevent neuronal loss caused by MPP⁺ when they were injected directly into the nigrostriatal system. Similar results have been reported in primates, but in these studies MPTP was administered systemically (Zuddas et al., 1992; Lange et al., 1993). Studies in mice have proved to be an entirely different matter. Most investigators, including our own group, have found that after systemic administration of MPTP, NMDA antagonists offer little or no protection against MPTP-induced striatal dopamine depletion and/or disappearance of TH immunoreactivity (Sonsalla et al., 1989; Kupsch et al., 1992; Chan et al., 1993b; Vaglini et al., 1994).

In the present study, we attempted to bridge these apparently divergent results by determining whether (+)MK-801 could protect dopaminergic neurons in the VME of mice as suggested by the primate and rat studies, even if it failed to prevent dopamine depletion of nigrostriatal nerve terminals. Our results clearly indicate that this is not the case. Three weeks after systemic administration of MPTP, there was no evidence that (+)MK-801 had prevented dopamine depletion in the striatum and VME of mice. Even more important, the number of dopaminergic neurons in the substantia nigra were the same regardless of whether the mice were treated with MPTP alone or in combination with (+)MK-801. To the best of our knowledge, only one other report has examined the effects of NMDA antagonists on nigral cell counts in mice after systemic administration of MPTP (Kupsch et al., 1992). In that study, a slight but not statistically significant increase in cell number was observed after combined treatment with MPTP and NMDA antagonists compared with MPTP alone. However, neurons were counted in only one to three sections from a single rostro-caudal level of the substantia nigra. Our study used an unbiased quantitative cell-counting technique to assess the entire nigral neuronal population. Neither rostro-caudal differences in nigral cell counting nor dimensional changes in nigral tissue and/or neurons within this tissue would have affected the results. It is also important to note that our results are likely to represent actual cell loss, rather than just a decline in TH immunoreactivity, because neurons were counted on the basis of both TH and Nissl. We think that these results leave little doubt that (+)MK-801, when given in maximally tolerated doses, does not protect the dopaminergic neurons in the mouse substantia nigra from the toxic effects of systemically administered MPTP.

The lack of protection by (+)MK-801 is unlikely to result from a lower dose and shorter dosing regimen, because similar results were obtained by Kupsch and colleagues (1992) with a higher dose (4 mg/kg) and a longer period of treatment (12 doses). Furthermore, Kupsch et al. (1992) have shown that when the competitive NMDA antagonist CGP 40116 was administered to mice, there was still no protection against MPTP-induced striatal dopamine depletion. In addition, it might be argued that (+)MK-801 may provide protection only when the MPTP-induced damage is less severe. However, the fact that inconsistent results of (+)MK-801 against MPTP toxicity were observed in studies with the same regimen of MPTP treatment (Brouillet and Beal, 1993; Sonsalla et al., 1989; 1992) argues against this interpretation.

The reasons why studies in rats and monkeys have shown protection, whereas those in mice have not, are unknown. One major difference between the investigations in mice and those in rats is that intracerebral injection of MPP⁺ was necessary in rats because they are not susceptible to systemically administered MPTP. The effects of direct exposure of nigral neurons to MPP⁺ could be quite different from the process that occurs after systemic administration of MPTP. For example, direct injection of MPP⁺ may cause nonspecific damage that could be at least partially prevented by NMDA antagonists. It is also possible that direct injection of (+)MK-801 into the rat brain, an approach used by Turski et al. (1991), has different effects than systemic administration. Clarke and Reuben (1995) have reported that high concentrations of (+)MK-801 (which are probably similar to those that can be achieved by injecting it directly into the brain) block the dopamine transporter in synaptosomal preparations. This would be likely to prevent MPP⁺ uptake into neurons, so (+)MK-801 could be providing transient protection via this mechanism. However, this would not explain the protection afforded by repeated systemic administration of (+)MK-801 against MPP⁺-induced loss of TH-positive neurons Turski et al. (1991) also reported. On the other hand, Sonsalla and colleagues (1992) did not achieve protection with systemically administered (+)MK-801 in a similar set of experiments.

Primate studies showing protection of nigral neurons after systemic administration of both MPTP and NMDA antagonists might at first glance appear to settle this issue of whether systemic administration of (+)MK-801 is protective, but close examination of these investigations raises several questions regarding how much protection the NMDA antagonists actually afforded. The study by Lange et al. (1993), using the competitive NMDA antagonist 3-[(±)-2-carboxypiperazin-4-yl]-propyl-1-phosphonic acid (CPP), examined animals at only one time-point, 2 days after exposure to MPTP. If NMDA antagonists provided only transient protection (as they appear to do in mice), lack of enduring protection would have been missed. Zuddas et al. (1992) used (+)MK-801 at a later time-point (7 days after the last injection), but cell counts were not done and TH density was only subjectively estimated.

It should be pointed out that none of the previous work showing protection of nigral neurons from the neurotoxic effect of MPTP by (+)MK-801 utilized the unbiased stereological cell-counting techniques employed in the present study. It would certainly be of interest to repeat the experiments in monkeys at later time-points using unbiased quantitative methods to determine to what degree (if any) counting techniques and/or time course influence the outcome of the studies in primates. If such studies do confirm protection by (+)MK-801 in primates, it will still be necessary to be certain that this protection is not due to the effects of (+)MK-801 on the dopamine transporter or to some other action of (+)MK-801, including its apparent effects on the biodistribution of MPP⁺ (see the following discussion).

Several other interesting observations emerged from this study. Somewhat surprisingly, the pattern of MPTP-induced dopamine depletion in the striatum and VME showed a biphasic pattern. This pattern was particularly evident in the VME, where, after an initial loss of dopamine (which fell to 30-40% of control values), dopamine concentrations returned to near normal at 1 day and then declined again at 1 and 3 weeks after MPTP administration. This biphasic pattern of dopamine depletion may reflect two different mechanisms, the initial loss representing a pharmacological/biochemical effect due either to an MPP⁺-induced release of dopamine (in the striatum) or to inhibition of TH (in the substantia nigra) (Pileblad et al., 1984; 1985; Hirata and Nagatsu, 1986; Nagatsu, 1990). This interpretation is supported by the fact that dopaminergic terminals remain intact at this stage (Sundström et al., 1988). Because the subsequent depletion or the second decrease occurs at a time when nigral neurons are thought to be degenerating, it seems likely that this phase represents actual neurotoxicity. Interestingly, the transition between the two phases occurs at a time when MPP⁺ has already been eliminated from the striatum and substantia nigra. These observations suggest that MPP⁺ triggers a cascade of events that continues after its elimination and may ultimately result in cell death. It should also be pointed out that the extent of damage caused by MPTP/MPP⁺ is substantially greater in the striatum than in the substantia nigra (at 1 and 3 weeks, 80% of dopamine was depleted in the striatum compared with only 30% in the substantia nigra). These results indicate that in mice, nigrostriatal nerve terminals are probably much more sensitive to the effects of MPTP than are their cell bodies.

These studies confirm and extend our earlier observations by showing that (+)MK-801 transiently prevents MPTP-induced dopamine depletion in the striatum and that the same phenomenon occurs in the VME. However, the time course differed between these two regions; the prevention of MPTP-induced dopamine depletion was observed in the striatum at 8 hr, whereas in the VME this was seen only at 1 week. In reporting our earlier results, we hypothesized that (+)MK-801 was blocking the removal of MPP⁺ by preventing vesicle release of dopamine, thereby delaying both MPP⁺ elimination from the striatum and striatal dopamine depletion (Chan et al., 1993b). This speculation was supported by the following observations. First, MPP⁺ can be sequestrated in dopamine vesicles after entering the neurons (Reinhard et al., 1987; 1988). Second, it has been shown that (+)MK-801 counters MPTP- and MPP⁺-induced dopamine release (Schmidt et al., 1984; Pileblad et al., 1984; 1985). Furthermore, there was a good temporary correlation between MPP⁺ retention in the striatum and the protection afforded by (+)MK-801 (Chan et al., 1993b). In contrast to the striatum, no such correlation between MPP⁺ retention and (+)MK-801 protection against dopamine depletion was seen in the VME. MPP⁺ elimination was delayed by only several hours in (+)MK-801-treated animals and was completely eliminated by 12 hr, whereas the effect of (+)MK-801 on MPTP-induced dopamine depletion was observed only at 1 week. Thus it appears that the delayed elimination of MPP⁺ is more likely to play a role in the effects of (+)MK-801 on MPTP-induced dopamine depletion in the striatum than in the substantia nigra.

In summary, although the data reported here only partially clarify the conflicting evidence on the role of excitotoxicity in the cascade of events that lead to MPTP-induced degeneration of dopaminergic neurons, we believe that they do resolve the issue of whether (+)MK-801 protects the mouse substantia nigra from the toxic effects of systemically administered MPTP. Additional experiments will be required to determine whether differing results in rats and primates are due to technical issues such as the methods of assessing nigral cell integrity, species differences and/or novel non-NMDA receptor effects of (+)MK-801. We believe such studies are warranted because they may offer insight into the process of nigral cell degeneration that underlies Parkinson's disease, a major human neurodegenerative disease of aging.