Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

METH is a synthetic drug that is closely related chemically to amphetamine but has a higher potential for abuse as a psychostimulant. Amphetamines cause a massive release of newly synthesized dopamine from presynaptic vesicles and an increase in extracellular dopamine concentration in the nigrostriatal system. High doses of amphetamines cause long-lasting neurotoxicity associated with a marked decrease in dopamine level (Kogan et al., 1976; Fuller and Hemrick-Luecke, 1980; Gibb et al., 1990; Robinson et al., 1990), dopamine transporter binding sites (Wagner et al., 1980) and tyrosine hydroxylase activity (Kogan et al., 1976; Gibb et al., 1990) in the striatum. Recent studies demonstrated a similar depletion in striatal dopamine nerve terminal markers in post-mortem, chronic METH users (Wilson et al., 1996). The damage to nigrostriatal dopaminergic neurons caused by METH and that caused by the neurotoxin MPTP represent experimental animal models of Parkinson's disease. The involvement of EAA transmission in METH-induced neurotoxicity is suggested by two major findings. First, blockade of the NMDA type of glutamate receptors by dizocilpine (MK-801) attenuates METH-induced neurotoxicity (Sonsalla et al., 1989; Farfel et al., 1992; Weihmuller et al., 1992; Ali et al., 1994). Second, repeated administration of METH causes an increase in glutamate release in the striatum (Nash and Yamamoto, 1992). The fact that EAA receptor antagonists modulate the development of amphetamine-induced behavioral sensitization further supports the role of glutamatergic neurotransmission in the effects of these psychostimulants (Karler et al., 1990).

NMDA receptor activation is thought to be associated with stimulation of the nNOS isoform. Interaction of glutamate, for instance, with the NMDA receptor complex opens channels that admit calcium into the cell; binding of calcium to calmodulin activates nNOS, which produces NO, and subsequently elicits the accumulation of cGMP (Garthwaite, 1991; Snyder, 1992). NO-mediated neurotoxicity may arise from the formation of free radicals (peroxynitrite and breakdown products) (Beckman et al., 1990) and activation of the nuclear enzyme poly (ADP-ribose) synthase that leads to energy depletion and cell death (Zhang et al., 1994). The relationship between NMDA receptor activation and the stimulation of NO production prompted us to investigate the role of nNOS first in METH-induced dopaminergic neurotoxicity and second in the behavioral sensitization caused by repeated administration of relatively low doses of METH. We reported that blockade of brain NOS by 7-NI prevented METH-induced dopaminergic neurotoxicity (Itzhak and Ali, 1996) and the development of locomotor sensitization (Itzhak, 1997) in Swiss Webster mice. The purpose of the present study was two-fold: to investigate the susceptibility to METH-induced dopaminergic neurotoxicity of mutant mice that lack the nNOS gene and to determine what the behavioral consequence of exposure to a potentially neurotoxic dose of METH is. We report that mice lacking the nNOS gene are protected against METH-induced dopaminergic neurotoxicity. In addition, despite the fact that wild-type animals displayed obvious dopaminergic deficit after exposure to a high dose of METH, they still developed a marked sensitized response to a challenge METH injection; however, nNOS knockout mice did not show such sensitized response to METH.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. (d)Methamphetamine-HCl, desipramine-HCl and benztropine mesylate were purchased from Sigma (St. Louis, MO). [³H]Mazindol (24.0 Ci/mmole) was purchased from New England Nuclear (Wilmington, DE). METH solution was prepared in saline (0.9% NaCl).

Animals. Homozygote neuronal NOS knockout mice [nNOS(/)] were obtained from a breeding colony established at the University of Miami School of Medicine using homozygote animals that were previously generated by homologous recombination (Huang et al., 1993) at Massachusetts General Hospital, Charlestown, MA. Because the nNOS gene mutation was made on a SV129 agouti mouse and C57BL/6 mouse background, three different animal strains (purchased from Jackson Laboratories) were used as wild-type controls: C57BL/6, SV129 and B6J/SV129 F2 strains; the latter represents F2 progeny of cross-breeding between C57BL/6 and SV129 strains. F1 progeny of heterozygote nNOS knockout mice [nNOS(+/)] were generated from the cross-breeding of wild-type C57BL/6 males and nNOS(/) females. For the present study, only male mice 10 to 12 weeks of age (23-28 g) were used. All animals were maintained on a 12-h light/dark lighting schedule, kept at a room temperature of 21°C and housed in groups of 4 to 5 with free access to food and water. The principles of laboratory animal care (NIH publication No. 85-23, revised 1985) were followed.

Administration of METH for the determination of dopaminergic neurotoxicity. Each of the five groups of mice, nNOS(/), nNOS(+/), B6/SV129, C57BL/6 and SV129, were divided into two subgroups each containing 10 to 15 mice. One group received METH (5 mg/kg) i.p. three times, every 3 h, and the second group received three saline injections. We have previously found that this schedule of METH administration to Swiss Webster mice produced 68, 44 and 55% decreases in the concentration of striatal dopamine, DOPAC and HVA, respectively, and 48% decrease in the number of DAT binding sites (Itzhak and Ali, 1996). Animals were sacrificed 72 h after the last METH injection. The brain was rapidly removed, and the caudate putamen was dissected and stored at 80°C. Striatal tissue from one hemisphere was used for monoamine assays, and the tissue from the other hemisphere was used for [³H]mazindol binding assays.

Determination of neurotransmitter concentrations. Concentrations of dopamine and its metabolites DOPAC and HVA were quantitated by a modified HPLC method combined with electrochemical detection as described by Ali et al. (1994). Each striatum was weighed in a measured volume (20% w/v) of 0.2 N perchloric acid containing 100 ng/ml of the internal standard 3,4-dihydroxybenzylamine. The tissue was next disrupted by ultrasonication and centrifuged at 4°C (15,000 × g; 7 min). Then 150 µl of the supernatant was removed and filtered through a 0.2-µm Nylon-66 microfilter (MF-1 centrifugal filter, Bioanalytical System, W. Lafayette, IN). Aliquots of 25 µl representing 2.5 mg of brain tissue were injected directly onto the HPLC/EC system for separation of the analytes. The concentrations of dopamine, DOPAC and HVA were calculated using standard curves that were generated by determining, in triplicate, the ratio between three different known amounts of the amine or its metabolites and a constant amount of internal standard.

Binding of [³H]mazindol to the dopamine transporter. Striatal tissue from one hemisphere of 3 to 4 mice was pooled together, homogenized in 20 volumes of Tris-HCl buffer (50 mM; pH 7.7) containing 300 mM NaCl and 5 mM KCl and centrifuged (40,000 × g; 15 min; 4°C). The pellet was resuspended in 6 volumes of sucrose (0.32 M), and aliquots were stored at 80°C. For binding assays, the tissue was resuspended in 15 volumes of the buffer to yield a protein concentration of 0.3 to 0.4 mg/ml. Saturation binding experiments were carried out in a final volume of 0.5 ml containing various concentrations of [³H]mazindol (1.0, 2.0, 4.0, 6.0, 8.0 and 10.0 nM) and desipramine (300 nM; to occlude binding to the norepinephrine transporter). Nonspecific binding was determined in the presence of 20 µM benztropine. The reaction mixture was incubated for 60 min at 4°C, and stopped by a rapid vacuum filtration (Brandel M-12) through Whatman GF/B filters presoaked for 15 min in buffer containing polyethylenimine (0.05%). The filters were washed twice with 4 ml ice-cold buffer, and radioactivity was determined by scintillation counting. Protein concentration was determined by the method of Lowry et al. (1951). The binding parameters of [³H]mazindol, e.g., the maximal number of binding sites (B_max) and the dissociation constant (K_d), were determined by the LIGAND program, version 2.3.10.

Measurement of body temperature. The animals' body temperature was measured before the first injection of METH (5 mg/kg) and every 30 min thereafter, for a total period of 120 min. The temperature measurement was taken by Thermoscan Instant Thermometer (Thermoscan, San Diego, CA), which was positioned in the animal's ear for only 1 s. The variation between the readings from the two ears usually did not exceed 0.3°C, and the average temperature determined from both ears was considered as n = 1.

Measurement of locomotor activity. To investigate the consequence of administration of the high dose of METH (5 mg/kg × 3) on the animals' locomotor activity, we used three groups of animals for this set of experiments: nNOS(/), B6/SV129 and C57BL/6 (n = 20 for each strain). Experiments were conducted in a manner similar to the procedure previously described (Itzhak, 1997). On day 1, before any drug treatment was initiated, each mouse was placed in the test cage; this was a standard transparent rectangular rodent cage 42 × 42 × 20 cm high. The animals' spontaneous activity was measured for a 60-min period. Then one group (n = 10) received a saline injection, and a second group (n = 10) received a relatively low dose of METH (1 mg/kg); the animals' locomotor activity was then measured for a 60-min period. One hour later, the animals that received 1 mg/kg METH received three additional injections of METH (5 mg/kg each) in 3-h intervals, as described above for the neurochemical experiments. On day 4 (68-72 h after the last METH injection), the animals' spontaneous activity was measured for a 60-min period. Then the animals were challenged with METH (1 mg/kg), and their locomotor activity was recorded for a 60-min period.

Statistical analysis. Analysis of the differences in outcome between saline and METH treatments within each mouse strain was performed by Student's t test. The results of locomotor activity were analyzed by a two-way ANOVA (drug treatment × time or strain × time) with time as the repeated measure (e.g., comparison between locomotor activity on day 1 and day 4). Bonferroni multiple comparison adjustment was performed to determine differences between specific groups.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Content of striatal dopamine and its metabolites. The concentrations of dopamine and its metabolites DOPAC and HVA in striatal tissue of mice, treated with either saline or METH (5 mg/kg × 3), are summarized in figure 1. The administration of METH to nNOS(/) mice produce no significant change in tissue content of dopamine, DOPAC or HVA compared with the levels obtained after saline administration to nNOS(/) mice. However, the same treatment delivered to heterozygote nNOS(+/) mice revealed a 35% decrease in dopamine concentration (P < .05) compared with control. Administration of METH to wild-type B6/SV129 mice resulted in 45, 37 and 35% decreases in the concentration of dopamine, DOPAC and HVA, respectively, compared with saline treatment (P < .01; fig. 1). The same treatment given to C57BL/6 mice caused 68, 61, and 45% losses in dopamine, DOPAC and HVA concentrations, respectively, compared with saline treatment (P < .01; fig. 1). Finally, administration of METH to SV129 mice produced 37, 35 and 39% decreases in the levels of dopamine, DOPAC and HVA, respectively, compared with the saline treatment (P < .01; fig. 1). These findings suggest that nNOS(/) mice are markedly protected against METH-induced depletion of dopamine and its metabolites.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Concentration of dopamine (panel A), DOPAC (panel B) and HVA (panel C) in mouse striatum. Homozygote nNOS(/), heterozygote nNOS(+/) and wild-type B6/SV129, C57BL/6 and SV129 mice were administered either saline (control; n = 10-12 for each mouse strain) or METH (5 mg/kg × 3; n = 10-15 for each mouse strain). The concentrations of dopamine, DOPAC and HVA were determined 72 h after the last METH injection. Results are presented as mean ± S.E.M. * P < .05; ** P < .01 Student's t test for the comparison between control and METH for each mouse strain.

Dopamine transporter binding sites. Saturation binding experiments of [³H]mazindol to the DAT were carried out to assess further the dopaminergic neurotoxicity caused by administration of high doses of METH. The binding parameters of [³H]mazindol (in the presence of 300 nM desipramine) to striatal tissue are summarized in table 1. Administration of METH to B6/SV129, C57BL/6 and SV129 mice resulted in 60, 66 and 49% decreases, respectively, in the number of [³H]mazindol binding sites, compared with saline administration. However, administration of METH to heterozygote nNOS(+/) and homozygote nNOS(/) mice resulted in only 32 and 12% decreases, respectively, in [³H]mazindol binding sites, compared with saline treatment (table 1). These findings suggest, again, that nNOS(/) mice are relatively protected against METH-induced loss of DAT binding sites.

Body temperature. We observed, 30 and 60 min after the administration of the first dose of METH (5 mg/kg) to heterozygote nNOS(+/), B6/SV129, C57BL/6 and SV129 mice, a marked increase in body temperature, compared with the animals' temperature measured before METH administration (36.11 ± 0.15°C). As indicated in figure 2, the maximal hyperthermic effect was observed 60 min after METH administration (38.23 ± 0.2 to 38.74 ± 0.15°C). However, METH administration had no significant effect on the body temperature of homozygote nNOS(/) mice (fig. 2). In some experiments, the animals' temperatures were also measured after the third injection of METH. Results indicated no significant differences between the effect of the first and the third METH injection on the animals' temperature (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effect of METH on animals' body temperature. The mice's body temperatures (n = 10-15 for each strain) were taken 0 to 10 min before the administration of METH (5 mg/kg) and every 30 min thereafter for a total period of 120 min. Results are presented as mean ± S.E.M. A significant (** P < .01 paired Student's t test) increase in body temperature was observed 30 and 60 min after METH administration in all mouse strains examined except the homozygote nNOS(/).

Locomotor activity. Because we observed that the administration of high doses of METH (5 mg/kg × 3) had differential effects on homozygote nNOS(/) and wild-type animals, we sought to determine how the absence and presence of nNOS and dopaminergic neurotoxicity would correlate with the animals' spontaneous locomotor activity, as well as with METH-induced locomotor stimulation. To this end, we chose to investigate 3 out of the 5 animal strains we had. Homozygote nNOS(/), B6/SV129, and C57BL/6 were chosen because the nNOS(/) mice displayed practically no dopaminergic neurotoxicity, whereas the B6/SV129 and C57BL/6 mice developed the greatest dopaminergic neurotoxicity (fig. 1; table 1). Moreover, except for the deletion of the nNOS gene, the genetic backgrounds of the B6/SV129 and nNOS(/) mice are the most similar. The spontaneous locomotor activity of the three groups of mice before the administration of any drug had a typical pattern. Initially, the animals' locomotor activity was relatively high, and then, as the animals became familiar with the new environment (the test cage), locomotor activity declined over time (fig. 3). A two-way ANOVA (strain × time) revealed nonsignificant strain effect (P = .09) [time effect (P = .0098) and interaction (P = .142)]. Spontaneous locomotor activity was determined again on day 4, after the administration of METH (1 mg/kg followed by 5 mg/kg × 3) on day 1. Results presented in figure 3 (bottom panel) indicate that spontaneous locomotor activity after the administration of METH on day 1 did not change compared with that recorded before the administration of METH. The following cumulative ambulation counts were registered for the 60-min period for the three animal strains on day 1 and day 4. nNOS(/): day 1, 4893 ± 381; day 4, 5125 ± 423. B6/SV129: day 1, 4186 ± 526; day 4, 4462 ± 582. C57BL/6: day 1, 4738 ± 320; day 4, 4269 ± 501. Thus ambulation counts on days 1 and 4 were not significantly different from each other for any of the groups tested.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Spontaneous locomotor activity of homozygote nNOS(/) and wild-type B6/SV129 and C57BL/6 mice (n = 20 for each strain) before (day 1) and after (day 4) the administration of a neurotoxic dose of METH. On day 1, each mouse was placed in a test cage, and ambulatory counts were registered every 10 min for a 60-min period. On day 1, a two-way ANOVA (strain × time) revealed a nonsignificant strain effect (P = .09), a significant time effect (comparison between the first and last 10-min intervals P = .0098) and nonsignificant interaction (P = .142). Spontaneous locomotor activity was measured again on day 4 (after the administration of 1.0 mg/kg and 5 mg/kg × 3 METH on day 1). A two-way ANOVA (strain × time) for comparison between locomotor activity on day 1 and that on 4 revealed nonsignificant strain (P = .12) and time (P = .18) effects.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effect of METH on locomotor activity of the mice. Homozygote nNOS(/) and wild-type B6/SV129 and C57BL/6 mice were divided into two groups (n = 10 each). On day 1, animals were placed in the test cage for 60 min and then received either saline or METH (1.0 mg/kg), and ambulation counts were recorded for a 60-min period. On day 4 (after exposure to the neurotoxic dose of METH on day 1), animals were placed in the test cage for 60 min and then received METH (1.0 mg/kg), and ambulation counts were recorded for a 60-min period. A two-way ANOVA (drug treatment × time) with time as the repeated measure yielded a significant interaction (P = .0001). Bonferroni multiple comparison adjustment was performed to determine specific group differences. On day 1, METH vs. saline: *a P = .0001. Comparison between day 1 and day 4 for the METH group: *b P = .0001.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study suggests that the nNOS isoform plays a major role in the dopaminergic neurotoxicity produced by the administration of a high dose of METH. Mice lacking the nNOS gene were protected against METH-induced depletion of dopamine, DOPAC and HVA and of dopamine transporter binding sites. nNOS(/) mice did not display any overt behavioral deficits compared with wild-type animals, and they were responsive to METH-induced locomotor stimulation. However, unlike wild-type animals, nNOS(/) mice did not develop behavioral sensitization as a consequence of the prior exposure to METH. Thus our data underscore the role of nNOS in METH-induced dopaminergic neurotoxicity and behavioral sensitization.

nNOS and dopaminergic neurotoxicity. Two major findings suggest that mice lacking the nNOS gene are protected against METH-induced dopaminergic neurotoxicity. First, the administration of a high dose of METH (5 mg/kg × 3) to nNOS(/) mice did not produce depletion in striatal content of dopamine, DOPAC and HVA, where administering the same treatment to heterozygote nNOS(+/) mice produced a 35% depletion in dopamine. A greater decrease in dopamine concentration and in that of its metabolites was observed after the administration of METH to wild-type animals (fig. 1). Second, METH administration to homozygote nNOS(/), heterozygote nNOS(+/) and wild-type animals resulted in 12, 35 and 49 to 66% depletion in DAT binding sites, respectively (table 1).

Behavioral consequences of the administration of a high dose of METH. The development of behavioral sensitization to amphetamines is usually investigated via the administration of intermittent relatively low, non-neurotoxic doses of amphetamines. The induction of behavioral sensitization to amphetamines has been linked to amphetamine-induced psychosis in humans (Robinson and Becker, 1986; Segal and Kuczenski, 1997) and to the development of drug craving (Robinson and Berridge, 1993).

Dopamine/glutamate/NO interactions. The mechanism underlying the protection against METH-induced dopaminergic neurotoxicity afforded by deletion of nNOS remains to be investigated. Because evidence suggests that METH-induced neurotoxicity is associated with an increase in glutamate release in the striatum (Nash and Yamamoto, 1992), and because NOS inhibitors attenuate NMDA receptor-mediated neurotoxicity (Garthwaite, 1991; Snyder, 1992; Dawson et al., 1993), it is possible that deletion of nNOS provides protection against the glutamatergic-mediated neurotoxicity that arises from the administration of a high dose of METH. Although dopaminergic neurotoxicity may correlate with increased glutamatergic activity, amphetamine-induced behavioral sensitization (by the administration of relatively low doses) may not necessarily overlap with an increase in extracellular glutamate level in rat ventral tegmental area and nucleus accumbens (Xue et al., 1996). Nevertheless, the interactions between nigrostriatal dopamine and corticostriatal glutamate transmission, and particularly the hypothesis that dopamine depletion is associated with a subsequent increase in glutamatergic transmission (Carlsson and Carlsson, 1990), may support the following theory: METH-induced dopamine depletion is associated with increased glutamatergic activity, which subsequently leads to stimulation of nNOS. The increase in NO levels may further modulate the release of dopamine and glutamate (Lonart et al., 1993; Montague et al., 1994), which contribute to both neurotoxicity and behavioral sensitization.