Introduction

Top Abstract Introduction Experimental Procedures Results Discussion Conclusions References

Attention deficit-hyperactivity disorder (ADHD) is a complex developmental behavioral and cognitive disorder that affects approximately 3 to 5% of school-aged boys (Barkley, 1997). For more than 50 years, the most widely prescribed pharmacological treatment for patients with ADHD has been psychostimulant drugs including methylphenidate and amphetamine. Stimulants facilitate the release and synaptic availability of dopamine (DA) (West et al., 1995; Pliszka et al., 1996; Barkley, 1997; Solanto, 1998; Fleckenstein et al., 2000). Stimulants also facilitate the release of other monoamine neurotransmitters, notably norepinephrine (NE) and serotonin (5-HT), which may also contribute to therapeutic mechanisms including improvements in response-delay, working memory, and attention, as well as modulation of motor activity (Pliszka et al., 1996; Jacobs and Fornal, 1997; Solanto, 1998; Biederman and Spencer, 1999). This view is strongly supported by the reported clinical usefulness of selective inhibitors of NE transport, including desipramine, nortriptyline, and atomoxetine in ADHD (Biederman and Spencer, 1999; Popper, 2000). Furthermore, venlafaxine and other even more selective serotonin reuptake inhibitors (SRIs) may also yield limited benefits in ADHD (Biederman and Spencer, 1999; Popper, 2000).

Salient features of ADHD can be simulated in juvenile rats with neonatal 6-hydroxydopamine (6-OHDA) lesions that selectively destroy DA projections to forebrain when combined with desipramine pretreatment to prevent loss of NE (Shaywitz et al., 1978; Zhang et al., 2001; Davids et al., 2002). The severe depletion of DA yields increased motor activity and learning deficits (Shaywitz et al., 1978; Takasuna and Iwasaki, 1996). Hyperactivity resulting from neonatal 6-OHDA lesions is dose dependently reversed by stimulants including d-amphetamine and both dl- and d-methylphenidate (Shaywitz et al., 1978; Zhang et al., 2001; Davids et al., 2002). Learning deficits exhibited by 6-OHDA lesioned rats also respond favorably to stimulants (Shaywitz et al., 1978). These actions parallel the effects of stimulants in patients with ADHD (Kostrzewa et al., 1994; Biederman and Spencer, 1999).

The hypothesis that stimulants act simply by activating forebrain dopaminergic systems is inconsistent with the typically severe depletion of cerebral DA in such lesioned rats (Joyce et al., 1996; Schwarting and Huston, 1996). On the other hand, support for contributions of NE and 5-HT to the actions of stimulants in 6-OHDA lesioned rats includes the following: 1) NE neurons are typically preserved in the lesioning model (Luthman et al., 1990; Ordway, 1995); 2) 5-HT neurons may actually overgrow in response to early removal of DA neurons (Towle et al., 1989; Kostrzewa et al., 1998); and 3) selective inhibitors of NE and 5-HT reuptake can exert inhibitory effects on behavioral activity in rats under some conditions (O'Connor and Leonard, 1988; Geyer, 1996). Remarkably, however, direct testing of the behavioral effects of selective NE or 5-HT inhibitors in 6-OHDA lesioned hyperactive rats and comparisons with selective inhibitors of DA reuptake remain to be carried out.

Accordingly, we tested the hypothesis that selective inhibitors of the inactivation of NE or 5-HT by neuronal transport would inhibit motor hyperactivity in juvenile male rats with selective neonatal lesions of the forebrain DA system, but that selective inhibitors of DA transport would not. We employed pairs of chemically dissimilar agents, including selective inhibitors of the transport of: 1) NE (desipramine, nisoxetine); 2) 5-HT (citalopram, fluvoxamine); and 3) DA [amfonelic acid, GBR-12909 (1-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-[3-phenylpropyl]piperazine dihydrochloride)] for comparison with d-methylphenidate.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion Conclusions References

Materials and Animal Subjects. Amfonelic acid, d-amphetamine sulfate, desipramine hydrochloride, GBR-12909 dihydrochloride, nisoxetine hydrochloride, and 6-hydroxydopamine hydrobromide were obtained from Sigma/RBI (Natick, MA). Donated drugs included d-methylphenidate (Celgene, Warren, NJ), (±)-citalopram hydrobromide (Lundbeck, Copenhagen, Denmark), and fluvoxamine maleate (Duphar, Amsterdam, Netherlands). We recently tested these drugs for potency and selectivity at monoamine transporters, using homogenates of rat forebrain and selective radioligands (Kula et al., 1999), as is summarized in Table 1. d-Methylphenidate and d-amphetamine showed >10-fold selectivity for DA and NE over 5-HT transporters. Amfonelic acid and GBR-12909 were highly selective for DA transporters (DATs) (350-fold compared with NE and 5-HT transporters). Desipramine and nisoxetine were highly selective for the NE transporter (340-fold compared with DA and 5-HT transporters), and citalopram and fluvoxamine were highly selective for the 5-HT transporter (1800-fold compared with DA and NE transporters; Table 1).

View this table: [in this window] [in a new window]   TABLE 1 Potency (Ki ± SE; nM) at monoamine transporters based on in vitro studies using rat forebrain tissue Data adapted from Kula et al., 1999; reprinted with permission from Elsevier Science.

Neonatal 6-Hydroxydopamine Lesioning. Neonatal 6-OHDA lesioning follows previously detailed methods (Zhang et al., 2001; Davids et al., 2002). On postnatal day (PD) 1, male rat pups were randomly assigned at 10 per lactating dam, with which both 6-OHDA lesioned and sham lesioned rat pups also were housed after lesioning on PD 5. Before lesioning, pups were given s.c. injections of desipramine hydrochloride (25 mg/kg body weight). After 45 min, subjects randomly received a 20-µl intracisternal injection of vehicle [0.9% (w/v) sodium chloride containing 0.1% (w/v) ascorbic acid] or 6-OHDA hydrobromide (100 µg free base), under hypothermal anesthesia (Zhang et al., 2001; Davids et al., 2002), and were returned to nursing dams after regaining consciousness.

Behavioral Testing of Responses to Selective Transport Inhibitors. Motor activity was recorded with an infrared photobeam activity-monitoring system (San Diego Instruments, San Diego, CA). Tests were conducted in a novel environment (43.2 × 20.3 × 20.3 cm transparent plastic cages within 4 × 8 horizontal infrared beams at 3.3 cm elevation), between 10:00 AM and 4:00 PM in the absence of food and water (Zhang et al., 2001; Davids et al., 2002). Locomotor activity was scored as breaking of consecutive photobeams. Data were collected at 5-min intervals for 90 min.

Verification of Neonatal 6-Hydroxydopamine Lesioning. Lesions were verified by autoradiographic analysis of DATs as an index for DA terminals in rat forebrain using the selective radioligand [³H]-CIT (Kula et al., 1999). Rats were sacrificed by decapitation at 72 h (PD 28) after behavioral testing on PD 25, and brains were quickly removed and frozen. Coronal sections (10 µm) of brain tissue were prepared in a cryostat at 17°C, thaw-mounted on gelatin-coated microscopic slides, and stored at 80°C until autoradiographic analysis. For this analysis, brain sections were preincubated for 60 min at room temperature in 50 mM Tris-citrate buffer (pH 7.4) containing 120 mM NaCl and 4 mM MgCl₂. Incubation continued for another 60 min in fresh buffer containing 2 nM [³H]-CIT. Specific binding was defined with excess GBR-12909 (1 µM). After incubation, slides were washed twice in ice-cold buffer for 5 min, rinsed in cold deionized water, and air-dried. Slides were exposed to ³H-sensitive Hyperfilm radiographic film at 4°C for 14 days with ³H-standards and developed using standard photographic procedures. Density of radioligand binding to DATs was quantified with a computerized MCID-M4 image analyzer (Imaging Research, Inc., St. Catherines, ON, Canada) and converted to nanocuries per milligram of tissue by use of the ³H-standards, with specific binding expressed as femtomoles per milligram of tissue. Radioligand density was quantified in lateral and medial caudate-putamen (CPu) and nucleus accumbens septi (NAc), all as detailed previously (Tarazi et al., 2001).

Data Analysis. Lesion effects on DAT binding were analyzed by two-way analysis of variance for overall changes across treatments and brain regions, followed by post hoc Dunnett's t tests for planned comparisons. Data are presented as means ± S.E.M. Since motor activity scores involved repeated measurements over 90 min, behavioral data were analyzed using a population-averaged panel-data model based on random-effects general estimating equations with Stata software for the Macintosh microcomputer (Stata Corp., College Station, TX) (Liang and Zeger, 1986). Differences between treatment groups are considered statistically significant at p  0.05 in two-tailed tests.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion Conclusions References

Effects of Lesioning on Dopamine Terminals. Neonatal 6-OHDA lesioning yielded major reductions in DAT binding in CPu as well as in NAc when tested at PD 28, following pharmacological assessments (Table 2). Average losses of specific DAT binding were 78% in both lateral and medial CPu and 70% in NAc in lesioned animals compared with sham controls.

View this table: [in this window] [in a new window]   TABLE 2 Effects of neonatal 6-OHDA lesions on DAT binding in caudate-putamen and nucleus accumbens septi, assayed autoradiographically with [3H]-CIT in brain sections from juvenile rats (PD 28) Data are specific binding (mean femtomoles per milligram of tissue ± S.E.M.) for N = 9-10 rats/group.

Lesion-Induced Hyperactivity. Neonatal 6-OHDA lesioning resulted in a marked overall increase in locomotor activity over 90-min testing sessions in a novel environment at PD 25 (p < 0.01 versus sham controls; Fig. 1). Motor activity within the initial 10 min was similar in lesioned (n = 12) and sham control subjects (n = 8), but in contrast to the controls in which locomotion declined rapidly to a stable level within 30 min, activity in lesioned rats failed to decline within the 90-min session. Cumulative locomotor scores for the entire session in lesioned versus sham subjects averaged 3109 ± 876 versus 285 ± 82 counts (an 11-fold difference; p < 0.01).

View larger version (38K): [in this window] [in a new window]   Fig. 1.   Effects of neonatal 6-OHDA lesioning on mean ± S.E.M. locomotor activity counts per 5 min versus time during testing at PD 25 in sham controls (open circles; n = 8) and neonatally 6-OHDA lesioned male rats (filled circles; n = 12) given vehicle i.p. immediately before testing. Locomotion increased markedly in lesioned rats (p < 0.01). Further demonstrated effects of d-methylphenidate, 10 mg/kg, i.p., on sham (open triangles; n = 6) and lesioned rats (filled triangles; n = 10) are shown. Locomotion in sham rats increased markedly (p < 0.001) and decreased significantly in lesioned rats (p < 0.01) after treatment with d-methylphenidate.

Behavioral Effects of Test Agents. Motor activity was markedly increased by d-methylphenidate (10 mg/kg, i.p.) in sham control subjects (n = 6, p < 0.001; Fig. 1). At the same dose, hyperactivity in lesioned rats was significantly antagonized by d-methylphenidate (n = 10, p < 0.01; Fig. 1).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Effects of amfonelic acid, 3 mg/kg (filled circles), and GBR-12909, 10 mg/kg, i.p. (open triangles), on locomotor activity counts per 5 min ± S.E.M. versus time during testing, in sham lesioned (n = 6-8; A) and 6-OHDA lesioned male rats (n = 9-12; B) at PD 25. Both treatments significantly increased locomotion in sham controls only, compared with vehicle-treated controls (open circles) (both p < 0.001).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Effects of citalopram, 10 mg/kg (filled circles), and fluvoxamine,10 mg/kg, i.p. (open triangles), on locomotor activity counts per 5 min ± S.E.M. over time, in sham lesioned (n = 7-8; A) and 6-OHDA lesioned male rats (n = 9-12; B) at PD 25. Both treatments significantly reduced motor activity in lesioned rats only, compared with vehicle-treated controls (open circles; citalopram, p < 0.01; fluvoxamine, p < 0.05).

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Effects of desipramine, 10 mg/kg (filled circles), and nisoxetine,10 mg/kg, i.p. (open triangles), on locomotor activity counts per 5 min ± S.E.M. over time during testing in sham lesioned (n = 7-8; A), and 6-OHDA lesioned male rats (n = 9-12; B) at PD 25. Both treatments significantly reduced motor activity in lesioned rats only, compared with vehicle-treated (open circles) controls (desipramine, p < 0.05; nisoxetine, p < 0.01).

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Summary of pharmacological effects on hyperactivity in neonatally 6-OHDA lesioned rats, based on data in Figs. 1-4. Data are mean ± S.E.M. percentage of control total locomotor activity levels over 90 min of testing at PD 25; controls are lesioned rats given a vehicle. All differences (except with amfonelic acid and GBR-12909) differed significantly (*, p < 0.01; **, p < 0.001) from vehicle-controls (indicated by the vertical dotted line).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion Conclusions References

In accord with previous studies (Shaywitz et al., 1978; Zhang et al., 2001; Davids et al., 2002), neonatal 6-OHDA lesioning of developing DA projections in rat forebrain resulted in robust hyperactivity when lesioned male rats were tested in a novel environment at a later developmental stage, PD 25 (Fig. 1). Autoradiographic analysis of the binding of a potent and selective radioligand to DAT sites in forebrain indicated major losses of DA innervation in the DA-rich CPu and NAc of lesioned subjects compared with sham controls (Table 2). In line with previous studies, d-methylphenidate increased locomotor activity in sham control subjects (Patrick et al., 1987; Davids et al., 2002) and antagonized hyperactivity in 6-OHDA lesioned rats (Davids et al., 2002) (Fig. 1).

Behavioral Effects of Inhibitors of Dopamine Transport. Selective inhibitors of DA reuptake, amfonelic acid and GBR-12909, both greatly increased locomotor activity in sham control rats (Fig. 2), as expected from previous studies (Mueller, 1993; Rothman and Glowa, 1995). Behavioral effects of stimulants are believed to be mediated, at least in part, by increased effects of DA, particularly in the behaviorally critical mixed limbic-motor region, NAc (Solanto, 1998). Infusions of DA or amphetamine within NAc can increase locomotion (Staton and Solomon, 1984), whereas destruction of DA terminals in NAc prevented motor hyperactivity in response to d-amphetamine (Kelly and Iversen, 1976). Stimulants enhance the extracellular, and presumed synaptic, concentrations of DA in rats in correlation with their behavioral effects (Kuczenski and Segal, 1997, 2001), as well as in brain tissue of human subjects diagnosed with ADHD and visualized with positron-emission tomography (Volkow et al., 2001).

Behavioral Effects of Inhibitors of Serotonin Transport. The selective SRIs citalopram and fluvoxamine (Table 1) did not affect motor activity in sham control rats, but both drugs strongly antagonized motor hyperactivity induced by neonatal 6-OHDA lesions (Figs. 3 and 5). These results are in accord with the reported ability of the 5-HT releasing agent fenfluramine and mixed 5-HT receptor agonist quipazine to reduce motor hyperactivity in 6-OHDA-treated rats (Heffner and Seiden, 1982). Furthermore, antihyperactivity effects of psychostimulants in 6-OHDA lesioned rats can be antagonized by the 5-HT antagonist methysergide (Heffner and Seiden, 1982). These results suggest that the locomotor-reducing effect of methylphenidate and amphetamine in 6-OHDA lesioned rats may reflect increased serotonergic neurotransmission.

Behavioral Effects of Inhibitors of Norepinephrine Transport. The selective NE reuptake inhibitors desipramine and nisoxetine (Table 1) strongly antagonized motor hyperactivity in lesioned rats without affecting activity in sham control subjects (Figs. 4 and 5). The NE system, including tissue concentrations of NE and expression of various adrenergic receptor types in forebrain tissue, appears to undergo more or less normal development in the neonatally DA-selective lesioned rat (Luthman et al., 1990; Ordway, 1995), making NE available for actions of stimulants as well as selective inhibitors of NE transport.

	Conclusions

Top Abstract Introduction Experimental Procedures Results Discussion Conclusions References

We found that selective inhibitors of NE and 5-HT transport consistently antagonized motor hyperactivity in juvenile male rats after selective neonatal lesioning of the DA system with 6-OHDA, suggesting an important role by NE or 5-HT in the effects of stimulant agents that act on neuronal release and reuptake of NE or 5-HT as well as DA in this widely employed model of clinical ADHD. Methylphenidate (d- and dl-methylphenidate) and d-amphetamine are relatively selective for NE and DA transporter, compared with 5-HT transporter (Kula et al., 1999; Fleckenstein et al., 2000; Table 1). Also, amphetamine can release DA, NE, and 5-HT in intact rats, with possibly the greatest action on NE (Rothman et al., 2001), whereas methylphenidate increases release of both DA and NE effectively, with much less effect on 5-HT (Kuczenski and Segal, 1997). These several observations provide consistent additional support for a contribution of NE to the antihyperactivity effects of both stimulants in 6-OHDA lesioned rats, with a parallel contribution to stimulant treatment in clinical ADHD remaining to be proved. The neuronal circuits and interactions underlying the beneficial effects of inhibitors of the uptake of both NE and 5-HT remain incompletely understood and may involve interactions between NE and 5-HT systems, and interactions of both with DA neurons (Heffner and Seiden, 1982; Jacobs and Fornal, 1997; Yeghiayan et al., 1997; Sasaki-Adams and Kelley, 2001).

Finally, despite inherent risks in generalizing from laboratory models to clinical therapeutics, the present findings encourage clinical consideration of alternatives to stimulants. Stimulants are highly effective in ADHD with low risk of tolerance, but they have significant adverse effects and a potential for abuse and illicit trafficking (Barkley, 1997; Popper, 2000). SRIs also might be expected to contribute to the treatment of ADHD, but their clinical efficacy appears to be inferior to that of stimulants (Popper, 2000), despite their strong hyperactivity-inhibiting actions in the lesioning model (Fig. 3). Moreover, 5-HT has complex actions on motor behaviors (Gerson and Baldessarini, 1980; Geyer, 1996; Jacobs and Fornal, 1997). Selective inhibitors of NE transport may be particularly likely to contribute to the clinical treatment of ADHD (Biederman and Spencer, 1999; Popper, 2000).