Dopamine neurons play an important modulatory role in the brain 71. Neurons that release dopamine influence behaviour by exerting modulatory effects on the transfer of information through neuronal circuits that connect functionally distinct cortical areas to specific regions of the striatum in parallel cortico-striato-thalamo-cortical pathways. Dopamine assists in reprogramming the brain by selectively reinforcing the weights of the synapses that are active around the time of behavioural reinforcement 472.

There are three major dopaminergic systems in the brain, the mesolimbic, mesocortical and nigrostriatal pathways. Mesolimbic dopamine neurons project from the ventral tegmental area of the midbrain (VTA) to limbic areas of the brain. The firing rate of dopamine neurons is increased in response to unexpected reward and decreased when a fully predicted reward is omitted 7374. It has been suggested that deficient reinforcement of appropriate behaviour and/or deficient extinction of previously reinforced behaviour can give rise to ADHD symptoms of delay aversion, hyperactivity in a familiar environment, impulsiveness, deficient sustained attention, increased behavioural variability and failure to extinguish previously acquired behaviour 4.

The mesocortical dopamine system originates in the VTA and projects to cortical areas, including the prefrontal, parietal and temporal cortex. These dopamine projections modulate circuits that are known to play an important role in a variety of executive functions, including motor control, behavioural inhibition, attention, and working memory 475. Dopamine activation of DRD2 selectively modulates neural activities associated with memory-guided motor activity in delayed response tasks whereas DRD1 are responsible for memory-related persistent activation of prefrontal cortex neurons 76. Deficient dopamine-mediated modulation of prefrontal cortical circuits has been suggested 4 to cause attention response deficiencies (impaired orienting responses, saccadic eye movements and responses towards a target) and impaired executive functions (poor behavioural planning).

Nigrostriatal dopamine neurons project from the substantia nigra pars compacta to the dorsal striatum (caudate nucleus and putamen). Impaired dopamine modulation of cortico-striato-thalamo-cortical circuits can impair motor function and cause deficient habit learning i.e. impaired nondeclarative memory formation 4. These impairments can give rise to apparent developmental delay, clumsiness, and neurological "soft signs" 4.

Dopamine uptake, storage and/or metabolism are disturbed in SHR 5877, as has been proposed for children and adults with ADHD 5253. Although the tissue concentration of dopamine, which reflects vesicle stores of dopamine, is similar in SHR and WKY 78, dopamine turnover which reflects release and metabolism of dopamine was lower in adult SHR substantia nigra, VTA, striatum and frontal cortex 7879. The dopamine metabolite, homovanillic acid, and the homovanillic acid /dopamine ratio are much lower in several brain areas of adult SHR compared to WKY. These findings are consistent with increased expression of DAT1 increasing dopamine re-uptake and reutilization thereby reducing metabolism. The results could also be interpreted to suggest that dopamine release is decreased and dopamine transmission is impaired in SHR.

In vitro stimulation-evoked (electrical and/or exposure to high K⁺ concentration) release of dopamine from terminals of mesocortical, mesolimbic and nigrostriatal dopamine neurons of SHR is significantly less than that of WKY 555658597780. A similar decrease in stimulus-evoked release of dopamine was observed in SHR nucleus accumbens shell in vivo 54. The elevated dopamine concentration in the nucleus accumbens shell may have increased activation of endogenous DRD2 autoreceptors and thus reduced stimulus-evoked dopamine release 54. In vitro DRD2-mediated inhibition of dopamine release was greater in SHR caudate-putamen and nucleus accumbens than WKY, while DRD2 function was unchanged in prefrontal cortex of SHR 77. Results are consistent with increased endogenous activation of DRD2 in the nucleus accumbens while increased efficacy of endogenous dopamine activation of DRD2 autoreceptors was suggested to account for the decreased stimulus-evoked release of dopamine in SHR striatum 77.

Increased DRD2 function was suggested to have occurred in a compensatory reaction to abnormally elevated dopamine levels at an early stage of development, perhaps as a result of exposure to stress and/or a genetic defect 80. A likely explanation could also be that elevated extracellular dopamine levels could have resulted from decreased DAT1 expression during the first postnatal month of SHR 50 which caused compensatory upregulation of both DAT1 and DRD2 receptors 515877. Increased d-amphetamine stimulated, DAT-mediated, release of dopamine from SHR prefrontal cortex and striatum is consistent with upregulation of DAT1 in adult SHR 58.

In agreement with decreased stimulus-evoked release of dopamine, postsynaptic DRD1 are increased in the caudate-putamen and nucleus accumbens of SHR 518182 suggesting compensatory upregulation of postsynaptic receptors.

Dopamine dysfunction may contribute to the altered reinforcement processes of ADHD 445. DRD1 mediates reinforcement by strengthening synaptic connections between neurons (long-term potentiation, LTP) or weakening synaptic connections (long-term depression, LTD) in neural circuits that involve prefrontal cortex and/or striatum (e.g. cortico-striato-thalamo-cortical circuits) 748384. LTP is regarded as a neuronal correlate of learning 68. It requires interplay between several factors. Among these is coincident glutamate stimulation of NMDA receptors and local membrane depolarization sufficient to remove the magnesium block of NMDA receptor channels. Glutamate activation of AMPA receptors allows influx of sodium ions and can thereby produce this depolarization. Activation of the DRD1-protein kinase A signalling pathway increases the mobilisation of AMPA and NMDA receptors to the cell surface, thereby promoting LTP 858687.

Calcium enters the cell through the NMDA receptor channel, activates various protein kinases including calcium/calmodulin-dependent protein kinase II, and mobilizes "silent" AMPA receptors required for LTP to take place 68. NMDA receptor-induced excitation is enhanced by DRD1 activation and attenuated by DRD2 activation 888990. Thus DRD1 activation may synergistically increase the excitatory actions of glutamate at NMDA receptors, increasing the open time of NMDA channels and therefore increase the calcium signal. Activation of DRD1 receptors is required for stimulation of cAMP formation, and subsequent activation of cAMP-dependent protein kinase, phosphorylation of CREB and gene transcription required for consolidation of memory traces 91.

NMDA receptors are required for LTP in the hippocampus 68, in cortico-striatal synapses 92, and in the nucleus accumbens 93. Phasic application of dopamine in the prefrontal cortex facilitates LTP, suggesting that dopamine can promote reinforcement processes by strengthening (or weakening) network connections in the prefrontal cortex 94. Bilateral infusion of a DRD1 agonist increased attentional accuracy and facilitated short-term spatial memory after delays of several seconds and impaired memory of the location of a visual target after short delays, thereby modulating short-term working memory in a delay-dependent manner 95. Effective decision making requires the ability to adapt behaviour on the basis of changes in emotional significance. Rats with lesions of the orbitofrontal cortex showed increased preference for larger but delayed rewards whereas rats with lesions of the basolateral amygdala showed increased choice of small immediate rewards 96.

Within the striatum, LTP (and LTD) only occurs in the presence of dopaminergic input 97. Consistent with decreased DRD1 activation in SHR giving rise to decreased facilitation of NMDA receptor function and impaired gene transcription, SHR have reduced expression of calcium/calmodulin-dependent protein kinase II and c-fos gene in the anterior striatum 9899100. Decreased DRD1 function will alter transmission of signals in cortical and striatal circuits of SHR and impair dopamine-mediated cognitive function and reinforcement of appropriate behaviour.

Deficient DRD1 function in SHR implies that only stimuli with strong reinforcer value will release enough DA to stimulate DRD1 sufficiently to facilitate NMDA receptor function and produce phosphorylation of CREB and other important proteins/peptides required for strengthening of synaptic connections in circuits representing goal-directed learning and memory consolidation. Weak stimuli will not cause behaviour to be reinforced in SHR because of reduced activation of the DRD1 signalling pathways. As suggested in the dynamic developmental behavioural theory (DDT) of Sagvolden et al 4, SHR have a steeper delay gradient, so decreased activation of DRD1 signalling pathways in response to an unexpected reinforcer would not strengthen synaptic connections in circuits that were activated by stimuli some time prior to the reinforcer with the result that more recently activated behavioural circuits would be preferentially strengthened and their memory consolidated.

The striatum is central to behavioural control and receives the greatest density of dopamine input of all central nervous structures. Striatal dopamine hypofunction may be associated with subtle motor control problems in children with ADHD and SHR 4. Impaired dopamine regulation of striatal function may contribute to the poor motor development associated with severe cases of ADHD 101. Response "disinhibition" may be due to impaired motor control associated with dopamine hypofunction in the striatum rather than frontal lobe dysfunction 4. Indeed, motor control problems may explain a number of effects including clumsiness, increased variability in speed, less accurate response re-engagement, "failure to inhibit" responses when quick reactions are required, impaired orienting responses, increased number of responses with extended reaction times, apparent developmental delay, neurological "soft signs", and language delays 4.

Papa et al. 100 found decreased calcium/calmodulin-dependent protein kinase II in the nucleus accumbens shell but not the core of SHR when compared to WKY. The mesolimbic dopamine projection to the shell subdivision of the nucleus accumbens is responsible for motivation and determines the amount of effort an animal is prepared to exert in order to obtain a reward. Hypofunction of the mesolimbic dopamine system will impair the function of the mesocortical and nigrostriatal dopamine systems by influencing dopamine release and the cortico-striato-thalamo-cortical circuits that dopamine modulates. This could impair learning and expression of goal-directed behaviour thereby contributing to the aspects of ADHD symptoms displayed by SHR 102.

In addition to the hypothesis that dopaminergic systems are hypofunctional in ADHD, noradrenergic neurons have been suggested to be poorly regulated and hyperfunctional in the prefrontal cortex of children with ADHD 40103104. Noradrenergic neurons appear to enhance the signal-to-noise ratio in prefrontal and parietal cortices, amplify responses to attended stimuli, and reduce responses to irrelevant stimuli 71105. Both of these functions are defective in ADHD 71.

The locus coeruleus diffusely innervates diverse regions throughout the central nervous system including the entire cerebral cortex, various subcortical areas, cerebellum and spinal cord, and plays an important role in attention, arousal, orienting, and vigilance 40. For example, locus coeruleus neurons respond selectively to attended (target) stimuli; tonic locus coeruleus activity corresponds to arousal state, and both very low and very high locus coeruleus activity are associated with impaired vigilance 103105. Noradrenergic neurons that project from the locus coeruleus to the prefrontal cortex release norepinephrine which helps to guide behaviour by modulating the transfer of information through neuronal circuits that are responsible for selective and sustained attention 40.

Norepinephrine, like dopamine, alters the strength of neural connections leading to adaptive changes in behaviour. This occurs through activation of β-adrenoceptors that stimulate cAMP formation, with subsequent activation of cAMP-dependent protein kinase, phosphorylation of CREB and gene transcription required for consolidation of memory traces in several brain areas, including amygdala and hippocampus 91106. Dopamine and norepinephrine act in concert to regulate prefrontal cortex function and thereby ensure appropriate behaviour 103. α₂-Adrenoceptors agonists enhance performance of tasks requiring prefrontal cortex function while α₁-adrenoceptor agonists impair prefrontal cortex function 103. Furthermore, α₁- and α₂-antagonists reverse these effects 103107. Either excessive activation of α₁-adrenoceptors or deficient α₂-adrenoceptor mediated modulation of prefrontal cortical circuits can impair prefrontal cortex function 103108. The prefrontal cortex projects to the VTA and locus coeruleus, thereby influencing the firing rate of both dopamine and norepinephrine neurons and impacting on many cognitive processes 103.

The highly specific antagonist of NET1, atomoxetine, is as effective as methylphenidate in treating ADHD 38109, further emphasizing an important role for the noradrenergic system in the disorder. However, atomoxetine also increases synaptic availability of dopamine in the prefrontal cortex 110 which may contribute to its beneficial effects. Drugs used to treat ADHD symptoms are likely to have different effects on different neurotransmitter systems. Drugs that act on the noradrenergic system, such as atomoxetine, tricyclic antidepressants like the NET1 blocker, desipramine, and α₂-adrenoceptor agonists such as clonidine and guanfacine, have a different therapeutic time-course compared to psychostimulants. Methylphenidate produces amelioration of ADHD symptoms within 30 minutes and is short-acting whereas noradrenergic drugs have to be administered for longer periods of time before a therapeutic effect is observed, and improvement is sustained for several months 38. This suggests that noradrenergic drugs cause long-term adaptive changes that are therapeutic. Chronic treatment with desipramine produces a series of changes in presynaptic and postsynaptic α- and β-adrenoceptors and causes long-term downregulation of cortical β-adrenoceptors 111112113. This suggests that behavioural improvement can perhaps be achieved by decreased noradrenergic activation of cortical β-adrenoceptors, thereby decreasing noradrenergic function, which is consistent with a hyperactive noradrenergic system and the dopamine/norepinephrine imbalance hypothesis of ADHD.

Disturbances in norepinephrine metabolism in SHR are suggested by the finding that tyrosine hydroxylase gene expression is higher in the ventrolateral medulla oblongata of SHR than WKY 114, consistent with elevated norepinephrine concentration in several brain areas of SHR including locus coeruleus, substantia nigra and prefrontal cortex 79. Increased norepinephrine concentrations in SHR brain is consistent with downregulation of β-adrenoceptors in cerebral cortex of SHR 115. Furthermore, NET1 function is increased in SHR cerebral cortex 115 which could increase uptake of dopamine into noradrenergic terminals and varicosities and deplete extracellular dopamine in the prefrontal cortex 116.

Evidence suggests that there is an imbalance between dopaminergic and noradrenergic neurotransmission in the prefrontal cortex of SHR 104. While dopamine release is decreased in SHR prefrontal cortex, norepinephrine concentrations are elevated. The noradrenergic system appears to be hyperactive as a result of impaired α₂-autoreceptor regulation 104.

Stimulus-evoked (electrically stimulated or K⁺-evoked) release of norepinephrine from prefrontal cortex slices of SHR was no different from that of WKY 117. However, α_2A-adrenoceptor mRNA levels were decreased in SHR compared to WKY and α₂-autoreceptor-mediated inhibition of norepinephrine release was less efficient in SHR than in WKY suggesting that α₂-adrenoceptor function is impaired 117118119. α_2A-Adrenoceptors are the subtype specifically expressed in the prefrontal cortex, so impaired α_2A-adrenoceptor function would be expected to impair cognition 103120.

Decreased α₂-autoreceptor-mediated regulation of norepinephrine neurons and impaired inhibition of norepinephrine release may be particularly disruptive to the function of target structures when the firing rate of locus coeruleus neurons is high, causing excessive spillover of norepinephrine. Repeatedly increased release of norepinephrine from sympathetic nerve terminals could give rise to the stress-dependent 60 development of hypertension in SHR. Expression of the gene encoding G_iα the G-protein subunit that inhibits cAMP formation from ATP by adenylyl cyclase is increased in SHR aorta at 2 weeks of age, possibly reflecting an attempt by a target organ to decrease the effect of increased norepinephrine release from sympathetic nerve endings. Poorly controlled norepinephrine release could also give rise to excessive activation of α₁-adrenoceptors in the prefrontal cortex impairing its function. Other noradrenergic terminal areas in the central nervous system may be similarly affected. These findings suggest that the noradrenergic system is hyperactive in SHR, particularly in response to stress, and supports the hypothesis that there is an imbalance between norepinephrine hyperfunction and dopamine hypofunction in ADHD.

Brain serotonin (5-hydroxytryptamine, 5-HT) function has been suggested to be altered in SHR 121. Higher serum testosterone, and lower amygdala serotonin content has been associated with a mutation in the non-pseudoautosomal region unique to the Y-chromosome of SHR 122. Administration of a serotonin transporter inhibitor, fenfluramine, evoked less prolactin secretion in SHR than WKY 121. Acute administration of the selective serotonin reuptake inhibitor, citalopram, reduced hyperactivity of SHR in an elevated plus-maze 123. However, there was no difference between SHR and WKY in midbrain, hippocampal, or striatal serotonin concentration or serotonin uptake kinetics 123. In addition, stressors released serotonin in the locus coeruleus of SHR and WKY rats to the same extent 124 and 5-HT_2C receptor function was reported to be unaltered in SHR compared to WKY 125. These findings do not support a role for serotonin in the aetiology of ADHD symptoms in SHR.

In addition to decreased autoreceptor-mediated inhibition of norepinephrine release from SHR prefrontal cortex slices, glutamate activation of AMPA receptors caused greater release of norepinephrine from SHR prefrontal cortex slices than WKY 126127. Glutamate is present in micromolar concentrations in the extracellular space outside the synaptic cleft and regulates tonic dopamine concentration in the extracellular fluid 128129130131132. Dopamine release is increased by activation of AMPA receptors in rat striatum 133134. Glutamate activation of NMDA receptors upregulates DRD1 function by a direct protein-protein interaction at the carboxy terminals of both receptors 135. As suggested by Seeman and Madras 136, the common defect in ADHD could be decreased extracellular dopamine levels. This deficiency could result from increased expression of DAT1, impaired dopamine synthesis, impaired release, or impaired regulation of extracellular dopamine by glutamate afferents from the prefrontal cortex, hippocampus, or amygdala 102. However, in vitro activation of AMPA receptors caused similar fractional release of dopamine from SHR and WKY nucleus accumbens core 134. Unlike WKY, glutamate-stimulated release of dopamine from the shell subdivision of SHR nucleus accumbens was significantly lower than from the core subdivision of SHR, suggesting that the shell may be particularly vulnerable to disturbances in dopamine release 134.

Neural circuits that use glutamate as a neurotransmitter are modulated by dopamine and norepinephrine. Future studies should be aimed at investigating glutamate function in the brains of SHR. In particular, measuring glutamate release in the prefrontal cortex and nucleus accumbens using in vivo microdialysis during reinforcement and extinction (hypothesized to be the major causative factors underlying ADHD 4) may provide useful information concerning afferent glutamate input to these brain areas as a result of the additional demands of such tasks.

Psychostimulants are the most effective drugs used in the treatment of ADHD and provide a powerful means to gain insight into the underlying disturbances of ADHD. d-Amphetamine and methylphenidate reduced the ADHD-like behaviour of SHR 137 [Sagvolden, unpublished; Russell, unpublished]. The increase in DRD1 density observed in SHR striatum is reversed by methylphenidate treatment suggesting that psychostimulants reduce ADHD-like behaviour of SHR by increasing dopamine activation of DRD1 518182 thereby enabling dopamine-mediated LTP and reinforcement mechanisms to take place.

Psychostimulants changed the performance of SHR in fixed-interval schedules of reinforcement of bar-presses by lengthening the delay-of-reinforcement gradient. However, WKY performance changed to a greater extent than SHR, suggesting that the effect of psychostimulant drugs was less pronounced in SHR than in WKY 138. The reduced reactivity to psychostimulants may be associated with abnormalities in DAT1 gene expression 5051 [Mill et al., unpublished] and dopamine hypofunction as a result of adaptation to increased availability of dopamine at an early stage of development with subsequent reorganization of neural mechanisms that control VTA dopamine neuron function 139. Further support for the hypothesis that regulation of midbrain dopamine neurons is altered in SHR was provided by the fact that repeated administration of methylphenidate (2.5 mg/kg) elicited locomotor sensitization i.e. increased locomotor response to the same dose of methylphenidate 3 days after cessation of treatment in Sprague-Dawley and WKY rats but not in SHR who were unaffected by the drug 140. In vitro findings provided further support, where methylphenidate released significantly less dopamine from SHR nucleus accumbens slices than WKY 58, and chronic methylphenidate treatment (3 mg/kg for 2 weeks) increased endogenous dopamine activation of DRD2 in WKY striatum but did not alter DRD2 function in SHR probably because DRD2 were already up-regulated in SHR and no longer responsive to increases in extracellular dopamine 141. These results suggest that neural circuits have been altered in SHR and that psychostimulant drugs affect SHR and WKY brains differently. This finding stresses the importance of studying animal models of ADHD as it shows that it cannot be assumed that drugs will have the same effect in children with ADHD and controls.

The SNAP-25 deficient mouse mutant coloboma (Cm/+) is of interest to ADHD because SNAP-25 polymorphisms have been associated with the disorder 153154. SNAP-25 regulates membrane trafficking and is involved in the release of all neurotransmitters as well as regulating translocation of proteins (e.g. NMDA receptor subunits) to the cell membrane. Altered expression of SNAP-25 will therefore have diffuse effects on neuronal function. The SNAP-25 deficient mouse mutant coloboma displays spontaneous hyperactivity 155 but lacks impulsiveness and has not been shown to have problems with sustained attention. Although the SNAP-25 deficient mouse does not model ADHD symptoms specifically, it may nevertheless serve as a useful model of non-specific brain dysfunction such as minimal brain disorder (MBD).

Depolarization-evoked (K⁺-evoked) release of glutamate from cortical synaptosomes is reduced in the coloboma mouse 156. DRD2 expression is increased in the VTA and substantia nigra, suggesting increased inhibition of dopamine neuron firing rate 157. Dopamine release and dopamine metabolites (DOPAC and HVA) are decreased in the striatum of coloboma mice which is consistent with decreased dopamine release and turnover 156158 i.e. a hypofunctional dopaminergic system, similar to SHR. Striatal DRD1 and DRD2 expression is unaltered in the coloboma mouse 157. Tyrosine hydroxylase expression is unaltered in VTA and substantia nigra whereas tyrosine hydroxylase and α_2A-adrenoceptor expression is increased in the locus coeruleus of the coloboma mouse 157. Noradrenergic function appears to be increased since experimental depletion of norepinephrine by DSP-4 reduces the hyperactivity of coloboma mice 159. This suggests that motor activity in coloboma mice is caused by a hyperactive noradrenergic system but the hyperactivity is not completely abolished by depletion of norepinephrine, suggesting that additional factors contribute to the mutant phenotype 159, perhaps the imbalance between noradrenergic hyperfunction and dopamine hypofunction as suggested for SHR.

Neonatal 6-OHDA-lesioned rats are not impulsive but they display hyperactivity and impaired learning in a spatial discrimination task, which improves after methylphenidate or d-amphetamine treatment 160161162163. Rat pups lesioned on postnatal day 1 displayed hyperactivity in adulthood 162. They showed an initial decrease in spontaneous motor behaviour when placed in a novel environment, but after repeated testing their activity was increased relative to controls 162. Hyperactivity was accompanied by decreased dopamine 162, increased DRD4 164, and increased serotonin transporter binding in striatum but not cerebral cortex 165. Hyperactivity was not altered by DAT1 inhibitors but was greatly reduced by DRD4 antagonists as well as inhibitors of SERT1 and NET1 160161164166. These findings suggest that psychostimulants reduce hyperactivity of 6-OHDA lesioned rats not by inhibiting DAT1 but by inhibiting norepinephrine and serotonin transporters. Inhibition of NET1 would reduce dopamine uptake into noradrenergic terminals in several brain areas including prefrontal cortex and nucleus accumbens.

DAT-knockout (DAT-KO) mice lack the gene that encodes DAT. These mice have been suggested as a model for ADHD because they are hyperactive in novel situations 167168169, have impaired extinction of responses in operant food reinforcement tasks 170. They are also impaired in learning and memory tasks 168169. Impulsiveness has not been systematically investigated in DAT-KO mice. Although the absence of DAT is an extreme model of reduced midbrain DAT binding in adolescents with ADHD 36 it also contrasts with several studies that found increased DAT in striatum of ADHD children and adults 101535. The DAT-KO mouse nevertheless provides useful information concerning the neurobiological consequences of impaired DAT function.

Released dopamine is cleared at a slow rate giving rise to a 5-fold elevation of extracellular tonic dopamine in the striatum of DAT-KO mice i.e. a hyperdopaminergic state 171. Electrically stimulated release of dopamine is decreased, suggesting that phasic release of dopamine is reduced i.e. the dopamine system is hypofunctional 171 similar to SHR and the coloboma mouse. However, unlike SHR, striatal DRD2 autoreceptors controlling dopamine synthesis are nonfunctional while DRD1 and DRD2 are downregulated by approximately 50% in the striatum of DAT-KO mice 171. Hyperactivity in the DAT1 knock-out mouse might be the result of increased dopamine tone or decreased phasic dopamine release with consequently impaired activation of postsynaptic DRD1 required for LTP (and LTD) to produce changes in synaptic strength necessary for associative learning and reinforcement of appropriate behaviour.

Whereas specific inhibitors of NET1 or DAT1 did not affect DAT-KO hyperactivity, inhibitors of the SERT1 as well as drugs that activate the serotonergic system, such as serotonin receptor agonists and serotonin precursors, dramatically reduced hyperactivity 168. DAT-KO mice provide convincing evidence that hyperactivity induced by high extracellular levels of dopamine can be reduced by enhancing serotonergic tone i.e. psychostimulants do not act via DAT1 to reduce hyperactivity in this model 168. Interestingly, antagonists of the 5-HT_2A receptor reverse the behavioural deficits of DAT-KO mice 172 and polymorphisms of the 5-HT_2A receptor gene have been associated with ADHD 173174. While this model provides invaluable insight into possible mechanisms of psychostimulant action, the relevance of these findings to ADHD is not clear since serotonin reuptake inhibitors are of limited value in treating ADHD as one of the side effects of serotonin uptake inhibitors is stimulation of motor activity 38168. However, serotonin acts on a large number of receptor subtypes each with different spatial location and behavioural effects. Evidence obtained with the DAT-KO mouse suggests that specific antagonists of the 5-HT_2A receptor deserve further investigation.

Interestingly, the DAT-knockdown (DAT-KD) mouse has been suggested as a model for obsessive compulsive disorder and Tourette's syndrome 175. DAT knockdown, achieved by reducing DAT promotor strength, reduces adult DAT expression to 10% of wild-type levels and raises extracellular dopamine levels in the striatum to 170% of wild-type controls 175. Hyperdopaminergic DAT-KD mice displayed excessive sequential stereotypy reflected as a complex serial pattern of grooming actions becoming more sequentially rigid and persistent. This type of behaviour is not characteristic of ADHD but may serve as a model for Tourette's and obsessive compulsive disorder 175. Consistent with enhanced dopamine activation of DRD1 receptors being responsible for the excessively rigid serial pattern of instinctive grooming behaviour, DRD1 agonists produced similarly enhanced sequential stereotypy of syntactic grooming chains 175. DAT-KD mice also tend to be hyperactive, to walk in perseverative straight paths, and to over-pursue certain incentive stimuli, consistent with obsessive compulsive disorder.

Rats that are selected for poor performance when trained in the 5-CSRT task provide a useful model of ADHD in that they are selected for deficient sustained attention, they show poor choice accuracy towards the end of testing sessions, and they demonstrate impulsiveness (premature responding) 176177. Methylphenidate treatment improved accuracy and reduced impulsiveness (at low doses) in poor performers 177. Poor 5-CSRT task performers are not hyperactive and therefore may serve as a model of the inattentive subtype of ADHD.

In normal animals, response accuracy is adversely affected by activation of serotonin 5-HT_1A receptors 178, while activation of 5-HT_2A receptors increases the number of premature responses, suggesting that increased serotonin tone could be responsible for impulsivity of poor performers 179. This is consistent with 5-HT_2A receptor antagonists reversing the behavioural deficits of DAT-KO mice 172.

Evidence supports a role for dopamine in regulating the level of performance in the 5-CSRT task. In normal animals, d-amphetamine-stimulated release of dopamine in the nucleus accumbens caused a dose-dependent increase in premature responding 178. Microinfusion of a DRD1 agonist into the medial prefrontal cortex selectively impaired the accuracy of attentional performance in high performers in the 5-CSRT task 180. In contrast, microinfusion of the DRD1 agonist into the medial prefrontal cortex of poor performers enhanced the accuracy of attentional performance; a low dose increased the speed of making correct responses 180. This finding once again emphasizes the need to study animal models of ADHD rather than normal animals in order to gain insight into the mechanisms that underlie the beneficial effects of drugs used to treat children with ADHD. Evidence suggests that the nervous system is altered in the 5-CSRT model of ADHD.

These results suggest that dopamine function is reduced in poor performers of the 5-CRST task and that 5-HT_2A antagonists may be beneficial in the treatment of ADHD.

Anoxia increases the risk of ADHD 181. Neonatal anoxia caused a sequence of acute and persistent neurochemical changes in rat monoaminergic systems as well as transient hyperactivity and spatial memory impairment that persisted into adulthood 182183184. Acute anoxia caused a transient decrease followed by an increase after 1 week in cerebellar norepinephrine levels 183. At the same time, serotonin levels decreased while its metabolite, 5-hydroxyindoleacetic acid (5-HIAA), increased 183. Striatal dopamine and metabolite concentrations decreased and then dopamine metabolites increased post ischaemia 183. The increase in serotonin and dopamine metabolites persisted into adulthood, suggesting that dopamine turnover is increased. Tyrosine hydroxylase mRNA levels were increased in VTA and substantia nigra of perinatally asphyxiated rats suggesting increased dopamine synthesis consistent with increased turnover. However, DRD1 and DRD2 mRNA levels were increased in the striatum suggesting impaired release of dopamine 185. These findings demonstrate the complex temporal sequence of compensatory changes that occur in monoaminergic systems following perinatal insult to the nervous system and implicate all three monoaminergic systems in spatial memory impairment.