Functional imaging provides opportunities for visualizing networks that contribute to cognitive dysfunction in the living brain. Radiotracer imaging has the potential to demonstrate specific neurotransmitter disturbances that correlate with cognitive impairment, while functional magnetic resonance imaging (MRI) has a high spatial and temporal resolution to detect changes in brain function that map onto specific components of task performance.
Given that the pathological hallmark of PD is degeneration of the ascending projections of midbrain dopaminergic neurons, the majority of functional imaging studies have concentrated on the role of frontostriatal circuitry in PD. Their findings have complemented decades of data from animal, pharmacological, and post-mortem studies emphasizing the contribution of dopaminergic disturbances to executive dysfunction in PD. Critically, the changes that underlie cognitive dysfunction have been shown to be distinct, such that executively impaired patients with early PD have differences in striatal and prefrontal activation compared with their cognitively unimpaired, but otherwise matched, counterparts . An issue that remains unresolved is whether cognitive functions associated with the frontal lobe are impaired as an indirect consequence of impaired nigrostriatal dopaminergic function or a direct consequence of impaired mesocortical dopaminergic transmission.
The concept that frontally based deficits in PD are a consequence of abnormal outflow from the basal ganglia is supported by fluorodopa ([18F]-DOPA) and [11C]-raclopride positron emission tomography (PET) data, which have demonstrated correlations between caudate dopamine depletion and executive impairments (e.g. [70, 71]). Likewise, H215O PET and functional MRI studies have shown regional reductions in blood flow in the basal ganglia in the context of preserved cortical responses during working memory tasks in PD [72, 73]. Detailed PET correlation studies have even suggested that different task requirements are lateralized within the striatum, implying that the pattern of cognitive changes manifested by a patient with PD may reflect the side of dopamine loss .
Other studies, however, suggest that dopamine disturbances within the cortex have a more prominent role, but indicate that the relationship between cortical dopamine and cognitive function is complex. [18F]-DOPA PET has showed increased dorsolateral prefrontal cortex dopamine in conjunction with reduced striatal dopamine in drug-naive patients compared with controls, possibly as a compensatory mechanism [40, 75]. These cortical changes are related to performance: reaction time in tests of sustained attention correlated positively with [18F]-DOPA uptake in the dorsolateral prefrontal cortex while performance in a test of suppressed attention correlated negatively with [18F]-DOPA uptake in the medial frontal cortex and anterior cingulate
. Performance on the Tower of London planning task and a working memory task has been associated with abnormally high blood flow as measured with H215O-PET in the right prefrontal and occipital cortices of patients off levodopa; levodopa normalized these disturbances to restore a pattern of blood flow similar to controls, and this correlated with change in performance . It has been suggested that dopamine acting within the frontal cortex enables a focusing of activity of glutamatergic output neurons, which as a result respond more efficiently .
Imaging studies in PD patients genotyped for COMT Val158Met have provided further insights into the relationship between prefrontal dopamine and executive dysfunction. Using functional MRI in patients with PD, we demonstrated that impaired performance on both the Tower of London planning task and an attentional-control task in COMT Met homozygotes was associated with reduced blood oxygen level-dependent (BOLD) activation in frontoparietal networks (Fig. 17.3) without corresponding changes in striatal activation [78, 79]. These differences in activation were proposed to reflect differences in dopamine concentrations in the cortex, where COMT is the key mechanism of dopamine clearance, and [18F]-DOPA PET imaging has now provided supporting evidence for this. A study of 20 patients with early PD demonstrated reduced dopamine turnover and higher pre-synaptic dopamine levels across several frontal cortical areas in Met homozygotes (lower COMT enzyme activity), relative to Val homozygotes, with no apparent differences in the striatum . These data collectively endorse a regionally specific effect of COMT on cortical dopamine and add weight to the idea that dopamine disturbances within the frontal cortex are capable of modulating the executive phenotype of PD.
Functional imaging also supports the concept that the inverted U-shaped relationship between dopamine levels and cognitive performance operates, at least in part, at a cortical level. Differential effects of COMT genotype on brain activation have been described in controls and PD patients on the same test of attentional control. In healthy volunteers, Val homozygotes had impaired set-formation ability and lower dorsolateral prefrontal cortex activation than Met homozygotes, while in PD patients, the opposite relationship was observed. This fits with the idea that PD patients, due to their cortical hyperdopaminergic state, sit further to the right of the inverted U-shaped curve where homozygosity for Met becomes detrimental due to presumed dopamine overload (Fig. 17.1) .
Although cortical and striatal disturbances are not mutually exclusive, the differing results of these cognitive neuroimaging studies in PD may reflect the heterogeneity observed within the patient population. The patients who have been included in studies vary over multiple dimensions that are known to impact on cognitive performance as well as brain function, including age, gender, dopaminergic medication, and, as we are now aware, genetic variation. It is also apparent that the relative impacts of cortical and striatal dopamine on performance depend on the specific demands of the task. It is theorized that cognitive stability benefits from increases in tonic prefrontal cortex dopaminergic transmission and reductions in phasic striatal dopaminergic transmission. In contrast, cognitive flexibility benefits from potentiated phasic striatal dopaminergic transmission and reduced tonic dopaminergic transmission in the prefrontal cortex. In addition, dorsal and ventral frontostriatal circuits contribute to the control of distinct types of representations . By considering these differences in future studies we may be able to resolve some of the apparent inconsistencies described in PD.
Fig. 17.3 Functional MRI was used to measure the effect of COMT Val158Met genotype on brain activation during performance of the Tower of London planning task in patients with early PD (n = 31). (A) BOLD activation during planning versus a control task (subtracting) in the whole cohort. Areas of signal change above a threshold of p = 0.05 after false discovery rate correction for whole brain volume are shown rendered onto a canonical brain image. PPC, posterior parietal cortex; DLPFC, dorsolateral prefrontal cortex. (B) Activity in val (n = 16) versus Met (n = 15) homozygotes in regions of interest as indicated in (A). There was significant underactivation of the frontoparietal network activated by the planning task in Met versus val homozygotes, as well as an impairment in behavioural performance (not shown).
Republished with permission of J Neurosci, from Catechol O-methyltransferase val158met genotype influences frontoparietal activity during planning in patients with Parkinson's disease, Williams-Gray CH, Hampshire A, Robbins TW, Owen AM, Barker RA., 27:4832-8 2007; permission conveyed through Copyright Clearance Center, Inc.