Dopaminergic Modulation of Medial Prefrontal Cortex Deactivation in Parkinson Depression

Abstract

Parkinson's disease (PD) is associated with emotional abnormalities. Dopaminergic medications ameliorate Parkinsonian motor symptoms, but less is known regarding the impact of dopaminergic agents on affective processing, particularly in depressed PD (dPD) patients. The aim of this study was to examine the effects of dopaminergic pharmacotherapy on brain activation to emotional stimuli in depressed versus nondepressed Parkinson disease (ndPD) patients. Participants included 18 ndPD patients (11 men, 7 women) and 10 dPD patients (7 men, 3 women). Patients viewed photographs of emotional faces during functional MRI. Scans were performed while the patient was taking anti-Parkinson medication and the day after medication had been temporarily discontinued. Results indicate that dopaminergic medications have opposite effects in the prefrontal cortex depending upon depression status. DPD patients show greater deactivation in the ventromedial prefrontal cortex (VMPFC) on dopaminergic medications than off, while ndPD patients show greater deactivation in this region off drugs. The VMPFC is in the default-mode network (DMN). DMN activity is negatively correlated with activity in brain systems used for external visual attention. Thus dopaminergic medications may promote increased attention to external visual stimuli among dPD patients but impede normal suppression of DMN activity during external stimulation among ndPD patients.

Figure 1

Maps thresholded at the same level of statistical significance (P < 0.001) showing medication-related variations in the response patterns among depressed and nondepressed PD patients.

Figure 2

Dopaminergic medication by depression interaction in the ventromedial prefrontal cortex (VMPFC) in PD patients; data points depict the average value across emotion categories from the 4th column of Table 2. Error bars represent the standard error of the mean.

Figure 3

Dopaminergic medication by depression interaction in the ventrolateral prefrontal cortex (VLPFC) in PD patients. Error bars represent the standard error of the mean.

Figure 4

Scatter plots illustrating the relationship between VMPFC deactivation and VLPFC activation for the emotional face processing task. Observations both on and off dopaminergic medication are included. Data points for an age-matched group of normal control subjects are added for comparison. Solid lines depict fits to data from each group separately. The dashed green lines indicate the 95% confidence interval for the linear curve fitting data from a separate group of age-matched normal controls scanned under the same protocol but not part of this study.

Figure 5

(a) Illustrating a hypothetical inverted-U dependence of frontal lobe function on dopaminergic medication associated with individual variation in baseline dopamine level. (b) Nondepressed PD patients (red) are positioned at the upstroke lower end of the curve: they reduce their level of VMPFC deactivation and increase VLPFC activation with levodopa medication. Depressed PD patients (blue) are positioned at the downstroke upper end of the curve: they increase their level of VMPFC deactivation and decrease VLPFC activation with levodopa medication. Labels a, b, c, and d indicate representative ndPD and dPD patients; open symbols denote the VMPFC and closed symbols the VLPFC brain regions; dashed lines represent the activity gap off dopaminergic medication, while solid lines represent that on levodopa.