Administration of the direct dopamine receptor agonist APO to healthy individuals resulted in increased plasma GH levels and disrupted performance on the AX-CPT as predicted. APO improved PPI in prepulse trials of low intensity (75/105 dB) in subjects with poor baseline sensorimotor gating and disrupted PPI in high-intensity prepulse trials (85/105 dB).
Elevated levels of plasma GH after administration of APO are in line with previous studies that have shown GH responses after comparable doses of APO in healthy controls and patients suffering from schizophrenia, major depression, and alcohol dependence (Schmidt et al. 2001
). The dose of APO currently administered appears to be sufficient to stimulate central dopamine receptors.
The observed deterioration in cognitive performance after APO is in line with other studies, showing decreased cognitive performance after administration of APO and the dopamine agonist pramipexole (Hamidovic et al. 2008
; Schellekens et al. 2009
). Since dopamine receptors in the prefrontal cortex (PFC) are located both at the presynaptic and postsynaptic neuron, one could speculate that the disruptive effect on cognitive performance is caused by decreased dopamine functioning after stimulation of presynaptic dopamine receptors or excessive dopamine functioning after stimulation of postsynaptic dopamine receptors (al-Tikriti et al. 1992
; Wang and Pickel 2002
In contrast with the hypotheses, participants with poor baseline cognitive performance were most sensitive to the deteriorating effects of APO on cognitive performance. That individuals with presumed lower levels of prefrontal dopamine functioning are more sensitive to the detrimental effects of APO on cognitive performance suggests that stimulation of presynaptic dopamine receptors, and subsequently reduced dopamine neurotransmission in the PFC is involved. Further studies are needed to explore the mechanisms underlying the effect of APO on cognitive performance.
The observation of PPI disruption in high-intensity prepulse trials after administration of APO, versus PPI enhancement in low intensity prepulse trials, is in line with previous findings with the dopamine agonists pergolide and 4-propyl-9-hydroxynaphthoxazine in rats (Martin-Iverson and Else 2000
; Swerdlow et al. 2001a
). This paradoxical finding has been hypothesized to result from either agonist action on dopamine receptors located pre and post synaptically, or at different parts in the startle neurocircuitry, or from agonist action on different neurochemical subclasses of dopamine receptors, e.g., D3 and D4 (Martin-Iverson and Else 2000
; Swerdlow et al. 2001b
Although dopamine receptors are located in several different brain areas, there is substantial evidence from preclinical research suggesting that the final pathway in the disruptive effect of APO on PPI is dependent on stimulation of dopamine receptors in the nucleus accumbens. Local application of dopamine agonists in the nucleus accumbens, but not in the neostriatum or the orbitofrontal cortex, reduces PPI (Swerdlow et al. 1992
; Wan et al. 1994
). Interestingly, within the prefrontal cortex, dopamine antagonists rather than agonists appear to decrease PPI (Ellenbroek et al. 1996
; Zavitsanou et al. 1999
). More importantly, Swerdlow and colleagues also showed that the PPI disruptive effect of systemically applied dopamine agonists can be blocked by blocking dopamine transmission in the nucleus accumbens (Swerdlow et al. 1990
). These data suggest that the effects of dopamine agonists on PPI are predominantly mediated via the mesolimbic, nucleus accumbens, and dopamine receptors. However, it cannot be ruled out that the disruptive effect of APO on PPI may be mediated by dopamine receptors in the medial prefrontal cortex (Broersen et al. 1999
; Lacroix et al. 2000
At low prepulse intensities, APO-improved sensorimotor gating in participants with suboptimal baseline levels of PPI with presumed suboptimal mesolimbic dopamine functioning. This suggests that stimulation of postsynaptic dopamine receptors and subsequently increased mesolimbic dopamine neurotransmission is involved. In contrast, baseline levels of PPI did not modify the response to APO in high-intensity prepulse trials. PPI on high-intensity prepulse trials is rather robust and likely less sensitive to the effect of baseline PPI on APO sensitivity. Yet, at low prepulse intensities, PPI may be dependent on a more narrow optimum of mesolimbic dopamine functioning.
The observed negative correlation between baseline cognitive performance and baseline PPI on low intensity prepulse trials supports a previously suggested reciprocal relation between prefrontal and subcortical dopamine neurotransmission (Kellendonk et al. 2006
; Wilkinson 1997
). Participants with high baseline cognitive performance, i.e., presumed optimal prefrontal dopamine functioning showed impaired levels of sensorimotor gating, i.e., presumed suboptimal mesolimbic dopamine functioning. Moreover, subjects sensitive to deteriorating effects of APO on cognitive performance appeared to be less sensitive to the PPI enhancing effects of APO. This again supports the idea that whereas cognitive performance is more related to cortical dopaminergic systems, PPI is more mediated via mesolimbic dopaminergic systems.
Yet, it has previously been suggested that sensorimotor gating is a pre-attentional function, important for the integrity of higher order cognitive processes (Braff and Geyer 1990
). Previous studies have indeed shown levels of PPI to correlate with attention and planning ability (Bitsios and Giakoumaki 2005
; Csomor et al. 2008
; Giakoumaki et al. 2006
). These findings merit further investigation with respect to the relation between sensorimotor gating and higher order cognitive processes.
Several additional issues regarding PPI need to be considered. First, the number of excluded startle trials in this study (20%) is high compared to literature (5%) (Bitsios et al. 2005
; Braff et al. 1992
; Liechti et al. 2001
; van der Linden et al. 2006
). Moreover, the number of excluded trials increased over consecutive startle sessions. This may result from both habituation and weariness of the participants due to repeated measurement sessions. Though there was no effect of treatment on the number of excluded trials, habituation effects may have influenced the observed effects of APO on PPI. Importantly, we used a low dose of APO (0.005 mg/kg s.c. or 2 mg s.l.) compared to animal literature (e.g., 0.5 mg/kg). Direct translation of the current findings to animal findings is, therefore, limited.
The effect of APO on the central outcome measures closely follows the kinetics of APO. In line with our hypotheses, we found the best correlation between central outcome measures and calculated APO levels in the second compartment (CSF) after s.c. administration. APO levels in the second compartment show an S-curve relation with central outcome measures, whereas plasma APO concentrations (first compartment) show a poor correlation due to a large counter-clock hysteresis curve. The use of two-compartmental pharmacokinetic/pharmacodynamic (PK/PD) modeling may provide further insight in true pharmacodynamic differences in dopamine receptor sensitivity between groups taking into account pharmacokinetic differences. Since correlation coefficients were rather low, factors other than APO levels may contribute to the observed changes in central outcome measures like weariness, motivational decline, habituation to the acoustic startle, and a learning effect on the AX-CPT. In addition, PK/PD analyses were based on a limited number of participants. Future studies are needed to confirm the current findings.