That study was published in a journal that is owned by Nature. Nice find, OP.
http://www.ncbi.nlm....les/PMC4462609/
Subjects
This study included 20 healthy male controls (38±8 years of age, body mass index 26±3; years of education 14±2) recruited through advertisements in local newspapers. Exclusion criteria included consumption of more than two caffeine beverages per day, current or past psychiatric disease as per DSM IV including any substance use disorder (smokers were excluded); past or present history of neurological, cardiovascular or endocrinological disease; history of head trauma with loss of consciousness greater than 30 min; and current medical illness. Seventeen of the participants reported that they did not drink coffee (or caffeinated beverages), one reported one cup a day and two reported two cups a day. Written informed consent was obtained from all the subjects and the studies were reviewed and approved by the Institutional Review Board at Stony Brook University Medical Center.
Discussion
Here we show that caffeine increases D2/D3R availability in striatum (evidenced as increases in BPND in dorsal putamen and VS) in a group of healthy controls with low levels of daily caffeine intake. These findings are consistent with findings from a prior PET [11C]raclopride study done in a small group of subjects (eight habitual coffee drinkers) that also reported increases in D2/D3R availability in striatum with caffeine (200 mg).24 The findings from these two studies thus suggest that caffeine at doses typically consumed by humans might enhance DA signaling by increasing D2/D3R levels or their affinity rather than by increasing DA release in the striatum.
Here we interpret our results of increases in BPND (in BPND availability) with caffeine to suggest that they reflect increases in D2/D3R levels rather than reflecting decreases in endogenous DA, which is the way that typically increases in BPND are interpreted (reduced competition from DA to bind to D2/D3R). The reasons for this interpretation follow. First, it is recognized that alerting drugs (amphetamine, methylphenidate and modafinil) increase DA release in the striatum.3, 25, 36 Second, clinical studies have shown that the DA increases in striatum induced by stimulant drugs are associated with increases in alertness.5 Finally, preclinical studies have shown that the increases in striatal DA induced by stimulants and modafinil is necessary for their wake-promoting actions.6 Thus, if caffeine had reduced DA in the striatum, this would have resulted in an increase in tiredness and sleepiness instead of the increases in alertness observed after caffeine administration. Our interpretation that the increases in striatal D2/D3R availability in VS with caffeine reflect an increase in D2/D3R levels is also consistent with our findings that downregulation of D2/D3R in VS after sleep deprivation is associated with reduced alertness.5
Striato-pallidal neurons adjust their excitability by changing D2R levels in the membrane.37 Thus, D2R downregulate with DA stimulation38 and upregulate with reduced DA signaling.39, 40 DA stimulation of D2R triggers their internalization,38 which can then be recycled or degraded.38, 41 Internalization of D2R is regulated by A2AR,42 agonists facilitate its internalization through the binding of β-arrestin 2 to A2AR-D2R receptor heteromers43 whereas A2AR antagonists interfere with D2R internalization in striatal neurons.44Thus, caffeine might interfere with a tonic A2AR-dependent internalization of D2R mediated by endogenous adenosine, which could contribute to its psychostimulant effects.14, 19, 45, 46 Indeed, our findings along with those previously reported showing that caffeine increased D2R availability in striatum,24 support this interpretation. As caffeine modulates DA signaling, in part, by its antagonism of A2AR,47 caffeine-induced D2R increases in striatum would be consistent with caffeine's antagonism of A2A-mediated D2R internalization. Indeed, A2A receptor knockout mice show increased D2R levels in striatum;48 though we cannot necessarily equate the chronic state of a knockout with the effects from acute caffeine exposure.
However, regardless of the mechanism responsible for the increases in striatal D2/D3R availability, our results indicate that in humans, caffeine at the doses typically consumed, does not increase DA in the striatum. This is consistent with findings from microdialysis studies in rodent showing that caffeine (0.25–5 mg kg−1 intravenously or 1.5 to 30 mg kg−1 intraperitoneally) did not increase DA in the nucleus accumbens,22, 23 though a study reported increases with a large (10 mg kg−1 intraperitoneally) but not a lower caffeine dose (3 mg kg−1 intraperitoneally).21 Thus, on the basis of the current and prior findings24and the preclinical results, caffeine at doses that are relevant to human consumption does not appear to increase DA in the nucleus accumbens. As the ability of drugs of abuse to increase DA is necessary for their rewarding effects and for the neuroadaptations associated with the addiction phenotype,49 this could explain why caffeine does not produce the compulsive administration and the loss of control that characterizes addiction.50
Caffeine-induced increases in D2/D3R in VS were associated with increases in alertness. This association between alertness and D2/D3R replicates our previous findings with sleep deprivation but in the opposite direction, in which we showed that the decreases in D2/D3R availability in VS with sleep deprivation were associated with reductions in alertness.5 In the prior PET study, caffeine-induced increases in striatal D2/D3R availability were associated with reduced tiredness.24 Thus this provides evidence that enhanced signaling through D2/D3R in striatal regions might enhance alertness or decrease tiredness, whereas reduced signaling might decrease alertness or increase fatigue.
Study limitations
Traditionally, increases in D2/D3R availability with [11C]raclopride, as observed here, have been interpreted to reflect decreases in DA release. Instead, our model leads us to interpret them as increases in D2/D3R levels and/or increases in affinity. However, our model cannot rule out the potential confound that more than one factor could be affecting the binding of [11C]raclopride. In this respect, preclinical experiments that use more selective compounds should be performed to investigate whether caffeine's effects on [11C]raclopride binding reflect changes in the expression or in the affinity of D2/D3R and whether these effects reflect caffeine's antagonism at A2AR. Also because [11C]raclopride binds to both D2R and D3R,51 we cannot distinguish whether caffeine-induced increases in striatal BPND reflects only increases in D2R or also in D3R. However, in putamen where the relative density of D3R is much lower than that of D2R,52 the effects of caffeine are likely to reflect D2R. Another potential confound in our study is that caffeine significantly reduces cerebral blood flow,53 which could interfere with the BPND measures as cerebral blood flow effects differ between cerebellum and striatum.54 However, because caffeine decreases cerebral blood flow in striatum to a greater extent than in cerebellum,54 this would lead to decreases in striatal BPND, whereas we showed the opposite; that is increases in striatal BPND with caffeine, indicating that our findings are not due to caffeine-induced changes in cerebral blood flow. Though the raclopride PET method cannot distinguish between presynaptic and postsynaptic D2/D3R, the fact that caffeine is an antagonist at A2A receptors, which are expressed in medium spiny neurons expressing D2R but not in DA neurons lead us to presume that the effects are postsynaptic. Another confound in our studies is the order effect as placebo was always given 2 h before caffeine. However, studies that have evaluated test–retest reproducibility for raclopride binding (including ours)55, 56 have reported no significant differences between measures even when the repeated measures were performed on the same day57 as per the current study, indicating that the order effect is unlikely to account for our findings. We are unable to assess if the participants were able to determine if they received caffeine or placebo as we did not query them at the end of the study. Finally, we did not collect blood samples for epinephrine and norepinephrine, which are increased by caffeine.58 Thus, we cannot rule out the contribution of caffeine's effects in the autonomic system on the behavioral effects of caffeine. Nonetheless, the significant association between increases in D2R availability in VS and alertness indicates that caffeine's effects on D2R signaling contribute to its alerting effects.
____
*going out for coffee*