Repeated Caffeine Intake Suppresses Cerebral Grey Matter Responses to Chronic Sleep Restriction in an A1 Adenosine Receptor-Dependent Manner.

Evidence has shown that both sleep loss and daily caffeine intake can induce changes in grey matter (GM). Caffeine is frequently used to combat sleepiness and impaired performance caused by insufficient sleep. It is unclear 1) whether daily use of caffeine could prevent or exacerbate the GM alterations induced by chronic sleep restriction, and 2) whether the potential impact on GM plasticity depends on individual differences in the availability of adenosine receptors, which are involved in mediating effects of caffeine on sleep and waking function. In this double-blind, randomized, controlled study, 36 healthy adults (aged 28.9 {+/-} 5.2 y/o; 15 females; habitual daily caffeine intake < 450 mg; 29 homozygous C/C allele carriers of the A2A adenosine receptor (A2AR) gene variant rs5751876 of ADORA2A) underwent a 9-day laboratory visit consisting of one adaption day, 2 baseline days (BL), 5-day sleep restriction (CSR, 5 h time-in-bed), and a recovery day (REC) after an 8-h sleep opportunity. Nineteen participants received 300 mg caffeine in coffee through the 5 days of CSR (CAFF group), while 17 matched participants received decaffeinated coffee (DECAF group). We measured the GM morphology on the 2nd BL Day, 5th CSR Day, and REC Day. Moreover, we used [18F]-CPFPX PET to quantify the baseline availability of A1 adenosine receptors (A1R) and their relation to GM plasticity. The voxel-wise multimodal whole-brain analysis on T1-weighted images controlled for variances of cerebral blood flow indicated a significant interaction between caffeine and CSR in four brain regions: 1) right temporal-occipital region, 2) right thalamus, 3) left dorsolateral, and 4) dorsomedial prefrontal region. The post-hoc analyses indicated increased GM intensity in the DECAF group in all four regions but decreased GM in the thalamus as well as dorsolateral and dorsomedial prefrontal regions in the CAFF group after sleep restriction. Furthermore, lower baseline subcortical A1R availability predicted larger reduction in the CAFF group after CSR of all brain regions except for the caffeine-associated thalamic reduction. In conclusion, our data suggest an adaptive upregulation in GM after 5-day CSR, while concomitant use of caffeine instead leads to a GM reduction. The lack of consistent association with individual A1R availability may suggest that CSR and caffeine affect GM plasticity predominantly by a different mechanism. Future studies on the role of adenosine A2A receptors (ADORA2A) in CSR-induced GM plasticity are warranted.