Fig. 7: Adenosine-A_2ARs mediate the 40 Hz light flickering-induced increase of the glymphatic flow.

a–c Representative western blot images (a, out of 3 similar experiments) and quantification (b, c) of A_2AR, A₁R, and β-tubulin in whole brain extracts after exposure to normal light or 40 Hz light flickering (n = 6 mice/group, mean ± SEM in the bar graphs, **P < 0.01, ns, not significant, unpaired Student’s t-test). d Representative photographs of fluorescence in coronal brain sections collected 30 min after injection of the Y39-1 fluorescent tracer in the cisterna magna of WT mice (top two rows) or A_2AR-KO mice (bottom two rows), previously exposed either to normal light or to a 30 min period of 40 Hz light flickering; numbers indicate the anteroposterior distance from bregma in mm (scale bars, 1 mm). e, f Quantification of intracisternally injected Y39-1 MPI (in arbitrary units, a.u.) and of the fluorescent area (expressed as % of section area), showed that exposure to 40 Hz light flickering increased the glymphatic influx in WT mice, but not in A_2AR-KO mice; analysis was performed on six sections per animal (n = 5–6 mice/group, mean ± SEM in the bar graphs, *P < 0.05, ns, not significant, one-way ANOVA with Tukey’s multiple comparison test). g Representative images of laser speckle flowmetry showing that 40 Hz light flickering increased cortical cerebral blood flow in WT mice, but not in A_2AR-KO mice, as quantified in h, i (n = 6 mice/group, mean ± SEM in the bar graphs, **P < 0.01, ns, not significant, paired Student’s t-test). j, k Representative photographs (scale bars, 50 μm) of the uptake by cultured cortical astrocytes (nuclei stained blue with Hoescht) of 5 kDa FITC-dextran (j, green) and 45 kDa TRITC-dextran (k, red), respectively, either in the absence of drugs (control, CTRL) or in the presence either of the AQP4 inhibitor TGN-020 (6 µM) or of the A_2AR antagonist SCH58261 (50 nM) or of both drugs. Quantification of the uptake by cultured astrocytes of the FITC/TRITC-tracers showed that the clearance of these tracers was decreased by the selective AQP4 inhibitor TGN-020, an effect reverted by the selective A_2AR antagonist, SCH58261 (data are mean ± SEM in the bar graphs, **P < 0.01, *P < 0.05, Student’s t-test). l, m The uptake by cultured astrocytes of the FITC/TRITC-tracers was enhanced by the selective AQP4 activator TGN-073 (10 µM), an effect blunted by the A_2AR antagonist SCH58261 (50 nM) (data are mean ± SEM, *P < 0.05, one-way ANOVA with Tukey’s multiple comparison test). n Representative western blot (out of 3 similar experiments) showing that the pull-down of AQP4 from cultured astrocytes retrieved A_2AR immunoreactivity (40 kDa and 42 kDa) (third lane from left) compared to non-manipulated membrane extracts (input, first lane from left) and this did not occur with a control IgG antibody (fifth lane from left), indicating that A_2AR co-immunoprecipitated with AQP4.