Search

Additional file 1: Figure S1. The nbs used and their staining procedure. For this study, different nb constructs were used. The 13A7 nb (P2X7-specific nb; see patent WO/2013/178783; [10]) was fused to the hinge, CH2, and CH3 domains of mouse IgG2c, resulting in a heavy chain format (nb A), whereas 1c81 (P2X7-specific nb; see patent WO/2013/178783 [10]) was dimerized and fused to the albumin-specific nb Alb8 (nb B). To prevent aggregation at high concentrations, we modified nb B (nb B-mod). For recognition of these nbs in FACS, we used the following staining protocols: After binding of nb A to P2X7, cells were stained with a biotinylated anti-mouse IgG2c-fused antibody followed by streptavidin BV421 conjugation. After binding of nb B-mod, cells were stained with an anti-Alb8-nb fused to the mouse IgG1 heavy chain backbone, followed by an anti-mouse IgG1 antibody conjugated with BV421. Figure S2. Schematic representation of icv surgery. The cranial burr hole was drilled 1.1 mm lateral and 0.5 mm posterior to bregma. Nbs were injected 2.3 mm deep directly into the left ventricular system. As a proof-of-concept, 2 μl of 5% Evans blue was injected into the ventricular system. Two hours after injection, Evans blue was distributed equally in the whole ventricular system. Figure S3. Gating strategy for brain resident microglia. Flow cytometry of brain cells. Three minutes before euthanasia, a CD45-fluorochrome-conjugated antibody was injected intravenously to separate intravascular from intraparenchymal cells. Brain resident microglia were identified as CD45intermed CD11bhigh cells, which were not labeled by the intravenously injected CD45-fluorochrome conjugated antibody. Figure S4. In vitro calibration of brain homogenates from pmeLUC mice. The panel shows the in vitro calibration of brain homogenates from pmeLUC mice, showing the luminescence response to the addition of exogenous ATP and the obliteration of luminescence in the presence of the ATP-hydrolyzing enzyme apyrase. With all the caveats due to the in vitro setting, this calibration suggests that the eATP concentration in the stroked brain may reach hundreds of micromoles/L. Figure S5. Negative control. In GFP-negative littermates, no P2X7 signal was found. Figure S6. High doses of P2X7-blocking nbs are necessary to cross the BBB efficiently. C57BL6J mice received different amounts of nb A intravenously. Four hours after iv injection mice were sacrificed and nbs bound to microglia were labeled by FACS (See MM). Full coverage of P2X7 was reached with 3200 μg, where 1000 μg and lower concentrations achieved less than 60% occupancy of microglial P2X7. These FACS data are representative images of two independent cohorts of 5 mice each. Figure S7. Low amounts of icv injectedP2X7-blocking nb B-mod showed high P2X7 occupancy on microglia. C57BL6J mice received different amounts of P2X7-specific nbs intracerebroventricularly. 18 h after icv injection mice were sacrificed and nbs bound to microglia were labeled by FACS (See MM). Low amounts of P2X7 blocking nb B-mod were needed to show almost full occupancy of microglial P2X7 receptor. These FACS data are representative images of two independent cohorts of 5 mice each. Figure S8. Icv injection of P2X7-blocking nb B-mod resulted in a long time P2X7 occupancy on microglia. C57BL6J mice received 30 μg P2X7-specific nb-B mod intracerebroventricularly. Mice were sacrificed at different time points after icv injection and nbs bound to microglia were labeled by FACS (See MM). After 2.5 h microglial P2X7 showed nearly complete coverage by P2X7 nbs. This high occupancy started to decrease 14 days after the icv injection, but still nearly 40% of microglial P2X7 was occupied after 21 days post icv injection. These FACS data are representative images of two independent cohorts of 6 mice each.

International audience; 11 ATP signaling and surface P2X4 receptors are upregulated selectively in neurons and/or glia in various CNS disorders 12 including anxiety, chronic pain, epilepsy, ischemia, and neurodegenerative diseases. However, the cell-specific functions of 13 P2X4 in pathological contexts remain elusive. To elucidate P2X4 functions, we created a conditional transgenic knock-in 14 P2X4 mouse line (Floxed P2X4mCherryIN) allowing the Cre activity-dependent genetic swapping of the internalization 15 motif of P2X4 by the fluorescent mCherry protein to prevent constitutive endocytosis of P2X4. By combining molecular, 16 cellular, electrophysiological, and behavioral approaches, we characterized two distinct knock-in mouse lines expressing 17 noninternalized P2X4mCherryIN either exclusively in excitatory forebrain neurons or in all cells natively expressing P2X4. 18 The genetic substitution of wild-type P2X4 by noninternalized P2X4mCherryIN in both knock-in mouse models did not 19 alter the sparse distribution and subcellular localization of P2X4 but increased the number of P2X4 receptors at the surface of 20 the targeted cells mimicking the pathological increased surface P2X4 state. Increased surface P2X4 density in the 21 hippocampus of knock-in mice altered LTP and LTD plasticity phenomena at CA1 synapses without affecting basal 22 excitatory transmission. Moreover, these cellular events translated into anxiolytic effects and deficits in spatial memory. Our 23 results show that increased surface density of neuronal P2X4 contributes to synaptic deficits and alterations in anxiety and 24 memory functions consistent with the implication of P2X4 in neuropsychiatric and neurodegenerative disorders. 25 Furthermore, these conditional P2X4mCherryIN knock-in mice will allow exploring the cell-specific roles of P2X4 in 26 various physiological and pathological contexts. 27