Fig 2: The C-terminal domain of CB1R, especially the Helix 9 motif, plays a role in axonal surface polarisation.(A) Amino acid sequence of the C-terminus of rat CB1R highlighting helical motifs (Helix 8 and Helix 9) predicted from NMR spectroscopy and computational modelling. The proposed interaction domains of CRIP1a and SGIP1, and the palmitoylated cysteine residue involved in membrane association and G-protein coupling, are also indicated. A potentially post-translationally modified cysteine in H9 is highlighted. (B) Schematic of the CD4-ctCB1R chimeric proteins used. (C) Representative confocal images of hippocampal neurons showing the distribution of expressed CD4 (left), CD4-ctCB1RWT (middle), or CD4-ctCB1R?H9 (right). Upper panels for each condition show a whole cell field of view and lower panels are enlargements of axonal (a) and dendritic (d) ROIs. Green = total; magenta = surface; blue = dendrite marker (MAP2). Merge: surface to total seen as white. Scale bar = 20 µm. (D) Quantification of data represented in (C) presented as the surface polarity index (A/D ratio). CD4-ctCB1RWT strongly favoured the axonal compartment compared to CD4 alone. CD4-ctCB1R?H9 favoured the axonal compartment significantly less than CD4-ctCB1RWT. One-way ANOVA with Tukey’s post hoc test. N = three independent experiments; n = 28–33 neurons per condition. CD4 vs. WT: mean ± SEM, 0.834 ± 0.0255 vs. 1.52 ± 0.0696; N = 3, n = 30 vs. N = 3, n = 33; ****p < 0.0001. CD4 vs. ?H9: mean ± SEM, 0.834 ± 0.0255 vs. 1.09 ± 0.0562; N = 3, n = 30 vs. N = 3, n = 28; **p = 0.0050. WT vs. ?H9: mean ± SEM, 1.52 ± 0.0696 vs. 1.09 ± 0.0562; N = 3, n = 33 vs. N = 3, n = 28; ****p < 0.0001.

Fig 3: Laser confocal microphotographs of hippocampal cultures from control (A), and tbl (B) mice illustrative of MPA2 (green) and CLT (red) immunocytochemistry. (C–F), graphical representation of the fluorescence intensities of MAP2 and CLT in absolute values per image (C,E); and relative to the % of measured area (D,F). In both absolute (C,E) and relative (D,F), values, MAP2 and CLT expressions are higher in control cultures than in tbl ones. The asterisks indicate significant differences in the Student’s t test: ***p = 0.0008 (C); **p = 0.005 (D); ***p = 0.0005 (E); and ***p = 4.4E-06 (F). Bars = 50 µm. a.u., arbitrary units.

Fig 4: Partial colocalization of GABAergic and glycinergic presynaptic markers throughout adulthood. (A,B) Confocal images of GAD65/67 (red) and GlyT2 (green) labeling showing the allocation of putative GABAergic and glycinergic synaptic terminals at P25 (A) and P60 (B). Additional staining with MAP-2 shows somatic localization of inhibitory terminals on the large SBCs (blue). Maps on the right show colocalized pixels (intensity above threshold in both channels) from the composite confocal images on the left. Partial colocalization of GAD65/67 and GlyT2 is present both at P25 and P60. (C) Population data showing GAD65/67 (red) and GlyT2 (green) pixels above respective threshold at P25 (n = 40 cells) and P60 (n = 24 cells). The amount of GAD65/67- and GLYT2-positive pixels is comparable within and between the age groups (p > 0.2, ANOVA). Black bars indicate the mean ± SEM number of colocalized pixels in both age groups. Percentages showing the relative fraction of colocalized pixels within the respective bar indicate a stable proportion of terminals releasing both GABA and glycine (P25 vs. P60, p = 0.8, ANOVA). (D) Gephyrin staining (white) indicates putative localizations of the postsynaptic GABAAR and/or GlyR. Merged image on the right shows gephyrin in apposition to GlyT2-positive terminals (arrows), to GAD65/67-positive terminals (white arrowhead), and to terminals colocalizing GlyT2 and GAD65/67 (empty arrowheads). Note the terminals expressing only GAD65/67 with no gephyrin staining on the postsynaptic site of the membrane (asterisks). Scale bars = 5 µm (A–D).

Fig 5: The C-terminal domain of CB1R plays a role in internalisation in dendrites independent of H9.(A) Representative confocal images of DIV 14 primary hippocampal neurons expressing CD4?CT (left), CD4-ctCB1RWT (middle), or CD4-ctCB1R?H9 (right). Neurons were subjected to 2 hr of antibody feeding followed by stripping of surface antibody to reveal the endocytosed pool of receptors. Upper panels for each condition show a whole cell field of view and lower panels are enlargements of axonal (a) and dendritic (d) ROIs. Green = total; red = endocytosed; blue = dendritic marker (MAP2). Merge: endocytosed to total seen as yellow. Scale bar = 20 µm. (B) Quantification of data presented in (A). Both CD4-ctCB1RWT and CD4-ctCB1R?H9 were significantly more internalised in dendrites, but not in axons, than CD4 alone. Three independent experiments; n = 24–26 neurons per condition. Two way ANOVA with Sidak’s post hoc test. Axons, CD4 vs. WT vs. ?H9: mean ± SEM, 1.00 ± 0.040 vs. 1.065 ± 0.039 vs. 1.148 ± 0.048; N = 3, n = 24 vs. N = 3, n = 26 vs. N = 3, n = 24; nsp = 0.3514. Dendrites, CD4 vs. WT: mean ± SEM, 1.042 ± 0.038 vs. 1.250 ± 0.051; N = 3, n = 24 vs. N = 3, n = 26; *p = 0.0279. Dendrites, CD4 vs. ?H9: mean ± SEM, 1.042 ± 0.038 vs. 1.355 ± 0.060; N = 3, n = 24 vs. N = 3, n = 24; *p < 0.0001. Dendrites, WT vs. ?H9: mean ± SEM, 1.250 ± 0.051 vs. 1.355 ± 0.060; N = 3, n = 26 vs. N = 3, n = 24; nsp = 0.8275.