Fig. 3: Aberrant functional rewiring and neuronal ensembles alongside globally reduced connectivity links in CA1-network under GluN1-Ab.
A Overall reduction in node degree under GluN1-Ab over different timescales, quantified based on STTC (top) and PCC (bottom) indices. Same format as in Fig. 2F,G. B Circular graphs with links (connections) and nodes (putative neurons [PNs]) of two example FOVs. Nodes were ordered in descending manner based on the ND of PNs, and links (unweighted) were drawn based on STTC matrices (tiling window of 0.2 s). To ease visualization, only links with a STTC above 95th percentile (separately for each FOV) were shown. C Functional connectivity patterns of example graphs in (B). Each dot is a color-coded node (PN) based on its betweenness centrality. To ease visualization of hub-like PNs (darker dots), the values were normalized to the maximum betweenness of the two FOVs. Line thickness encodes the STTC-based connection strengths, normalized to maximum STTC for each FOV separately. D Distribution of node degree normalized to the number of nodes per network (left, tiling window of 0.2 s), its Lorenz curve (middle), and the corresponding Gini coefficients at different timescale. Lorenz curves demonstrate the cumulative proportion of node degrees against the cumulative proportion of PNs rank-ordered by their node degree. The line of equality represents the case where all PNs have equal node degrees. Gini quantifies the deviation from equality. E Similarly to (D), but for betweenness. F Increased clustering coefficient of PNs under GluN1-Ab. G Example FOVs with detected PN ensembles. For clarity, ensemble overlaps were not shown. H Rastergram of the FOV shown in (G) under GluN1-Ab, with color-coded and re-arranged CaT-trains of PNs based on their affiliated ensemble. I Number of detected PN ensembles (top) and the ratio of PNs in ensembles to unaffiliated PNs (bottom), per FOV. J Intracellular recording of sEPSCs from CA1-PNs in hippocampal slices. Left, schematic of recording configuration. Created in BioRender; Geis, C. https://BioRender.com/9qs1fwa (2026). Middle, distribution of sEPSC amplitude. Right, the median of all sEPSC amplitudes (top) and those above 95th percentile (bottom), computed separately per neuron. K Stronger long-term depression (LTD) under GluN1-Ab. Left, schematic of recording configuration. Created in BioRender; Geis, C. https://BioRender.com/9qs1fwa (2026). Right, slope of field EPSP relative to that of baseline period (gray). Yellow-colored period was analyzed for between-group difference after LTD induction. Curves represent mean ± SEM. Boxplots show median and interquartile range; dots represent individual mice, cells, or slices. Sample sizes: (A, D–F, I) n = 9 mice/group (total cells: 4279 Ctrl-Ab, 3200 GluN1-Ab); (J) n = 11 (Ctrl-Ab) / 14 (GluN1-Ab) cells; (K) n = 16 (Ctrl-Ab) / 21 (GluN1-Ab) slices. Statistical comparisons: two-sample t-tests (E [left], J [top]), Mann-Whitney U tests (D [left], J [bottom]), permutation tests (A, D [right], E [right], F, I), or mixed-model analysis (K); see Supplementary Table 1.
