Fig. 1: Dentate granule cell input is required for intact phase precession in CA3.
a Intact phase precession in control CA3 cells. Left, schematic of the major theta-modulated excitatory inputs to CA3. Right, Two example CA3 cells and their spatial firing patterns (gray lines, path; black dots, spike locations; red dots, spike locations of an example spike train; scale bar, 50 cm), an example spike train and corresponding LFP trace (red ticks, spikes; solid line, 6–10 Hz filtered LFP; gray line, raw LFP; scale bars, 250 ms and 500 μV), and phase-versus-normalized distance plot (black dots, spikes; red dots, spikes of example train, 1 cycle = 360°). Solid green lines in the plot indicate significant phase precession (p < 0.05; circular-linear regressions). b Data as in (a) but for CA3 cells without DG inputs. Dashed green lines indicate the lack of phase precession (p > 0.05; circular-linear regressions; LFP scale bars, 250 ms and 500 μV; Path scale bar, 50 cm). c Phase precession slopes and proportions of phase precessing cells. One slope per cell was obtained by pooling the spikes of all trains and by fitting a circular-linear regression to this pool (slope-by-cell analysis). The median magnitude of the slopes (violin plots; n = 84 control (CTRL) and 68 DG lesion (LESION) CA3 cells, z-statistic = − 3.62, p = 2.8 × 10-4, MW test) and the proportion of negative slopes (bar plots; only negative slopes, χ2 = 8.34, p = 0.0039; negative and significant slopes, χ2 = 8.43, p = 0.0037, chi-square test) were reduced by the DG lesion. d Slopes for all trains from control (CTRL(DG), left) and DG-lesion (LESION(DG), right) CA3 cells (slope-by-train analysis). Each row depicts the slope values from each of the trains of one cell (blue ticks), and cells are sorted from top to bottom by their trains’ median slope (black tick when negative, purple tick when positive). Shaded regions correspond to negative values. e Phase precession slopes (violin plots; n = 84 control and 68 DG lesion CA3 cells, z-statistic = − 4.39, p = 1.2 × 10−5, MW test) and proportions of phase precessing cells (bar plots; χ2 = 4.28, p = 0.038, chi-square test) from the slope-by-train analysis. For analysis of proportions, a cell was considered phase precessing if the median slope was negative. Violin plots in panels (c, e): Outline, distribution; shading, negative slopes (inner shading in c, negative and significant slopes); error bars, 1.5 times the interquartile interval above the third and below the first quartile. *p < 0.05,**p < 0.01, ***p < 0.001. Source data are provided as a Source Data file.
