Figure 4.
Theta-Coherent Synaptic Excitation in Hippocampal GCs in Awake Rats In Vivo
(A) Recording of LFP in an awake rat. Top: minimally filtered trace (150 Hz low-pass; green); middle: same trace filtered in the theta range (3–8 Hz; red); bottom: same trace filtered in the gamma range (30–90 Hz; red).
(B) Top: frequency-time representation of normalized power of the unfiltered LFP recording (same experiment shown in A) on a compressed timescale. Data from a representative experiment (top: 0–100 Hz frequency range; middle: 0–10 Hz frequency range; dashed lines indicate the time interval corresponding to the traces in A) are shown. Normalized power is color coded (calibration bar on the right). Notice the presence of multiple theta and gamma power peaks along 2 min of continuously recorded LFP activity. Bottom left: average power spectrum (green, single experiments; black, average; vertical dashed lines indicate the theta band). Bottom right: fraction of experimental time in which the ratio of theta to nontheta power is >1. Note that theta rhythm is a predominant form or activity under our experimental conditions.
(C) Average cross-frequency coherence analysis of LFP activity. Note the existence of coupling between LFP amplitude envelope at 30–90 Hz and LFP phase at 3–10 Hz. Coupling strength is color coded (calibration bar on the right).
(D) Simultaneous recording of EPSCs and LFP in awake rats (same experiment as in A). EPSCs were recorded in the WC voltage-clamp configuration at –70 mV, close to the GABAAR reversal potential. Green, LFP recording; black, WC recording.
(E) Top: average coherence between EPSCs and the LFP. To evaluate the statistical significance of coherence, we compared the original data (blue) to shuffled data (brown). Shaded areas indicate SEM. Control data were significantly different from shuffled data (p < 0.05; Kruskal-Wallis test). Bottom: summary bar graph of maximum coherence and corresponding frequency.
(F) Phase relationship between EPSCs and LFP during a theta cycle. Left: polar plot illustrates distribution of onset points of EPSCs detected by deconvolution. As the LFP recording electrode was located in the molecular layer of the dentate gyrus, the trough of the theta cycle was taken as a reference point for phase measurement (0°). Distribution of angular deviations differed significantly from a uniformity (p < 0.005; Rayleigh test; see Experimental Procedures). Data from a representative experiment are shown. Right: angular lag of EPSCs to LFP trough. All experiments were performed in awake rats (13 simultaneous LFP-WC recordings; LFP analysis in A–C, combined LFP-WC analysis of D–F). In summary graphs, bars indicate mean ± SEM, circles represent data from individual experiments. See also Figures S4 and S5.