Theta and beta oscillatory dynamics in the dentate gyrus reveal a shift in network processing state during cue encounters

Abstract

The hippocampus is an important structure for learning and memory processes, and has strong rhythmic activity. Although a large amount of research has been dedicated toward understanding the rhythmic activity in the hippocampus during exploratory behaviors, specifically in the theta (5-10 Hz) frequency range, few studies have examined the temporal interplay of theta and other frequencies during the presentation of meaningful cues. We obtained in vivo electrophysiological recordings of local field potentials (LFP) in the dentate gyrus (DG) of the hippocampus as rats performed three different associative learning tasks. In each task, cue presentations elicited pronounced decrements in theta amplitude in conjunction with increases in beta (15-30 Hz) amplitude. These changes were often transient but were sustained from the onset of cue encounters until the occurrence of a reward outcome. This oscillatory profile shifted in time to precede cue encounters over the course of the session, and was not present during similar behaviors in the absence of task relevant stimuli. The observed decreases in theta amplitude and increases in beta amplitude in the DG may thus reflect a shift in processing state that occurs when encountering meaningful cues.

Keywords: beta; dentate gyrus; hippocampus; oscillations; theta.

Figure 1

Decreases in theta (4–12 Hz) amplitude and increases in beta (15–30 Hz) amplitude in response to conditioned reinforcement in a circular track paradigm. (A) Upper: Raw LFP trace as the rat approaches and stops at the reward location. The zero time point indicates a stop at the reward location. Middle: Instantaneous theta amplitude (blue) or beta amplitude (red) during the same time interval as the raw LFP trace. Lower: Gabor spectrogram during the same time interval as the raw LFP trace. (B) Cartoon schematic of the circular track apparatus with up to three example reward locations shown. (C) Mean velocity during intervals surrounding stops at conditioned reward locations. (D) Mean spectrogram across all sessions and all rats. Large amplitude 16 Hz oscillations prior to stops at conditioned reward locations likely reflect a theta harmonic. (E) Mean theta (blue) and mean beta (red) amplitude (mV) for recording sessions in which there was one (left) or multiple (right) conditioned reward locations. The zero point marks the rat's encounter with a conditioned reward location. The upper gray bar indicates 250 ms bins with significant decreases in theta amplitude and the lower gray bar indicates 250 ms bins with significant increases in beta amplitude. Error bars indicate standard error for theta (cyan) and beta (magenta) means. (F) Same as in (E), for the first three (left) and last three (right) laps of sessions in which there was more than one shift in reward location, with gray traces indicating the mean velocity; significant via Tukey's HSD, ^*p < 0.05.

Figure 2

Random stops on the circular track do not elicit changes in theta and beta amplitude. (A) Mean theta (red) and mean beta (blue) amplitude (mV) during intervals surrounding stops longer than 3 s. Gray traces outline the mean theta and mean beta amplitude during stops at conditioned reward locations shown in Figure 1E, right. The zero point indicates the onset of a stop. The upper gray bar indicates 250 ms bins with significant decreases in theta amplitude (blue) and the lower gray bar indicates 250 ms bins with significant increases in beta amplitude (red). Error bars indicate standard error for theta (cyan) and beta (magenta) means. (B) Mean velocity during intervals surrounding random stops. (C) Distribution of random stops with respect to reward location. The plot does not depict the physical location of a stop on the circular track, but rather the location of a stop with respect to a given condition reward location; significant via Tukey's HSD, ^*p < 0.05.

Figure 3

Conditioned cue paradigm elicits similar changes in theta and beta amplitude. (A) Cartoon schematic of the task. (B) Mean theta (blue) and mean beta (red) amplitude as the rat encounters the conditioned cue (weigh boat) at the zero time point. The upper gray bar indicates 250 ms bins with significant decreases in theta amplitude (blue) and the lower gray bar indicates 250 ms bins with significant increases in beta amplitude (red). Error bars indicate standard error for theta (cyan) and beta (magenta) means. (C) Mean velocity during intervals surrounding stops at the conditioned cue; significant via Tukey's HSD, ^*p < 0.05.

Figure 4

Changes in theta and beta amplitude during encounters with each object in an object association task. (A) Cartoon schematic of the task in which rats were trained to approach and push over three Lego objects in order to receive a 100%, (best), 25% (good), or 0.002% quinine (bad) pellet underneath. Each object was conditioned to one of the three pellet types. (B) Mean theta (blue) and mean beta (red) amplitude for encounters with the bad (left), good (middle), and best objects (right). The upper gray bar indicates 250 ms bins with significant decreases in theta amplitude (blue) and the lower gray bar indicates 250 ms bins with significant increases in beta amplitude (red). Error bars indicate standard error for theta (cyan) and beta (magenta) means. (C) Mean velocity during intervals surrounding object encounters in the object association task; significant via Tukey's HSD, ^*p < 0.05.

Figure 5

CSD analysis of beta local field potentials along the D/V Axis. (A) Mean theta (blue) and mean beta (red) amplitude for intervals surrounding stops at the water port for a single session (top: upper blade of the granule cell layer, bottom: lower blade of the granule cell layer). The upper gray bar indicates 250 ms bins with significant decreases in theta amplitude (blue) and the lower gray bar indicates 250 ms bins with significant increases in beta amplitude (red). Error bars indicate standard error for theta (cyan) and beta (magenta) means. (B) Average CSD (current source density) from one rat for the 200 ms surrounding the trough of each beta cycle that met the following conditions: (1) cycle occurred within the 5 s intervals surrounding stops at the water port and (2) beta amplitude was 2 STD above the mean amplitude of the recording session. Given the large range in beta frequencies (15–30 Hz), CSD analysis was performed for two smaller frequency ranges: Beta 1 (15–18 Hz) and Beta 2 (24–27 Hz). (C) Schematic of the dendritic and somatic layers of the dentate gyrus, including a cartoon representation of a single silicon probe shank. (D) Summary diagram outlining the main findings across experiments. Decreases in theta amplitude are observed prior to stops during initial cue encounters, whereas increases in beta amplitude are observed after stops. Stops at familiar cues elicit similar changes in theta, but increases in beta can be observed prior to stops at familiar cues. When there are no cues present, decreases in theta can be observed after stops, but there are no increases in beta; significant via Tukey's HSD, ^*p < 0.05.