Subthreshold Oscillating Waves in Neural Tissue Propagate by Volume Conduction and Generate Interference

Abstract

Subthreshold neural oscillations have been observed in several brain regions and can influence the timing of neural spikes. However, the spatial extent and function of these spontaneous oscillations remain unclear. To study the mechanisms underlying these oscillations, we use optogenetic stimulation to generate oscillating waves in the longitudinal hippocampal slice expressing optopatch proteins. We found that optogenetic stimulation can generate two types of neural activity: suprathreshold neural spikes and subthreshold oscillating waves. Both waves could propagate bidirectionally at similar speeds and go through a transection of the tissue. The propagating speed is independent of the oscillating frequency but increases with increasing amplitudes of the waves. The endogenous electric fields generated by oscillating waves are about 0.6 mV/mm along the dendrites and about 0.3 mV/mm along the cell layer. We also observed that these oscillating waves could interfere with each other. Optical stimulation applied simultaneously at each slice end generated a larger wave in the middle of the tissue (constructive interference) or destructive interference with laser signals in opposite phase. However, the suprathreshold neural spikes were annihilated when they collided. Finally, the waves were not affected by the NMDA blocker (APV) and still propagated in the presence of tetrodotoxin (TTX) but at a significantly lower amplitude. The role of these subthreshold waves in neural function is unknown, but the results show that at low amplitude, the subthreshold propagating waves lack a refractory period allowing a novel analog form of preprocessing of neural activity by interference independent of synaptic transmission.

Keywords: hippocampus; neural oscillation; optogenetics; propagation; subthreshold oscillation.

Figure 1

Neural activity induced by optogenetic stimulation in the longitudinal hippocampal slice. (A) The experimental setup to trigger neural activity by optogenetic stimulation. Two electrodes were positioned on the cellular layer and an optic fiber was placed near the recording electrodes to deliver the laser pulse train. (B) The laser pulse train at 10 Hz can trigger two types of neural activity. One was the suprathreshold spikes with an amplitude similar to that of spontaneous spikes. The other was an oscillating wave with a smaller amplitude. (C) The optogenetic induced spikes by single laser pulse can propagate through the longitudinal slice. The expanded window shows that there are delays between two spikes recorded from REC1 and REC2 electrodes. (D) Similarly, two oscillating waves triggered by 5 Hz laser pulse train and recorded from REC1 and REC2 electrodes have delays in each cycle. (E) The speeds of the spikes and oscillating waves are not significantly different. (F) 5 Hz pulse train can trigger a 5 Hz oscillating wave. The wave circles followed the laser pulse train. Similarly, 10 Hz, 20 Hz, and 50 Hz pulse trains can trigger 10 Hz, 20 Hz, and 50 Hz oscillating waves individually. (G) The amplitude of the induced spikes was higher than those of the oscillating waves at 5, 10, 20, and 50 Hz (*: p < 0.05, n = 6 slices).

Figure 2

Propagation of subthreshold wave across the transection in the slice. (A) A transection was made in the middle of the slice. Two electrodes (REC1 and REC2) were placed on both sides of the slice to monitor the propagation. The laser pulse train was applied on one side of the slice to initiate the waves. (B) The 5 Hz wave can cross the transection while the subthreshold wave can be recorded in both electrodes with the presence of the transection. The expanded window shows that the delay was observed between two waves from REC1 and REC2. (C) The 10 Hz wave can propagate through the transection. The expanded window shows the wave can cross the transection with a delay. (D) The speeds of the waves at various frequency bands were similar across the frequency (n = 6 slices).

Figure 3

The wavelength and amplitude-dependent propagation of the subthreshold waves. (A) The wavelength of the subthreshold waves decreased with frequency (**: p < 0.01, n = 6 slices). (B) The speed of the subthreshold wave decreased when the amplitude of the waves reduced (*: p < 0.05, n = 5 slices). (C) The speed decayed with the amplitude drop at 10 Hz subthreshold waves (*: p < 0.05, n = 5 slices). (D) A similar trend was also observed in the 20 Hz subthreshold waves (*: p < 0.05, n = 5 slices).

Figure 4

The endogenous electric fields of the subthreshold waves. (A) The electric field measurement between somatic and dendritic layers. Two electrodes were placed in the somatic and dendritic layers distal to the initiation of the wave to measure the voltage difference. (B) The field measurement along the longitudinal direction. Two electrodes were placed in the cellular layer to measure the voltage and calculate the electric field. (C) An example of the measured electric field between somatic and dendritic layers when the 10 Hz wave was present. (D) An example of the electric field in the longitudinal direction at 10 Hz wave. (E) The electric fields between somatic and dendritic layers were measured at 5 Hz, 10 Hz, and 20 Hz waves (n = 5 slices). (F) The electric fields along the longitudinal direction were measured at 5 Hz, 10 Hz, and 20 Hz waves (n = 5 slices).

Figure 5

Interference of the subthreshold waves. (A) The subthreshold wave at 10 Hz was triggered in the temporal region of the slice and was recorded in the middle of the slice. (B) Similarly, the subthreshold wave can be triggered in the septal area and recorded in the middle of the slice. (C) When the laser was applied on both sides of the slice and generated two subthreshold with the same phase, larger subthreshold waves were recorded in the middle of the slice. (D) When the laser triggered two waves with a 180-degree phase shift, a smaller subthreshold wave was observed in the middle of the slice. (E) The amplitude of subthreshold waves reconstructed by adding two encountering waves was similar to that measured from the recording signal. It shows that the waves can constructively interfere with each other (n = 5 slices). (F) The amplitude of subthreshold waves from the estimated waves was similar to that from the recorded waves. It shows that the waves can also destructively interfere with each other (n = 5 slices).

Figure 6

Interference effect related to the phase shift between two waves. (A) Waves were initiated at either the temporal area or septal area. (B) Interference between two waves was observed in the middle area of the slice with a phase shift of 0, 90, 180, and 270 degrees between the two waves. (C) The amplitude of the interference between the waves with a phase shift between two waves (**: p < 0.01, n = 6 slices).

Figure 7

Collision of the suprathreshold spikes. (A) Both suprathreshold spikes and subthreshold waves were triggered in the temporal area of the slice or initiated in the septal area of the slice. (B) The constructive interference can only be observed outside the window of the suprathreshold spike duration. (C) Similarly, destructive interference can only happen without the presence of the suprathreshold spikes. (D) The suprathreshold spike was triggered in the temporal region of the slice and propagated to the septal region. (E) Similarly, the suprathreshold spikes can be triggered in the septal area and traveled to the temporal area. (F) When the laser was applied on both sides of the slice and generated suprathreshold spikes simultaneously, only one spike was recorded through the slice instead of two spikes. (G) The normalized amplitudes of the spikes triggered by the laser from the left side, right side, or both sides were not significantly different. (H) The pulse widths of the spikes triggered by the laser from the left side, right side, or both sides were similar. (I) By analyzing the amplitudes from the signals in each condition, the amplitudes of these suprathreshold spikes were similar. However, the estimated amplitude was higher than the real amplitude of the spike when assuming the spikes can interfere with each other (**: p < 0.05, n = 3 slices).

Figure 8

NMDA receptor properties of the subthreshold waves. (A,B) The subthreshold waves can be triggered at 5 Hz and 10 Hz before and after the application of the NMDA receptor blocker, APV. (C) The amplitudes of the subthreshold waves were similar with and without the presence of APV (n = 5 slices).

Figure 9

Sodium channel properties of the subthreshold waves. (A,B) Subthreshold waves at 5 Hz and 10 Hz can be triggered before the application of TTX but the amplitude was greatly reduced following the application of TTX. (C) Amplitudes of the subthreshold waves decreased after the slice was perfused with TTX solution (**: p < 0.01, n = 5 slices). (D) Induced oscillating wave can still propagate through the slice with smaller amplitude after the application of TTX. (E) Expanded window of (D) show the delays between two waves from different electrodes. (F) Speeds of the oscillating waves at both 5 Hz and 10 Hz decreased after the blocking the sodium channels (n = 5 slices).