Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Apr;115(4):1821-35.
doi: 10.1152/jn.00137.2015. Epub 2015 Dec 30.

Sensory-driven and Spontaneous Gamma Oscillations Engage Distinct Cortical Circuitry

Affiliations
Free PMC article

Sensory-driven and Spontaneous Gamma Oscillations Engage Distinct Cortical Circuitry

Cristin G Welle et al. J Neurophysiol. .
Free PMC article

Abstract

Gamma oscillations are a robust component of sensory responses but are also part of the background spontaneous activity of the brain. To determine whether the properties of gamma oscillations in cortex are specific to their mechanism of generation, we compared in mouse visual cortex in vivo the laminar geometry and single-neuron rhythmicity of oscillations produced during sensory representation with those occurring spontaneously in the absence of stimulation. In mouse visual cortex under anesthesia (isoflurane and xylazine), visual stimulation triggered oscillations mainly between 20 and 50 Hz, which, because of their similar functional significance to gamma oscillations in higher mammals, we define here as gamma range. Sensory representation in visual cortex specifically increased gamma oscillation amplitude in the supragranular (L2/3) and granular (L4) layers and strongly entrained putative excitatory and inhibitory neurons in infragranular layers, while spontaneous gamma oscillations were distributed evenly through the cortical depth and primarily entrained putative inhibitory neurons in the infragranular (L5/6) cortical layers. The difference in laminar distribution of gamma oscillations during the two different conditions may result from differences in the source of excitatory input to the cortex. In addition, modulation of superficial gamma oscillation amplitude did not result in a corresponding change in deep-layer oscillations, suggesting that superficial and deep layers of cortex may utilize independent but related networks for gamma generation. These results demonstrate that stimulus-driven gamma oscillations engage cortical circuitry in a manner distinct from spontaneous oscillations and suggest multiple networks for the generation of gamma oscillations in cortex.

Keywords: GABAA receptor; cortical column; gamma oscillation; spontaneous activity; visual cortex.

Figures

Fig. 1.
Fig. 1.
Local field potential (LFP) activity during the presentation of a drifting grating. A: 16 LFPs, filtered at 0.1–300 Hz, showing the evoked response to a single presentation of a full-screen vertical drifting grating (95% contrast, 3 Hz, 0.08 cycles/°) for 1 s, as indicated by green rectangle. An evoked response and corresponding fast frequency activity following stimulus onset are clearly visible during this single trial. B: average evoked response for 50 presentations of the drifting grating. The maximum amplitude of the evoked response occurs 90 ms after stimulus onset within L4. C: current source density (CSD) analysis of the average response in B shows a large initial sink in L4 at 90 ms followed by sinks in L2/3 (182 ms) and L5/6 (201 ms). D: same data as presented in A but filtered for the gamma frequency range, 20–50 Hz. A stimulus-driven increase in gamma activity occurs during the presentation of the drifting grating, and a spontaneous bout of gamma activity occurs several seconds later. E: average evoked response as in B filtered for gamma frequency range. Because gamma activity is not time-locked to stimulus onset, these averages only represent a small portion of the total gamma activity, as reflected by the smaller amplitude of the averages compared with the single-trial activity in D. F: CSD analysis of the average evoked gamma. Note difference in y-axis scale. Unlike the CSD analysis of the broadband LFP (C), there is not one initial sink in L4 followed by distinct sinks in the other layers but instead nonsignificant, multiple sinks and sources distributed over space and time.
Fig. 2.
Fig. 2.
Laminar characteristics of stimulus-driven and spontaneous gamma activity. A, left: power spectra from each of the 16 channels for 1 experiment. Power spectra are calculated for the 1-s window during the presentation of the stimulus (green) and for the baseline period 1 s immediately before the stimulus onset (black). Spectra are calculated for each single trial and then averaged across all trials. Right: for each channel, the stimulus spectrum (green) is divided by the baseline spectrum (black) to reveal the frequencies with the greatest increase above baseline during the presentation of the stimulus. The largest difference is within the gamma frequency band (20–30 Hz). B: values of the stimulus-to-baseline ratio (A, right) plotted with respect to cortical depth. Left: fold increase over baseline is represented by pseudocolor, with red indicating the largest increase (∼6-fold increase). Right: values for the frequency with the largest increase (26 ± 1 Hz, designated by arrow in A, right) plotted with respect to depth. Width of the gamma peak is 11.3 ± 1.8 Hz. C, left: spontaneous bouts of gamma are identified manually and confirmed by threshold crossing (2.5 SD + mean). Power spectra are calculated in the same manner as in A for the 1-s window during the spontaneous bout (blue) and the 1-s baseline period immediately preceding each bout (black). Right: the spectra during the spontaneous bout are then divided by the corresponding baseline spectra for each channel to reveal a ratio measurement. D: ratio measurements calculated in C plotted with respect to depth. Pseudocolor scale indicates increase over baseline and is the same as in B. Maximum increase (∼6-fold) occurs at frequency 25.2 ± 1 Hz, and width of the gamma peak is 18.5 ± 1.4 Hz. E: ratio measurements shown in A and C for stimulus and spontaneous gamma, respectively, were quantified (stimulus n = 18 probes, 13 animals; spontaneous n = 9 probes, 7 animals). Increase over baseline was quantified as the maximum amplitude for each channel, indicated by “a” in A, right. To compare between stimulus and spontaneous, these values were normalized to the value in L5/6. *Increase in visually driven gamma was significantly higher in L2/3 (1.83-fold greater, P = 0.022) and L4 (1.86-fold greater, P = 0.012) than in L5/6 only during stimulus driven gamma. F: width of the gamma peak includes those frequencies in which the ratio measurement is greater than baseline spectral noise, demonstrated by “w” in A, right. Width of the gamma peak was similar throughout the layers of cortex but encompassed a wider range of frequencies during spontaneous gamma (23.2 ± 1.9 Hz) than stimulus (13.4 ± 2.7 Hz, *P = 0.02). G: peak frequency was designated as the frequency with the largest increase over baseline, demonstrated by “f” in A, right. Peak frequency was constant through the layers of cortex and was the same during stimulus-driven (29.5 ± 3.4 Hz) and spontaneous (29.1 ± 1.8 Hz) bouts of gamma. H: latency to the first bout of gamma was only calculated for stimulus-driven gamma; since this measurement is made from stimulus onset it cannot be calculated for spontaneous gamma. Latency was determined by finding the first gamma peak to cross the significance threshold after the onset of the stimulus. The population average showed no difference in onset latency between L2/3 (524 ± 246 ms) and L4 (495 ± 252 ms), but the latency in L5/6 was significantly longer than in L4 (652 ± 279 ms, *P = 0.04).
Fig. 3.
Fig. 3.
Coherence and phase measurements between cortical laminae. A: coherence ± jackknife error measurements for 50 trials of drifting grating presentation. Coherence of L4 and L2/3 between 20 and 50 Hz (0.83 ± 0.1) was greater than the coherence of L4 and L5/6 (0.64 ± 0.17, *P < 0.0001); Coherence between L2/3 and L5/6 was lower than both of the other pairs (0.46 ± 0.23). These differences were significant over the population (right; n = 5 animals) and were similar between stimulus-driven and spontaneous gamma (not shown). B: wave-triggered averages (WTAs) using the peaks of gamma oscillations in L4 as time stamps show a distinct phase relationship through the cortical depth. The phase relationship is seen even more clearly in the CSD plots on right. Similar phase shifts are seen in both stimulus-driven (left) and spontaneous (right) gamma activity. C: detailed view of WTA averages seen in B for 1 channel from each lamina. L2/3 and L5/6 are equidistant to L4. D: quantification of the phase relationships shown in B and C for the population shows a larger phase precession through the layers during stimulus-driven gamma than spontaneous (n = 5 animals). The positive delay between L2/3 and L4 is larger for visually driven (2.7 ± 1.7 ms) than for spontaneous (1.9 ± 1.6 ms; *P = 0.029) gamma. The average phase between L4 and L5/6, in contrast, showed a negative delay (visually driven = −2.2 ± 1.5 ms; spontaneous = −1.95 ± 1.8 ms).
Fig. 4.
Fig. 4.
Single units are modulated by gamma oscillations differently through the cortex. A: raster and peristimulus histogram (PSTH) from a regular-spiking (RS) cell in L2/3 during presentation of a drifting grating. The evoked response in the LFP (black trace, inverted polarity) follows a time course similar to the PSTH (gray) of the unit. B: perievent histograms for stimulus-driven (green) and spontaneous (blue) spiking events, calculated with respect to the positive peaks of gamma oscillations that crossed the significance threshold (2.5 SD above mean). C: distribution of rhythmicity indexes (RIs) for 176 cells during stimulus-driven (green) and spontaneous (blue) gamma. RIs for 1/3 (38%) of the neurons were greater than the significance threshold during stimulus-driven gamma (mean RI 0.99 ± 0.6), and RI for 15% of the neurons crossed threshold during spontaneous gamma (mean RI 0.742 ± 0.4). Box and whisker plots demonstrate the median and interquartile values for each condition (stimulus median = 0.79, lower quartile = 0.62, upper quartile = 1.16; spontaneous median = 0.66, lower quartile = 0.51, upper quartile = 0.84). D: RI did not correlate with the magnitude of gamma oscillations in the LFP from the same tetrode for either spontaneous or stimulus-driven gamma. E: distinct laminar differences in rhythmicity were observed during stimulus-driven but not spontaneous gamma. Neurons in both L2/3 (1 ± 0.6, *P = 0.025) and L5/6 (1.18 ± 0.6, *P = 0.0039) had significantly higher mean RI than L4 neurons (0.78 ± 0.3) (left). During spontaneous gamma there were no differences between the mean RIs for each layer (L2/3 = 0.76 ± 0.45; L4 = 0.67 ± 0.33; L5/6 = 0.77 ± 0.4). Similarly, a greater percentage of neurons in L2/3(38%) and L5/6 (58.5%) were rhythmic compared with L4 (20.9%) during stimulus-driven gamma (right). There was a similar trend during spontaneous gamma, but the differences between layers were smaller than during stimulus-driven gamma (L2/3 = 17.1%, L4 = 9.3%, L5/6 = 26.8%).
Fig. 5.
Fig. 5.
Fast-spiking (FS) and RS cell classification and modulation through the cortex. A: FS and RS cells were well separated on the basis of their peak-to-trough amplitude ratio, peak-to-trough time, and ending slope of the trough. B: FS cells showed significantly greater rhythmicity than RS cells in terms of both mean RI (FS 1.42 ± 0.7; RS 0.95 ± 0.5, *P = 0.002), and % of rhythmic cells (75% FS, 35% RS). C: FS cells displayed a laminar structure to their rhythmicity during both stimulus-driven and spontaneous gamma. During stimulus-driven gamma (dark lines) FS cells in L5/6 were more rhythmic (2.03 ± 0.7) than those in L4 (1.16 ± 0.6, *P = 0.022) or L2/3 (1.12 ± 0.6; *P = 0.02). FS cells had a similar laminar profile during spontaneous gamma (light lines) (L5/6 1.24 ± 0.7; L4 0.75 ± 0.7, *P = 0.038; L2/3 0.82 ± .042, *P = 0.033). The rhythmicity in L5/6 was greater during stimulus-driven than spontaneous gamma activity (*P = 0.0217). D: RS cells in L2/3 (0.998 ± 0.6) and L5/6 (1.06 ± 0.46) showed greater rhythmicity than cells in L4 (0.734 ± 0.25, *P < 0.0001). During spontaneous gamma, RS cells showed no distinct changes in rhythmicity through the depth of the cortex (L2/3 0.76 ± 0.5, L4 0.66 ± 0.27, L5/6 0.71 ± 0.34). RS cell rhythmicity in both L2/3 and L5/6 was greater during stimulus-driven than spontaneous gamma activity (*P < 0.0001). E: similar to mean RI shown in C, more FS cells were rhythmic in L5/6 (100%) than in L2/3 (66.7%) or L4 (60%) during stimulus-driven gamma. This same relationship held true for spontaneous gamma (L5/6 = 80%; L2/3 = 33.3%; L4 = 20%). F: like the data presented in D, more RS cells were rhythmic in L2/3 (36%) and L5/6 (52.8%) than in L4 (15.8%) during stimulus-driven gamma. During spontaneous gamma there was a trend in the same direction, but the difference between layers was much smaller (L2/3 = 17%, L4 = 7.9%, L5/6 = 19.4%).
Fig. 6.
Fig. 6.
Gamma activity is reduced after the application of picrotoxin. A: GABAA receptor antagonist picrotoxin increases the amplitude of the evoked response in the LFP during presentation of the drifting grating through all layers of cortex, with the largest increase in L2/3 (n = 3 animals, *P = 0.0215). B: single traces of evoked responses per layer showing that the effect of picrotoxin is to increase the amplitude of the visually evoked response only in supragranular layers. C: picrotoxin reduces the gamma power (20–50 Hz) during the presentation of the visual stimulus in L2/3 (*P = 0.0313). D: % of rhythmic RS cells (n = 30) in both L2/3 and L5/6 decreases after the application of picrotoxin (100% and 88% decrease); % of rhythmic FS cells (n = 7) falls in L2/3 (33% decrease) but not in L5/6 (0% decrease).

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback