Internally mediated developmental desynchronization of neocortical network activity

Abstract

During neocortical development, neurons exhibit highly synchronized patterns of spontaneous activity, with correlated bursts of action potential firing dominating network activity. This early activity is eventually replaced by more sparse and decorrelated firing of cortical neurons, which modeling studies predict is a network state that is better suited for efficient neural coding. The precise time course and mechanisms of this crucial transition in cortical network activity have not been characterized in vivo. We used in vivo two-photon calcium imaging in combination with whole-cell recordings in both unanesthetized and anesthetized mice to monitor how spontaneous activity patterns in ensembles of layer 2/3 neurons of barrel cortex mature during postnatal development. We find that, as early as postnatal day 4, activity is highly synchronous within local clusters of neurons. At the end of the second postnatal week, neocortical networks undergo a transition to a much more desynchronized state that lacks a clear spatial structure. Strikingly, deprivation of sensory input from the periphery had no effect on the time course of this transition. Therefore, developmental desynchronization of spontaneous neuronal activity is a fundamental network transition in the neocortex that appears to be intrinsically generated.

Figure 1.

Rapid decorrelation of neuronal activity after P12. A, Average intensity projection of a time-lapse “movie” stack of two-photon images of OGB-loaded cells in L2/3 of barrel cortex from a representative in vivo experiment (P9 mouse). B, Contours of cells automatically detected in the image using custom-written routines in NIH ImageJ and MATLAB. Not all OGB-labeled cells in A have a corresponding profile in B, because some of the cells were astrocytes. C, ΔF/F calcium traces of 15 individual L2/3 neurons from representative experiments at P5, P9, P13, and P26. Five representative traces for each age group are shown in expanded form in supplemental Figure S2 (available at www.jneurosci.org as supplemental material). D, Top, Extrapolated firing rates from deconvolved calcium traces of all neurons in the above experiments. Bottom, Percentage of neurons active across time. Dashed line shows the threshold at a significance level of p = 0.05 for detection of synchronous network events. E, Correlation matrices displaying the correlation coefficients between the deconvolved calcium traces of all possible cell pairs. Cells were ordered sequentially from top to bottom and then from right to left.

Figure 2.

The spatial structure of network activity is developmentally regulated. A, Cumulative probability plots of correlation coefficients, pooled according to distance between cell pairs. All data obtained from experiments with OGB in unanesthetized mice at P4–P7 (n = 4 mice), P8–P11 (n = 5 mice), P12–P15 (n = 6 mice), and older than P16 (n = 4 mice). B, Mean correlation coefficient versus distance separating neuronal pairs. All mice were unanesthetized. C, Mean correlation coefficient of cells located within 100 μm from each other at different postnatal ages. D, Proportion of significantly correlated neuronal pairs located 10–100 μm from each other (left) or 200–500 μm from each other (right) at different postnatal ages. E, Representative spatial correlograms showing the correlation coefficient of cells located within 50 μm of each other for a P6 (left) and a P14 (right) mouse. Lines connecting neuronal pairs are pseudo colored to reflect the amplitude of their correlation coefficient. For B–D, correlation coefficients from all cell pairs within an animal were averaged, and the mean of means was calculated for each developmental stage and treatment group. Error bars in B–D reflect the SEM. Asterisks in C and D demonstrate statistical significance using a one-way ANOVA, followed by a Tukey's post hoc test.

Figure 3.

Relationship of network activity to barrel boundaries in L4. A, Post hoc reconstruction of the L4 barrel cytoarchitecture (from cytochrome oxidase staining) in relationship to the OGB-filled neurons in L2/3 previously imaged with in vivo two-photon microscopy in a P6 mouse. B, Correlation matrix displaying the correlation coefficients between the estimated firing rate vectors for all possible cell pairs. L2/3 neurons are grouped according to whether they are situated above one barrel or another. Cells residing above septae are not displayed. C, Mean correlation coefficient versus distance separating pairs of cells located above the same barrel (black) or above different barrels (red) at P6–P8 (n = 3 mice). Error bars in C and D reflect the SEM. D, Mean correlation coefficient versus distance separating pairs of cells located above the same row of barrels (black) or above the same arc of barrels (red) at P6–P8 (n = 3 mice).

Figure 4.

Developmental decorrelation persists under anesthesia. A, Average somatic ΔF/F traces of the same cortical region in the same P9 mouse in the unanesthetized state (left) and under light isoflurane anesthesia (right). B, Correlation matrices of deconvolved somatic calcium traces of an ensemble of L2/3 neurons from the same P9 mouse in the anesthetized state (left) and under light isoflurane anesthesia (right). C, Mean correlation coefficient versus distance between cell pairs obtained from the same animal depicted in A and B in the unanesthetized state (black) and under light isoflurane anesthesia (red). D, Mean firing rates estimated from the deconvolved calcium traces in unanesthetized and isoflurane-anesthetized mice, at P8–P11, P12–P15, and older than P16 stages. Error bars in D and F reflect the SEM. E, Mean correlation coefficients of cell pairs located 10–100 μm from each other at different postnatal ages. Pooled data from experiments in unanesthetized (red circles), urethane-anesthetized (green triangles), or isoflurane-anesthetized (black triangles) mice, with each point representing data obtained from a different animal. F, Mean correlation coefficients of neuronal pairs separated by 10–100 μm at P9–P11 (n = 5), P12–P15 (n = 9), and older than P16 (n = 22), in isoflurane-anesthetized mice. The asterisk demonstrates statistical significance using a one-way ANOVA, followed by a Tukey's post hoc test. Correlation coefficients for all cell pairs within an animal were averaged, and the mean of means was calculated for each developmental stage.

Figure 5.

Lower neuronal firing probability per network event after the second postnatal week. A, Representative in vivo whole-cell recordings from L2/3 neurons at P6, P13, and P18 under light isoflurane anesthesia (0.5–0.8%). B, Representative in vivo whole-cell recordings from L2/3 neurons in unanesthetized mice at P10, P15, and P20. C, Frequency of network events, proportion of network events resulting in the firing of action potentials, resting membrane potential, and action potential threshold across different postnatal ages, in mice under light isoflurane anesthesia. Each point represents data from a single neuron. *p < 0.05.

Figure 6.

Sensory deprivation does not affect the decorrelation of spontaneous network activity. A, Post hoc reconstruction of the L4 barrel cytoarchitecture (from cytochrome oxidase staining) in relationship to the OGB-stained neurons in L2/3 previously imaged with in vivo two-photon microscopy. The mouse was deprived of whisker input by plucking all contralateral whiskers from P2 until the day of imaging, in this case P15. B, Scatter plot and bar graph (inset) of mean correlation coefficients of neuronal pairs located within 10–100 μm of each other for control (black) and deprived (red) mice at P9–P11 (n = 3), P14–P16 (n = 3), and P19–P22 (n = 4). Each point represents a different animal. There are no significant differences. C, Mean correlation coefficient versus distance separating pairs of cells located above barrels (black) or above septae (red) at P14–P16, in control (top) or sensory-deprived (bottom) animals. Error bars reflect the SEM.