Size-dependent regulation of synchronized activity in living neuronal networks

We study the effect of network size on synchronized activity in living neuronal networks. Dissociated cortical neurons form synaptic connections in culture and generate synchronized spontaneous activity within 10 days in vitro. Using micropatterned surfaces to extrinsically control the size of neuronal networks, we show that synchronized activity can emerge in a network as small as 12 cells. Furthermore, a detailed comparison of small (∼20 cells), medium (∼100 cells), and large (∼400 cells) networks reveal that synchronized activity becomes destabilized in the small networks. A computational modeling of neural activity is then employed to explore the underlying mechanism responsible for the size effect. We find that the generation and maintenance of the synchronized activity can be minimally described by: (1) the stochastic firing of each neuron in the network, (2) enhancement in the network activity in a positive feedback loop of excitatory synapses, and (3) Ca-dependent suppression of bursting activity. The model further shows that the decrease in total synaptic input to a neuron that drives the positive feedback amplification of correlated activity is a key factor underlying the destabilization of synchrony in smaller networks. Spontaneous neural activity plays a critical role in cortical information processing, and our work constructively clarifies an aspect of the structural basis behind this.

Figure 1

Primary rat cortical neurons grown on micropatterned substrates at 10 DIV. The size of the micropatterns were as follows: (a) $200\times 200\mu {\mathrm{m}}^2$ (small), (b) $500\times 500\mu {\mathrm{m}}^2$ (medium), and (c) $1000\times 1000\mu {\mathrm{m}}^2$ (large). Scale bars: (a) and (b) 100 μm and (c) 200 μm. (d) The number of cells on each micropattern. The boxes indicate the span from the median to the first and third quartiles, the whiskers indicate the whole data spread, and circular plots indicate the mean. The number of cells was determined from phase-contrast micrographs.

Figure 2

Fluorescence Ca imaging of micropatterned neuronal networks. The cells were loaded with the fluorescence Ca indicator Fluo-4. (a) Time lapse images of a synchronized network burst observed in a 12-cell network on the small micropattern. (b) and (c) Cortical networks on (b) medium and (c) large micropatterns loaded with Fluo-4.

Figure 3

Analysis of spontaneous neural activity in micropatterned cortical networks. (a) Relative fluorescence intensity of the Ca indicator Fluo-4 in a large network. Traces from five representative cells are shown. (b) Raster plot of the spontaneous neural activity for a large network derived from the relative fluorescence intensity. Each point corresponds to a bursting activity in a neuron, determined from the derivative of the fluorescence trace. In this particular example, synchronized network bursts were detected six times during a 360-s recording session. Note that a fraction of the neurons randomly selected from the whole population was analyzed. (c) A closeup view of a network burst. (d) Frequency of network bursts for the three network sizes. The boxes indicate the span from the median to the first and third quartiles, the whiskers indicate the whole data spread, and the circular plots indicate the mean.

Figure 4

Effect of network size on synchrony. (a) Raster plot of the spontaneous neural activity for a small network. (b) Matrix plot of correlation coefficients of the network shown in (a). (c) Average correlation coefficient calculated for each network size. The boxes indicate the span from the median to the first and third quartiles, the whiskers indicate the whole data spread, and the circular plots indicate the mean.

Figure 5

Computational simulation of spontaneous neural activity in cultured cortical networks of different sizes. (a) and (b) Schematic of network models. The blue squares are the nodes (neurons), and the gray lines are the links. (c) and (d) Raster plot derived from the model network. Number of neurons N and average node degree $k$ were as follows: (a) and (c) $N=20,k=4.5$ and (b) and (d) $N=400,k=20$.

Figure 6

Dependence of the average correlation coefficient on network size. The blue line represents the mean of the simulation data. For comparison, the experimental results are plotted in red. The simulation data over a wider range are shown in the inset.

Figure 7

Dependence of the network burst frequency on network size. The blue line represents the mean of the simulation data. For comparison, the experimental results are plotted in red. The simulation data over a wider range are shown in the inset.