Geon The Mechanism of Local Neural Synchrony MT

 

In the brain, a group of neurons tend to fire together. Their activity may oscillate between ON and OFF states at a certain frequency. Like other types of oscillations or waves, the brain wave is also characterized by a phase, which specifies the initial point of the oscillation. If two groups of neurons oscillate at the same frequency and phase, they are called "phase locked" or "synchronized".

In the past two decades, significant progress has been made on the mechanism underlying local circuit oscillations, but little is known about the biophysical basis for long-range synchronization. This article will discuss the local synchronization.

The Mechanism of Local Synchronization

Most central neurons respond selectively to inputs at a preferred frequency. This feature is known as "resonance", which arises from both passive and active membrane properties. The passive property always filters out high frequencies, while the active property may filter out low frequencies, thereby resulting in a preferred frequency band (Hutcheon and Yarom, 2000; PDF). The active property is mainly determined by various types of potassium (K+) channels. Experiments have shown that the theta band (4 - 8 Hz) depends on HCN and M/KCNQ K+ channels (Hu et al., 2002; Yan et al., 2012). The gamma band (30 - 80 Hz) is regulated by Kv1 (Sciamanna and Wilson, 2011). Another K+ channel, Kv3, is important for faster spiking (Erisir et al., 1999). Hence, a group of neurons can oscillate at the same frequency, as long as they have the same intrinsic property.

Image

Figure 1. The gap junction. It is an intercellular channel formed by head-to-head connection of two connexons. Each connexon is an ion channel composed of six connexin proteins. The gap junctions allow direct exchange of ions (and thus membrane voltage) among connected cells. They are also called "electrical synapses". [Image source: Wikipedia]

An oscillation is characterized not only by the frequency, but also by the phase. To oscillate at the same frequency and phase (i.e., synchronize), neurons must interact with the same oscillating source, and/or with themselves. Two forms of interactions have been well established: (1) via chemical synapses (mediated by neurotransmitters) and (2) via electrical synapses (gap junctions). Signal transmission through electrical synapses is much faster than chemical synapses. Furthermore, the electrical coupling is bidirectional, which allows rapid reciprocal exchange of membrane voltage among connected cells. For these reasons, the electrical coupling is commonly used in the synchronization of parvalbumin-containing GABAergic neurons which are densely connected by gap junctions (Fukuda et al., 2006).

A brain region may contain several types of neurons, including both excitatory and inhibitory neurons. Different types of neurons have different resonant frequencies. Synchronization among heterogeneous neurons is mostly inhibition based (Whittington et al., 2011; Sohal et al., 2009). That is, the neurons (e.g., pyramidal cells) without electrical coupling can be synchronized by forming chemical synapses with an inhibitory GABAergic neuron (e.g., basket cell). In the absence of inhibition, a group of pyramidal cells will oscillate at certain frequency with random phases. When the inhibitory neuron fires, it can exert hyperpolarizing inhibition on the activity of connected pyramidal cells. As the inhibition decays, the pyramidal cells will resume their oscillations. Consequently, the aggregated activity of connected pyramidal cells will be paced by the inhibitory neuron (Gonzalez-Burgos et al., 2011, Figure 2). A basket cell may connect up to 1,000 pyramidal cells (Halasy et al., 1996). When a group of basket cells are synchronized by electrical coupling, they can recruit a much larger population of neurons into synchronization.

Limitation of Synaptic Coupling

The synaptic coupling, either chemical or electrical, requires close contact between neurons. To mediate long-range synchronization, the signal would have to travel along the axon. It takes about 30 milliseconds (ms) for a nerve impulse to travel from one hemisphere to the other through myelinated axons in the corpus callosum, and 150 - 300 ms through unmyelinated axons (Fields, 2008). Yet, synchronization can be established within a fraction of the oscillation period, about 1 ms (Nobili, 2009).

Synchronization among widely separated brain regions is prevalent. During the maintenance of visual working memory, alpha synchronization was observed in fronto-parietal, cingulate, and insular cortices concurrently with synchronization in the beta- and gamma-band networks (Palva and Palva, 2011). When the stimulus became perceptually ambiguous, beta synchronization was enhanced across bilateral frontal eye field, posterior parietal cortex and visual area MT. Its strength predicted the subject’s percept (Siegel et al., 2012). The theta-modulated gamma-band synchronization was found among activated regions during a verb generation task (Doesburg et al., 2012) and auditory attention control (Doesburg et al., 2012). Due to the axonal conduction delay, the synaptic coupling is unlikely to account for these large-scale synchronizations. The next article will show how the long-distance synchronization can be achieved by the electromagnetic (EM) coupling, namely, via EM waves which travel at the speed of light.

 

Author: Frank Lee
First published: February 13, 2013
Last updated: June 18, 2014