Mixed effects on neurons

Therapeutic applications of ESB like deep brain stimulation were used before serious work had been done with animal models, unlike most new treatments, which undergone intensive testing before use. Consequently the exact mechanisms behind ESB based therapies are still being uncovered and debated.

ESB in all its forms has been claimed to excite neurons, to inhibit them, or to do both. in fact all these can occur, depending on the stimulation profile used, and the neurons affected. As research continues therapies can be designed with specific actions in mind, harnessing the power of ESB to modulate neural function in a precise manner.

Linking macro to micro

Applying electrical stimuli to the brain can produce observable effects, but what exactly is occurring between the pulse of electrons and the changes in behavior? This problem is reminiscent of Skinner's Black Box, a long-standing caveat of psychology. Being pragmatic it doesn't matter how therapies work, as long as they do. Still from deeper understanding will come more effective therapies.

Almost all in vivo ESB involves giving field stimulation to a group of neurons and neurites. The field is the area around the electrode, either a nucleus or a bundle of fibres, and stimulation influences them all.

In vitro effects of electrical impulses on neurons have been studied intensely, revealing the cellular mechanisms of action. However in vitro single-neuron studies are hard to reconcile in relation to the gross, whole organism effects of ESB. When complex neuronal feedback systems and long time scales are considered, it becomes clear that there is much to be discovered in this area.

Charge transfer

A metal electrode placed inside bodily tissues forms an interface between the two phases. In the metal electrode phase and it's electrical circuits charge is carried by electrons. In the bodily medium, the electrolyte, charge is carried by ions such as sodium, potassium, and chloride.

There are two primary mechanisms of charge injection from a metal electrode into an electrolyte.
The first consists of charging and discharging the 'double layer capacitance', causing a redistribution of charge, but no permanent electron transfer from the electrode to the electrolyte. A plane is formed over the surface of the metal electrode, along with a plane of opposite charge in the electrolyte. Also polar molecules such as water may be orientated by the charge field at the interface; the net orientation of polar molecules separating charge. The accumulation of charge develops a potential difference between the two phases that repels further current transfer.

For a small total injected charge, all current transfer is by charging and discharging the double layer. Above some injected charge density, a second mechanism occurs consisting of Faradaic reactions. When charge is injected beyond the capitance of the system, these Faradaic reactions occur, and electrons are transferred between the electrode and electrolyte. Thereby the chemical composition of the electrolyte is changed by reduction or oxidation reactions.
Products of Faradaic reactions usually remove energy from the system by precipitating or dissipating into the electrolyte. These products are often damaging to cells, disrupting the normal equilibrium of the extracellular fluid.

In addition to the the normal capacitance of the electrode-circuit, some metals have the property of pseudo-capacity, where a Faradaic reaction occurs, but because the product remains bound to the electrode surface, the reactant may be recovered if the direction of current is reversed. Platinum is commonly used for stimulating electrodes, as it has a pseudo-capacity, reducing dissipated charge.
[Merrill et al, 2005]

Electrical pulses

With small electrical changes away from equilibrium, current flows primarily into capacitive transfer, charging the electrode's capacitance, with minimal Faradaic reactions. As more charge is delivered through an electrode interface, the electrode capacitance continues to charge, and Faradaic reactions begin to become a significant fraction of the total injected current.

Generally one desires to use small currents to reduce Faradaic loss of energy, as it contributes to injury of the bodily tissue, and speeds the degeneration of the electrode. However this isn't always possible, as a larger current may be needed to for the desired effect - such as initiating action potentials. In this case steps can be taken to discourage irreversible reactions, and encourage the reversal of Faradaic reactions. As mention before certain metals are better for this, especially platinum.
One can also use a reversal potential after the initial potential, swapping electrode poles, which returns ionised species near the electrode to their original state. Naturally the sooner the reversal phase, the better, as reaction products will diffuse away from the electrode with time. Although an immediate inversion will suppress the pulse's ability to invoke effects on cells. Thus a delay is often used between the initial, stimulating phase and the reversal phase. A variety of pulse waveforms have been developed to satisfy the needs of experimenters.
[Merrill et al, 2005]

Electrical fields and neurons

Electrical stimulation of a neuron may depolarise it's membrane by introducing an electrical charge gradient into the extracellular medium. Membrane channels sensitive to potential difference will react by changing their permeability. Once the membrane reaches a certain threshold an action potential is triggered; voltage gated cation channels open in a feed-back cascade, flooding the cytoplasm with sodium. Tight membrane junctions with other neurons allow the action potential to be propagated to other cells directly by the electrostatic flux.

At synaptic densities the action potential will initiate exocytosis of membrane associated vesicles through the increasing intracellular calcium concentrations. Exocytosed vesicles spread their contents into the vicinity, the transmitter substances exerting further actions on nearby cells; exciting, inhibiting or modifying them. All these effects act in concert during ESB.

The ability of neurons to summate all incoming stimuli and integrate them into a response is important for ESB. Even if the ESB stimulus is sub-threshold, i.e. it doesn't invoke an action potential, it can still alter the net firing activity of the field neurons.

As well as invoking changes in membrane potential ESB will affect dissolved molecules in the field. Ionisation of molecules occurs, due to the electrostatic pressure, although the effects of such changes on signalling are scarcely known. It is assumed they can contribute to the effects of the electrical pulse, yet ionised reaction products are known to damage cells, and so may of limited use in altering signalling.
Constant electrical fields, in other words direct current fields, also produce interesting changes in cell behaviour. Direct currents are rarely used for ESB, as they cause electrolysis in the tissue fluid. See a video of a DC field affecting cell movement.
[McCaig et al,2005]

There are two types of response seen in neurons subjected to continual electrical stimulation - one in which cells depolarise and repolarise rapidly, reaching a stable and slightly depolarised potential. The other involves a longer repolarisation, reaching a more depolarised stable potential.
[Anderson et al,2004]

Models of neural resonance and noise

The macro-scale functions of a nucleus of neurons is a concert of each individual neuron's activity, which is in turn regulated by a host of factors. Neurons are constantly receiving chemical and electrical inputs from neighboring and innervating cells, which are integrated by the common transduction mechanisms to yield responses. This has been described in models such as the empirical biological neuron firing model. Naturally neurons can receive many inputs, and their responses can be modified by phosphorylation of ion-channels involved in signal transduction, making such models quite plausible and useful.

Models of neural activity can be useful in describing the effects of stimuli on neuronal activity (both individually and collectively). Some researchers have suggested that stochastic resonance is an integral feature of neuronal activity, where relaying of inputs is enhanced by background noise. For instance a sub-threshold low frequency stimulus can be detected if combined with a basal level of interference (random stimuli), and can be further tuned by adjusting the refractory and excitatory characteristics of a neuron. The signal enhancement could be useful in filtering or selectively enhancing/removing a signal from a set of signals in the field of the neuron.
[Osma et al,2004]

Perhaps ESB exerts an effect on neuronal signaling related to that modeled by stochastic resonance. The stimulation could act to sensitise neurons to sub-threshold stimuli, or to exclude certain stimuli from inducing excitation.

Long term effects of modifying signaling