Why do neurones use chemical signalling at synaptic junctions?

Question

When a neurone fires, it sends an electrical signal that jumps down the axon via the nodes of Ranvier very rapidly. At a synaptic junction, chemical Brownian diffusion signalling with receptor surface proteins is relatively slow and is often exploited by venoms and susceptible to toxins (on the plus side it's the reason a lot of medical drugs work). It seems flawed for evolution to have selected for this rather than some alternative quicker and more direct electrical interface.

Question. Why are chemical synaptic interfaces used in higher organisms at the synaptic junction?

Answer 1

This is the coolest part. Those synapses are the reason the brain is so complex! Basically you've got the first part right, the neurones are quicker and they transmit messages from one end to the other. The other thing you have to do is analyse and calculate. Signals from multiple neurones feed into a single neurone using a chemical synapse. Similarly the converse is true, a single neurone can feed multiple synapses. So when we for example walk, receptors monitoring our balance can feed this to our conscious and unconscious mind, but I'm grossly simplifying this. What our brain does is take hundreds and thousands of ports of information, with where the chemical synapse is and the type of synapse and the receptors and a million other things just slightly changing the information until it's perfect and then send that information a million ways however it likes.

The most important thing is this allows inhibition. Most of the brain uses inhibitory GABA rather than activating signals. Furthermore, the system uses the delay created. Different transmitters act for different durations and send a different volume of signal. An impulse is all or nothing, it's of a fixed amplitude, however a chemical signal can be fine tuned. Neurones can not only interact to inhibit themselves or other neurones to increase the resolution of a signal etc. Conversely a system of just physical connections allows little regulation, in fact it'd be practically a seizure.

Does that make sense?

Answer 2

There are both chemical and electrical synapses in many organisms. The electrical synapses are called gap junctions.

As you point out, the primary advantage of gap junctions is their speed, and they are commonly used in systems involving defensive reflexes.

However, as AndroidPenguin indicates, chemical synapses allow for greater computational abilities (changing the gain, integrating multiple inputs, etc). Gap junctions are also disadvantageous because they often (but not always) transmit signals in both directions; chemical synapses tend to be more unidirectional (of course, there's always backpropagation).

Answer 3

There are indeed 'gap junctions' which pass current directly from one cell to the next. So what advantages do we get out of chemical synapses that gap junctions do not provide?

1. Asymmetry. Synapses do not operate in reverse, thus the postsynaptic cell cannot generate currents in the presynaptic terminal (although there are secondary forms of communication which may operate in the reverse direction).
2. Ion Selectivity. A synapse allows one set of ionic currents to be transformed into any other set. Thus sodium and calcium influx at the presynaptic terminal might translate to chloride influx to make the synapse inhibitory. Furthermore, the ability of the postsynaptic cell to control calcium levels is critical for plasticity mechanisms.
3. Insulation. Gap junctions increase the conductance of the cell membrane, which can reduce the fidelity of signals as they propagate along the dendrite. In contrast, an inactive synapse might have virtually no effect on the postsynaptic cell--it does not pass currents.
4. Kinetics. Because chemical synapses generate their own ionic currents, they are also free to modify the speed of those currents (how quickly they rise and fall). This is much more difficult to achieve with gap junctions.

There is a very long list of functionality provided only by synapses which I have not listed here, but I think most of those are things which could, in principle, also be achieved with gap junctions (for example: amplitude control, metabolic responses, short- and long-term plasticity).