The goal of understanding the workings of the brain is
still very far from being reached. One area where significant
progress has been made is the study of the detailed mechanism
by which the connections between two individual brain
cells operate. For the great majority of these connections-called
synapses-the signaling is achieved by the incoming message
causing the release of a chemical, the-"transmitter."
Once released, the transmitter diffuses across a very
narrow cleft and binds on to specific molecules, the receptors,
on the surface of the target cell. Although there are
a variety of mechanisms by which the transmitter receptor
complex can influence the target cell, the best studied
mechanism is one in which membrane channel opens, allowing
the flow of ions into or out of the cell. The net ion
flow leads to a direct change in the membrane potential
of the cell, affecting its excitability, and hence the
likelihood of the incoming message being relayed.
Historically, the understanding of the working of the
synapse has been influenced by the technical ease with
which these connections can be studied. Much of the early
work was performed on the junction, in vertebrates, between
the motor nerve and muscle. Subsequently, the connections
to the large nerve cells in the spinal cord, giving rise
to the motor nerve to muscle, became the "model" synapse.
It is only in the last two decades that the techniques
have developed so that the most numerous type of brain
synapse, that between two nerve cells in regions of the
brain such as the cerebral cortex, can be studied in adequate
detail. It still remains a problem that much of the work
on this last type of synapse has, for reasons of technical
tractability, been performed on immature, developing brains
rather than those of the adult. The importance of drawing
attention to these issues is that the functional task
performed at these three types of synapse can be quite
different, and hence some features of the operation of
the synapse may reflect its specific job (or state of
development).
In a series of papers starting 50 years ago, Bernard
Katz and his colleagues revealed a then startling feature
of the way in which the transmitter was released at the
nerve-muscle junction. Instead of an expected mechanism
in which variations in the amount of transmitter released
was continuous, it was found the amount was in units of
approximately 5,000 molecules. Taking a term from physics
for an "irreducible minimum," Katz called these units
quanta. The underlying structural mechanism turned out
to be that the transmitter was concentrated inside small
intracellular organelles, called vesicles, and release
was achieved by enabling these vesicles to empty their
content into the intracellular space, adjacent to the
receiving cell's specialized receptors. Quantal analysis
refers to the methods used to deduce, for a particular
synapse, how many quanta of transmitter are released and
what the average effect of each quantum is on the target
cell, measured in terms of amount of ions flowing (charge)
or change in the membrane potential. The importance of
such an analysis is partly related to the insight it gives
into the detailed mechanism of the signaling process and
partly because it provides an effective method to quantify
the extent to which the change in the strength of a synapse
(for example, as a memory mechanism) reflects primarily
a change in the number of quanta released (presynaptic)
or a change in the response of the target cell to each
quantum (postsynaptic change).
What have we learnt from the last 50 years? At the synapse
between nerve and muscle, under normal conditions, a large
number of quanta of transmitter are released producing
a very large effect on the membrane potential, which is
more than sufficient to activate the muscle. At a more
detailed level, the released transmitter in a single quantum
has ample opportunity to bind to the adjacent postsynaptic
receptors, so that the factor which is dominant in setting
the size of the postsynaptic effect is the exact number
of molecules shared in the particular vesicle which releases
its content. By contrast, when brain cells are the postsynaptic
target, there is a restricted number of such receptors
so that the limitation on postsynaptic affect is not necessarily
the number of molecules, but can be the number of receptors.
Why this difference? There is good evidence for the motor
nerve cell (motoneurone) and for other brain cells that
this feature has been used functionally in two ways. In
the motoneurone, where the synapses show no long-term
plasticity (i.e., no "memory" mechanism), it is used exclusively
to ensure a synaptic "democracy." In nerve cells, unlike
muscle, the synapses can end at various distances from
the zone where the cell makes the decision to fire, sending
the message onwards. Everything else being equal, those
synapses further away have less of a voice. However, the
postsynaptic cell has found a mechanism, still mysterious,
by which it adjusts the number of postsynaptic receptors,
with larger numbers of receptors the further the synapse
is from the decision point. This adjustment is made in
such a way that, quantitatively, the efficacy of the synapse
in causing firing of the cell are equal, whatever their
location. This result was first reported for the motoneurone
20 years ago and it is an indication of how difficult
and slow it has been to achieve progress that the same
result has only now been found in a looser form for brain
cells in the cortex. Apart from technical difficulties,
there is an additional factor in these latter cells; that
the synapses also show long-term changes in their efficacy.
One of the mechanisms by which this occurs is postsynaptic,
as an increase in the number of response receptors. Thus,
depending on the "memory-state" of the synapse, there
are more or less postsynaptic receptors than would be
expected if the only factor operating was the mechanism
of synaptic "democracy."
One issue, about which there has been considerable controversy,
is whether the only memory mechanism at the synapse is
a change in the postsynaptic receptors. There is also
evidence that there is an additional, presynaptic mechanism
viz. a greater likelihood of releasing more quanta. This
evidence was reviewed and it was concluded that it is
compelling, not just from quantal analysis but from other
methods as well. Thus, in principle, the synapse is modifiable,
both presynaptically and postsynaptically. Although there
is some preliminary evidence about the rules governing
which process predominates, further research is required.
What the result does emphasize is that the synapse is
a unified device, with memory mechanisms being available
for changes in both the amount of transmitter released
and the postsynaptic response to a fixed amount of transmitter.
