Synapses learning and memory

Animal cells exchange signals with one another throughout life, not just during development. These exchanges are necessary if the organism is to function as an integrated unit and survive. Neurones are no exceptions, though their method of communicating - by chemical signals at synapses - is specialised. Any cell might alter its internal state and gene expression pattern in response to a signal from another cell. Again, neurones are no exceptions; but again, their signal receptors are confined to membranes bordering the synapse.

43 This point is widely misunderstood. One example will serve for illustration. Some years ago, newspaper headlines heralded the discovery of a "gene for the nurturing instinct" in mice. What had really been discovered was that elimination of a particular immediate-early gene known as fos-B produced poor mothers. These genetically damaged mice did not groom their pups normally, or show any urgency in recovering them when they wandered. Mice lacking fos-B develop abnormally in a brain region known as the preoptic area, which in normal mice becomes active in many stimulus situations - including presentation of pups. So the fos-B deficient mice had impaired development in a significant part of the brain, and in consequence had a number of behavioural abnormalities. One of these was poor maternal behaviour. A phrase such as "gene for the nurturing instinct" betokens a misunderstanding of biology.

After the embryonic brain has developed, the responsiveness of neurones to signals becomes important in learning and memory. Two main types of process seem to account for at least some aspects of learning and memory: structural remodelling of the synapse, i.e. the formation of new axonal branches and perhaps the removal of old ones; and functional changes in synaptic strength. Both these mechanisms depend on transmitter chemicals that alter the postsynaptic cell's internal state and its pattern of gene expression. These signalling chemicals usually occupy their receptors for longer periods than neurotransmitters do.

Synaptic remodelling works roughly as follows. Suppose a chemical signal released from the presynaptic neurone causes one or more long-term changes the postsynaptic cell (1 in Fig. 16-3). This cell responds by secreting a factor that enters the presynaptic neurone (2), where it is carried back to the cell body along the cytoskeletal transport system (3). Here, it alters the gene expression pattern of the presynaptic cell. One consequence might be a redistribution of membrane material at the axon termini (4) and growth of new axon branches (5), perhaps leading to the formation of new synapses with other neurones. This kind of remodelling has been described in large animal brains. The implications for learning are considerable. Although mature neurones cannot divide or be replaced, they can continue to grow and to make new branches and new connections. The phrase "life-long learning" makes biological sense.

The alternative to remodelling is to change the strength of an existing synapse. Long-term potentiation (LTP) and long-term depression (LTD) seem to account for some aspects of memory. LTP depends on a chemical, secreted from the axon terminus, which only occupies its postsynaptic receptors when it is released in sufficient quantity. This happens when the axon transmits rapid successions of action potentials at frequent intervals. Occupation of the receptor causes the postsynaptic cell to take in calcium, which changes certain signalling pathways, activating a succession of genes that, ultimately, make the postsynaptic membrane permanently more likely to depolarise when it is stimulated.

It is easy to imagine that LTP provides a mechanism for Hebbian associative learning and for more complicated learning processes. If two stimulus-response circuits interact, then LTP of the synapses that link them makes it easier to transmit an impulse from one circuit to the other.

  1. 16-3: synaptic remodelling, the possible cellular basis of learning and memory.
  2. 16-3: synaptic remodelling, the possible cellular basis of learning and memory.

However, the best available evidence suggests that LTP is relevant only to long-term memory storage, which is established only an hour or two after the learning event has taken place. It is not relevant to immediate and short-term memory. As an explanation for associative learning, therefore, it is not entirely adequate.

LTD, which makes the postsynaptic membrane becomes less liable to depolarisation, involves an analogous mechanism. Whereas LTP is a response to rapid repeated use of the circuit, LTD results when the use is slow and prolonged. Thus, LTD might enable neuronal circuits to adapt to continuous, and therefore uninteresting, stimuli. It stops the brain paying attention to them. Without such adaptation, a brain could not function effectively. It could not focus on significant changes in the animal's environment.

Was this article helpful?

0 0

Post a comment