Affiliations: Department of Biology and Biochemistry, University of
Houston, Houston TX 77204-5513, USA
Note: [] Corresponding address: Department of Biology and Biochemistry,
University of Houston, 4800 Calhoun Rd., Houston TX, 77204-5513, USA. Tel.: +1
713 743 2658; Fax: +1 713-743-2636; E-mail: ccolbert@uh.edu.
Abstract: A hallmark of synaptic plasticity is the associative, or Hebbian,
nature of its induction. By associative, we mean that the timing relationships
between activity of the pre- and postsynaptic elements of a synapse determine
whether synaptic strengths are modified. lt is well-established that
associativity results, in large part, from the dual requirements for activation
of the N-methyl-D-aspartate receptor-ionophore, namely presynaptic
neurotransmitter release and postsynaptic depolarization. However, the specific
dendritic events that provide the postsynaptic depolarization have been
relatively unexplored. Increasing evidence suggests that back-propagating
(i.e., antidromic) Na^+ action potentials provide the necessary postsynaptic
depolarization to allow induction of associative synaptic plasticities. In
hippocampal CAI and neocortical layer V pyramidal neurons, these action
potentials provide much greater levels of dendritic depolarization than would
be expected from synaptic currents alone. Moreover, they provide a relatively
brief and synchronous depolarization throughout the dendritic arbor, allowing
timing relationships to more directly reflect pre- and postsynaptic cell
firing. Interestingly, certain properties of the back-propagating actions
potentials differ from axonal or somatic action potentials in ways that seem to
reflect their function. For example, the all-or-none property of action
potential amplitude does not hold in the dendrites. In this review we discuss
the back-propagating action potential as a dendritic signal that provides
information to synapses about the firing state of the postsynaptic neuron.
First, we consider the evidence that action potentials propagate back from the
axon. Second, we describe the characteristics of the back-propagating action
potential in terms of interactions of its underlying ionic currents. Third, we
describe how these properties contribute to the timing aspects of the induction
of long-term potentiation. Finally, we discuss modulation of the underlying ion
channels by neurotransmitter systems and other agents and speculate on their
roles in learning and memory.
Keywords: hippocampus, learning and memory, neural computation
