When is an inhibitory synapse effective?
Ning Qian and Terrence J. Sejnowski, Proc. Natl. Acad. Sci. USA 1990,
87:8145-8149. Download the
full paper (PDF file)
Abstract
Interactions between excitatory and inhibitory synaptic inputs on
dendrites determine the level of activity in neurons. Models based on
the cable equation predict that silent shunting inhibition can
strongly veto the effect of an excitatory input. The cable model
assumes that ionic concentrations do not change during the electrical
activity, which may not be a valid assumption, especially for small
structures such as dendritic spines. We present here an analysis and
computer simulations to show that for large Cl- conductance changes,
the more general Nernst-Planck electrodiffusion model predicts that
shunting inhibition on spines should be much less effective than that
predicted by the cable model. This is a consequence of the large
changes in the intracellular ionic concentration of Cl- that can occur
in small structures, which would alter the reversal potential and
reduce the driving force for Cl-. Shunting inhibition should therefore
not be effective on spines, but it could be significantly more
effective on the dendritic shaft at the base of the spine. In contrast
to shunting inhibition, hyperpolarizing synaptic inhibition mediated
by K+ currents can be very effective in reducing the excitatory
synaptic potentials on the same spine if the excitatory conductance
change is less than 10 nS. We predict that if the inhibitory synapses
found on cortical spines are to be effective, then they should be
mediated by K+ through GABAB receptors.
Back to Qian Lab Home Page