|
Membrane potential fluctuations lower the detection threshold of neocortical pyramidal neurons in vivo.Alain Destexhe and Nicolas HôSociety for Neuroscience Abstracts 25: 360, 1999. AbstractNeocortical neurons in vivo experience high levels of background synaptic activity responsible for continuous membrane potential (Vm) fluctuations. How such fluctuations affect the response of these neurons is however unknown. The impact of Vm fluctuations was investigated using biophysical models of morphologically-reconstructed pyramidal neurons, containing voltage-dependent conductances in soma and dendrites, and GABAergic and glutamatergic synapses whose densities were estimated from morphological data. The release conditions at these synapses were estimated based on intracellular recordings in vivo, yielding a model of background synaptic activity consistent with experimental measurements. In these conditions, the response to constant synaptic stimuli showed a high variability similar to in vivo recordings. The probability of synaptically-evoked spikes revealed significant differences with and without background activity: Vm fluctuations enabled pyramidal neurons to detect inputs that would normally be subthreshold, and to discriminate between inputs normally indistinguishable. We show that this effect is not due to the tonically-active synaptic conductances but is specifically attributable to Vm fluctuations. This finding may have important implications at the network level: a population of “fluctuating” pyramidal neurons has a much lower detection threshold, which may be important for implementing arousal mechanisms. Intracellularly-recorded neocortical neurons in awake animals indeed show continuous Vm fluctuations (Steriade et al., this meeting). In conclusion, contrary to the intuitive idea that fluctuations would restrain information processing, we show here that Vm fluctuations similar to what is observed in vivo allows neocortical neurons to process a considerably wider range of inputs, therefore enhancing their computational efficiency. Supported by MRC of Canada (MT-13724). |