Dendritic Voltage and Calcium-Gated Channels Amplify the Variability of Postsynaptic Responses in a Purkinje Cell Model

De Schutter, Erik. Dendritic voltage and calcium-gated channels dendrites, thus affecting synaptic integration (Bernander et amplify the variability of postsynaptic responses in a Purkinje cell al. 1991; Holmes and Woody 1989; Rapp et al. 1992). It model. J. Neurophysiol. 80: 504–519, 1998. The dendrites of most also has been suggested that synaptic integration by a passive neurons express several types of voltage and Ca-gated channels. dendrite cannot explain the irregular firing observed in neuThese ionic channels can be activated by subthreshold synaptic rons in vivo (Softky and Koch 1993). Recently, it has beinput, but the functional role of such activations in vivo is unclear. come clear that most dendrites are not passive. Instead, denThe interaction between dendritic channels and synaptic backdritic Ca channels have been demonstrated in many neuground input as it occurs in vivo was studied in a realistic computer rons, including cerebellar Purkinje cells (Llinás and model of a cerebellar Purkinje cell. It previously was shown using Sugimori 1980a) and neocortical (Amitai et al. 1993) and this model that dendritic Ca channels amplify the somatic response to synchronous excitatory inputs. In this study, it is shown hippocampal (Regehr and Tank 1992) pyramidal neurons. that dendritic ion channels also increased the somatic membrane Several groups have used the slice preparation to show that potential fluctuations generated by the background input. This amsubthreshold excitatory inputs can activate dendritic Ca plification caused a highly variable somatic excitatory postsynaptic channels (Denk et al. 1995; Eilers et al. 1995; Magee and potential (EPSP) in response to a synchronous excitatory input. Johnston 1995; Markram and Sakmann 1994; Regehr and The variability scaled with the size of the response in the model Tank 1992). A combined experimental and modeling study with excitable dendrite, resulting in an almost constant coefficient demonstrated that background synaptic input can modify the of variation, whereas in a passive model the membrane potential bursting discharges caused by dendritic Ca channels in fluctuations simply added onto the EPSP. Although the EPSP amthalamic reticular neurons (Destexhe et al. 1996). plitude in the active dendrite model was quite variable for different Despite all these experimental studies, the role of dendritic patterns of background input, it was insensitive to changes in the timing of the synchronous input by a few milliseconds. This effect channels in synaptic integration remains unclear. Therefore was explained by slow changes in dendritic excitability. This excita modeling approach was used to study the interaction of ability was determined by how the background input affected the dendritic voltageand Ca-gated channels with background dendritic membrane potentials in the preceding 10–20 ms, causing input in Purkinje cells. The realistic Purkinje cell model with changes in activation of voltage and Ca-gated channels. The active dendritic properties (De Schutter and Bower 1994a,b) most important model variables determining the excitability at the is based on all available experimental data and is not specifitime of a synchronous input were the Ca-activation of K chancally tuned to simulate the responses to synaptic inputs. nels and the inhibitory synaptic conductance, although many other Because it is actually quite successful in reproducing such model variables could be influential for particular background patresponses, the model can be considered to have predictive terns. Experimental evidence for the amplification of postsynaptic power (De Schutter and Bower 1994b). For example, in the variability by active dendrites is discussed. The amplification of the variability of EPSPs has important functional consequences in model focal parallel fiber activation leads to localized voltgeneral and for cerebellar Purkinje cells specifically. Subthreshold, age-gated Ca inflow in the spiny dendrite (De Schutter and background input has a much larger effect on the responses to Bower 1993, 1994c). This model prediction was confirmed coherent input of neurons with active dendrites compared with experimentally by others who used confocal imaging (Eilers passive dendrites because it can change the effective threshold et al. 1995) or dual photon microscopy (Denk et al. 1995) for firing. This gives neurons with dendritic calcium channels an to demonstrate Ca inflow in the spiny dendrite caused by increased information processing capacity and provides the Purparallel fiber activation. The same model also showed that kinje cell with a gating function. background inhibition is essential to obtain the normal irregular in vivo firing pattern of Purkinje cells (De Schutter and Bower 1994b) and that, in fact, on average the inhibitory I N T R O D U C T I O N synaptic current in the Purkinje cell dendrite exceeds the Single neuron physiology and the functional properties total excitatory parallel fiber synaptic current (Jaeger et al. of voltage-gated channels usually are studied in the slice 1997). These predictions were confirmed by blocking inhibipreparation. It is not always clear, however, how such studies tion in vivo, which changes the Purkinje cell firing pattern apply to the in vivo situation, where neurons may very well to the typical regular in vitro one (Jaeger and Bower 1994) behave differently due to the presence of continuous backand by the use of a dynamic clamp in vitro (Jaeger and ground synaptic input (for Purkinje cells, see Fig. 1 of Jaeger Bower 1996). Recently others also showed the importance and Bower 1994; for other neurons, see Destexhe et al. 1996; of inhibition for the irregular spiking of Purkinje cells (HäusParé and Lebel 1997). Theoretical studies have shown that ser and Clark 1997). The present study extends these prior findings by an invesbackground input changes the passive cable properties of