Active dendrites regulate spatio-temporal synaptic integration in hippocampal dentate granule cells

doi:10.1016/S0925-2312(99)00087-9

Neurocomputing

Volumes 26–27, June 1999, Pages 45-51

https://doi.org/10.1016/S0925-2312(99)00087-9 Get rights and content

Abstract

Compartmental modeling experiments were carried out to compare spatio-temporal synaptic integration in a model of a fully reconstructed hippocampal dentate granule (DG) cell containing either passive dendrites or dendrites with voltage-dependent conductances. The presence of active channels in dendrites increased the elastic properties of dendritic membrane, increased the sensitivity of the response to synapse clustering (up to an optimal degree of clustering), decreased the input location dependent variability of the response at the soma, and facilitated coincidence detection in the dendritic arbor. We conclude that active channels in dendrites may alter significantly both spatial and temporal integrative processes.

Introduction

Dendrites have been treated traditionally as passive cables and synaptic integration in passive dendrites has been studied extensively [7], [8]. However, recent experimental evidence suggests that dendrites are not passive and this has inspired new modeling studies of synaptic integration [6], [3]. These new studies have focused on the spatial aspect of synaptic integration. However, temporal summation must also be influenced by active channel properties. In this study, compartmental modeling experiments were carried out to investigate spatio-temporal synaptic integration in a fully reconstructed hippocampal dentate granule cell. We compared simulation results obtained from models with active or passive dendrites and found that the existence of voltage-dependent conductances in dendrites may alter significantly both spatial and temporal integrative processes.

Section snippets

Methods

A model with a fully reconstructed dendritic tree supplemented with 4000 dendritic spines was used to study the possible mechanisms underlying the dendritic integration processes in dentate granule cells. To simulate the active properties of DG cells, nine channel types were included in the model: Na, fast and slow K $_{DR}, K_{A}$ , BK, SK, and T-, N-, and L-type calcium channels. Channel densities were chosen to reproduce the characteristic responses of DG cells to current injections and synaptic

Active channels reduce variability of response to different strength stimuli

Responses of models with active or passive dendrites were compared following single pulse or tetanic (8 pulse 400 Hz) synaptic activation of 270 synapses located within the middle third of the dendritic tree. Dendritic spine membrane was either passive or contained 5 T-type or 10 N-type calcium channels. Models with active or passive dendrites and dendritic spines responded very differently to applied stimuli. A short current stimulus or a single synaptic activation (Fig. 1A) evoked an action

Discussion

Models of hippocampal dentate gyrus granule cells with active or passive dendrites were created to study the role of dendritic voltage-activated conductances in the integrative processes of this cell type. We found that active dendrites, by enhancing weak synaptic inputs and attenuating strong synaptic inputs, can reduce the variability that synaptic stimulations with different strength have on somatic depolarization. In agreement with previous studies for hippocampal pyramidal cells [3], input

Acknowledgments

This work was supported by National Institute of Mental Health Grant MH-51081 to W.R. Holmes. We thank W.B. Levy and N.L. Desmond for the anatomical data of the dentate granule cell used in the simulations.

Ildikó Aradi received an M.Sc. degree in Physics and a Ph.D. degree in Neurobiology from Eötvös University, Hungary. Her doctoral thesis was on modeling of single cells and neural circuits of the olfactory bulb. She has been a postdoctoral fellow at Ohio University since 1997. Her research interests include computational modeling of hippocampal neurons and neural networks. Her current work is focusing on detailed modeling of different cell types of dentate gyrus in order to study and

References (8)

I. Aradi, W.R. Holmes, Role of multiple calcium and calcium-dependent conductances in regulation of hippocampal dentate...
J.M. Bower, D. Beeman, The Book of GENESIS, Santa Clara, CA, TELOS, The Electronic Library of Science, Springer,...
E.P. Cook et al.
Active dendrites reduce location-dependent variability of synaptic input trains
J. Neurophysiol.
(1997)
M. Hines, Computer modeling methods for neurons, in: M.A. Arbib (Ed.), The Handbook of Brain Theory and Neural...

There are more references available in the full text version of this article.

Cited by (4)

Modeling Circuit Alterations in Epilepsy: A Focus on Mossy Cell Loss and Mossy Fiber Sprouting in the Dentate Gyrus
2008, Computational Neuroscience in Epilepsy
This chapter provides an overview of the general approach and considerations for developing a large-scale model to study epileptogenesis with the focus on simulating the effects of mossy fiber sprouting and hilar cell loss in the dentate circuits. It presents the results of simulation studies to highlight the applicability of the technique and insights obtained based on such large-scale modeling approaches. Understanding a complex network phenomenon like epilepsy requires a conceptual framework that links the biophysical properties of individual neurons and the circuit where the disease manifests. Computational modeling provides a means to bridge these levels and is ideally suited to study dynamical diseases such as epilepsy. There is increasing physiological evidence for a role for mossy cells in spreading hyperexcitability through their long-range connections. Implementation of large-scale models of the dentate gyrus allows for examining the net effect of these distinctive network alterations on the function of the dentate “gate” independent of changes intrinsic, extracellular or ephaptic properties. The novel insights into the development and progression of epilepsy revealed by simulation studies offer a compelling case for the use of data-driven simulations in understanding complex network diseases such as epilepsy. It is anticipated that network modeling methods will make it possible to test the circuit effects of specific therapeutic manipulations and allow for uniquely tailoring treatment to the distinct underlying cause of a seizure disorder.
Modeling the dentate gyrus
2007, Progress in Brain Research
Citation Excerpt :
First, since GCs (Desmond and Levy, 1985) and MCs (Amaral, 1978) are rich in dendritic spines, it is necessary to account for the membrane area contribution of the spines (Rall et al., 1992). This can be accomplished by decreasing membrane resistivity (increasing leak conductance) and increasing membrane capacitance (Aradi and Holmes, 1999b). An additional consideration is the presence of spontaneous activity in some cell types such as MCs (Ishizuka et al., 1995; Ratzliff et al., 2004), as well as their lower input resistance in the presence of background synaptic activity (Ratzliff et al., 2004), compared to the input resistance in ionotropic glutamate and GABA receptor antagonists (Lubke et al., 1998).
Computational modeling has become an increasingly useful tool for studying complex neuronal circuits such as the dentate gyrus. In order to effectively apply computational techniques and theories to answer pressing biological questions, however, it is necessary to develop detailed, data-driven models. Development of such models is a complicated process, akin to putting together a jigsaw puzzle with the pieces being such things as cell types, cell numbers, and specific connectivity. This chapter provides a walkthrough for the development of a very large-scale, biophysically realistic model of the dentate gyrus. Subsequently, it demonstrates the utility of a modeling approach in asking and answering questions about both healthy and pathological states involving the modeled brain region. Finally, this chapter discusses some predictions that come directly from the model that can be tested in future experimental approaches.
Temporal sequence detection with spiking neurons: Towards recognizing robot language instructions
2006, Connection Science
Modulation of network behaviour by changes in variance in interneuronal properties
2002, Journal of Physiology

Bill Holmes received his Ph.D. from the Department of Biomathematics at UCLA. Currently he is an Associate Professor in the Neurobiology Program in the Department of Biological Sciences at Ohio University. His research interests include the development of mathematical and computational models of individual neurons of the hippocampus that will be appropriate for use in network models to explore hippocampal function. The immediate focus is to develop models of dentate granule cells that describe how computation and synaptic modification occur in these cells.

View full text