Modeling action potential generation during single and dual electrode stimulation of CA3 axons in hippocampal slice

https://doi.org/10.1016/j.compbiomed.2010.03.003Get rights and content

Abstract

Despite widespread use of electrical stimulation of hippocampal brain slices, the precise sites of action potential initiation and propagation are unknown. To better understand these sites, we modeled axon recruitment by coupling stimulation-induced electric potential fields to biophysical models of CA3 axon arbors in mouse hippocampal brain slices. Geometrically realistic CA3 axon arbors were generated from prior anatomic observations. For single electrode stimulation, 95% of initiation sites were no further than 135 μm away and propagation sites were no further than 875 μm. Using dual electrode stimulation, additional axon arbor pieces were recruited, an effect that increased with distance between stimulating electrodes.

Introduction

Investigation of the hippocampal formation has been critical for understanding aspects of neurophysiology such as synaptic plasticity, neural networks, memory formation and disease states like temporal lobe epilepsies and Alzheimer's disease [1], [2], [3], [4]. A common approach to studying the hippocampal formation is to observe the evoked response of neuronal membranes to an applied electric potential field in the hippocampal brain slice preparation. A prominent target of stimulation is the Schaffer collaterals, a largely unmyelinated axon tract. Experiments involving the stimulation of the Schaffer collaterals are fundamental to our understanding of hippocampus function, yet action potential initiation and propagation sites following stimulation at the resolution of individual axons remain unclear. This uncertainty largely results from the technical difficulty of obtaining intra axonal recordings of the small (<1 μm) CA3 axons [5]. Additionally, the use of microelectrode arrays for brain slice experiments presents the opportunity to stimulate the Schaffer collaterals at multiple sites which may allow for improved recruitment relative to single electrode stimulation paradigms. It is presently unclear to what extent the recruitment of axons can be improved using multiple stimulating electrodes. Computational modeling is well suited for resolving these issues.

Several models of CA3 axon arbors of varying complexity have been presented [6], [7], [8], [9], [10], [11], [12]. The axon geometry in these models range from simple unbranched cables to highly branched tree like structures. Bernard et al. [9] presented the most realistic model of CA3 axon arbors using detailed observations from anatomic data [13], [14]. In addition, the electrogenic properties of membranes in these models range from non-existent to Hodgkin–Huxley ion dynamics which match experimental observations. Prior modeling studies indicate that axon orientation relative to the stimulating electrode [15] and the electrical properties of axon membranes [7], [11], [12] are important determinants for activation. The aforementioned CA3 axon models are likely limited in their ability to predict sites of action potential initiation and propagation following extracellular stimulation because they have either a simplified branch geometry [6], [7], [8], [10], [11], [12] or a simplified model of membrane activation [9]. To date, a model of CA3 axons with accurate branch structure and electrogenic properties has not been created to study the direct effects of extracellular electric stimulation in the slice.

The model presented here fulfills these requirements. This model extends the work of Bernard et al. [9] with the incorporation of branching in the radial direction and associational branches (CA3 to CA3 projections). Also, this model includes electric potential fields around the stimulating electrode and a biophysical membrane model to determine axon membrane responses to the electric potential field [15], [16], [17], [18].

To summarize, the precise sites of action potential initiation and propagation of CA3 axons following extracellular stimulation of the Schaffer collaterals in the hippocampal brain slice are unclear. Also, the potential of using dual stimulating electrode paradigms to improve recruitment is unknown. The model presented here addresses these issues by examining the effects of electrical stimulation applied to a model with realistic membrane geometry and electrogenic properties. The specific goals were to (i) create a set of axon arbors with axon geometry and electrogenic properties resembling those found in CA3 axons in the hippocampal brain slice, (ii) compute the sites of initiation and propagation of action potentials following single electrode extracellular electrical stimulation in these axons and (iii) investigate the differences in action potential initiation and propagation between single and dual electrode stimulation paradigms to illuminate the benefits of using multiple stimulating electrodes in the experimental setting.

Section snippets

Materials and methods

The general structure of the model coupled a finite element model of the stimulation induced electric field to a biophysical model of CA3 axon arbors. The model simulated stimulation of a mouse transverse brain slice using electrodes in a commercially available multi-electrode probe (Med64 system, Panasonic, Japan). The mouse was chosen since this animal has been frequently used as a source of hippocampal brain slices. The CA3 axon arbors used in the model were generated using an algorithm

Results

A model was created to study action potential generation in CA3 axons in the hippocampal brain slice and the effect of multielectrode stimulation paradigms on axon recruitment. First, a set of 10 000 axon arbors were generated which resembled axon arbors found in the hippocampal brain slice. The geometry of generated axon arbors was validated by comparing the branching tendencies of the generated arbors to branching tendencies observed in vivo. The electrogenic properties of generated axons were

Discussion

This study aimed to investigate action potential generation in CA3 axons of the hippocampal brain slice using multielectrode arrays. It was unclear where activated axonal membrane was located in the slice following a single electrode stimulation pulse, and the extent to which dual stimulating electrodes could improve axon recruitment. To address these issues, a novel model of CA3 axon activation was developed, which combined computational modeling of electric potential fields with biophysical

Summary

Despite widespread use of electrical stimulation of hippocampal brain slices, the precise sites of action potential initiation and propagation during stimulation are unknown. Also, the potential of using dual stimulating electrode paradigms to improve recruitment is unknown. The model presented here addresses these issues by examining the effects of electrical stimulation applied to models of CA3 axon arbors with realistic membrane geometry and electrogenic properties. The specific goals were

Conflict of interest statement

None declared.

Acknowledgment

The authors would like to thank the Barrow Neurological Foundation for funding this work.

References (30)

  • J.B. Ranck

    Which elements are excited in electrical stimulation of mammalian central nervous system: a review

    Brain Res.

    (1975)
  • L. Lopez-Aguado et al.

    Activity-dependent changes of tissue resistivity in the CA1 region in vivo are layer-specific: modulation of evoked potentials

    Neuroscience

    (2001)
  • P.A. Schwartzkroin et al.

    Orthodromic and antidromic effects of a cortical epileptiform focus on ventrolateral nucleus of the cat

    J. Neurophysiol.

    (1975)
  • X.J. Wang et al.

    Gamma oscillation by synaptic inhibition in a hippocampal interneuronal network model

    J. Neurosci.

    (1996)
  • J.R. Whitlock et al.

    Learning induces long-term potentiation in the hippocampus

    Science

    (2006)
  • A. Andersen et al.

    The Hippocampus Book

    (2007)
  • G.M. Shepherd et al.

    Three-dimensional structure and composition of CA3–>CA1 axons in rat hippocampal slices: implications for presynaptic connectivity and compartmentalization

    J. Neurosci.

    (1998)
  • F. Pongracz et al.

    A model of NMDA receptor-mediated activity in dendrites of hippocampal CA1 pyramidale neurons

    J. Neurophysiol.

    (1992)
  • R.D. Traub et al.

    A model of a CA3 hippocampal pyramidale neuron incorporating voltage-clamp data on intrinsic conductances

    J. Neurophysiol.

    (1991)
  • R.D. Traub et al.

    A branching dendritic model of a rodent CA3 pyramidale neurone

    J. Physiol.

    (1994)
  • C. Bernard et al.

    Model of spatio-temporal propagation of action potentials in the Schaffer collateral pathway of the CA1 area of the rat hippocampus

    Hippocampus

    (1997)
  • D. Debanne et al.

    Action-potential propagation gated by an axonal I(A)-like K+ conductance in hippocampus

    Nature

    (1997)
  • I.L. Kopysova et al.

    Critical role of axonal A-type K+ channels and axonal geometry in the gating of action potential propagation along CA3 pyramidale cell axons: a simulation study

    J. Neurosci.

    (1998)
  • J.P. Meeks et al.

    Action potential initiation and propagation in CA3 pyramidale axons

    J. Neurophysiol.

    (2007)
  • N. Ishizuka et al.

    Organization of intrahippocampal projections originating from CA3 pyramidale cells in the rat

    J. Comp. Neurol.

    (1990)
  • Cited by (2)

    • Computational Modeling of Subdural Cortical Stimulation: A Quantitative Spatiotemporal Analysis of Action Potential Initiation in a High-Density Multicompartment Model

      2015, Neuromodulation
      Citation Excerpt :

      Studies at the network level, which are difficult in biological systems, are needed in order to fully understand the mechanisms of stimulation. In the past decade, several computer models of the modulatory effects of electrical stimulation have been developed as attempts to understand various factors influencing stimulation and the recruitment of neurons in cortical (9–13), hippocampal (14), and thalamic networks (15,16), as well as in dorsal column fibers (17). All of these studies have dealt mostly with nonactive networks and have focused on current distribution and voltage threshold effects for action potential induction in myelinated or unmyelinated axons (i.e., AIS, nodes of Ranvier).

    View full text