Elsevier

NeuroImage

Volume 64, 1 January 2013, Pages 43-60
NeuroImage

Variations in the temporal pattern of perforant pathway stimulation control the activity in the mesolimbic pathway

https://doi.org/10.1016/j.neuroimage.2012.09.001Get rights and content

Abstract

Signal processing in the hippocampal formation and resultant signal propagation to cortical and subcortical structures during high frequency stimulation (i.e. 100 Hz) of the perforant pathway was studied in medetomidine anesthetized rats by functional magnetic resonance imaging (fMRI) and electrophysiological recordings. The perforant pathway was stimulated with bursts of 20 pulses, one burst per second, or with continuously applied pulses. The stimulation duration was adjusted to 8 s (short) or 30 s (long). In general, extending the stimulation duration only caused a local spreading of the fMRI response, but no changes in the magnitude of the fMRI response. This was in agreement with the electrophysiological responses, which also remained unchanged. In contrast, increasing the number of pulses in one stimulus train (i.e. changing from burst to continuous stimulation), caused both spreading and an increase in local fMRI responses that were accompanied by an altered neuronal response pattern. Continuous stimulation also triggered additional fMRI responses in the septum, nucleus accumbens, anterior cingulate cortex/medial prefrontal cortex, and ventral tegmental area/substantia nigra. The appearance of fMRI responses outside the hippocampal formation required at least 3 consecutive stimulation trains, characterized by region specific hemodynamic response functions. Thus, once triggered, continuous stimulation caused a sequential appearance in fMRI responses starting in the hippocampal formation, followed by signal changes in the ventral tegmental area/substantia nigra and anterior cingulate cortex/medial prefrontal cortex and eventually in the nucleus accumbens. These results indicate that high frequency stimulation of the hippocampal formation can activate the mesolimbic pathway, provided that repetitive stimulations are applied.

Highlights

► Combination of hippocampal micro-stimulation with fMRI and electrophysiology ► Study the activation pattern caused by burst and continuous 100 Hz stimulation ► Repetitive burst stimulation (20 pulses) trains activate the hippocampal formation. ► Repetitive continuous 100 Hz stimulation trains also activate the mesolimbic system. ► Different hemodynamic response functions in the mesolimbic system and hippocampus.

Introduction

The perforant pathway is a central fiber bundle that connects the entorhinal cortex with the dentate gyrus, CA1–CA3, and subiculum. In particular, cells from layer II of the entorhinal cortex projects to the dentate gyrus and CA3, whereas cells from layer III projects to CA1 and the subiculum. Fibers originating from the lateral entorhinal cortex area project via the lateral perforant pathway to the septal half of the dentate gyrus and there, to the most superficial third of the molecular layer. In contrast, fibers originating from the medial entorhinal cortex project via the medial perforant pathway to the more temporal part of the dentate gyrus and there to the middle third of the molecular layer (Amaral and Lavenex, 2007, Hjorth-Simonsen and Jeune, 1972, Witter, 2007). Thus, the entorhinal cortex is one of the major sources of afferent information to the hippocampus and by that a coordinated activation of the hippocampus by the perforant pathway fibers is crucial for normal hippocampal function.

Electrical stimulation of the perforant pathway with a single pulse will, dependent on the intensity, elicit a response in neurons in the dentate gyrus, hippocampus proper (i.e. CA1–CA3), and subiculum, all of which are targeted by this fiber bundle. The incoming activity will either elicit a restricted variation in the postsynaptic potential or the generation of an action potential and by that a propagation of the activity. Based on the time constants of these processes, a subsequent identical pulse will either elicit a similar or an altered response: i.e. either a stronger (facilitated) or weaker (inhibited) response. Because these neurons form a constituent part of local circuits, they will, as soon as the repetitive stimulus arrives, influence each other. Thus, the magnitude and quality of broad-spectrum neuronal activity during the application of several consecutive pulses is difficult to predict based on the stimulation pattern alone. Based on the prevailing circuit interactions, i.e. feedforward or feedback inhibition, recurrent excitation, or mutual inhibition (Buhl and Whittington, 2007), prolongation of a pulse series can induce both an increase or a decrease in the overall neuronal activity. As a result of such time-dependent variation, target regions of these neuronal circuits eventually become co-activated, or the neuronal activity within the initially activated region becomes reduced. While electrophysiological measurements of field potentials are well suited to monitoring fast changes in neuronal activities in a restricted region (i.e. where the recording electrode is located), they are not very helpful in determining local variations in neuronal activity, particularly when the developing spatial activation pattern is vague and, therefore, an exact placement of the recording electrode is challenging. In addition to commissural connections linking mainly CA3 pyramidal cells with neurons within the contralateral hippocampus, the main output activity originates in the subiculum and reaches a variety of cortical and subcortical structures (Witter, 2006). Thus, monitoring the activation in one or several target regions will give information regarding how incoming pulses, i.e. perforant pathway stimulation, are processed in and eventually propagated to target regions in the hippocampus.

An excellent method for monitoring spatial distribution of altered neuronal activations is functional magnetic resonance imaging (fMRI). However, fMRI visualizes changes in neuronal activity indirectly, via measuring hemodynamic parameters such as local changes in blood oxygen level (BOLD, blood oxygen level dependent), blood flow, or blood volume; therefore, the temporal resolution of fMRI is only in the range of seconds (Attwell and Iadecola, 2002, Ekstrom, 2010, Heeger and Ress, 2002, Nair, 2005). Recently, both approaches were combined and used to monitor changes in neuronal activity, mainly to study the neurophysiological basis of fMRI in the visual cortex of monkeys (Logothetis et al., 2001). An even more direct approach to relate specific variations in neuronal activity to the measured fMRI response was developed for the hippocampal formation (Angenstein et al., 2007). The implementation of a direct electrical stimulation electrode in an afferent fiber bundle, i.e. the perforant pathway, enabled the application of specific stimulation patterns and by that a very specific and adjustable activation pattern. This approach enables not only a co-registration of neuronal activities and resulting variations in hemodynamics, but also offers a visualization of regions that become monosynaptically or polysynaptically activated (Angenstein et al., 2007, Canals et al., 2008).

So far, this method has been used to study how variations in incoming activities control the development of fMRI responses in the hippocampus. Here, we employ this approach to visualize the affect of temporal variations in the perforant pathway stimulation on fMRI responses in brain regions that are connected to the hippocampus.

Section snippets

Animals and surgical procedure

Animals were cared for and used according to a protocol approved by the animal experiment committee, and in conformity with the European convention for the protection of vertebrate animals used for experimental purposes and institutional guidelines 86/609/CEE, November 24, 1986. The experiments were approved by the animal care committee of the State Saxony-Anhalt (No.: 203.h.-42502-2-705IfN). Twenty-eight healthy male Wistar–Han rats (age 9 weeks, body weight 250–300 g) were divided into four

Results

Simultaneous fMRI, microstimulation and electrophysiological measurements were used to monitor how temporal variations of high frequency stimulation (100 Hz) affect hippocampal brain circuits, and by that BOLD responses within and outside the hippocampal formation. The stimuli were modified in two ways, first by increasing the train duration from 8 to 30 s (while keeping the burst duration constant) and second by increasing the burst duration from 200 to 1000 ms (and keeping the train duration

Discussion

This study was aimed at determining how temporal variations in perforant pathway stimulation control the spatial distribution of BOLD responses within the hippocampal formation and other targeted neural regions. Two distinct 100 Hz stimulation paradigms (i.e. burst and continuous stimulation) were applied for either a short (8 s) or a long period (30 s). The main results of this study are: (I) longer presentation of the same stimulus causes a spreading of the BOLD response in the ipsilateral

Acknowledgments

We would like to thank Dr. Jonathan Lovell for critical reading of the manuscript. F.A. was supported by a grant of the Deutsche Forschungsgemeinschaft (DFG – An200-06).

References (29)

Cited by (0)

View full text