The effects of single and multiple episodes of theta patterned or high frequency stimulation on synaptic transmission from hippocampal area CA1 to the subiculum in rats

Synaptic organisation and behaviour-dependent activity of mGluR8a-innervated GABAergic trilaminar cells projecting from the hippocampus to the subiculum

In the hippocampal CA1 area, the GABAergic trilaminar cells have their axon distributed locally in three layers and also innervate the subiculum. Trilaminar cells have a high level of somato-dendritic muscarinic M2 acetylcholine receptor, lack somatostatin expression and their presynaptic inputs are enriched in mGluR8a. But the origin of their inputs and their behaviour-dependent activity remain to be characterised. Here we demonstrate that (1) GABAergic neurons with the molecular features of trilaminar cells are present in CA1 and CA3 in both rats and mice. (2) Trilaminar cells receive mGluR8a-enriched GABAergic inputs, e.g. from the medial septum, which are probably susceptible to hetero-synaptic modulation of neurotransmitter release by group III mGluRs. (3) An electron microscopic analysis identifies trilaminar cell output synapses with specialised postsynaptic densities and a strong bias towards interneurons as targets, including parvalbumin-expressing cells in the CA1 area. (4) Recordings in freely moving rats revealed the network state-dependent segregation of trilaminar cell activity, with reduced firing during movement, but substantial increase in activity with prolonged burst firing (> 200 Hz) during slow wave sleep. We predict that the behaviour-dependent temporal dynamics of trilaminar cell firing are regulated by their specialised inhibitory inputs. Trilaminar cells might support glutamatergic principal cells by disinhibition and mediate the binding of neuronal assemblies between the hippocampus and the subiculum via the transient inhibition of local interneurons.

Theta-burst LTP

This review covers the spatial and temporal rules governing induction of hippocampal long-term potentiation (LTP) by theta-burst stimulation. Induction of LTP in field CA1 by high frequency stimulation bursts that resemble the burst discharges (complex-spikes) of hippocampal pyramidal neurons involves a multiple-step mechanism. A single burst is insufficient for LTP induction because it evokes both excitatory and inhibitory currents that partially cancel and limit postsynaptic depolarization. Bursts repeated at the frequency (~5 Hz) of the endogenous theta rhythm induce maximal LTP, primarily because this frequency disables feed-forward inhibition and allows sufficient postsynaptic depolarization to activate voltage-sensitive NMDA receptors. The disinhibitory process, referred to as "priming", involves presynaptic GABA autoreceptors that inhibit GABA release. Activation of NMDA receptors allows a calcium flux into dendritic spines that serves as the proximal trigger for LTP. We include new data showing that theta-burst stimulation is more efficient than other forms of stimulation for LTP induction. In addition, we demonstrate that associative interactions between synapses activated during theta-bursts are limited to major dendritic domains since such interactions occur within apical or basal dendritic trees but not between them. We review evidence that recordings of electrophysiological responses during theta burst stimulation can help to determine if experimental manipulations that affect LTP do so by affecting events antecedent to the induction process, such as NMDA receptor activation, or downstream signaling cascades that result from postsynaptic calcium fluxes. Finally, we argue that theta-burst LTP represents a minimal model for stable, non-decremental LTP that is more sensitive to a variety of experimental manipulations than is LTP induced by other stimulation paradigms. This article is part of a Special Issue entitled SI: Brain and Memory.

Neurophysiological correlates of object recognition in the dorsal subiculum

The medial temporal lobe (MTL) encompasses a network of interconnected cortical areas that is considered the neural substrate for some types of memory, such as spatial, episodic, recognition, and associative memory. Within the MTL, the subiculum has been well characterized in terms of its connectivity and structure, but its functional role remains elusive. A long-held view is that the subiculum is mainly involved in spatial encoding because it exhibits spatially selective firing and receives prominent projections from the CA1 field, which is an essential substrate for spatial memory. However, the dorsal subiculum (DS) is also reciprocally connected to the perirhinal and postrhinal cortices, which are critically involved in recognition memory. This connectivity pattern suggests that DS might encode not only spatial signals but also recognition signals. Here, we examined this hypothesis by recording with multi-electrodes in DS and CA1 of freely behaving mice, as they performed the novel object recognition (NOR) task. Analysis of network oscillations revealed that theta power was significantly higher in DS when mice explored novel objects as compared to familiar objects and that this theta modulation was absent in CA1. We also found significant differences in coherence between DS and CA1, in the theta and gamma bands, depending on whether mice examined objects or engaged in spatial exploration. Furthermore, single-unit recordings revealed that DS cells did not exhibit phase-locked firing to theta and differed from CA1 place cells in that they had multiple peaks of spatially selective firing. We also detected DS units that were responsive specifically to novel object exploration, indicating that a subset of DS neurons were tuned to novelty during the NOR task. We have thus identified clear neurophysiological correlates for recognition within the DS, at the network and single-unit levels, strongly suggesting that it participates in encoding recognition-related signals.

Naturalistic stimulus trains evoke reproducible subicular responses both within and between animals in vivo

Previous investigation of CA1-evoked subicular responses has used either single low-frequency pulses (LF), paired-pulses (PP), or high-frequency bursts. Here we test for the first time how subiculum responds to naturalistic stimulation trains (NSTs). We recorded CA1-evoked field potentials from dorsal rat subiculum in response to LF, PP, and two NST patterns. The latter were derived from CA1 place cell activity; NST1 contained bursts of stimuli presented in two main episodes, while the burst-patterned stimuli in NST2 were spaced more evenly. NSTs generated significantly greater field responses compared with LF or PP patterns. Response patterns to either NST were significantly correlated across trial repeats in 9 out of 10 rats, supporting a robust postsynaptic encoding of CA1 input by subiculum. Correlations between NST responses were also observed across experiments; however, these were more variable than those within experiments. The relationship between response magnitude and activation history revealed a strong correlation between magnitude and NST instantaneous frequency for NST1 but was weaker for NST2. In addition, the number of stimuli within a prior 500 ms window was a determining factor for response magnitude for both NSTs. Overall, the robust reproducibility in subicular responses within rats suggests that information within NSTs is faithfully transmitted through the CA1-subiculum axis. However, variation in response sequences across rats suggests that encoding patterns to the same input differ across the subiculum. Changes in the ratio of target bursting and regularly spiking neurons along the subicular proximodistal axis may account for this variation. The activation history of this connection also appears to be a strong determining factor for response magnitude.

The subiculum comes of age

The subiculum has long been considered as a simple bidirectional relay region interposed between the hippocampus and the temporal cortex. Recent evidence, however, suggests that this region has specific roles in the cognitive functions and pathological deficits of the hippocampal formation. A group of 20 researchers participated in an ESF-sponsored meeting in Oxford in September, 2005 focusing on the neurobiology of the subiculum. Each brought a distinct expertise and approach to the anatomy, physiology, psychology, and pathologies of the subiculum. Here, we review the recent findings that were presented at the meeting.

The subiculum: a review of form, physiology and function

We review the neuroanatomical, neurophysiological and functional properties of the mammalian subiculum in this paper. The subiculum is a pivotal structure positioned between the hippocampus proper and entorhinal and other cortices, as well as a range of subcortical structures. It is an under-investigated region that plays a key role in the mediation of hippocampal-cortical interaction. We argue that on neuroanatomical, physiological and functional grounds, the subiculum is properly part of the hippocampal formation, given its pivotal role in the hippocampal circuit. We suggest that the term "subicular complex" embraces a heterogenous range of distinct structures and this phrase does not connote a functionally or anatomically meaningful grouping of structures. The subiculum has a range of electrophysiological and functional properties which are quite distinct from its input areas; given the widespread set of cortical and subcortical areas with which it interacts, it is able to influence activity in quite disparate brain regions. The rules which govern the plasticity of synaptic transmission are not well-specified; it shares some properties in common with the hippocampus proper, but behaves quite differently in other respects. Equally, its functional properties are not well-understood, it plays an important but ill-defined role both in spatial navigation and in mnemonic processing. The important challenges for the future revolve around the theoretical specification of its unique contribution to hippocampal formation processing on the one hand, and the experimental investigation of the many open questions (anatomical, physiological, pharmacological, functional) regarding its properties, on the other.

Long-term potentiation of single subicular neurons in mice

Subicular neurons receive direct afferent connections from the vast majority of CA1 pyramidal cells and send their axons to the various brain areas. Because of this strategic position, subicular cells can modulate output of the hippocampus and, thus, play a significant part in memory, spatial processing, and seizure amplification and propagation from the hippocampus. Despite its important role as a hippocampal interface with different brain regions, present knowledge of the subiculum and the plastic properties of the synapses on the subicular neurons is rather limited. By using IR-DIC videomicroscopy and whole-cell patch-clamp recordings in mouse hippocampal slices, I demonstrated that long-term potentiation (LTP) in CA1-subicular cell synapses can be readily induced by high-frequency stimulation (HFS) of the afferents, but not by pairing of low-frequency stimulation with depolarization of postsynaptic cells. This tetanus-induced LTP is input specific, insensitive to the N-methyl-D-aspartate (NMDA) receptor antagonist 3-[(R)-2Carboxipiperazin-4-yl]-propyl-1-phosphonic acid (R-CPP), and reduces paired-pulse facilitation in potentiated synapses. Subsequent morphologic analysis of the recorded cells, which were filled either with Lucifer Yellow or Biocytin, revealed pyramidal-shaped neurons localized predominantly in the deep layer of the subiculum, close to the CA1 border. Axons of the majority of these neurons extended to the alveus and on toward the hippocampus, probably exiting it via the fornix. These data indicate that CA1-subicular cell synapses in mice exhibit LTP, which can be expressed presynaptically, and its induction does not require NMDA-receptor activation. The observed activity-dependent plasticity might play an important role in the integrative mechanisms of the subiculum and may influence transfer of information from the hippocampus to subcortical and cortical brain areas.

Synaptic plasticity in the hippocampal area CA1-subiculum projection: implications for theories of memory

This paper reviews investigations of synaptic plasticity in the major, and underexplored, pathway from hippocampal area CA1 to the subiculum. This brain area is the major synaptic relay for the majority of hippocampal area CA1 neurons, making the subiculum the last relay of the hippocampal formation prior to the cortex. The subiculum thus has a very major role in mediating hippocampal-cortical interactions. We demonstrate that the projection from hippocampal area CA1 to the subiculum sustains plasticity on a number of levels. We show that this pathway is capable of undergoing both long-term potentiation (LTP) and paired-pulse facilitation (PPF, a short-term plastic effect). Although we failed to induce long-term depression (LTD) of this pathway with low-frequency stimulation (LFS) and two-pulse stimulation (TPS), both protocols can induce a "late-developing" potentiation of synaptic transmission. We further demonstrate that baseline synaptic transmission can be dissociated from paired-pulse stimulation of the same pathway; we also show that it is possible, using appropriate protocols, to change PPF to paired-pulse depression, thus revealing subtle and previously undescribed mechanisms which regulate short-term synaptic plasticity. Finally, we successfully recorded from individual subicular units in the freely-moving animal, and provide a description of the characteristics of such neurons in a pellet-chasing task. We discuss the implications of these findings in relation to theories of the biological consolidation of memory.

Responses of rat subicular neurons to convergent stimulation of lateral entorhinal cortex and CA1 in vivo

There has been little electrophysiological examination of the afferent projection from lateral entorhinal cortex to dorsal subiculum. Here we provide evidence that synaptic inputs from lateral entorhinal cortex and CA1 converge onto single dorsal subicular neurons in vivo. Subicular responses to CA1 stimulation consisted of excitation and/or long-duration inhibition. Neurons excited by CA1 activation usually showed inhibition to entorhinal stimulation. The latter inhibition was usually of short duration, however, long duration inhibition was seen in a significant proportion of responses. Entorhinal stimulation produced excitatory responses in four bursting cells and it was these cells that also tended to show the longest inhibition. Only bursting cells could be driven antidromically by entorhinal stimulation. Biocytin-filled multipolar and pyramidal cells displayed excitation-inhibition sequences to CA1 and inhibition to entorhinal stimulation. These data strongly suggest that subicular inhibitory neurons receive excitatory input from CA1 and display mutual inhibition. The source of entorhinal-evoked inhibition is less clear. The relative sparseness of observed entorhinal-evoked responses suggests that the input to dorsal subiculum from any one part of lateral entorhinal cortex is spatially restricted. These data show that excitation-inhibition sequences can be seen in subicular pyramidal and multipolar cells and that single subicular neurons receive convergent inputs from CA1 and entorhinal cortex. We show for the first time that bursting cells can be driven both orthodromically and antidromically by direct entorhinal stimulation. These data support the existence of a reciprocal excitatory connection between lateral entorhinal cortex and dorsal subiculum and suggest further that this connection may involve only bursting subicular neurons.

Acetylcholinesterase-positive fiber deafferentation and cell shrinkage in the septohippocampal pathway of aged amyloid precursor protein london mutant transgenic mice

Several lines of evidence implicate a cholinergic deficit in Alzheimer's disease (AD). Transgenic mice that overexpress clinical mutants of the human amyloid precursor protein (APP) have been generated that recapitulate many aspects of AD. We now analyzed the cholinergic system in aged APP/London transgenic mice. The major finding was the reorganization of acetylcholinesterase-positive fibers within the hippocampus and the reduced size of cholinergic cells in the medial septum. The reduction of acetylcholinesterase-positive fibers in the subiculum together with increased fiber density in the CA1 and in the dentate gyrus suggests a synaptic sprouting compensatory mechanism within the hippocampus. In the cortex, amyloid plaques were associated with intense acetylcholinesterase activity and surrounded by dystrophic acetylcholinesterase-positive fibers. Nevertheless, the overall pattern of cholinergic innervation was unchanged. These results demonstrate that overexpression of APP/London caused, besides amyloid plaques in aged mouse brain, also cholinergic deafferentation and cholinergic cell shrinkage.

The effects of low frequency and two-pulse stimulation protocols on synaptic transmission in the CA1-subiculum pathway in the anaesthetized rat

The downregulation of synaptic efficacy is referred to as long-term depression (LTD). Recent work has shown that a two-pulse stimulation (TPS) protocol is successful at inducing LTD in vivo in area CA1 of the hippocampus. Here, we examine the ability of two TPS protocols and two low-frequency stimulation (LFS) protocols to induce LTD in the projection from hippocampal area CA1 to the subiculum in the anaesthetized rat. We find no evidence of LTD induction with TPS or LFS protocols. Instead, with three of the protocols (both TPS protocols and 1 Hz LFS), a late-developing potentiation is observed.