Faculty Bibliography

Glutamatergic dentate gyrus (DG) mossy cells (MCs) innervate the primary DG cell type, granule cells (GCs). Numerous MC synapses are on GC proximal dendrites in the inner molecular layer (IML). However, field recordings of the GC excitatory postsynaptic potential (fEPSPs) have not been used to study this pathway selectively. Here we describe methods to selectively activate MC axons in the IML using mice with Cre recombinase expressed in MCs. Slices were made after injecting adeno-associated virus (AAV) encoding channelrhodopsin (ChR2) in the DG. In these slices, we show that fEPSPs could be recorded reliably in the IML in response to optogenetic stimulation of MC axons. Furthermore, fEPSPs were widespread across the septotemporal axis. However, fEPSPs were relatively weak because they were small in amplitude and did not elicit a significant population spike in GCs. They also showed little paired pulse facilitation. We confirmed the extracellular findings with patch clamp recordings of GCs despite different recording chambers and other differences in methods. Together the results provide a simple method for studying MC activation of GCs and add to the evidence that this input is normally weak but widespread across the GC population.

For many years, the hilus of the dentate gyrus (DG) was a mystery because anatomical data suggested a bewildering array of cells without clear organization. Moreover, some of the anatomical information led to more questions than answers. For example, it had been identified that one of the major cell types in the hilus, the mossy cell, innervates granule cells (GCs). However, mossy cells also targeted local GABAergic neurons. Furthermore, it was not yet clear if mossy cells were glutamatergic or GABAergic. This led to many debates about the role of mossy cells. However, it was clear that hilar neurons, including mossy cells, were likely to have very important functions because they provided strong input to GCs. Hilar neurons also attracted attention in epilepsy because pathological studies showed that hilar neurons were often lost, but GCs remained. Vulnerability of hilar neurons also occurred after traumatic brain injury and ischemia. These observations fueled an interest to understand hilar neurons and protect them, an interest that continues to this day. This article provides a historical and personal perspective into the ways that I sought to contribute to resolving some of the debates and moving the field forward. Despite several technical challenges the outcomes of the studies have been worth the effort with some surprising findings along the way. Given the growing interest in the hilus, and the advent of multiple techniques to selectively manipulate hilar neurons, there is a great opportunity for future research.

Neurogenesis occurs in the adult brain in the hippocampal dentate gyrus, an area that contains neurons which are vulnerable to insults and injury, such as severe seizures. Previous studies showed that increasing adult neurogenesis reduced neuronal damage after these seizures. Because the damage typically is followed by chronic life-long seizures (epilepsy), we asked if increasing adult-born neurons would prevent epilepsy. Adult-born neurons were selectively increased by deleting the pro-apoptotic gene Bax from Nestin-expressing progenitors. Tamoxifen was administered at 6 weeks of age to conditionally delete Bax in Nestin-CreER^T2

Maternal choline supplementation (MCS) improves cognition in Alzheimer's disease (AD) models. However, the effects of MCS on neuronal hyperexcitability in AD are unknown. We investigated the effects of MCS in a well-established mouse model of AD with hyperexcitability, the Tg2576 mouse. The most common type of hyperexcitability in Tg2576 mice are generalized EEG spikes (interictal spikes [IIS]). IIS also are common in other mouse models and occur in AD patients. In mouse models, hyperexcitability is also reflected by elevated expression of the transcription factor ∆FosB in the granule cells (GCs) of the dentate gyrus (DG), which are the principal cell type. Therefore, we studied ΔFosB expression in GCs. We also studied the neuronal marker NeuN within hilar neurons of the DG because reduced NeuN protein expression is a sign of oxidative stress or other pathology. This is potentially important because hilar neurons regulate GC excitability. Tg2576 breeding pairs received a diet with a relatively low, intermediate, or high concentration of choline. After weaning, all mice received the intermediate diet. In offspring of mice fed the high choline diet, IIS frequency declined, GC ∆FosB expression was reduced, and hilar NeuN expression was restored. Using the novel object location task, spatial memory improved. In contrast, offspring exposed to the relatively low choline diet had several adverse effects, such as increased mortality. They had the weakest hilar NeuN immunoreactivity and greatest GC ΔFosB protein expression. However, their IIS frequency was low, which was surprising. The results provide new evidence that a diet high in choline in early life can improve outcomes in a mouse model of AD, and relatively low choline can have mixed effects. This is the first study showing that dietary choline can regulate hyperexcitability, hilar neurons, ΔFosB, and spatial memory in an animal model of AD.

Brain-derived neurotrophic factor (BDNF) is important in the development and maintenance of neurons and their plasticity. Hippocampal BDNF has been implicated in Alzheimer's disease (AD) because hippocampal levels in AD patients and AD animal models are often downregulated, suggesting that reduced BDNF contributes to AD. However, the location where hippocampal BDNF protein is most highly expressed, the mossy fiber (MF) axons of dentate gyrus granule cells (GCs), has been understudied, and not in controlled conditions. Therefore, we evaluated MF BDNF protein in the Tg2576 mouse model of AD. Tg2576 and wild-type (WT) mice of both sexes were examined at 2-3 months of age, when amyloid-β (Aβ) is present in neurons but plaques are absent, and 11-20 months of age, after plaque accumulation. As shown previously, WT mice exhibited high levels of MF BDNF protein. Interestingly, there was no significant decline with age in either the genotype or sex. Notably, MF BDNF protein was correlated with GC ΔFosB, a transcription factor that increases after 1-2 weeks of elevated neuronal activity. We also report the novel finding that Aβ in GCs or the GC layer was minimal even at old ages. The results indicate that MF BDNF is stable in the Tg2576 mouse, and MF BDNF may remain unchanged due to increased GC neuronal activity, since BDNF expression is well known to be activity dependent. The resistance of GCs to long-term Aβ accumulation provides an opportunity to understand how to protect vulnerable neurons from increased Aβ levels and therefore has translational implications.

UNLABELLED:Interictal spikes (IIS) and seizures are well-documented in Alzheimer's disease (AD). IIS typically outnumber seizures, supporting their role as a prominent EEG biomarker in AD. In preclinical models, we showed that high frequency oscillations (HFOs>250Hz) also occur, but it is currently unknown how HFOs compare to IIS. Therefore, we asked whether the incidence of HFOs and IIS differed and if they are differentially affected by behavioral state. We used three mouse lines that simulate aspects of AD: Tg2576, presenilin 2 knockout, and Ts65Dn mice. We recorded and quantified HFOs and IIS in the hippocampus during wakefulness, slow-wave sleep, and rapid eye movement sleep. In all three mouse lines, HFOs were more frequent than IIS. High numbers of HFOs correlated with fewer IIS, suggesting for the first time possible competing dynamics among them in AD. Notably, HFOs occurred in more behavioral states than IIS. In summary, HFOs were the most abundant EEG abnormality when compared to IIS, and occurred in all behavioral states, suggesting they are a better biomarker than IIS. These findings pertained to three mouse lines, which is important because they simulate different aspects of AD. We also show that HFOs may inhibit IIS. SHORT SUMMARY/OBJECTIVE:Interictal spikes (IIS) and seizures are common in Alzheimer's disease (AD). IIS are more frequent than seizures and occur during earlier disease stages. In preclinical models, we showed that high frequency oscillations (HFOs>250Hz) occur, but a comparison between IIS and HFOs is lacking. Here we used 3 mouse lines with AD features and local field potential recordings to quantify IIS and HFOs. We found that HFOs outnumbered IIS and that their total numbers were inversely correlated with IIS. HFOs occurred during more behavioral states than IIS. Therefore, HFOs were the most abundant EEG abnormality, and this was generalizable across 3 types of preclinical AD.

Interictal spikes (IIS) are a common type of abnormal electrical activity in Alzheimer's disease (AD) and preclinical models. The brain regions where IIS are largest are not known but are important because such data would suggest sites that contribute to IIS generation. Because hippocampus and cortex exhibit altered excitability in AD models, we asked which areas dominate the activity during IIS along the cortical-CA1-dentate gyrus (DG) dorso-ventral axis. Because medial septal (MS) cholinergic neurons are overactive when IIS typically occur, we also tested the novel hypothesis that silencing the MS cholinergic neurons selectively would reduce IIS. We used mice that simulate aspects of AD: Tg2576 mice, presenilin 2 (PS2) knockout mice and Ts65Dn mice. To selectively silence MS cholinergic neurons, Tg2576 mice were bred with choline-acetyltransferase (ChAT)-Cre mice and offspring were injected in the MS with AAV encoding inhibitory designer receptors exclusively activated by designer drugs (DREADDs). We recorded local field potentials along the cortical-CA1-DG axis using silicon probes during wakefulness, slow-wave sleep (SWS) and rapid eye movement (REM) sleep. We detected IIS in all transgenic or knockout mice but not age-matched controls. IIS were detectable throughout the cortical-CA1-DG axis and occurred primarily during REM sleep. In all 3 mouse lines, IIS amplitudes were significantly greater in the DG granule cell layer vs. CA1 pyramidal layer or overlying cortex. Current source density analysis showed robust and early current sources in the DG, and additional sources in CA1 and the cortex also. Selective chemogenetic silencing of MS cholinergic neurons significantly reduced IIS rate during REM sleep without affecting the overall duration, number of REM bouts, latency to REM sleep, or theta power during REM. Notably, two control interventions showed no effects. Consistent maximal amplitude and strong current sources of IIS in the DG suggest that the DG is remarkably active during IIS. In addition, selectively reducing MS cholinergic tone, at times when MS is hyperactive, could be a new strategy to reduce IIS in AD.

A significant proportion of temporal lobe epilepsy (TLE) patients experience drug-resistant seizures associated with mesial temporal sclerosis, in which there is extensive cell loss in the hippocampal CA1 and CA3 subfields, with a relative sparing of dentate gyrus granule cells and CA2 pyramidal neurons (PNs). A role for CA2 in seizure generation was suggested based on findings of a reduction in CA2 synaptic inhibition (Williamson and Spencer, 1994) and the presence of interictal-like spike activity in CA2 in resected hippocampal tissue from TLE patients (Wittner et al., 2009). We recently found that in the pilocarpine-induced status epilepticus (PILO-SE) mouse model of TLE there was an increase in CA2 intrinsic excitability associated with a loss of CA2 synaptic inhibition. Furthermore, chemogenetic silencing of CA2 significantly reduced seizure frequency, consistent with a role of CA2 in promoting seizure generation and/or propagation (Whitebirch et al., 2022). In the present study, we explored the cellular basis of this inhibitory deficit using immunohistochemical and electrophysiological approaches in PILO-SE male and female mice. We report a widespread decrease in the density of pro-cholecystokinin-immunopositive (CCK⁺) interneurons and a functional impairment of CCK⁺ interneuron-mediated inhibition of CA2 PNs. We also found a disruption in the perisomatic perineuronal net in the CA2 stratum pyramidale. Such pathologic alterations may contribute to an enhanced excitation of CA2 PNs and CA2-dependent seizure activity in the PILO-SE mouse model.SIGNIFICANCE STATEMENT Impaired synaptic inhibition in hippocampal circuits has been identified as a key feature that contributes to the emergence and propagation of seizure activity in human patients and animal models of temporal lobe epilepsy (TLE). Among the hippocampal subfields, the CA2 region is particularly resilient to seizure-associated neurodegeneration and has been suggested to play a key role in seizure activity in TLE. Here we report that perisomatic inhibition of CA2 pyramidal neurons mediated by cholecystokinin-expressing interneurons is selectively reduced in acute hippocampal slices from epileptic mice. Parvalbumin-expressing interneurons, in contrast, appear relatively conserved in epileptic mice. These findings advance our understanding of the cellular mechanisms underlying inhibitory disruption in hippocampal circuits in a mouse model of spontaneous recurring seizures.

HIGHLIGHTS/CONCLUSIONS:Interictal spikes (IIS) occur in 3 mouse lines with Alzheimer's disease featuresIIS in all 3 mouse lines were most frequent during rapid eye movement (REM) sleepThe dentate gyrus showed larger IIS and earlier current sources vs. CA1 or cortexChemogenetic silencing of medial septum (MS) cholinergic neurons reduced IIS during REMMS silencing did not change REM latency, duration, number of bouts or theta power. UNLABELLED:Interictal spikes (IIS) are a common type of abnormal electrical activity in Alzheimer's disease (AD) and preclinical models. The brain regions where IIS are largest are not known but are important because such data would suggest sites that contribute to IIS generation. Because hippocampus and cortex exhibit altered excitability in AD models, we asked which areas dominate the activity during IIS along the cortical-CA1-dentate gyrus (DG) dorso-ventral axis. Because medial septal (MS) cholinergic neurons are overactive when IIS typically occur, we also tested the novel hypothesis that silencing the MS cholinergic neurons selectively would reduce IIS.We used mice that simulate aspects of AD: Tg2576 mice, presenilin 2 (PS2) knockout mice and Ts65Dn mice. To selectively silence MS cholinergic neurons, Tg2576 mice were bred with choline-acetyltransferase (ChAT)-Cre mice and offspring were injected in the MS with AAV encoding inhibitory designer receptors exclusively activated by designer drugs (DREADDs). We recorded local field potentials along the cortical-CA1-DG axis using silicon probes during wakefulness, slow-wave sleep (SWS) and rapid eye movement (REM) sleep.We detected IIS in all transgenic or knockout mice but not age-matched controls. IIS were detectable throughout the cortical-CA1-DG axis and occurred primarily during REM sleep. In all 3 mouse lines, IIS amplitudes were significantly greater in the DG granule cell layer vs. CA1 pyramidal layer or overlying cortex. Current source density analysis showed robust and early current sources in the DG, and additional sources in CA1 and the cortex also. Selective chemogenetic silencing of MS cholinergic neurons significantly reduced IIS rate during REM sleep without affecting the overall duration, number of REM bouts, latency to REM sleep, or theta power during REM. Notably, two control interventions showed no effects.Consistent maximal amplitude and strong current sources of IIS in the DG suggest that the DG is remarkably active during IIS. In addition, selectively reducing MS cholinergic tone, at times when MS is hyperactive, could be a new strategy to reduce IIS in AD.

There has been considerable speculation regarding the function of the dentate gyrus (DG) - a subregion of the mammalian hippocampus - in learning and memory. In this Perspective article, we compare leading theories of DG function. We note that these theories all critically rely on the generation of distinct patterns of activity in the region to signal differences between experiences and to reduce interference between memories. However, these theories are divided by the roles they attribute to the DG during learning and recall and by the contributions they ascribe to specific inputs or cell types within the DG. These differences influence the information that the DG is thought to impart to downstream structures. We work towards a holistic view of the role of DG in learning and memory by first developing three critical questions to foster a dialogue between the leading theories. We then evaluate the extent to which previous studies address our questions, highlight remaining areas of conflict, and suggest future experiments to bridge these theories.