Hippocampal Interneurons and Their Role in the Control of Network Excitability

Chris J. McBain, PhD, Head, Section on Cellular and Synaptic Physiology
J. Joshua Lawrence, PhD, Staff Scientist
Kenneth Pelkey, PhD, Senior Fellow
Michael Daw, PhD, Visiting Fellow
Brian Erkkila, PhD, Visiting Fellow
Christine Torborg, PhD, Visiting Fellow
Ludovic Tricoire, PhD, Visiting Fellow
Brian Jefferies, BS, Biologist
Xiaoqing Yuan, MSc, Biologist
Christian Cea del Rio, BS, Graduate Student
Tsz-wan Michelle Ho, BS, Graduate Student

GABAergic inhibitory interneurons constitute a population of hippocampal neurons whose high degree of anatomical and functional divergence make them suitable candidates for controlling the activity of large populations of principal neurons. Interneurons (INT) receive strong excitatory glutamatergic innervation via numerous anatomically distinct afferent projections; recent evidence has demonstrated that the molecular composition of the AMPA-preferring class of glutamate receptors expressed at INT synapses is often distinct from that of receptors found at principal cell synapses. Our goal is to identify the types of ligand-gated ion- and voltage-gated ion channels—and the roles they play—found on neurons of the mammalian hippocampus and cortex. Specifically, we strive to understand the functions of these receptors in regulating synaptic and neuronal excitability, short- and long-term plasticity, and the development and pathophysiology of the mammalian cortical structure. Through this research, we envision a better understanding of the specific mechanisms of the regulation of excitability in precise cohorts of central nervous system cells and thus more targeted and valid methods of treatment of a wide array of central nervous system disease states.

State-dependent cAMP sensitivity of presynaptic function underlies metaplasticity in a hippocampal feedforward inhibitory circuit

Pelkey; in collaboration with Lacaille, Topolnik

Synapses may reside along a continuum of discrete plastic competency states dictated by their activation history; in turn, the states govern the polarity, magnitude, duration, and cell-signaling cascades invoked during future plasticity episodes. Metaplasticity is the term describing such higher-order regulation of the ability to induce synaptic plasticity; it may preserve homeostasis by preventing neural connections from accumulating within saturated states of potentiation or depression while enabling synaptic integration across prolonged temporal domains. Metaplasticity is well documented for postsynaptically expressed NMDAR-dependent forms of long-term plasticity; however, metaplasticity of presynaptically expressed forms of long-term plasticity has received less attention. Recently, we discovered a striking form of presynaptic metaplasticity at mossy fiber–stratum lucidum INT (MF-SLIN) synapses of the hippocampus (Pelkey et al., Neuron 2005;46:89). At naive MF-SLIN terminals, presynaptic mGluR7 activation during high-frequency stimulation (HFS) yields long-term depression (LTD). However, following agonist-induced internalization of mGluR7, the same HFS protocol results in a presynaptically expressed form of long-term potentiation (LTP). Thus, at these synapses, mGluR7 functions as a metaplastic switch whose surface expression and activation govern bidirectional plasticity. Surprisingly, our recent findings indicate that mGluR7 activation/internalization controls the polarity of MF-SLIN plasticity by gating the cAMP sensitivity of MF-SLIN presynaptic release. While naive surface mGluR7–expressing MF-SLIN synapses are completely insensitive to cAMP elevation, synapses that have internalized mGluR7 robustly potentiate following cAMP increases. Moreover, HFS-induced LTP at MF-SLIN synapses that have internalized mGluR7 requires adenylate cyclase (AC) activity. Intriguingly, we also discovered an in vivo association between mGluR7 and RIM1α—a PKA substrate required for presynaptic LTP; the association is regulated by mGluR7 activation/internalization. These findings reveal that a state-dependent cAMP sensitivity controlled by mGluR7-RIM1α interactions underlies presynaptic MF-SLIN metaplasticity. This mGluR7-controlled reversal in polarity of MF-SLIN plasticity has dramatic consequences for hippocampal information propagation, as the MF-SLIN pathway provides a highly efficient feedforward inhibitory circuit controlling dentate granule cell–mediated recruitment of the auto-associative CA3 pyramidal cell network.

Pelkey KA, McBain CJ. Differential regulation of functionally divergent release sites along a common axon. Curr Opin Neurobiol 2007;17:366-373.
Pelkey KA, McBain CJ. Target-cell dependent plasticity within the mossy fiber-CA3 circuit reveals compartmentalized regulation of presynaptic function at divergent release sites. J Physiol
Pelkey KA, Topolnik L, Lacaille JC, McBain CJ. Compartmentalized Ca²⁺ channel regulation at functionally divergent release sites of single mossy fibers underlies target cell-dependent plasticity. Neuron 2006;52:497-510.

Co-requirements of PICK1 binding and PKC phosphorylation for stable surface expression of mGluR7

Pelkey; in collaboration with Roche, Huganir

As described above, the presynaptic metabotropic glutamate receptor (mGluR) mGluR7 modulates excitatory neurotransmission by regulating neurotransmitter release and plays a critical role in certain forms of synaptic plasticity at the mossy fiber synapse. Although the dynamic regulation of mGluR7 surface expression governs a form of metaplasticity in the hippocampus, little is known about the molecular mechanisms regulating mGluR7 trafficking. We showed that mGluR7 cell-surface expression is stabilized by both protein kinase C phosphorylation and receptor binding to the PDZ domain–containing protein PICK1. Phosphorylation of mGluR7 on serine 862 (S862) inhibits CaM binding, thereby increasing mGluR7 surface expression and receptor binding to PICK1. Furthermore, in mice lacking PICK1, PKC-dependent increases in mGluR7 phosphorylation and surface expression are diminished, and mGluR7-dependent plasticity at mossy fiber-INT hippocampal synapses is impaired. These data support a model in which PICK1 binding and PKC phosphorylation act together to stabilize mGluR7 on the cell surface in vivo.

Pelkey KA, Yuan X-Q, Lavezzari G, Roche KW, McBain CJ. mGluR7 undergoes rapid internalization in response to activation by the allosteric agonist AMN082. Neuropharmacology 2007;52:108-117.
Suh YH, Pelkey KA, Lavezzari G, Roche PA, Huganir RL, McBain CJ, Roche KW. Co-requirements of PICK1 binding and PKC phosphorylation for stable surface expression of the metabotropic glutamate receptor mGluR7. Neuron 2008;58:736-748.

Depolarization-induced LTD (DiLTD) at immature mossy fiber–CA3 pyramid synapses produced by postsynaptic burst-firing–like protocols

Ho, Pelkey; in collaboration with Huganir

Elucidation of the cellular mechanisms of synapse maturation is crucial to understanding central nervous system development. Mossy fiber–CA3 pyramid (MF–PYR) synapses develop postnatally, providing a tractable model of synapse maturation in a defined circuit. Recently, we demonstrated that, in early postnatal development, AMPARs supporting MF-PYR transmission undergo a subunit switch: immature synapses contain mixed populations of GluR2-lacking and GluR2-containing AMPARs and then mature to a state dominated by GluR2-containing AMPARs. A similar subunit switch occurs during DiLTD, a postsynaptic form of plasticity observed only at immature MF-PYR synapses. Sustained PYR depolarization (3 to 5 minutes) triggers DiLTD to promote Ca²⁺ influx through L-type voltage-gated Ca²⁺ channels (L-VGCCs). Interestingly, the developmental window of DiLTD competence coincides with the emergence of action potential (AP) burst firing (BF) in PYRs, suggesting an attractive physiological mechanism to induce DiLTD. Using BF-like protocols in acute hippocampal slices, we therefore investigated whether phasic AP-dependent L-VGCC activation can also trigger DiLTD. To elicit PYR BF, we repetitively injected a depolarizing current pulse (125 ms, 0.5 nA) at 1 Hz for 5 minutes in current clamp mode. The protocol reliably produced AP bursts (4 to 5 APs per pulse) and persistently depressed MF-PYR synapses. However, AP blockade by intracellular QX-314 did not prevent this LTD, suggesting that the protocol directly activates L-VGCCs independently of APs. Indeed, in the absence of APs, the current step typically depolarized PYRs sufficiently to activate L-VGCCs while the L-VGCC blocker nifedipine prevented LTD. To avoid direct L-VGCC activation, we delivered brief current pulses (5 pulses of 0.8 nA, 5 ms duration, 25 ms apart) at 1 Hz for 5 minutes. This paradigm produced robust LTD that was prevented by either intracellular QX-314 or the L-VGCC blocker nifedipine. Like DiLTD, AP-induced LTD exhibited postsynaptic expression. We conclude that DiLTD is induced by phasic L-VGCC activation driven by trains of PYR APs, suggesting a previously unsuspected role for PYR BF in MF-PYR synapse plasticity and maturation.

Ho TW, Pelkey KA, Topolnik L, Petralia RS, Takamiya K, Xia J, Huganir RL, Lacaille J-C, McBain CJ. Developmental expression of Ca²⁺-permeable AMPARs underlies depolarization-induced LTD at mossy fiber-CA3 pyramid synapses. J Neurosci 2007;27:11663-11675.

The balance of excitatory and inhibitory synaptic currents is maintained during short-term plasticity in the CA3 hippocampus

Torborg

Strong facilitation during repetitive stimulation is a defining feature of the synapse between the excitatory granule cell (GC) and CA3 pyramidal cell (PC) (Lawrence et al., J Physiol 2004;554.1:175). However, such a strong excitatory drive makes the network susceptible to hyperexcitability. Inhibition is one mechanism for tempering this potentially catastrophic circuit arrangement. PCs receive both GC-driven feedforward and PC-driven feedback inhibition from local inhibitory INTs. Using whole-cell voltage-clamp techniques, we tested the hypothesis that feedforward and feedback inhibition temporally overlap to regulate circuit excitability during short-term plasticity. Low-intensity extracellular stimulation of the GC layer elicited both monosynaptic EPSCs (excitatory postsynaptic currents) and polysynaptic IPSCs (inhibitory postsynaptic currents) in PCs. Surprisingly, the total inhibitory charge transfer progressively increased during a 20 Hz train of stimuli, balancing the increase in excitatory charge transfer and maintaining a constant excitation:inhibition (E:I). Given that the IPSC onset was delayed relative to the EPSC, the latter’s peak amplitude was able to increase progressively during the train. Thus, excitation can facilitate an increase in spike probability even as the constant E:I ratio maintains a narrow temporal window and prevents hyperexcitability. Next, we sought to determine where within the inhibitory circuit such facilitation occurred. Consistent with previous results, we found that GC-INT synapses were either facilitating or depressing in response to the short train of stimuli. Post hoc morphological analysis revealed no correlation between the type of short-term plasticity and INT morphology. We then examined the INT-PC synapse and observed that the total inhibitory charge transfer was either depressed or did not change during the train for stimulation in all layers, suggesting that depression is the predominant form of short-term plasticity at that synapse. Taken together, our results suggest that INTs that receive facilitating synapses from GCs provide a larger contribution to polysynaptic inhibition than those receiving depressing synapses; moreover, such facilitation is sufficient to mask any depression at the INT-PC synapse. However, given the high degree of facilitation onto PC, feedback inhibition is likely to contribute to the facilitation of inhibition during repetitive stimulation.

A transgenic approach to elucidating hippocampal mossy fiber function

Pelkey; in collaboration with Tonegawa

As the primary connection between dentate gyrus granule cells and the CA3 region of the hippocampus, mossy fibers (MF) provide a critical pathway linking cortical activity with intrinsic hippocampal circuits. Thus, MF activity dramatically affects overall hippocampal excitability and information processing. In collaboration with Susumo Tonegawa, we are attempting to determine the role of MF transmission in hippocampal-dependent learning paradigms; we are using transgenic mice that conditionally lack synaptic transmission specifically in the MF pathway. The mice were genetically engineered to express tetanus toxin light chain (TeTX) specifically in dentate gyrus granule cells under the control of a doxycycline-sensitive promoter region. As tetanus toxin effectively blocks presynaptic release of transmitter by cleaving the vesicle-associated protein VAMP, the mice should allow us to functionally knock out the mossy fiber pathway in a temporally controlled and reversible manner while leaving all other hippocampal circuitry intact. Indeed, our preliminary in vitro electrophysiological characterization of the mice confirms that MF-CA3 pyramidal cell transmission is efficiently blocked in mice induced to express TeTX and that the blockade is fully reversible upon treatment of the animals with a diet to repress TeTX expression. Moreover, synaptic transmission in all other hippocampal pathways (perforant path, Schaffer collateral, and associational commissural pathways) remains indistinguishable between MF-TeTX–expressing and wild-type control animals. Future behavioral work with the mice will greatly expand our understanding of MF function in hippocampus-related learning and disease.

Varying proportions of asynchronous transmitter release from identified hippocampal GABAergic INTs

Daw

Chemical synaptic transmission classically occurs when neurotransmitter is released from a presynaptic terminal in response to an AP in a tightly time-locked fashion. Previously, it has been shown at several central synapses that, in addition to synchronous release, some transmitter may be released in a longer and less precise time window such that release is asynchronous and not locked to individual presynaptic triggers. A variety of classes of local GABAergic INTs mediate inhibitory synaptic transmission in the hippocampus. In the hippocampal dentate gyrus, basket cells (BC) that contain cholecystokinin (CCK) show a high proportion of asynchronous release onto principal cells in response to long presynaptic trains while those that contain parvalbumin (PV) exhibit highly synchronous release. Using paired INT-INT and INT-PC recordings, we further studied this cell-type specificity in terms of both pre- and postsynaptic cell type in the CA1 and CA3 regions of the hippocampus. We identified non–fast-spiking INTs by expression of GFP in a mouse line that expresses GFP under the control of the GABA-synthetic enzyme GAD-65 (GAD-65 GFP) promoter. We recorded fast-spiking INTs in either wild-type mice or mice expressing GFP under the promoter for PV (B-13 line).

As in the dentate gyrus, presumed CCK-containing, non–fast-spiking BCs showed a much higher proportion of asynchronous release than that shown by presumed PV-containing, fast-spiking BCs in a manner unaffected by the identity of the postsynaptic cell (INT or pyramid). We studied two other classes of INT: bistratified and trilaminar cells. Both cell types showed an even higher degree of asynchronous release than either BC class, again regardless of postsynaptic cell type. To study the mechanisms of the two modes of release, we lowered the release probability by decreasing the extracellular Ca²⁺ concentration. This manipulation reduced each mode of release equally, suggesting a common mechanism. Our observation that the N-type calcium channel blocker ω-conotoxin GVIA completely blocked both modes of release supports a common mechanism.

Cell type–specific kainate receptor modulation of GABAergic transmission in the hippocampus

Daw

A variety of classes of local GABAergic INTs mediate inhibitory synaptic transmission in the hippocampus. Activation of the kainate (KA) class of ionotropic glutamate receptors reduces inhibitory transmission onto hippocampal PCs but augments inhibitory transmission onto INTs. However, the classes of INT affected by KA receptor activation have yet to be investigated. Therefore, we used paired INT-INT and INT-PC recordings to explore the INTs affected by KA receptor activation.

KA application had no effect on unitary inhibitory postsynaptic current (uIPSC) amplitude between fast-spiking BC INTs and PCs while uIPSC amplitude in PCs mediated by non–fast-spiking BCs and bistratified cells was halved. Changes in the paired-pulse ratio and coefficient of variation suggest that the reduction was presynaptic in origin; a similar depression of uIPSC amplitude by KA in the presence of CGP 55845 (30 percent) showed that the reduction was not attributable to indirect activation of GABAB receptors, as previously reported.

The effects of KA synaptic transmission between non–fast-spiking INTs were more complex but appeared to be related to the short-term plasticity displayed by the synapse irrespective of INT class. Kainate application inhibited synapses that showed pronounced facilitation throughout long bursts of presynaptic stimulation (25 APs) while synapses that depressed during bursts either did not respond to kainate or did not potentiate. As with PCs, changes in the paired-pulse ratio were inversely correlated with changes in the first IPSC amplitude, suggesting a presynaptic site of action. Surprisingly, however, plasticity throughout longer bursts did not change. Experiments in which presynaptic release probability was directly reduced by lowering the extracellular calcium concentration showed that only large (greater than 60 percent) reductions in the amplitude of the first IPSC produced changes in plasticity during bursts. The results suggest that, in many experiments, the kainate-induced depression was too small to affect plasticity in bursts. Our experiments show that activation of presynaptic KA receptors on hippocampal INTs can modulate inhibitory transmission in a manner that is dependent on both pre- and postsynaptic cell type.

Hippocampal INT subtypes are associated with distinct muscarinic receptor mRNA expression profiles

Lawrence, Tricoire, Cea del Rio

INTs in neocortex and hippocampus differentially undergo muscarinic receptor (mAChR) activation, but it is not clear whether muscarinic phenotypes of INTs arise from differences in the expression or localization of mAChRs, intrinsic voltage-gated ion channels, calcium-buffering capacity, or some combination of these factors. Using single-cell PCR, biocytin labeling, and transgenic mice expressing GFP in parvalbumin- (PV, line B13) or GAD65-expressing INTs, we determined mAChR mRNA expression profiles (M1R–M5R) in morphologically, electrophysiologically, and neurochemically identified INT populations. We based neurochemical profiles for each INT on nine neurochemical markers: GAD67, GAD65, calretinin, PV, calbindin (CB), somatostatin (SOM), neuropeptide Y (NPY), vasoactive intestinal peptide (VIP), and CCK. In all CA1 PV-GFP cells identified as fast-spiking BCs, we detected PV mRNA but not CB, CCK, VIP, or SOM mRNA. PV⁺ BCs expressed M1 (100 percent), M2 (40 percent), M3 (20 percent), and M4 (60 percent) muscarinic receptor mRNA. In contrast, CA1 GAD65 GFP⁺ INTs expressed abundant CCK (92 percent), and CB (69 percent) mRNA, as did non–fast-spiking Schaeffer collateral–associated and CCK BC populations. We probed a subset of GAD65 GFP⁺ cells for mAChR mRNAs. Similar to PV⁺ BCs, M1 mRNA was present in all GAD65 GFP⁺ cells. In contrast, M2 mRNA was absent while M3 mRNA was expressed in higher relative abundance (71 percent) than in PV-GFP⁺ cells. The data are consistent with M1 antagonist sensitivity to mAChR responses in PV-GFP and GAD65 cell types. Immunocytochemical evidence for M2Rs on PV⁺ boutons prompted us to examine the functional consequence of M2Rs on PV-GFP INTs. In PV⁺ BC–pyramidal cell pairs, bath application of the muscarinic receptor agonist muscarine reduced the unitary IPSC amplitude in wild-type, but not in PV-GFPxM2 KO, mice. Thus, our data provide evidence not only of the differential expression of mAChR subtypes between INT subtypes but also for differential targeting within an INT subtype.

Neurochemical identity governs cholinergic phenotype across hippocampal BC networks

Cea del Rio, Lawrence, Tricoire

Inhibitory INT BCs release GABA perisomatically onto PCs in order to synchronize PCs during gamma (γ; 40 Hz) oscillations. Two major BC subtypes are present in the hippocampus: CCK- and PV-BCs. In the neocortex, CCK- and PV-BCs possess distinct cholinergic phenotypes; however, the cholinergic neuromodulation of hippocampal CCK- and PV-BCs remains poorly understood. To examine differential cholinergic modulation in BC subtypes, we performed whole-cell recordings in hippocampal slices from GAD65- or PV-GFP transgenic mice and morphologically identified CCK- and PV-BCs, respectively. To determine neurochemical profiles and muscarinic-receptor (M1–M5) content, we also performed single-cell RT-PCR experiments. First, upon application of muscarine, CCK-BCs underwent a transient hyperpolarization while PV-BCs underwent depolarization. Second, mAChR activation induced an increase in firing frequency in response to depolarizing current steps in both CCK- and PV-BCs. However, the offset of the current stimulus elicited a larger mAChR-induced afterdepolarization (ADP) in CCK-BCs. To determine whether cholinergic neuromodulation tunes CCK- and PV-BCs differently, we systematically varied the frequency (1to 100 Hz) and number of APs (1 to 100 APs at 20 Hz) elicited in both BC subtypes. In CCK-BCs, the mAChR-induced ADP (mADP) was largest at 20 Hz and could be generated with as few as 10 APs. In contrast, the mADP of PV-BCs exhibited no frequency or AP preference.

The specific M1-mAChR antagonist telenzepine prevented the mADP but not the rise in firing frequency, suggesting the presence of both M1 and non–M1 mAChR subtypes. In contrast, in PV-BCs, telenzepine completely blocked both the mADP and an increase in firing. In addition, GAD65 x M3KO transgenic mice demonstrated that the mAChR-induced change in firing frequency and mAChR-induced ADP were lower than in wild-type animals. Finally, consistent with pharmacological data, single-cell RT-PCR confirmed that M1 receptor transcripts were present in all CCK- and PV-BCs and revealed that M3 receptors were more abundant in CCK-BCs cells than in PV-BCs. The data suggest that M1 and M3 mAChRs play synergistic roles in mAChR-induced changes in firing and the generation of ADP.

Lawrence JJ. Cholinergic control of GABA release: emerging parallels between neocortex and hippocampus. Trends Neurosci 2008;31:317-327.
Lawrence JJ, Grinspan ZM, Statland JM, McBain CJ. Muscarinic receptor activation tunes mouse stratum oriens interneurons to amplify spike reliability. J Physiol 2006;571:555-562.
Lawrence JJ, Statland JM, Grinspan ZM, McBain CJ. Cell type-specific dependence of muscarinic signaling in mouse hippocampal stratum oriens interneurons. J Physiol 2006;570:595-610.

Genetic fate mapping of hippocampal inhibitory INTs

Tricoire; in collaboration with Fishell, Cauli

Hippocampal INTs arise from progenitor cells of the caudal and medial ganglionic eminences (CGE and MGE), which are located in the ventral part of the telencephalon. Given the INTs’ diversity, the precise birth date of INT subtypes remains unknown. Indeed, few data exist on the respective postnatal fate of the cohorts of INTs generated during embryogenesis. Olig2 is a basic helix-loop-helix transcription factor highly expressed in the MGE. To perform an inducible genetic fate mapping of INT precursors, we used transgenic mice that expressed a tamoxifen-inducible form of Cre recombinase under the control of the Olig2 promoter. After a single administration of tamoxifen, a Z/EG reporter allele allowed the expression of GFP upon the Cre-mediated removal of a stop casset as well as the irreversible labeling of precursors expressing Olig2. This configuration permitted cell labeling in a temporally and spatially precise manner and allowed for subsequent study of the cells in their mature state after birth and long after the Olig2 promoter switched off.

We examined hippocampal fate-mapped INTs in 3- to 4-week-old mouse pups after administering tamoxifen to pregnant females between E9.5 and E15.5. Early data revealed that hippocampal INTs are generated earlier than their cortical INT counterparts. The vast majority of hippocampal GFP⁺ INTs localized to the CA1–3 subfields. Using immunofluorescence, we observed that the occurrence of INT markers such as PV, SOM, and NPY in GFP⁺ cells was dependent on the time of tamoxifen administration. Few GFP⁺ cells expressed calretinin or VIP, suggesting a different origin for hippocampal INTs expressing these markers. We then used patch clamp recording in combination with single-cell RT-PCR to study the electrophysiological, molecular, and morphological properties of the fate-mapped cells in acute hippocampal slices. Single-cell RT-PCR revealed the presence of GAD65 and GAD67, confirming the GABAergic phenotype of fate-mapped cells as well as the prevalence of PV, SOM, and NPY in these cells. Our observations suggest that select cohorts of hippocampal INTs are generated within the MGE at specific times throughout embryonic development. Importantly, the genetic approach used here permits the indelible marking of INT subtypes and links early developmental events to the physiological profiles of mature cells.

Developmental origin of hippocampal neuronal nitric oxide synthase–expressing INTs

Tricoire; in collaboration with Fishell

The hippocampus widely expresses the neuronal isoform of nitric oxide synthase (nNOS), an enzyme frequently found in NPY-expressing GABAergic INTs. So far, we have identified two distinct cell types in the mammalian cortex that are positive for nNOS/NPY: the Ivy cell and neurogliaform cell. While Ivy cells are located mainly in the stratum oriens and stratum radiatum, neurogliaform cells are typically found in the stratum lacunosum moleculare. Both cell types have a dense axonal arbor and are involved in slow synaptic transmission. The axon of neurogliaform cells overlaps with the glutamatergic input from the entorhinal cortex while the axonal field of Ivy cells is aligned with the CA3 input to stratum oriens and stratum radiatum.

Although Ivy cells have been described as one of the most abundant INT subtypes, it is unclear whether they represent a population of cells distinct from neurogliaform cells. To resolve this issue, we employed a genetic fate-mapping approach, using the Olig2CreER:Z/EG mouse (described above) to examine the origin of nNOS-expressing INTs. We found that a large portion of fate-mapped cells that expressed nNOS were located in both the stratum oriens and stratum pyramidale, but we observed few cells in the stratum lacunosum-moleculare. The occurrence of nNOS in fate-mapped cells shows maxima when tamoxifen is administered at E9.5 and E13.5, with a trough at E11.5 suggesting two successive waves of nNOS⁺ INT precursors. Electrophysiological recordings, combined with morphological reconstruction and single-cell RT-PCR, allowed positive identification of Ivy cells in fate-mapped cells in Olig2CreER:Z/EG, suggesting an MGE origin for Ivy cells. We will extend these initial studies to include transgenic mice that express GFP in INTs exclusively derived from either the MGE or CGE. These experiments will determine where and when Ivy and neurogliaform cells are generated and whether they share a common or divergent developmental and genetic program.

Collaborators

Bruno Cauli, PhD, Université Pierre et Marie Curie, Paris, France
Graham Collingridge, PhD, University of Bristol, Bristol, UK
Gordon Fishell, PhD, New York University, New York, NY
Richard Huganir, PhD, Howard Hughes Medical Institute, The Johns Hopkins University, Baltimore, MD
John Isaac, PhD, Porter Neuroscience Research Center, NINDS, Bethesda, MD
Jean-Claude Lacaille, PhD, Université de Montréal, Montréal, Canada
Ronald Petralia, PhD, Laboratory of Neurochemistry, NIDCD, Bethesda, MD
Katherine Roche, PhD, Porter Neuroscience Research Center, NINDS, Bethesda, MD
Gabor Szabo, PhD, Institute of Experimental Medicine, Budapest, Hungary
Susumu Tonegawa, PhD, Picower Institute, MIT, Cambridge, MA
Lisa Topolnik, PhD, Université de Montréal, Montréal, Canada

For further information, contact mcbainc@mail.nih.gov.