Optogenetic investigation of Chrna2 cells in the subiculum and their
role in modulating entorhinal cortex input
Heather Nichol
Integrated Program in Neuroscience
McGill University, Montreal
July 2015
A thesis submitted to McGill University in partial fulfillment of the requirements of the degree
of Master of Science.
© Heather Nichol 2015
Table of Contents
Abstract ……………………………………………………………………………………….......5
Résumé …………………………………………………………………………………………....7
Acknowledgements ……………………………………………………………………………….9
Contributions ……………………………………………………………………………………10
List of Abbreviations ……………………………………………………………………………11
List of Tables and Figures ……………………………………………………………………….12
1. Introduction …………………………………………………………………………………...14
2. Literature Review …………………………………………………………………………......16
2.1 Theta rhythm in the hippocampus …………………………………………………..16
2.1.1 Defining Theta ………………………………………………………….....16
2.1.2 Functions of Hippocampal Theta Rhythm ………………………………...16
2.1.3 Hippocampal Theta Generation …………………………………………..17
2.2 Entorhinal Cortex (EC) ……………………………………………………………..18
2.2.1 EC Connections with Hippocampus ……………………………………....18
2.2.2 Role of EC in Theta Generation …………………………………………..19
2.2.3 EC Contributions to Hippocampal Functions ………………………….....19
2.3 OLM Interneurons …………………………………………………………………..20
2.3.1 Characteristics of CA1 OLM Interneurons ……………………………….20
2.3.2 Role of OLM Interneurons in Modulating EC Input ………………………22
2.3.3 Role of OLM Interneurons in Modulating Theta ………………………….23
2
2.4 Subiculum ……………………………………………………………………………23
2.4.1 Subiculum Cytoarchitecture and Neuron Populations ……………………24
2.4.2 Role of Subiculum in Hippocampal Function ……………………………..25
3. Objectives …………………………………………………………………………………….26
4. Methods ……………………………………………………………………………………....27
4.1 Animals ……………………………………………………………………………...27
4.2 Quantification of Somatostatin Expression …………………………………………27
4.3 Electrophysiological Properties of Chrna2 Cells …………………………………..28
4.4 Chrna2 Cell Reconstruction ………………………………………………………...30
4.5 Optogenetic Activation of Chrna2 Cells …………………………………………….30
4.6 Dual Optogenetics …………………......……………………………………………33
4.7 Statistical Analysis …………………………………………………………………..37
5. Results …………………………………………………………………………………….......37
5.1 Are subiculum Chrna2 cells a distinct population? ………………………………....37
5.1.1 Do subiculum Chrna2 cells express somatostatin? ……………………….38
5.1.2 What are their electrophysiological properties? ………………………….38
5.1.3 What is their dendritic and axonal morphology? …………………………39
5.1.4 What post-synaptic responses do Chrna2 cells elicit in pyramidal cells?...40
5.2 What is the role of Chrna2 cells in the modulation of EC input? …………………...44
5.2.1 How do Chrna2 cells modulate post-synaptic responses to EC input to the
pyramidal cells? …………………………………………………………………44
3
6. Discussion …………………………………………………………………………………….48
6.1 Characterization of Subiculum Chrna2 Cells ……………………………………….48
6.2 Optogenetic Investigation of Chrna2 Cell-Mediated Post-Synaptic Responses …….53
6.3 Optogenetic Investigation of EC Input-Mediated Post-Synaptic Responses ………..58
6.4 Dual Optogenetic Control …………………………………………………………...60
6.5 Implications for the Role of Chrna2 Cells in Subicular Function …………………..61
6.6 Limitations …………………………………………………………………….........63
7. Final Summary ………………………………………………………………………………..64
Figures …………………………………………………………………………………………...66
References …………………………………………………………………………………….....82
4
Abstract
The hippocampus is an instrumental brain region for a number of important functions,
including spatial information processing, learning, and memory. To fully understand the
mechanisms underlying these functions, it is critical to understand the interaction between
hippocampal inputs and local interneurons. One input of particular interest is direct input from
the entorhinal cortex (EC), which contributes to the generation of hippocampal theta rhythm and
which transfers vital sensory information to the hippocampus. An interneuron population likely
to play a role in modulating this input is the oriens lacunosum-moleculare (OLM) interneuron. A
number of characteristics, including the position of their axon terminals and association with
theta rhythm, make OLM interneurons prime candidates to regulate EC input. The recent
discovery of a genetic marker specific to OLM interneurons, the nicotinic acetylcholine receptor
alpha2 subunit (Chrna2), enables the investigation of such a role. In the hippocampus, Chrna2 is
exclusively expressed in OLM interneurons in CA1, as well as in interneurons in subiculum
which have yet to be described. The subiculum is an ideal sub-region in which to focus an
investigation into Chrna2 cells as it is believed to make an important contribution to
hippocampal function, in both theta rhythm generation and spatial learning and memory.
This study aimed to characterize Chrna2 cells in the subiculum, and to investigate their
role in the regulation of EC input. Characterization of subiculum and CA1 Chrna2 cells found
similarities in somatostatin (Som) expression, electrophysiological properties, and axon
directionality and differences in soma and dendritic morphology and in the proportion of the
Som-expressing cells represented by Chrna2 cells in each region. An investigation of the postsynaptic responses elicited by Chrna2 cells revealed inhibitory responses, composed of both
GABAA receptor (GABAAR) and GABAB receptor (GABABR) mediated components, which
5
were not significantly different between CA1, subiculum regular firing and subiculum bursting
firing pyramidal cells. Overall, these findings suggest that subiculum and CA1 Chrna2 cells are
largely similar. Characterization of EC input to subiculum pyramidal cells found the postsynaptic responses to be similar in both regular and burst firing pyramidal cells. However, a
direct demonstration of how Chrna2 cell-mediated inhibition modulates this EC input was not
possible due to the inability to achieve dual optogenetic control of Chrna2 cells and EC input.
Overall, this study has provided the first in-depth characterization of Chrna2 cells in the
subiculum, finding that subiculum and CA1 Chrna2 cells are generally equivalent and are likely
to play similar roles both sub-regions, roles which may include the regulation of theta rhythm
and a contribution to learning and memory functions.
6
Résumé
L'hippocampe est une région du cerveau critique pour plusieurs fonctions importantes,
telles que le traitement de l'information spatiale, l'apprentissage et la mémoire. Pour bien
comprendre les mécanismes sous-jacents de ces fonctions, il est essentiel de comprendre
l'interaction entre les entrées hippocampiques et les interneurones locaux. L’entrée du cortex
entorhinale (EC) est d’un intérêt particulier, car elle contribue à la génération du rythme thêta et
transfert des informations sensorielles cruciales pour l'hippocampe. Les interneurones oriens
moleculare-lacunosum (OLM) constituent l’une des populations susceptibles de jouer un rôle
dans la modulation de cette entrée corticale. Un certain nombre de caractéristiques - en
particulier la position de leurs terminaisons axonales et leur association au rythme thêta - suggère
que les interneurones OLM sont des candidats excellents pour réguler l’entrée du EC. La
découverte récente d’un marqueur génétique spécifique aux interneurones OLM, la sous-unité
alpha2 du récepteur nicotinique à l’acétylcholine (Chrna2), permet l'étude d’un tel rôle. Dans
l'hippocampe, Chrna2 est exprimée exclusivement dans les interneurones OLM de CA1, ainsi
que dans les interneurones du subiculum qui n’ont pas encore été décrits. Le subiculum est une
sous-région idéale pour étudier les cellules Chrna2, car il contribue significativement à la
fonction de l’hippocampe, à la fois dans la génération du rythme thêta, mais également dans
l’apprentissage et la mémoire spatiale.
La présente étude visait à caractériser les cellules Chrna2 dans le subiculum, et
comprendre leur rôle dans la régulation de l’entrée du EC. La caractérisation des cellules Chrna2
dans le subiculum et CA1 a montré des similitudes dans l’expression de la somatostatine (Som),
dans les propriétés électrophysiologiques et dans la directionnalité des axones. Cette
caractérisation a aussi montré des différences dans la morphologie des dendrites et du soma et
7
dans la proportion de cellules Chrna2 dans la population de cellules exprimant la Som pour
chaque région. Une étude des réponses post-synaptiques induites par les cellules Chrna2 a révélé
que les réponses inhibitrices, composées de réponses GABAA et GABAB, n'étaient pas
significativement différentes entre les cellules pyramidales de CA1, ainsi que les cellules
toniques et phasiques du subiculum. Dans l'ensemble, ces résultats suggèrent que les cellules
Chrna2 du subiculum et de CA1 sont en grande partie similaires. La caractérisation de l’entrée
du EC aux cellules pyramidales du subiculum a révélé des réponses post-synaptiques semblables
dans les cellules toniques et phasiques. Cependant, en raison de l'incapacité d'atteindre la
contrôle optogenetic des cellules Chrna2 et l’entrée du EC en même temps, une démonstration
directe de la manière dont l’inhibition modulée par les cellules Chrna2 influence l’entrée du EC
n'était pas possible.
Dans l'ensemble, cette étude a fourni la première caractérisation en profondeur des
cellules Chrna2 dans le subiculum, en établissant que les cellules Chrna2 du subiculum et de
CA1 sont généralement semblables et sont susceptibles de jouer des rôles similaires dans les
deux sous-régions - en particulier de réguler le rythme thêta et contribuer à l'apprentissage et à la
mémoire.
8
Acknowledgements
First, I would like to thank my supervisor, Dr. Sylvain Williams, for providing me with
the opportunity to pursue this research project in his lab. I benefitted greatly from his guidance
and support throughout the project’s progression and his confident composure through its ups
and downs. It has been an invaluable experience. I would also like to thank the members of my
advisory committee, Dr. Lalit Srivastava and Dr. Philippe Séguéla, and my mentor, Dr. Naguib
Mechawar, for their feedback and support.
I have had the privilege of working with a dedicated and knowledgeable group of people
in the Williams lab and this project would not have been possible without their training, advice
and support. These include (in rough order of appearance): Jennifer Robinson, Dr. Frédéric
Manseau, Dr. Bénédicte Amilhon, Dr. Guillaume Ducharme, Dr. Ning Gu, Dr. Siddhartha
Mondragon-Rodriguez, Richard Boyce, Dr. Chris Kortleven, Dr. Caroline Fasano, Leah LaScala,
Eva Vico, Dr. Jean-Bastien Bott, Dr. Guillaume Etter, Camille Gola, Dr. Amy Chee, Dr. Jun
Kang, and Polina Reynolds. I would like to give special thanks to a few lab members in
particular: Frédéric Manseau, for teaching me to patch, and the numerous associated techniques,
and for setting up my rig, among many other things; Bénédicte Amilhon, for her greatly
beneficial instruction and advice on injections, patching, immunos, analysis and more, and for
reviewing my thesis; Jennifer Robinson, for her immensely appreciated training and advice on
slicing, injections, patching, colony management and just about anything else I needed help with,
and for many encouraging conversations; and Guillaume Ducharme, for helping me build and
use the equipment needed for dual optogenetics.
9
I would like to thank several labs at the Douglas Mental Health University Institute for
the use of their equipment: the lab of Dr. Giamal Luheshi for the use of their cryostat, the lab of
Dr. Bruno Giros for the use of their microscope, and the lab of Dr. Naguib Mechawar for the use
of their microscope and reconstruction software, as well as helpful instruction. A thank you also
goes to Dr. Joseph Rochford for his statistical advice.
An important thank you also goes to a significant mentor in my scientific life, my
undergraduate thesis supervisor, Dr. Ken Rose. His philosophy on science and teaching are
forever steeped in my own scientific “ways” and his ongoing guidance is deeply appreciated.
Finally, my acknowledgements would not be complete without recognition of my
amazing support system of family and friends outside of academia. I would especially like to
thank my fiancé, who knows just how to make me laugh after a long day at the lab, and my
mom, who was, and has always been, an indispensable source of encouragement, advice and
support.
Contributions
Dr. Sylvain Williams supervised this project, providing guidance in the design and
analysis of experiments and presentation of the results. Heather Nichol performed all
experiments described in this study. This thesis was written by Heather Nichol and edited by
Bénédicte Amilhon and Sylvain Williams. The translation of the abstract was edited by
Guillaume Etter.
10
List of Abbreviations
AHP – afterhyperpolarization
Ra – access resistance
cc – current clamp
Rm – membrane resistance
Chrna2 – nicotinic acetylcholine receptor
SEM – standard error of the mean
alpha2 subunit
SLM – stratum lacunosum-moleculare
EC – entorhinal cortex
Som – somatostatin
EPSC – excitatory post-synaptic current
Tom – tdTomato
EPSP – excitatory post-synaptic potential
vc – voltage clamp
GABAAR – GABAA receptor
Vr – resting membrane potential
GABABR – GABAB receptor
h.p. – holding potential
IPSC – inhibitory post-synaptic current
IPSP – inhibitory post-synaptic potential
ISI – interspike interval
OLM – oriens lacunosum-moleculare
PBS – phosphate buffered saline
PFA – paraformaldehyde
PSP – post-synaptic potential
11
List of Tables and Figures
Table 1. Electrophysiological properties of Chrna2 cells recorded in subiculum and CA1. ........67
Figure 1. Characterization somatostatin (Som) expression and electrophysiological properties of
Chrna2 cells. .................................................................................................................................66
Figure 2. Chrna2 cell morphology. ...............................................................................................68
Figure 3. Anatomical and electrophysiological characterization of ChETA-YFP expression in
Chrna2 cells. .................................................................................................................................70
Figure 4. Example post-synaptic responses elicited in pyramidal cells by optogenetic activation
of Chrna2 cells. .............................................................................................................................72
Figure 5. Quantification of post-synaptic responses elicited in pyramidal cell by optogenetic
activation of Chrna2 cells. ............................................................................................................73
Figure 6. Paired pulse recordings of post-synaptic responses elicited by optogenetic activation of
Chrna2 cells. .................................................................................................................................74
Figure 7. Changes in post-synaptic responses to optogenetic activation of Chrna2 cell as a
function of holding potential..........................................................................................................75
Figure 8. Quantification of post-synaptic responses elicited in pyramidal cell by optogenetic
activation of Chrna2 cells during application of GABAAR and GABABR antagonists. ..............76
Figure 9. Patch clamp recordings in a C1V1-expressing Chrna2 cell. .........................................77
12
Figure 10. Anatomical and electrophysiological characterization of Chrimson-Tom expression in
EC cells. ........................................................................................................................................78
Figure 11. Post-synaptic responses elicited in subiculum pyramidal cells by optogenetic
activation of EC input. ..................................................................................................................79
Figure 12. Responses of ChETA-expressing Chrna2 cells and Chrimson-expressing EC cells to
blue light stimulation. ...................................................................................................................80
Figure 13. Post-synaptic responses in pyramidal cells in ChETA- or Chrimson-expressing
animals in response to blue or yellow light stimulation. ..............................................................81
13
1. Introduction
The hippocampus is a multi-modal structure long known to serve critical functions in
spatial information processing, learning and memory. The complex organization in the
hippocampus, from its distinct sub-regions, to its numerous inputs, to the large number of
different neuron populations, allows it to perform these functions, and renders achieving a
complete understanding of the mechanisms underlying these functions an intricate task.
Interneurons are believed to have a central role in hippocampal function, being at the interface
between the hippocampus network which they can powerfully modulate, and inputs to this
network. An important area of investigation toward the understanding of hippocampus function
is, thus, the interaction between hippocampal inputs and local interneurons.
Hippocampal network-input interactions can be studied in the framework of rhythmic
oscillations, activity states which are believed to play an important role in hippocampal function.
A frequency band of particular interest is theta rhythm: oscillations of 4-12 Hz postulated to
contribute to information processing and transfer, and learning and memory (Berry and
Thompson, 1978, O'Keefe and Recce, 1993, Buzsáki, 2002, Colgin et al., 2009, Colgin, 2013).
Inputs from the medial septum to the hippocampus have generally been regarded as the main
theta generator (Green and Arduini, 1954, Buzsáki et al., 1983, Buzsáki, 2002, Colgin, 2013),
but other inputs also play important roles. The entorhinal cortex (EC) provides excitatory inputs
which contribute to theta generation (Buzsáki et al., 1983, Kamondi et al., 1998, Buzsáki, 2002),
as well as provide critical sensory information to the hippocampus (Fyhn et al., 2004, Hafting et
al., 2005, Moser et al., 2008). However, the intra-hippocampal mechanisms which modulate EC
input and regulate theta oscillations have not yet been fully elucidated.
14
A particular population of hippocampal interneurons are well-positioned to play a role in
the modulation of EC input: oriens lacunosum-moleculare (OLM) interneurons. These Somexpressing interneurons target the same dendritic location as EC inputs in CA1 (Cajal, 1911,
Lorente de Nò, 1934, McBain et al., 1994, Sik et al., 1995, Katona et al., 1999, Maccaferri et al.,
2000) and have been associated with theta rhythm (Maccaferri and McBain, 1996a, Pike et al.,
2000, Gillies et al., 2002, Klausberger et al., 2003, Gloveli et al., 2005, Varga et al., 2012,
Katona et al., 2014). Recently, Leão et al. (2012) discovered a specific marker for OLM
interneurons, the nicotinic acetylcholine receptor alpha2 subunit (Chrna2). In the hippocampus,
Chrna2 is exclusively expressed in OLM interneurons in CA1, as well as in interneurons in
subiculum which have yet to be described. The goal of this study is to characterize Chrna2 cells
in the subiculum and investigate their role in modulating EC input to this sub-region; a largely
under-investigated hippocampal sub-region which is poised to make a significant contribution to
hippocampal function. The subiculum is the major output structure of the hippocampus (O'Mara
et al., 2001, Anderson et al., 2007), and recent work suggest it acts as an intrinsic theta generator
capable of setting the timing of endogenous hippocampal theta oscillations (Jackson et al., 2014).
The characterization and assessment of the role of Chrna2 cells in modulating EC input to
the subiculum will help complete the understanding of the mechanisms underlying hippocampal
function, including theta rhythm. In turn, this may improve the understanding of the pathology
underlying disorders of learning and memory, especially given that lesions to or dysfunction of
the subiculum and EC input to hippocampus have been linked to deficits in these important
functions (Galani et al., 1997, Laxmi et al., 1999, Oswald and Good, 2000, Brun et al., 2002,
Remondes and Schuman, 2004, Steffenach et al., 2005). This work may also be of particular
importance to Alzheimer’s disease as the subiculum is one of the first hippocampal regions to
15
display degeneration and dysfunction in this disorder, including the loss of Som-expressing
interneurons (Adachi et al., 2003, de la Prida et al., 2006, Scher et al., 2011, Goutagny et al.,
2013, Trujillo-Estrada et al., 2014).
2. Literature Review
2.1 Theta rhythm in the hippocampus
2.1.1 Defining Theta
Local field potential recordings in the hippocampus reveal rhythmic oscillations in
activity which are believed to play an important role in hippocampal function. A frequency band
of particular interest is theta rhythm, oscillations ranging from 4-12 Hz [values in rodents;
(O'Keefe, 1993, Colgin, 2013)]. Theta rhythm was first described in the hippocampus by Green
and Arduini (1954), who found hippocampal theta to be associated with arousal. They observed
theta during periods of alertness and when the recorded animal appeared interested in its
surroundings (Green and Arduini, 1954). Hippocampal theta rhythm has now been recorded both
in animals, including rodents, rabbits, and monkeys (Green and Arduini, 1954, Vanderwolf,
1969, Stewart and Fox, 1991, Colgin, 2013, Jutras et al., 2013), and in humans (Tesche, 1997,
Tesche and Karhu, 2000, Lega et al., 2012). Theta rhythm is known be prominent during
behavioural states including REM sleep, running and exploration and is postulated to serve a
number of functions (Buzsáki, 2002, Colgin, 2013, Hasselmo and Stern, 2013).
2.1.2 Functions of Hippocampal Theta Rhythm
Theta rhythm is believed to play several important roles in hippocampal function. First,
theta rhythm may provide a mechanism for the packaging and coding of information received by
the hippocampus. For example, theta cycles are believed to contain compressed representations
16
of spatial location based on the timing of pyramidal cell firing relative to the theta cycle (O'Keefe
and Recce, 1993, Skaggs et al., 1996). They may also serve as a mechanism to package related
information into discrete units (Jezek et al., 2011, Gupta et al., 2012). Second, theta rhythm is
postulated to contribute to learning and memory. Lesions which disrupt hippocampal theta
rhythm have been found to disrupt memory function (Berry and Thompson, 1978, Winson,
1978). The presence of hippocampal theta rhythm has been associated with faster learning rates
in rabbits (Seager et al., 2002, Griffin et al., 2004) and theta power has been correlated with
performance on memory tasks in rodents, rabbits, monkeys and humans (Berry and Thompson,
1978, Givens and Olton, 1990, Lega et al., 2012, Jutras et al., 2013). Modelling studies have also
suggested that the processes of memory retrieval and encoding could be restricted to different
phases of theta (Hasselmo et al., 2002, Hasselmo, 2005). Third, theta rhythm is believed
contribute to the synchronization of the hippocampus with other brain areas, such as the EC,
coordination which may improve information transfer (Colgin et al., 2009, Colgin, 2013).
2.1.3 Hippocampal Theta Generation
A number of mechanisms are believed to underlie the generation of theta rhythm in the
hippocampus. Inputs from the medial septum have long been seen as the major theta generator
(Green and Arduini, 1954, Buzsáki et al., 1983, Buzsáki, 2002, Colgin, 2013). Manipulations
which disrupt these inputs have been found to reduce or abolish theta rhythm in the hippocampus
(Petsche et al., 1962, Givens and Olton, 1990, Buzsáki, 2002). However, other inputs also play
significant roles. One of these inputs is direct EC input to the distal dendrites of CA1 pyramidal
cells (Buzsáki et al., 1983, Kamondi et al., 1998, Buzsáki, 2002) (see below for further details:
2.2 Entorhinal Cortex). The hippocampus is also capable of generating theta through intrinsic
mechanisms, as has been demonstrated with the intact hippocampal preparation (Goutagny et al.,
17
2009) (see below for further details: 2.4 Subiculum). Hippocampal interneurons are believed to
play a crucial role in the regulation of rhythmic oscillations (Freund and Buzsáki, 1996, McBain
and Fisahn, 2001, Buzsáki, 2002). For example, in a recent study, Amilhon et al. (2015)
demonstrated that optogenetically driving PV-expressing interneurons in the intact hippocampal
preparation could regulate the power and frequency of intrinsic hippocampus oscillators. The
interneurons which are the focus of this study, OLM interneurons, are also believed to play an
important role in theta regulation (see below for further details: 2.3 OLM Interneurons).
2.2 Entorhinal Cortex (EC)
The EC is a critical brain region in the larger hippocampal formation, generally defined
as including the dentate gyrus, CA3, CA1, subiculum, parasubiculum, presubiculum and EC
(Anderson et al., 2007). It is a hub for the transfer of sensory information to and from the
hippocampus (Anderson et al., 2007), providing critical spatial information (Moser et al., 2008)
and participating in the generation of hippocampal theta (Buzsáki, 2002).
2.2.1 EC Connections with Hippocampus
The EC can be divided into 6 layers (I-VI) with different patterns of connectivity. Layers
II and III are the main source of EC projections to the hippocampus with ECII projecting to
dendate gyrus and CA3, and ECIII projecting to CA1 and subiculum (van Groen et al., 2003,
Chevaleyre and Siegelbaum, 2010, Rowland et al., 2013). The deep layers of the EC (V-VI)
receive projections from CA1 and subiculum (Köhler, 1985, Naber et al., 2001, van Groen et al.,
2003). The reciprocal connections between ECIII and CA1/subiculum are aligned
topographically. Lateral ECIII projects to distal CA1 and proximal subiculum whereas medial
ECIII projects to proximal CA1 and distal subiculum (Witter et al., 2000, Naber et al., 2001, van
18
Groen et al., 2003). Accordingly, lateral ECV-VI receive projections from distal CA1 and
proximal subiculum whereas medial ECV-VI receive projections from proximal CA1 and distal
subiculum (Witter et al., 2000, Naber et al., 2001, van Groen et al., 2003). The projections from
ECIII to CA1 and subiculum are also localized to the superficial strata of each hippocampal subregion. ECIII projections terminate in the stratum lacunosum moleculare (SLM) of CA1 and the
superficial molecular layer of the subiculum, thus targeting the distal dendrites of CA1 and
subiculum pyramidal cells (Naber et al., 2001, van Groen et al., 2003, Anderson et al., 2007).
2.2.2 Role of EC in Theta Generation
The direct EC input to the distal dendrites of hippocampal pyramidal cells is believed to
play an important role in theta generation. It evokes dendritic currents which are critical for the
generation of extracellular oscillations (Kamondi et al., 1998, Buzsáki, 2002, Mizuseki et al.,
2009). The theta dipole created by this input in the SLM in CA1 is abolished by lesions to the EC
(Kamondi et al., 1998), lesions which are also associated with memory deficits (Remondes and
Schuman, 2004). Such lesions to or deafferentation of the EC change the nature of hippocampal
theta to one that is atropine-sensitive and that displays depth versus voltage profiles which are
significantly different from those observed in the awake animals but similar to those observed
under urethane anesthesia (Buzsáki et al., 1983, Ylinen et al., 1995). Furthermore, work in the
intact hippocampal preparation has shown that optically driving EC input to the hippocampus
can entrain hippocampal oscillators (Amilhon et al., 2015).
2.2.3 EC Contributions to Hippocampal Functions
In addition to playing a role in the generation of theta rhythm in the hippocampus, EC
input contributes to hippocampal functions including spatial information processing, learning
19
and memory. The EC provides critical sensory information to the hippocampus. Of particular
importance is spatial information, provided in large part by EC grid cells (Fyhn et al., 2004,
Hafting et al., 2005, Moser et al., 2008). Spatial information provided by EC grid cells is
believed to play an important role in the generation of the place fields associated with
hippocampal place cells (Solstad et al., 2006, Moser et al., 2008). In fact, accurate spatial firing
patterns can develop in CA1 and spatial recognition is maintained after isolation of the direct
EC-CA1 pathway (Brun et al., 2002). These inputs are also involved in spatial recall. EC lesions
or isolation have been found to disrupt the consolidation and retention of spatial memories (Brun
et al., 2002, Remondes and Schuman, 2004, Steffenach et al., 2005). The timing of EC input to
the hippocampus is also believed to be important for learning and memory. Modelling studies
have suggested that EC input may be active at the peak of the theta cycle to promote memory
encoding, and inhibited at the trough of the theta cycle to promote memory retrieval (Hasselmo
et al., 2002). The interneurons of interest in this study, OLM interneurons, are prime candidates
to regulate and time this important EC input, contributing to its effects on hippocampal function.
2.3 OLM Interneurons
The hippocampus boasts a large diversity of interneuron populations (Freund and
Buzsáki, 1996, McBain and Fisahn, 2001). Among them, the OLM interneuron appears ideally
suited for a role in the modulation of direct EC input and theta rhythm.
2.3.1 Characteristics of CA1 OLM Interneurons
OLM interneurons have been largely characterized in CA1. The name “oriens
lacunosum-moleculare” interneuron is derived from their morphology, first described by Cajal
(1911) and Lorente de Nò (1934). In CA1, the soma and dendritic trees of OLM interneurons are
20
located in the stratum oriens and their axons extend directly out to arbourize in the stratum
lacunosum-moleculare, with sparse collaterals in stratum oriens (McBain et al., 1994, Sik et al.,
1995, Maccaferri et al., 2000, Losonczy et al., 2002, Leão et al., 2012). This morphology aligns
their axon terminals with the direct EC input to the distal dendrites of CA1 pyramidal cells.
OLM interneurons are estimated to compose 4.3% of the total interneuron population in
CA1 (Bezaire and Soltesz, 2013). The majority of excitatory inputs to OLM interneurons are
local pyramidal axon collaterals (Lacaille et al., 1987, Blasco-Ibáñez and Freund, 1995,
Maccaferri and McBain, 1995, Ali and Thomson, 1998, Sun et al., 2014). Correspondingly, the
majority of OLM interneuron axon terminals target local pyramidal cell dendrites, with a small
percentage targeting other interneurons (Katona et al., 1999, Bezaire and Soltesz, 2013). Thus,
OLM interneurons are well positioned to contribute to feedback inhibition. Moreover, it has been
shown that, in CA1, they participated exclusively in feedback, and not feedforward inhibition
(Maccaferri and McBain, 1995, 1996b).
OLM interneurons are characterized, electrophysiologically, by spontaneous, slow
frequency firing at rest (McBain et al., 1994, Maccaferri et al., 2000, Leão et al., 2012), as well
as regular firing, with accommodation in frequency, in response to depolarizing current pulses
(Sik et al., 1995, Maccaferri and McBain, 1996a, Gloveli et al., 2005, Chittajallu et al., 2013).
OLM interneurons also display an Ih current, a hyperpolarization-activated cation current, which
produces a sag in their responses to hyperpolarizing current pulses (Sik et al., 1995, Maccaferri
and McBain, 1996a, Gloveli et al., 2005, Leão et al., 2012, Chittajallu et al., 2013). This
conductance is believed to play a role in the theta frequency firing observed in these interneurons
(Maccaferri and McBain, 1996a, Gillies et al., 2002, but see: Kispersky et al., 2012).
21
OLM interneurons are GABAergic and are known to co-express somatostatin (Som)
(Morrison et al., 1982, Somogyi et al., 1984, Naus et al., 1988, Katona et al., 1999, Losonczy et
al., 2002, Klausberger et al., 2003, Leão et al., 2012). Som, as well as metabotropic glutamate
receptor 1alpha (Baude et al., 1993), have been used as markers for OLM interneurons, but are
not specific to this interneuron population alone. It was recently discovered that OLM
interneurons in CA1 exclusively express the nicotinic acetylcholine receptor alpha 2 subunit
(Chrna2) (Ishii et al., 2005, Nakauchi et al., 2007, Leão et al., 2012), allowing Leão et al. (2012)
to develop the first transgenic mouse line to specifically target OLM interneurons. Chrna2 is also
expressed in interneurons in the subiculum, but the properties of these subiculum Chrna2 cells
have yet to be investigated and compared to those of their counterparts in CA1.
2.3.2 Role of OLM Interneurons in Modulating EC Input
The axonal morphology of OLM interneurons makes them prime candidates for a role in
modulating EC input. Field recordings have demonstrated that the activation of OLM
interneurons decreases the excitatory effects of EC input (Maccaferri and McBain, 1995). The
optogenetic activation of CA1 Chrna2 cells has been shown to decrease the voltage spread
associated with electrical stimulation of EC input (Leão et al., 2012). Som-expressing
interneurons, postulated to be specifically OLM interneurons, have been found to inhibit direct
EC input to CA1 pyramidal cells, inhibition which was necessary to the formation of contextual
fear memories (Lovett-Barron et al., 2014). Furthermore, optogenetic silencing of Somexpressing interneurons has been found to disrupt the effects of direct EC input on theta
oscillations in CA1/subiculum in the intact hippocampal preparation (Amilhon et al., 2015).
However, there have been no investigations of the modulatory effects of Chrna2 cells on EC
input to the subiculum.
22
2.3.3 Role of OLM Interneurons in Modulating Theta
Hippocampal interneurons have long been thought to play a role in the regulation of theta
oscillations (Cobb et al., 1995, Freund and Buzsáki, 1996, McBain and Fisahn, 2001, Buzsáki,
2002), and OLM interneurons have a number of properties which are suggestive of such a role.
OLM interneurons display slow membrane conductance resonating at theta frequency and an
intrinsic capacity to fire at theta frequency (Maccaferri and McBain, 1996b, Pike et al., 2000,
Gillies et al., 2002, Gloveli et al., 2005, but see: Kispersky et al., 2012). The firing of OLM
interneurons has been found to increase during theta-associated behaviour and to be phaselocked to the theta cycle, with highest firing during the trough in stratum pyramidale
(Klausberger et al., 2003, Varga et al., 2012, Katona et al., 2014). Accordingly, they may phasemodulate the direct excitatory input from the EC to the distal dendrites of pyramidal cells
(Klausberger et al., 2003). This timing corresponds to the model of memory encoding and
retrieval segregation according to theta phase, inhibiting EC input at the theta trough (Hasselmo
et al., 2002). While no studies have investigated the role of OLM interneurons, in particular, in
modulating theta rhythm, some studies have assessed the role of Som-expressing interneurons.
While activating or silencing Som-expressing interneurons has been found to have only a minor
or no influence on theta oscillations (Royer et al., 2012, Amilhon et al., 2015), silencing Somexpressing interneurons during the activation of EC projections has been found to significantly
attenuate the effect of EC input on theta rhythm in CA1/subiculum (Amilhon et al., 2015).
2.4 Subiculum
The subiculum can be considered the final step of information flow through the
hippocampus. Despite its prominent position in the hippocampal circuit, the subiculum has been
significantly under-investigated in comparison to other hippocampal sub-regions.
23
2.4.1 Subiculum Cytoarchitecture and Neuron Populations
The subiculum possesses a unique cytoarchitecture of three layers (Lorente de Nò, 1934,
O'Mara et al., 2001, de la Prida et al., 2006, Anderson et al., 2007). The polymorphic layer is the
deepest layer and is continuous with stratum oriens of CA1. Next, the pyramidal cell layer is
expanded and less densely packed than in neighbouring CA1. Most superficial is the molecular
layer which is continuous with both the stratum radiatum and SLM of CA1.
The pyramidal cells in the subiculum can be divided into two groups based on their firing
properties: regular and burst firing (Stewart and Wong, 1993, Taube, 1993). The distribution of
regular and burst firing pyramidal cells varies along the proximal-distal and superficial-deep
axes. The proportion of burst firing cells has been found to increase distally on the proximaldistal axis (Staff et al., 2000, Jarsky et al., 2008, Kim and Spruston, 2012) and deeper on the
superficial-deep axis (Greene and Mason, 1996, Harris et al., 2001, but see: Kim and Spruston,
2012). Different proportions of regular and bursting firing cells have also been found to project
to different brain regions, owing to an organization of subiculum projections along the proximaldistal axis (Kim and Spruston, 2012). For example, the lateral EC receives a greater proportion
of its projections from regular firing pyramidal cells, whereas the medial EC receives the
opposite as these two regions receive projections from the proximal and distal subiculum,
respectively. However, no projection target has been found to receive input from only one cell
type or the other (Kim and Spruston, 2012). Differences have also been found in the morphology
of their local axonal collaterals, with regular firing cells having a more laminar organization and
burst firing cells have a more columnar organization (Harris et al., 2001).
24
With regard to interneurons, very little is known. To the best of this author’s knowledge,
no studies have systematically characterized a specific interneuron population in the subiculum.
Such studies warrant investigation, however, because the subiculum is poised to play a pivotal
role in hippocampal function.
2.4.2 Role of Subiculum in Hippocampal Function
The conventional hippocampal circuit flows from dentate gyrus → CA3 → CA1 →
subiculum, with the subiculum acting as a major hippocampal output structure (O’Mara, 2001;
Anderson et al., 2007). The subiculum’s targets include the EC, as described above,
presubiculum, parasubiculum, neocortex, amygdala, hypothalamus, thalamus, septum and
nucleus accumbens (Swanson and Cowan, 1977, Köhler, 1985, 1990, Witter et al., 1990,
Canteras and Swanson, 1992, O'Mara et al., 2001, Anderson et al., 2007). These numerous
projections demonstrate the pivotal role played by the subiculum in transmitting information
from the hippocampus to the rest of the brain. Recent work using the intact hippocampal
preparation has shown that the subiculum also provides a novel backwards signalling pathway
through the hippocampus (subiculum→CA1→CA3) through inhibitory connections (Jackson et
al., 2014), thus also contributing to information processing in other hippocampal sub-regions.
These projections may also contribute to the modulation of theta rhythm in the hippocampus.
Jackson et al. (2014) demonstrated that the subiculum can act as an intrinsic theta generator in
the intact hippocampal preparation and can set the timing of endogenous theta oscillations.
The subiculum is also believed to have an important function in spatial learning and
memory (O'Mara et al., 2009). Like in CA1, pyramidal cells in subiculum show a locational
signal (Sharp and Green, 1994). However, unlike CA1 place cells, subiculum cells tend to fire
25
throughout the environment and show a gradation in firing rate with localized regions of high
firing (Sharp and Green, 1994). The loss of this spatially modulated activity may relate to the
deficits in spatial learning and memory observed after subicular lesions (Galani et al., 1997,
Laxmi et al., 1999, Oswald and Good, 2000). Furthermore, studies of Alzheimer’s disease, a
form of dementia which is associated with spatial memory deficits, have linked this disease to
dysfunction in the subiculum. Studies in human patients and animal models of Alzheimer’s
disease have found that the subiculum is among the earliest brain regions to show degeneration,
accumulation of neurofibrillary tangles and neuritic plaques, and impairments in activity, such as
in theta rhythm (Adachi et al., 2003, de la Prida et al., 2006, Scher et al., 2011, Goutagny et al.,
2013, Trujillo-Estrada et al., 2014).
Overall, the subiculum appears to play a significant role in hippocampal function, a role
that has only begun to be understood. It is an ideal sub-region in which to perform the
investigations of this study.
3. Objectives
The main goal of this project is to characterize Chrna2 cells in the subiculum, and to
investigate their role in the regulation of EC input. The specific aims of the project can be
divided as follows:
1. Are subiculum Chrna2 cells a distinct population?
1.1. Do subiculum Chrna2 cells express somatostatin?
1.2. What are their electrophysiological properties?
1.3. What is their dendritic and axonal morphology?
1.4. What post-synaptic responses do Chrna2 cells elicit in pyramidal cells?
26
2. What is the role of Chrna2 cells in the modulation of EC input to the subiculum?
2.1. How do Chrna2 cells modulate post-synaptic responses to EC input in pyramidal
cells?
4. Methods
4.1 Animals
All procedures follow protocols and guidelines approved by the McGill University
Animal Care Committee and the Canadian Council on Animal Care. Animals were housed in a
temperature-controlled room with a 12/12 hour light/dark cycle and food and water ad libitum.
This project made use of Chrna2-cre transgenic C57BL/6 mice (Chrna2 mice, Lab of
Richardson Leão, Uppsala, Sweden). In this recently developed mouse line, cre recombinase is
expressed under the Chrna2 promoter (Leão et al., 2012). As described above, Chrna2 has been
demonstrated to be expressed exclusively in OLM interneurons in CA1 (Leão et al., 2012); thus,
this mouse line allows for the specific identification and targeting of this distinct interneuron
population. Chrna2 mice were also crossed with cre-reporter R26-tdTomato homozygote mice
(Jackson Laboratory, stock number 007905, Bar Harbor, ME) to generate Chrna2-Tom mice in
which tdTomato (Tom) is exclusively expressed in cells expressing cre recombinase. Tom
fluorescence was used to identify Chrna2 cells.
4.2 Quantification of Somatostatin Expression
Immunohistochemistry experiments were used to determine whether Chrna2 cells in the
subiculum express Som. Chrna2-Tom animals were intracardially perfused with 4%
paraformaldehyde (PFA) in phosphate buffered saline (PBS). Brains were dissected and
transferred to the same fixative solution for 24 hours of post-fixation. The brains were
27
transferred to 15% sucrose in PBS for cryoprotection for 24 hours, and then frozen in isopentane
at -30ᴼC. Brains were sectioned into 25 µm thick coronal slices using a cryostat (Leica
CM3050S, Germany). Slices (1 slice per 100 µm for the full rostrocaudal extent of the
hippocampus) were incubated with a rabbit anti-Som primary antibody (1:250, Santa Cruz
Biotechnology, Santa Cruz, CA) for 16 hours at 4⁰C, followed by an anti-rabbit secondary
antibody coupled to Alexa488 (1:1000, Life Technologies, Eugene, OR) for 2 hours at room
temperature to visualize Som+ cells. Slices were mounted and one image was acquired at 10x
magnification in both CA1 and subiculum for each slice using an Axio Observer microscope
(Carl Zeiss, Germany). Som+ and Tom+ cells were counted using ImageJ software (Rasband,
W.S., ImageJ, U. S. National Institutes of Health, http://imagej.nih.gov/ij/).
4.3 Electrophysiological Properties of Chrna2 Cells
To assess the electrophysiological properties of subiculum Chrna2 cells, patch clamp
experiments were performed in Chrna2-Tom animals. Mice were killed by decapitation, and the
brain was dissected and placed in an ice-cold high-sucrose solution (252 mM sucrose, 24 mM
NaHCO3, 10 mM glucose, 3 mM KCl, 2 mM MgCl2, 1.25 mM NaH2PO4, and 1 mM CaCl2,
continuously oxygenated with 95% O2/5% CO2; pH 7.3). Horizontal or coronal slices (300 or
400 µm) were cut using a vibratome (Leica VT1000S, Germany) and transferred to an artificial
cerebrospinal fluid (aCSF) solution at room temperature (solution as above with 126 mM NaCl
replacing sucrose, 4.5 mM KCl and 2 mM CaCl2) for 30 minutes before recording. Slices were
recorded in a bath of aCSF perfused at a rate of 2 ml/min and heated to 30ᴼC (Temperature
controller TC-324B, Warner Instruments, Hamden, CT). Chrna2 cells were identified by Tom
fluorescence using an upright BX51WI Olympus microscope with a 60x immersion objective
(Olympus Canada, Richmond Hill, ON) and an X-cite Series 120Q fluorescence system (Lumen
28
Dynamics, Mississauga, ON). Glass patch pipettes (Warner Instruments, Hamden, CT) had a
resistance 2.0-6.0 MΩ and were filled with intra-pipette solution (144 mM K-gluconate, 10 mM
HEPES, 3 mM MgCl2, 2 mM Na2 ATP, 0.3 mM GTP, and 0.2 mM EGTA; adjusted to pH 7.2
with KOH) with neurobiotin (3-5 mg/ml, Vector Laboratories, Burlingame, CA). Chrna2 cells
were recorded using a visually guided whole-cell patch clamp technique, a MultiClamp 700B
amplifier, a DigiData 1440A digitizer and pClamp10 software, and analyzed with Clampfit10
Software (for all: Molecular Devices, Sunnyvale, CA). Recordings were kept for analysis only if
spikes overshot 0 mV and access resistance was <30 MΩ. Membrane resistance (Rm) and access
resistance (Ra) were measured in voltage clamp (vc) using pClamp10 software. The following
properties were assessed in current clamp (cc). Resting membrane potential (Vr) and spontaneous
spiking were assessed over a 30 second recording with no holding current. To assess spike
properties, cells were held at a holding potential (h.p.) of -60 mV and a series of 600 ms
depolarizing current steps was applied. The step which elicited the first spike was used to assess
spike amplitude and half width, and after-hyperpolarization (AHP) amplitude and time. The
same series of current steps was applied at a h.p. of -80 mV. The current step which elicited the
first spike at -80 mV was noted and the response to a step of equal current applied at -60 mV was
analyzed to determine firing rate, peak interspike interval (ISI) (ISI for the first two spikes),
steady state ISI (ISI for the last two spikes) and accommodation [(steady state ISI – peak
ISI)/steady state ISI]. To assess sag, if present, a series of hyperpolarizing current steps was
applied at a h.p. of -60 mV. The step which hyperpolarized the cell to -120 mV was used to
calculate sag amplitude, measured as the difference between peak and steady state
hyperpolarization, and to determine the presence of a rebound spike. The properties of subiculum
and CA1 cells were compared.
29
4.4 Chrna2 Cell Reconstruction
To examine the dendritic and axonal morphology of Chrna2 cells in subiculum and CA1,
the intracellular label neurobiotin was included in the intrapipette solution for the previously
described patch clamp experiments. Following patch clamp recording, slices were incubated in
PFA (4% in PBS) for 24 hours, and then stored in a PBS for 1-3 days. To visualize the
neurobiotin staining, slices were incubated with streptavidin-Alexa647 or streptavidin-Alexa488
(for both: 1:1000, Jackson ImmunoResearch, West Grove, PA) for 2 hours at room temperature.
The labelled Chrna2 cells were imaged and reconstructed using an upright Zeiss Imager.M2
microscope with a 20x objective (Carl Zeiss, Germany) and Neurolucida 11 software (MBF
Bioscience, Williston, VT). Axons were identified based on size and position.
4.5 Optogenetic Activation of Chrna2 Cells
To investigate the pyramidal cell post-synaptic responses elicited by Chrna2 cell
activation, an optogenetics approach was employed. This allowed the precise activation of a
population of Chrna2 cells. The opsin ChETA, a humanized variant of the excitatory opsin ChR2
[hChR2(E123T/T159C)] with faster on/off kinetics and improved spike fidelity at higher
frequency stimulation (Berndt et al., 2011) was delivered using a cre-dependent adeno-associated
viral vector (AAVdj). In this vector, the ChETA-eYFP gene is inverted and flanked by lox
sequences. Cre recombinase recognizes these sequences and flips the gene, placing it in the
correct orientation to be expressed. Thus, the opsin will only be expressed in cells expressing cre
recombinase; in this case, Chrna2 cells. To deliver the opsin, post-natal day 15 Chrna2 or
Chrna2-Tom mice were anesthetized with isoflurane and placed in a stereotaxic frame (Stoeling,
Wood Dale, IL). AAVdj-EF1α-DIO-ChETA-eYFP (Oregon Health & Science University,
30
Portland, OR) was injected at the dorsal CA1/subiculum border (0.6 µl at rate of 0.06 µl/min,
needle left in place for 5 additional minutes; coordinates (from Bregma): lateral: ±3.00,
anteroposterior: -2.70, dorsoventral: -2.05). Following surgery, animals were returned to their
home cage. Experiments were performed 2-4 weeks post-surgery to allow for virus expression.
At post-injection day 22-24, Chrna2-Tom animals were perfused (protocol as described
above) for use in immunohistochemistry experiments to validate virus expression. The
immunohistochemistry protocol was as described above with the following antibodies: a goat
anti-GFP primary antibody, reactive with eYFP (1:5000, Novus Biotechnologies, Littleton, CO),
and a rabbit anti-RFP primary antibody, reactive with Tom (1:10,000, Rockland (VWR),
Limerick, PA), followed by an anti-goat secondary antibody coupled to Alexa488 and an antirabbit secondary antibody coupled to Alexa568 (for both: 1:1000, Life Technologies, Eugene,
OR). Slices were imaged using a Zeiss Imager.M2 microscope with a 10x objective (Carl Zeiss,
Germany) and StereoInvestigator 13 software (MBF Bioscience, Williston, VT). The injection
site was located, and 1 slice per 100 µm was analyzed for 500 µm rostral and caudal to this
location. One image stack was acquired at 10x magnification in both CA1 and subiculum for
each slice, and YFP+ and Tom+ cells were counted using ImageJ software (Rasband, W.S.,
ImageJ, U. S. National Institutes of Health, http://imagej.nih.gov/ij/).
To characterize the optogenetic activation of Chrna2 cells, animals were used in patch
clamp experiments at post-injection day 16-25. (Patch clamp protocol as described above. No
neurobiotin in intrapipette solution for this and all subsequent experiments.) A 1 mm optic fibre
connected to a blue LED (474 nm or 447 nm, Luxeon Start LEDs, Brantford, ON) was
positioned over the hippocampus (maximum light power 49 mW for 474 nm or 43 mW for 447
nm, at fibre tip). To first ensure that Chrna2 cells could be driven by blue light, ChETA31
expressing Chrna2 cells, as identified by YFP expression, were patched and responses to blue
light pulses were recorded. Photocurrent was measured at the end of a 500 ms light pulse in vc at
a h.p. of -70 mV (mean of 10 recordings per cell). Spiking parameters were assessed in cc during
light pulses of 1, 5, 10, 20, 50 and 500 ms at a h.p. of -60 mV. Frequency of firing was measured
from 500 ms light pulses (mean of 10 recordings per cell). The mean delay between light onset
and spike start was measured for the first spike elicited by a 500 ms light pulse (mean of 3
recordings per cell). To assess spike fidelity, light pulses of 1, 5 and 10 ms were delivered in
trains of 1, 2, 4, 6, 8, 10, 20, 40, 60, 80, and 100 Hz (mean of 10 recordings per cell per
condition, 100 Hz not assessed for 10ms pulses).
To characterize pyramidal cell post-synaptic responses elicited by Chrna2 cell activation,
putative pyramidal cells, as identified by shape, position and firing properties, were patched in
subiculum and CA1 during optogenetic stimulation of Chrna2 cells. Regular versus burst firing
subiculum pyramidal cells were identified by responses to 600 ms depolarizing current steps.
Burst firing pyramidal cells displayed at least 1 burst during the current step (burst = 2 or more
spikes at a frequency ≥100 Hz). Responses to blue light pulses of 5, 10, 20 and 50 ms were
recorded in both vc and cc at h.p.’s from -45 to -80 mV in steps of 5 mV. Analyses were
conducted on the mean of 10 recording per condition for each cell. If a spontaneous event, such
as a spike, obscured the post-synaptic response, this recording was omitted from the mean. Any
condition for which fewer than 3 recordings remained after these omissions was not included in
the analysis. Amplitude, half width, decay time and delay between light onset and post-synaptic
response start or peak were analyzed and compared between pulse widths and pyramidal cell
types at a h.p. of -60 mV. This h.p. was chosen as it was the potential closest to mean Vr (-56.3
mV ±1.2 mV) that was far enough from spike threshold to allow little to no spiking in all cells.
32
Amplitude of the response was examined in vc across different h.p.’s to determine the reversal
potential. Paired pulse recordings were also performed in vc for 10 ms light pulses at a h.p. of 60 mV at delays of 50, 100, 200, 500 and 1000 ms. Paired pulse ratio was calculated as
amplitude of 2nd/1st response from the mean of 10 recordings per condition.
To determine what receptors play a role in post-synaptic responses to optogenetic
activation of Chrna2 cells, pharmacological methods were employed to block GABA receptors.
GABAARs were blocked using SR95531 (gabazine, 5µM, Abcam, United Kingdom). GABABRs
were blocked using CGP-55845 hydrochloride (CGP-55845, 1µM, Abcam, United Kingdom).
Responses to blue light pulses of 5, 10, 20 and 50 ms were recorded in cc at h.p.’s from -45 to 80 mV (protocols and analysis as described above) after the addition of gabazine, and again after
the subsequent addition of CGP-55845. Amplitude, half width, decay time and delay between
light onset and post-synaptic response start or peak were analyzed at a h.p. of -60 mV and pulse
width of 50 ms. This pulse width was chosen as, in all but 1 cell (n = 14/15), a component of the
response remained after gabazine application; a larger proportion than for other pulse widths.
4.6 Dual Optogenetics
A dual optogenetic technique was chosen to independently and precisely control both
Chrna2 cells and EC input to the subiculum. Optogenetic manipulation offers greater specificity
in activating EC input than commonly used electrical stimulation. Electrical stimulation of direct
EC input generally involves stimulation in the SLM where inputs from other regions, such as the
thalamus (Wouterlood et al., 1990), are also found. In the paradigm described below, optogenetic
stimulation allows only projections from the EC to be activated through expression of the opsin
in the EC and optogenetic stimulation in the hippocampus. To achieve dual optogenetic control,
33
2 different opsins must be used. The opsin expressed in EC projections must be activated by a
different wavelength of light than that expressed in Chrna2 cells. Two different opsin
combinations were tested in this project.
First, C1V1, a red-shifted excitatory opsin activated by yellow light (Yizhar et al., 2011),
was expressed in Chrna2 cells using a cre-dependent adeno-associated viral vector, AAVdjEF1α-DIO-C1V1-mCherry (Stanford University Gene Vector and Virus Core, Stanford, CA),
following the same virus injection protocol described above for CA1/subiculum. This was to be
paired with the expression of ChETA in EC projections using an adeno-associated viral vector
which allows expression under the CaMKIIα promotor, AAVdj-CaMKIIa-ChETA-eYFP
(Oregon Health & Science University, Portland, OR), yielding expression in excitatory EC
principle cells (Nathanson et al., 2009). This virus was injected according to the above described
injection protocol but using EC coordinates (from Bregma: lateral: ±4.50, anteroposterior: -3.15,
dorsoventral: -2.95).
To determine whether Chrna2 cells could be driven independently by yellow light,
Chrna2 animals injected with AAVdj-EF1α-DIO-C1V1-mCherry were used in patch clamp
experiments 18-24 days post-injection. C1V1-expressing Chrna2 cells, as identified by mCherry
expression, were patched and responses to yellow and blue light (474 nm) pulses were recorded
(protocols as described above for direct responses). Yellow light pulses were delivered by a 1
mm optic fibre connected to a yellow LED (593 nm, Luxeon Start LEDs, Brantford, ON, max
power 36 mW at tip). It was determined that C1V1-expressing Chrna2 cells were activated by
both yellow and blue light (see Results). The combination of C1V1 and ChETA was, thus,
unsuitable for dual optogenetic control and a different opsin combination was tested.
34
The second opsin combination employed the recently discovered opsin Chrimson, a redshifted excitatory opsin also activated by yellow light (Klapoetke et al., 2014), to activate EC
projections, and ChETA to activate Chrna2 cells (as described above). The excitation spectrum
for Chrimson is red-shifted 45 nm further than that for C1V1, making it better suited for dual
optogenetic control (Klapoetke et al., 2014). Chrimson was expressed in EC under the synapsin
promoter, and thus in all neurons in the injection area, using the adeno-associated viral vector
AAVdj-Syn-Chrimson-tdTomato (Oregon Health & Science University, Portland, OR).
To characterize the optogenetic activation of EC projections to subiculum, Chrna2
animals were injected with AAVdj-Syn-Chrimson-tdTomato following the above described
injection protocol and EC coordinates. At 19-23 days post-injection, animals were used in patch
clamp experiments. Chrimson-associated Tom fluorescence was visualized during these
experiments to assess virus expression using an upright BX51WI Olympus microscope with a
60x immersion objective (Olympus Canada, Richmond Hill, ON) and an X-cite Series 120Q
fluorescence system (Lumen Dynamics, Mississauga, ON). To ensure that EC cells could be
driven by yellow light, Chrimson-expressing EC cells, as identified by Tom expression, were
recorded during yellow light pulses to characterize their response to optogenetic stimulation
(protocols as described above for direct responses). To characterize the post-synaptic responses
to optogenetic activation of EC input, responses to yellow light pulses of 1, 5, 10, 20, 50 and 500
ms were recorded in subiculum pyramidal cells in both vc and cc at h.p.’s from -45 to -80 mV
(protocols and analysis as described above for post-synaptic responses). As this stimulation often
elicited post-synaptic spikes, spike timing was measured in cc from 500 ms pulses (as describe
above) and spike probability was measured for 5, 10, 20 and 50 ms pulses width and at all h.p.’s
in cc (mean of 10 recordings per condition). Spike probability in regular and burst firing
35
pyramidal cells was compared at a h.p. of -55 mV, chosen to approach spike threshold. Trains of
10 ms pulses were also delivered at frequencies of 2 to 12 Hz at a h.p. of -55 mV to assess postsynaptic spike fidelity (mean of 10 recordings per frequency).
To achieve precise, independent control of Chrimson-expressing EC input and ChETAexpressing Chrna2 cells using yellow and blue light, respectively, it is important that there is no
cross-activation of the two opsins by the opposite wavelength of light. At high light power,
Chrimson will be activated by blue light, but it has been shown that it is possible to use an
intensity low enough that Chrimson-expressing cells will not fire in response to blue light pulses
(Klapoetke et al., 2014). To maximize separation between the activation spectrum of Chrimson
and the wavelength of blue light, a 447 nm LED was used (Luxeon Start LEDs, Brantford, ON,
maximum light power 13 mW at optic fibre tip). Patch clamp experiments were performed to
assess cross-activation. Direct responses were recorded in either ChETA-expressing Chrna2 cells
or in Chrimson-expressing EC cells during pulses of increasing blue light intensity from 1.5-13
mW. Post-synaptic responses were also recorded in pyramidal cells as a second measure of
cross-activation. This is of particular importance as this project aims to determine how Chrna2
cells modulate post-synaptic responses to EC inputs in subiculum pyramidal cells. To determine
if it was possible to elicit Chrna2-cell mediated post-synaptic responses without eliciting EC
input-mediated post-synaptic responses, subiculum pyramidal cells were recorded during blue
light pulses of increasing intensity (1.5-13 mW) in animals expressing either ChETA in Chrna2
cells or Chrimson in EC projections. Recordings were analyzed to identify the lowest light
intensity able to elicit a post-synaptic potential in cc following a 5, 10, 20 or 50 ms light pulse at
a h.p. of -60 mV. These minimum intensities were compared between ChETA and Chrimson
expressing animals. The same procedure was also performed using yellow light pulses of
36
increasing intensity (1.5-20 mW) (LED 593 nm, Luxeon Start LEDs, Brantford, ON, maximum
light power 20 mW at optic fibre tip) to determine if it was possible to elicit EC input-mediated
post-synaptic responses without eliciting Chrna2-cell mediated post-synaptic responses.
4.7 Statistical Analysis
All data were tested for normality using the Shapiro-Wilks test. Normally distributed data
were compared with the following parametric tests: 2-tailed 1-sample t-test for paired-pulse
ratios, 2-tailed paired or unpaired t-tests for 2 group comparisons and 2-way ANOVAs with
Bonferroni post-hoc tests for multi-factor comparisons. ANOVAs were corrected for sphericity
with the Greenhouse-Geisser correction. Data that were not normally distributed were compared
with the following non-parametric tests: 2-tailed Mann-Whitney (unpaired) or Wilcoxon (paired)
tests for 2 group comparisons and a combination of Kruskal-Wallis (interactions and cell type
main effect) and Friedman tests (pulse width main effect) with Bonferroni corrected Wilcoxon 2tailed post-hoc tests for multi-factor comparisons. In cases of multiple comparisons, p values for
1-sample tests, 2 group tests or for interaction effects and main effects in multi-factor tests were
adjusted using the Bonferroni-Holm correction. Statistical tests were performed using SPSS
(version 22, IBM, Markham, ON), VassarStats (Dr. Richard Lowry, Vassar College,
Poughkeepsie, NY) or custom statistics programs (Dr. Joseph Rochford, McGill University,
Montreal, QC). Means are listed with ±standard error of the mean (SEM).
5. Results
5.1 Are subiculum Chrna2 cells a distinct population?
To answer this question, Chrna2 cells in subiculum were characterized according to their
Som expression, electrophysiological properties and morphology, and compared to their
37
counterparts in CA1. Post-synaptic responses elicited by Chrna2 cells in subiculum and CA1
pyramidal cells were also characterized and compared.
5.1.1 Do subiculum Chrna2 cells express somatostatin?
To address this first objective, immunohistochemistry experiments were performed with
Chrna2-Tom mice (n=4) (Fig. 1Ai). Tom+ and Som+ cells were counted in subiculum and CA1
(Subiculum n=2522, CA1 n=1733) (Fig. 1Aii). In subiculum, 88 ±1% of Tom+ cells were Som+.
In CA1, this proportion was significantly higher at 93 ±1% (Two proportion t-test with
Bonferroni- Holm correction, t =3.7, p <0.01). Overall however, as in CA1, the majority of
subiculum Chrna2 cells express Som (Fig. 1Aii). The proportion of Som+ cells that are Tom+
was also assessed, and found to be significantly greater in subiculum than in CA1 at 59 ±2% and
47 ±3%, respectively (Two proportion t-test with Bonferroni- Holm correction, t=7.2, p <0.01).
This suggests that Chrna2 cells represent a greater proportion of the Som+ interneuron
population in subiculum than in CA1 (Fig. 1Aii).
5.1.2 What are their electrophysiological properties?
To assess the electrophysiology properties of subiculum Chrna2 cells, patch clamp
experiments were performed in Chrna2-Tom animals. Chrna2 cells were patched in subiculum
and CA1 (n=18 and 16, respectively), and a number of properties, including resting membrane
properties, responses to depolarizing and hyperpolarizing current steps, and spike properties,
were assessed (Fig. 1B and Table 1). Chrna2 cells in both subiculum and CA1 fired
spontaneously at rest at slow frequency, responded to depolarizing current steps with regular
firing, and displayed a sag in their response to hyperpolarizing current steps (Table 1). No
significant differences were found for any of the electrophysiological properties examined [Two38
tailed t-tests (t’s≤1.7), Mann-Whitney tests (U’s≤201) and two-proportion t-tests (t’s≤0.9) with
Bonferroni-Holm correction, p>0.05 for all.]
5.1.3 What is their dendritic and axonal morphology?
To assess the morphology of Chrna2 cell in subiculum and compare this morphology to
CA1, cells were filled with neurobiotin in the above described patch clamp experiments. Four
subiculum Chrna2 cells and 2 CA1 Chrna2 cells were labelled sufficiently to reconstruct the
axon and dendritic arbour (Fig. 2). Unfortunately, in no cases was the distal axonal arbourization
visible. However, these reconstructions did allow assessment of general features of Chrna2 cell
dendritic and axonal morphology. For CA1 Chrna2 cells, the soma and horizontally-oriented
dendrites were located in the stratum oriens (Fig. 2A,B). The axon extended directly out toward
stratum lacunosum-moleculare, with few collaterals to stratum oriens (Fig. 2A,B). In subiculum
Chrna2 cells, the axon also extended out superficially, toward the molecular layer (Fig. 2C-F).
Their axons appeared to branch earlier than in CA1, generally resulting in more than 1 branch
extending to the molecular layer. As in CA1, subiculum Chrna2 cell had few axon collaterals to
their deep layer, the polymorphic layer. The somas of subiculum Chrna2 cells appeared to be
more widely dispersed than in CA1 (Fig. 2C-F), a feature apparent when visualizing Tom
fluorescence in Chrna2-Tom animals (Fig. 1Ai). Rather than the narrow band of stratum oriens
occupied by Chrna2 cell somas in CA1, the somas of Chrna2 cells in subiculum occupied the full
extent of the polymorphic layer, sometimes extending into the deep pyramidal layer. The
dendrites of subiculum Chrna2 cells also appeared to occupy a broader area of subiculum, with
dendrites that extended beyond the horizontal plane occupied by the dendrites of CA1 Chrna2
cells (Fig. 2C-F). Overall, the axons of both subiculum and CA1 Chrna2 cells extend
39
superficially, with those in subiculum appearing to branch earlier, and the soma and dendrites of
subiculum Chrna2 cells appear to occupy a broader area than those in CA1.
5.1.4 What post-synaptic responses do Chrna2 cells elicit in pyramidal cells?
To investigate the post-synaptic responses Chrna2 cells elicit in pyramidal cells, an
optogenetics approach was used. The excitatory opsin ChETA, tagged with eYFP, was
specifically expressed in Chrna2 cells using a cre-dependent viral vector such that Chrna2 cells
could be precisely activated with blue light.
To first validate opsin expression, the brains of Chrna2-Tom animals injected with
AAVdj-EF1α-DIO-ChETA-eYFP were examined in immunohistochemistry experiments (n=3).
Tom+ Chrna2 cells and YFP+ ChETA-expressing cells were counted in subiculum and CA1
(Subiculum n=522, CA1 n=308) (Fig. 3A,B). The YFP+ population was assessed to determine
the proportion of cells that were either Tom+ or Tom-. The majority of YFP+ cells in both
subiculum and CA1 were also Tom+ at 95 ±2% and 93 ±2%, respectively (Fig. 3B). These
proportions were not significantly different (Two-proportion t-test, t = 1.9, p>0.05). This
suggests that virus expression was specific to Chrna2 cells in both the subiculum and CA1.
It was next critical to ensure that ChETA-expressing Chrna2 cells could be driven to
spike with blue light. YFP+ cells were patched and responses to blue light pulses were recorded
(Fig. 3 C,D) (n=20: 15 subiculum and 5 CA1). Parameters were not significantly different
between subiculum and CA1, therefore the following means were taken for all Chrna2 cells from
both regions (Unpaired t-tests (t’s≤2.2) and a Mann-Whitney U test (U=55.5) with BonferroniHolm correction for multiple comparisons, p>0.05 for all)]. Mean photocurrent was found to be
-293.5 ±42.3 pA. The mean minimum pulse width required to elicit a spike was 1.4 ±0.3 ms and
40
the mean firing frequency over a 500 ms light pulse was 50.4 ±6.7 Hz. Mean delay between light
pulse onset and spike onset was found to be 3.5 ±0.3 ms. Spike fidelity was robust (mean >90%)
for frequencies up to and including 20 Hz (n=7-9) (Fig. 3D).
To characterize the post-synaptic responses elicited in pyramidal cells by Chrna2 cell
activation, putative pyramidal cells were patched in subiculum and CA1 [n=39: 29 subiculum
(17 regular and 12 burst firing), 10 CA1]. The first aim of these experiments was to compare the
responses elicited in CA1, subiculum regular firing or subiculum burst firing pyramidal cells.
Optogenetic activation elicited inhibitory post-synaptic responses in all 3 cell types (Fig. 4). The
amplitude of the IPSC was not significantly different between cell types, but did increase
significantly with increasing pulse width, with the exception of between 10 and 20 ms (Fig.
5Ai,B) (2-way ANOVA with Bonferroni-Holm corrections. Interaction and cell type main effect:
F’s≤1.1, p’s>0.05. Pulse width main effect: F=18.6, p < 0.01 with post-hoc t-test with Bonferroni
correction: t’s≥2.9, p < 0.05 for all comparisons except 10vs20ms, t=0.9, p>0.05). The results
were similar when amplitude was measured in cc, with no difference between cell type and with
amplitude increasing for every increase in pulse width (2-way ANOVA with Bonferroni-Holm
corrections. Interaction and cell type main effect: F’s≤2.4, p>0.05. Pulse width main effect:
F=60.8, p < 0.01 with post-hoc t-test with Bonferroni correction: t’s≥5.6, p < 0.001 for all
comparisons). To determine whether the kinetics of the response differed between cell types or
pulse widths, half width was assessed. The half width of the IPSP was not significantly different
between cell types, but did increase significantly with increasing pulse width (Fig. 5Aii,C)
(Kruskal-Wallis test with Bonferroni-Holm corrections for Interaction and cell type main effect:
H’s≤5.7, p>0.05. Friedman Test with Bonferroni-Holm correction for pulse width main effect:
csqr=25.7, p < 0.01 with post-hoc Wilcoxon with Bonferroni correction: W’s≤247, p < 0.05 for
41
all comparisons). Comparison of IPSP decay time yielded similar results with no significant
difference between cell type, and a longer decay time for 50 ms pulse than for 5, 10 or 20 ms
pulses (Kruskal-Wallis with Bonferroni-Holm corrections for interaction and cell type main
effect: H’s≤5.36, p>0.05, Friedman test with Bonferroni-Holm correction for pulse width main
effect: csqr=15.3, p < 0.01, with post-hoc Wilcoxon with Bonferroni correction: W’s≤72, p <
0.05 for 50 ms vs 5, 10 and 20 ms; W’s≥208, p>0.05 for remaining comparisons). To determine
if response timing differed between cell types or pulse widths, delay between light onset and
IPSP start or IPSP peak was assessed (Fig. 5Aii,D,E). For both parameters, the delay was not
significantly different between cell types (For both: 2-way ANOVA with Bonferroni-Holm
corrections. Interaction and cell type main effect: F’s≤4.1, p’s>0.05). For delay to IPSP start,
there was also no significant difference between pulse widths (2-way ANOVA with BonferroniHolm correction. Pulse width main effect: F=2.7, p>0.05). For delay to IPSP peak, the delay
increased significantly with increasing pulse width, with the exception of between 5 and 10 ms
(2-way ANOVA with Bonferroni-Holm correction. Pulse width main effect: F=84.6, p < 0.01,
with post-hoc t-test with Bonferroni correction: t’s≥4.7, p < 0.05 for all comparisons except
5vs10ms, t=1.5, p>0.05). Overall, there was no significant difference in the amplitude, kinetics
or timing of the inhibitory post-synaptic responses elicited by the activation of Chrna2 cells in
CA1, subiculum regular firing or subiculum burst firing pyramidal cells.
To further characterize the inhibitory post-synaptic responses elicited by Chrna2 cell
activation, paired-pulse recordings were performed over a range of delays to investigate any
short-term plasticity in the response (n=9 for 50 and 100ms, n=6 for 200,500 and 1000 ms).
Modest, but statistically significant, paired-pulse depression was observed for all pulse delays
42
tested from 50 to 1000 ms (Fig. 6A,B) (One-sample t-tests with Bonferroni correction: t ≥ -3.0,
p<0.05 for all.)
IPSCs were also assessed over a series of h.p.’s from -45 to -80 mV to determine the
reversal potential (Fig. 7) (n=16). IPSC amplitude decreased from -45 mV to -80 mV for all
holding potentials, but never reached zero or reversed, despite reaching h.p.’s in the range of the
reversal for the chloride, the ion associated with GABA ARs. Thus, it was of interest to next
investigate which receptors play a role in this response.
To determine what receptors may mediate the post-synaptic response to Chrna2 cell
activation, responses were recorded after bath application of gabazine, a GABAAR antagonist,
and CGP-55845, a GABABR antagonist. Application of gabazine appeared to block a significant
component of the response, leaving a smaller, slower component in the majority of cells and
completely abolishing the response in a minority (Fig. 8A). The proportion of cells in which a
component remained after gabazine application was greatest for the longest pulse width
examined (50 ms) (Cells with a response remaining after gabazine: 5 ms n=10/15, 10 ms
n=11/15, 20 ms n=10/15, 50 ms n=14/15). The smaller, slower response was completely
abolished by subsequent application of CGP-55845, and thus is likely GABABR mediated.
Response amplitude, kinetics and timing were assessed to examine the GABABR-mediated
component in further detail and to determine whether it might be different between subiculum
regular firing, subiculum burst firing and CA1 pyramidal cells (n=14: subiculum regular n=7,
subiculum burst n=4, CA1 n=3) (Fig. 8B-D). IPSP amplitude was significant smaller after
gabazine application compared to control conditions, but was not significantly different between
cell types (Fig. 8B) (2-way ANOVA with Bonferroni-Holm corrections. Interaction and cell type
main effect: F’s≤2.3, p’s>0.05. Drug main effect: F=46.7, p < 0.001). Response half width (Fig.
43
8C) and decay time were significantly longer after gabazine application compared to control
conditions, but were not significantly different between cell types (For both – 2-way ANOVA
with Bonferroni-Holm corrections. Interaction and cell type main effect: F’s≤0.5, p’s>0.05. Drug
main effect: F’s≥8.1, p’s < 0.05). Delay from light onset to IPSP start (Fig. 8D) and to IPSP peak
were also significantly longer after gabazine application compared to control conditions, but was
not significantly different between cell types (For both – 2-way ANOVA. Interaction and cell
type main effect: F’s≤2.3, p’s>0.05. Drug main effect: F’s≥44.9, p < 0.001.) Overall, the
GABABR-mediated component of the post-synaptic response elicited by Chrna2 cell activation is
similar in amplitude, kinetics and timing for CA1 and subiculum regular and burst firing
pyramidal cells.
5.2 What is the role of Chrna2 cells in the modulation of EC input to the subiculum?
To answer this question, this project aimed to independently activate EC inputs and
Chrna2 cells in the subiculum to determine how responses to excitatory EC input might change
in the presence of inhibitory input from Chrna2 cells. To achieve precise, independent control, a
dual optogenetic approach was employed.
5.2.1 How do Chrna2 cells modulate post-synaptic responses to EC input in pyramidal
cells?
To achieve control of both Chrna2 cells and EC input to the subiculum using a dual
optogenetic technique, the opsin used to activate EC projections must be activated by a different
wavelength of light than the opsin used to activate Chrna2 cells. The first opsin combination
tested was the yellow-light activated excitatory opsin C1V1 (Yizhar et al., 2011), expressed in
44
Chrna2 cells, paired with the blue-lighted activated excitatory opsin ChETA, expressed in EC
projections.
C1V1 was expressed in Chrna2 cells a cre-dependent manner through injection of the
viral vector AAVdj-EF1α-DIO-C1V1-mCherry in CA1/subiculum. To ensure that C1V1expressing Chrna2 cells could be driven to spike with yellow light, mCherry+ cells were patched
and responses to yellow light pulses were recorded (Fig. 9) (n=3). C1V1-expressing Chrna2 cells
could be driven to spike with yellow light pulses (Fig. 9B). However, they were also driven to
spike with blue light (Fig. 9C), making this opsin combination unsuitable for dual optogenetic
control.
The second opsin combination tested made use of Chrimson, an excitatory opsin also
activated by yellow light but further red-shifted than C1V1 (Klapoetke et al., 2014). This opsin
was expressed in EC projections under the synapsin promotor through injection of the viral
vector AAVdj-Syn-Chrimson-tdTomato in EC. Chrimson expression in EC projections was
paired with ChETA expression in Chrna2 cells as described previously.
To characterize Chrimson expression and to ensure that EC cells could be driven to spike
by yellow light, animals injected with AAVdj-Syn-Chrimson-tdTomato were used in patch
clamp experiments. Robust Tom fluorescence was observed throughout the EC injection site and
in the hippocampal locations expected for EC projections: the SLM in CA1, superficial
molecular layer in the subiculum, as well as the dentate gyrus (Fig. 10A,B). Chrimsonexpressing EC cells were recorded during stimulation with yellow light (n=16) (Fig. 10C,D).
Mean photocurrent was found to be -625.2 ±125.5 pA. The minimum pulse width required to
elicit a spike was 1 ms for all cells and over half of the cells (9/16) spiked only once at the
45
beginning of the light pulse. Mean time delay between light pulse onset and spike onset was
found to be 3.8 ±0.6 ms. Spike fidelity was robust (mean >90%) for frequencies up to and
including 10 Hz (n=4-5) (Fig. 10D).
To characterize the post-synaptic responses elicited in subiculum pyramidal cell by EC
input, putative pyramidal cells were patched in subiculum during optogenetic activation of
Chrimson-expressing EC projections. EC input elicited a range of responses in subiculum
pyramidal cells (n=12) (Fig. 11). In 5/12 cells, EC input elicited primarily post-synaptic spiking
for h.p.’s from -45 to -80 mV (Fig. 11A). In 7/12 cells, EC elicited a biphasic post-synaptic
response, composed of an early EPSP followed by an IPSP (Fig. 11B). For most of these
biphasic cells (6/7), EC input was able to elicit spiking, to varying degrees, at h.p.’s of -60 mV
and above. For all cells, EC input elicited spiking with a probability of >0.6 for all pulse widths
at h.p.’s from -45 mv to -60 mV, at or above resting potential for these subiculum pyramidal
cells (n=10-12) (Fig 11C). There was no significant difference in spike probability between
regular and burst firing cells or between pulse widths (Friedman’s test: pulse width main effect
p>0.05 and Mann-Whitney: cell type main effect p>0.05). Mean delays between light pulse onset
and PSP start or spike onset were found to be 5.4 ±0.4 ms and 9.6 ±0.9 ms, respectively. Spike
fidelity for pulse trains was found to be >80% for 2, 4 and 6 Hz and ≥60% for 8, 10 and 12 Hz
(n=4) (Fig 11D).
For the dual optogenetic technique to be effective, it is important that there is no crossactivation of the two opsins by the opposite wavelength of light. Previous work has shown that at
higher lighter intensities, Chrimson can be driven by blue light (Klapoetke et al., 2014), so patch
clamp experiments were performed to assess cross-activation. ChETA-expressing Chrna2 cells
and Chrimson-expressing EC cells were recorded during blue light pulses of increasing intensity
46
and both photocurrent and spike probability were quantified (Fig. 12). In ChETA-expressing
Chrna2 cells, photocurrent plateaued at 4 mW (n=3) (Fig. 12A). In Chrimson-expressing EC
cells, photocurrent did not plateau from 1.5-13 mW, but was considerable in amplitude at lower
light intensities, reaching the plateau amplitude for ChETA-expressing cells by 6 mW (n=5)
(Fig. 12B). However, more important to dual optogenetic control than photocurrent is the
separation of spike driving, the functional output of optogenetic stimulation. Spike probability in
ChETA-expressing Chrna2 cells reached 1.0 for all pulse widths tested at 1.5 to 2.5 mW (n=4)
(Fig. 12C). Chrimson-expressing EC cells began to spike at 4 mW for 50 ms pulses, but at 6 to 9
mW for shorter pulse widths (5, 10, and 20 ms) (n=6-7).These recordings suggest that light
power of 2 to 6 mW may be sufficient to elicit spiking in ChETA-expressing Chrna2 cells but
not in Chrimson-expressing EC cells for pulse widths from 5 to 20 ms. Post-synaptic responses
were also recorded in putative pyramidal cells as a second measure of cross-activation (Fig. 13).
Recordings were analyzed to identify the lowest light intensity able to elicit a post-synaptic
response in subiculum pyramidal cells in animals expressing either ChETA in Chrna2 cells or
Chrimson in EC projections. The minimum blue light intensity able to elicit a post-synaptic
response in either ChETA- or Chrimson-expressing animals overlapped for pulse widths of 5, 10
and 20 ms and was only narrowly separated for 50 ms (ChETA: n=2, Chrimson: n =4) (Fig.
13A). These results suggest that ChETA-expressing Chrna2 cells cannot be activated without
also activating Chrimson-expressing EC projections.
Recordings were also done to assess cross-activation of ChETA by yellow light. Previous
work in this lab had suggested that ChETA would not be activated by yellow light (Amilhon et
al., 2015), but experiments in this study found that ChETA-expressing Chrna2 cells could be
driven to spike with yellow light. Thus, pyramidal cells were patched, as described above, in
47
either animals expressing ChETA in Chrna2 cells or Chrimson in EC projections and responses
to yellow light were recorded (ChETA: n=5, Chrimson: n=4) (Fig. 13B). As with blue light, the
minimum intensity of yellow light able to elicit a post-synaptic response overlapped in ChETA
and Chrimson expressing animals. These results suggest that Chrimson-expressing EC
projections cannot be activated without also activating ChETA-expressing Chrna2 cells.
Overall, neither of the above described opsin combinations were suitable for dual
optogenetic control. Therefore, the direct effect of Chrna2 cell activation on EC input to
pyramidal cells was not tested.
6. Discussion
Chrna2 was recently discovered to be a specific genetic marker for CA1 OLM
interneurons (Leão et al., 2012), interneurons which are ideally positioned to modulate direct EC
input. The goal of this study was to characterize Chrna2 cells in the subiculum and investigate
their role in regulating EC input. The findings allow characterization and comparison of Chrna2
cells and their input to pyramidal cells in the subiculum and CA1, and an investigation of EC
input to the subiculum. The following will discuss these findings and their implications for future
work and for the understanding of the role of Chrna2 cells in subicular function.
6.1 Characterization of Subiculum Chrna2 Cells
Chrna2 cells in the subiculum were characterized according to their Som expression,
electrophysiological properties, and morphology and these properties were compared to those of
Chrna2 cells in CA1. The majority of subiculum Chrna2 cells were found to express Som, as
were those in CA1. This finding is consistent with the characteristic Som expression associated
with OLM interneurons in CA1 and CA3 (Morrison et al., 1982, Somogyi et al., 1984, Naus et
48
al., 1988, Katona et al., 1999, Losonczy et al., 2002, Klausberger et al., 2003, Leão et al., 2012).
Subiculum Chrna2 cells were also found to represent a larger proportion of the Som-expressing
neuron population than in CA1, which may confer differences in the modulation and effects of
Som-expressing interneurons. First, a larger proportion of Chrna2 cells in the Som-expressing
population may result in a more potent cholinergic modulatory role in the subiculum than CA1.
Among the sub-types of Som-expressing interneurons, OLM interneurons in CA1 appear to be
particularly strongly activated by cholinergic input from the medial septum (Yamano and Luiten,
1989, McQuiston and Madison, 1999, Widmer et al., 2006, Leão et al., 2012). This excitatory
drive is believed to be mediated predominantly by the nicotinic acetylcholine receptor alpha7
subunit, as well as by Chrna2 itself (Leão et al., 2012). The higher proportion of Som-expressing
cells represented by Chrna2 cells may suggest that medial septum cholinergic input excites a
larger proportion of the Som-expressing interneuron population in the subiculum than in CA1.
Given that the cholinergic input from the medial septum may play a role in theta generation
(Buzsáki, 2002), Som-expressing interneurons may differentially contribute to theta rhythm in
each sub-region. Second, it has been suggested that the peptide Som may have a physiological
role on its own, mostly characterized by inhibitory pre- and post-synaptic actions (Baraban and
Tallent, 2004, Kluge et al., 2008, Katona et al., 2014). The effects of Som-mediated inhibition
may be different in subiculum owing to the different dendritic targets of Som-expressing
interneurons. Whereas Chrna2 cells target the distal dendrites of pyramidal cells, other Somexpressing neurons, such as bistratified cells, target more proximal dendritic locations
(Maccaferri et al., 2000, Katona et al., 2014). A larger proportion of Chrna2 cells within the
Som-expressing population in subiculum suggests that a larger proportion of Som-mediated
inhibition may target the distal versus proximal dendrites. This could enhance a previously
49
described role of Chrna2 cells in biasing the effects of excitatory input to pyramidal cells to
favour input to proximal dendrites over that to distal dendrites (Leão et al., 2012). In subiculum,
this could mean enhancing the effects of CA1 input (Amaral et al., 1991) over EC input. Overall,
the expression of Som in both subiculum and CA1 Chrna2 cells highlights a similarity between
these interneurons, and the higher proportion of Som-expressing cells represented by Chrna2
cells in subiculum indicates their contribution to global Som-mediated effects that may differ
between sub-regions.
The electrophysiological properties of Chrna2 cells in the subiculum and CA1 were next
characterized and compared. Resting membrane properties, responses to depolarizing and
hyperpolarizing current steps, and spike properties were investigated. Overall, the properties of
Chrna2 cells in subiculum and CA1 were not significantly different. Chrna2 cells in both subregions displayed slow, spontaneous firing at resting membrane potential, a characteristic feature
of CA1 OLM interneurons (McBain et al., 1994, Maccaferri et al., 2000, Leão et al., 2012).
Chrna2 cells in both sub-regions responded to depolarizing current steps with regular spiking
which displayed some frequency accommodation over the duration of the pulse, a property
consistent with the known features of CA1 OLM interneurons (Sik et al., 1995, Maccaferri and
McBain, 1996a, Gloveli et al., 2005, Chittajallu et al., 2013). A sag was observed in their
responses to hyperpolarizing current steps, another characteristic feature of CA1 OLM
interneurons (Sik et al., 1995, Maccaferri and McBain, 1996a, Gloveli et al., 2005, Leão et al.,
2012, Chittajallu et al., 2013). This sag is associated with Ih current, a hyperpolarizationactivated conductance believed to play a role in the theta frequency firing recorded in OLM
interneurons (Maccaferri and McBain, 1996a, Gillies et al., 2002, but see: Kispersky et al.,
50
2012). Taken together, these results show that Chrna2 cells in both subiculum and CA1 display
comparable electrophysiological properties.
Finally, the morphology of Chrna2 cells in subiculum and CA1 was characterized and
compared. CA1 Chrna2 cells displayed a morphology that was similar to that described in
previous studies of CA1 OLM interneurons (Cajal, 1911, Lorente de Nò, 1934, McBain et al.,
1994, Sik et al., 1995, Maccaferri et al., 2000, Losonczy et al., 2002, Leão et al., 2012).
Subiculum Chrna2 cells displayed some similarities to those in CA1. In both subiculum and
CA1, the axon extended out along the deep-superficial axis toward the molecular layer or SLM,
respectively, with few collaterals to their respective deep layer. Though the Chrna2 neurons were
insufficiently labeled and distal axonal arbours difficult to visualize, the results did suggest that
axons predominantly target the distal dendrites of pyramidal cells. This is also supported by the
dense terminals seen in the most superficial layer of both the subiculum and CA1 in Chrna2-Tom
animals injected with AAVdj-DIO-ChETA-eYFP. A general difference was noted, however, in
the proximal axonal branching. Subiculum Chrna2 cell axons displayed more proximal axonal
branching than those in CA1, often resulting in more than one branch entering the molecular
layer. This could result in the axonal arbour occupying a wider expanse of the molecular layer
and potentially targeting more pyramidal cells. However, this remains to be stated definitively
without labeling complete axonal arbours. Differences were also observed in the morphology of
subiculum Chrna2 cell somas and dendrites, which generally occupied a broader area than in
CA1. This difference may be reflective of the area occupied by pyramidal cell axon collaterals in
each sub-region. In CA1, pyramidal cell axon collaterals are largely restricted to stratum oriens
(Knowles and Schwartzkroin, 1981, Tamamaki et al., 1987). These axon collaterals have been
shown to be the primary source of excitatory inputs to CA1 OLM interneuron dendrites (Lacaille
51
et al., 1987, Blasco-Ibáñez and Freund, 1995, Maccaferri and McBain, 1995, Ali and Thomson,
1998, Sun et al., 2014), which are similarly restricted horizontally in stratum oriens (McBain et
al., 1994, Sik et al., 1995, Leão et al., 2012). In subiculum, pyramidal cell axon collaterals
occupy a wider area, covering multiple layers (Harris et al., 2001). This may be associated with
the broader area occupied by Chrna2 cell dendrites found in this study. A similar arrangement
has been described in CA3 where the broader dendritic tree of OLM interneurons is mirrored by
the broader expanse of CA3 pyramidal cell axon collaterals (Gulyás et al., 1993, Müller and
Remy, 2014). It is important to note, however, that the inputs to Chrna2 cells have yet to be
specifically studied in subiculum, so this hypothesis is made on the assumption that they also
receive substantial input from pyramidal cell axon collaterals. It would be of interest to test this
assumption in future work. In general, morphological comparisons suggest that the axons of both
subiculum and CA1 Chrna2 cells extend toward their superficial layer, consistent with the
targeting of the distal dendrites of pyramidal cells. Additionally, the somas and dendrites of
subiculum Chrna2 cells occupy a broader area than those in CA1, which may reflect the
organization of pyramidal cell collaterals.
Overall, the characterization of subiculum Chrna2 cells has identified both similarities to
and distinctions from their counterparts in CA1. Chrna2 cells in subiculum and CA1 both
express Som, are identical in terms of their electrophysiological properties, and both have axons
which extend toward their superficial layers. In contrast, subiculum Chrna2 cells represent a
greater proportion of the Som-expression population and display somas and dendrites which
occupy a broader area than in CA1. These comparisons suggest that subiculum Chrna2 cells are
largely similar to those in CA1, but differences in morphology and proportion may confer subtle
differences in their role in modulating subicular activity. To assess this notion more directly, this
52
study next characterized the post-synaptic responses elicited in pyramidal cells by Chrna2 cell
activation.
6.2 Optogenetic Investigation of Chrna2 Cell-Mediated Post-Synaptic Responses
An optogenetic technique was used to activate Chrna2 cells. The excitatory opsin ChETA
was specifically expressed in Chrna2 cells in subiculum and CA1, and stimulation with blue light
elicited robust spiking. Delay between light pulse onset and spike onset was consistent with a
direct response (Amilhon et al., 2015). Spike fidelity was also found to be robust for stimulation
frequencies from 1 to 20 Hz, inclusive, demonstrating that optogenetic activation of Chrna2 cells
can be used to drive firing through the theta frequency range. As optogenetic activation of
ChETA-expressing Chrna2 cells was shown to be successful, post-synaptic responses elicited in
pyramidal cells could be investigated.
Post-synaptic responses from optogenetically activated Chrna2 cells were recorded in
CA1 pyramidal cells as well as from subiculum regular and burst firing pyramidal cells.
Activation of Chrna2 cells elicited inhibitory post-synaptic responses in all pyramidal cells from
both regions as expected from a previous report (Leão et al., 2012). The responses obtained from
both regions were compared between different light pulse widths and cell types. For all cell types
and pulse widths, the response started 5-7 ms after light pulse onset, a delay which is consistent
with a post-synaptic response (Amilhon et al., 2015). Generally, longer pulse widths produced
larger amplitude responses with longer half widths, decay times, and delays between light pulse
onset and response peak. This result corresponds to the above described recordings from Chrna2
cells during optogenetic stimulation which indicated that longer pulse widths would elicit more
spikes in Chrna2 cells. The amplitude, kinetics, and timing of responses in CA1, subiculum
53
regular firing, and subiculum burst firing pyramidal cells were not significantly different
between cell types. The similarity in the responses in subiculum and CA1 suggests that Chrna2
cells target pyramidal neurons to a similar extent and may mediate similar modulatory effects in
the pyramidal cells of these sub-regions.
It has been suggested that IPSC size and kinetics, as measured in the soma, are associated
with an interneuron’s target region on the dendritic tree (Maccaferri et al., 2000). Similar postsynaptic response size and kinetics between CA1 and subiculum further support this study’s
findings in regards to Chrna2 cell morphology which suggest that Chrna2 cells in subiculum, like
those in CA1, likely target the distal dendrites of pyramidal cells. CA1 Chrna2 cells have been
found to attenuate direct EC excitatory inputs to pyramidal cell distal dendrites while facilitating
Schaffer collateral excitatory inputs to more proximal dendrites (Leão et al., 2012). If subiculum
Chrna2 cells are playing a similar role, they could also serve to attenuate direct EC input and to
facilitate proximal dendritic input, which consists largely of projections from CA1 (Amaral et al.,
1991). Since the Chrna2 cell actions in modulating Schaffer collateral input may be mediated by
disinhibition of stratum radiatum interneurons (Leão et al., 2012), future work could investigate
any potential difference between the responses subiculum and CA1 Chrna2 cells elicit in local
interneurons.
The lack of any significant difference in the post-synaptic responses recorded in
subiculum regular and burst firing pyramidal cells suggests that Chrna2 cells do not
preferentially target one pyramidal cell type over the other. Greene and Mason (1996)
demonstrated that Som had a more robust effect on bursting than regular firing neurons, but the
results of this study suggest that, for at least one sub-population of Som-expressing neurons, the
direct GABAergic effects do not differ between cell types. However, it is important to also
54
consider that dendritic inhibition is particularly effective at inhibiting burst firing (Royer et al.,
2012), in part due to its ability to modulate plateau potentials (Lovett-Barron et al., 2012). Thus,
though the inhibitory response itself may be similar, Chrna2 cell-mediated inhibition to distal
dendrites may have a more significant functional effect on burst firing than regular firing
pyramidal cells owing to its potential to modulate the output mode of burst firing neurons; an
interesting possibility to investigate in future work.
To further characterize the post-synaptic response elicited by Chrna2 cells, paired-pulse
recordings were performed. Modest paired-pulse depression was observed for all pulse delays
tested. This is consistent with previous work that has found small paired-pulse depression in
paired recordings of CA1 OLM interneurons and pyramidal cells (Maccaferri et al., 2000,
Minneci et al., 2007). This suggests that the response to Chrna2 cell input in pyramidal cells may
decrease over time during repetitive stimulation. If this was the case during theta frequency
firing, Chrna2 cells may lose some of their modulatory effect over time. However, the depression
was minor and may also be counterbalance by facilitation in the activation of Chrna2 cells by
pyramidal cell axon collaterals. Indeed, excitatory input from CA1 pyramidal cells to OLM
interneurons has been found to facilitate over repeated stimulation (Ali and Thomson, 1998,
Losonczy et al., 2002, Kim, 2014), an effect which could enhance their output and help maintain
their inhibitory influence despite modest short term depression in the post-synaptic responses
they elicit in pyramidal cells.
Post-synaptic responses were also recorded at h.p.’s from -45 to -80 mV to determine
their reversal potential. The inhibitory response decreased as the h.p. was more negative, but did
not reverse, nor did it reach zero amplitude. This was surprising as the range of h.p.’s examined
were close to the reversal potential for chloride, approximately -80 mV for the solutions used,
55
the ion conductance associated with GABAARs (Olsen and DeLorey, 1999). There are several
possible explanations for this result. First, the membrane potential recorded and held at the soma
may have been different from the membrane potential in the distal dendrites. The
hyperpolarization produced in vc in the soma is likely smaller further out in the dendritic tree.
Thus, the Chrna2 cell inhibitory input arriving at distal dendrites may have activated chloride
conductances in a milieu with a membrane potential that was, in fact, higher than the reversal
potential for chloride, despite the soma being held close to this reversal potential. Second, there
may be receptors other than GABAARs involved in the response. For example, the conductance
associated with GABABRs is potassium-mediated, making its reversal potential lower than
GABAARs at closer to -90 mV (Sodickson and Bean, 1996, Olsen and DeLorey, 1999, Degro et
al., 2015). Aiming to determine whether this was the case, GABA receptor antagonists were used
to tease apart the components of the response.
Post-synaptic responses to Chrna2 cell activation were recorded in subiculum and CA1
after the application of the GABAAR antagonist gabazine. In the majority of cells, gabazine
attenuated a significant component of the response, but a smaller amplitude, slower component,
consistent with a GABABR-mediated response, remained (Sodickson and Bean, 1996, Booker et
al., 2013, Degro et al., 2015). This remaining component was abolished by the addition of the
GABABR antagonist CGP-55845. To characterize the GABABR-mediated component remaining
after gabazine application and to investigate whether it differed between CA1, subiculum regular
firing and subiculum burst firing pyramidal cells, responses recorded pre- and post-gabazine
application were compared. The responses after gabazine application were significantly smaller
in amplitude and displayed slower kinetics and longer delays to response start and peak. No
significant differences were found for GABABR-mediated response amplitude, kinetics or timing
56
between cell types. This further solidifies this study’s previous findings showing that Chrna2
cells elicit similar responses in both CA1 and subiculum and both subiculum pyramidal cell
types. Of note, differences between the GABABR conductance between regular and burst firing
cells may be observable in certain activity states, in particular, repetitive, high frequency activity
able to elicit Som release. Som has been found to reduce GABABR conductance to a greater
extent in burst firing than in regular firing cells (Greene and Mason, 1996). Thus, in cases of
repetitive, high frequency Chrna2 cell activity, when Som may be co-released with GABA, the
GABABR component of the post-synaptic response may be smaller in burst firing cells.
Overall, these results suggest that Chrna2 cell-mediated post-synaptic responses in both
subiculum and CA1 are composed of both GABAAR and GABABR mediated components. This
study is the first to demonstrate that there is a GABABR component to OLM interneuron
postsynaptic inhibition. Maccaferri et al. (2000) demonstrated the involvement of GABAARs in
the input from OLM interneurons to pyramidal cells in CA1, but used an intrapipette solution
which blocked GABABR-mediated conductance and, thus, were unable to investigate whether
these receptors also played a role. The optogenetic manipulation used here is also more likely to
uncover a GABABR component, as activation of these receptors generally requires higher levels
of interneuron activity (Baraban and Tallent, 2004), and whereas as experimental designs such as
paired-patch activate only a single cell, the optogenetic stimulation here activates a population of
Chrna2 cells. Future work investigating the role of Chrna2 cells should consider the potential
physiological role of these two GABA receptor-mediated responses.
57
6.3 Optogenetic Investigation of EC Input-Mediated Post-Synaptic Responses
The second aim of the study was to investigate how Chrna2 cells modulate post-synaptic
responses to EC input. To determine this, it was first necessary to investigate the post-synaptic
responses elicited by EC input. EC projections to subiculum were activated using an optogenetic
technique, which allows more specificity than typically used method of electrical stimulation to
activate EC inputs in hippocampus. The goal was to activate EC input using the red-shifted
excitatory opsin Chrimson, such that Chrna2 cells and EC input could be activated independently
by blue and yellow light, respectively.
Chrimson-expressing EC cell were consistently activated by stimulation with yellow light
pulses as short as 1 ms, with most cells firing a single spike regardless of pulse width. Delay
between light pulse onset and spike onset was consistent with a direct response (Amilhon et al.,
2015). Spike fidelity was found to be robust for stimulation at frequencies from 1 to 10 Hz,
inclusive, demonstrating that optogenetic activation of EC cells can be used to drive firing
through the theta frequency range. As optogenetic activation of Chrimson-expressing EC cells
was shown to be successful, post-synaptic responses elicited in subiculum pyramidal cells by EC
input could be investigated.
EC input elicited a range of responses in subiculum pyramidal cells. In approximately
half the pyramidal cells recorded, EC input elicited a spike regardless of h.p., suggesting strong
excitatory input. In the other half, EC input elicited a biphasic response and began to elicit
spiking for h.p’s of -60 mV and above. This is suggestive of an EPSP elicited directly from EC
input followed by an IPSP likely mediated by feedforward inhibition, consistent with results
from a previous report (Amilhon et al., 2015). This inhibition likely decreased the effect of the
58
direct EC input such that it was only able to elicit spiking when the pyramidal cell was held
closer to its spiking threshold. It is unlikely that this feedforward inhibition is mediated by
Chrna2 cells, as previous work has suggested that Chrna2 cells do not received input from the
EC (Leão et al., 2012) and participate exclusively in feedback inhibition in CA1 (Maccaferri and
McBain, 1995, 1996b).
Spike probability was assessed, and no significant difference was found between regular
and burst firing pyramidal cells. Using in vivo electrical stimulation in rats, Gigg et al. (2000)
found preferential excitatory EC input to burst versus regular firing subiculum pyramidal cells.
But this study suggests that, as was found for Chrna2 cell input, EC projections provide
excitatory input to both cells types and do not preferentially target one subiculum pyramidal cell
type over the other. Differences between these findings may be due to differences in species, an
in vitro versus in vivo paradigm or optogenetic versus electrical stimulation. Overall, the results
of this study suggest that, in regards to EC input to subiculum and the potential modulation of the
input by Chrna2 cells, there are no differences between regular and burst firing pyramidal cells.
To test hypothesis directly, this study aimed to activate both inputs together in dual optogenetic
experiments, but this was unachievable (see below).
EC input was able to elicit trains of post-synaptic spiking with spike fidelity >60% for
frequencies of 2-12 Hz suggesting that optogenetic stimulation of EC projections to subiculum
could drive spiking in subiculum pyramidal cells throughout the theta range. Furthermore, the
delay between light onset and spike onset in pyramidal cells was suggestive of a post-synaptic
response (Amilhon et al., 2015). Although post-synaptic spiking was likely the result of a direct
EC input, one limitation of this study is that indirect EC input from the CA3→CA1→subiculum
pathway cannot be excluded and may also have contributed to the depolarization and spiking.
59
However, the delay between light onset and PSP start was consistent with that observed for the
direct post-synaptic responses elicited by Chrna2 cell activation. Thus, one can be confident that
the post-synaptic response elicited by optogenetic stimulation of EC projections is mediated, at
least in part, by direct EC input.
Overall, these results suggest that EC projections to subiculum can directly evoke postsynaptic responses capable of eliciting spikes in subiculum pyramidal cells. With the success of
the activation of EC input to subiculum, the project next aimed to pair Chrna2 cell and EC input
activation to investigate the modulatory effects of Chrna2 cells.
6.4 Dual Optogenetic Control
To precisely and independently control both Chrna2 cells and EC input, it was essential
that ChETA-expressing Chrna2 cells and Chrimson-expressing EC projections be exclusively
activated by blue or yellow light, respectively. Unfortunately, after investigating two different
opsin pairings, robust separation of the activation of each population was not achievable. In the
most promising opsin paring, low intensity blue light, within a narrow range (2 to 6 mW), was
able to elicit spiking in ChETA-expressing Chrna2 cells while not eliciting spiking in Chrimsonexpressing EC cells. However, when cross-activation was assessed in pyramidal cells by
examining post-synaptic responses, no intensity of blue light was able to elicit exclusive Chrna2mediated post-synaptic responses while excluding EC activation. Likewise, there was no
intensity of yellow light able to elicit only EC-mediated post-synaptic responses while excluding
Chrna2-mediated activation. These results show that dual independent optogenetic control was
not possible and, thus, a direct demonstration of how Chrna2 cell-mediated inhibition modulates
EC input was not achieved. However, the investigation of cross-activation has laid the
60
groundwork for future studies which could pair Chrimson with another newly discovered
excitatory opsin, Chronos (Klapoetke et al., 2014). Chronos is excited by blue light, but has
faster kinetics and higher sensitivity than ChETA (Klapoetke et al., 2014). It can, thus, be
activated by lower intensities of blue light, allowing a greater separation between the minimum
light intensity able to elicit activation of Chrna2 cells and EC projections if it were paired with
Chrimson. As this study found that ChETA can be activated by yellow light, future work could
also be improved by using a red LED to activate Chrimson, again enhancing the separation
between the minimum light intensity able to elicit activation of Chrna2 cells and EC projections.
6.5 Implications for the Role of Chrna2 Cells in Subicular Function
Despite the inability to achieve dual optogenetic control, the results of this study have
enhanced the understanding of Chrna2 cells in subiculum and have implications for the
understanding of their role in subicular function. In general, the results of this study indicate that
subiculum Chrna2 cells are similar to those in CA1, particularly in their effects on pyramidal
cells. These similarities suggest that they may also serve similar roles. Two areas of particular
interest, worthy of discussion and future work, are roles in theta rhythm and memory function.
OLM interneurons in CA1 are believed to play a role in the modulation of theta
oscillations. As described above, they are known to fire spontaneously at theta frequency, to
increase their firing during theta rhythm and phase lock to the trough of the theta cycle
(Maccaferri and McBain, 1996a, Klausberger et al., 2003, Gloveli et al., 2005, Varga et al.,
2012). These properties may allow CA1 OLM interneurons to phase-modulate EC input
(Klausberger et al., 2003), which also plays a role in theta generation (Kamondi et al., 1998,
Buzsáki, 2002, Mizuseki et al., 2009). Given the similarities this study has found in the firing,
61
axonal projections and inhibitory post-synaptic responses elicited by Chrna2 cells in the
subiculum and CA1, subiculum Chrna2 cells may also phase-modulate EC input. A recent study
by Amilhon et al. (2015) found that silencing Som-expressing interneurons during the activation
of EC projections significantly attenuated the effect of EC input on theta rhythm in
CA1/subiculum. Future work focusing specifically on Chrna2 cells could determine what role
this sub-population of Som-expressing interneurons plays in this effect. Additionally, the finding
that Chrna2 cells activate GABABR-mediated conductances in subiculum and CA1 pyramidal
cells may suggest a novel way in which these cells could modulate theta rhythm. Scanziani
(2000) found that during methacholine-induced theta rhythm, increasing or decreasing GABABR
activation could decrease or increase the frequency of theta oscillations, respectively. As
GABABR activation can depend on the level of interneuron activation (Nurse and Lacaille, 1997,
Scanziani, 2000), the level of activity in Chrna2 cells may affect the pacing of theta frequency,
an effect worth considering in future studies investigating the role of Chrna2 cells in theta
modulation.
OLM interneurons in CA1 are also believed to play a role in memory, a prime function of
the hippocampus. Lovett-Barron et al. (2014) found that inhibition of direct EC input to the distal
tuft of CA1 pyramidal cells was necessary for the formation of contextual fear memories,
inhibition they suggest is provided by OLM interneurons. Furthermore, Som has also been
shown to be necessary for the acquisition of contextual fear memories (Kluge et al., 2008). Given
the similarity in their post-synaptic responses, subiculum Chrna2 cells could also contribute to
memory formation by excluding the discrete sensory information provided by direct EC input or
through Som-mediated modulation, particularly since the subiculum is also known to play a role
in spatial learning and memory (Galani et al., 1997, Laxmi et al., 1999, Oswald and Good, 2000,
62
O'Mara et al., 2009). CA1 OLM interneurons have also been associated with cognitive decline in
aging. In a study of aging rats, Stanley et al. (2012) observed a selective loss of OLM
interneuron and a decrease in inhibition of direct EC input in CA1. Considering their similarities
to CA1, the loss of Chrna2 cells in subiculum may also give rise to cognitive impairments. This
idea is supported by studies which find that the subiculum is among the earliest brain regions
affected in Alzheimer’s disease (Adachi et al., 2003, de la Prida et al., 2006, Scher et al., 2011,
Goutagny et al., 2013, Trujillo-Estrada et al., 2014), and that Som-expressing interneurons in
subiculum are particularly vulnerable (Trujillo-Estrada et al., 2014).
Overall, more work is needed to fully understand the role of Chrna2 cells in subicular
function, but their similarities to CA1 Chrna2 cells suggest that investigations of the modulation
of theta rhythm and of memory function may be an interesting starting point.
6.6 Limitations
The main limitation of this study was the inability to perform the dual optogenetic
technique. Without the capacity to independently control Chnra2 cells and EC input, there could
be no direct demonstration of the potential modulatory effects of Chrna2 cells on EC input.
However, given the similarities demonstrated in the post-synaptic responses to Chrna2 cell
activation in CA1 and subiculum pyramidal cells, one can suggest that Chrna cells in subiculum
could attenuate the effects of direct EC input as has been observed in CA1 (Leão et al., 2012). As
described above, future studies making use of the opsin pair of Chronos and Chrimson may
achieve dual optogenetic control and answer the question of how Chrna2 cells modulate EC
input in the subiculum directly.
63
Another limitation of this study is that all optogenetic investigation was restricted to the
dorsal hippocampus. Given the proposed division of functional roles between dorsal and ventral
subiculum (O'Mara, 2005, de la Prida et al., 2006), these findings may not be generalizable to the
ventral subiculum.
7. Final Summary
The goal of this study was to characterize Chrna2 cells in the subiculum, and to
investigate their role in the regulation of EC input. Characterization of subiculum and CA1
Chrna2 cells found similarities in Som expression, electrophysiological properties, and axon
directionality. While differences in soma and dendritic morphology and in proportion of the
Som-expressing population may confer subtle differences, these results suggest that subiculum
and CA1 Chrna2 cells are predominantly similar. This idea was reinforced by investigation of
the post-synaptic responses elicited by Chrna2 cells in subiculum and CA1 pyramidal cells.
These post-synaptic responses were inhibitory, composed of both GABAAR- and GABABRmediated components, and not significantly different between CA1, subiculum regular firing and
subiculum bursting firing pyramidal cells. Overall, these findings suggest that subiculum and
CA1 Chrna2 cells may play similar modulatory roles.
To investigate the role of Chrna2 cells in the context of EC input modulation, postsynaptic responses to EC input to subiculum pyramidal cells was characterized and found to be
similar in both regular and burst firing pyramidal cells. Unfortunately, independent optogenetic
control of Chrna2 cells and EC input was not achieved, and thus, the direct effect of Chrna2 cellmediated inhibition on EC input to pyramidal cells could not be demonstrated.
64
Despite the inability to achieve dual optogenetic control, these findings suggest that
subiculum and CA1 Chrna2 cells are largely equivalent and are likely to play similar roles both
sub-regions, roles which may include the regulation of theta rhythm and a contribution to
learning and memory functions. Overall, this study has provided the first in-depth
characterization of Chrna2 cells in subiculum and has enhanced the understanding of how this
interneuron population may contribute to subicular function.
65
Figures
Figure 1. Characterization of somatostatin (Som) expression and electrophysiological properties
of Chrna2 cells. A) i) Immunohistochemistry: 10x images from a Chrna2-Tom animal treated
with an anti-Som antibody. Chrna2 cell are visualized with Tom fluorescence in red. Som+ cells
are in green. Co-expressing cells are shown in yellow. Left: subiculum. Right: CA1. ML:
molecular layer. PL: polymorphic layer. SLM: stratum lacunosum-moleculare. SR: stratum
radiatum. SP: stratum pyramidale. SO: stratum oriens. Scale bar = 100 µm. ii) Venn diagrams
depicting proportions of Som+ and Tom+. Tom+Som- cells in red, Tom+Som+ cells in yellow
and Tom-Som+ cells in green. Exact number of cells as indicated. Left: subiculum. Right: CA1.
Animals: n=4. Cells: subiculum n=2522, CA1 n=1733. B) Representative Chrna2 cells recorded
66
in subiculum (left) and CA1 (right). i) Traces showing current clamp (cc) recordings of
responses to depolarizing and hyperpolarizing current steps. ii) Top: 60x DIC images showing
recorded cell in Bi. Bottom: Corresponding 60x image of Tom fluorescence. Arrowhead denotes
position of recorded cell. Left: subiculum. Right: CA1. Scale bar = 15 µm.
Table 1. Electrophysiological properties of Chrna2 cells recorded in subiculum and CA1. ±SEM.
Subiculum CA1
n=18
n=16
387.8
±47.2
378.1 ±35.6
Rm (MΩ)
Ra (MΩ) 15.7 ±1.3
Vr -49.9 ±1.5
Spontaneous spiking 83%
Frequency of spontaneous 4.5 ±1.9
spiking (Hz)
Firing rate (Hz) 26 ±2.1
14.3 ±1.1
-52.3 ±1.3
94%
6.4 ±1.9
23.4 ±2.4
Max firing ISI (ms) 35.8 ±3.1
45.7 ±5.6
Steady state ISI (ms) 49.8 ±3.6
54.3 ±5.5
Accommodation % 26.7 ±4.2
15.9 ±4.8
Sag amplitude (mV) 11.5 ±1.7
11.7 ±1.7
Rebound spike 61%
63%
Spike amplitude (mV) 91.8 ±1.4
91.2 ±2.5
Spike half width (ms) 1.5 ±0.1
1.2 ±0.1
AHP Amplitude (mV) -17.2 ±0.7
-18.0 ±0.6
AHP time (ms) 136.5 ±13.1
145 ±13.1
67
68
Figure 2. Chrna2 cell morphology. A) Reconstruction of a Chrna2 cell patched in CA1. Axon in
red. Soma and dendrites in black. SLM: stratum lacunosum-moleculare. SP: stratum pyramidale.
SO: stratum oriens. B) Same as in A for another CA1 Chrna2 cell. C) Same as in A for a Chrna2
cell patched in subiculum. ML: molecular layer. PL: polymorphic layer. D-F) Same as in C for 3
other subiculum Chrna2 cells. Scale bar = 50 µm for all. Process diameter is not to scale.
69
Figure 3. Anatomical and electrophysiological characterization of ChETA-YFP expression in
Chrna2 cells. A) 10x images from a Chrna2-Tom animal injected with AAVdj-DIO-ChETAeYFP in CA1/subiculum. Chrna2 cells are visualized with Tom fluorescence in red. ChETAexpressing cells are visualized with YFP fluorescence in green. Co-expressing cells are shown in
yellow. Top: subiculum. Bottom: CA1. ML: molecular layer. PL: polymorphic layer. SLM:
stratum lacunosum-moleculare. SO: stratum oriens. Scale bar = 100 µm. B) Mean proportion of
YFP+ cells that were Tom+ or Tom- in subiculum (red) and CA1 (blue). Animal: n = 3. Cells:
subiculum n=522, CA1 n=308. C) Patch clamp recordings from a subiculum Chrna2 cell
expressing ChETA. i) Responses to depolarizing and hyperpolarizing current steps in cc. ii)
Mean recorded photocurrent produced by a 500 ms light pulse (blue box) in voltage-clamp (vc)
at a holding potential (h.p.) of -70 mV. Spike peak was truncated to expand view of plateau
70
photocurrent. iii) Spiking in response to a 500 ms light pulse (blue box) in cc at a h.p. of -60 mV.
iv) Spiking in response to an 8 Hz train of 5 ms light pulses (blue boxes) in cc at a h.p. of 60mV. D) Mean spike fidelity for trains of 1, 5 and 10 ms light pulses at frequencies of 1 to 100
Hz. n=7-9. Error bars = SEM.
71
Figure 4. Example post-synaptic responses elicited in pyramidal cell by optogenetic activation of
Chrna2 cells. A) Patch clamp recording from a CA1 pyramidal cell. i) Responses to depolarizing
and hyperpolarizing current steps in cc. ii) Mean inhibitory post-synaptic potential (IPSP) in
response to a 50 ms light pulse (blue box) recorded in cc at a h.p. of -60 mV. iii) Mean inhibitory
post-synaptic current (IPSC) in response to a 50ms light pulse (blue box) recorded in vc at a h.p.
of -60 mV. B) Same as in A for a subiculum burst firing pyramidal cell. C) Same as in A for a
subiculum regular firing pyramidal cell. Scales consistent through i, ii and iii.
72
Figure 5. Quantification of post-synaptic responses elicited in pyramidal cells by optogenetic
activation of Chrna2 cells. Ai) Example mean inhibitory post-synaptic current (IPSC) in
response to a 50ms light pulse (blue box) recorded in vc at a h.p. of -60 mV in a subiculum
73
regular firing (orange), subiculum burst firing (red) or CA1 (blue) pyramidal cell. ii) Same as in
Ai for mean inhibitory post-synaptic potential (IPSP) recorded in cc. B) Mean IPSC amplitude
for all cells recorded across cell type and pulse width. Subiculum regular n=17, subiculum burst
n=12, CA1 n=10. Colour legend maintained throughout the figure. C) Same as in B for mean
half width. D) Same as in B for mean delay between light pulse onset and PSP start. E) Same as
in B for mean delay between light pulse onset and IPSP peak. Error bars = SEM. * = p<0.05. ns
= Not significantly different, p>0.05.
Figure 6. Paired pulse recordings of post-synaptic responses elicited by optogenetic activation of
Chrna2 cells. A) Example mean IPSCs recorded at a h.p. of -60 in response to two 10 ms blue
light pulses (blue boxes) with a delay of 100 ms (i) or 200 ms (ii). B) Mean paired-pulse ratio for
delays for 50 to 1000 ms. n=9 for 50 and 100ms, n=6 for 200, 500 and 1000 ms. Error bars =
SEM. * = p<0.05.
74
Figure 7. Changes in post-synaptic responses to optogenetic activation of Chrna2 cell as a
function of holding potential. A) Example mean IPSC in response to a 50 ms light pulse (blue
box) at h.p.’s of -50 to -80 mV. B) Mean IPSCs recorded at h.p.’s from -45 to -80 mV for 5 to 50
ms light pulses. n = 16. Error bars = SEM.
75
Figure 8. Quantification of post-synaptic responses elicited in pyramidal cell by optogenetic
activation of Chrna2 cells during application of GABAAR and GABABR antagonists. Ai)
Example mean IPSP in response to a 50ms light pulse (blue box) at a h.p. of -60 mV recorded in
a control aCSF (CTRL), after the addition of gabazine (GABAAR antagonist) (GBZ) , and after
the subsequent addition of CGP-55845 (GABABR antagonist)(GBZ+CGP). B) Mean IPSP
amplitude for all cells recorded in both control conditions and after gabazine application across
cells type. Subiculum regular n=7, subiculum burst n=4, CA1 n=3. Colour legend maintained
throughout the figure. C) Same as in B for mean half width. D) Same as in B for mean delay
between light pulse onset and PSP start. Error bars = SEM. * = p<0.05. *** = p<0.001.
76
Figure 9. Patch clamp recordings in a C1V1-expressing Chrna2 cell. A) Responses to
depolarizing and hyperpolarizing current steps in cc. B) Responses to yellow light. i) Mean
recorded EPSP produced by a 500 ms light pulse (yellow box) in cc at a h.p. of -60 mV. ii) Mean
recorded photocurrent produced by a 500 ms light pulse (yellow box) in vc at a holding potential
h.p. of -70 mV. iii) Spiking in response to a 500 ms light pulse (yellow box) in cc at a h.p. of -55
mV. C) Same as in B for blue light.
77
Figure 10. Anatomical and electrophysiological characterization of Chrimson-Tom expression in
EC cells. A) 4x images of a coronal hippocampal slice showing: subiculum (Sub), CA1 and
dentate gyrus (DG) (i) and CA1 and dentate gyrus (ii). Left: DIC image. Right: Corresponding
Tom fluorescence. Note that fluorescence in the hippocampus is restricted to the SLM of CA1,
molecular layer of subiculum and dentate gyrus, the expected pattern for EC projections. B)
Same as in A showing the EC injection site and CA1. C) Recording from a Chrimson-expressing
EC cell. i) Responses to depolarizing and hyperpolarizing current steps in cc. ii) Mean recorded
photocurrent produced by a 500 ms light pulse (yellow box) in vc at a h.p. of -70 mV. Spike peak
was truncated to expand view of plateau photocurrent. iii) Spiking in response to a 500 ms light
pulse (yellow box) in cc at a h.p. of -60mV. iv) Spiking in response to an 8 Hz train of 5 ms light
pulses (yellow boxes) in cc at a h.p. of -60 mV. D) Mean spike fidelity for trains of 1, 5 and 10
ms light pulses at frequencies of 1 to 100 Hz (n=4-5). Error bars = SEM.
78
Figure 11. Post-synaptic responses elicited in subiculum pyramidal cells by optogenetic
activation of EC input. A) Patch clamp recording from a subiculum pyramidal cell. i) Responses
to depolarizing and hyperpolarizing current steps in cc. ii) Spiking in response to a 10ms light
pulse (yellow box) in cc at a h.p. of -55 mV. Note that this cell is representative of cells which
responded primarily with post-synaptic spikes (n=5/12). iii) Corresponding response in vc to a
10ms light pulse (yellow box) at a h.p. of -55 mV. B) Recording from another subiculum
pyramidal cell. i) Same as in A. ii) Mean biphasic post-synaptic potential in response to a 10 ms
light pulse (yellow box) in cc at a h.p. of -55 mV. Note that this cell is representative of cells
which responded primarily with biphasic post-synaptic potentials (n=7/12). iii) Corresponding
response in vc to a 10ms light pulse (yellow box) at a h.p. of -55 mV. C) Mean post-synaptic
spike probability for responses elicited in pyramidal cells by light pulses of 5, 10, 20 and 50 ms
at h.p. from -45 to -80 mV (n=10-12). D) Mean spike fidelity for trains of 5 ms light pulses at
frequencies of 2 to 12 Hz (n=4). All error bars = SEM.
79
Figure 12. Responses of ChETA-expressing Chrna2 cells and Chrimson-expressing EC cells to
blue light stimulation. A) Response in ChETA-expressing Chrna2 cells. i) Diagram depicting
experimental setup: patch clamp recording of a Chrna2 cell during blue light stimulation. ii)
Mean photocurrent elicited by 500 ms blue light pulses of increasing intensity recorded in vc at a
h.p. of -70 ms. n=3. B) Same as in A for a Chrimson-expressing EC cells. n=5. C) Mean spike
probability in ChETA-expressing Chrna2 cells or Chrimson-expressing EC cells at increasing
80
blue light intensity for light pulse of 5, 10, 20, and 50 ms. ChETA Chrna2: n=4. Chrimson EC:
n=6-7. Error bars = SEM.
Figure 13. Post-synaptic responses in pyramidal cells in ChETA- or Chrimson-expressing
animals in response to blue or yellow light stimulation. A) Post-synaptic responses elicited by
blue light. i) Diagram depicting experimental setup: patch clamp recording of a subiculum
pyramidal cell during blue light stimulation. ii) Mean minimum light intensity able to elicit a
post-synaptic response in pyramidal cells in animals expressing either ChETA in Chrna2 cells
(n=2) or Chrimson in EC projections (n=4). B) Same as in A for responses to yellow light.
ChETA: n=5, Chrimson: n=4.
81
References
Adachi M, Kawakatsu S, Hosoya T, Otani K, Honma T, Shibata A, Sugai Y (2003) Morphology
of the inner structure of the hippocampal formation in Alzheimer disease. Am J
Neuroradiol 24:1575-1581.
Ali AB, Thomson AM (1998) Facilitating pyramid to horizontal oriens-alveus interneurone
inputs: dual intracellular recordings in slices of rat hippocampus. J Physiol 507 (Pt
1):185-199.
Amaral DG, Dolorfo C, Alvarez-Royo P (1991) Organization of CA1 projections to the
subiculum: a PHA-L analysis in the rat. Hippocampus 1:415-435.
Amilhon B, Huh CY, Manseau F, Ducharme G, Nichol H, Adamantidis A, Williams S (2015)
Parvalbumin Interneurons of Hippocampus Tune Population Activity at Theta Frequency.
Neuron 86:1277-1289.
Anderson P, Morris R, Amaral D, Bliss T, O'Keefe J (2007) The Hippocampus Book. New York,
NY: Oxford.
Baraban SC, Tallent MK (2004) Interneuron Diversity series: Interneuronal neuropeptides-endogenous regulators of neuronal excitability. Trends Neurosci 27:135-142.
Baude A, Nusser Z, Roberts JD, Mulvihill E, McIlhinney RA, Somogyi P (1993) The
metabotropic glutamate receptor (mGluR1 alpha) is concentrated at perisynaptic
membrane of neuronal subpopulations as detected by immunogold reaction. Neuron
11:771-787.
Berndt A, Schoenenberger P, Mattis J, Tye KM, Deisseroth K, Hegemann P, Oertner TG (2011)
High-efficiency channelrhodopsins for fast neuronal stimulation at low light levels. Proc
Natl Acad Sci U S A 108:7595-7600.
82
Berry SD, Thompson RF (1978) Prediction of learning rate from the hippocampal
electroencephalogram. Science 200:1298-1300.
Bezaire MJ, Soltesz I (2013) Quantitative assessment of CA1 local circuits: knowledge base for
interneuron-pyramidal cell connectivity. Hippocampus 23:751-785.
Blasco-Ibáñez JM, Freund TF (1995) Synaptic input of horizontal interneurons in stratum oriens
of the hippocampal CA1 subfield: structural basis of feed-back activation. Eur J Neurosci
7:2170-2180.
Booker SA, Gross A, Althof D, Shigemoto R, Bettler B, Frotscher M, Hearing M, Wickman K,
Watanabe M, Kulik Á, Vida I (2013) Differential GABAB-receptor-mediated effects in
perisomatic- and dendrite-targeting parvalbumin interneurons. J Neurosci 33:7961-7974.
Brun VH, Otnass MK, Molden S, Steffenach HA, Witter MP, Moser MB, Moser EI (2002) Place
cells and place recognition maintained by direct entorhinal-hippocampal circuitry.
Science 296:2243-2246.
Buzsáki G (2002) Theta oscillations in the hippocampus. Neuron 33:325-340.
Buzsáki G, Leung LW, Vanderwolf CH (1983) Cellular bases of hippocampal EEG in the
behaving rat. Brain Res 287:139-171.
Cajal SR (1911) Histology of the Nervous System. New York, NY: Oxford University Press.
Canteras NS, Swanson LW (1992) Projections of the ventral subiculum to the amygdala, septum,
and hypothalamus: a PHAL anterograde tract-tracing study in the rat. J Comp Neurol
324:180-194.
Chevaleyre V, Siegelbaum SA (2010) Strong CA2 pyramidal neuron synapses define a powerful
disynaptic cortico-hippocampal loop. Neuron 66:560-572.
83
Chittajallu R, Craig MT, McFarland A, Yuan X, Gerfen S, Tricoire L, Erkkila B, Barron SC,
Lopez CM, Liang BJ, Jeffries BW, Pelkey KA, McBain CJ (2013) Dual origins of
functionally distinct O-LM interneurons revealed by differential 5-HT(3A)R expression.
Nat Neurosci 16:1598-1607.
Cobb SR, Buhl EH, Halasy K, Paulsen O, Somogyi P (1995) Synchronization of neuronal
activity in hippocampus by individual GABAergic interneurons. Nature 378:75-78.
Colgin LL (2013) Mechanisms and functions of theta rhythms. Annu Rev Neurosci 36:295-312.
Colgin LL, Denninger T, Fyhn M, Hafting T, Bonnevie T, Jensen O, Moser MB, Moser EI
(2009) Frequency of gamma oscillations routes flow of information in the hippocampus.
Nature 462:353-357.
de la Prida LM, Totterdell S, Gigg J, Miles R (2006) The subiculum comes of age. Hippocampus
16:916-923.
Degro CE, Kulik A, Booker SA, Vida I (2015) Compartmental distribution of GABAB receptormediated currents along the somatodendritic axis of hippocampal principal cells. Front
Synaptic Neurosci 7:6.
Freund TF, Buzsáki G (1996) Interneurons of the hippocampus. Hippocampus 6:347-470.
Fyhn M, Molden S, Witter MP, Moser EI, Moser MB (2004) Spatial representation in the
entorhinal cortex. Science 305:1258-1264.
Galani R, Jarrard LE, Will BE, Kelche C (1997) Effects of postoperative housing conditions on
functional recovery in rats with lesions of the hippocampus, subiculum, or entorhinal
cortex. Neurobiol Learn Mem 67:43-56.
Gigg J, Finch DM, O'Mara SM (2000) Responses of rat subicular neurons to convergent
stimulation of lateral entorhinal cortex and CA1 in vivo. Brain Res 884:35-50.
84
Gillies MJ, Traub RD, LeBeau FE, Davies CH, Gloveli T, Buhl EH, Whittington MA (2002) A
model of atropine-resistant theta oscillations in rat hippocampal area CA1. J Physiol
543:779-793.
Givens BS, Olton DS (1990) Cholinergic and GABAergic modulation of medial septal area:
effect on working memory. Behav Neurosci 104:849-855.
Gloveli T, Dugladze T, Saha S, Monyer H, Heinemann U, Traub RD, Whittington MA, Buhl EH
(2005) Differential involvement of oriens/pyramidale interneurones in hippocampal
network oscillations in vitro. J Physiol 562:131-147.
Goutagny R, Gu N, Cavanagh C, Jackson J, Chabot JG, Quirion R, Krantic S, Williams S (2013)
Alterations in hippocampal network oscillations and theta-gamma coupling arise before
Aβ overproduction in a mouse model of Alzheimer's disease. Eur J Neurosci 37:18961902.
Goutagny R, Jackson J, Williams S (2009) Self-generated theta oscillations in the hippocampus.
Nat Neurosci 12:1491-1493.
Green JD, Arduini AA (1954) Hippocampal electrical activity in arousal. J Neurophysiol 17:533557.
Greene JR, Mason A (1996) Neuronal diversity in the subiculum: correlations with the effects of
somatostatin on intrinsic properties and on GABA-mediated IPSPs in vitro. J
Neurophysiol 76:1657-1666.
Griffin AL, Asaka Y, Darling RD, Berry SD (2004) Theta-contingent trial presentation
accelerates learning rate and enhances hippocampal plasticity during trace eyeblink
conditioning. Behav Neurosci 118:403-411.
85
Gulyás AI, Miles R, Sík A, Tóth K, Tamamaki N, Freund TF (1993) Hippocampal pyramidal
cells excite inhibitory neurons through a single release site. Nature 366:683-687.
Gupta AS, van der Meer MA, Touretzky DS, Redish AD (2012) Segmentation of spatial
experience by hippocampal θ sequences. Nat Neurosci 15:1032-1039.
Hafting T, Fyhn M, Molden S, Moser MB, Moser EI (2005) Microstructure of a spatial map in
the entorhinal cortex. Nature 436:801-806.
Harris E, Witter MP, Weinstein G, Stewart M (2001) Intrinsic connectivity of the rat subiculum:
I. Dendritic morphology and patterns of axonal arborization by pyramidal neurons. J
Comp Neurol 435:490-505.
Hasselmo ME (2005) What is the function of hippocampal theta rhythm?--Linking behavioral
data to phasic properties of field potential and unit recording data. Hippocampus 15:936949.
Hasselmo ME, Bodelón C, Wyble BP (2002) A proposed function for hippocampal theta rhythm:
separate phases of encoding and retrieval enhance reversal of prior learning. Neural
Comput 14:793-817.
Hasselmo ME, Stern CE (2013) Theta rhythm and the encoding and retrieval of space and time.
Neuroimage 85 (Pt 2):656-666.
Ishii K, Wong JK, Sumikawa K (2005) Comparison of alpha2 nicotinic acetylcholine receptor
subunit mRNA expression in the central nervous system of rats and mice. J Comp Neurol
493:241-260.
Jackson J, Amilhon B, Goutagny R, Bott JB, Manseau F, Kortleven C, Bressler SL, Williams S
(2014) Reversal of theta rhythm flow through intact hippocampal circuits. Nat Neurosci
17:1362-1370.
86
Jarsky T, Mady R, Kennedy B, Spruston N (2008) Distribution of bursting neurons in the CA1
region and the subiculum of the rat hippocampus. J Comp Neurol 506:535-547.
Jezek K, Henriksen EJ, Treves A, Moser EI, Moser MB (2011) Theta-paced flickering between
place-cell maps in the hippocampus. Nature 478:246-249.
Jutras MJ, Fries P, Buffalo EA (2013) Oscillatory activity in the monkey hippocampus during
visual exploration and memory formation. Proc Natl Acad Sci U S A 110:13144-13149.
Kamondi A, Acsády L, Wang XJ, Buzsáki G (1998) Theta oscillations in somata and dendrites
of hippocampal pyramidal cells in vivo: activity-dependent phase-precession of action
potentials. Hippocampus 8:244-261.
Katona I, Acsády L, Freund TF (1999) Postsynaptic targets of somatostatin-immunoreactive
interneurons in the rat hippocampus. Neuroscience 88:37-55.
Katona L, Lapray D, Viney TJ, Oulhaj A, Borhegyi Z, Micklem BR, Klausberger T, Somogyi P
(2014) Sleep and movement differentiates actions of two types of somatostatinexpressing GABAergic interneuron in rat hippocampus. Neuron 82:872-886.
Kim S (2014) Action potential modulation in CA1 pyramidal neuron axons facilitates OLM
interneuron activation in recurrent inhibitory microcircuits of rat hippocampus. PloS one
9:e113124.
Kim Y, Spruston N (2012) Target-specific output patterns are predicted by the distribution of
regular-spiking and bursting pyramidal neurons in the subiculum. Hippocampus 22:693706.
Kispersky TJ, Fernandez FR, Economo MN, White JA (2012) Spike resonance properties in
hippocampal O-LM cells are dependent on refractory dynamics. J Neurosci 32:36373651.
87
Klapoetke NC, Murata Y, Kim SS, Pulver SR, Birdsey-Benson A, Cho YK, Morimoto TK,
Chuong AS, Carpenter EJ, Tian Z, Wang J, Xie Y, Yan Z, Zhang Y, Chow BY, Surek B,
Melkonian M, Jayaraman V, Constantine-Paton M, Wong GK, Boyden ES (2014)
Independent optical excitation of distinct neural populations. Nat Methods 11:338-346.
Klausberger T, Magill PJ, Márton LF, Roberts JD, Cobden PM, Buzsáki G, Somogyi P (2003)
Brain-state- and cell-type-specific firing of hippocampal interneurons in vivo. Nature
421:844-848.
Kluge C, Stoppel C, Szinyei C, Stork O, Pape HC (2008) Role of the somatostatin system in
contextual fear memory and hippocampal synaptic plasticity. Learn Mem 15:252-260.
Knowles WD, Schwartzkroin PA (1981) Axonal ramifications of hippocampal Ca1 pyramidal
cells. J Neurosci 1:1236-1241.
Köhler C (1985) Intrinsic projections of the retrohippocampal region in the rat brain. I. The
subicular complex. J Comp Neurol 236:504-522.
Köhler C (1990) Subicular projections to the hypothalamus and brainstem: some novel aspects
revealed in the rat by the anterograde Phaseolus vulgaris leukoagglutinin (PHA-L)
tracing method. Prog Brain Res 83:59-69.
Lacaille JC, Mueller AL, Kunkel DD, Schwartzkroin PA (1987) Local circuit interactions
between oriens/alveus interneurons and CA1 pyramidal cells in hippocampal slices:
electrophysiology and morphology. J Neurosci 7:1979-1993.
Laxmi TR, Bindu PN, Raju TR, Meti BL (1999) Spatial memory impairment in ventral subicular
lesioned rats. Brain Res 816:245-248.
88
Leão RN, Mikulovic S, Leão KE, Munguba H, Gezelius H, Enjin A, Patra K, Eriksson A, Loew
LM, Tort AB, Kullander K (2012) OLM interneurons differentially modulate CA3 and
entorhinal inputs to hippocampal CA1 neurons. Nat Neurosci 15:1524-1530.
Lega BC, Jacobs J, Kahana M (2012) Human hippocampal theta oscillations and the formation
of episodic memories. Hippocampus 22:748-761.
Lorente de Nò R (1934) Studies of the structure of the cerebral cortex. II. Continuation of the
study of the ammonic system. J Psychol Neurol 113-177.
Losonczy A, Zhang L, Shigemoto R, Somogyi P, Nusser Z (2002) Cell type dependence and
variability in the short-term plasticity of EPSCs in identified mouse hippocampal
interneurones. J Physiol 542:193-210.
Lovett-Barron M, Kaifosh P, Kheirbek MA, Danielson N, Zaremba JD, Reardon TR, Turi GF,
Hen R, Zemelman BV, Losonczy A (2014) Dendritic inhibition in the hippocampus
supports fear learning. Science 343:857-863.
Lovett-Barron M, Turi GF, Kaifosh P, Lee PH, Bolze F, Sun XH, Nicoud JF, Zemelman BV,
Sternson SM, Losonczy A (2012) Regulation of neuronal input transformations by
tunable dendritic inhibition. Nat Neurosci 15:423-430.
Maccaferri G, McBain CJ (1995) Passive propagation of LTD to stratum oriens-alveus inhibitory
neurons modulates the temporoammonic input to the hippocampal CA1 region. Neuron
15:137-145.
Maccaferri G, McBain CJ (1996a) The hyperpolarization-activated current (Ih) and its
contribution to pacemaker activity in rat CA1 hippocampal stratum oriens-alveus
interneurones. J Physiol 497 (Pt 1):119-130.
89
Maccaferri G, McBain CJ (1996b) Long-term potentiation in distinct subtypes of hippocampal
nonpyramidal neurons. J Neurosci 16:5334-5343.
Maccaferri G, Roberts JD, Szucs P, Cottingham CA, Somogyi P (2000) Cell surface domain
specific postsynaptic currents evoked by identified GABAergic neurones in rat
hippocampus in vitro. J Physiol 524 (Pt 1):91-116.
McBain CJ, DiChiara TJ, Kauer JA (1994) Activation of metabotropic glutamate receptors
differentially affects two classes of hippocampal interneurons and potentiates excitatory
synaptic transmission. J Neurosci 14:4433-4445.
McBain CJ, Fisahn A (2001) Interneurons unbound. Nat Rev Neurosci 2:11-23.
McQuiston AR, Madison DV (1999) Nicotinic receptor activation excites distinct subtypes of
interneurons in the rat hippocampus. J Neurosci 19:2887-2896.
Minneci F, Janahmadi M, Migliore M, Dragicevic N, Avossa D, Cherubini E (2007) Signaling
properties of stratum oriens interneurons in the hippocampus of transgenic mice
expressing EGFP in a subset of somatostatin-containing cells. Hippocampus 17:538-553.
Mizuseki K, Sirota A, Pastalkova E, Buzsáki G (2009) Theta oscillations provide temporal
windows for local circuit computation in the entorhinal-hippocampal loop. Neuron
64:267-280.
Morrison JH, Benoit R, Magistretti PJ, Ling N, Bloom FE (1982) Immunohistochemical
distribution of pro-somatostatin-related peptides in hippocampus. Neurosci Lett 34:137142.
Moser EI, Kropff E, Moser MB (2008) Place cells, grid cells, and the brain's spatial
representation system. Annu Rev Neurosci 31:69-89.
90
Müller C, Remy S (2014) Dendritic inhibition mediated by O-LM and bistratified interneurons in
the hippocampus. Front Synaptic Neurosci 6:23.
Naber PA, Lopes da Silva FH, Witter MP (2001) Reciprocal connections between the entorhinal
cortex and hippocampal fields CA1 and the subiculum are in register with the projections
from CA1 to the subiculum. Hippocampus 11:99-104.
Nakauchi S, Brennan RJ, Boulter J, Sumikawa K (2007) Nicotine gates long-term potentiation in
the hippocampal CA1 region via the activation of alpha2* nicotinic ACh receptors. Eur J
Neurosci 25:2666-2681.
Nathanson JL, Yanagawa Y, Obata K, Callaway EM (2009) Preferential labeling of inhibitory
and excitatory cortical neurons by endogenous tropism of adeno-associated virus and
lentivirus vectors. Neuroscience 161:441-450.
Naus CC, Morrison JH, Bloom FE (1988) Development of somatostatin-containing neurons and
fibers in the rat hippocampus. Brain Res 468:113-121.
Nurse S, Lacaille JC (1997) Do GABAA and GABAB inhibitory postsynaptic responses
originate from distinct interneurons in the hippocampus? Can J Physiol Pharmacol
75:520-525.
O'Keefe J (1993) Hippocampus, theta, and spatial memory. Curr Opin Neurobiol 3:917-924.
O'Keefe J, Recce ML (1993) Phase relationship between hippocampal place units and the EEG
theta rhythm. Hippocampus 3:317-330.
O'Mara S (2005) The subiculum: what it does, what it might do, and what neuroanatomy has yet
to tell us. J Anat 207:271-282.
O'Mara SM, Commins S, Anderson M, Gigg J (2001) The subiculum: a review of form,
physiology and function. Prog Neurobiol 64:129-155.
91
O'Mara SM, Sanchez-Vives MV, Brotons-Mas JR, O'Hare E (2009) Roles for the subiculum in
spatial information processing, memory, motivation and the temporal control of
behaviour. Prog Neuropsychopharmacol Biol Psychiatry 33:782-790.
Olsen R, DeLorey T (1999) GABA Receptor Physiology and Pharmacology In: Basic
Neurochemistry: Molecular, Cellular and Medical Aspects (GJ, S. et al., eds)
Philadelphia: Lippincott-Raven.
Oswald CJ, Good M (2000) The effects of combined lesions of the subicular complex and the
entorhinal cortex on two forms of spatial navigation in the water maze. Behav Neurosci
114:211-217.
Petsche H, Stumpf C, Gogolak G (1962) The significance of the rabbit's septum as a relay station
between the midbrain and the hippocampus. I. The control of hippocampus arousal
activity by the septum cells. Electroencephalogr Clin Neurophysiol 14:202-211.
Pike FG, Goddard RS, Suckling JM, Ganter P, Kasthuri N, Paulsen O (2000) Distinct frequency
preferences of different types of rat hippocampal neurones in response to oscillatory input
currents. J Physiol 529 Pt 1:205-213.
Remondes M, Schuman EM (2004) Role for a cortical input to hippocampal area CA1 in the
consolidation of a long-term memory. Nature 431:699-703.
Rowland DC, Weible AP, Wickersham IR, Wu H, Mayford M, Witter MP, Kentros CG (2013)
Transgenically targeted rabies virus demonstrates a major monosynaptic projection from
hippocampal area CA2 to medial entorhinal layer II neurons. J Neurosci 33:14889-14898.
Royer S, Zemelman BV, Losonczy A, Kim J, Chance F, Magee JC, Buzsáki G (2012) Control of
timing, rate and bursts of hippocampal place cells by dendritic and somatic inhibition.
Nat Neurosci 15:769-775.
92
Scanziani M (2000) GABA spillover activates postsynaptic GABA(B) receptors to control
rhythmic hippocampal activity. Neuron 25:673-681.
Scher AI, Xu Y, Korf ES, Hartley SW, Witter MP, Scheltens P, White LR, Thompson PM, Toga
AW, Valentino DJ, Launer LJ (2011) Hippocampal morphometry in population-based
incident Alzheimer's disease and vascular dementia: the HAAS. J Neurol Neurosurg
Psychiatry 82:373-376.
Seager MA, Johnson LD, Chabot ES, Asaka Y, Berry SD (2002) Oscillatory brain states and
learning: Impact of hippocampal theta-contingent training. Proc Natl Acad Sci U S A
99:1616-1620.
Sharp PE, Green C (1994) Spatial correlates of firing patterns of single cells in the subiculum of
the freely moving rat. J Neurosci 14:2339-2356.
Sik A, Penttonen M, Ylinen A, Buzsáki G (1995) Hippocampal CA1 interneurons: an in vivo
intracellular labeling study. J Neurosci 15:6651-6665.
Skaggs WE, McNaughton BL, Wilson MA, Barnes CA (1996) Theta phase precession in
hippocampal neuronal populations and the compression of temporal sequences.
Hippocampus 6:149-172.
Sodickson DL, Bean BP (1996) GABAB receptor-activated inwardly rectifying potassium
current in dissociated hippocampal CA3 neurons. J Neurosci 16:6374-6385.
Solstad T, Moser EI, Einevoll GT (2006) From grid cells to place cells: a mathematical model.
Hippocampus 16:1026-1031.
Somogyi P, Hodgson AJ, Smith AD, Nunzi MG, Gorio A, Wu JY (1984) Different populations
of GABAergic neurons in the visual cortex and hippocampus of cat contain somatostatinor cholecystokinin-immunoreactive material. J Neurosci 4:2590-2603.
93
Staff NP, Jung HY, Thiagarajan T, Yao M, Spruston N (2000) Resting and active properties of
pyramidal neurons in subiculum and CA1 of rat hippocampus. J Neurophysiol 84:23982408.
Stanley EM, Fadel JR, Mott DD (2012) Interneuron loss reduces dendritic inhibition and GABA
release in hippocampus of aged rats. Neurobiol Aging 33:431.e431-413.
Steffenach HA, Witter M, Moser MB, Moser EI (2005) Spatial memory in the rat requires the
dorsolateral band of the entorhinal cortex. Neuron 45:301-313.
Stewart M, Fox SE (1991) Hippocampal theta activity in monkeys. Brain Res 538:59-63.
Stewart M, Wong RK (1993) Intrinsic properties and evoked responses of guinea pig subicular
neurons in vitro. J Neurophysiol 70:232-245.
Sun Y, Nguyen AQ, Nguyen JP, Le L, Saur D, Choi J, Callaway EM, Xu X (2014) Cell-typespecific circuit connectivity of hippocampal CA1 revealed through Cre-dependent rabies
tracing. Cell Rep 7:269-280.
Swanson LW, Cowan WM (1977) An autoradiographic study of the organization of the efferent
connections of the hippocampal formation in the rat. J Comp Neurol 172:49-84.
Tamamaki N, Abe K, Nojyo Y (1987) Columnar organization in the subiculum formed by axon
branches originating from single CA1 pyramidal neurons in the rat hippocampus. Brain
Res 412:156-160.
Taube JS (1993) Electrophysiological properties of neurons in the rat subiculum in vitro. Exp
Brain Res 96:304-318.
Tesche CD (1997) Non-invasive detection of ongoing neuronal population activity in normal
human hippocampus. Brain Res 749:53-60.
94
Tesche CD, Karhu J (2000) Theta oscillations index human hippocampal activation during a
working memory task. Proc Natl Acad Sci U S A 97:919-924.
Trujillo-Estrada L, Dávila JC, Sánchez-Mejias E, Sánchez-Varo R, Gomez-Arboledas A, Vizuete
M, Vitorica J, Gutiérrez A (2014) Early neuronal loss and axonal/presynaptic damage is
associated with accelerated amyloid-β accumulation in AβPP/PS1 Alzheimer's disease
mice subiculum. J Alzheimers Dis 42:521-541.
van Groen T, Miettinen P, Kadish I (2003) The entorhinal cortex of the mouse: organization of
the projection to the hippocampal formation. Hippocampus 13:133-149.
Vanderwolf CH (1969) Hippocampal electrical activity and voluntary movement in the rat.
Electroencephalogr Clin Neurophysiol 26:407-418.
Varga C, Golshani P, Soltesz I (2012) Frequency-invariant temporal ordering of interneuronal
discharges during hippocampal oscillations in awake mice. Proc Natl Acad Sci U S A
109:E2726-2734.
Widmer H, Ferrigan L, Davies CH, Cobb SR (2006) Evoked slow muscarinic acetylcholinergic
synaptic potentials in rat hippocampal interneurons. Hippocampus 16:617-628.
Winson J (1978) Loss of hippocampal theta rhythm results in spatial memory deficit in the rat.
Science 201:160-163.
Witter MP, Naber PA, van Haeften T, Machielsen WC, Rombouts SA, Barkhof F, Scheltens P,
Lopes da Silva FH (2000) Cortico-hippocampal communication by way of parallel
parahippocampal-subicular pathways. Hippocampus 10:398-410.
Witter MP, Ostendorf RH, Groenewegen HJ (1990) Heterogeneity in the Dorsal Subiculum of
the Rat. Distinct Neuronal Zones Project to Different Cortical and Subcortical Targets.
Eur J Neurosci 2:718-725.
95
Wouterlood FG, Saldana E, Witter MP (1990) Projection from the nucleus reuniens thalami to
the hippocampal region: light and electron microscopic tracing study in the rat with the
anterograde tracer Phaseolus vulgaris-leucoagglutinin. J Comp Neurol 296:179-203.
Yamano M, Luiten PG (1989) Direct synaptic contacts of medial septal efferents with
somatostatin immunoreactive neurons in the rat hippocampus. Brain research bulletin
22:993-1001.
Yizhar O, Fenno LE, Prigge M, Schneider F, Davidson TJ, O'Shea DJ, Sohal VS, Goshen I,
Finkelstein J, Paz JT, Stehfest K, Fudim R, Ramakrishnan C, Huguenard JR, Hegemann
P, Deisseroth K (2011) Neocortical excitation/inhibition balance in information
processing and social dysfunction. Nature 477:171-178.
Ylinen A, Soltész I, Bragin A, Penttonen M, Sik A, Buzsáki G (1995) Intracellular correlates of
hippocampal theta rhythm in identified pyramidal cells, granule cells, and basket cells.
Hippocampus 5:78-90.
96