ALTERED METABOTROPIC GLUTAMATE RECEPTOR FUNCTION IN THE
NEOCORTEX OF A FRAGILE X MOUSE
By
Jacqueline Ann Fenn

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Neuroscience-Doctor of Philosophy
2017

	

ABSTRACT
ALTERED METABOTROPIC GLUTAMATE RECEPTOR FUNCTION IN THE
NEOCORTEX OF A FRAGILE X MOUSE
By
Jacqueline Ann Fenn
Fragile X Syndrome (FXS) is the leading cause of inherited intellectual disability. It is
characterized by a wide array of symptoms, including cognitive impairments, attention deficit
and hyperactivity disorder (ADHD), hypersensitivity to sensory stimuli, autistic features, mood
lability, and seizures. Since the creation of the Fmr1 KO mouse more than 20 years ago, a
wealth of studies have uncovered a role for group I metabotropic glutamate receptors (mGluRs)
in mediating many FXS phenotypes, leading to development of “The mGluR Theory of Fragile
X”. However, studies supporting this theory have focused on impairments in the hippocampus,
amygdala, and other structures of the allocortex. The isocortex remains largely uninvestigated,
despite its major role in sensory integration, attentional processes, and executive function.
MgluRs are also highly expressed in the neocortex, where they can modulate neuronal
excitability and synaptic transmission. Using electrophysiological methods, I investigated the
role of group I and group II mGluRs in modulating neocortical circuits in primary visual cortex.
Humans with FXS show severe visual-motor deficits, and perform poorly on global motion
tasks. Autopsy studies also reveal abnormal dendritic spine morphologies in layer 5/6 of the
visual neocortex, and this is substantiated in the Fmr1 KO mouse. Using whole cell patch clamp
recordings of different neuron subtypes in layer 5/6 of primary visual cortex, I found that Fmr1
KO layer 5/6 somatostatin expressing neurons (SST+) have a significant decrease in input
resistance compared to wild type (WT), indicating that they are less intrinsically excitable than
WT SST+ neurons. Further, I show that activation of group II mGluRs leads to disinhibition of

	

excitatory pyramidal neurons. I also discover that suppression of GABAergic transmission by
group II mGluRs is normal in the Fmr1 KO, leading to my hypothesis that increased
disinhibition of pyramidal neurons by group II mGluRs is due to their altered modulation of fast
glutamatergic transmission onto layer 5/6 interneurons in the Fmr1 KO mouse.
Using recordings from fluorescently labeled interneurons in layer 5/6 visual neocortex, I
discover that group II mGluR-mediated suppression of fast excitatory glutamatergic transmission
onto inhibitory neurons is exaggerated in the Fmr1 KO mouse. I show that this effect is cell
specific, as it only occurs in SST+ interneurons and not parvalbumin expressing interneurons.
Finally, I show that this deficit in excitatory drive onto SST+ interneurons is mediated by
presynaptic mGluRs, and that these defects are specific to layer 5/6 visual neocortex. Both the
decreased excitability of SST+ interneurons and exaggerated group II-mGluR mediated
suppression of excitatory drive onto these cells would lead to neocortical circuit hyperexcitability
in the Fmr1 KO mouse. A hyperexcitable neocortical circuit would be anticipated to negatively
impact sensory integration, a requirement for both stimulus encoding and attentional processes.
Lastly, a hyperexcitable neocortical network could give rise to epileptiform activity.
The results of these studies are fascinating, as group II mGluRs have not previously been
implicated in studies of the neocortical pathogenesis in the Fmr1 KO mouse. This investigation
also showed that group I mGluR-mediated modulation of membrane excitability and fast
synaptic transmission is unaltered in the Fmr1 KO mouse, suggesting that “The mGluR theory of
Fragile X” may in fact be more or less valid contingent on the brain region under investigation.
It is my hope that these circuit studies will inform scientific investigations on autism spectrum
disorders and epilepsy syndromes, as both show high comorbidity in individuals with FXS.

	

Copyright by
JACQUELINE ANN FENN
2017

	

This work is dedicated to my mother, who continues to work relentlessly to help me achieve
my goals. Thank you for your endless love and for pushing me to never give up.

	

v	

ACKNOWLEDGEMENTS

I would first like to recognize my advisor, Dr. Charles “Lee” Cox, for allowing me to
complete my degree in his lab and for his financial support throughout my graduate career. Lee
has taught me to troubleshoot, think independently, and to never accept published science as fact
without first asking questions. He has instilled in me that conducting research means being
wrong, and that’s okay. Still, he’s taught me to treasure the moments when my data confirms my
hypothesis and adds to the body of scientific knowledge. This experience has helped me
appreciate the endless complexities of the brain. I am continuously perplexed and amazed by the
nuances of neural circuits. I owe this to my advisor, who remains as impassioned by
electrophysiology today as I can presume he was during his earlier career. I would like to thank
him for his patience whilst teaching me to be a scientist.
I would also like to thank Dr. Joseph Beatty, who is an extremely qualified
electrophysiologist that seldom says no when asked for help. Since his arrival in the lab, my
research technique has exponentially improved and I attribute this to the countless number of
times he has assisted me in troubleshooting my equipment and reviewing fundamentals. I deeply
value the input he has given me regarding my research technique and theory.
I want to thank all of the members of my research committee, Dr. Hongbing Wang, Dr.
James Galligan, and Dr. Peter Cobbett, who have advised me since my first committee meeting
at Michigan State University. Not only have they been generous with their time, but they have
continued to sharpen my scientific acumen in our meetings. I greatly appreciate their
perspectives on my research, as it has helped bring to my attention both the gaps in my
knowledge and my data so that I may continue to develop my scientific competency.

	

vi	

I would like to thank my colleagues in Cox Lab, past and present, who have all helped
with my research. I would like to thank Dr. Paul, who helped me navigate my initial patchclamp learning curve and also taught me how to do histology post hoc. I would be remiss not to
thank Dawn Autio. She maintains all the mouse colonies, performs histology, maintains lab
materials, and still manages to conduct her own research. She is a phenomenal technician and
my mom away from home. Finally, Kathleen Louis, who became one of my best friends during
our shared lab experiences. I thank her for commiserating with me during this journey.
I would like to thank the Neuroscience Program for their continued support of me during
my educational endeavors. The staff is remarkably knowledgeable about the logistical
components of grad school, and have always been readily available to me when I needed help.
Lastly, I would like to thank my family and friends who have consistently helped me
throughout my journey. I want to thank Rachael White, who finds the time to call me every
week. She is the closest thing I have ever had to a sister, and for her I am grateful. I would like
to thank Dr. James Anderson for being my teacher and spiritual advisor since high school. I
would like to thank Dr. Erica Wehrwein for being an astonishing and strong female role model
for me in academia and life. I would also like to thank my mother Sylvia Maldonado. I am not
me without her, and there are literally no words to describe her importance to me. I would like
to thank Jeremy Otero, who would travel day or night to do whatever I asked of him without
complaint. Since he came into my life I have never lacked the presence of a father figure. I also
want to thank Jimmy, who has loved me during some of the most trying times of my degree
program. He has never stopped believing in me or pushing me to work. Most importantly, I
would like to thank God for the hands, head, and heart he has given me. Hands to collect the
data, a head to understand it, and a heart to keep the faith!

	

vii	

TABLE OF CONTENTS
LIST OF TABLES.............................................................................................................xi
LIST OF FIGURES..........................................................................................................xii
KEY TO ABBREVIATIONS..........................................................................................xiv
CHAPTER I: INTRODUCTION .......................................................................................1
1.1 Organization of the Visual System .............................................................................2
A. Feed-forward visual neocortical circuitry ..........................................................6
B. Feed-back visual neocortical circuitry ................................................................8
C. Murine model of the visual system ......................................................................9
D. Interneuron contributions to local microcircuits ..............................................11
D1. Fast-spiking interneurons ...................................................................11
D2. Non-fast-spiking interneurons.............................................................13
D3. 5HT3a receptor-expressing interneurons ...........................................14
1.2 Metabotropic Glutamate Receptors .........................................................................16
A. General features of metabotropic glutamate receptors.....................................16
B. Group I mGluRs ................................................................................................17
C. Group II mGluRs ...............................................................................................19
D. Group III mGluRs .............................................................................................21
E. Distribution and functional role of mGluRs in the visual neocortex.................22
1.3 Fragile X Syndrome ...................................................................................................24
A. Physical and behavioral phenotype...................................................................25
B. Neuroanatomical findings in FXS individuals ..................................................28
C. FMRP function and sub-cellular localization ...................................................29
D. The Fmr1 KO as an animal model of FXS ........................................................31
E. Altered glutamate signaling and “The mGluR Theory” ...................................34
F. GABAergic dysfunction in the Fmr1 KO ...........................................................38
G. Neocortical pathogenesis in FXS ......................................................................42
G1. Anatomical studies ..............................................................................43
G2. Intrinsic excitability of neocortical neurons .......................................44
G3. Synaptic plasticity defects and other network alterations ..................45
G4. Developmental changes in the neocortex of the Fmr1 KO mouse ......48
1.4 Concluding Remarks .................................................................................................50
APPENDIX .......................................................................................................................53
REFERENCES.................................................................................................................59
CHAPTER II: MGLUR-MEDIATED MODULATION OF VISUAL NEOCORTICAL
PYRAMIDAL CELL EXCITATION IN A FRAGILE X MOUSE..................................78
2.1 Introduction ................................................................................................................78
2.2 Materials and Methods ..............................................................................................80
A. Membrane Potential Drug Application Protocol ..............................................81
B. Stimulation Protocol..........................................................................................82

	

viii	

2.3 Results .........................................................................................................................83
A. The intrinsic properties of Fmr1 KO pyramidal neurons are similar to wild
type in layer 5/6 of the visual neocortex ................................................................83
B. mGluR-mediated modulation of pyramidal neuron membrane potential is
intact in the Fmr1 KO mouse .................................................................................84
C. Group I mGluRs do not modulate excitatory synaptic transmission onto
visual neocortical pyramidal neurons in layer 5/6 ................................................86
D. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 5/6 visual neocortical pyramidal neurons in the WT
and Fmr1 KO mouse ..............................................................................................87
E. The intrinsic properties of Fmr1 KO pyramidal neurons are similar to wild
type in layer 2/3 of the visual neocortex ................................................................87
F. Group I mGluRs do not modulate excitatory synaptic transmission onto
visual neocortical pyramidal neurons in layer 2/3 ................................................88
G. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 2/3 visual neocortical pyramidal neurons in the WT
and Fmr1 KO mouse ..............................................................................................89
2.4 Discussion ...................................................................................................................89
APPENDIX .......................................................................................................................93
REFERENCES...............................................................................................................108
CHAPTER III: MGLUR-MEDIATED MODULATION OF VISUAL NEOCORTICAL
PYRAMIDAL CELL INHIBITION IN A FRAGILE X MOUSE ..................................112
3.1 Introduction ..............................................................................................................112
3.2 Materials and Methods ............................................................................................114
3.3 Results .......................................................................................................................116
A. Group II mGluR-mediated suppression of IPSCs on layer 5/6 pyramidal
neurons is exaggerated in the Fmr1 KO mouse...................................................116
B. In the absence of excitatory drive onto interneurons, group II mGluRmediated suppression of IPSCs is normal in the Fmr1 KO mouse ......................117
C. Suppression of IPSCs onto pyramidal neurons is due to either pre- or
post- synaptic group II mGluRs ...........................................................................118
3.4 Discussion .................................................................................................................120
APPENDIX .....................................................................................................................123
REFERENCES...............................................................................................................133
CHAPTER IV: FMR1 KO NON FAST-SPIKING INTERNEURONS EXHIBIT
EXAGGERATED RESPONSE TO GROUP II MGLUR MEDIATED SUPPRESSION OF
EXCITATORY TRANSMISSION .................................................................................137
4.1 Introduction ..............................................................................................................137
4.2 Materials and Methods ............................................................................................139
4.3 Results .......................................................................................................................142
A. The intrinsic properties of Fmr1 KO layer 5/6 FS interneurons are similar
to WT ....................................................................................................................142
B. Group I mGluRs do not modulate EPSCs in FS interneurons ........................143
C. Group II mGluRs modulate EPSCs on FS interneurons similarly between

	

ix	

WT and Fmr1 KO ................................................................................................144
D. Fmr1 KO layer 5/6 NFS interneurons have a decreased input resistance
compared to WT ........................................................................................................... 145
E. Fmr1 KO NFS interneurons show an exaggerated response to group II
mGluR activation .................................................................................................147
F. Group II mGluR-mediated suppression of excitatory transmission onto NFS
interneurons is mediated presynaptically ............................................................147
G. The intrinsic properties of layer 2/3 NFS interneurons are similar between
WT and Fmr1 KO mice ........................................................................................150
H. Exaggerated group II mGluR-mediated suppression of Fmr1 KO NFS
interneuron EPSCs is layer specific ....................................................................151
4.4 Discussion .................................................................................................................152
APPENDIX .....................................................................................................................157
REFERENCES...............................................................................................................177
CHAPTER V: GENERAL CONCLUSIONS ................................................................181
APPENDIX .....................................................................................................................190
REFERENCES...............................................................................................................193

	

x	

LIST OF TABLES
Table 1.1. Three groups of metabotropic glutamate receptors ...................................55
Table 2.1. Intrinsic properties of layer 5/6 pyramidal neurons in WT and Fmr1
KO mice ............................................................................................................................84
Table 2.2. Layer 5/6 pyramidal neuron membrane potential responses to
application of general mGluR agonist (ACPD) and specific mGluR I (DHPG) and
II (APDC) agonists ...........................................................................................................86
Table 2.3. Intrinsic properties of layer 2/3 pyramidal neurons in WT and Fmr1
KO mice ............................................................................................................................88
Table 4.1. Intrinsic properties of layer 5/6 FS interneurons in WT and Fmr1 KO
mice..................................................................................................................................143
Table 4.2. Intrinsic properties of layer 5/6 SST+ interneurons in WT and Fmr1
KO mice ..........................................................................................................................146
Table 4.3. Intrinsic properties of layer 2/3 SST+ interneurons in WT and Fmr1
KO mice ..........................................................................................................................151

	

xi	

LIST OF FIGURES
Figure 1.1. Primary visual neocortex circuitry and parallel visual pathways. ..........54
Figure 1.2. Development of FXS premutation and full mutation. ...............................56
Figure 1.3. The mGluR Theory of Fragile X Syndrome. .............................................57
Figure 2.1. Schematic diagram of stimulation protocol. ..............................................94
Figure 2.2. The intrinsic excitability of Fmr1 KO pyramidal neurons is unaltered
in layer 5/6 visual neocortex. ...........................................................................................95
Figure 2.3. mGluR-mediated modulation of pyramidal neuron membrane
potential is intact in layer 5/6 of the Fmr1 KO mouse. .................................................96
Figure 2.4. Group I mGluRs do not suppress excitatory synaptic transmission of
layer 5/6 pyramidal neurons. ..........................................................................................98
Figure 2.5. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 5/6 visual neocortical pyramidal neurons in the WT and
Fmr1 KO mouse. ............................................................................................................100
Figure 2.6. The intrinsic properties of Fmr1 KO layer 2/3 pyramidal neurons
are similar to wild type.. ................................................................................................102
Figure 2.7. DHPG does not alter EPSCs in layer 2/3 pyramidal neurons from WT
and Fmr1 KO mice. .......................................................................................................103
Figure 2.8. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 2/3 visual neocortical pyramidal neurons in the WT and
Fmr1 KO mouse. ............................................................................................................105
Figure 2.9. Group II mGluR-mediated suppression of EPSCs on pyramidal
neurons is normal in the Fmr1 KO mouse. .................................................................107
Figure 3.1. Group II mGluR-mediated modulation of IPSCs in layer 5/6
pyramidal neurons is exaggerated in the Fmr1 KO mouse. ......................................124
Figure 3.2. Group II mGluRs suppress fast inhibitory transmission onto WT and
Fmr1 KO pyramidal neurons in the absence of excitatory drive onto neocortical
interneurons....................................................................................................................126
Figure 3.3. Group II mGluRs modulate sIPSC amplitude and frequency similarly
between WT and Fmr1 KO layer 5/6 pyramidal neurons. ........................................128

	

xii	

Figure 3.4. Presynaptic mGluRs at the interneuron terminals modulate inhibition
similarly between WT and Fmr1 KO layer 5/6 pyramidal neurons. ........................130
Figure 3.5. Group II mGluR-mediated suppression of IPSCs on pyramidal
neurons is normal in the Fmr1 KO mouse. .................................................................132
Figure 4.1. Intrinsic excitability is decreased in layer 5/6 Fmr1 KO visual
neocortical NFS interneurons, but not FS interneurons. ...........................................158
Figure 4.2. Group I mGluRs do not modulate excitatory synaptic transmission
onto WT and Fmr1 KO layer 5/6 FS interneurons. ....................................................160
Figure 4.3. Group II mGluR-mediated modulation of excitatory transmission
onto FS interneurons is unaltered in the visual neocortex of the Fmr1 KO
mouse...............................................................................................................................162
Figure 4.4. Group II mGluR-mediated suppression of excitatory transmission onto
NFS interneurons is exaggerated in the visual neocortex of the Fmr1 KO
mouse...............................................................................................................................164
Figure 4.5. Group II mGluR activation does not alter miniature EPSC amplitude or
frequency in WT and Fmr1 KO layer 5/6 visual neocortical NFS interneurons. ....166
Figure 4.6. Group II mGluR-mediated suppression of EPSCs in layer 5/6 NFS
interneurons leads to a change in paired pulse ratio. .................................................168
Figure 4.7. EPSCs evoked in strontium show a decrease in frequency in response
to application of group II mGluR activation in both WT and Fmr1 KO layer 5/6
NFS interneurons. ..........................................................................................................170
Figure 4.8. The intrinsic properties Fmr1 KO layer 2/3 NFS interneurons are
unaltered. ........................................................................................................................171
Figure 4.9. Altered group II mGluR-mediated suppression of NFS interneuron
EPSCs in the Fmr1 KO is layer specific.......................................................................172
Figure 4.10. Group II mGluR-mediated suppression of EPSCs on NFS
interneurons is increased in the Fmr1 KO mouse. .....................................................174
Figure 4.11. Activation of presynaptic group II mGluRs leads to exaggerated
suppression of excitatory drive onto NFS interneurons and disinhibition of
GABAergic signaling in the visual neocortex of the Fmr1 KO mouse. .....................176
Figure 5.1. Altered presynaptic processes in the Fmr1 KO mouse upon activation
of group II mGluRs........................................................................................................191

	

xiii	

KEY TO ABBREVIATIONS

	

ACPD

(1S,3R)-1-Aminocyclopentane-1,3-dicarboxylic acid

AHP

after-hyperpolarizing potential

AMPA

α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid

APDC

(2R,4R)-4-aminopyrrolidine-2,4-dicarboxylate

ASD

autism spectrum disorder

cAMP

cyconucleotide aminophosphate

CHR2

channelrhodopsin 2

DAP

depolarizing afterpotential

DHPG

(RS)-3,5-dihydroxyphenylglycine

DNQX

6,7-dinitroquinoxaline-2,3-dione

EPSC

excitatory postsynaptic current

ERK

extracellular signal regulated kinase

FMRP

fragile x mental retardation protein

FS

fast spiking

FXS

fragile x syndrome

GABA

γ-aminobutyric acid

GFP

green fluorescent protein

GPCR

g-protein coupled receptor

IB

intrinsic bursting

IPSC

inhibitory postsynaptic current

LGN

lateral geniculate nucleus

LTD

long term depression

xiv	

	

LTP

long term potentiation

MAP1B

microtubule associated protein 1B

MAPK

mitogen activated protein kinase

mGluR

metabotropic glutamate receptor

mEPSC

miniature excitatory postsynaptic current

mIPSC

minitature inhibitory postsynaptic current

mTOR

mammalian target of rapamycin

NFS

non-fast-spiking

NMDA

N-methyl-D-aspartate

pI3K

Phosphatidylinositol-4,5-bisphosphate 3-kinase

PPR

paired pulse ratio

PV+

parvalbumin

RGC

retinal ganglion cells

RS

regular spiking

sIPSC

spontaneous inhibitory postsynaptic current

SNARE

Soluble N-ethylmaleimide-sensitive factor activating
protein receptor

SST+

somatostatin positive

STEP

striatal-enriched protein tyrosine phosphotase

Tsc ½

tuberous sclerosis ½

V1

primary visual neocortex

WT

wild type

xv	

CHAPTER I: INTRODUCTION
In 1943, Martin and Bell, described a group of patients characterized by a common set of
features including intellectual disability and social withdrawal (Bhakar, Dolen, & Bear, 2012).
These individuals carried a fragile site on the X chromosome, giving rise to the name Fragile X
Syndrome (FXS). The features of FXS are both physical and neurological, but the mechanisms
underlying these abnormalities were completely speculative until the development of a murine
animal model that recapitulated several of the human phenotypes (Dutch Belgium Consortium,
1994). Since the development of the Fmr1 KO mouse, scientists have been fortunate enough to
uncover several cellular changes in the brains of these mice attributable to the loss of the fragile
X mental retardation protein (FMRP). Absent FMRP leads to defects in cell signaling pathways
and altered mRNA/protein expression of a variety of critical receptors, ion channels, and proteins
involved in different synaptic processes. Changes in synaptic events presumably lead to
anatomical and behavioral changes, which are also observed in Fmr1 KO mice. What are these
synaptic changes? This is a question that continues to plague the field, although remarkable
progress has been made. Several brain regions show changes in synaptic transmission and/or
plasticity, and in many instances this is instigated by altered signaling downstream of group I
metabotropic glutamate receptors (mGluRs). Armed with this knowledge, my dissertation
studies are aimed at investigating the role that mGluRs play in cellular excitability and synaptic
transmission in the primary visual neocortex, a region that has been documented to exhibit both
anatomical and electrophysiological alterations in the Fmr1 KO mouse. In the following pages, I
first provide an overview of the nonhuman primate model of the visual system. I spend a
disproportionate amount of time discussing the region under my investigation, the primary visual
cortex (V1). Additionally, I will then compare the visual system of the mouse to the nonhuman

	

1	

primate model. With this foundation of the visual system, I next provide an overview of the
structure and function of the three groups of mGluRs, one of which has been in the spotlight of
Fragile X research since the advent of the Fmr1 KO mouse and was a logical choice to
investigate in my dissertation research. Finally, I provide an overview of FXS in humans,
including how it occurs, what the clinical manifestations are, and what human studies have
revealed about the FXS brain. I then discuss the Fmr1 KO animal model, including what we
have learned regarding alterations in glutamatergic and GABAergic transmission. Finally, I
discuss findings from the limited functional studies that have been done in the neocortex of the
Fmr1 KO mouse. Armed with this introductory material, I can make broader speculations about
my research findings and how they fit into what we already know about the visual system and
the pathogenesis of FXS.
1.1 Organization of the Visual System
The visual system is highly structured. Similar to other sensory systems, it is functionally
organized such that neurons having similar response properties are located in close proximity.
However, this topographical arrangement becomes less defined as information moves from the
retina to the visual cortex and each successive level of information processing leads to increased
integration of visual stimuli. Your visual world can be divided into a left and right visual field.
Information from the left visual field is transferred to the right side of each retina, and
subsequently to the right hemisphere of your brain. The opposite is true for the right half of your
visual field, which is detected on the left side of each retina and transferred to the left hemisphere
of your brain. In this way, both hemispheres of the brain process visual information from both
eyes.

	

2	

Extraction of the features of your external environment begins at the level of the eye. Visual
information is integrated by retinal ganglion cells (RGCs). Smaller RGCs located predominantly
in the central region (fovea) of the retina have smaller receptive fields and are color sensitive,
making them good for detecting fine details of the visual environment as well as color contrast
(Callaway 1998). However, these cells are also slowly adapting and have axons with slower
conduction velocities, making it difficult for these neurons to detect rapid movement. For these
reasons, these RGCs give rise to a visual pathway that carries information specific to the “what”
details of the visual environment. This pathway, known as the parvocellular (P) pathway, carries
its information more ventrally through the brain and is otherwise known as the ventral or
temporal stream. In contrast, another subset of RGCs reside disproportionately in more
peripheral locations on the retina and possess some qualities that are diametrically opposed to
those of the P Pathway. These RGCs have larger receptive fields, are color insensitive, rapidly
adapting, and have larger diameter axons with faster conduction velocities. They are also
extremely sensitive to luminance contrast, making them optimally suited for detecting the form
and movement of objects. These RGCs give rise to the M pathway of the visual system. The M
pathway is also known as the dorsal or parietal stream because it carries visual information
dorsally from the retina to higher order visual areas in the parietal lobe. It is commonly known
as the “where” pathway because it carries information specific to the location of objects. A third
subset of RGCs, giving rise to the koniocellular (K) pathway, have the smallest diameter axons
of the three RGC subtypes and have been suggested to play a role in modulating the information
received through the M and P pathways (Casagrande 1994). The features of these different
RGCs are preserved in their respective recipient neurons throughout the visual system. For
example, the primary visual cortex (V1) has specific neurons that are recipients of P pathway

	

3	

information. These neurons are smaller in size and have smaller receptive fields, while V1
neurons that are recipients of M pathway information are typically larger and have larger
receptive fields.
Almost all RGC axons travel through the optic nerve to the visual thalamus, or lateral
geniculate nucleus (LGN). Once these fibers reach the LGN, they disperse to synapse onto six
different layers of the thalamus. The four most dorsal layers are termed the parvocellular layers
while the two most ventral layers are termed the magnocellular layers, and these layers receive
and deliver visual information specific to the P pathway and M pathway respectively. The K
pathway thalamic neurons are intercalated between the P and M layers of the LGN. Each layer
of the LGN responds to information from either the left or right eye, so at the level of the
thalamus we can say that cells exhibit ocular dominance. Axons from the six layers of the LGN
will travel together along the optic radiations to ascend to V1.
V1 is one of the most extensively studied and anatomically understood neocortical brain
regions. Also known as the striate cortex, V1 contains approximately 200 million neurons, and
has six layers that are intralaminarly subdivided in humans and other primates (Lund 1988,
Callaway 1998, Balaram and Kaas 2014). The first two layers have no subdivisions. Layers 3
and 5 are divided into sublayers 3A/3B and 5A/5B respectively. Layer 4 is divided into three
sublayers 4A, 4B, and 4C. Finally, layer 6 is traditionally referred to as a single layer (Lund
1988, Callaway, 1998), although some scientists prefer to divide it into sublayer 6A and 6B
based on cell density (Balaram and Kaas, 2014). This delineation, however, is less obvious due
to the heterogeneity of cell types and sizes. For this reason, I will refer to it as one layer. The
architecture of V1 is highly systematic, with its cells residing in what are known as ocular
dominance columns. These columns run vertically through the visual neocortex. If you imagine

	

4	

a recording electrode penetrating the neocortical surface perpendicularly such that it crosses
through all six layers, you would expect the cells in that column to respond strongly to visual
information from either the right or left eye. All of these cells would also respond most strongly
to a visual stimulus of one particular orientation. Thus visual neocortical cells also exhibit
orientation columns, with the exception of layer 4 where cells have no orientation preference.
Within a V1 layer, every 50 µM the orientation of a cells receptive field shifts 10 degrees in
either a counterclockwise or clockwise direction. Thus, intralaminar organization is also highly
structured. As previously mentioned, the journey of information processing through V1 begins
at Layer 4C, where the majority of axons leaving the LGN synapse will synapse. Generally
speaking, V1 cells that are closest to Layer 4C exhibit less complicated behavior then cells in the
superficial and deep neocortical layers. For this reason, the majority of Layer 4C cells are
known as “simple” cells. These cells are similar to LGN cells in that they exhibit ocular
dominance and have on-off center-surround receptive fields. However, some of these simple
cells will also respond more strongly to stimuli of a particular orientation and are thus said to be
orientation selective. Outside of Layer 4C, simple cells will synapse onto cells that exhibit more
complex receptive field properties. These cells, known as complex cells, comprise the majority
of V1 cells. Complex cells exhibit orientation selectivity like simple cells, but they lack
positional selectivity. This means that complex cells can respond to a properly oriented line no
matter where it resides in the receptive field. Thus, as the line moves complex cells have a
sustained response and are considered to be movement selective. Depending on the layer, these
complex cells can respond to additional characteristics of visual stimuli. For example, 10 to 20
percent of complex cells in layer 2/3 show direction selectivity (Hubel, 1988), preferring the
oriented stimulus to move in one direction over its opposing direction. Some complex cells also

	

5	

respond strongly to specific line lengths and curvatures. Now that I have covered some general
features of V1, I will provide a review of the feed-forward and feed-back circuitry of the visual
neocortex, as well as a discussion of the different neuron subtypes distributed throughout V1.
For a schematic of the V1 circuitry, please see the appendix at the end of this chapter (Figure
1.1).
A. Feed-forward visual neocortical circuitry
The beginning of neocortical feed-forward processing begins in Layer 4C, where the
majority of thalamic afferents of the lateral geniculate nucleus (LGN) synapse onto both
excitatory spiny stellate neurons and inhibitory interneurons of V1 (Hubel and Wiesel 1972,
Callaway 1998, Balaram and Kaas 2014). As previously stated, the cells of layer 4C maintain
ocular dominance but may or may not also possess orientation selectivity. The magnocellular
and parvocellular LGN afferents selectively target different divisions of Layer 4C, although they
also weakly innervate layer 6. Koniocellular afferents synapse onto neurons in layer 1-3. Thus,
it has been determined that LGN afferents synapse in almost every layer of V1, excluding Layer
4B and Layer 5.
Magnocellular afferents synapse onto cells in Layer 4Cα (Callaway 1998). These cells
can, but do not always, exhibit orientation selectivity. Layer 4Cα spiny stellate cells send their
axonal projections either to excitatory spiny stellate and pyramidal neurons of layer 4B or to
pyramidal neurons in the lower third of layer 2/3 (3B) (Martin 2002). Magnocellular afferents
from 4B then synapse onto “blob” regions in layers 2-3A, and 3B. Thus it can be said that the
magnocellular information in layer 4Cα is inevitably carried to the blob regions of layers 2 and 3,
where information about spatial relationships and object location is processed (Callaway 1998).
Although monocular dominance columns are prevalent throughout the cortical layers, in layer

	

6	

2/3 we see the first signs of binocularity in the visual cortex, with over half of all cells being
influenced by both the eyes. Thus it can be said that V1 architecture becomes less structured as
information moves out of layer 4 and the receptive fields of neurons begin to overlap and
integrate more complicated stimuli. From layer 2/3, cells that received M pathway information
travel to layer 4 of higher order visual cortical areas outside of V1. These areas include V2,
medial temporal area (V5), and medial superior temporal area (MST). Axons from layer 2/3 that
do not leave V1, will synapse onto layer 5 and 6 excitatory neurons. Axons from Layer 5 cells
will then leave V1 through the white matter to synapse onto subcortical structures important in
visual processing. Axons from layer 6 will predominantly synapse onto thalamic neurons.
Parvocellular LGN afferents take a different path through V1. These afferents synapse
onto excitatory spiny stellate neurons in Layer 4Cß and to a lesser extent in layer 4A. Cells in
Layer4Cß also exhibit center-surround receptive fields and ocular dominance, but they lack
orientation selectivity. From Layer 4Cß, axonal projections travel to layer 3B. Pyramidal
neurons in layers 2/3 are selective to orientation, but can exhibit binocularity. These neurons are
also clustered together in regions known as blob and interblob areas. Cells in the blob area are
not selective for orientation, but may be selective for color. Inputs from layer 2/3 leave V1 and
synapse in layer 4 of the following higher order areas: V2, V3, V4, and the inferotemporal (IT)
areas. These regions are located ventrally, and code information about the shape of an object.
In summary, studies looking at the density of axonal arborizations in the different cortical
layers have determined that layer 4 spiny stellate cells project predominantly to layer 2/3, with a
small subset of inputs synapsing in layer 5. Upon receiving layer 4 inputs, layer2/3 excitatory
neurons then project to layer 5 and 6 neurons in V1, and send inputs outside of V1 to accessory
cortical regions. Layer 5 neurons project to layer 6 and back to layer 2/3. Layer 6 neurons will

	

7	

then send projections back to layer 2/3 and layer 4C V1. Thus, layer 5/6, which I will discuss in
detail in the next section, is traditionally referred to as the feedback layer, although it also sends
axonal projections outside of V1.
B. Feed-back visual neocortical circuitry
Layer 5/6 is composed of layers 5 and 6. Layer 5 neurons receive inputs from layers 24B and they project inputs back to these layers with no specificity for blob or interblob regions
(Callway 1998). This is a characteristic feature of feedback projections, which typically exhibit
more generically distributed axonal arborizations. Layer 5 of the visual neocortex contains at
least three types of pyramidal neurons, each with distinct patterns of axonal arborization (Wiser
& Callaway, 1996). The most common are type A, which account for 2/3rds of the population.
Their basal and apical dendrites are confined almost exclusively to layer 5, and they receive the
majority of their input from layers 2-4B. The axons of these pyramidal neurons have a plethora
of arborizations in layers 2-4B. Another set of pyramidal neurons, have an apical dendrite
extending to layer 1-3, but an axon that remains predominantly localized to layer 5. These cells
receive inputs from layer 2-4B, but modulate mostly other layer 5 cells, with the exception of a
subset whose axons project through the white matter. It is speculated that due to the size of these
projecting axons, that they predominantly synapse onto the pulvinar nucleus of the thalamus.
However, this has not been definitively shown.
The third class of excitatory neurons have extremely large cell bodies relative to the other
pyramidal cells in layer 5, and these pyramids have an apical dendrite that extends to layer 1.
They receive input from both layer 2/3 and potentially from K pathway cells in layer one. Due
to their size, it is likely that they project to the superior colliculus, a region in the brain that
controls saccadic eye movements (Lund et al., 1988).

	

8	

Layer 6 excitatory neurons are the most diverse of all the cell layers. There are 2 classes
of excitatory neurons in this layer and they each comprise about half of all layer 6 pyramidal
neurons. Class I excitatory neurons have very few dendritic branches in layer 5 and they
predominantly send dense feedback axonal projections to layer 4C. Class II neurons have
abundant dendritic arbor throughout layer 5, but their axonal destination varies based on whether
they are in Class IIA or Class IIB excitatory neurons. Class IIA neurons, receive the majority of
their input from layers 2-4B. Their axons stay predominantly in layer 5 and 6 and are speculated
to thus predominantly mediate intralaminar excitation. Class IIB neurons keep their axons in V1,
but have dendrites in both layer 5 and layer 6. Thus, they receive input from thalamic afferents
and layer 2-4B. They predominantly send feedback projections to layer 2-4B. The discussion of
feedback and top-down circuits included explicit detail about excitatory neurons, so I will now
discuss the role that the various interneuron subtypes play in V1.
C. Murine model of the visual system
Although the murine model of the visual system is in many ways remarkably similar to
nonhuman primates, it retains some key distinctions. Visual integration also begins at the level of
the retinal ganglion cells in mice, however they have a smaller central zone dedicated to
binocular vision (Priebe and McGee, 2014). Also different than primates, RGCs with receptive
fields in the binocular zone only send minor projections to the LGN, while the contralateral eye
provides the majority of input to mouse LGN. The mouse thalamic neurons that receive
ipsilateral inputs differentially project their axons to binocular regions of V1 (Priebe and McGee,
2014). The mouse LGN also has a disproportionate number of relay neurons that are orientation
and direction selective, and these cells are speculated to be homologous to monkey koniocellular
cells (Cheong et al., 2013). Once in V1, mouse neurons share remarkable similarity to primate

	

9	

neurons, exhibiting orientation-selectivity as well as the presence of both simple and complex
cells (Niell and Stryker, 2008). Also similar to primates, as information moves to the deep layers
of V1 cells exhibit broader orientation tuning compared to more superficial layers (Kashube,
2014). One large difference, however, lies in the spatial resolution of V1 cells, which is
significantly lower in mice. Thus they prefer stimuli that have lower spatial frequencies. They
also exhibit a higher proportion of simple cells relative to complex cells, while the majority of
primate V1 is composed of complex cells (Niell, 2015). While the architecture of V1 in the
mouse is less structured than primates, there also exists ocular dominance columns (Dolen et al.,
2007). Also similar to primates, there is ample evidence in support of separate parallel
processing streams in the visual neocortex of the mouse. Once information leaves V1, it gets
processed in one of 9 extrastriate regions (Katzner et al. 2013) that form topically organized
representations of the contralateral eye. Information specific to the ventral stream is processed in
the following areas: laterointermediate (LI), posterior (P), postrhinal (POR). The dorsal stream
is processed by the following areas: rostrolateral (RL), anteriorlateral (A), posteromedial (PM)
and anteromedial (AM) (Wang et al., 2011,2012; Glickfeld et al., 2013; Laramee and Borie,
2015). Furthermore, the feedforward and feedback circuitry is also extremely comparable to the
aforementioned circuits I described above in nonhuman primates. However, one key difference
is that inputs into V1 not only target layer 4, but also target both layers 2/3 and 5/6.
Nevertheless, the majority of layer 2/3 neurons project to layer 5 pyramidal neurons, and layer 5
pyramidal neurons also project to the thalamic nuclei and superior colliculus, as well as to
specific neurons in layers 2-4. In vivo recordings on V1 putative interneurons have also
discovered properties of these neurons that are similar to primates, including their broader tuning
(Metin et al., 1988). In the subsequent paragraphs, I discuss some properties of the varies mouse

	

10	

interneuron subtypes in V1.
D. Interneuron contributions to local microcircuits
Inhibition is a critical element of proper circuit function in the brain. Interneurons are
abundant throughout all of the neocortical layers, where they play a pivotal role in modifying the
aforementioned excitatory circuits of the visual cortex as they integrate visual information.
There are several different subtypes of interneurons in the visual neocortex that synthesize γaminobutyric acid (GABA), the most abundant inhibitory neurotransmitter in the CNS. These
interneuron subtypes are composed of different biochemical, morphological, and synaptic
properties dependent on their location in the CNS. The three major neocortical interneuron
subtypes include 1) fast-spiking interneurons expressing the Ca+ binding protein parvalbumin
(PV), 2) non fast-spiking interneurons expressing the neuropeptide somatostatin, and 3)
interneurons expressing the serotonin receptor 5HT3a (Rudy et al., 2011). These three subtypes,
together, comprise 20-30% of all neocortical neurons (Markram et al., 2004).
D1. Fast-spiking interneurons
Fast-spiking interneurons (FS) comprise the most abundant interneuron subtype in the
CNS, make up 40% of all inhibitory neocortical neurons (Rudy, B. et al., 2011, Markram, H et
al., 2004; Ascoli, et al., 2008). These interneurons express the calcium-binding protein
parvalbumin (PV+), and are thus also commonly referred to as PV+ interneurons. Typically, FS
interneurons exhibit a basket cell morphology with aspiny, multipolar dendrites (Rudy et al.,
2011). This morphology comes from the fact that these cells preferentially target the somata
and proximal dendrites in such a way that the convergent inputs of several FS cells give the
appearance of a “basket” around pyramidal cell somata. A second common morphology adopted
by FS cells is that of the chandelier shape (Rudy et al., 2011). Chandelier cells most commonly

	

11	

target the axon initial segment (AIS) of pyramidal neurons (Kawaguchi and Kubota, 1997;
Markram, H. et al 2004). FS interneuron axonal arborizations are typically restricted to the layer
in which their soma resides, although a small subset of FS cells have been shown to extend their
axons translaminarly (Rudy, B. et al 2011). FS cells exhibit fast-spiking electrophysiological
properties which include the ability to fire sustained high-frequency trains of action potentials
(APs) with both large and fast afterhyperpolarizations (AHPs) and very little spike frequency
adaptation (Kawaguchi and Kubota., 1997; Beierlein and Connors, 2000). They can be observed
to “stutter”, or fire abrupt episodes of intermittent, nonadapting, repetitive discharges. Of all the
neocortical interneurons they have the lowest input resistance and the fastest membrane time
constant (Rudy et al., 2011; Markram et al., 2004). FS interneurons provide significant lateral
inhibition across neocortical columns, and receive the majority of thalamocortical afferents in the
input layer of the neocortex (Beierlein and Connors, 2003). Neocortical pyramidal neurons
synapse predominantly on the dendrites of these interneurons. Research suggests that 90% of
local glutamatergic inputs synapse onto the dendrites, while the other 10% of inputs synapse on
the soma (Markram, H. et al., 2004). Although FS cells reside in most of the neocortical layers,
their greatest numbers lie in layer 4 where they comprise the largest distribution of the three
interneuron subtypes (Rudy et al., 2011). They also comprise the second largest distribution of
interneurons in layers 2/3 and layers 5/6, while they are absent in layer 1 of the neocortex. Thus,
the distribution of FS interneurons exhibits an inside-out laminar gradient. In vivo optogenetic
studies of the role of FS interneurons in the neocortical circuitry, indicate that they are the major
participants in the generation of γ oscillations (30-80Hz) in the neocortical microcircuit (Sohal,
V. et al 2009), as well as active participants in neocortical UP-DOWN states.

	

12	

D2. Non-fast-spiking interneurons
The second largest group of neocortical interneurons are the non-fast-spiking (NFS)
interneurons, expressing somatostatin (SST+) neuropeptide (Rudy et al., 2011; Markram et al.,
2004). NFS interneurons comprise 30% of neocortical interneurons (Rudy et al., 2011). The
most commonly discussed NFS interneurons are those of the martinotti cell morphology
(Palusckiewicz et al., 2011; Oliva et al., 2000). Martinotti cells are present throughout layers 2-6
of the neocortex, and their axons ascend vertically through the cortical layers, arborizing onto the
dendritic tufts of pyramidal neurons (Kawaguchi and Kubota, 1997; Rudy et al., 2011). Unlike
FS interneurons, NFS interneurons make their synaptic contacts onto the dendritic shafts and
spines of both basal and apical dendrites of excitatory pyramidal neurons (Markram et al., 2004).
The electrophysiological properties of these interneurons also differ greatly in comparison to FS
cells. Martinotti cells have a regular adapting firing pattern, and as such have frequently been
termed regular spiking non-pyramidal (RSNP) cells (Rudy et al., 2011; Kawaguchi and Kubota,
1997; Beierlein and Connors, 2000; Markram et al., 2004). They have higher input resistances
than other interneuron subtypes, and in layer 5/6 are frequently seen firing bursts of action
potentials when they are depolarized from hyperpolarized membrane potentials (Beirlein and
Connors 2000; Beierlein and Connors 2003). Local glutamatergic inputs from pyramidal
neurons onto Martinotti cells are facilitating, unlike those of FS interneurons which are
depressing. This facilitating capability, allows Martinotti cells to increase their feedback
inhibition as the rate and duration of presynaptic discharge increases (Tai et al., 2014). It is
speculated that this unique feature, makes Martinotti cells most likely to be activated during
periods of high network activity. Of mention, although Martinotti cells are the most well
documented NFS interneuron in the neocortex, NFS interneurons are a heterogeneous population

	

13	

of neurons with a variety of morphologies and biochemical profiles (Rudy et al., 2011). The
advent of transgenic mouse lines, allowed scientists to investigate the heterogeneity of this
population for the first time. In the X94 mouse, a unique NFS interneuron was discovered that
exhibited many of the electrophysiological properties of a FS interneuron, with the exception that
these cells did have significant spike adaptation (Ma et al., 2015). Similar to Martinotti cells,
these neurons also received facilitating inputs from local glutamatergic cells. They have axonal
arborizations that synapse in layer 4 of the neocortex rather than layer 1. Only time will tell
whether other NFS interneurons are present in the neocortex, however given the diversity of
molecular marker expression and connectivity of this population of interneurons studied thus far
(Gonchar et al., 2007; McGarry et al., 2010), it is highly likely that other NFS interneuron
subtypes exist. Most recently, it was shown that NFS interneurons that co-express calretinin
Ca2+ binding protein, preferentially reside in layer 2/3 while those NFS interneurons without
calretinin expression target infragranular layers preferentially (Rudy et al., 2011). These cell
types received varying inputs from the cortical layers, with NFS/ calretinin + cells receiving
excitatory inputs mainly from layers 2/3, while the deeper layer NFS/calretinin - neurons
received strong excitation from both layers 2/3 and layer 4. Interestingly, both of these subtypes
share a similar electrophysiological profile to the widely studied Martinotti cell. NFS neurons are
most abundant in layers 5 and 6, while they are the second largest population of interneurons in
layer 4. They are the least abundant of the three cell types in layers 2 and 3, and have almost no
occupants in layer 1.
D3. 5HT3a receptor-expressing interneurons
Interneurons expressing the 5HT3a serotonin receptor (5HT3aR) comprise the remaining
30% of neocortical interneurons (Rudy et al., 2011). Of all the interneuron subtypes, these

	

14	

interneurons are the most heterogeneous population (Rudy et al., 2011, Markram et al., 2004). It
is because of this, that 5HT3aR interneurons are the least understood of the three neocortical
subtypes. One biochemical finding in this subtype of interneurons, showed that some 5HT3aR
expressing interneurons also express vasoactive intestinal peptide (VIP), while neither NFS or
FS interneurons express this neuropeptide (Rudy et al., 2011). Using a transgenic mouse that
expresses GFP in interneurons expressing the 5HT receptor, it was shown that the VIP+
interneurons account for up to 40% of all interneurons expressing the 5HT receptor (Rudy et al.,
2011; Kawaguchi and Kubota, 1997). Another difference between this cell population and both
NFS and FS cells is their developmental origin. Unlike the other two subtypes, almost 90% of
5HT3aR interneurons have been shown to develop from the caudal ganglionic eminence (Rudy
et al., 2011). One study (Myoshi et al., 2010), found 9 different electrophysiological subtypes of
interneurons originating from the caudal ganglionic eminence. The majority of VIP+ neurons
form synapses with dendrites, although a small subset synapse onto the soma of pyramidal
neurons, similar to FS cells. Interestingly, this subset of neurons also shares the basket cell
morphology of FS cells, but its morphologies are more heterogeneous than the neurons (Rudy et
al., 2011). VIP+ neurons have the highest input resistance of all the cell types, and are thus the
most excitable. They are the predominant interneuron subtype in layer 2/3, and comprise almost
the entire interneuron population in layer 1.
The understanding of these different neocortical interneuron subtypes is imperative for
making broader generalizations about the results of my dissertation studies, as it is highly likely
that the differences I discussed above in the laminar distributions, electrophysiological
properties, and biochemical profiles of these interneurons are likely to lead to very different
modulation of the V1 excitatory neocortical circuitry. In the subsequent section, I will provide

	

15	

an overview of the three major classes of mGluRs, as this will also be pivotal to understanding
my research results.
1.2 Metabotropic Glutamate Receptors
A. General features of metabotropic glutamate receptors
L-Glutamate is the most abundant excitatory neurotransmitter in the nervous system. The
majority of excitatory CNS synapses utilize glutamate transmission (Petralia et al., 1996) and
express a unique composition of both ionotropic and metabotropic glutamate receptors. Fast
excitatory neurotransmission is predominantly mediated by three ionotropic glutamate receptor
types: N-methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionate
(AMPA; GluR1-GluR4) and kainate (GluR5-GluR7, KA1, KA2) receptors. Metabotropic
glutamate receptors (mGluRs) are the second type of glutamate receptors, comprising a
heterogeneous family of G protein-coupled receptors (GPCRs). A GPCR is a membrane-bound
protein that is activated by extracellular ligand binding, and interacts with G proteins to
transduce intracellular signals leading to a variety of downstream effects (Niswender and Conn,
2010). Upon ligand binding, a conformational change in the GPCR leads to dissociation of the
G-protein complex, consisting of α, ß, and γ subunits. The dissociated units continue on to exert
the various mGluR effects. The mGluRs belong to family C GPCRs, and these receptors are
constitutive dimers. Each dimer has 7 transmembrane spanning domains. The mGluRs are
divided into three distinct subtypes: group I mGluRs, group II mGluRs, and group III mGluRs.
These receptors are coupled to a variety of intracellular second messenger signaling cascades.
MGluRs are expressed throughout the nervous system, where they have been shown to modulate
intrinsic excitability, synaptic plasticity, and neurotransmission, at a wide array of both
excitatory and inhibitory synapses. While they most often function as neuromodulators of

	

16	

synaptic activities, sometimes they are directly involved in neurotransmission (Nakanishi, 1994).
The exact distribution of the various mGluR subtypes show regional, developmental, synaptic,
and cell specificity. Activation of mGluR second messenger systems leads to a variety of
downstream effects, including phosphoinositide hydrolysis, activation of phospholipase D,
changes in cAMP formation, and changes in ion channel function (Schoepp and Conn, 1993).
Modulation of the CNS by mGluRs has been implicated to have therapeutic potential in treating
several pathological conditions including schizophrenia, major depression, anxiety, addiction,
neurodegenerative disease, epilepsy, and FXS. A table summarizing the information provided
below is located at the end of this chapter (Table 1.1).
B. Group I mGluRs
Group I mGluRs include mGluR1 and mGluR5. There are several splice variants of each of
these receptors. Group I mGluRs have been shown to be highly expressed throughout the brain,
with sub-type specific differences in the location of expression. For example, mGluR1 has been
observed at high levels in hippocampus, globus pallidus, substantia nigra, thalamus, cerebellum,
and the olfactory bulb (Hovelso, et al., 2012; Bhattacharyya, 2016, Martin et al., 1992).
MGluR1 has also been found in the striatum, neocortex, and hypothalamus, albeit the expression
is less than that in the aforementioned structures (Hovelso et al 2012). The second group I
mGluR receptor, mGluR5, has high expression in hippocampus and olfactory bulb, as well as in
the neocortex, striatum, caudate nucleus, and nucleus acumbens (Hovelso et al., 2012).
Compared to mGluR1, mGluR5 has lower expression levels in the cerebellum (Bordi and
Ugolini, 1999) and thalamus (Hovelso 2012). Group I mGluRs are also expressed in astrocytes.
In particular, astrocytic mGluR5 has been shown to affect neurotransmission of adjacent
synapses (Hovelso et al., 2012). The distribution of these mGluR receptors changes throughout

	

17	

the course of development, indicating that mGluRs may be crucial for early circuit maturation.
The expression of mGluR1 continues to increase throughout development in both neocortex and
hippocampus (Shigemoto et al.,1992; Catania et al., 1994). However, expression of splice
variant mGluR5a has some developmental time limitations. In particular, mGluR5a expression
reaches its peak in the second postnatal week, and then decreases further on in development
(Reid et al., 1995). MGlur5b expression continues to increase postnatally and becomes the
predominant form of mGluR5 in the adult brain. It is suggested that the differential expression
of these mGluRs throughout development is important for synaptogenesis and/or neurogenesis of
the neocortex (Munoz et al., 1999; Martinez-Galan et al., 2001). Electron microscopy has
demonstrated that group I mGluRs are located mostly perisynaptically on the postsynaptic side
of the synapse (Tamaru et al., 2001). Binding of glutamate to group I mGluR receptors occurs
on the extracellular domain while allosteric modulators bind group I mGluRs along the seven
transmembrane spanning domain. Group I mGluRs are associated with the Gq/G11 G-protein, and
this protein positively couples group I mGluRs to phospholipase Cß (PLCß). PLCß activation
leads to hydrolysis of phosphoinositides which forms both inositol 1,4,5-triphosphate (IP3) and
diacyl-glycerol (DAG). IP3 mobilizes intracellular calcium and this in conjunct with DAG is
sufficient to activate protein kinase C (PKC). Activation of protein kinase C leads to the
production of new protein synthesis. Group I mGluRs activate several downstream effectors,
including phospholipase D, protein kinase pathways (casein kinase 1, cyclin-dependent protein
kinase 5, Jun kinase, and components of the mitogen-activated protein kinase/extracellular
receptor kinase (MAPK/ERK) pathway, and the mammalian target of rapamycin pathway
(MTOR/p70 S6 kinase pathway). The latter two pathways have both been implicated in Fragile
X pathology. The MAPK pathway can be activated by group I mGluRs in both hippocampal and

	

18	

neocortical cells, and is involved in group I mGluR-mediated long term depression (LTD) in
hippocampal excitatory neurons. This mGluR-LTD is exaggerated in Fragile X Syndrome. For
more details about MAP kinase and MTOR signaling, see section 1.3 (“The MgluR Theory of
Fragile X”). Furthermore, group I mGluRs have been shown to alter neurotransmitter release
through regulation of K+ and Ca2+ channels. Inhibition of presynaptic K+ channels by group I
mGluRs delays repolarization to increase neurotransmitter release whereas bidirectional
regulation of presynaptic Ca2+ channels leads to either increases or decreases in release. Further,
they have been shown to decrease neurotransmitter release through postsynaptic stimulation of
endocannabinoids, which are retrograde neurotransmitters that bind to CB1 receptors
presynaptically. Finally, group I mGluR-mediated LTD has been reported in many different
brains regions (Huber et al., 2002; Anwyl, 1999; Citri and Malenka, 2008), although group I
mGluRs have also been shown to modulate NMDA-dependent long term potentiation (LTP).
Upon activation of their second messenger systems, group I mGluRs undergo desensitization via
phosphorylation. A similar process is observed for the other mGluR groups. This
desensitization is important for preventing chronic overstimulation of the signaling pathway.
C. Group II mGluRs
Similar to group I mGluRs, group II mGluRs are widely expressed in the CNS. Group II
mGluRs consist of mGlur2 and mGluR3. Although splice variants have not been identified for
mGluR2, mGluR3 RNA is alternatively spliced into at least 4 variants (Niswender and Conn,
2010). Group II mGluRs are highly expressed in several brain regions including hippocampus,
neocortex, nucleus acumbens, striatum, and amygdala (Holvelso et al., 2012; Petralia et al.,
1996). mGluR2 is also highly expressed in olfactory bulb, cerebellum, caudate nucleus, and
thalamic reticular nucleus, while mGluR3 is particularly abundant in septum, and substantia

	

19	

nigra (Tamaru et al., 2001). MGluR3 is also the predominant group II mGluR receptor
expressed in microglial cells (Petralia et al., 1996), where it has been shown to induce an
apoptotic pathway (Taylor et al., 2005). The expression of group II mGluRs is also
developmentally regulated. The location of group II mGluRs varies depending on the region of
investigation. In the hippocampus, presynaptic mGluRs are predominantly type 2, while
postsynaptic receptors are most often mGluR3 (Shigemoto, et al 1997; Tamaru et al., 2001;
Petralia et al., 1996). However, this is not always the case as brain structures including
neocortex, hippocampus, and striatum may have both mGluR2 and mGluR3 receptors located
pre- and post- synaptically (Hovelso et al., 2012). Presynaptic mGluR2s are typically located
extrasynaptically at axons and pre-terminal regions, while postsynaptic mGluR2s are
concentrated at the soma and dendritic shafts. In contrast, postsynaptic mGluR3s are abundant
on dendritic spines, where they are located perisynaptically. This is similar to the distribution of
group I mGluRs (Tamaru et al., 2001; Zou, Jiang & Yang, 2017). Both group II mGluR
subtypes are coupled to Gi/o G-proteins. Dissociation of Gi/o leads to inhibition of adenylate
cyclase (AC) and reduction of cyclic AMP (cAMP) production. Decreased cAMP leads to
subsequent decreases in protein kinase A (PKA) activity and a decrease in protein translation.
Additionally, these receptors can also directly regulate ion channels when the ßγ subunit
dissociates from the mGluR and directly binds the ion channel (Chavis, et al 1994). Through this
mechanism, group II mGluRs have been shown to inhibit P/Q-type Ca2+ channels, effectively
decreasing neurotransmitter release (O’Connor et al., 1999). They have also been shown to
regulate the neurotransmitter release machinery downstream of Ca2+ entry (Hovelso, et al. 2012).
Furthermore, the ßγ subunits can directly interact with the SNARE complex (Hovelso et al.,
2012). Thus, group II mGluR activation can lead to a reduction in the number of active release

	

20	

sites. Group II mGluRs also play a role in synaptic plasticity, mediating both long term
depression and long term potentiation, although group II mGluR LTP has not been widely
documented (Hovelso, et al., 2012). Group II mGluRs are present at both glutamatergic an
GABAergic synapses, where they have been shown to inhibit neurotransmission (Doi et al.,
2002; Flavin et al., 2000, Wan, et al., 2007).
D. Group III mGluRs
Although my dissertation studies focus solely on the role of group I and II mGluRs in Fragile
X pathology, for the purpose of completeness I will briefly discuss some features of group III
mGluRs. Group III mGluRs consist of mGluR4, mGluR6, mGluR7, and mGluR8. There are
several splice variants of mGluR6, mGluR8, and mGluR7, although not all studies agree on
whether variants exist for mGluR4 (Niswender and Conn, 2010). Expression of group III
mGluRs varies depending on brain region and receptor subtype. While mGluR4 has a wide
expression in the CNS, it is only highly expressed in hippocampus and cerebellum. MGluR7 is
the most highly expressed group III mGluR overall, being located in large quantities in the
following brain structures: hippocampus, neocortex, globus pallidus, amygdala, colliculi, and
olfactory bulb (Hovelso et al., 2012). It is also located in striatum, substantia nigra, caudate
nucleus, and nucleus acumbens, albeit to a lesser extent. Group III mGluRs are also present in
microglia, where they have been shown to activate a neuroprotective pathway (Taylor et al.,
2005). All of the group III mGluRs are linked to the Gi/o subunit and decrease cAMP through the
same pathway as group II mGluRs. Group III mGluRs are mainly localized within the
presynaptic active zone of glutamatergic synapses where they act as autoreceptors (Kinoshita et
al., 1998; Shigemoto et al., 1997, Hovelso et al., 2012). Presynaptic group III mGluRs are
similar to presynaptic group II mGluRs with regard to their effects on hippocampal synaptic

	

21	

plasticity. In particular, these receptors have been shown to activate a form of long term
depression at synapses between hippocampal mossy fibers and stratum lucidum interneurons
(Hovelso et al. 2012). They have also been shown to have an indirect effect on LTP at these
synapses mediated by mGluR7. Current literature suggests that group II and group III mGluRs
can also activate both the MAPK pathway as well as the phosphotidyl inositol 3-kinase (PI3)
pathway (Niswender and Conn, 2010), making the understanding of the role of these receptors in
modulating synaptic transmission increasingly complicated.
E. Distribution and functional role of mGluRs in the visual neocortex
MGluRs are located in all layers of the visual neocortex and are known to play a role in the
developmental maturation of the visual circuitry, while also maintaining a role in the modulation
of synaptic events in mature vertebrate circuits. Although there is some layer variation,
generally speaking, the critical period for plasticity in the visual cortex starts in the third week,
peaks during weeks 4-6, and ends by one year after birth (Daw et al. 1992). During this time,
mGluRs play a very important role in the formation of eye-specific connections in the visual
neocortex. Immunohistochemical studies indicate that the quantity of group I mGluRs
diminishes over time during development (Reid et al., 1997), however, changes in the laminar
distribution of group I mGluRs is subtype specific. Postnatal expression of both group I mGluRs
is predominant in layers 1 and 5/6 immediately after birth (Reid et al., 1997). While this
distribution remains unchanged for mGluR1a, by the peak of the critical period for ocular
dominance plasticity, mGluR5 changes its major laminar distribution to layer 4, where it remains
abundant for the rest of the lifespan (Reid et al., 1997). Sensory deprivation during the critical
period postpones the laminar change in mGluR5 and produces increased levels of mGluR5 up to
6 weeks postnatally, without affecting mGluR1 expression and laminar distribution. Thus it is

	

22	

evident that mGluR5 is especially critical for normal sensory-dependent maturation of the visual
neocortex. Group I mGluRs also mediate synaptic plasticity during circuit maturation and after
the critical period ends.
Group II mGluRs show a unique laminar distribution throughout the course of development.
Immunohistochemical studies show that by 3-4 weeks, group II receptors are present in all layers
(Reid et al. 1995). However, over the next 3-5 weeks the distribution changes such that group II
mGluRs disappear from layer 4 in the adult brain (Reid et al., 1995). Sensory-deprivation during
this time postpones the disappearance of group II mGluRs from layer 4. These findings are
validated with functional studies observing the response of neocortical neurons to visual
stimulation in vivo (Beaver, Ji, & Daw, 1999). Iontophoretic application of group II mGluR
agonist, APDC, at three weeks led to suppression of visually evoked responses as well as
spontaneous activity in individual neurons from all layers. This effect was absent from layer 4
past 6 weeks of age. Furthermore, over the course of the lifespan group II mGluR expression
decreases, similar to what is seen with group I mGluRs (Beaver, Ji, & Daw, 1999). The time
course of the disappearance of group II mGluRs coincides with the development of ocular
dominance columns, and subsequent studies have shown that group II mGluRs are involved in
ocular dominance plasticity.
Group I and group II mGluRs affect mature visual neocortical circuits in a variety of ways,
including modulation of intrinsic properties, synaptic transmission, and synaptic plasticity.
Visual cortical neuron excitatory postsynaptic currents from adolescent mice (p14-p20), are
reversibly suppressed by a general mGluR I/II agonist in layers 4-6 (Sladeczek et al., 1993).
This effect can be blocked by potassium channel blocker 4-aminopyridine (4-AP), indicating that
mGluRs can act to open presynaptic potassium channels and effectively decrease transmitter

	

23	

decrease. A later study looking at a specific group II mGluR agonist, APDC, showed that group
II mGluRs were responsible for attenuating excitatory postsynaptic currents in pyramidal
neurons of all cortical layers (Flavin et al., 2000). This function of group II mGluRs was present
throughout development, as these changes were observed in both young (17-31) and old (53-67)
rats. Additionally, group II mGluRs have been shown to suppress GABAergic transmission in
layer 4 of the visual neocortex (Liu, Petrof & Sherman, 2014). From these studies, it appears
that the major role of group II mGluRs in mature circuits is to limit neurotransmitter release.
Due to the variation in group I and group II mGluR expression in V1 throughout
development, it is likely that alterations in signaling of either of these receptor subtypes in
individuals with FXS, will affect both the developmental maturation of the V1 circuitry as well
as the functionality of the circuits thereafter. For example, in the Fmr1 KO changes in group I
mGluR signaling have a significant effect on development of ocular dominance plasticity during
earlier V1 circuit maturation (Dolen et al., 2007), while altered signaling in adult mice leads to
absent plasticity in layer 5/6 of V1 (Wilson and Cox, 2007). These changes, and others, will be
discussed in the following paragraphs about Fragile X Syndrome.
1.3 Fragile X Syndrome
FXS is a heritable disorder of intellectual disability. Current consensus places the prevalence
of FXS at 1 in 4,000 males and 1 in 8,000 females (Hagerman, 2008). In the United States alone,
it is estimated that 1 million Americans carry the Fragile X mutation, and 100,000 of those
individuals have a full FXS diagnosis (National Fragile X Foundation, 2017). The mutation
leading to FXS is located on the X chromosome in a region coding for the FMR1 gene.
Translation of this gene leads to production of FMRP protein, an mRNA binding protein present
in a variety of human tissues including brain, testes, ovaries, lung, and kidney (Hinds et al.,

	

24	

1993). In the 5’ untranslated region of the FMR1 gene, there is a CGG sequence with a variable
number of repeats located in the first exon of the FMR1 gene. It resides 250 base pairs
downstream of the CpG island that serves as the promotor region of the gene. In healthy
individuals, this CGG sequence is <55 repeats. Individuals with a repeat number between 55200 are considered premutation carriers, and can exhibit unique phenotypes including Fragile X
Associated Tremor Ataxia Syndrome (FXTAS) and Fragile X Associated Primary Ovarian
Insufficiency (FXPOI). When the repeat number exceeds 200, the CpG island of the FMR1 gene
becomes hypermethylated, effectively silencing gene transcription. Subsequent production of
FMRP protein is almost nonexistent, leading to the development of a full FXS phenotype
(Figure 1.2). Severity of this phenotype is directly correlated to the expression level of FMRP,
which can vary in individuals that exhibit germline mosaicism and in females that undergo X
inactivation of the mutated chromosome with compensation by the second X-chromosome (De
Boulle et al., 1993). These individuals will typically have less severe symptoms than those with
a fully hypermethylated gene.
A. Physical and behavioral phenotype
The Fragile X phenotype includes a wide array of physical abnormalities in addition to
behavioral disturbances. Physical features of FXS vary throughout the lifespan and include wide
and protruding ears, long narrow faces with a prominent jaw and forehead, long and narrow eye
openings, excessively large head relative to body size, epicanthal folds, and strabismus. Growth
patterns in FXS children are also altered, with FXS children more commonly exhibiting
precocious puberty and thus being larger in size until adulthood, at which point these individuals
are smaller in stature than healthy adults (Loesch and Hay, 1988). This disrupted growth pattern
is believed to be secondary to endocrine dysfunction as is the exhibition of macroorchidism in

	

25	

more than 80% of FXS adult males (Merenstein et al 1996). Further, although less is known
about how loss of FMRP protein leads to connective tissue dysplasia, additional clinical
manifestations suggestive of connective tissue disorder in FXS include hyperextensible finger
joints, velvet-like skin, a high-arched palate, flat feet, recurrent otitis media, and increased
incidence of mitral-valve prolapse and other cardiac involvement. Unlike other developmental
disorders where the physical features are highly fixed and thus easily distinguished, the
remarkable variability of the physical features in FXS leads to a diagnosis made most frequently
by observation of neurological and/or behavioral manifestations.
The neurological symptoms of FXS include learning impairments, hypersensitivity to
sensory stimuli, hyperactivity and attention deficit, impulsivity, aggressiveness, anxiety, sleep
disturbances, epilepsy, and autistic features (D’Hulst 2007; Hagerman 2000). FXS always
includes some degree of cognitive impairment, with as many as 85% of males and 25% of
females having an intelligence quotient less than 70 (Stafstrom, 2009). This intelligence quotient
does not indicate a general handicap of intellect, as FXS children show cognitive strength in
vocabulary, long-term memory for meaningful and learned information, and face emotion
perception (Kogan et al., 2004). They do, however, have relative weaknesses in attentional
control, working memory, linguistic processing, and visual spatial cognition (Kogan et al., 2004).
However, the intellectual and behavioral disturbances typically vary in both type and severity.
Male children with FXS most commonly exhibit attention deficit hyperactivity disorder and
aggressive behaviors, while females with FXS most commonly show social avoidance and
shyness (Hagerman and Stafstrom, 2009). However, two other relatively common neurological
disturbances in FXS include epilepsy and autistic behaviors. It is estimated that up to 25% of
children with FXS full mutation will have seizures (Stafstrom, 2009), and as many as 50% of

	

26	

pre-pubertal fragile X children exhibit atypical electroencephalographic (EEG) patterns. Thus,
hyperexcitability of the CNS is a hallmark feature in humans with FXS. The seizures most
frequently resemble benign focal epilepsy of childhood (BFEC) (Berry-Kravis, 2002). EEG
patterns for BFEC commonly show abnormal centrotemporal spike patterns. BFEC typically
resolves before adulthood, making the observation of centrotemporal spike patterns on the EEG
of children with FXS a positive prognostic indicator (Hagerman & Hagerman, 2002). Of the
variable types of seizure patterns, complex partial seizures are most common in FXS, although
some children with FXS will have generalized tonic-clonic seizures (Wisniewski et al 1991;
Berry-Kravis, 2002). It is currently unknown how loss of FMRP leads to epilepsy in FXS,
however several potential mechanisms have been speculated based on animal studies (see 1D
and 1F). Autistic features, including gaze aversion, hand flapping or biting, shyness, and social
avoidance, are extremely common in FXS. As many as 30% of children with FXS will also
receive a full autism diagnosis (Gibson et al. 2008; Hagerman et al. 2005). Additionally, 2-5%
of autistic children have FXS, making it the leading genetic cause of autism (Kaufman et al.
2004). It is hypothesized that hyperexcitable cortical circuitry underlies the pathology of autism,
as inundation of the neocortical circuitry with sensory information that cannot be inhibited
properly may lead to impaired sensory integration and subsequent alterations in physical and/or
behavioral responses. Finally, hypersensitivity to sensory stimuli is another key feature of
neurological pathology in FXS and is believed to be due to impaired sensory integration.
Impaired performance on tasks requiring sensory processing is suggestive of impaired sensory
integration. Impaired performance has been observed in males with FXS performing a global
motion task that probes the M pathway (Kogan et al., 2004). As previously mentioned, the
magnocellular pathway is specialized for extrapolating information about object motion. When

	

27	

FXS males where presented with visual stimuli that probe the magnocellular pathway (low
spatial frequency, high temporal frequency), they performed significantly worse than mental-age
matched and chronological-age matched controls in their ability to detect contrast (Kogan et al.,
2004). While it could be argued that attentional deficits led to the impaired performance,
scientists also probed for the ability to detect stimuli that probe the parvocellular pathway (color
stimuli, motionless gratings). Individuals with FXS had no trouble with these tasks.
Additionally, immunohistochemical studies confirmed that FMRP expression is elevated in areas
of the thalamus that carry magnocellular pathway visual information, providing an anatomical
correlate to suggest that the magnocellular pathway is more susceptible to pathology in the
absence of FMRP. Other studies have also observed deficits in processing magnocellular
pathway information (Kogan et al., 2004b; Cornish et al. 1998). Furthermore, individuals with
FXS are also hypersensitive to somatosensory stimuli, suggesting potentially impaired
integration of somatosensory information. Presumably, this hypersensitivity has neocortical
origins, as higher integration of sensory stimuli occurs in the neocortex. However, the specific
mechanisms have yet to be determined, although animal studies are beginning to uncover
answers (See 1F “Neocortical Pathogenesis in FXS”).
B. Neuroanatomical findings in FXS individuals
Neuroimaging studies in humans have documented a brain that is grossly normal in FXS. A
handful of studies have shown increases in hippocampal volume (Kates et al., 1997b; Reiss et al
1994), as well as an enlarged caudate nucleus and thalamus (Reiss et al., 1995), although these
changes may be age dependent as they are not always observed. One structural finding that has
been shown more consistently is a decrease in posterior cerebellar vermis size (Mazzocco et al.,
1997; Mostofsky et al., 1998; Reiss et al., 1994). The vermis has a role in modulating sensory

	

28	

input to the auditory and visual cortices, as well as modulating higher executive functions
through its connections with the frontal lobe (Ivry, 1993). Additional neuroanatomical studies
using Golgi impregnation of individual neurons have also revealed changes in the dendritic
morphology of both neocortical and hippocampal neurons in FXS (Hinton et al., 1991; Irwin,
Galvez, & Greenough, 2000; Irwin et al., 2001; Rudelli et al., 1985). Specifically, these studies
have shown an increase in immature dendritic morphologies in layers 3-5 of temporal and visual
cortices (Hinton et al., 1991; Irwin et al., 2001; Rudelli et al., 1985), and the pyramidal layer of
allocortex (Rudelli et al., 1985; Wisniewski et al., 1991). One study also documented an
increased spine density (Irwin et al., 2001), although this finding has not been documented in the
majority of human studies. These changes in spine maturation have also been documented in the
mouse model of FXS (Comery et al., 1997; Irwin et al., 2002), and are discussed below (see
Section 1F “Neocortical Pathogenesis in FXS”).
C. FMRP function and sub-cellular localization
FMRP is a cytoplasmic RNA-binding protein expressed in most neurons and glia,
although there is some regional specificity in expression level within the CNS. In particular,
higher FMRP expression levels are observed in the granular cell layers of the hippocampus and
cerebellum, habenula, and cerebral cortex (Hinds et al., 1993). FMRP is highly expressed
cytoplasmically, where it serves to regulate protein translation in the cell body, dendrites, and
synapses. However, dimerized FMRP can also shuttle into the nucleus through a nuclear
localization signal. It has three RNA-binding domains, two of which can bind RNAs with
“kissing-complex” tertiary motifs (Bhakar et al., 2012), and the third of which preferentially
binds RNAs with a G-quartet structure (Bassell and Warren, 2008). Inside the nucleus, FMRP
can be found bound to nuclear mRNAs, pre-mRNAs while they are being transcribed, and a

	

29	

nuclear exporter protein (Bhakar et al 2012). Together, these components comprise a
ribonucleoprotein complex (mRNP). This complex contains a nuclear export signal that allows
it to be translocated out of the nucleus, where it can be packaged into RNA granules. The
granules are shuttled to synapses, and release their contents upon stimulation. This allows the
translation of several different proteins including those implicated in spine morphogenesis and
synaptic plasticity. FMRP is estimated to bind 4% of total brain mRNA (Ashley et al., 1993;
Darnell et al., 2011), and has been shown to regulate the translation of 4-8% of synaptic proteins
(Bassell and Warren, 2008). The vast majority of rodent work to-date has focused on the role of
FMRP in regulating postsynaptic processes including protein translation (Pfeiffer and Huber,
2009). However, FMRP also targets mRNAs that comprise one-third of the presynaptic
proteome (Darnell et. al 2011) and is expressed presynaptically in granules (Antar et al., 2006;
Christie et al., 2009; Till et al., 2012, Akins et al., 2009; Akins et al., 2012). Thus, it seems
highly likely that loss of FMRP would lead to impaired presynaptic functions. Oddly, to-date
very few functional studies have been published about the role of presynaptic mechanisms of
FMRP regulation, so many questions remain about the pathology that results when FMRP is
silenced presynaptically. Nevertheless, postsynaptic studies have provided a wealth of
information about the general functions of FMRP. When FMRP is bound to its mRNA target,
protein translation does not occur. Thus it is considered to negatively regulate protein
translation, potentially through blockade of both the initiation (Centonze et al 2008, Lacoux et al
2012) and the elongation phase of translation (Ceman et al 2003, Darnell et al. 2011). Evidence
in support of negative regulation by FMRP includes FMRP dose-dependent suppression of
mRNA that is removed when FMRP is not allowed to bind its mRNA targets (Li et al., 2001), as
well as increased levels of protein synthesis occurring in the absence of FMRP (Huber et al.,

	

30	

2002). Deletion of FMRP impacts the expression of several different types of mRNAs and
proteins, including those responsible for a variety of cation channels, adhesion molecules,
neurotransmitter receptors, and components of the vesicular transport and release machinery
(Brown et al. 2001; Qiu et al 2008, Schutt et al 2009, Paluszkiewicz 2011).
D. The Fmr1 KO as an animal model of FXS
The murine homolog of FMRP has a 97% identical amino acid sequence with human
FMRP (Ashley et al., 1993), which makes it an optimal animal model for studying the pathology
of FXS. In 1994, the Dutch Belgium Consortium implanted embryonic stem cells containing a
viral vector with a disrupted Fmr1 DNA sequence into pseudo-pregnant C57BL/6J (B6) female
mice (Dutch Belgium Consortium, 1994). Offspring were crossed again with B6 mice resulting
in the first mouse model of FXS. These mice did not produce FMRP protein but did contain
detectable levels of Fmr1 mRNA, setting up an animal model that would hopefully resemble
human FXS phenotypes. While the gross physical features of Fmr1 KO mice were normal
relative to control mice (Bakkar et al., 1994), these investigators found enlarged testes in Fmr1
KO mice that has been corroborated in later studies (Slegtenhorst-Eegdeman et al., 1998).
Several studies also detected morphological changes in neocortical dendritic spines resembling
the observations made in the neocortex of human autopsy brains (Svoboda, K et al., 2001; Irwin
et al., 2002; Galvez and Greenough, 2005). Abnormal hippocampal spine morphologies, albeit
different than neocortical immature spines, were also observed in the Fmr1 KO mouse
(Grossman et al., 2006), and are discussed below (See 1D “The MGluR Theory of Fragile X”).
The Dutch Belgium Consortium (1994) conducted the first behavioral studies in the Fmr1 KO
mouse, and found that these mice show normal behaviors during passive avoidance tasks,
increased exploratory behavior and motor activity, and impaired acquisition of novel spatial

	

31	

information during platform reversal trials in the Morris Water Maze task. Although the authors
were uncertain about what increased motor activity and exploratory behaviors mean, they
proposed that these changes could be the result of hyperactivity or less efficient learning of the
environment, requiring a need to continue exploring (Dutch Belgium Consortium, 1994). They
also suggested that impaired acquisition of novel spatial information during reversal trials may
be due to deficits in hippocampal plasticity, a mechanism required for spatial learning.
Nevertheless, these studies were the first to demonstrate that the Fmr1 KO mouse did model
some of the features of human FXS. A surplus of behavioral studies conducted since then,
appear to have more or less conflicting results depending on the behavior under investigation.
While not necessarily a “behavior”, one of the most agreed upon features of the Fmr1 KO is
increased seizure susceptibility. Although spontaneous seizures have not been documented, the
Fmr1 KO has increased susceptibility to audiogenic seizures, induced by a 120-125 decibel,
high-intensity siren (Bakkar et al., 1994; Pacey et al., 2009, Yan et al., 2005; Ding, Sethna, &
Wang, 2014; Musumeci et al., 2000). Interestingly, the degree to which Fmr1 KO mice express
audiogenic seizure sensitivity is dependent on a variety of factors, including developmental age
and background strain (Pietropaolo s et al 2011, Veeraragavan et al., 2011; Goebel-Goody et al
2012). Another fairly consistent behavioral finding in the Fmr1 KO mouse is hyperactivity.
Fmr1 KO mice placed in an open-field have increased locomotor activity and/or enter the center
of the open field more often than wild type mice (Bakkar et al., 1994; Ding et al., 2014,
Pietropaolo et al., 2011; Kazdoba et al., 2014). Some behaviors observed in humans that have
also been observed in Fmr1 KO mice are repetitive behaviors (Pietropaolo et al., 2011; Spencer
et al., 2011), impaired sociability (Dahlhaus et al., 2010; McNaughton et al., 2008), and delayed
language development (Mineur et al., 2002), but these observations are not reproducible across

	

32	

all labs. For example, repetitive behaviors such as higher levels of self-grooming and excessive
marble-burying have also been observed in the Fmr1 KO mouse (McNaughton et al., 2008),
although not all studies show significant differences (Veeraragavan et al., 2011; Thomas et al.,
2011). The same can be said of behavioral assays for anxiety and cognitive deficits. Studies
done in mice to look at anxiety behaviors have produced more conflicting results with animals
showing reduced anxiety, measured as more time in open arms in an elevated plus-maze (Peier et
al., 2000; Heulens et al., 2012). Alternatively, others show either no change or reduced time in
open arms, indicative of increased anxiety (Bilousova et al., 2009; Mineur et al., 2002; Yan et
al., 2004). While it may seem like the conflicting results of several of these assays would
diminish the usefulness of the Fmr1 KO mouse as an animal model for the human disease, it is
important to keep in mind that behavioral assays are highly sensitive to several factors,
including, but not limited to: testing conditions, animal housing, genetic strain and age (Kazdoba
et al., 2014). Thus, it is possible that any one of these factors could be contributing to the
variable results produced from some of these assays. Furthermore, scientists may interpret the
behavior of these animals in different ways. For example, some scientists may consider more
time spent entering the center of an open field as a measure of decreased anxiety, whereas others
may consider it a measure of increased hyperactivity. However, extrapolating the reason for an
observed behavior is extremely difficult, as many behaviors are not limited to the activation of a
single brain region and some assessments may test several behaviors at once. Nevertheless, the
consensus that Fmr1 KO mice exhibit anatomical defects, hyperactivity, and increased seizure
susceptibility, suggests that this mouse is indeed a valid model of human FXS. All of the
aforementioned deficits have been rescued by an mGluR5 antagonist (Dolen et al., 2007), and
this has helped develop broad support of “The mGluR Theory” of FXS, which I will next discuss

	

33	

in detail.
E. Altered glutamate signaling and “The mGluR Theory”
Around the time that the Fmr1 KO mouse was developed, it was shown that activation of
group I mGluRs leads to increased protein synthesis (Weiler et al., 1993) and that FMRP is
translated at synaptosomes in response to stimulation of group I mGluRs (Weiler et al., 1997).
Additionally, it was determined that synaptic events such as long-term depression (LTD) and
long-term potentiation (LTP) require local protein synthesis for their maintenance (Kang and
Schuman, 1996). These studies provided a rationale to study mGluR-dependent LTD in the
hippocampus of the Fmr1 KO mouse, a form of plasticity that was known to depend on protein
translation at the synapse (Huber, Roder & Bear, 2001; Huber and Bear, 2000). The results of
this study determined that group I mGluR-mediated LTD in the CA1 region of the hippocampus
was elevated in the Fmr1 KO (Huber et al. 2002). This original study prompted the development
of “The mGluR Theory of Fragile X” (Figure 1.3), which postulated that several of the
neurological and psychiatric aspects of FXS are due to exaggerated responses downstream of
group I mGluR activation (Bear, Huber, and Warren, 2004). This theory maintains that any
processes requiring synaptic protein translation upon activation of group I mGluRs will be
exaggerated in the absence of FMRP. Subsequent hippocampal studies built support for this
theory. In particular, it was shown that activation of group I mGluRs in the CA3 region led to
prolonged epileptiform discharges in the Fmr1 KO mouse (Chuang et al., 2005), and that
stimulation of endogenous glutamate was sufficient to activate these epileptiform discharges in
the Fmr1 KO. This was not the case for wild type animals, which required pharmacological
activation of group I mGluRs before any epileptiform activity was observed. Exaggerated
mGluR-LTD has also been observed in the cerebellum of the Fmr1 KO mouse (Koekkoek et al.,

	

34	

2005), while reduce mGluR LTD has been reported in the neocortex (Martin, Lasalle &
Manzoni, 2016). Nevertheless, it appears that blocking mGluR5 signaling may be
therapeutically advantageous in the Fmr1 KO mouse. A pivotal study supporting the mGluR
theory used genetic reduction of mGluR5 in the Fmr1 KO mouse to rescue several of the
observed defects in the Fmr1 KO mouse, including increased seizure susceptibility, aberrant
spine morphology, elevated hippocampal protein synthesis, exaggerated hippocampal mGluRLTD, and aberrant experience-dependent plasticity in the visual neocortex (Dolen et al., 2007).
Pharmacological administration of the mGluR5 antagonist was also able to rescue behavior on an
open field test and increased susceptibility to audiogenic seizures (Yan et al., 2005), prolonged
epileptiform discharges (Chuang et al., 2005), hypersensitive acoustic startle (De Vrij et al.,
2008), and decreased mRNA granule expression (Aschrafi et al., 2005). In addition, other labs
have been able to replicate the original study using group I mGluR antagonism to restore excess
hippocampal protein synthesis (Osterweil et al., 2010) and hippocampal spine density (de Vrij et
al., 2008). Restoration of plasticity deficits in the neocortex have also been reported (Martin,
Lasalle & Manzoni, 2016). In addition to restoration of synaptic plasticity, another study in the
amygdala showed that application of a group I mGluR antagonist MPEP, could rescue decreased
frequency of fast ionotropic glutamate-receptor-mediated miniature excitatory postsynaptic
currents (mEPSCs) in the Fmr1 KO (Suvrathan et al., 2010). Therefore, although it appears that
the changes in group I mGluR-mediated signaling vary regionally, in many cases blockade of
these mGluRs is effective in restoring several of these defects.
At the time that the original hippocampal mGluR-LTD study was conducted in the Fmr1
KO mouse, it was uncertain how FMRP was interacting with group I mGluR signaling to
regulate synaptic protein translation. As I previously mentioned group I mGluRs not only

	

35	

activate their canonical Gq mediated pathway, but they also activate multiple kinase pathways,
all of which lead to local protein translation (Tang and Schuman, 2002). The PI3K-mTOR and
ERK/MAPK signaling pathways, however, can both be activated in the dendrites by group I
mGluR signaling, and have been shown to play a role in the exaggerated mGluR-LTD occurring
due to loss of FMRP (Bhakar, Dolen & Bear, 2012). When group I mGluRs are stimulated in the
WT mouse, PI3K-mToR pathway is activated via formation of a protein complex consisting of a
scaffolding protein (homer), mGluR5, and PIKE. PIKE activates phosphoinositide 3-kinase
(PI3K) resulting in phosphorylation of PIP2. Newly created PIP3, with the assistance of another
kinase called phosphoinositide-dependent kinase (PDK) phosphorylates serine/threonine kinase
(Akt). Akt directly phosphorylates mTOR, leading to translation of proteins at the synapse that
are important for cell growth, proliferation, autophagy, and synaptic plasticity. FMRP

normally functions to negatively repress mTOR signaling by binding PIKE and other
effectors of mTOR. Upon stimulation of group I mGluRs, protein phosphatase 2A
(PP2A) is activated, which leads to dephosphorylation of FMRP, freeing PIKE to initiate
mTOR-mediated protein translation (Narayanan et al., 2007). mTOR signaling will
eventually activate S6K, which will rephosphorylate FMRP to shutdown FMRPdependent translation once again (Santor et al., 2011). In fact, blockade of mTOR
signaling by rapamycin leads to loss of mGluR-LTD at CA1 synapses in the
hippocampus of wild-type mice (Sharma et al., 2010). However, loss of FMRP in the
Fmr1 KO mouse leads to enhanced mTOR signaling at rest and exaggerated mGluR-LTD
that becomes mTOR-pathway independent (Sharma et al., 2010).
The extracellular signal regulated kinase (ERK) pathway is also important for group I
mGluR-mediated LTD. The ERK pathway is a mitogen-activated protein kinase signaling
	

36	

cascade (MAPK), which means that it characteristically involves the following sequential steps:
GTPase activation (Ras), activation of a MAPK kinase (Raf), phosphorylation of MAPK kinase
(MEK), phosphorylation of ERK, and ERK-mediated protein translation. ERK activates protein
translation directly via activation of ribosomal protein S6K (RSK) as well as indirectly through
phosphorylation of MAPK interacting protein kinase (MNK ½), which leads to activation of
eukaryotic elongation initiation factor (eIF4E). eIF4E binds the 5’ mRNA cap and initiates
translation of proteins involved in AMPA receptor internalization (Bhakar, Dolen & Bear, 2012).
These include, but are not limited to, Arc/Arg3.1, oligophrenin-1 (OPHN1), microtubuleassociated protein 1B (MAP1B), and STEP (Sanderson et al., 2016). AMPA receptor
internalization is the main mechanism mediating mGluR-LTD. Several of these proteins are
elevated in the Fmr1 KO (Zalfa et al. 2003; Darnell et al. 2011), providing a potential
explanation for exaggerated mGluR LTD in FXS. Furthermore, the internalization of AMPA
receptors is also implicated in dendritic spine stability, which is aberrant in the Fmr1 KO.
AMPA receptor endocytosis is correlated with actin reorganization (Vanderklish et al., 2002;
Eales et al., 2014), and it is this remodeling that changes the morphology of dendritic spines
(Cingolani et al., 2008). Finally, ERK signaling has been shown to be responsible for the
elevated protein synthesis in the Fmr1 KO, as inhibition of ERK ½ pharmacologically leads to
normalization of excessive protein levels (Osterweil et al., 2010). This is not observed by
blocking the mTOR pathway.
In summary, Group I mGluRs have been implicated in the pathogenesis of FXS since the
creation of the Fmr1 KO mouse over 20 years ago. The pathogenic mechanisms downstream of
group I mGluR activation occur due to loss of FMRP regulation downstream of two pathways
connected to group I mGluR signaling, the pI3-mTOR and ERK/MAPK pathways. These

	

37	

defects led to the development of “The mGluR theory of Fragile X” which postulates that any
processes requiring protein translation downstream of group I mGluR signaling will be altered in
FXS, and this may be the hallmark alteration implicated in the wide array of FXS phenotypes.
Intriguingly, group II and III mGluRs have also been suggested to activate the pathways that are
aberrantly regulated upon group I activation in the Fmr1 KO (Niswender and Conn, 2010), so it
is possible that signaling of these receptors is also dysregulated in FXS. This warrants their
inclusion in our investigation of mGluR-mediated defects in the neocortex of the Fmr1 KO
mouse.

Finally, it is worth mentioning that the wealth of basic science research implicating

exaggerated group I mGluR signaling in the Fmr1 KO led to the investigation of an mGluR
antagonist for therapeutic usefulness in humans with FXS. Unfortunately, the clinical trials have
published disappointing results (Davenport et al., 2016). Phase II clinical trials were conducted
with the mGluR5 antagonists mavoglurant and basimglurant to assess their therapeutic potential
on a variety of different outcome measures. However, in both cases individuals with FXS were
comparable to placebo groups (Clapp, 2016; Santarelli, 2016), and the trials were subsequently
discontinued. At the very least, these results suggest that alterations in group I mGluR signaling
cannot alone explain the clinical manifestations of FXS. Clinical trials using agonists of
GABAergic signaling, however, have shown some promise. In the next section, I will discuss
several of the observed alterations in GABAergic signaling in the Fmr1 KO.
F. GABAergic dysfunction in the Fmr1 KO
Proper brain function depends on a correct balance between excitatory and inhibitory
signaling (Sabanov, 2016). While several of the aforementioned studies discovered changes in
glutamatergic function, other studies have uncovered defects in GABAergic inhibitory signaling.
GABA is the predominant inhibitory neurotransmitter in the CNS, and it binds to GABA

	

38	

receptors, of which there are two types: fast ionotropic GABAA receptors and metabotropic
GABAB receptors coupled to a G-protein signaling cascade. Both GABAA and GABAB receptors
are capable of hyperpolarizing the cell membrane, albeit through different mechanisms. GABAA
receptors are ion channels that permit influx of chloride into the cell upon GABA binding.
GABAB receptors mediate hyperpolarization of the membrane through opening of inward
rectifying K+ channels linked to the G-protein signaling cascade. Furthermore, GABAA
receptors mediate two types of inhibition: fast, phasic, inhibition that occurs upon release of a
high concentration of neurotransmitter, and tonic inhibition which is a slow, persistent, GABAA
current occurring when low concentrations of GABA reside in the extrasynaptic space. All
interneuron subtypes release GABA into the synaptic cleft, leading to inhibition of the
postsynaptic cell.
In the Fmr1 KO mouse, both biochemical and electrophysiological studies have
demonstrated defects in GABAergic function. Changes at GABAergic synapses have currently
been reported in the hippocampus (Hong et al., 2012; Sabanov et al., 2016), subiculum (Curia et
al., 2009), amygdala (Olmos-Serrano et al., 2010; Vislay et al., 2013), striatum (Centonze et al.,
2008), and neocortex (El Idrissi et al., 2005; Adusei, et al., 2010). Some of the first studies to
uncover deficits in inhibitory neurotransmission in Fragile X, found regional-specific changes in
the expression of both mRNAs and proteins comprising several GABAA receptor subunits, as
well as changes in expression of compounds important for the synthesis, transportation, and
catabolism of GABA. Fmr1 KO mice show decreased mRNA levels of GABAA d and ß3
subunits in the hippocampus and neocortex (Gantois, I. et al, 2006; Hong et al., 2012).
Additionally, 8 out of 18 known subunits of the GABAA receptor have been shown to be
underexpressed in the neocortex of Fragile X mice. This underexpression includes the subunits

	

39	

comprising the most abundant GABA receptor in the CNS, namely α1, β2, and γ2. These GABA
subunits comprise 60% of the GABAA receptors in the brain (D’Hulst et. al, 2006).
Interestingly, these results are corroborated in a drosophila model of FXS, indicating that
dysfunctional inhibition is an evolutionarily conserved pathology in Fragile X Syndrome.
Protein expression of several of the aforementioned GABAA subunits have also been shown to be
reduced in the neocortex, hippocampus, diencephalon, and brainstem of Fragile X mice (El
Idrissi et al., 2005, Adusei, D. et al 2010). Deficits in the expression of these various subunits
point to a potentially therapeutic intervention, as these subunits are sensitive to either and/or both
benzodiazepines and neuroactive steroids like alphaxalone (D’Hulst et. al, 2006). Interestingly,
there appears to be developmentally specific changes in the expression levels of these mRNAs
and proteins. While many of the above mentioned mRNAs had correspondingly decreased
protein expression at early time points in development, by adulthood (2-3months) the majority of
these mRNAs show normal protein expression. The exception to this lies in the protein
expression of ß2 subunit, which remains decreased into adulthood, and has been linked to
disrupted locomotor behavior in the mouse model of Fragile X (Adusei, D et al., 2010).
Furthermore, not only has decreased GABAA subunit expression been extensively documented,
but mRNA expression of Gad1, Gat1 and Gat4, and ssadh have been shown to be decreased in
both the neocortex and cerebellum. These mRNAs create proteins important for GABA
synthesis, transport, and degradation, respectively (D’Hulst, C. et al 2008). A protein
responsible for the clustering of GABAA receptors to the postsynaptic membrane, gephyrin, also
showed reduced expression in the neocortex (D’Hulst, C. et al 2008). Presynaptically, changes
in the expression of GABA-synthesizing enzyme GAD have also been documented (Adusei et
al., 2010, D’Hulst et al., 2009, El Idrissi et al. 2005, Olmos-Serrano et al 2010), and these

	

40	

changes appear to be dependent on the brain region examined. Given that underexpression of
several elements of the GABAergic system have been shown abundantly in the neocortex, it
seems highly likely that inhibition, more so than excitation, is mediating regional specific
changes in the neocortex (D’Hulst, 2009).
In addition to the extensive mRNA and protein expression studies, a growing body of
electrophysiological evidence is building to confirm the functional defects in inhibitory
neurotransmission in FXS. Not surprisingly, these changes show both increases and decreases in
inhibitory transmission that are, to some degree, regional specific. Fragile X mice showed a
marked increase in the frequency of spontaneous and miniature inhibitory currents, suggesting
that phasic GABAergic transmission is elevated in the striatum due to the loss of FMRP
(Centonze, et al. 2008). This study also shows decreased GABAergic synapse density in the
subiculum, contrary to their electrophysiology results. It is plausible that increased GABAergic
transmission is a compensatory mechanism induced by the reduction in synapse number,
although this remains to be proven. Later studies, also show decreases in tonic inhibition in the
subiculum (Curia, 2009), a region critical to spatial encoding and retrieval of short-term
memories (D’Antuono, M. et al. 2003). Inhibitory deficits were later documented in the
amygdala, a limbic area that processes emotionally salient information, especially those
pertaining to fear. Both phasic and tonic inhibition, mediated by GABAA receptors, were
reduced in the amygdala of the Fmr1 KO mouse. Electron microscopy in the amygdala showed
decreases in the number of inhibitory synapses in Fmr1 KO mice, similar to what was observed
in the subiculum (Olmos-Serrano, et al. 2010). The authors attribute the decrease in synapse
number as well as a decrease in GABA production to the electrophysiological defects they
observe. A recent study done in the hippocampus of the Fmr1 KO also demonstrated decreased

	

41	

amplitude in evoked inhibitory currents of CA1 pyramidal neurons, as well as decreased
amplitude and frequency of sIPSCs and mIPSCs (Sabanov et al., 2016). Although the authors
provide no mechanism to explain these changes, they show that there is decreased protein
expression of several GABA subunits, indicating a potential decrease in the number of GABA
receptors in CA1 pyramidal neurons, or a change in their composition and function. Although
all of these studies implicate GABAA abnormalities, a selective few studies have also shown
changes with GABAB in the Fmr1 KO. Henderson et al. (2012), found that several diseaserelated pathologies in the Fmr1 KO mouse could be reversed in slice preparations, cultured
neurons, or in vivo, by activation of arbaclofen, an agonist at GABAB receptors. These
pathologies include increased basal protein synthesis, elevated AMPA internalization, and
abnormal dendritic spine morphology. Abnormal auditory EEG patterns as well as increased
audiogenic seizures can also be corrected by arbaclofen (Pacey et al., 2009; Sinclair et al., 2017),
in addition to rescue of behavioral performances on a T-maze task and anxiety behaviors
(Sinclair et al, 2017). Finally, an electrophysiological study recording IPSPs in hippocampal
pyramidal neurons, showed that application of arbaclofen could restore GABAB deficits in
inhibition (Wahlstrom-Helgren & Klyachko, 2015). Unfortunately, just this month phase III trial
results from an arbaclofen study have recently been published and have reported no
improvement in symptoms (Berry-Kravis et al., 2017). It is evident from both this clinical trial
as well as the failed clinical studies of mGluR5 antagonists, that several different receptor
pathways converge on FMRP, such that any drug treatment targeting one receptor is unlikely to
be enough to rescue abnormal physical and/or neuropsychiatric phenotypes.
G. Neocortical pathogenesis in FXS
Studies in the neocortex of the Fmr1 KO provide evidence for changes in both spine

	

42	

morphology and intrinsic excitability. Network level studies also show changes in neocortical
excitability measured as differences in synaptic plasticity, spontaneous activity, and
synchronicity of firing. Although single cell electrophysiological studies are limited, they too
provide support for increased synaptic excitability, that appears to be contingent on the synapse
under investigation. Furthermore, some of these changes can be rescued by glutamatergic
receptors. In the following paragraphs, I will provide details about each of these investigated
changes, and will conclude with a discussion about neocortical changes specific to development.
G1. Anatomical studies
Of all the investigations into neocortical pathology in FXS, perhaps the most widely
identified disruption is with the spine morphology of excitatory pyramidal neurons. Dendritic
spines in the neocortex are the major postsynaptic recipients of excitatory glutamatergic inputs
(Pan et al.; 2010), and so it seems likely that aberrant spine morphology would affect the
excitability of the neocortical circuitry as well as its ability to integrate sensory information. As
previously discussed, Golgi-stain impregnations in autopsy brains of males and females with
Fragile X have uncovered significant changes in the spine morphology of the human neocortex
(Irwin et al, 2000). All of these studies have documented the presence of long, thin, immature
spines on the apical dendrites of pyramidal neurons in layers 3 and 5 of the parieto-occipital
cortex (Rudelli et al.,1985; Wisniewski et al., 1991; Hinton et al., 1991; Irwin et al., 2001).
Immature neocortical spines typically exhibit longer lengths with irregular dilations and smaller
spine head sizes relative to the shaft. These spines are presumed to be less functional than
mature spines. These pathological findings have been extensively studied in the neocortex of the
Fmr1 KO. Layer 5 visual neocortical neurons from brains of both younger (p21) and older mice
(p60-120) show an increased number of long spines relative to short spines as well as an

	

43	

increased number of immature spine morphologies (Comery et al., 1997; Irwin et al., 2002; Su et
al., 2011; Restivo et al., 2005). Some labs show an increase in spine density as well (Dolen et
al., 2007; Comery et al.; 1997; Su et al., 2011), but this finding is more controversial as it is not
always seen (Irwin et al.,2002) and is speculated by some to be dependent on developmental
stage (Nimchinsky et al., 2001). Similar spine pathology has been documented in the
somatosensory cortex of neonatal and adult mice (Pan et al.; 2010; Nimchinsky et al.; 2001).
Several labs have seen partial restoration of morphological deficits with chronic application of
antagonists of the group I mGlu5 receptor (Dolen et al.; 2007, Pan et al.; 2010), further
implicating group I mGluRs in neocortical pathology.
G2. Intrinsic excitability of neocortical neurons
Changes in the intrinsic excitability of neocortical neurons in the Fmr1 KO have not been
widely documented, although a handful of studies have found some differences. Gibson and
associates (2008) report that layer 4 excitatory neurons of the somatosensory cortex have
elevated intrinsic membrane excitability. Specifically, these neurons show an increase in input
resistance and a decrease in cell capacitance. These authors reported no changes in the intrinsic
excitability and action potential (AP) properties of FS interneurons in this layer. Another study
also showed that NFS interneurons of layer 2/3 somatosensory cortex (S1) have no change in
their intrinsic properties (Paluskiewicz et al., 2011). No other studies have assessed the intrinsic
properties of GABAergic neurons of the neocortex. Another study looking at the excitability of
cortical pyramidal neurons in layer 5B of S1 found no change in resting membrane potential,
input resistance, sag ratio, or membrane time constant in the Fmr1 KO, as well as no change in
the properties of single APs (Zhang et al., 2014). This same group did show increases in
excitability of the dendritic compartments of these neurons, however, and they attributed this to a

	

44	

change in Bk and HCN 1 (Ih) channels. Yet another study also determined that somatic
measures of input resistance, membrane time constant, and firing threshold are normal in layer 5
pyramidal neurons (Desai et al., 2006), demonstrating more evidence in favor of normal intrinsic
excitability of neocortical neurons. Whether changes in intrinsic excitability of other neocortical
brain regions are present, remains to be determined.
G3. Synaptic plasticity defects and other network alterations
Given that group I mGluRs are implicated in altered hippocampal plasticity in the Fmr1
KO, it seemed likely that they would play a role in neocortical synaptic plasticity as well. In
support of this, several studies have shown defective neocortical synaptic plasticity. In the visual
neocortex, loss of long-term potentiation was observed in layer 5/6 of the Fmr1 KO (Wilson &
Cox, 1997). This plasticity was partially mediated by group I mGluR5 receptor. Absent or
diminished LTP, as well as increased threshold for long term potentiation has also been observed
in the prefrontal neocortex (Meredith et al., 2007; Martin et al., 2016), somatosensory cortex (Li
et al., 2002; Desai et al., 2006,) auditory cortex (Yang et al., 2014), and anterior cingulate cortex
(Tao Chen et al., Zhao, et al., 2005; Koga et al., 2015). Group I mGluRs were either mediators
of this potentiation or rescued plasticity defects in the Fmr1 KO. These studies provide
compelling evidence to support the idea that there are population level changes in the excitability
of the neocortex. Additional network level studies in the neocortex have determined this to be
the case.
Additional studies show that Fmr1 KO mice exhibit a reduced threshold for audiogenic
seizures (Chen and Toth, 2001; Musumeci et al., 2000; Yan et al., 2005; Ding, Sethna, & Wang,
2014). In slice preparations, scientists have further shown that both thalamically driven and
spontaneous upstates show increased durations in layer 4 and layer 5 of the somatosensory cortex

	

45	

of Fmr1 KO mice (Gibson et al., 2008; Hays et al., 2011). These findings have also been
validated in vivo (Hays et al., 2011). Authors of the Gibson et al. (2008) study, found that the
duration of a layer 4 upstates in the Fmr1 KO was almost two-fold longer than control animals.
They further discovered that inhibitory network synchrony during upstates was significantly
decreased. Increasing inhibition of network activity is suggested to terminate upstates (Brunel
and Wang 2001), so loss of inhibitory synchrony would be expected to result in longer upstate
duration. Further, it was shown that genetic reduction of mGluR 5 signaling could restore
defects in upstate duration, and that this was not dependent on protein translation (Hays et al.,
2011). What exactly are upstates? Upstates are periods of higher neuronal activity, where cells
are more depolarized and more likely to fire action potentials. Upstates are frequently rhythmic,
vacillating with downstates (periods where cells are more hyperpolarized and exhibit little to no
firing) at a frequency of < 1Hz. While the exact function of upstates is unknown, it is speculated
that they comprise the slow oscillatory rhythm seen during slow wave sleep and quiet
wakefulness (Steriade et al., 1993). However, given that neuronal activity is heightened during
upstates, it is logical to use the duration spent in an upstate as a measure of cortical excitability.
Increased frequency of upstates in layer 2/3 of the somatosensory cortex of Fmr1 KO mice, with
normal upstate durations, has also been shown in vivo (Goncalves et al., 2013). These authors
noted layer 2/3 neurons have 3-fold higher firing rates during states of sleep or quiet
wakefulness, but this effect is not observed in awake behaving animals. Periods of rest are
extremely critical for memory consolidation (Ji and Wilson, 2007), and individuals with FXS are
known to struggle with memory tasks. Further, they are also known to exhibit sleep
disturbances, including shorter sleep durations and frequent awakenings (Hagerman &
Hagerman, 2002). Thus, the increased neuronal firing observed in the Fmr1 KO mouse during

	

46	

rest may be indicative of hyperarousal during sleep. Additionally, developmentally transient
alterations in the neocortical network of Fmr1 KO mice have also been reported (Bureau et al.,
2008; Goncalves et al., 2013; Harlow et al., 2010). For example, as early as postnatal day 4, two
photon Ca2+ imaging has shown that there is more frequent, synchronous firing of neuronal
ensembles in the Fmr1 KO that lasts until the 4th week of life (Goncalves et al., 2013). This is
significant, as these defects are occurring during the critical period of mouse brain development
(Anis, Vitaly, & Portera-Cailliau, 2015), and are likely to cause changes in circuit maturation.
All of these studies suggest that the neocortex in FXS is hyperexcitable, and may be
hyperresponsive to incoming sensory input. Not surprisingly, the somatosensory cortex of the
Fmr1 KO mouse, shows a faster spread of depolarization and increased sensory evoked activity
in response to whisker stimulation and forepaw stimulation respectively (Zhang et al., 2014).
While the above network studies provided sufficient evidence in favor of a
hyperexcitable cortical network, the individual synapses and cell types that are responsible for
this cortical excitability remain mostly unknown. Are all neocortical synapses affected by loss
of FMRP, or does this loss preferentially impact specific synapses, cell types, and/or layers? A
handful of studies have begun to answer this question, although single cell electrophysiological
investigations remain underrepresented in the neocortex. Paired recordings from layer 4 of the
somatosensory cortex show both a reduction in connection frequency and strength of pyramidal
neuron synapses onto FS interneurons, indicating a 51% reduced excitatory drive onto these cells
(Gibson et al., 2008). Furthermore, there was also a decrease, albeit smaller, in excitatory drive
onto pyramidal neurons of around 18%. The authors hypothesized that these changes should
lead to a net imbalance of excitatory/inhibitory circuitry in favor of hyperexcitability, and this
was confirmed with the upstate studies I discussed previously. A later study using transgenic

	

47	

mice that selectively express FMRP in either GABAergic cells or excitatory cells, showed that
deletion of Fmr1 in excitatory neurons mimics the longer upstates seen in the Fmr1 KO (Hays et
al., 2011). This was not the case for deletion of Fmr1 in GABAergic cells, indicating that
defective excitatory drive onto FS interneurons is the major contributor to neocortical circuit
hyperexcitability. Fast-spiking interneurons are not the only inhibitory cells that have been
implicated in neocortical FXS pathology. Specifically, another study showed that group I
mGluR mediated activation of NFS interneuron firing in layer 2/3 was reduced, and that this
diminished activation of NFS neurons led to impaired synchronization of spontaneous inhibitory
currents (sIPSCs) in the Fmr1 KO (Paluskiewicz et al., 2011). NFS interneurons receive
facilitating inputs from excitatory neurons, and so their inhibitory output increases with
increasing network activity (Paluszkiewicz et al., 2011; Rudy et al., 2011). Therefore, defective
inhibition by SST interneurons would specifically be detrimental in situations where cortical
activity was abnormally high, such as during epileptiform situations or conditions of unrelenting
sensory input.
G4. Developmental changes in the neocortex of the Fmr1 KO mouse
Further, single cell electrophysiological studies have determined that loss of FMRP
impacts the timing of the critical period for synaptic plasticity (Bureau et al., 2008; Harlow et al.,
2010). Strontium experiments looking at isolated Sr2+-AMPA mEPSCs and Sr2- NMDA
mEPSCs determined that quantal amplitude of both the AMPA and NMDA components at
developing thalamocortical synapses was unaltered. However, looking at the ratio of NMDA to
AMPA EPSC failures during minimal stimulation experiments at these synapses determined the
NMDA/AMPA ratio was increased in the Fmr1 KO due to a decrease in NMDA EPSC failures.
They suggest this indicates an abundance of silent synapses during the critical period (P7) in

	

48	

layer 4 of the somatosensory cortex in these animals. Further, they show that this leads to a shift
in the critical period for synaptic plasticity. In normal animals, LTP is strongest at P3 and P4,
and gradually decreases over the course of development (Barth and Malenka, 2001). However,
in Fmr1 KO mice LTP is significantly decreased at these ages and subsequently increases until a
maximum is reached at P6-7, wherein it starts to decline again and resembles the control mice
developmental time course. Interestingly, another study reported that during immediate
postnatal development, there was a decrease in connection probability of layer 4 spiny stellate
cell synapses onto layer 2/3 neurons in the Fmr1 KO compared to WT, and this alteration was
normalized by the third postnatal week (Bureau et al., 2008). Thus, these reports support the
idea that the neocortical critical period is delayed in the Fmr1 KO, and that FMRP plays a
critical role in circuit maturation. These functional studies are supported with anatomical spine
studies showing transient changes in spine morphologies during similar time frames
(Nimchinsky et al., 2001; Galvez and Greenough et al., 2005). These results are especially
intriguing in light of the fact that children with FXS experience developmental delays in motors
skills (Kau et al., 2002). Similar changes in critical period plasticity have been observed for
ocular dominance in the Fmr1 KO mouse (Dolen et al., 2007), providing additional evidence that
FMRP is required for the normal developmental progression of neocortical synaptic maturation.
In summary, the neocortical pathogenesis of the Fmr1 KO mouse is diverse. The most
widely accepted findings include those showing abnormal dendritic spine morphology and
absent or diminished LTP. In contrast, the jury is out regarding whether intrinsic excitability is
altered in neocortical neurons, although most studies suggest that it is not. Further, while several
network level studies show alterations in the Fmr1 KO leading to neocortical hyperexcitability,
studies of the individual synapses contributing to these network defects are rare. Finally, in the

	

49	

neocortex and potentially in other brain regions, FMRP is important during the early postnatal
development of the neocortex, as several deficits are seen during this time frame that are
transient but nonetheless important for the normal maturation of neocortical circuitry. Studies in
older mice are almost nonexistent, so it is difficult to say at this point how changes during
development affect adult animals.
1.4 Concluding Remarks
While the development of a murine model of FXS has provided us with invaluable
insight into the pathogenic neural mechanisms leading to the neuropsychiatric profile of FXS,
these studies have left us with more questions than answers, especially where it involves
neocortical involvement. My dissertation studies were conducted in an attempt to bridge the gap
between what we know about neocortical FXS pathology and what we still need to understand.
Three particularly interesting patterns have been observed in the neocortex of Fmr1 KO
mice. First, dendritic spine morphology is aberrant and includes immature spine morphologies
that may be contributing to the defective long term synaptic dynamics that have been
documented in the neocortex. Second, group I mGluRs have been implicated in the neocortical
pathology of this syndrome. However, it appears that there is regional specificity in the role of
FMRP downstream of group I mGluR signaling, as the plasticity defects seen in the neocortex in
most cases are diametrically opposed to the plasticity changes seen in the hippocampus. Thus,
more research assessing the role that group I mGluRs play in neocortical pathology is required so
that we may better understand how mGluR signaling in the neocortex differs from other brain
regions. Third, while electrophysiological investigations at the level of a single neocortical
synapse are limited in the Fmr1 KO (Gibson et al., 2008), it is suggested that excitation of
interneurons, in particular, seems to be the source of neocortical deficits. Defective excitation of

	

50	

interneurons has been shown in all layers of somatosensory cortex layer 1 (Gibson et al., 2008;
Hays et al., 2013; Paluskiewicz et al., 2011), and affects both FS and NFS interneurons.
Changes such as these have not been investigated in V1 of the Fmr1 KO despite several lines of
evidence suggesting that human patients suffer from significant visual-motor deficits, and >90%
of males with FXS are gaze aversive (Merenstein et al., 1996). These findings indicate sensory
hypersensitivity in response to eye contact (Cohen et al., 1989), and implicate V1 as a brain
region that is functionally altered in humans. The major alterations that have been observed in
V1 of the Fmr1 KO mice are layer 5/6 abnormal dendritic spine morphology, absent layer 5/6
LTP, and altered ocular dominance plasticity (Wilson and Cox, 2007; Dolen et al., 2007). Group
I mGluRs are involved in all of these processes, but no one knows how group I mGluRs are
modulating the individual synapses in these layers. Armed with the knowledge that mGluRs
have been shown to modulate both glutamatergic and GABAergic transmission in the visual
neocortex (Liu, Petro & Sherman, 2014)), it was my goal to investigate the role of mGluRs in the
local V1 circuitry of the Fmr1 KO. I provide evidence showing that group I mGluRs do not
modulate glutamatergic synaptic transmission in layers 2/3 and 5/6 of V1, while group II
mGluRs strongly suppress excitatory glutamatergic transmission in these layers. I also show that
group II mGluR-mediated modulation of glutamatergic synaptic transmission onto NFS
interneurons in layer 5/6 is altered. These results are significant, as they implicate an additional
mGluR subtypes in the pathology of FXS. Further, my research findings support previously
published research suggesting that reduced excitatory drive onto interneurons is a major
dysfunction in neocortical pathogenic mechanisms of FXS. I show that this defect is presynaptic
implicating FMRP in processes occurring at the axonal active zone, also supporting a previous
study showing that deletion of presynaptic FMR1 specifically replicates deficits in excitatory

	

51	

drive onto interneurons in somatosensory cortex (Hays et al., 2013). Hopefully, with mounting
electrophysiological support for the presynaptic role of FMRP, scientists will begin to uncover
new processes that are regulated by FMRP intracellularly. Finally, I show that changes in the
group II mGluR modulation of neurotransmission leads to disinhibition of the excitatory neurons
in layer 5/6, indicating circuit hyperexcitability in V1. This hyperexcitability may also be due to
the decreased intrinsic excitability I observe in layer 5/6 somatostatin interneurons. This finding
is the first to demonstrate changes in the intrinsic excitability of neocortical interneurons in layer
5/6 of the visual neocortex.

	

52	

APPENDIX

	

53	

Figure 1.1. Primary visual neocortex circuitry and parallel visual pathways.
In the top half of the diagram is a schematic drawing of the microcircuits in V1. K pathway
synapses modulate M (red) and P pathway (green), and are not shown for simplicity purposes.
Below the diagram are flow charts depicting the sequential flow of information through the
parallel pathways of the visual system.

	

54	

Table 1.1. Three groups of metabotropic glutamate receptors.
Metabotropic glutamate receptors belong to class C GPCRs. Group I mGluRs include receptor
subtypes 1 and 5, and are positively coupled to the Gq/G11 signaling pathway. Activation of the
Gq pathway leads to PLCß activation and hydrolysis of phosphoinositides. This leads to
formation of IP3 and DAG. IP3 mobilizes intracellular calcium and this in conjunct with DAG is
sufficient to activate PKC. Activation of PKC leads to the production of new protein synthesis.
Group II and III mGluRs are negatively coupled to the Gi/o signaling pathway. Dissociation of
Gi/o leads to inhibition of adenylate cyclase (AC) and reduction of cAMP production. Decreased
cAMP leads to subsequent decreases in PKA activity and a decrease in protein translation. All
of these receptors can also modulate ion channel activity directly, and have been shown to reside
upstream from MAPK/ERK and PI3-mTOR signaling.

	

mGluR groups

Receptor subtypes

Group I mGluRs
Group II mGluRs
Group III mGluRs

mGluR1, mGluR5
mGluR2, mGluR3
mGluR4, mGluR6-8

55	

G-protein signaling
pathway
Gq/G11
Gi/o
Gi/o

Figure 1.2. Development of FXS premutation and full mutation.
This diagram displays an X chromosome with a “fragile site” where the Fmr1 gene is located. In
healthy individuals, a CGG repeat region downstream from the promotor CpG island is normal
repeated no more than 55 times. In individuals with the permutation, CGG repeats vary from 55200, and this leads to excess production of FMRP and gain-of-function phenotypes, including
FXPOI and FXTAS. Individuals with a CGG repeat number > 200 will have hypermethylated
promotor regions and subsequent loss of FMRP production. This leads to the full FXS
phenotype.

	

56	

Figure 1.3. The mGluR Theory of Fragile X Syndrome.

	

57	

Figure 1.3. (cont’d)
The mGluR theory of FXS proposes that loss of FMRP downstream of group I mGluR signaling,
leads to increased translation of proteins implicated in synaptic plasticity, In the absence
of negative regulation of mRNA translation, proteins that cause internalization of AMPA
receptors are produced in excess, leading to the synaptic phenotype of exaggerated mGluR-LTD
in the hippocampus of the Fmr1 KO mouse. Loss of FMRP downstream of group I mGluR
signaling also leads to increased translation of proteins important for normal spine development,
leading to aberrant spine morphologies in the Fmr1 KO. Alterations in mGluR signaling have
subsequently been found in other brain regions, implicating a role for mGluRs in Fragile X
Syndrome pathogenic mechanisms.

	

58	

REFERENCES

	

59	

REFERENCES
Abrams, M. T., Reiss, A. L., Freund, L. S., Baumgardner, T. L., Chase, G. A., & Denckla, M. B.
(1994). Molecular-neurobehavioral associations in females with the fragile X full
mutation. Am J Med Genet, 51(4), 317-327. doi:10.1002/ajmg.1320510407
Adusei, D. C., Pacey, L. K., Chen, D., & Hampson, D. R. (2010). Early developmental
alterations in GABAergic protein expression in fragile X knockout mice.
Neuropharmacology, 59(3), 167-171. doi:10.1016/j.neuropharm.2010.05.002
Akins, M. R., Berk-Rauch, H. E., & Fallon, J. R. (2009). Presynaptic translation: stepping out of
the postsynaptic shadow. Front Neural Circuits, 3, 17. doi:10.3389/neuro.04.017.2009
Akins, M. R., Leblanc, H. F., Stackpole, E. E., Chyung, E., & Fallon, J. R. (2012). Systematic
mapping of fragile X granules in the mouse brain reveals a potential role for presynaptic
FMRP in sensorimotor functions. J Comp Neurol, 520(16), 3687-3706.
doi:10.1002/cne.23123
Antar, L. N., Li, C., Zhang, H., Carroll, R. C., & Bassell, G. J. (2006). Local functions for FMRP
in axon growth cone motility and activity-dependent regulation of filopodia and spine
synapses. Mol Cell Neurosci, 32(1-2), 37-48. doi:10.1016/j.mcn.2006.02.001
Anwyl, R. (1999). Metabotropic glutamate receptors: electrophysiological properties and role in
plasticity. Brain Res Brain Res Rev, 29(1), 83-120.
Aschrafi, A., Cunningham, B. A., Edelman, G. M., & Vanderklish, P. W. (2005). The fragile X
mental retardation protein and group I metabotropic glutamate receptors regulate levels
of mRNA granules in brain. Proc Natl Acad Sci U S A, 102(6), 2180-2185.
doi:10.1073/pnas.0409803102
Ashley, C. T., Jr., Wilkinson, K. D., Reines, D., & Warren, S. T. (1993). FMR1 protein:
conserved RNP family domains and selective RNA binding. Science, 262(5133), 563566.
Bassell, G. J., & Warren, S. T. (2008). Fragile X syndrome: loss of local mRNA regulation alters
synaptic development and function. Neuron, 60(2), 201-214.
doi:10.1016/j.neuron.2008.10.004
Baumgardner, T. L., Reiss, A. L., Freund, L. S., & Abrams, M. T. (1995). Specification of the
neurobehavioral phenotype in males with fragile X syndrome. Pediatrics, 95(5), 744-752.
Beaver, C. J., Ji, Q., & Daw, N. W. (1999). Effect of the group II metabotropic glutamate
agonist, 2R,4R-APDC, varies with age, layer, and visual experience in the visual cortex.
J Neurophysiol, 82(1), 86-93.

	

60	

Beierlein, M., Gibson, J. R., & Connors, B. W. (2000). A network of electrically coupled
interneurons drives synchronized inhibition in neocortex. Nat Neurosci, 3(9), 904-910.
doi:10.1038/78809
Beierlein, M., Gibson, J. R., & Connors, B. W. (2003). Two dynamically distinct inhibitory
networks in layer 4 of the neocortex. J Neurophysiol, 90(5), 2987-3000.
doi:10.1152/jn.00283.2003
Beltramo, R., D'Urso, G., Dal Maschio, M., Farisello, P., Bovetti, S., Clovis, Y., . . . Fellin, T.
(2013). Layer-specific excitatory circuits differentially control recurrent network
dynamics in the neocortex. Nat Neurosci, 16(2), 227-234. doi:10.1038/nn.3306
Berger, T. K., Perin, R., Silberberg, G., & Markram, H. (2009). Frequency-dependent disynaptic
inhibition in the pyramidal network: a ubiquitous pathway in the developing rat
neocortex. J Physiol, 587(Pt 22), 5411-5425. doi:10.1113/jphysiol.2009.176552
Berry-Kravis, E. (2002). Epilepsy in fragile X syndrome. Dev Med Child Neurol, 44(11), 724728.
Berry-Kravis, E., Hagerman, R., Visootsak, J., Budimirovic, D., Kaufmann, W. E., Cherubini,
M., . . . Carpenter, R. L. (2017). Arbaclofen in fragile X syndrome: results of phase 3
trials. J Neurodev Disord, 9, 3. doi:10.1186/s11689-016-9181-6
Bhakar, A. L., Dolen, G., & Bear, M. F. (2012). The pathophysiology of fragile X (and what it
teaches us about synapses). Annu Rev Neurosci, 35, 417-443. doi:10.1146/annurev-neuro060909-153138
Bhattacharyya, S. (2016). Inside story of Group I Metabotropic Glutamate Receptors (mGluRs).
Int J Biochem Cell Biol, 77(Pt B), 205-212. doi:10.1016/j.biocel.2016.03.003
Bilousova, T. V., Dansie, L., Ngo, M., Aye, J., Charles, J. R., Ethell, D. W., & Ethell, I. M.
(2009). Minocycline promotes dendritic spine maturation and improves behavioural
performance in the fragile X mouse model. J Med Genet, 46(2), 94-102.
doi:10.1136/jmg.2008.061796
Blackmer, T., Larsen, E. C., Takahashi, M., Martin, T. F., Alford, S., & Hamm, H. E. (2001). G
protein betagamma subunit-mediated presynaptic inhibition: regulation of exocytotic
fusion downstream of Ca2+ entry. Science, 292(5515), 293-297.
doi:10.1126/science.1058803
Bordi, F., & Ugolini, A. (1999). Group I metabotropic glutamate receptors: implications for
brain diseases. Prog Neurobiol, 59(1), 55-79.
Brager, D. H., & Johnston, D. (2014). Channelopathies and dendritic dysfunction in fragile X
syndrome. Brain Res Bull, 103, 11-17. doi:10.1016/j.brainresbull.2014.01.002

	

61	

Brown, V., Jin, P., Ceman, S., Darnell, J. C., O'Donnell, W. T., Tenenbaum, S. A., . . . Warren,
S. T. (2001). Microarray identification of FMRP-associated brain mRNAs and altered
mRNA translational profiles in fragile X syndrome. Cell, 107(4), 477-487.
Bureau, I., Shepherd, G. M., & Svoboda, K. (2008). Circuit and plasticity defects in the
developing somatosensory cortex of FMR1 knock-out mice. J Neurosci, 28(20), 51785188. doi:10.1523/JNEUROSCI.1076-08.2008
Cahusac, P. M., & Wan, H. (2007). Group II metabotropic glutamate receptors reduce excitatory
but not inhibitory neurotransmission in rat barrel cortex in vivo. Neuroscience, 146(1),
202-212. doi:10.1016/j.neuroscience.2007.01.049
Cartmell, J., & Schoepp, D. D. (2000). Regulation of neurotransmitter release by metabotropic
glutamate receptors. J Neurochem, 75(3), 889-907.
Catania, M. V., De Socarraz, H., Penney, J. B., & Young, A. B. (1994). Metabotropic glutamate
receptor heterogeneity in rat brain. Mol Pharmacol, 45(4), 626-636.
Catania, M. V., Landwehrmeyer, G. B., Testa, C. M., Standaert, D. G., Penney, J. B., Jr., &
Young, A. B. (1994). Metabotropic glutamate receptors are differentially regulated
during development. Neuroscience, 61(3), 481-495.
Cea-Del Rio, C. A., & Huntsman, M. M. (2014). The contribution of inhibitory interneurons to
circuit dysfunction in Fragile X Syndrome. Front Cell Neurosci, 8, 245.
doi:10.3389/fncel.2014.00245
Ceman, S., O'Donnell, W. T., Reed, M., Patton, S., Pohl, J., & Warren, S. T. (2003).
Phosphorylation influences the translation state of FMRP-associated polyribosomes. Hum
Mol Genet, 12(24), 3295-3305. doi:10.1093/hmg/ddg350
Centonze, D., Rossi, S., Mercaldo, V., Napoli, I., Ciotti, M. T., De Chiara, V., . . . Bagni, C.
(2008). Abnormal striatal GABA transmission in the mouse model for the fragile X
syndrome. Biol Psychiatry, 63(10), 963-973. doi:10.1016/j.biopsych.2007.09.008
Chang, Y. M., & Luebke, J. I. (2007). Electrophysiological diversity of layer 5 pyramidal cells in
the prefrontal cortex of the rhesus monkey: in vitro slice studies. J Neurophysiol, 98(5),
2622-2632. doi:10.1152/jn.00585.2007
Chattopadhyaya, B., Di Cristo, G., Higashiyama, H., Knott, G. W., Kuhlman, S. J., Welker, E.,
& Huang, Z. J. (2004). Experience and activity-dependent maturation of perisomatic
GABAergic innervation in primary visual cortex during a postnatal critical period. J
Neurosci, 24(43), 9598-9611. doi:10.1523/JNEUROSCI.1851-04.2004
Chavis, P., Shinozaki, H., Bockaert, J., & Fagni, L. (1994). The metabotropic glutamate receptor
types 2/3 inhibit L-type calcium channels via a pertussis toxin-sensitive G-protein in
cultured cerebellar granule cells. J Neurosci, 14(11 Pt 2), 7067-7076.

	

62	

Choi, C. H., Schoenfeld, B. P., Bell, A. J., Hinchey, P., Kollaros, M., Gertner, M. J., . . .
McBride, S. M. (2011). Pharmacological reversal of synaptic plasticity deficits in the
mouse model of fragile X syndrome by group II mGluR antagonist or lithium treatment.
Brain Res, 1380, 106-119. doi:10.1016/j.brainres.2010.11.032
Christie, S. B., Akins, M. R., Schwob, J. E., & Fallon, J. R. (2009). The FXG: a presynaptic
fragile X granule expressed in a subset of developing brain circuits. J Neurosci, 29(5),
1514-1524. doi:10.1523/JNEUROSCI.3937-08.2009
Chuang, S. C., Bianchi, R., & Wong, R. K. (2000). Group I mGluR activation turns on a voltagegated inward current in hippocampal pyramidal cells. J Neurophysiol, 83(5), 2844-2853.
Chuang, S. C., Zhao, W., Bauchwitz, R., Yan, Q., Bianchi, R., & Wong, R. K. (2005). Prolonged
epileptiform discharges induced by altered group I metabotropic glutamate receptormediated synaptic responses in hippocampal slices of a fragile X mouse model. J
Neurosci, 25(35), 8048-8055. doi:10.1523/JNEUROSCI.1777-05.2005
Citri, A., & Malenka, R. C. (2008). Synaptic plasticity: multiple forms, functions, and
mechanisms. Neuropsychopharmacology, 33(1), 18-41. doi:10.1038/sj.npp.1301559
Comery, T. A., Harris, J. B., Willems, P. J., Oostra, B. A., Irwin, S. A., Weiler, I. J., &
Greenough, W. T. (1997). Abnormal dendritic spines in fragile X knockout mice:
maturation and pruning deficits. Proc Natl Acad Sci U S A, 94(10), 5401-5404.
Conn, P. J., & Pin, J. P. (1997). Pharmacology and functions of metabotropic glutamate
receptors. Annu Rev Pharmacol Toxicol, 37, 205-237.
doi:10.1146/annurev.pharmtox.37.1.205
Cornish, K. M., Munir, F., & Cross, G. (1998). The nature of the spatial deficit in young females
with Fragile-X syndrome: a neuropsychological and molecular perspective.
Neuropsychologia, 36(11), 1239-1246.
Cox, C. L., & Sherman, S. M. (1999). Glutamate inhibits thalamic reticular neurons. J Neurosci,
19(15), 6694-6699.
Crandall, S. R., & Cox, C. L. (2013). Thalamic microcircuits: presynaptic dendrites form two
feedforward inhibitory pathways in thalamus. J Neurophysiol, 110(2), 470-480.
doi:10.1152/jn.00559.2012
Curia, G., Papouin, T., Seguela, P., & Avoli, M. (2009). Downregulation of tonic GABAergic
inhibition in a mouse model of fragile X syndrome. Cereb Cortex, 19(7), 1515-1520.
doi:10.1093/cercor/bhn159
D'Antuono, M., Merlo, D., & Avoli, M. (2003). Involvement of cholinergic and gabaergic
systems in the fragile X knockout mice. Neuroscience, 119(1), 9-13.

	

63	

D'Hulst, C., De Geest, N., Reeve, S. P., Van Dam, D., De Deyn, P. P., Hassan, B. A., & Kooy, R.
F. (2006). Decreased expression of the GABAA receptor in fragile X syndrome. Brain
Res, 1121(1), 238-245. doi:10.1016/j.brainres.2006.08.115
D'Hulst, C., Heulens, I., Brouwer, J. R., Willemsen, R., De Geest, N., Reeve, S. P., . . . Kooy, R.
F. (2009). Expression of the GABAergic system in animal models for fragile X syndrome
and fragile X associated tremor/ataxia syndrome (FXTAS). Brain Res, 1253, 176-183.
doi:10.1016/j.brainres.2008.11.075
D'Hulst, C., & Kooy, R. F. (2007). The GABAA receptor: a novel target for treatment of fragile
X? Trends Neurosci, 30(8), 425-431. doi:10.1016/j.tins.2007.06.003
Dahlhaus, R., & El-Husseini, A. (2010). Altered neuroligin expression is involved in social
deficits in a mouse model of the fragile X syndrome. Behav Brain Res, 208(1), 96-105.
doi:10.1016/j.bbr.2009.11.019
Darnell, J. C., Van Driesche, S. J., Zhang, C., Hung, K. Y., Mele, A., Fraser, C. E., . . . Darnell,
R. B. (2011). FMRP stalls ribosomal translocation on mRNAs linked to synaptic function
and autism. Cell, 146(2), 247-261. doi:10.1016/j.cell.2011.06.013
Daw, N. W., Fox, K., Sato, H., & Czepita, D. (1992). Critical period for monocular deprivation
in the cat visual cortex. J Neurophysiol, 67(1), 197-202.
De Boulle, K., Verkerk, A. J., Reyniers, E., Vits, L., Hendrickx, J., Van Roy, B., . . . Willems, P.
J. (1993). A point mutation in the FMR-1 gene associated with fragile X mental
retardation. Nat Genet, 3(1), 31-35. doi:10.1038/ng0193-31
De Diego Otero, Y., Severijnen, L. A., van Cappellen, G., Schrier, M., Oostra, B., & Willemsen,
R. (2002). Transport of fragile X mental retardation protein via granules in neurites of
PC12 cells. Mol Cell Biol, 22(23), 8332-8341.
Desai, N. S., Casimiro, T. M., Gruber, S. M., & Vanderklish, P. W. (2006). Early postnatal
plasticity in neocortex of Fmr1 knockout mice. J Neurophysiol, 96(4), 1734-1745.
doi:10.1152/jn.00221.2006
Ding, Q., Sethna, F., & Wang, H. (2014). Behavioral analysis of male and female Fmr1 knockout
mice on C57BL/6 background. Behav Brain Res, 271, 72-78.
doi:10.1016/j.bbr.2014.05.046
Doi, A., Ishibashi, H., Jinno, S., Kosaka, T., & Akaike, N. (2002). Presynaptic inhibition of
GABAergic miniature currents by metabotropic glutamate receptor in the rat CNS.
Neuroscience, 109(2), 299-311.
Dolen, G., Osterweil, E., Rao, B. S., Smith, G. B., Auerbach, B. D., Chattarji, S., & Bear, M. F.
(2007). Correction of fragile X syndrome in mice. Neuron, 56(6), 955-962.
doi:10.1016/j.neuron.2007.12.001

	

64	

Fanselow, E. E., & Connors, B. W. (2010). The roles of somatostatin-expressing (GIN) and fastspiking inhibitory interneurons in UP-DOWN states of mouse neocortex. J Neurophysiol,
104(2), 596-606. doi:10.1152/jn.00206.2010
Ferron, L., Nieto-Rostro, M., Cassidy, J. S., & Dolphin, A. C. (2014). Fragile X mental
retardation protein controls synaptic vesicle exocytosis by modulating N-type calcium
channel density. Nat Commun, 5, 3628. doi:10.1038/ncomms4628
Flavin, H. J., Jin, X. T., & Daw, N. W. (2000). 2R,4R-4-Aminopyrrolidine-2,4-dicarboxylate
(APDC) attenuates cortical EPSPs. Brain Res, 873(2), 212-217.
Funauchi, M., Haruta, H., & Tsumoto, T. (1994). Effects of an inhibitor for calcium/calmodulindependent protein phosphatase, calcineurin, on induction of long-term potentiation in rat
visual cortex. Neurosci Res, 19(3), 269-278.
Gantois, I., Vandesompele, J., Speleman, F., Reyniers, E., D'Hooge, R., Severijnen, L. A., . . .
Kooy, R. F. (2006). Expression profiling suggests underexpression of the GABA(A)
receptor subunit delta in the fragile X knockout mouse model. Neurobiol Dis, 21(2), 346357. doi:10.1016/j.nbd.2005.07.017
Gibson, J. R., Bartley, A. F., Hays, S. A., & Huber, K. M. (2008). Imbalance of neocortical
excitation and inhibition and altered UP states reflect network hyperexcitability in the
mouse model of fragile X syndrome. J Neurophysiol, 100(5), 2615-2626.
doi:10.1152/jn.90752.2008
Glickfeld, L. L., Andermann, M. L., Bonin, V., & Reid, R. C. (2013). Cortico-cortical
projections in mouse visual cortex are functionally target specific. Nat Neurosci, 16(2),
219-226. doi:10.1038/nn.3300
Glickfeld, L. L., Histed, M. H., & Maunsell, J. H. (2013). Mouse primary visual cortex is used to
detect both orientation and contrast changes. J Neurosci, 33(50), 19416-19422.
doi:10.1523/JNEUROSCI.3560-13.2013
Goebel-Goody, S. M., Wilson-Wallis, E. D., Royston, S., Tagliatela, S. M., Naegele, J. R., &
Lombroso, P. J. (2012). Genetic manipulation of STEP reverses behavioral abnormalities
in a fragile X syndrome mouse model. Genes Brain Behav, 11(5), 586-600.
doi:10.1111/j.1601-183X.2012.00781.x
Goncalves, J. T., Anstey, J. E., Golshani, P., & Portera-Cailliau, C. (2013). Circuit level defects
in the developing neocortex of Fragile X mice. Nat Neurosci, 16(7), 903-909.
doi:10.1038/nn.3415
Govindaiah, G., & Cox, C. L. (2006). Metabotropic glutamate receptors differentially regulate
GABAergic inhibition in thalamus. J Neurosci, 26(52), 13443-13453.
doi:10.1523/JNEUROSCI.3578-06.2006

	

65	

Grossman, A. W., Elisseou, N. M., McKinney, B. C., & Greenough, W. T. (2006). Hippocampal
pyramidal cells in adult Fmr1 knockout mice exhibit an immature-appearing profile of
dendritic spines. Brain Res, 1084(1), 158-164. doi:10.1016/j.brainres.2006.02.044
Hagerman, P. J. (2008). The fragile X prevalence paradox. J Med Genet, 45(8), 498-499.
doi:10.1136/jmg.2008.059055
Hagerman, P. J., & Stafstrom, C. E. (2009). Origins of epilepsy in fragile X syndrome. Epilepsy
Curr, 9(4), 108-112. doi:10.1111/j.1535-7511.2009.01309.x
Hagerman, R., Hoem, G., & Hagerman, P. (2010). Fragile X and autism: Intertwined at the
molecular level leading to targeted treatments. Mol Autism, 1(1), 12. doi:10.1186/20402392-1-12
Hagerman, R. J., Ono, M. Y., & Hagerman, P. J. (2005). Recent advances in fragile X: a model
for autism and neurodegeneration. Curr Opin Psychiatry, 18(5), 490-496.
doi:10.1097/01.yco.0000179485.39520.b0
Harlow, E. G., Till, S. M., Russell, T. A., Wijetunge, L. S., Kind, P., & Contractor, A. (2010).
Critical period plasticity is disrupted in the barrel cortex of FMR1 knockout mice.
Neuron, 65(3), 385-398. doi:10.1016/j.neuron.2010.01.024
Hays, S. A., Huber, K. M., & Gibson, J. R. (2011). Altered neocortical rhythmic activity states in
Fmr1 KO mice are due to enhanced mGluR5 signaling and involve changes in excitatory
circuitry. J Neurosci, 31(40), 14223-14234. doi:10.1523/JNEUROSCI.3157-11.2011
Henderson, C., Wijetunge, L., Kinoshita, M. N., Shumway, M., Hammond, R. S., Postma, F. R.,
. . . Healy, A. M. (2012). Reversal of disease-related pathologies in the fragile X mouse
model by selective activation of GABAB receptors with arbaclofen. Sci Transl Med,
4(152), 152ra128. doi:10.1126/scitranslmed.3004218
Hensch, T. K. (2005). Critical period plasticity in local cortical circuits. Nat Rev Neurosci, 6(11),
877-888. doi:10.1038/nrn1787
Heulens, I., D'Hulst, C., Van Dam, D., De Deyn, P. P., & Kooy, R. F. (2012). Pharmacological
treatment of fragile X syndrome with GABAergic drugs in a knockout mouse model.
Behav Brain Res, 229(1), 244-249. doi:10.1016/j.bbr.2012.01.031
Hinds, H. L., Ashley, C. T., Sutcliffe, J. S., Nelson, D. L., Warren, S. T., Housman, D. E., &
Schalling, M. (1993). Tissue specific expression of FMR-1 provides evidence for a
functional role in fragile X syndrome. Nat Genet, 3(1), 36-43. doi:10.1038/ng0193-36
Hinton, V. J., Brown, W. T., Wisniewski, K., & Rudelli, R. D. (1991). Analysis of neocortex in
three males with the fragile X syndrome. Am J Med Genet, 41(3), 289-294.
doi:10.1002/ajmg.1320410306

	

66	

Hong, A., Zhang, A., Ke, Y., El Idrissi, A., & Shen, C. H. (2012). Downregulation of GABA(A)
beta subunits is transcriptionally controlled by Fmr1p. J Mol Neurosci, 46(2), 272-275.
doi:10.1007/s12031-011-9531-5
Hovelso, N., Sotty, F., Montezinho, L. P., Pinheiro, P. S., Herrik, K. F., & Mork, A. (2012).
Therapeutic potential of metabotropic glutamate receptor modulators. Curr
Neuropharmacol, 10(1), 12-48. doi:10.2174/157015912799362805
Huber, K. M., Gallagher, S. M., Warren, S. T., & Bear, M. F. (2002). Altered synaptic plasticity
in a mouse model of fragile X mental retardation. Proc Natl Acad Sci U S A, 99(11),
7746-7750. doi:10.1073/pnas.122205699
Ivry, R. (1993). Cerebellar involvement in the explicit representation of temporal information.
Ann N Y Acad Sci, 682, 214-230.
Kang, H., & Schuman, E. M. (1996). A requirement for local protein synthesis in neurotrophininduced hippocampal synaptic plasticity. Science, 273(5280), 1402-1406.
Kasper, E. M., Larkman, A. U., Lubke, J., & Blakemore, C. (1994). Pyramidal neurons in layer 5
of the rat visual cortex. I. Correlation among cell morphology, intrinsic
electrophysiological properties, and axon targets. J Comp Neurol, 339(4), 459-474.
doi:10.1002/cne.903390402
Kates, W. R., Abrams, M. T., Kaufmann, W. E., Breiter, S. N., & Reiss, A. L. (1997). Reliability
and validity of MRI measurement of the amygdala and hippocampus in children with
fragile X syndrome. Psychiatry Res, 75(1), 31-48.
Kato, N. (1993). Dependence of long-term depression on postsynaptic metabotropic glutamate
receptors in visual cortex. Proc Natl Acad Sci U S A, 90(8), 3650-3654.
Katz, L. C., & Shatz, C. J. (1996). Synaptic activity and the construction of cortical circuits.
Science, 274(5290), 1133-1138.
Katzner, S., & Weigelt, S. (2013). Visual cortical networks: of mice and men. Curr Opin
Neurobiol, 23(2), 202-206. doi:10.1016/j.conb.2013.01.019
Kaufmann, W. E., Cortell, R., Kau, A. S., Bukelis, I., Tierney, E., Gray, R. M., . . . Stanard, P.
(2004). Autism spectrum disorder in fragile X syndrome: communication, social
interaction, and specific behaviors. Am J Med Genet A, 129A(3), 225-234.
doi:10.1002/ajmg.a.30229
Kawaguchi, Y., & Kubota, Y. (1997). GABAergic cell subtypes and their synaptic connections
in rat frontal cortex. Cereb Cortex, 7(6), 476-486.

	

67	

Kazdoba, T. M., Leach, P. T., Silverman, J. L., & Crawley, J. N. (2014). Modeling fragile X
syndrome in the Fmr1 knockout mouse. Intractable Rare Dis Res, 3(4), 118-133.
doi:10.5582/irdr.2014.01024
Kinoshita, A., Shigemoto, R., Ohishi, H., van der Putten, H., & Mizuno, N. (1998).
Immunohistochemical localization of metabotropic glutamate receptors, mGluR7a and
mGluR7b, in the central nervous system of the adult rat and mouse: a light and electron
microscopic study. J Comp Neurol, 393(3), 332-352.
Koekkoek, S. K., Yamaguchi, K., Milojkovic, B. A., Dortland, B. R., Ruigrok, T. J., Maex, R., . .
. De Zeeuw, C. I. (2005). Deletion of FMR1 in Purkinje cells enhances parallel fiber
LTD, enlarges spines, and attenuates cerebellar eyelid conditioning in Fragile X
syndrome. Neuron, 47(3), 339-352. doi:10.1016/j.neuron.2005.07.005
Kogan, C. S., Bertone, A., Cornish, K., Boutet, I., Der Kaloustian, V. M., Andermann, E., . . .
Chaudhuri, A. (2004). Integrative cortical dysfunction and pervasive motion perception
deficit in fragile X syndrome. Neurology, 63(9), 1634-1639.
Kogan, C. S., Boutet, I., Cornish, K., Zangenehpour, S., Mullen, K. T., Holden, J. J., . . .
Chaudhuri, A. (2004). Differential impact of the FMR1 gene on visual processing in
fragile X syndrome. Brain, 127(Pt 3), 591-601. doi:10.1093/brain/awh069
Lacoux, C., Di Marino, D., Boyl, P. P., Zalfa, F., Yan, B., Ciotti, M. T., . . . Bagni, C. (2012).
BC1-FMRP interaction is modulated by 2'-O-methylation: RNA-binding activity of the
tudor domain and translational regulation at synapses. Nucleic Acids Res, 40(9), 40864096. doi:10.1093/nar/gkr1254
Laramee, M. E., & Boire, D. (2014). Visual cortical areas of the mouse: comparison of
parcellation and network structure with primates. Front Neural Circuits, 8, 149.
doi:10.3389/fncir.2014.00149
Larkum, M. E., Nevian, T., Sandler, M., Polsky, A., & Schiller, J. (2009). Synaptic integration in
tuft dendrites of layer 5 pyramidal neurons: a new unifying principle. Science, 325(5941),
756-760. doi:10.1126/science.1171958
Larson, J., Jessen, R. E., Kim, D., Fine, A. K., & du Hoffmann, J. (2005). Age-dependent and
selective impairment of long-term potentiation in the anterior piriform cortex of mice
lacking the fragile X mental retardation protein. J Neurosci, 25(41), 9460-9469.
doi:10.1523/JNEUROSCI.2638-05.2005
Lee, C. C., & Sherman, S. M. (2009). Glutamatergic inhibition in sensory neocortex. Cereb
Cortex, 19(10), 2281-2289. doi:10.1093/cercor/bhn246
Li, J., Pelletier, M. R., Perez Velazquez, J. L., & Carlen, P. L. (2002). Reduced cortical synaptic
plasticity and GluR1 expression associated with fragile X mental retardation protein
deficiency. Mol Cell Neurosci, 19(2), 138-151. doi:10.1006/mcne.2001.1085

	

68	

Li, Z., Zhang, Y., Ku, L., Wilkinson, K. D., Warren, S. T., & Feng, Y. (2001). The fragile X
mental retardation protein inhibits translation via interacting with mRNA. Nucleic Acids
Res, 29(11), 2276-2283.
Liu, T., Petrof, I., & Sherman, S. M. (2014). Modulatory effects of activation of metabotropic
glutamate receptors on GABAergic circuits in the mouse cortex. J Neurophysiol, 111(11),
2287-2297. doi:10.1152/jn.00730.2013
Loesch, D. Z., & Hay, D. A. (1988). Clinical features and reproductive patterns in fragile X
female heterozygotes. J Med Genet, 25(6), 407-414.
Lozano, R., Hare, E. B., & Hagerman, R. J. (2014). Modulation of the GABAergic pathway for
the treatment of fragile X syndrome. Neuropsychiatr Dis Treat, 10, 1769-1779.
doi:10.2147/NDT.S42919
Mann, K., Kiefer, F., Spanagel, R., & Littleton, J. (2008). Acamprosate: recent findings and
future research directions. Alcohol Clin Exp Res, 32(7), 1105-1110. doi:10.1111/j.15300277.2008.00690.x
Markram, H., Toledo-Rodriguez, M., Wang, Y., Gupta, A., Silberberg, G., & Wu, C. (2004).
Interneurons of the neocortical inhibitory system. Nat Rev Neurosci, 5(10), 793-807.
doi:10.1038/nrn1519
Martin, H. G., Lassalle, O., & Manzoni, O. J. (2016). Differential Adulthood Onset mGlu5
Signaling Saves Prefrontal Function in the Fragile X Mouse. Cereb Cortex.
doi:10.1093/cercor/bhw328
Martin, H. G. S., Lassalle, O., Brown, J. T., & Manzoni, O. J. (2016). Age-Dependent LongTerm Potentiation Deficits in the Prefrontal Cortex of the Fmr1 Knockout Mouse Model
of Fragile X Syndrome. Cereb Cortex, 26(5), 2084-2092. doi:10.1093/cercor/bhv031
Martin, L. J., Blackstone, C. D., Huganir, R. L., & Price, D. L. (1992). Cellular localization of a
metabotropic glutamate receptor in rat brain. Neuron, 9(2), 259-270.
Martinez-Galan, J. R., Lopez-Bendito, G., Lujan, R., Shigemoto, R., Fairen, A., &
Valdeolmillos, M. (2001). Cajal-Retzius cells in early postnatal mouse cortex selectively
express functional metabotropic glutamate receptors. Eur J Neurosci, 13(6), 1147-1154.
Mason, A., & Larkman, A. (1990). Correlations between morphology and electrophysiology of
pyramidal neurons in slices of rat visual cortex. II. Electrophysiology. J Neurosci, 10(5),
1415-1428.
Mateo, Z., & Porter, J. T. (2007). Group II metabotropic glutamate receptors inhibit glutamate
release at thalamocortical synapses in the developing somatosensory cortex.
Neuroscience, 146(3), 1062-1072. doi:10.1016/j.neuroscience.2007.02.053

	

69	

Mazzocco, M. M., Kates, W. R., Baumgardner, T. L., Freund, L. S., & Reiss, A. L. (1997).
Autistic behaviors among girls with fragile X syndrome. J Autism Dev Disord, 27(4),
415-435.
McNaughton, C. H., Moon, J., Strawderman, M. S., Maclean, K. N., Evans, J., & Strupp, B. J.
(2008). Evidence for social anxiety and impaired social cognition in a mouse model of
fragile X syndrome. Behav Neurosci, 122(2), 293-300. doi:10.1037/0735-7044.122.2.293
Meredith, R. M., Holmgren, C. D., Weidum, M., Burnashev, N., & Mansvelder, H. D. (2007).
Increased threshold for spike-timing-dependent plasticity is caused by unreliable calcium
signaling in mice lacking fragile X gene FMR1. Neuron, 54(4), 627-638.
doi:10.1016/j.neuron.2007.04.028
Merenstein, S. A., Sobesky, W. E., Taylor, A. K., Riddle, J. E., Tran, H. X., & Hagerman, R. J.
(1996). Molecular-clinical correlations in males with an expanded FMR1 mutation. Am J
Med Genet, 64(2), 388-394. doi:10.1002/(SICI)1096-8628(19960809)64:2<388::AIDAJMG31>3.0.CO;2-9
Metin, C., Godement, P., & Imbert, M. (1988). The primary visual cortex in the mouse: receptive
field properties and functional organization. Exp Brain Res, 69(3), 594-612.
Mineur, Y. S., & Crusio, W. E. (2002). Behavioral and neuroanatomical characterization of
FVB/N inbred mice. Brain Res Bull, 57(1), 41-47.
Mineur, Y. S., Sluyter, F., de Wit, S., Oostra, B. A., & Crusio, W. E. (2002). Behavioral and
neuroanatomical characterization of the Fmr1 knockout mouse. Hippocampus, 12(1), 3946. doi:10.1002/hipo.10005
Mostofsky, S. H., Mazzocco, M. M., Aakalu, G., Warsofsky, I. S., Denckla, M. B., & Reiss, A.
L. (1998). Decreased cerebellar posterior vermis size in fragile X syndrome: correlation
with neurocognitive performance. Neurology, 50(1), 121-130.
Munoz, A., Liu, X. B., & Jones, E. G. (1999). Development of metabotropic glutamate receptors
from trigeminal nuclei to barrel cortex in postnatal mouse. J Comp Neurol, 409(4), 549566.
Musumeci, S. A., Bosco, P., Calabrese, G., Bakker, C., De Sarro, G. B., Elia, M., . . . Oostra, B.
A. (2000). Audiogenic seizures susceptibility in transgenic mice with fragile X syndrome.
Epilepsia, 41(1), 19-23.
Nakanishi, S. (1994). Metabotropic glutamate receptors: synaptic transmission, modulation, and
plasticity. Neuron, 13(5), 1031-1037.
Niell, C. M. (2015). Cell types, circuits, and receptive fields in the mouse visual cortex. Annu
Rev Neurosci, 38, 413-431. doi:10.1146/annurev-neuro-071714-033807

	

70	

Niell, C. M., & Stryker, M. P. (2008). Highly selective receptive fields in mouse visual cortex. J
Neurosci, 28(30), 7520-7536. doi:10.1523/JNEUROSCI.0623-08.2008
Nimchinsky, E. A., Oberlander, A. M., & Svoboda, K. (2001). Abnormal development of
dendritic spines in FMR1 knock-out mice. J Neurosci, 21(14), 5139-5146.
Niswender, C. M., & Conn, P. J. (2010). Metabotropic glutamate receptors: physiology,
pharmacology, and disease. Annu Rev Pharmacol Toxicol, 50, 295-322.
doi:10.1146/annurev.pharmtox.011008.145533
Ohishi, H., Neki, A., & Mizuno, N. (1998). Distribution of a metabotropic glutamate receptor,
mGluR2, in the central nervous system of the rat and mouse: an immunohistochemical
study with a monoclonal antibody. Neurosci Res, 30(1), 65-82.
Oliva, A. A., Jr., Jiang, M., Lam, T., Smith, K. L., & Swann, J. W. (2000). Novel hippocampal
interneuronal subtypes identified using transgenic mice that express green fluorescent
protein in GABAergic interneurons. J Neurosci, 20(9), 3354-3368.
Olmos-Serrano, J. L., Paluszkiewicz, S. M., Martin, B. S., Kaufmann, W. E., Corbin, J. G., &
Huntsman, M. M. (2010). Defective GABAergic neurotransmission and pharmacological
rescue of neuronal hyperexcitability in the amygdala in a mouse model of fragile X
syndrome. J Neurosci, 30(29), 9929-9938. doi:10.1523/JNEUROSCI.1714-10.2010
Osterweil, E. K., Chuang, S. C., Chubykin, A. A., Sidorov, M., Bianchi, R., Wong, R. K., &
Bear, M. F. (2013). Lovastatin corrects excess protein synthesis and prevents
epileptogenesis in a mouse model of fragile X syndrome. Neuron, 77(2), 243-250.
doi:10.1016/j.neuron.2012.01.034
Osterweil, E. K., Krueger, D. D., Reinhold, K., & Bear, M. F. (2010). Hypersensitivity to
mGluR5 and ERK1/2 leads to excessive protein synthesis in the hippocampus of a mouse
model of fragile X syndrome. J Neurosci, 30(46), 15616-15627.
doi:10.1523/JNEUROSCI.3888-10.2010
Pacey, L. K., Heximer, S. P., & Hampson, D. R. (2009). Increased GABA(B) receptor-mediated
signaling reduces the susceptibility of fragile X knockout mice to audiogenic seizures.
Mol Pharmacol, 76(1), 18-24. doi:10.1124/mol.109.056127
Paluszkiewicz, S. M., Olmos-Serrano, J. L., Corbin, J. G., & Huntsman, M. M. (2011). Impaired
inhibitory control of cortical synchronization in fragile X syndrome. J Neurophysiol,
106(5), 2264-2272. doi:10.1152/jn.00421.2011
Pan, F., Aldridge, G. M., Greenough, W. T., & Gan, W. B. (2010). Dendritic spine instability
and insensitivity to modulation by sensory experience in a mouse model of fragile X
syndrome. Proc Natl Acad Sci U S A, 107(41), 17768-17773.
doi:10.1073/pnas.1012496107

	

71	

Patel, A. B., Hays, S. A., Bureau, I., Huber, K. M., & Gibson, J. R. (2013). A target cell-specific
role for presynaptic Fmr1 in regulating glutamate release onto neocortical fast-spiking
inhibitory neurons. J Neurosci, 33(6), 2593-2604. doi:10.1523/JNEUROSCI.244712.2013
Paul, K., Venkitaramani, D. V., & Cox, C. L. (2013). Dampened dopamine-mediated
neuromodulation in prefrontal cortex of fragile X mice. J Physiol, 591(4), 1133-1143.
doi:10.1113/jphysiol.2012.241067
Peier, A. M., McIlwain, K. L., Kenneson, A., Warren, S. T., Paylor, R., & Nelson, D. L. (2000).
(Over)correction of FMR1 deficiency with YAC transgenics: behavioral and physical
features. Hum Mol Genet, 9(8), 1145-1159.
Petilla Interneuron Nomenclature, G., Ascoli, G. A., Alonso-Nanclares, L., Anderson, S. A.,
Barrionuevo, G., Benavides-Piccione, R., . . . Yuste, R. (2008). Petilla terminology:
nomenclature of features of GABAergic interneurons of the cerebral cortex. Nat Rev
Neurosci, 9(7), 557-568. doi:10.1038/nrn2402
Petralia, R. S., Wang, Y. X., Niedzielski, A. S., & Wenthold, R. J. (1996). The metabotropic
glutamate receptors, mGluR2 and mGluR3, show unique postsynaptic, presynaptic and
glial localizations. Neuroscience, 71(4), 949-976.
Pfeiffer, B. E., & Huber, K. M. (2009). The state of synapses in fragile X syndrome.
Neuroscientist, 15(5), 549-567. doi:10.1177/1073858409333075
Pietropaolo, S., Guilleminot, A., Martin, B., D'Amato, F. R., & Crusio, W. E. (2011). Geneticbackground modulation of core and variable autistic-like symptoms in Fmr1 knock-out
mice. PLoS One, 6(2), e17073. doi:10.1371/journal.pone.0017073
Pin, J. P., & Duvoisin, R. (1995). The metabotropic glutamate receptors: structure and functions.
Neuropharmacology, 34(1), 1-26.
Priebe, N. J., & McGee, A. W. (2014). Mouse vision as a gateway for understanding how
experience shapes neural circuits. Front Neural Circuits, 8, 123.
doi:10.3389/fncir.2014.00123
Qiu, L. F., Hao, Y. H., Li, Q. Z., & Xiong, Z. Q. (2008). Fragile X syndrome and epilepsy.
Neurosci Bull, 24(5), 338-344. doi:10.1007/s12264-008-1221-0
Reid, S. N., & Daw, N. W. (1997). Activation of metabotropic glutamate receptors has different
effects in different layers of cat visual cortex. Vis Neurosci, 14(1), 83-88.
Reid, S. N., & Romano, C. (2001). Developmental and sensory-dependent changes of group II
metabotropic glutamate receptors. J Comp Neurol, 429(2), 270-276.

	

72	

Reid, S. N., Romano, C., Hughes, T., & Daw, N. W. (1995). Immunohistochemical study of two
phosphoinositide-linked metabotropic glutamate receptors (mGluR1 alpha and mGluR5)
in the cat visual cortex before, during, and after the peak of the critical period for eyespecific connections. J Comp Neurol, 355(3), 470-477. doi:10.1002/cne.903550311
Reid, S. N., Romano, C., Hughes, T., & Daw, N. W. (1997). Developmental and sensorydependent changes of phosphoinositide-linked metabotropic glutamate receptors. J Comp
Neurol, 389(4), 577-583.
Reiss, A. L., Lee, J., & Freund, L. (1994). Neuroanatomy of fragile X syndrome: the temporal
lobe. Neurology, 44(7), 1317-1324.
Ronesi, J. A., & Huber, K. M. (2008). Homer interactions are necessary for metabotropic
glutamate receptor-induced long-term depression and translational activation. J Neurosci,
28(2), 543-547. doi:10.1523/JNEUROSCI.5019-07.2008
Rudelli, R. D., Brown, W. T., Wisniewski, K., Jenkins, E. C., Laure-Kamionowska, M., Connell,
F., & Wisniewski, H. M. (1985). Adult fragile X syndrome. Clinico-neuropathologic
findings. Acta Neuropathol, 67(3-4), 289-295.
Rudy, B., Fishell, G., Lee, S., & Hjerling-Leffler, J. (2011). Three groups of interneurons
account for nearly 100% of neocortical GABAergic neurons. Dev Neurobiol, 71(1), 4561. doi:10.1002/dneu.20853
Sabanov, V., Braat, S., D'Andrea, L., Willemsen, R., Zeidler, S., Rooms, L., . . . Balschun, D.
(2017). Impaired GABAergic inhibition in the hippocampus of Fmr1 knockout mice.
Neuropharmacology, 116, 71-81. doi:10.1016/j.neuropharm.2016.12.010
Sabatini, B. L., Maravall, M., & Svoboda, K. (2001). Ca(2+) signaling in dendritic spines. Curr
Opin Neurobiol, 11(3), 349-356.
Sanderson, T. M., Hogg, E. L., Collingridge, G. L., & Correa, S. A. (2016). Hippocampal
metabotropic glutamate receptor long-term depression in health and disease: focus on
mitogen-activated protein kinase pathways. J Neurochem, 139 Suppl 2, 200-214.
doi:10.1111/jnc.13592
Santoro, M. R., Bray, S. M., & Warren, S. T. (2012). Molecular mechanisms of fragile X
syndrome: a twenty-year perspective. Annu Rev Pathol, 7, 219-245. doi:10.1146/annurevpathol-011811-132457
Scanziani, M., Salin, P. A., Vogt, K. E., Malenka, R. C., & Nicoll, R. A. (1997). Use-dependent
increases in glutamate concentration activate presynaptic metabotropic glutamate
receptors. Nature, 385(6617), 630-634. doi:10.1038/385630a0
Schoepp, D. D., & Conn, P. J. (1993). Metabotropic glutamate receptors in brain function and
pathology. Trends Pharmacol Sci, 14(1), 13-20.

	

73	

Schutt, J., Falley, K., Richter, D., Kreienkamp, H. J., & Kindler, S. (2009). Fragile X mental
retardation protein regulates the levels of scaffold proteins and glutamate receptors in
postsynaptic densities. J Biol Chem, 284(38), 25479-25487.
doi:10.1074/jbc.M109.042663
Selby, L., Zhang, C., & Sun, Q. Q. (2007). Major defects in neocortical GABAergic inhibitory
circuits in mice lacking the fragile X mental retardation protein. Neurosci Lett, 412(3),
227-232. doi:10.1016/j.neulet.2006.11.062
Sherman, S. M. (2014). The function of metabotropic glutamate receptors in thalamus and
cortex. Neuroscientist, 20(2), 136-149. doi:10.1177/1073858413478490
Shigemoto, R., Kinoshita, A., Wada, E., Nomura, S., Ohishi, H., Takada, M., . . . Mizuno, N.
(1997). Differential presynaptic localization of metabotropic glutamate receptor subtypes
in the rat hippocampus. J Neurosci, 17(19), 7503-7522.
Shigemoto, R., Nakanishi, S., & Mizuno, N. (1992). Distribution of the mRNA for a
metabotropic glutamate receptor (mGluR1) in the central nervous system: an in situ
hybridization study in adult and developing rat. J Comp Neurol, 322(1), 121-135.
doi:10.1002/cne.903220110
Silberberg, G., & Markram, H. (2007). Disynaptic inhibition between neocortical pyramidal cells
mediated by Martinotti cells. Neuron, 53(5), 735-746. doi:10.1016/j.neuron.2007.02.012
Silverman, J. L., Pride, M. C., Hayes, J. E., Puhger, K. R., Butler-Struben, H. M., Baker, S., &
Crawley, J. N. (2015). GABAB Receptor Agonist R-Baclofen Reverses Social Deficits
and Reduces Repetitive Behavior in Two Mouse Models of Autism.
Neuropsychopharmacology, 40(9), 2228-2239. doi:10.1038/npp.2015.66
Sinclair, D., Featherstone, R., Naschek, M., Nam, J., Du, A., Wright, S., . . . Siegel, S. J. (2017).
GABA-B Agonist Baclofen Normalizes Auditory-Evoked Neural Oscillations and
Behavioral Deficits in the Fmr1 Knockout Mouse Model of Fragile X Syndrome. eNeuro,
4(1). doi:10.1523/ENEURO.0380-16.2017
Sladeczek, F., Momiyama, A., & Takahashi, T. (1993). Presynaptic inhibitory action of a
metabotropic glutamate receptor agonist on excitatory transmission in visual cortical
neurons. Proc Biol Sci, 253(1338), 297-303. doi:10.1098/rspb.1993.0117
Slegtenhorst-Eegdeman, K. E., de Rooij, D. G., Verhoef-Post, M., van de Kant, H. J., Bakker, C.
E., Oostra, B. A., . . . Themmen, A. P. (1998). Macroorchidism in FMR1 knockout mice
is caused by increased Sertoli cell proliferation during testicular development.
Endocrinology, 139(1), 156-162. doi:10.1210/endo.139.1.5706
Stafstrom, C. E. (2009). Severe epilepsy syndromes of early childhood: the link between genetics
and pathophysiology with a focus on SCN1A mutations. J Child Neurol, 24(8 Suppl),
15S-23S. doi:10.1177/0883073809338152

	

74	

Stern, E. A., Maravall, M., & Svoboda, K. (2001). Rapid development and plasticity of layer 2/3
maps in rat barrel cortex in vivo. Neuron, 31(2), 305-315.
Su, T., Fan, H. X., Jiang, T., Sun, W. W., Den, W. Y., Gao, M. M., . . . Yi, Y. H. (2011). Early
continuous inhibition of group 1 mGlu signaling partially rescues dendritic spine
abnormalities in the Fmr1 knockout mouse model for fragile X syndrome.
Psychopharmacology (Berl), 215(2), 291-300. doi:10.1007/s00213-010-2130-2
Szabadics, J., Lorincz, A., & Tamas, G. (2001). Beta and gamma frequency synchronization by
dendritic gabaergic synapses and gap junctions in a network of cortical interneurons. J
Neurosci, 21(15), 5824-5831.
Tahvildari, B., Wolfel, M., Duque, A., & McCormick, D. A. (2012). Selective functional
interactions between excitatory and inhibitory cortical neurons and differential
contribution to persistent activity of the slow oscillation. J Neurosci, 32(35), 1216512179. doi:10.1523/JNEUROSCI.1181-12.2012
Tai, C., Abe, Y., Westenbroek, R. E., Scheuer, T., & Catterall, W. A. (2014). Impaired
excitability of somatostatin- and parvalbumin-expressing cortical interneurons in a mouse
model of Dravet syndrome. Proc Natl Acad Sci U S A, 111(30), E3139-3148.
doi:10.1073/pnas.1411131111
Tamaru, Y., Nomura, S., Mizuno, N., & Shigemoto, R. (2001). Distribution of metabotropic
glutamate receptor mGluR3 in the mouse CNS: differential location relative to pre- and
postsynaptic sites. Neuroscience, 106(3), 481-503.
Tanabe, Y., Masu, M., Ishii, T., Shigemoto, R., & Nakanishi, S. (1992). A family of
metabotropic glutamate receptors. Neuron, 8(1), 169-179.
Tassone, F., Hagerman, R. J., Loesch, D. Z., Lachiewicz, A., Taylor, A. K., & Hagerman, P. J.
(2000). Fragile X males with unmethylated, full mutation trinucleotide repeat expansions
have elevated levels of FMR1 messenger RNA. Am J Med Genet, 94(3), 232-236.
Taylor, D. L., Jones, F., Kubota, E. S., & Pocock, J. M. (2005). Stimulation of microglial
metabotropic glutamate receptor mGlu2 triggers tumor necrosis factor alpha-induced
neurotoxicity in concert with microglial-derived Fas ligand. J Neurosci, 25(11), 29522964. doi:10.1523/JNEUROSCI.4456-04.2005
Thomas, A. M., Bui, N., Graham, D., Perkins, J. R., Yuva-Paylor, L. A., & Paylor, R. (2011).
Genetic reduction of group 1 metabotropic glutamate receptors alters select behaviors in a
mouse model for fragile X syndrome. Behav Brain Res, 223(2), 310-321.
doi:10.1016/j.bbr.2011.04.049
Thomas, A. M., Bui, N., Perkins, J. R., Yuva-Paylor, L. A., & Paylor, R. (2012). Group I
metabotropic glutamate receptor antagonists alter select behaviors in a mouse model for

	

75	

fragile X syndrome. Psychopharmacology (Berl), 219(1), 47-58. doi:10.1007/s00213011-2375-4
Till, S. M., Wijetunge, L. S., Seidel, V. G., Harlow, E., Wright, A. K., Bagni, C., . . . Kind, P. C.
(2012). Altered maturation of the primary somatosensory cortex in a mouse model of
fragile X syndrome. Hum Mol Genet, 21(10), 2143-2156. doi:10.1093/hmg/dds030
Turk, J., & Cornish, K. (1998). Face recognition and emotion perception in boys with fragile-X
syndrome. J Intellect Disabil Res, 42 ( Pt 6), 490-499.
Veeraragavan, S., Bui, N., Perkins, J. R., Yuva-Paylor, L. A., Carpenter, R. L., & Paylor, R.
(2011). Modulation of behavioral phenotypes by a muscarinic M1 antagonist in a mouse
model of fragile X syndrome. Psychopharmacology (Berl), 217(1), 143-151.
doi:10.1007/s00213-011-2276-6
Vislay, R. L., Martin, B. S., Olmos-Serrano, J. L., Kratovac, S., Nelson, D. L., Corbin, J. G., &
Huntsman, M. M. (2013). Homeostatic responses fail to correct defective amygdala
inhibitory circuit maturation in fragile X syndrome. J Neurosci, 33(17), 7548-7558.
doi:10.1523/JNEUROSCI.2764-12.2013
Wahlstrom-Helgren, S., & Klyachko, V. A. (2015). GABAB receptor-mediated feed-forward
circuit dysfunction in the mouse model of fragile X syndrome. J Physiol, 593(22), 50095024. doi:10.1113/JP271190
Wakita, M., Oyama, Y., Takase, Y., & Akaike, N. (2015). Modulation of excitatory synaptic
transmission in rat hippocampal CA3 neurons by triphenyltin, an environmental
pollutant. Chemosphere, 120, 598-607. doi:10.1016/j.chemosphere.2014.09.073
Wang, Q., Gao, E., & Burkhalter, A. (2011). Gateways of ventral and dorsal streams in mouse
visual cortex. J Neurosci, 31(5), 1905-1918. doi:10.1523/JNEUROSCI.3488-10.2011
Wang, Q., Sporns, O., & Burkhalter, A. (2012). Network analysis of corticocortical connections
reveals ventral and dorsal processing streams in mouse visual cortex. J Neurosci, 32(13),
4386-4399. doi:10.1523/JNEUROSCI.6063-11.201
Wang, L. W., Berry-Kravis, E., & Hagerman, R. J. (2010). Fragile X: leading the way for
targeted treatments in autism. Neurotherapeutics, 7(3), 264-274.
doi:10.1016/j.nurt.2010.05.005
Weiler, I. J., & Greenough, W. T. (1993). Metabotropic glutamate receptors trigger postsynaptic
protein synthesis. Proc Natl Acad Sci U S A, 90(15), 7168-7171.
Weiler, I. J., Irwin, S. A., Klintsova, A. Y., Spencer, C. M., Brazelton, A. D., Miyashiro, K., . . .
Greenough, W. T. (1997). Fragile X mental retardation protein is translated near synapses
in response to neurotransmitter activation. Proc Natl Acad Sci U S A, 94(10), 5395-5400.

	

76	

Wilson, B. M., & Cox, C. L. (2007). Absence of metabotropic glutamate receptor-mediated
plasticity in the neocortex of fragile X mice. Proc Natl Acad Sci U S A, 104(7), 24542459. doi:10.1073/pnas.0610875104
Yan, Q. J., Asafo-Adjei, P. K., Arnold, H. M., Brown, R. E., & Bauchwitz, R. P. (2004). A
phenotypic and molecular characterization of the fmr1-tm1Cgr fragile X mouse. Genes
Brain Behav, 3(6), 337-359. doi:10.1111/j.1601-183X.2004.00087.x
Yan, Q. J., Rammal, M., Tranfaglia, M., & Bauchwitz, R. P. (2005). Suppression of two major
Fragile X Syndrome mouse model phenotypes by the mGluR5 antagonist MPEP.
Neuropharmacology, 49(7), 1053-1066. doi:10.1016/j.neuropharm.2005.06.004
Zhang, Y., Bonnan, A., Bony, G., Ferezou, I., Pietropaolo, S., Ginger, M., . . . Frick, A. (2014).
Dendritic channelopathies contribute to neocortical and sensory hyperexcitability in
Fmr1(-/y) mice. Nat Neurosci, 17(12), 1701-1709. doi:10.1038/nn.3864
Zhao, M. G., Toyoda, H., Ko, S. W., Ding, H. K., Wu, L. J., & Zhuo, M. (2005). Deficits in trace
fear memory and long-term potentiation in a mouse model for fragile X syndrome. J
Neurosci, 25(32), 7385-7392. doi:10.1523/JNEUROSCI.1520-05.2005
Zou, J., Jiang, J. Y., & Yang, J. J. (2017). Molecular Basis for Modulation of Metabotropic
Glutamate Receptors and Their Drug Actions by Extracellular Ca2. Int J Mol Sci, 18(3).
doi:10.3390/ijms18030672

	

77	

CHAPTER II: MGLUR-MEDIATED MODULATION OF VISUAL NEOCORTICAL
PYRAMIDAL CELL EXCITATION IN A FRAGILE X MOUSE
2.1 Introduction
Fragile X syndrome is a heritable disease of intellectual disability, and also the most
common genetic cause of autism (Kaufman et al., 2004). It affects 1 in 4000 males and 1 in
8000 females (Hagerman et al., 2008), and causes moderate to severe intellectual impairment in
all full mutation carriers. FXS also leads to a variety of neurological and psychiatric clinical
manifestations include hypersensitivity to tactile stimuli, anxiety, mood instability, sleep
disturbances, seizure disorder, and autistic features (Huber et al., 2002). Development of FXS
occurs when the FMR1 gene, located on the X chromosome, is hypermethylated due to
replication of CGG repeats (>200) in the 5’ untranslated region. Hypermethylation of the gene
subsequently leads to little or no production of the gene product FMRP. FMRP is expressed
throughout most neurons, and is known to localize abundantly in dendrites, where it serves as a
negative regulator of protein translation at the synapse (Antar et al., 2004). It has been shown to
affect long term changes in synaptic efficacy, including long term potentiation (LTP) and long
term depression (LTD). Exaggerated LTD, mediated by group I metabotropic glutamate
receptors (mGluRs), in the Fmr1 KO hippocampus and cerebellum is one of the hallmark
pathophysiological changes observed in the Fmr1 KO mouse (Huber et al., 2002). These
plasticity studies led to the hypothesis that loss of FMRP leads to exaggeration of group I mGluR
signaling, and this is the likely mediator of several seemingly different symptoms in FXS.
Subsequently, over a decade of studies have shown that reduction of group I mGluR activation,
both genetically and pharmacologically, leads to the restoration of hippocampal-mediated
learning in mice on a variety of behavioral assessments (Dolen et al., 2007), as well as recovery

	

78	

of abnormal synaptic morphology in the neocortex. Group I mGluRs have also been associated
with epileptogenic activity in the Fmr1 KO mouse. In mice, loss of FMRP leads to enhanced
epileptogenic activity in hippocampal slices, and this activity is mediated by overactive group I
mGluRs (Chuang et al., 2005). Taken together, these studies have built strong scientific support
in favor of the role of group I mGluRs in the pathogenesis of Fragile X syndrome.
However, the extent to which group I mGluRs, and for that matter other mGluRs,
mediate pathological neocortical mechanisms in Fragile X is currently unknown. Although
several neocortical studies have demonstrated that loss of FMRP leads to altered synaptic
plasticity it does so in a way that differs from that seen in the hippocampus and cerebellum.
Thus, it is highly likely that the pathophysiological mechanisms plaguing the hippocampus, do
not directly translate to other brain regions.
Understanding that a gap in knowledge exists regarding the role of neocortical mGluRs in
the pathophysiology of FXS, I began my investigation of metabotropic receptor modulation in
the visual neocortex of Fmr1 KO mice. Previous studies have shown that the dendritic spine
morphology in layer 5/6 visual neocortical neurons of the Fmr1 KO mouse, mimic the immature
spine morphology seen in Golgi-impregnated stains of human visual neocortical spines (Hinton
et al., 1991; Irwin, Galvez, & Greenough, 2000; Irwin et al., 2001; Rudelli et al., 1985); this
immature spine morphology can be rescued with genetic reduction of group I mGluRs in the
Fmr1 KO mouse (Dolen 2007). Further, published findings from our lab have shown that group
I mGluR-mediated synaptic plasticity in layer 5/6 visual neocortical neurons is absent in the
Fmr1 KO mouse (Wilson & Cox, 2007). With this in mind, I hypothesized that group I mGluRmediated modulation of layer 5/6 neocortical microcircuits are altered in FXS. Additionally,
given the ever growing support of FMRP as a regulator of presynaptic mRNA translation (Akins

	

79	

2009), I also investigated the effects of group II metabotropic glutamate receptors, as they are
well-documented to exist abundantly at presynaptic terminals (Anwyl et al., 1999). Here, I show
that group I and group II mGluRs modulate the postsynaptic membrane potential of visual
neocortical pyramidal neurons, and that both of these glutamate receptors regulate synaptic
transmission in the layer 5/6. I then determine that mGluR-mediated modulation of excitatory
transmission onto pyramidal neurons is intact in the Fmr1 KO mouse in both layer 5/6 and 2/3 of
the primary visual neocortex.
2.2 Materials and Methods
All experimental procedures were performed in accordance with the National Institutes of
Health Guide for Care and Use if Laboratory Animals and were approved by the Michigan State
University Institutional Animal Care and Use Committee.
Mice on the FVB background (postnatal age 16-21 days) of either sex were anesthetized
with isoflurane and perfused with cold, oxygenated (5% CO2-95% O2), modified saline prior to
decapitation. Brains were removed and transferred to cold (4°C), oxygenated slicing solution
containing the following (in mM): 2.5 KCl, 1.25 NaH2PO4, 10.0 MgSO4, 0.5 CaCl2, 26.0
NaHCO3, 10.0 glucose, and 234.0 sucrose. Slices (300 µm thickness) were collected in a
coronal plane were collected using a Leica vibrating tissue slicer, and then transferred to a
holding chamber filled with oxygenated physiological saline containing (in mM): 126.0 NaCl,
2.5 KCl, 1.25 NaH2PO4, 2.0 MgCl2, 2.0 CaCl2, 26.0 NaHCO3, and 10.0 glucose. Slices were
incubated at 32 ± 1°C for at least 30 minutes, and then left to rest at room temperature for at least
30 minutes prior to recording.
Whole-cell intracellular recordings were obtained using a Zeiss microscope equipped
with differential interface contrast optics (Zeiss, Thornwood, NY). Single slices were transferred

	

80	

to a submersion chamber continuously superfused at 2.5ml/min with oxygenated physiological
saline heated to 32◦C. Visual neocortex was identified with a low power (5x) objective and
individual pyramidal neurons were identified using a water-immersion objective (63x).
Pyramidal neurons were targeted by visual identification of pyramidal-shaped soma with a
diameter greater than 20 µm, and confirmed by their characteristic electrophysiological
properties as determined using an 800-2000 ms current step injection protocol. The majority of
pyramidal neurons (62%) exhibit a regular spiking (RS) slowly adapting firing pattern, with a
moderate sized late after-hyperpolarizing potential (AHP) and most often no visible early (AHP)
or depolarizing afterpotential (DAP) (Chang and Luebke, 2007; Larkman and Mason; 1990).
Another less common (5%), but easily distinguished electrophysiological profile of layer 5/6
pyramids was that of intrinsic bursting firing (IB) of 3-6 bursts of APs (Chang and Luebke,
2007; Larkman and Mason; 1990). For recordings, pipettes made of borosilicate capillary glass
were pulled to tip resistances of 4-6MΩ and then filled with intracellular solution containing (in
mM): 117.0 K-gluconate, 13.0 KCl, 1.0 MgCl2, 0.07 CaCl2, 0.1 EGTA, 10.0 HEPES, 2.0 Na2ATP, 0.4 Na-GTP, and 0.2% biocytin. Only cells with a membrane potential ≥ -50 mV and
overshooting action potentials were included for analysis. Access resistance was monitored
throughout the experiments and recordings were not included if the access resistance increased to
>20MΩ.
A. Membrane Potential Drug Application Protocol
For the membrane studies, membrane potential was recorded in current clamp mode. A
series of metabotropic glutamate agonists were applied via bolus injection to determine the
effects of mGluRs on the postsynaptic cell membrane potential. First, a general mGluR agonist
(1S,3R)-1-Aminocyclopentane-1,3-dicarboxylic acid (ACPD) was applied (75 µM, 40 s). Upon

	

81	

complete washout of the drug, specific mGluR agonists were applied for group I or group II
mGluRs. The order of drug application of the specific agonists was varied to prevent convoluted
drug effects. (RS)-3,5-dihydroxyphenylglycine (DHPG), a group I mGluR agonist, was applied
for 30 seconds at 50 µM and allowed to completely washout before application of group II
mGluR agonist, (2R,4R)-4-aminopyrrolidine-2,4-dicarboxylate (APDC, 75µM, 40 s). All
recordings were done in the presence of tetrodotoxin (1µM) to isolate the postsynaptic response.
B. Stimulation Protocol
In layer 5/6 neuron recordings, excitatory postsynaptic currents (EPSCs) were evoked by
electrical stimulation of afferents in layer 2/3 (Figure 2.1, left panel). For layer 2/3 recordings,
a stimulating electrode was placed in layer 4 of the visual neocortex (Figure 2.1, right panel).
For EPSC recordings, the chloride channel blocker, 4,4'-dinitrostilbene-2,2'-disulfonic acid
(DNDS,1 mM) was included in the recording pipette to block postsynaptic chloride channels.
GABAA receptors are chloride channels, and so DNDS blocks inhibitory postsynaptic currents
(IPSCs) generated from these channels. We are thus able to effectively isolate EPSCs in the
recorded neurons. Stimuli were delivered at 0.1 Hz, and the peak amplitudes of the evoked EPSC
was measured before, during, and after washout of the selective group I and group II mGluR
agonists, (RS)-3,5-dihydroxyphenylglycine (DHPG) and (2R,4R)-4-aminopyrrolidine-2,4dicarboxylate (APDC) respectively.
All compounds used in this study were prepared from stock solutions and diluted to the final
concentration with physiological saline prior to administration. All agonists were delivered by
bolus injection to the central line of the recording chamber using a syringe pump. Based upon
the calculated flow rate of the main line, the delivered drug concentration was estimated to be

	

82	

one-fourth of the initial drug concentration. DNDS was purchased from Life Technologies,
while all other compounds were purchased from Tocris and Sigma-Aldrich.
All recordings were acquired using a Multiclamp 700B amplifier (Molecular Devices). Data
was filtered at 2.5 kHz and digitized at 10 kHz using pClamp software. For membrane potential
experiments, 12 consecutive sweeps were averaged to determine the membrane potential at
baseline and 3 consecutive sweeps were average during the maximal depolarizing or
hyperpolarizing drug response. A positive drug response was defined as a change in membrane
potential that was greater than 2 standard deviations of the baseline response. For evoked
experiments, 12 consecutive sweeps (2 minutes) were averaged to determine the baseline peak
amplitude of the synaptic response. The drug effect was determined as the average of 3
consecutive sweeps during the maximal drug response. A positive drug response was defined as
a change in amplitude that was greater than 2 standard deviations of the baseline response. All
data are reported as the mean ± SEM and analyzed using either paired or independent t-test were
applicable. A p-value <0.05 was considered statistically significant.
2.3 Results
A. The intrinsic properties of Fmr1 KO pyramidal neurons are similar to wild type in
layer 5/6 of the visual neocortex.
Prior to exploring the synaptic integrity of the visual neocortex in Fragile X Syndrome,
we first set out to determine whether any differences in baseline intrinsic properties exist
between WT and Fmr1 KO mice pyramidal neurons. We performed current step protocols in
layer 5/6 WT (Figure 2.2Ai) and Fmr1 KO (Figure 2.2Aii) pyramidal neurons and looked at the
current-voltage relationships to determine the input resistance. We also measured the resting
membrane potential prior to current step injections. In control mice, layer 5/6 visual neocortical

	

83	

pyramidal neurons have an average input resistance of 144.6 ± 10.8 MW (n=48; Figure 2.2B).
Fmr1 KO animals, showed a small increase in input resistance (175.149 ± 16.4 MW, n=47,
Figure 2.2B), however, this increase was determined not to be significant (p > 0.05, t-test,
Figure 2.2B). Next, we quantified the membrane potential of these neurons. In WT conditions,
the average membrane potential at rest was -73.8 ± 0.5 mV (n=48, Figure 2.2C). Fmr1 KO
pyramidal neuron input resistance was -74.3 ± 0.7 mV, and this was statistically similar to layer
5/6 WT pyramidal cells (p> 0.05, independent t-test, Figure 2.2C). These results are
summarized in the following table (Table 2.1).
Table 2.1. Intrinsic properties of layer 5/6 pyramidal neurons in WT and Fmr1 KO mice.

Input Resistance (MΩ)

Membrane Potential
(mV)

Fmr1 KO

144.6 ± 10.8
(n=48)
175.1 ± 16.4
(n=47)

-73.8 ± 0.5
(n=48)
-74.3 ± 0.7
(n=47)

p

0.13
t-test

0.58
t-test

Wild Type

B. mGluR-mediated modulation of pyramidal neuron membrane potential is intact in the
Fmr1 KO mouse.
We next asked how pharmacological activation of metabotropic glutamate receptors
could modulate membrane potential. Specifically, we were interested in quantifying changes in
the membrane potential of layer 5/6 pyramidal neurons in response to brief pharmacological
activation of both general and specific mGluR agonists. Application of the general agonist,
ACPD (75µM, 40s) in the presence of tetrodotoxin (TTX, 1µM) led to either depolarizations or
hyperpolarizations in both WT and Fmr1 KO pyramidal neurons (Figure 2.3 Ai-ii).

In WT

neurons, 11 of 17 cells depolarized in the presence of ACPD, with the average depolarizing
response amplitude measuring 3.3 ± 0.8 mV (Figure 2.3Ai and 2.3B). In Fmr1 KO mice,
	

84	

pyramidal neuron depolarizations were of similar amplitude to WT neurons (8 of 20 cells, 3.3 ±
0.5 mV, p >0.05 t-test, Figure 2.3Aii and 2.3B). Hyperpolarizations to ACPD in WT neurons
were larger in amplitude (6 of 17 cells, 7.1 ± 1.5 mV) on average than WT depolarizations but
were seen less often than hyperpolarizations (Figure 2.3B). This is in contrast to Fmr1 KO
neurons, where hyperpolarizations were more likely to be observed (11 of 20 cells). However,
the amplitude of these hyperpolarizations (6.6 ± 1.2 mV) was not significantly different than WT
cells (p>.05, independent t-test, (Figure 2.3B).
Next, to determine the specific metabotropic receptors underlying the variable responses
to ACPD, we applied specific agonists of both group I and group II mGluRs (Figure 2.3Ai-ii).
Brief application of the specific group I mGluR agonist, (RS)-3,5-Dihydroxyphenylglycine
(DHPG, 50 µM, 30s) led to depolarizations in both the WT (n=13) and Fmr1 KO (n=16)
pyramidal neurons (Figure 2.3 Ai-ii). Comparison of response amplitudes across the two
genotypes led to similar results (WT: 4.8 ± 0.8 mV, Fmr1 KO: 4.5 ± 0.9 mV, p>.05, independent
t-test, Figure 2.3 B). Interestingly, in the Fmr1 KO mouse 5 cells failed to show a response to
DHPG although all WT neurons had a significant drug response to DHPG.
Finally, to confirm the metabotropic source of the ACPD-induced membrane
hyperpolarizations in these pyramidal neurons, we briefly applied specific group II mGluR
agonist, (2R,4R)-4-Aminopyrrolidine-2,4-dicarboxylate (APDC, 75 µM, 40s). In the WT mouse,
neurons hyperpolarized with a response amplitude of 2.4 ± 0.6 mV (n=11, Figure 2.3 Ai). The
Fmr1 KO pyramidal cells also hyperpolarized to APDC application, with an average response of
3.4 ± 0.7mV (n=15, Figure 2.3 Aii). Similar to our observations with group I agonists,
application of APDC to Fmr1 KO neurons led to a population of cells that failed to respond
(n=6). In contrast, all WT cells tested had a significant drug response. These results determined

	

85	

that group I and group II mGluRs postsynaptically alter the membrane potential of layer 5/6
pyramidal neurons. However, these postsynaptic effects of mGluRs are unaltered in the Fmr1
KO. A summary of these results is reported in the table below.
Table 2.2. Layer 5/6 pyramidal neuron membrane potential responses to application of
general mGluR agonist (ACPD) and specific mGluR I (DHPG) and II (APDC) agonists.

ACPD

DHPG

APDC

WT

KO

WT

KO

WT

KO

Depolarizations

65%
(11/17)

40%
(8/20)

100%
(13/13)

76%
(16/21)

0%
(0/11)

0%
(0/21)

Hyperpolarizations

35%
(6/17)

55%
(11/20)

0%
(0/13)

0%
(0/21)

100%
(11/11)

71%
(15/21)

No change

0%
(0/17)

5%
(1/20)

0%
(0/13)

24%
(5/21)

0%
(0/11)

29%
(6/21)

C. Group I mGluRs do not modulate excitatory synaptic transmission onto visual neocortical
pyramidal neurons in layer 5/6.
After determining whether intrinsic properties are affected in FXS, we next sought to
investigate the mGluR-mediated modulation of synaptic transmission in layer 5/6 visual
neocortical pyramidal neurons. To that end, we recorded evoked excitatory currents (EPSCs)
from pyramidal neurons in both WT (Figure 2.4 Ai) and Fmr1 KO mice (Figure 2.4 Bi). We
then quantified peak amplitude in the presence of specific mGluR agonist DHPG (25 µM, 30s).
The baseline amplitude of evoked EPSCs in the WT was 181.5 ± 28.2 pA, and application of
DHPG did not significantly alter EPSC amplitude (n=24, DHPG: 150.7 ± 21.8 pA, p>0.05,
paired t-test, Figure 2.4 Aii-iii). In Fmr1 KO neurons, DHPG also failed to alter EPSC
amplitude (n=21, Pre: 230.8 ± 26.1 pA, DHPG: 210.4 ± 25.2 pA, p>0.05, paired t-test). Figure
2.4 Bii-iii). A comparison of the average EPSC amplitude response to DHPG between WT and

	

86	

Fmr1 KO was not statistically significant. (WT: 86.8 ± 2.9 %, KO: 90.1 ± 3.0 %, p > 0.05, t-test,
Figure 2.4 C).
D. Group II mGluR activation leads to suppression of excitatory synaptic transmission onto
layer 5/6 visual neocortical pyramidal neurons in the WT and Fmr1 KO mouse.
The second mGluR agonist we tested, APDC (75µM, 40s), also decreased EPSC amplitude
in WT (Figure 2.5 Ai-iii) and Fmr1 KO (Figure 2.5 Bi-iii) pyramidal neurons, albeit the
reduction was larger in magnitude than the decrease we observed upon activating group I
mGluRs. In WT mice, the average baseline amplitude of the EPSC was 248.7 ± 28.7 pA (n=16,
Figure 2.5 C), and APDC decreased this peak amplitude to 80.1 ± 9.8 pA. Fmr1 KO neurons
exhibited a similar decrease in EPSC amplitude that was not statistically different from control
neurons (Pre: 221.4 ± 31.2 pA, APDC: 77.1 ± 13.3 pA, p>0.05, t-test, Figure 2.5 C).
E. The intrinsic properties of Fmr1 KO pyramidal neurons are similar to wild type in
layer 2/3 of the visual neocortex.
We next recorded from layer 2/3 WT (Figure 2.6 Ai) and Fmr1 KO (Figure 2.6 Aii)
pyramidal neurons and quantified their input resistance and membrane potential from a currentstep protocol. In the layer 2/3 of the visual neocortex, wild type pyramidal neurons had an
increased input resistance relative to layer 5/6 WT pyramidal neurons (279.0 ± 32.2 MW, n=9,
Figure 2.6B). A similar trend was observed in layer 2/3 Fmr1 KO pyramidal neurons when
compared to Fmr1 KO layer 5/6 neurons (209.1 ± 30.2 MW, n=10, Figure 2.6 B). No notable
differences were observed between WT and Fmr1 KO layer 2/3 pyramidal neurons (p>0.05, ttest, Figure 2.6 B). The membrane potential of the superficial pyramidal neurons was more
hyperpolarized relative to the layer 5/6 pyramidal neurons in both WT (n=9) and Fmr1 KO
(n=10) animals (WT: -83.5 ± 1.3 mV, Fmr1 KO: -87.1 ± 1.4 mV, Figure 2.6C). Compared to

	

87	

WT cells, the membrane potential observed in the Fmr1 KO neurons was more hyperpolarized,
and this was just outside of significance (p=0.07, t-test, Figure 2.6 C). For a summary of these
results, see Table 2.3.
Table 2.3. Intrinsic properties of layer 2/3 pyramidal neurons in WT and Fmr1 KO mice.

Input Resistance (MΩ)

Membrane Potential mV

Fmr1 KO

279.7 ± 32.2
(n=9)
209.1 ± 30.2
(n=10)

-83.5 ± 1.3
(n=9)
-87.1 ± 1.4
(n=10)

p

0.13
t-test

0.07
t-test

Wild Type

F. Group I mGluRs do not modulate excitatory synaptic transmission onto visual
neocortical pyramidal neurons in layer 2/3.
To determine whether the mGluR-mediated modulation of synaptic transmission we
observed was unique to layer 5/6 pyramidal neurons, we next tested whether group I mGluRs
modulate glutamatergic synaptic transmission in layer2/3 pyramidal neurons of WT and Fmr1
KO mice (Figure 2.7 Ai and Bi). We recorded electrically evoked EPSCs in layer 2/3 of the
visual neocortex with an electrode placed in layer 4 (Figure 2.1). The average amplitude of the
evoked response in wild type animals was 159.0 ± 17.7 pA (n=5), and brief application of
specific mGluR I agonist DHPG (25 µM, 30s) reduced the peak amplitude of the response to
147.0 ± 17.9 pA (n=7, Figure 2.7Aii-iii). However, this decrease was not statistically significant
(p>0.05, paired t-test). A similar trend was observed in the Fmr1 KO mouse. The average predrug amplitude in the Fmr1 KO was 333.0 ± 148.5 pA (n=8, Figure 2.7Bii-iii) and DHPG
significantly reduced the EPSC amplitude (311.7 ± 138.8 pA, n=8, p < 0.05, paired t-test). We
conclude that group I mGluRs do not modulate synaptic transmission in the layer 2/3.

	

88	

G. Group II mGluR activation leads to suppression of excitatory synaptic transmission onto
layer 2/3 visual neocortical pyramidal neurons in the WT and Fmr1 KO mouse.
In the case of specific group II mGluR agonist APDC, both WT and Fmr1 KO pyramidal
neurons EPSC were responsive (2.8 Ai and Bi). In WT neurons, pre-drug EPSC amplitude was
160.8 ± 35.4 pA (n=8) and application of APDC (75µM, 40s) produced a significant decrease in
the response (41.3 ± 8.5 pA, n=8, p <0.05, paired t-test, Figure 2.8ii-iii). For the Fmr1 KO,
average pre-drug amplitude of the EPSC was 350.8 ± 65.2 pA (n=7) and the post-drug response
was significantly reduced (109.5 ± 30.3 pA, n=7, p <0.05, paired t-test, Figure 2.8 Bii-iii).
Statistical analysis determined that there were no significant differences in WT vs Fmr1 KO drug
response (p>0.05, t-test, Figure 2.8 C).
2.4 Discussion
In our studies, we show that membrane potential and input resistance are similar in WT and
Fmr1 KO pyramidal neurons of the visual neocortex. A few studies have shown that juvenile
layer 4 regular-spiking pyramidal neurons of the somatosensory cortex exhibit increased input
resistance relative to wild type controls, with no difference in membrane potential (Zhang et al.,
2016; Gibson et al., 2008). However, another study done in layer 2/3 determined that there were
no changes in input resistance or membrane potential in pyramidal cells of the medial pre-frontal
cortex (Meredith et al., 2007). We also report that neither the membrane potential or input
resistance of pyramidal neurons in layer 5/6 or 2/3 are altered in the neocortex of the Fmr1 KO
mouse. While it may still be possible that pyramidal neuron intrinsic excitability is a
contributing factor leading to visual neocortical hyperexcitability in FXS, we indicate that input
resistance and membrane potential are intrinsic properties of these excitatory neurons that do not
contribute to visual neocortical microcircuit alterations in the Fmr1 KO.

	

89	

We go on to demonstrate that activation of group I mGluRs lead to depolarizations of layer
5/6 pyramidal neuron postsynaptic membrane potentials. The magnitude of these depolarizations
is normal in the Fmr1 KO mouse. Although, previous literature suggests that group I mGluRmediated postsynaptic LTD is exaggerated in the Fmr1 KO mouse hippocampus (Huber et al.,
2002), our data show that not all mGluR-mediated postsynaptic mechanisms are altered in the
Fmr1 KO. Thus we provide evidence of regional specificity of group I mGluRs in mediating
dysfunction in the absence of FMRP. We also showed that group II mGluRs hyperpolarize the
postsynaptic pyramidal cell membrane, and that this function is intact in the Fmr1 KO mouse.
Interestingly, in our membrane studies we found that a population of layer 5/6 pyramidal neurons
in the Fmr1 KO mouse were not responsive to the pharmacological glutamate receptor agonists
that we used, even though all the WT cells we tested responded to the pharmacological agents.
Any number of reasons could be proposed to explain these results, including impaired binding at
mGluRs, preferential binding of the agonist to other neuron subtypes, or better clearing of
mGluR agonist at the synapse in Fmr1 KO animals.
We also show that group I mGluR agonist DHPG does not modulate fast excitatory
transmission onto layer 5/6 pyramidal neurons in either wild type or Fmr1 KO mice. In a
minority of cells tested, DHPG was seen to reduce excitatory transmission by less than 25%,
however the majority of cells tested showed no change. This is interesting given that DHPG
yields long term depression in several brain regions, including the visual neocortex (Kirkwood
2005). We speculate that brief application of DHPG does not affect the dynamics of the
excitatory response, but that it is the prolonged activation of group I mGluRs that leads to longlasting alterations in fast excitatory transmission. We also tested the effects of group II mGluRs
on the excitatory response, and activation of these receptors led to a strong suppression of

	

90	

excitation onto pyramidal neurons. Group II mGluRs modulate fast excitatory transmission in
several brain regions (Anwyl et al., 1999), and this mechanism is one means by which
hyperexcitability can be prevented (Hovelso et al., 2012). Thus, we were interested in whether
this mechanism is intact in the Fmr1 KO mouse, and we determined that group II mGluRmediated suppression of excitatory currents onto pyramidal neurons is comparable to that
observed in the WT.
Although none of the parameters we tested in layer 5/6 pyramidal neurons were
significantly altered in Fmr1 KO mice, we set out to determine whether this was the case in layer
2/3. Indeed, it has been shown that Fmr1 KO mice pyramidal neurons show greater spine
densities in layer 2/3 in addition to layer 5/6 (Dolen et al., 2007, Irwin et al., 2000), and deficits
in group I mGluR-mediated modulation of synaptic transmission onto neurons in layer 2/3 have
been reported in the Fmr1 KO (Paluskiewicz et al., 2011). With this information we set out to
determine whether mGluR-mediated modulation of layer 2/3 pyramidal neurons is intact in the
Fmr1 KO, and we report that none of the parameters we tested were observably altered.
These experiments have shown the extent to which activation of group I and group II
mGluRs alter the membrane potential of pyramidal neurons and the extent to which they
modulate excitatory transmission onto these cells. We further demonstrated that the effects of
both group I and group II mGluRs are intact in the Fmr1 KO mouse. We add to the body of
literature on the effects of group I mGluRs in the neocortex, while simultaneously investigating
group II mGluR modulation in the neocortex in Fragile X. This investigation has not been done
previously, despite evidence suggesting that group II mGluR pharmacology can restore
exaggerated LTD in the hippocampus (Choi et al., 2007). The observation that these experiments
yield no change between WT and Fmr1 KO pyramidal neurons, does not exclude the possibility

	

91	

that alterations of group I and group II mGluR signaling may still be contributing to
hyperexcitability of V1 neuronal circuits. Inhibition is equally important in maintaining the
excitatory/inhibitory balance of neocortical circuitry, and there are multiple different interneuron
subtypes in the visual neocortex that accomplish this task. Thus, further studies must be done to
investigate the role of mGluRs in modulating GABAergic cells in the visual neocortex (Chapter
3).

	

92	

APPENDIX

	

93	

Figure 2.1. Schematic diagram of stimulation protocol.
For layer 5/6 recordings, a stimulating electrode was placed into layer 2/3 off-column, and
neurons in layer 5/6 were recorded from in whole-cell patch clamp configuration. For layer 2/3
recordings, a stimulating electrode was placed into layer 4 off-column, and neurons in layer 2/3
were recorded from also in whole-cell patch clamp configuration.

	

94	

Figure 2.2. The intrinsic excitability of Fmr1 KO pyramidal neurons is unaltered in layer
5/6 visual neocortex.
Figure Ai. Representative tracing of a wild type pyramidal cell during a current-step injection
protocol. Aii. Similar to figure Ai but recorded in pyramidal neurons from the Fmr1 KO mouse.
B. Population data showing the average input resistance in wild type and Fmr1 KO cells
(WT=48, KO=47 p > 0.05, t-test). C. Population data showing the average membrane potential
of wild type and Fmr1 KO cells (WT=48, KO=47, p > 0.05, t-test)

	

95	

Figure 2.3. mGluR-mediated modulation of pyramidal neuron membrane potential is intact
in layer 5/6 of the Fmr1 KO mouse.
Ai. Wild type layer 5/6 pyramidal neuron membrane potential response to application of a
general group I/II mGluR agonist (APDC) and specific agonists for either group I mGluRs
(DHPG) or group II mGluRs (APDC). All experiments were done in the presence of TTX
(1µM) to isolate the postsynaptic response. Aii. Same as Ai, but for layer 5/6 Fmr1 KO
pyramidal neurons. B. Graph showing population data for the change in membrane potential to
various mGluR agonists in both WT an Fmr1 KO mouse. Numbers are reported as absolute
	

96	

Figure 2.3. (cont’d)
values. ACPD produced both depolarizations and hyperpolarizations in the WT and Fmr1 KO.
Group I mGluR agonist DHPG produced depolarizations, and group II mGluR agonist APDC
produced hyperpolarizations. No statistically significant differences were observed between
genotypes for any of the drug applications. (p > 0.05, t-test).

	

97	

Figure 2.4. Group I mGluRs do not suppress excitatory synaptic transmission of layer 5/6
pyramidal neurons.

	

98	

Figure 2.4. (cont’d)
Ai. Representative trace showing the response of a layer 5/6 wild type pyramidal neuron EPSC
to brief application of group I mGluR agonist DHPG (25µM). Aii. Raw time course of the
changes in peak amplitude in a single wild type pyramidal cell. Aiii. Wild type normalized time
course of the population peak amplitude change (n=24). Bi-iii. Same as Ai-iii, but for Fmr1 KO
pyramidal neurons. C. Average EPSC amplitude expressed as % of control for all cells tested in
both WT and Fmr1 KO (WT=24, KO=21). A comparison of the EPSC amplitude of WT and
Fmr1 KO pyramidal neurons in the presence of DHPG found no significant differences (P >
0.05, t-test).

	

99	

Figure 2.5. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 5/6 visual neocortical pyramidal neurons in the WT and Fmr1 KO
mouse.
Ai. Representative trace showing a wild type EPSC evoked by electrical stimulation of layer 2/3.
Traces show wild type EPSC response before, during, and after brief application of APDC
(75µM). Aii. Raw time course of the change in wild type EPSC amplitude before, during, and

	

100	

Figure 2.5. (cont’d)
after APDC application. Aiii. WT population data showing the normalized time course of the
change in peak amplitude in response to APDC (n=16). Bi-iii. Same as Ai-iii but for Fmr1 KO.
C. Average population EPSC amplitude in the presence of APDC expressed as percent of control
for both wild type and Fmr1 KO (WT=16, KO=16). No statistical differences were found
between peak amplitude of WT and Fmr1 KO in the presence of APDC (p > 0.05, t-test).
	

	

101	

Figure 2.6. The intrinsic properties of Fmr1 KO layer 2/3 pyramidal neurons are similar to
wild type.
Ai. Representative tracing of a layer 2/3 wild type pyramidal cell during a current-step injection
protocol. Aii. Similar to figure Ai but recorded in layer 2/3 pyramidal neurons from the Fmr1
KO mouse. B. Population data showing the average input resistance in wild type and Fmr1 KO
cells (WT=9, KO=10, p > 0.05, t-test). C. Population data showing the average membrane
potential of wild type and Fmr1 KO cells (WT=9, KO=10 p > 0.05, t-test). No statistical
differences between genotypes were observed for either input resistance or membrane potential
(p>0.05, t-test).

	

102	

Figure 2.7. DHPG does not alter EPSCs in layer 2/3 pyramidal neurons from WT and
Fmr1 KO mice.

	

103	

Figure 2.7. (cont’d)
Ai. Representative trace showing the response of a layer 2/3 wild type pyramidal neuron EPSC
to brief application of group I mGluR agonist DHPG (25µM). Aii. Raw time course of the
changes in peak amplitude in a single wild type pyramidal cell. Aiii. Wild type normalized time
course of the population peak amplitude change (n=5). Bi-iii. Same as Ai-iii, but for Fmr1 KO.
C. Average EPSC amplitude expressed as % of control for all cells tested in both WT and Fmr1
KO (WT=5, KO=8). Statistical comparison of peak amplitude change in response to DHPG
shows no difference between wild type and Fmr1 KO pyramidal neurons (P > 0.05, t-test).

	

104	

Figure 2.8. Group II mGluR activation leads to suppression of excitatory synaptic
transmission onto layer 2/3 visual neocortical pyramidal neurons in the WT and Fmr1 KO
mouse.
Ai. Representative trace showing a layer 2/3 wild type pyramidal EPSC evoked by electrical
stimulation of layer 2/3. Traces show wild type EPSC response before, during, and after brief
application of APDC (75µM). Aii. Raw time course of the change in wild type EPSC amplitude
before, during, and after APDC application. Aiii. WT population data showing the normalized

	

105	

Figure 2.8. (cont’d)
time course of the change in peak amplitude in response to APDC (n=8). Bi-iii. Same as Ai-iii
but for Fmr1 KO (n=7). C. Average population EPSC amplitude in the presence of APDC
expressed as % of control for both wild type and Fmr1 KO (WT=8, KO=7). No statistical
differences were found between peak amplitude of WT and Fmr1 KO in the presence of APDC
(p > 0.05, t-test).

	

106	

Figure 2.9. Group II mGluR-mediated suppression of EPSCs on pyramidal neurons is
normal in the Fmr1 KO mouse.
In chapter 2, excitatory postsynaptic currents (EPSCs) were evoked using a stimulating electrode
placed in layer 2/3 (S). We show that activation of group II mGluRs leads to suppression of
EPSCs at glutamatergic synapses onto layer 5/6 pyramidal neurons in both the WT and Fmr1
KO (black circle). The magnitude of this suppression is similar between WT and Fmr1 KO
mouse (black checkmark).

	

107	

REFERENCES

	

108	

REFERENCES
Akins, M. R., Berk-Rauch, H. E., & Fallon, J. R. (2009). Presynaptic translation: stepping out of
the postsynaptic shadow. Front Neural Circuits, 3, 17. doi:10.3389/neuro.04.017.2009
Antar, L. N., Afroz, R., Dictenberg, J. B., Carroll, R. C., & Bassell, G. J. (2004). Metabotropic
glutamate receptor activation regulates fragile x mental retardation protein and FMR1
mRNA localization differentially in dendrites and at synapses. J Neurosci, 24(11), 26482655. doi:10.1523/JNEUROSCI.0099-04.2004
Anwyl, R. (1999). Metabotropic glutamate receptors: electrophysiological properties and role in
plasticity. Brain Res Brain Res Rev, 29(1), 83-120.
Chang, Y. M., & Luebke, J. I. (2007). Electrophysiological diversity of layer 5 pyramidal cells in
the prefrontal cortex of the rhesus monkey: in vitro slice studies. J Neurophysiol, 98(5),
2622-2632. doi:10.1152/jn.00585.2007
Choi, C. H., Schoenfeld, B. P., Bell, A. J., Hinchey, P., Kollaros, M., Gertner, M. J., . . .
McBride, S. M. (2011). Pharmacological reversal of synaptic plasticity deficits in the
mouse model of fragile X syndrome by group II mGluR antagonist or lithium treatment.
Brain Res, 1380, 106-119. doi:10.1016/j.brainres.2010.11.032
Chuang, S. C., Zhao, W., Bauchwitz, R., Yan, Q., Bianchi, R., & Wong, R. K. (2005). Prolonged
epileptiform discharges induced by altered group I metabotropic glutamate receptormediated synaptic responses in hippocampal slices of a fragile X mouse model. J
Neurosci, 25(35), 8048-8055. doi:10.1523/JNEUROSCI.1777-05.2005
Dolen, G., Osterweil, E., Rao, B. S., Smith, G. B., Auerbach, B. D., Chattarji, S., & Bear, M. F.
(2007). Correction of fragile X syndrome in mice. Neuron, 56(6), 955-962.
doi:10.1016/j.neuron.2007.12.001
Gibson, J. R., Bartley, A. F., Hays, S. A., & Huber, K. M. (2008). Imbalance of neocortical
excitation and inhibition and altered UP states reflect network hyperexcitability in the
mouse model of fragile X syndrome. J Neurophysiol, 100(5), 2615-2626.
doi:10.1152/jn.90752.2008
Hagerman, P. J. (2008). The fragile X prevalence paradox. J Med Genet, 45(8), 498-499.
doi:10.1136/jmg.2008.059055
Hagerman, P. J., & Stafstrom, C. E. (2009). Origins of epilepsy in fragile X syndrome. Epilepsy
Curr, 9(4), 108-112. doi:10.1111/j.1535-7511.2009.01309.x
Hinton, V. J., Brown, W. T., Wisniewski, K., & Rudelli, R. D. (1991). Analysis of neocortex in
three males with the fragile X syndrome. Am J Med Genet, 41(3), 289-294.
doi:10.1002/ajmg.1320410306

	

109	

Hovelso, N., Sotty, F., Montezinho, L. P., Pinheiro, P. S., Herrik, K. F., & Mork, A. (2012).
Therapeutic potential of metabotropic glutamate receptor modulators. Curr
Neuropharmacol, 10(1), 12-48. doi:10.2174/157015912799362805
Huber, K. M., Gallagher, S. M., Warren, S. T., & Bear, M. F. (2002). Altered synaptic plasticity
in a mouse model of fragile X mental retardation. Proc Natl Acad Sci U S A, 99(11),
7746-7750. doi:10.1073/pnas.122205699
Irwin, S. A., Idupulapati, M., Gilbert, M. E., Harris, J. B., Chakravarti, A. B., Rogers, E. J., . . .
Greenough, W. T. (2002). Dendritic spine and dendritic field characteristics of layer V
pyramidal neurons in the visual cortex of fragile-X knockout mice. Am J Med Genet,
111(2), 140-146. doi:10.1002/ajmg.10500
Irwin, S. A., Galvez, R., & Greenough, W. T. (2000). Dendritic spine structural anomalies in
fragile-X mental retardation syndrome. Cereb Cortex, 10(10), 1038-1044.
Irwin, S. A., Patel, B., Idupulapati, M., Harris, J. B., Crisostomo, R. A., Larsen, B. P., . . .
Greenough, W. T. (2001). Abnormal dendritic spine characteristics in the temporal and
visual cortices of patients with fragile-X syndrome: a quantitative examination. Am J
Med Genet, 98(2), 161-167.
Kaufmann, W. E., Cortell, R., Kau, A. S., Bukelis, I., Tierney, E., Gray, R. M., . . . Stanard, P.
(2004). Autism spectrum disorder in fragile X syndrome: communication, social
interaction, and specific behaviors. Am J Med Genet A, 129A(3), 225-234.
doi:10.1002/ajmg.a.30229
Kirkwood, A., Rozas, C., Kirkwood, J., Perez, F., & Bear, M. F. (1999). Modulation of longterm synaptic depression in visual cortex by acetylcholine and norepinephrine. J
Neurosci, 19(5), 1599-1609.
Larkman, A., & Mason, A. (1990). Correlations between morphology and electrophysiology of
pyramidal neurons in slices of rat visual cortex. I. Establishment of cell classes. J
Neurosci, 10(5), 1407-1414.
Meredith, R. M., Holmgren, C. D., Weidum, M., Burnashev, N., & Mansvelder, H. D. (2007).
Increased threshold for spike-timing-dependent plasticity is caused by unreliable calcium
signaling in mice lacking fragile X gene FMR1. Neuron, 54(4), 627-638.
doi:10.1016/j.neuron.2007.04.028
Paluszkiewicz, S. M., Olmos-Serrano, J. L., Corbin, J. G., & Huntsman, M. M. (2011). Impaired
inhibitory control of cortical synchronization in fragile X syndrome. J Neurophysiol,
106(5), 2264-2272. doi:10.1152/jn.00421.2011
Rudelli, R. D., Brown, W. T., Wisniewski, K., Jenkins, E. C., Laure-Kamionowska, M., Connell,
F., & Wisniewski, H. M. (1985). Adult fragile X syndrome. Clinico-neuropathologic
findings. Acta Neuropathol, 67(3-4), 289-295.

	

110	

Zhang, Y., Bonnan, A., Bony, G., Ferezou, I., Pietropaolo, S., Ginger, M., . . . Frick, A. (2014).
Dendritic channelopathies contribute to neocortical and sensory hyperexcitability in
Fmr1(-/y) mice. Nat Neurosci, 17(12), 1701-1709. doi:10.1038/nn.3864

	

111	

CHAPTER III: MGLUR-MEDIATED MODULATION OF VISUAL NEOCORTICAL
PYRAMIDAL CELL INHIBITION IN A FRAGILE X MOUSE
3.1 Introduction
Fragile X Syndrome is a genetic form of intellectual disability (Hagerman et al., 2009).
Studies using an Fmr1 KO mouse, have uncovered a wide array of defects with glutamatergic
transmission. However, less is known about the effects of loss of FMRP on GABAergic
transmission, particularly as it pertains to neocortex, where interneurons comprise 20% of all
cells (Rudy et al., 2011). Most neocortical studies on GABA dysfunction have looked at mRNA
expression of GABAA subunits. GABAA receptors mediate fast ionotropic inhibitory
transmission, and any changes in the expression level of GABAA subunits would be
hypothesized to affect inhibitory function at the synapse. Several studies in the Fmr1 KO show
decreased mRNA levels of GABAA subunits in the hippocampus and neocortex of Fragile X
mice (Gantois, I. et al, 2006; Hong, et al 2012). This underexpression includes the subunits α1,
β2, and γ2, which together comprise the most abundant GABAA receptor in the CNS. Protein
expression of GABAA subunits has also been shown to be reduced in the neocortex,
hippocampus, diencephalon, and brainstem of Fmr1 KO mice (El Idrissi et al., 2005, Adusei et
al., 2010). Additionally, mRNA expression of proteins important for GABA synthesis, transport,
and degradation, have been shown to be decreased in both the neocortex and cerebellum
(D’Hulst et al. 2008). Despite these results, we still have no idea what the functional
consequences are of underexpression of neocortical GABAergic compounds in the Fmr1 KO.
The electrophysiological studies investigating functional changes in GABAergic signaling todate have documented both increases and decrease in tonic inhibition that appear to be dependent
on the brain region under examination. Increases in the frequency of spontaneous and miniature

	

112	

inhibitory currents have been reported in the subiculum (Centonze et al., 2008), indicating an
increase in tonic inhibition. However, decreased tonic inhibition has also been reported in the
subiculum (Curia et al., 2009), amygdala (Olmos-Serrano, et al., 2010; Martin et al., 2014), and
hippocampus (Sabonov et al., 2017). Phasic inhibition has also been shown to be decreased in
the amygdala (Olmos-Serrano et al., 2010). Although we would expect that the decreased
expression of GABAA subunits in the neocortex might lead to decreased inhibition in the
neocortex, that question remains unanswered due to the lack of electrophysiological studies on
GABAergic transmission in neocortex. While deficits in excitatory drive onto interneurons in
the somatosensory neocortex have been reported (Gibson et al., 2008; Paluszkiewicz et al.,
2011), no changes in neocortical GABAergic transmission have been reported (Gibson et al.,
2008). This is interesting considering another study showed a significant decrease in the density
of FS interneurons in the somatosensory cortex, as well as altered laminar distributions of these
neurons (Selby, Zhang & Sun, 2006). Clearly, more electrophysiological studies need to be done
before we understand what is happening to GABAergic transmission in the neocortex.
In Chapter 2, I previously determined that mGluR-mediated modulation of glutamatergic
transmission onto pyramidal neurons is unaltered in the layer 5/6 visual neocortex of the Fmr1
KO mouse. Given that several human studies suggest that the neocortex is hyperexcitable in
FXS, it is possible that hyperexcitable neocortical circuitry occurs due to defective inhibition
rather than enhanced excitation. In the neocortex, mGluRs have been shown to modulate
GABAergic transmission (Sladeczek et al., 2016; Liu, Petrof & Sherman, 2014). Since mGluRs
are heavily implicated in the pathogenesis of the Fragile X mouse model, I hypothesize that
mGluR-mediated modulation of GABAergic neurotransmission in the layer 5/6 neocortex is
altered in the Fmr1 KO mouse. Group II mGluRs, in particular seem to be extremely effective at

	

113	

suppressing neurotransmission at both glutamatergic and GABAergic synapses (Liu, Petrof, &
Sherman, 2014; Flavin, Jin & Daw., 2000). Consequently, in this chapter I focused on the
effects of group II mGluRs in modulating GABAergic transmission. I recorded inhibitory
currents from pyramidal neurons before, during, and after application of a group II mGluR
agonist in an attempt to observe and quantify changes in the amplitude of this response in the
Fmr1 KO. I also determined the location of group II mGluRs at GABAergic interneuron
synapses onto pyramidal cells.
3.2 Materials and Methods
All experimental procedures were performed in accordance with the National Institutes of
Health Guide for Care and Use if Laboratory Animals and were approved by the Michigan State
University Institutional Animal Care and Use Committee.
Mice on the FVB background (postnatal age 16-21 days) of either sex were anesthetized
with isofluorane and perfused with cold, oxygenated (5% CO2-95% O2), modified saline prior to
decapitation. Brains were removed and transferred to cold (4°C), oxygenated slicing solution
containing the following (in mM): 2.5 KCl, 1.25 NaH2PO4, 10.0 MgSO4, 0.5 CaCl2, 26.0
NaHCO3, 10.0 glucose, and 234.0 sucrose. Slices (300 µm thickness) were collected in a
coronal plane were collected using a Leica vibrating tissue slicer, and then transferred to a
holding chamber filled with oxygenated physiological saline containing (in mM): 126.0 NaCl,
2.5 KCl, 1.25 NaH2PO4, 2.0 MgCl2, 2.0 CaCl2, 26.0 NaHCO3, and 10.0 glucose. Slices were
incubated at 32 ± 1°C for at least 30 minutes, and then left to rest at room temperature for at least
30 minutes prior to recording.
Whole-cell recordings were obtained using a microscope equipped with differential
interface contrast optics (Zeiss, Thornwood, NY). Single slices were transferred to a submersion

	

114	

chamber continuously superfused at 2.5ml/min with oxygenated physiological saline heated to
32◦C. Visual neocortex was identified with a low power (5x) objective and individual neurons
were identified using a water-immersion objective (63x). Pyramidal neurons were identified
through preferential targeting of pyramidal-shaped soma with a diameter greater than 20 µm.
The majority of pyramidal neurons (62%) exhibit a regular spiking (RS) slowly adapting firing
pattern, with a moderate sized late after-hyperpolarizing potential (AHP) and most often no
visible early (AHP) or depolarizing afterpotential (DAP) (Chang and Luebke, 2007; Larkman
and Mason; 1990). Another less common (5%), but easily distinguished electrophysiological
profile of layer 5/6 pyramids was that of intrinsic bursting firing (IB) of 3-6 bursts of APs
(Chang and Luebke, 2007; Larkman and Mason; 1990). For recordings, pipettes made of
borosilicate capillary glass were pulled to tip resistances of 4-6MΩ and then filled with a cesiumbased intracellular solution containing (in mM): 117.0 Cs-gluconate, 13.0 CsCl, 1.0 MgCl2, 0.07
CaCl2, 0.1 EGTA, 10.0 HEPES, 2.0 Na2-ATP, and 0.4 Na-GTP. Only cells with a membrane
potential ≥ -50 mV and overshooting action potentials were included for analysis. Access
resistance was monitored throughout the experiments and recordings were not included if the
access resistance increased to >20MΩ.
In recordings from layer 5/6 pyramidal neurons, inhibitory postsynaptic currents (IPSCs)
were evoked by electrical stimulation of afferents in layer 2/3 (Figure 2.1, left panel). Stimuli
were delivered at 0.1 Hz, and the peak amplitude of the evoked IPSC was measured before,
during, and after washout of the selective group II mGluR agonist, (2R,4R)-4-aminopyrrolidine2,4-dicarboxylate (APDC). Spontaneous IPSCs (sIPSCs) and miniature IPSCs (mIPSCs) were
recorded in the absence of electrical stimulation while the cell was held at 0 mV. mIPSCs were
recorded in the presence of tetrodotoxin (1µM).

	

115	

All compounds used in this study were prepared from stock solutions stored at the
appropriate temperature and diluted to the correct concentration with physiological saline prior to
administration. APDC was delivered by bolus injection to the central line of the recording
chamber using a syringe pump. Antagonist application was applied in the bath physiological
saline. Based upon the calculated flow rate of the main line, the delivered drug concentration
was estimated to be one-fourth of the initial drug concentration. All compounds were purchased
from Tocris and Sigma-Aldrich.
All recordings were acquired using a Multiclamp 700B amplifier (Molecular Devices).
Data was filtered at 2.5 kHz and digitized at 10 kHz using pClamp software. For evoked
experiments, 12 consecutive sweeps (2 minutes) were averaged to determine the IPSC baseline
peak amplitude of the synaptic response. The drug effect was determined as the average of 3
consecutive sweeps during the maximal drug response. A positive drug response was defined as
a change in amplitude that was greater than 2 standard deviations of the baseline response. The
frequency and amplitude of spontaneous and miniature IPSCS were analyzed using
MiniAnalysis software (Synaptosoft). Kolmogorov-Smirnoff tests were used within cells to
determine if sIPSC and mIPSC frequency and/or amplitude were altered by drug application. All
data are reported as the mean ± SEM and analyzed using either paired or independent t-test were
applicable. A p-value <0.05 was considered statistically significant.
3.3 Results
A. Group II mGluR-mediated suppression of IPSCs on layer 5/6 pyramidal neurons is
exaggerated in the Fmr1 KO mouse.
To determine whether activation of group II mGluRs leads to altered GABAergic
inhibition onto pyramidal neurons in the Fmr1 KO, whole cell patch clamp recordings were

	

116	

obtained from layer 5/6 pyramidal neurons in the visual neocortex of WT (Figure 3.1 Ai) and
Fmr1 KO animals (Figure 3.1 Bi). A cesium based internal solution and a holding potential of 0
mV were used to optimize recordings of inhibitory postsynaptic currents (IPSCs). Synaptic
responses were evoked using a bipolar stimulating electrode inserted into layer 2/3. In WT
animals (n=16), the baseline ISPC averaged 1333.2 ± 143.9 pA. Following application of APDC
(75 µM, 40 s), the IPSC was reduced to 685.4 ± 117.1 pA, and this decrease was statistically
significant (p<0.001, paired t-test, Figure 3.1 Aii-iii). In neurons from Fmr1 KO animals,
APDC also significantly suppressed the IPSC (Pre-drug: 959.7 ± 121.3 pA; APDC: 272.5 ± 65.6
pA, n=8, p<0.001, paired t-test, Figure 3.1 Bii-iii). The average amplitude of the WT IPSC
upon application of APDC was 45.3 ± 5.5 % of the control IPSC amplitude, while the Fmr1 KO
was 26.4 ± 5.3 % of the control response, indicating a suppression of the IPSC in the Fmr1 KO
that was nearly 20% larger than the WT. This difference was determined to be statistically
significant (p=0.023, t-test, Figure 3.1C). Thus, we show that activation of group II mGluRs
leads to increased suppression of inhibition onto pyramidal neurons in the Fmr1 KO.
B. In the absence of excitatory drive onto interneurons, group II mGluR-mediated
suppression of IPSCs is normal in the Fmr1 KO mouse.
Is the increased suppression of inhibitory currents we observe in Fmr1 KO pyramidal
neurons due to the presence of group II mGluRs at GABA synapses? To answer this question,
we next sought to record evoked inhibitory currents in the presence of 3-(2- carboxypiperazin-4yl)-propyl-1-phosphonic acid (CPP, 20 µM) and 6,7-dinitroquinoxaline-2,3-dione (DNQX, 40
µM) to isolate interneuron synapses onto pyramidal neurons. In the presence of CPP and
DNQX, APDC significantly reduced the amplitude of the IPSCs in WT cells (Figure 3.2 Ai).
The average baseline response of the IPSC was 439.3 ± 104.4 pA. Following application of

	

117	

APDC, this response decreased to 314.1 ± 83.8 pA (n=9, p<0.05, paired t-test, Figure 3.2 Aiiiii). In Fmr1 KO pyramidal neurons, the average baseline response in the presence of CPP and
DNQX was 122.2 ± 27.7 pA, and APDC decreased this amplitude to 60.7 ± 9.3 pA (n=4, p<0.05,
paired t-test, Figure 3.2 Bii-iii). A comparison of the IPSC amplitude in response to APDC
expressed as a % of the control response, was made between WT and Fmr1 KO. No statistical
differences were observed between genotypes (WT: 69.5 ± 5.9 %; Fmr1 KO: 52.9 ± 6.0%,
p>0.05, t-test, Figure 3.2 C). These data indicate that group II mGluRs directly modulate
GABAergic transmission in layer 5/6 of the visual neocortex. Furthermore, the exaggerated
inhibition of pyramidal neurons by group II mGluRs is not caused by direct changes in
GABAergic transmission onto pyramidal neurons.
C. Suppression of IPSCs onto pyramidal neurons is due to either pre- or post- synaptic
group II mGluRs.
We show group II mGluRs directly modulate GABAergic transmission in layer 5/6 of the
visual neocortex. However, we do not know where these mGluRs are located at GABAergic
synapses. To locate the group II mGluRs we first recorded spontaneous IPSCs in layer 5/6
pyramidal neurons (Figure 3.3 Ai and Bi). In the WT mouse, 3 of 5 cells showed a statistically
significant change in sIPSC amplitude in response to brief application of APDC (K.S. test, p <
0.05, Figure 3.3 Aii). The average baseline amplitude of sIPSCs was 27.5 ± 2.1 pA, and
application of APDC reduced the average amplitude to 26.5 ± 2.6 pA (n=5). This results in a
response amplitude that is 96.1 ± 2.7 % of the control amplitude (Figure 3.3 C). In the Fmr1
KO, a similar observation was made. In 6 of 10 cells, APDC significantly decreased the
amplitude of the sIPSC, from an average baseline of 29.5 ± 2.0 pA, to a post-drug average of
27.2 ± 1.7 pA (K.S. test, p < 0.05, Figure 3.3 Bii). This led to a response amplitude that was, on

	

118	

average, 92.7 ± 2.0 % of the control response (Figure 3.3C). A comparison of WT and Fmr1
KO sIPSCs determined that there was no significant difference in the amplitude between
genotypes (WT: 96.1 ± 2.7 %, Fmr1 KO: 92.7 ± 2.0 %, p > 0.05, t-test, Figure 3.3 C). The
average frequency of WT sIPSCs was also significantly decreased by APDC application (4 of 5
cells, p < 0.05, K.S. test, Figure 3.3 Aiii). APDC reduced sIPSC frequency from 17.6 ± 5.5 Hz
to 14.4 ± 4.9 Hz (n=5). In the Fmr1 KO, APDC reduced the average sIPSC frequency from 17.0
± 4.7 Hz to 14.4 ± 4.1 Hz (n=10, p < 0.05, K.S. test, Figure 3.3 Biii). Comparison of the
average event frequency expressed as % of control for both WT and Fmr1 KO determined that
there were no significant differences between genotypes (WT: 77.1 ± 6.6 %, Fmr1 KO: 84.4 ±
2.0 %, p > 0.05, t-test, Figure 3.3 D). From these data, we cannot conclude where the
suppression of sIPSCs is occurring at the GABA synapse upon activation of group II mGluRs.
To better assess the location of the group II mGluR effect at GABA synapses, we next
recorded miniature IPSCs in the presence of TTX to isolate AP-independent release from the
terminals of interneuron inputs onto pyramidal neurons of WT and Fmr1 KO mice. In the
presence of TTX, APDC significantly decreased mIPSC amplitude in 3 of 8 wild type cells (K.S.
test, p<0.05, Figure 3.4Ai). Pooled data show that the average mIPSC amplitude of all wild type
cells was 31.1 ± 3.6 pA, and after APDC this amplitude was not significantly different (n=8,
APDC: 31.4 ± 4.7 pA, p>0.05, paired t-test, Figure 3.4 Aii). In the presence of APDC, all 8 WT
cells tested showed a significant decrease in the frequency of the response (p<0.05, K.S. test). A
pooled average from all WT pyramidal neurons, showed that mIPSC frequency decreased in
response to group II mGluR activation (Pre-drug: 9.9 ± 1.7 Hz, APDC: 7.8 ± 1.6 Hz, p<0.05,
paired t-test, Figure 3.4 Aiii). In Fmr1 KO pyramidal neurons, mIPSC amplitude also decreased
in the presence of APDC (Figure 3.4 Bi). Specifically, 5 of 12 Fmr1 KO cells show a

	

119	

statistically significant decrease in mIPSC amplitude (p<0.05, K.S. test). The pooled amplitude
of all mIPSCs for the Fmr1 KO was 27.7 ± 2.2 pA, and APDC did not change the amplitude of
this response (n=12, APDC: 26.8 ± 2.1 pA, p>0.05, paired t-test, Figure 3.4Bii). However,
presence of APDC decreased the frequency of Fmr1 KO mIPSCs in 9 of 12 cells (p<0.05, K.S.
test). Pooled data showed a statistically significant decrease in the frequency of Fmr1 KO
pyramidal mIPSCs in response to APDC application (n=12, Pre-drug: 16.1 ± 3.9 Hz, APDC:
14.1 ± 3.4 Hz, p<0.05, paired t-test, Figure 3.4Biii). From the above data, we conclude that
group II mGluRs are predominantly acting at presynaptic interneuron terminals to decrease
GABAergic transmission. We also observe that this GABAergic modulation is intact in the
Fmr1 KO mouse, indicating that exaggerated suppression of inhibition onto pyramidal neurons
in the Fmr1 KO is due to changes in group II mGluR processes upstream of the GABA synapse.
3.4 Discussion
In this chapter, we determine if activation of group II mGluRs could differentially
modulate GABAergic inhibition in layer 5/6 pyramidal neurons in WT and Fmr1 KO animals.
In addition to modulating glutamatergic transmission across all layers of the visual neocortex
(Flavin, Jin & Daw, 2000), group II mGluRs have been shown to mediate inhibition of
GABAergic currents in hippocampus (Jouvenceau et al. 1995), thalamus (Salt and Eaton, 1995;
Salt and Turner, 1998), olfactory bulb (Hayashi et al., 1993), striatum (Hanania and Johnson,
1999), tectum (Neale and Salt, 2006), and certain regions of the brainstem (Chen and Bonham,
2005; Liu, Petrof & Sherman, 2014). Upon activation, these group II mGluRs have also been
shown to decrease GABAergic transmission onto both layer 2/3 and layer 4 excitatory pyramidal
neurons in the auditory and visual cortices (Liu, Petrof & Sherman). In this chapter, we report
that group II mGluRs also inhibit GABAergic transmission onto layer 5/6 pyramidal neurons, as

	

120	

measured by a decrease in evoked IPSC amplitude is response to pharmacological activation of
group II mGluRs. We further demonstrate that this effect is exaggerated by 20% in the Fmr1
KO mouse, leading to a decrease in inhibition onto layer 5/6 pyramidal neurons. This is the first
electrophysiological study to show a functional defect in group II mGluR signaling in FXS.
Additionally, we show that group II mGluRs are present at both GABAergic terminals and at
postsynaptic pyramidal neurons. Interestingly, neither of these group II mGluR populations are
responsible for mediating the exaggerated suppression of inhibition that we see in the Fmr1 KO.
Blocking glutamatergic excitation of interneurons led to a suppression of the evoked IPSC by
group II mGluRs that was similar between WT and Fmr1 KO. We also recorded both sIPSC and
mIPSC recordings and confirmed that group II mGluR-mediated suppression of IPSC frequency
and amplitude is similar in WT and Fmr1 KO mice, further supporting the idea that there is
normal group II mGluR-mediated modulation of GABAergic transmission onto pyramidal
neurons in the layer 5/6. Together these results suggest that the changes we observe are due to
group II mGluRs residing upstream of the GABAergic terminals. We hypothesize that the group
II mGluR-mediated suppression of inhibition in Fmr1 KO pyramidal neurons is actually
disinhibition caused by exaggerated suppression of excitation onto layer 5/6 interneurons upon
activation of group II mGluRs. We also add to the literature by showing that group II mGluRmediated modulation of GABA synapses are intact in the Fmr1 KO. Interestingly, the only
functional studies that have shown impairments in inhibition of the neocortex, both find that
excitatory drive onto interneurons and/or modulation of excitatory drive is deficient (Gibson et
al., 2008; Paluszkiewicz et al., 2011). This is also what our data suggest. While defective
GABAergic transmission appears to be a major defect in other brain regions such as the
amygdala (Olmos-Serrano et al., 2010; Vislay et al., 2013; Martin et al., 2014), it may turn out

	

121	

that selective loss of excitatory drive onto and/or modulation of interneurons is the major circuit
defect in neocortical brain regions. Unfortunately experiments such as these will need to be
completed in other neocortical brain regions and layers, before we can be certain. In the
following chapter, I will test the hypothesis that activation of group II mGluRs leads to
exaggerated suppression of excitatory glutamatergic transmission onto layer 5/6 interneurons
(Chapter 4).

	

122	

APPENDIX

	

123	

Figure 3.1. Group II mGluR-mediated modulation of IPSCs in layer 5/6 pyramidal neurons
is exaggerated in the Fmr1 KO mouse.

	

124	

Figure 3.1. (cont’d)
Ai. Representative trace showing the response of a layer 5/6 wild type pyramidal neuron IPSC to
brief application of group II mGluR agonist, APDC (75 µM). ii. Raw time course of the change
in peak amplitude in a single wild type pyramidal cell. iii. Wild type normalized time course of
the population peak amplitude change (n=16). Bi-iii. Same as Ai-iii, but for the Fmr1 KO (n=8).
C. Average IPSC amplitude immediately following APDC application, expressed as % of control
for all cells tested in both WT and Fmr1 KO. Comparison of peak amplitude following APDC
application determined that the Fmr1 KO IPSC suppression mediated by group II mGluR
activation is exaggerated relative to wild type pyramidal neuron IPSCs (WT=16, KO=8,
p=0.023, t-test).

	

125	

Figure 3.2. Group II mGluRs suppress fast inhibitory transmission onto WT and Fmr1 KO
pyramidal neurons in the absence of excitatory drive onto neocortical interneurons.

	

126	

Figure 3.2. (cont’d)
Ai. Representative trace showing the presence of a layer 5/6 wild type pyramidal neuron IPSC to
brief application of group II mGluR agonist APDC (75 µM), in the presence of bath applied
NMDA and AMPA receptor antagonists, CPP (20 µM) and DNQX (40 µM). Aii. Raw time
course of the changes in peak amplitude in a single wild type pyramidal cell. Aiii. Wild type
normalized time course of the population peak amplitude change (n=9). Bi-iii. Same as Ai-iii,
but for Fmr1 KO. C. Average IPSC amplitude immediately following APDC application,
expressed as % of control for all cells tested in both WT and Fmr1 KO (WT=9, KO=4). A
comparison of the population peak amplitude determined that there were no significant
differences between WT and Fmr1 KO (WT=9, KO=4, p > 0.05, t-test).

	

127	

Figure 3.3. Group II mGluRs modulate sIPSC amplitude and frequency similarly between
WT and Fmr1 KO layer 5/6 pyramidal neurons.
Ai. Raw data showing spontaneous IPSCs recorded from WT pyramidal neurons before, during
and after application of group II mGluR agonist APDC (75 µM). Aii. Representative cumulative
distribution showing a significant decrease in WT amplitude in response to APDC (n=1, p< 0.05,
	

128	

Figure 3.3. (cont’d)
K.S. test). Aiii. Representative cumulative distribution showing a decrease in frequency upon
APDC application (n= 1, p < 0.05, K.S. test). Bi-iii. Same as Ai-iii, but for Fmr1 KO. Fmr1 KO
pyramidal neurons also show a shift in the cumulative distribution for amplitude and frequency
in response to APDC (p < 0.05, K.S. test). C. Comparison of the average population sIPSC
response amplitude between WT and Fmr1 KO was not significant (WT=5, KO=10, p > 0.05, ttest). D. Population data for sIPSC frequency was also not statistically different between
genotypes (WT=5, KO=10, p > 0.05, t-test).

	

129	

Figure 3.4. Presynaptic mGluRs at the interneuron terminals modulate inhibition similarly
between WT and Fmr1 KO layer 5/6 pyramidal neurons.
Ai. Raw data showing miniature IPSCs recorded from WT pyramidal neurons before, during and
after application of group II mGluR agonist APDC (75 µM). Aii. Average wild type population
change in mIPSC peak amplitude in response to APDC application (n=8). Aiii. Average wild
type population change in mIPSC frequency in response to APDC application (n=8). Bi-iii.
Same as Ai-iii, but for the Fmr1 KO. APDC did not change the amplitude of mIPSCs for either
WT or Fmr1 KO (p > 0.05, t-test). Frequency of mIPSCs was statistically decreased in the
	

130	

Figure 3.4. (cont’d)
presence of APDC for both WT and Fmr1 KO pyramidal neurons, indicating that group II
mGluRs are located presynaptically at GABAergic terminals. No differences were seen between
WT and Fmr1 KO (p > 0.05, t-test).

	

131	

Figure 3.5. Group II mGluR-mediated suppression of IPSCs on pyramidal neurons is
normal in the Fmr1 KO mouse.
In chapter 3, inhibitory postsynaptic currents (IPSCs) were evoked using a stimulating electrode
placed in layer 2/3 (S). We show that activation of group II mGluRs leads to suppression of
IPSCs at glutamatergic synapses onto layer 5/6 pyramidal neurons in both the WT and Fmr1 KO
(black circle). The magnitude of this suppression is similar between WT and Fmr1 KO mouse
(black checkmark).

	

132	

REFERENCES

	

133	

REFERENCES
Adusei, D. C., Pacey, L. K., Chen, D., & Hampson, D. R. (2010). Early developmental
alterations in GABAergic protein expression in fragile X knockout mice.
Neuropharmacology, 59(3), 167-171. doi:10.1016/j.neuropharm.2010.05.002
Centonze, D., Rossi, S., Mercaldo, V., Napoli, I., Ciotti, M. T., De Chiara, V., . . . Bagni, C.
(2008). Abnormal striatal GABA transmission in the mouse model for the fragile X
syndrome. Biol Psychiatry, 63(10), 963-973. doi:10.1016/j.biopsych.2007.09.008
Chang, Y. M., & Luebke, J. I. (2007). Electrophysiological diversity of layer 5 pyramidal cells in
the prefrontal cortex of the rhesus monkey: in vitro slice studies. J Neurophysiol, 98(5),
2622-2632. doi:10.1152/jn.00585.2007
Chen, C. Y., & Bonham, A. C. (2005). Glutamate suppresses GABA release via presynaptic
metabotropic glutamate receptors at baroreceptor neurones in rats. J Physiol, 562(Pt 2),
535-551. doi:10.1113/jphysiol.2004.076885
D'Hulst, C., Heulens, I., Brouwer, J. R., Willemsen, R., De Geest, N., Reeve, S. P., . . . Kooy, R.
F. (2009). Expression of the GABAergic system in animal models for fragile X syndrome
and fragile X associated tremor/ataxia syndrome (FXTAS). Brain Res, 1253, 176-183.
doi:10.1016/j.brainres.2008.11.075
El Idrissi, A., Ding, X. H., Scalia, J., Trenkner, E., Brown, W. T., & Dobkin, C. (2005).
Decreased GABA(A) receptor expression in the seizure-prone fragile X mouse. Neurosci
Lett, 377(3), 141-146. doi:10.1016/j.neulet.2004.11.087
Flavin, H. J., Jin, X. T., & Daw, N. W. (2000). 2R,4R-4-Aminopyrrolidine-2,4-dicarboxylate
(APDC) attenuates cortical EPSPs. Brain Res, 873(2), 212-217.
Gantois, I., Vandesompele, J., Speleman, F., Reyniers, E., D'Hooge, R., Severijnen, L. A., . . .
Kooy, R. F. (2006). Expression profiling suggests underexpression of the GABA(A)
receptor subunit delta in the fragile X knockout mouse model. Neurobiol Dis, 21(2), 346357. doi:10.1016/j.nbd.2005.07.017
Gibson, J. R., Bartley, A. F., Hays, S. A., & Huber, K. M. (2008). Imbalance of neocortical
excitation and inhibition and altered UP states reflect network hyperexcitability in the
mouse model of fragile X syndrome. J Neurophysiol, 100(5), 2615-2626.
doi:10.1152/jn.90752.2008
Hagerman, P. J., & Stafstrom, C. E. (2009). Origins of epilepsy in fragile X syndrome. Epilepsy
Curr, 9(4), 108-112. doi:10.1111/j.1535-7511.2009.01309.x

	

134	

Hanania, T., & Johnson, K. M. (1999). Regulation of NMDA-stimulated [14C]GABA and
[3H]acetylcholine release by striatal glutamate and dopamine receptors. Brain Res,
844(1-2), 106-117.
Hayashi, Y., Momiyama, A., Takahashi, T., Ohishi, H., Ogawa-Meguro, R., Shigemoto, R., . . .
Nakanishi, S. (1993). Role of a metabotropic glutamate receptor in synaptic modulation
in the accessory olfactory bulb. Nature, 366(6456), 687-690. doi:10.1038/366687a0
Hong, A., Zhang, A., Ke, Y., El Idrissi, A., & Shen, C. H. (2012). Downregulation of GABA(A)
beta subunits is transcriptionally controlled by Fmr1p. J Mol Neurosci, 46(2), 272-275.
doi:10.1007/s12031-011-9531-5
Larkman, A., & Mason, A. (1990). Correlations between morphology and electrophysiology of
pyramidal neurons in slices of rat visual cortex. I. Establishment of cell classes. J
Neurosci, 10(5), 1407-1414.
Liu, T., Petrof, I., & Sherman, S. M. (2014). Modulatory effects of activation of metabotropic
glutamate receptors on GABAergic circuits in the mouse cortex. J Neurophysiol, 111(11),
2287-2297. doi:10.1152/jn.00730.2013
Martin, B. S., Corbin, J. G., & Huntsman, M. M. (2014). Deficient tonic GABAergic
conductance and synaptic balance in the fragile X syndrome amygdala. J Neurophysiol,
112(4), 890-902. doi:10.1152/jn.00597.2013
Neale, S. A., & Salt, T. E. (2006). Modulation of GABAergic inhibition in the rat superior
colliculus by a presynaptic group II metabotropic glutamate receptor. J Physiol, 577(Pt
2), 659-669. doi:10.1113/jphysiol.2006.119073
Olmos-Serrano, J. L., Paluszkiewicz, S. M., Martin, B. S., Kaufmann, W. E., Corbin, J. G., &
Huntsman, M. M. (2010). Defective GABAergic neurotransmission and pharmacological
rescue of neuronal hyperexcitability in the amygdala in a mouse model of fragile X
syndrome. J Neurosci, 30(29), 9929-9938. doi:10.1523/JNEUROSCI.1714-10.2010
Paluszkiewicz, S. M., Olmos-Serrano, J. L., Corbin, J. G., & Huntsman, M. M. (2011). Impaired
inhibitory control of cortical synchronization in fragile X syndrome. J Neurophysiol,
106(5), 2264-2272. doi:10.1152/jn.00421.2011
Salt, T. E., & Eaton, S. A. (1995). Distinct presynaptic metabotropic receptors for L-AP4 and
CCG1 on GABAergic terminals: pharmacological evidence using novel alpha-methyl
derivative mGluR antagonists, MAP4 and MCCG, in the rat thalamus in vivo.
Neuroscience, 65(1), 5-13.
Selby, L., Zhang, C., & Sun, Q. Q. (2007). Major defects in neocortical GABAergic inhibitory
circuits in mice lacking the fragile X mental retardation protein. Neurosci Lett, 412(3),
227-232. doi:10.1016/j.neulet.2006.11.062

	

135	

Sladeczek, F., Momiyama, A., & Takahashi, T. (1993). Presynaptic inhibitory action of a
metabotropic glutamate receptor agonist on excitatory transmission in visual cortical
neurons. Proc Biol Sci, 253(1338), 297-303. doi:10.1098/rspb.1993.0117
Vislay, R. L., Martin, B. S., Olmos-Serrano, J. L., Kratovac, S., Nelson, D. L., Corbin, J. G., &
Huntsman, M. M. (2013). Homeostatic responses fail to correct defective amygdala
inhibitory circuit maturation in fragile X syndrome. J Neurosci, 33(17), 7548-7558.
doi:10.1523/JNEUROSCI.2764-12.2013

	

136	

CHAPTER IV: FMR1 KO NON FAST-SPIKING INTERNEURONS EXHIBIT
EXAGGERATED RESPONSE TO GROUP II MGLUR MEDIATED SUPPRESSION OF
EXCITATORY TRANSMISSION
4.1 Introduction
Fragile X Syndrome (FXS) is the most common inherited form of autism. It is primarily a
disorder of intellectual impairment, with approximately 85% of males and 25% of females
experiencing severe intellectual difficulties (<70 IQ). However, the syndrome also features a
wide array of physical and neuropsychiatric symptoms. Neurological features include
hypersensitivity to sensory stimuli across all modalities, autistic features, and seizure disorder
(Hagerman, 2008). All of these phenotypes have been speculated to result from hyperexcitable
neocortical circuitry. In support of this, functional neocortical studies suggest that neocortical
interneurons are dysfunctionally regulated in the neocortex of Fmr1 KO mice. There are three
major interneuron subtypes in the neocortex (Markram, 2004). Of these subtypes, the two
largest are the parvalbumin-(PV+) and somatostatin-(SST+) expressing interneurons, together
comprising 70% of all neocortical interneurons (Rudy et al., 2001). These interneuron subtypes
differ in their morphology, laminar expression, and in the role they play in modulating excitation
of the neocortical circuitry. PV+ interneurons exhibit a fast-spiking profile (FS) and synapse
onto the soma and proximal dendrites of their target neurons, while SST+ interneurons exhibit
non fast-spiking properties (NFS) with low threshold spikes (LTS) and are dendritic targeting
(Kawaguchi and Kubota, 1997; Beierlein and Connors, 2000). FS interneurons are particularly
well suited for rapid inhibition of their target cell output, whereas NFS interneurons receive
highly facilitating excitatory inputs and are speculated to play a more important role during
elevated levels of network activity, where they can act to synchronize inhibition (Paluskiewicz et
al., 2011; Beierlein and Connors, 2003). Both FS and NFS interneurons are dysfunctionally

	

137	

regulated in the neocortex of the Fmr1 KO mouse. In layer 4 of the somatosensory cortex,
excitatory drive onto FS interneurons is decreased by 51%, and this contributes to a prolonged
duration of upstates observed in the Fmr1 KO somatosensory cortex (Gibson et al., 2008).
Additionally, group I mGluR-mediated modulation of NFS interneurons in layer 2/3 of the
somatosensory cortex is decreased, leading to reduced synchronization of inhibition
(Paluszkiewicz et al., 2011). These studies provide compelling evidence in favor of decreased
excitatory drive of inhibition as a major culprit of neocortical pathogenesis in FXS, albeit it
appears that the extent to which different interneuron subtypes are affected may be laminar
specific. Further, they specifically implicate aberrant mGluR signaling in the function of these
interneurons in FXS. In support of this, we have previously shown an inhibitory defect of layer
5/6 excitatory pyramidal neurons in the visual neocortex of the Fmr1 KO mouse triggered by
pharmacological activation of group II mGluRs. Specifically, activation of these mGluRs leads
to enhanced suppression of GABAergic neurotransmission onto pyramidal neurons in the Fmr1
KO mouse. This would be expected to cause hyperexcitability of visual neocortical circuitry.
Interestingly, we also discovered that group II mGluR-mediated modulation of GABA synapses
onto excitatory cells is normal in the Fmr1 KO mouse, suggesting that the alterations we
observed in group II mGluR-mediated modulation of inhibition is occurring upstream from the
GABA synapse. We hypothesize that group II mGluRs are modulating excitatory transmission
onto inhibitory neurons, and that this is the likely location of dysfunction in the Fmr1 KO mouse.
Given that both FS and NFS interneurons have previously been implicated in FXS neocortical
pathogenesis, it is possible that one or both of these interneurons are involved in the exaggerated
suppression of inhibition that we observe onto layer 5/6 pyramidal neurons. To that end, we
performed whole-cell recordings from both FS and NFS interneurons in the layer 5/6, and

	

138	

quantified the changes we observed in group II mGluR-mediated modulation of excitation onto
these neurons. We demonstrate that group II mGluR-mediated suppression of glutamatergic
transmission onto FS interneurons remains unaltered in the Fmr1 KO. In contrast, group II
mGluR-mediated modulation of glutamatergic transmission onto NFS interneurons is
exaggerated in the Fmr1 KO mouse, leading to decreased excitation of these interneurons
relative to WT NFS cells. We further show that group II mGluRs are mediating this effect
presynaptically, presenting new evidence in support of a presynaptic role for FMRP. Finally, we
show that this alteration in the Fmr1 KO mouse is layer specific, as excitatory synapses onto
NFS interneurons in the layer 2/3 do not show the same alterations that we observe in the layer
5/6.
4.2 Materials and Methods
All experimental procedures were performed in accordance with the National Institutes of
Health Guide for Care and Use if Laboratory Animals and were approved by the Michigan State
University Institutional Animal Care and Use Committee.
Mice on the FVB background (postnatal age 16-21 days) of either sex were anesthetized
with isoflurane and perfused with cold, oxygenated (5% CO2-95% O2), modified saline prior to
decapitation. Brains were removed and transferred to cold (4°C), oxygenated slicing solution
containing the following (in mM): 2.5 KCl, 1.25 NaH2PO4, 10.0 MgSO4, 0.5 CaCl2, 26.0
NaHCO3, 10.0 glucose, and 234.0 sucrose. Slices (300 µm thickness) were collected in a
coronal plane were collected using a Leica vibrating tissue slicer, and then transferred to a
holding chamber filled with oxygenated physiological saline containing (in mM): 126.0 NaCl,
2.5 KCl, 1.25 NaH2PO4, 2.0 MgCl2, 2.0 CaCl2, 26.0 NaHCO3, and 10.0 glucose. Slices were

	

139	

incubated at 32 ± 1°C for at least 30 minutes, and then left to rest at room temperature for at least
30 minutes prior to recording.
Whole-cell recordings were obtained using a microscope equipped with differential
interface contrast optics (Zeiss, Thornwood, NY). Single slices were transferred to a submersion
chamber continuously superfused at 2.5ml/min with oxygenated physiological saline heated to
32◦C. Visual neocortex was identified with a low power (5x) objective and individual neurons
were identified using a water-immersion objective (63x). For a subset of experiments,
fluorescent microscopy was used to visualize GFP-expressing neurons in two different
transgenic mouse lines. These transgenic lines have been shown to express green fluorescent
protein (GFP) in a subset of either parvalbumin positive (CB6-Tg(Gad1-EGFP)G42Zjh/J/Jax
labs) or somatostatin positive interneurons (FVBTg(GadGFP)45704Swn/Jax labs) under the
control of the mouse Gad1 gene. Neuron subtypes were further distinguished by their
characteristic electrophysiological properties as determined using an 800-2000 ms current step
injection protocol. PV+ interneurons exhibit the following intrinsic properties: more
hyperpolarized membrane potentials, higher threshold to fire action potentials, nonaccommodating spike firing, and abrupt episodes of nonadapting repetitive discharges
(Kawaguchi and Kubota, 1997). Relative to parvalbumin interneurons, somatostatin
interneurons have higher input resistances, more depolarized membrane potentials, both early
and late after-hyperpolarizing potentials (AHPs), and moderate spike frequency adaptation
(Beierlein et al., 2000; Beierlein et al., 2003). Pyramidal neurons were identified through
preferential targeting of pyramidal-shaped soma with a diameter greater than 20 µm, in addition
to observing their electrophysiological properties. The majority of pyramidal neurons (62%)
exhibit a regular spiking (RS) slowly adapting firing pattern, with a moderate sized late after-

	

140	

hyperpolarizing potential (AHP) and most often no visible early (AHP) or depolarizing
afterpotential (DAP) (Chang and Luebke, 2007; Larkman and Mason; 1990). In contrast, SST+
interneurons always display a prominent early AHP and a DAP, making this an important factor
for distinguishing between the two cell types. Another less common (5%), but easily
distinguished electrophysiological profile of layer 5/6 pyramids was that of intrinsic bursting
firing (IB) of 3-6 bursts of APs (Chang and Luebke, 2007; Larkman and Mason; 1990). For
recordings, pipettes made of borosilicate capillary glass were pulled to tip resistances of 4-6MΩ
and then filled with intracellular solution containing (in mM): 117.0 K-gluconate, 13.0 KCl, 1.0
MgCl2, 0.07 CaCl2, 0.1 EGTA, 10.0 HEPES, 2.0 Na2-ATP, 0.4 Na-GTP, and 0.2% biocytin.
Only cells with a membrane potential ≥ -50 mV and overshooting action potentials were
included for analysis. Access resistance was monitored throughout the experiments and
recordings were not included if the access resistance increased to >20MΩ.
In recordings from layer 5/6 FS and NFS interneurons, excitatory postsynaptic currents
(EPSCs) were evoked by electrical stimulation of afferents in layer 2/3 (Figure 2.1). For layer
2/3 recordings, a stimulating electrode was placed in layer 4 of the visual neocortex. For EPSC
recordings, the chloride channel blocker, 4,4'-dinitrostilbene-2,2'-disulfonic acid (DNDS, 1 mM)
was included in the recording pipette to block GABAA–mediated IPSCs, effectively isolating
EPSCs in the recorded neurons. Stimuli were delivered at 0.1 Hz, and the peak amplitudes and/or
charge of the evoked EPSCs was measured before, during, and after washout of the selective
group I and group II mGluR agonists, (RS)-3,5-dihydroxyphenylglycine (DHPG) and (2R,4R)-4aminopyrrolidine-2,4-dicarboxylate (APDC) respectively. Miniature EPSCs (mEPSCs) were
recorded in the presence of tetrodotoxin (1µM). During strontium experiments, the calcium
chloride in the physiological saline was replaced with strontium chloride (2 mM).

	

141	

All compounds used in this study were prepared from stock solutions stored at the
appropriate temperature and diluted to the correct concentration with physiological saline prior to
administration. All agonists were delivered by bolus injection to the central line of the recording
chamber using a syringe pump. Based upon the calculated flow rate of the main line, the
delivered drug concentration was estimated to be one-fourth of the initial drug concentration.
DNDS was purchased from Life Technologies, while all other compounds were purchased from
Tocris and Sigma-Aldrich.
All recordings were acquired using a Multiclamp 700B amplifier (Molecular Devices).
Data was filtered at 2.5 kHz and digitized at 10 kHz using pClamp software. For evoked
experiments, 12 consecutive sweeps (2 minutes) were averaged to determine the baseline peak
amplitude and/or charge of the synaptic response. The drug effect was determined as the average
of 3 consecutive sweeps during the maximal drug response. A positive drug response was
defined as a change in amplitude or charge that was greater than 2 standard deviations of the
baseline response. The frequency and amplitude of mEPSCs or evoked responses in the
presence of strontium were analyzed using MiniAnalysis software (Synaptosoft). All data are
reported as the mean ± SEM and analyzed using either paired or independent t-test were
applicable. Kolmogorov-Smirnoff tests were used to determine if mEPSC frequency and/or
amplitude were altered within cells by drug application. A p-value <0.05 was considered
statistically significant.
4.3 Results
A. The intrinsic properties of Fmr1 KO layer 5/6 FS interneurons are similar to WT.
After investigating the intrinsic properties of excitatory cells in the layer 5/6, we next
questioned whether the intrinsic properties of interneurons in the Fmr1 KO mouse differ from

	

142	

those of the wild type. We began our investigation quantifying both input resistance and
membrane potential of WT and Fmr1 KO FS interneurons (Figure 4.1Ai and ii), which
comprise the major interneuron subtype in the visual neocortex (Markram et al., 2004). In the
WT mouse, the average input resistance was 171.7 ± 19.3 MΩ (n=22, Figure 4.1 Ci). The Fmr1
KO FS interneurons had an input resistance similar to that seen in WT FS interneurons. (146.5 ±
21.5 MΩ, n=10) and thus not statistically significant (p>0.05, t-test, Figure 4.1 Ci). Next, we
quantified membrane potentials of WT and Fmr1 KO interneurons. WT FS interneurons had an
average membrane potential of -73.8 ± 1.2 mV (n=22, Figure 4.1 Cii), and Fmr1 KO cells had a
slightly more hyperpolarized potential (-74.1 ± 1.5 mV, n=10, Figure 4.1 Cii), however this was
not statistically different from WT neurons (p>0.05, t-test, Figure 4.1Cii). These results are
summarized in the following table (Table 2.4).
Table 4.1. Intrinsic properties of layer 5/6 FS interneurons in WT and Fmr1 KO mice.

Input Resistance

Membrane Potential

Fmr1 KO

171.7 ± 19.3
(n=22)
146.5 ± 21.5
(n=10)

-73.8 ± 1.2
(n=22)
-74.1 ± 1.5
(n=10)

p

0.39
t-test

0.87
t-test

Wild Type

B. Group I mGluRs do not modulate EPSCs in FS interneurons.
Prior to investigating the modulation of FS interneuron EPSCs by group II mGluRs, we
first wanted to determine whether group I mGluRs modulate excitatory synaptic transmission in
layer 5/6 FS interneurons. To accomplish this task, we obtained whole cell recordings from
layer 5/6 FS interneurons in WT and Fmr1 KO mice. We chose to investigate these interneurons
first because they have previously been implicated in neocortical Fragile X pathology (Gibson et
al., 2008; Patel et al., 2013), and because they are the most abundant neocortical interneuron

	

143	

subtype (Rudy et al., 2011). FS interneurons were identified based on their unique physical and
electrophysiological properties. Specifically, they are smaller in size, relatively hyperpolarized,
exhibit non-accommodating action potential firing to depolarizing current steps, and receive a
multitude of spontaneous excitatory inputs visible throughout a series of current steps delivered
upon patching the cell (Kawaguchi and Kubota, 2006). Additionally, in some instances FS
interneurons were visualized using transgenic mice that express GFP in PV+ interneurons. We
evoked EPSCs in WT and Fmr1 KO FS interneurons by electrical stimulation of layer 2/3 visual
neocortex (20-100µA, 0.1Hz, Figure 4.2 Ai and Bi). The baseline EPSC amplitude in control
animals averaged 520.4 ± 86.6 pA, and application of DHPG (25µM, 30s) had no effect on
EPSC amplitude (n=11, DHPG: 491.8 ± 81.4 pA, p>0.05, paired t-test, Figure 4.2 Aii-iii). In
Fmr1 KO FS interneurons (Figure 4.2 Bi), the average pre-drug amplitude of the EPSC was
343.1 ± 46.4 pA, and application of DHPG to these cells also failed to produce significant
changes in the amplitude of the response (n=9, DHPG: 348.9 ± 45.9, p >0.05, paired t-test,
Figure 4.2 Bii-iii). The average amplitude of the EPSC response to DHPG in WT FS
interneurons was 95.2 ± 1.6 % of control, and the Fmr1 KO FS interneuron EPSC response was
102.2 ± 2.5% of control (WT=11, Fmr1 KO=9, Figure 4.2C). From this data set, we conclude
that group I mGluRs do not modulate glutamatergic transmission in layer 5/6 FS interneurons.
C. Group II mGluRs modulate EPSCs on FS interneurons similarly between WT
and Fmr1 KO.
Recall from the previous chapter that we observed an exaggerated group II mGluRdependent suppression of the IPSC amplitude in pyramidal neurons from Fmr1 KO mice. We
also determined that activating group II mGluRs while simultaneously blocking excitatory drive
onto neocortical interneurons leads to IPSC suppression in pyramidal neurons that is similar in

	

144	

amplitude between Fmr1 KO and WT. With these results, we are now certain that the change we
observe in the IPSC amplitude between WT and Fmr1 KO pyramidal neurons is due to
alterations in group II mGluR-mediated modulation of excitatory drive onto interneurons.
Which interneurons are responsible for this effect, however, remains unknown. To that end, we
next sought to determine how group II mGluRs modulate excitatory transmission onto specific
neocortical interneuron subtypes. We obtained whole cell recordings from layer 5/6 FS
interneurons in WT and Fmr1 KO mice and tested the effects of APDC on the evoked responses
of WT and Fmr1 KO FS interneurons (Figure 4.3 Ai and Bi). In WT animals, the average
control response was 523.7 ± 89.7 pA, and this was significantly attenuated by APDC (204.4 ±
34.8 pA, n=11, p<0.05, paired t-test, Figure 4.3Aii-iii). In Fmr1 KO interneurons, the average
response prior to APDC was 319.1 ± 39.1 pA, and this response was also significantly attenuated
by the group II agonist (135.0 ±18.7 pA, n=7, p<0.05, paired t-test, Figure 4.3Bii-iii).
Comparison of the change in amplitude between WT and Fmr1 KO, expressed as % of the
control response, determined that the APDC suppression of EPSC amplitude was not different
between phenotypes (WT: 40.8 ± 3.3 %, Fmr1 KO: 43.1 ± 5.5p>0.05, t-test, Figure 4.3C).
These results indicate that group II mGluR-mediated suppression of EPSCs in FS interneurons is
intact in the Fmr1 KO, and that this interneuron subtype is not responsible for the exaggerated
suppression of inhibition by APDC that we see in Fmr1 KO layer 5/6 pyramidal neurons.
D. Fmr1 KO layer 5/6 NFS interneurons have a decreased input resistance compared to
WT.
NFS interneurons are the second largest subgroup of inhibitory cells in the neocortex
(Kawaguchi and Kubota, 1997; Markram et al., 2004). For this reason, we also included them in
our analysis of the intrinsic properties of neocortical interneurons. We recorded a series of

	

145	

current-step injections from both WT and Fmr1 KO NFS interneurons (Figure 4.1 Bi and ii),
and compared both input resistance and membrane potential. The average input resistance of
wild type layer 5/6 NFS interneurons in the visual neocortex was 359.8 ± 18.1 MΩ (n=39,
Figure 4.1Di), and this was significantly larger than the input resistance we observed in FS
interneurons (Figure 4.1Ci), supporting previously published literature (Beierlein, Gibson, &
Connors, 200; Kawaguchi and Kubota, 1997). In Fmr1 KO interneurons, the average input
resistance was smaller than that observed in WT NFS interneurons (306.9 ± 17.3 MΩ, n=32,
Figure 4.1Di), and this difference was statistically significant (p = 0.038, t-test, Figure 4.1Di).
We also measured membrane potential in both WT and Fmr1 KO cells (WT: -65.9 ± 0.9 mV, n=
39; Fmr1 KO: -67.1 ± 0.9 mV, n=32, Figure 4.1Dii), but found no significant differences
between genotypes (p = 0.36, t-test, 4.1Dii). However, the average membrane potential of NFS
interneurons was more depolarized than FS interneurons, and this is what we expected to see
based on previous studies (Beierlein, Gibson, & Connors, 2000). The decrease in input resistance
of Fmr1 KO NFS interneurons that we report, indicates that these interneurons are less
intrinsically excitable than layer 5/6 wild type NFS interneurons. These results are summarized
below (Table 2.5)
Table 4.2. Intrinsic properties of layer 5/6 SST+ interneurons in WT and Fmr1 KO mice.

Input Resistance

Membrane Potential

Fmr1 KO

359.8 ± 18.1
(n=39)
306.9 ± 17.3
(n=32)

-65.9 ± 0.9
(n=39)
-67.1 ± 0.9
(n=47)

p

0.038*
t-test

0.928
t-test

Wild Type

	

146	

E. Fmr1 KO NFS interneurons show an exaggerated response to group II mGluR
activation.
After determining that modulation of excitatory drive onto Fmr1 KO FS interneurons by
group II mGluRs is comparable to wild type, we next wondered whether the same would be true
of modulated excitatory drive onto NFS interneurons. Do group II mGluRs modulate the
excitatory drive of NFS interneurons similarly in WT and Fmr1 KO mice? For these
experiments, we recorded from NFS interneurons in layer 5/6 and electrically evoked EPSCs in
these neurons by stimulating layer 2/3 (Figure 2.1). In these cells we observed a multi-peaked
EPSC response to electrical stimulation (Figure 4.4 Ai and Bi), and so we measured the area of
the evoked response (charge) in these neurons. In WT animals, the control EPSC averaged 2.9 ±
0.4 pC, and application of APDC led to a significantly attenuated response (APDC: 1.1 ± 0.2 pC,
n=13, p<0.001, paired t-test, Figure 4.4 Ai-iii). In Fmr1 KO animals, the NFS EPSC was also
suppressed by APDC (Figure 4.4 Bi), and this was statistically significant (Pre: 3.2 ± 0.6 pC,
APDC: 0.7 ± 0.2 pC, n=10, p<0.001, paired t-test, Figure 4.4 Bii-iii). In the presence of APDC,
the average WT EPSC amplitude was 38.2 ± 2.9 % of the control response, while the Fmr1 KO
was 22.7 ± 2.3 % of control, indicating a 15% increase in suppression of glutamatergic
transmission by group II mGluRs in the Fmr1 KO mouse. Statistical analysis determined this
difference to be significant (p<0.001, t-test, Figure 4.4 C). Thus, Fmr1 KO mice show
exaggerated suppression of NFS interneuron excitatory currents, mediated by group II mGluR
activation.
F. Group II mGluR-mediated suppression of excitatory transmission onto NFS
interneurons is mediated presynaptically.
The suppression of EPSCs onto NFS interneurons could be due to activation of either

	

147	

pre- or post-synaptic group II mGluRs. To determine the site of action, we next assessed
miniature EPSCs (mEPSCs) in the presence of APDC (75µM, 40s). Changes in the frequency of
mEPSCs are indicative of a change in release probability, and hence a presynaptic mechanism of
origin. Alternatively, changes in mEPSC amplitude reflect a change in pre- and/or post-synaptic
mechanisms. We recorded mEPSCs from NFS interneurons in both WT (4.5 Ai) and Fmr1 KO
mice (4.5 Bi). The average baseline mEPSC amplitude in WT mice was 21.2 ± 0.7 pA, and
application of APDC did not alter mEPSC amplitude in 3 of 4 cells tested (APDC: 20.5 ± 1.1
pA, p >0.05, K.S test). For mEPSC frequency, the baseline response was 3.7 ± 1.4 Hz, and
application of APDC did not alter frequency in any cells tested (n=4, APDC: 0.8 ± 0.3 Hz, p>
0.05, K.S.test). Population averages of mEPSC amplitude and frequency were unaltered by
APDC (p > 0.05, paired t-test, Figure 4.5 Aii and iii). Activation of group II mGluRs had no
effect on mEPSC amplitude in 3 of 4 Fmr1 KO cells tested (Pre-drug: 26.0 ± 2.8 pA, APDC:
21.7 ± 1.1 pA, p > 0.05, K.S test). Similarly, there was no change in KO mEPSC frequency in 3
of 4 cells (Pre-drug: 1.2 ± 0.4 Hz, APDC: 0.8 ± 0.4 Hz, p>0.05, K.S. test). The average
population mEPSC response amplitude upon application of APDC, was not statistically
significant (p> 0.05, paired t-test, Figure 4.5 Bii), and this was also true for mEPSC frequency
(p > 0.05, paired t-test, Figure 4.5 Biii). We believe that no significant differences were
observed in either mEPSC frequency or amplitude because the baseline number of excitatory
events was minimal, indicating that pyramidal neurons are not participating in significant levels
of AP-independent neurotransmitter release in a slice preparation. Subsequently, two alternative
methods to localize group II mGluRs at the excitatory synapse onto SST+ interneurons were
investigated.

	

148	

We next quantified the paired-pulse ratio (PPR) of two evoked excitatory currents with
an interstimulus interval of 30-100ms (Figure 4.6). Any changes in paired-pulse ratio have been
shown to be indicative of a change in presynaptic release probability. In 3 of 6 wild-type cells
tested (50%, Figure 4.6 C), application of APDC (15-75µM, 40s) led to a significant change in
the paired pulse ratio (Figure 4.6 Ai-iii). Of these three cells, two showed a decrease in the ratio
while one cell showed an increase. The paired pulse ratio was also significantly altered by
APDC application in 3 of 5 Fmr1 KO NFS interneurons (60%, Figure 4.6 Bi-iii). In all 3 cells,
the paired pulse ratio increased (Figure 4.6 C). The mixed effects we see in paired pulse ratio
did not allow us to determine the site of action of the group II mGluRs.
Finally, we quantified changes in excitatory events evoked in the presence of modified
extracellular solution containing strontium (2mM) in place of calcium (Figure 4.7). Strontium
prolongs the duration of neurotransmitter release and creates asynchronous release, thereby
reducing the temporal summation of events postsynaptically. This occurs due to strontiumbinding of voltage gated calcium channels at the axon terminal, effectively prolonging the
postsynaptic excitatory response and separating it into smaller “quantal-like” events from which
frequency and amplitude measurements can be quantified. In WT NFS interneurons (Figure
4.7Ai), the average baseline number of events was 13.4 ± 1.6, and APDC significantly reduced
the number of events to 4.6 ± 1.4 (n=11, p < 0.05, paired t-test). A similar trend was observed in
Fmr1 KO animals (Figure 4.7 Bi). Prior to drug application, the average baseline number of
events in Fmr1 KO NFS cells was 10.6 ± 4.8, and these events were significantly decreased in
the presence of APDC (n=5, APDC: 6.0 ± 4.4, p<0.05, paired t-test). APDC decreased the
number of events in the WT to 31.7 ± 7.7 % of control, while it decreased the Fmr1 KO number
of events to 42.8 ± 15.9 % of control (Figure 4.7D). We also assessed changes in amplitude of

	

149	

the strontium excitatory events to ascertain potential postsynaptic effects. In WT animals
(Figure 4.7 Ai), 6 of 8 NFS interneurons show no difference in amplitude of the events in
response to APDC (K.S. test, p > 0.05). The pooled amplitude of all WT NFS neurons in the
presence of APDC was 78.2 ± 6.0 % of control (Figure 4.7 C). In the Fmr1 KO mouse (Figure
4.7 Bi), 4 of 5 NFS interneurons show no change in amplitude after application of APDC
(K.S.test, p > .05). The average amplitude for all Fmr1 KO NFS neurons upon application of
group II mGluRs was 92.7 ± 3.5 % of control (Figure 4.7 C). Using these results, we conclude
that although group II mGluRs are present both pre- and post- synaptically at glutamatergic
synapses onto NFS interneurons, the major site of action of group II mGluR-mediated
suppression of evoked EPSCs is likely presynaptic.
G. The intrinsic properties of layer 2/3 NFS interneurons are similar between WT and
Fmr1 KO mice.
Is the group II mGluR-mediated suppression of excitatory drive onto Fmr1 KO NFS
interneurons specific to layer 5/6? To answer this question, we performed whole cell recordings
from NFS interneurons in layer 2/3 of the visual neocortex. We first quantified both input
resistance and membrane potential of WT and Fmr1 KO NFS interneurons (Figure 4. 8Ai and
ii). WT NFS interneurons had an average input resistance of 321.2 ± 47.7 MΩ (n=16, Figure
4.8 B). In Fmr1 KO mice (Figure 4.8 Aii), the input resistance was less than that seen in WT
neurons (235.1 ± 17.8 MΩ, n=9), but this decrease was not significant (p>0.05, t-test, Figure
4.8B). We also measured resting membrane potential, and determined the average value to be 64.6 ± 1.6 mV in wild type mice (n=16, Figure 4.8C). The average resting membrane potential
in Fmr1 KO NFS interneurons was -67.1 ± 1.5 mV, and this was not significantly different from

	

150	

the wild type cells (n=9, p>0.05, independent t-test, Figure 4.8C). These results are
summarized below (Table 2.6).
Table 4.3. Intrinsic properties of layer 2/3 SST+ interneurons in WT and Fmr1 KO mice.

Input Resistance (MΩ)

Membrane Potential
(mV)

Fmr1 KO

321.2 ± 47.7
(n=16)
235.1 ± 17.8
(n=9)

-64.6 ± 1.6
(n=16)
-67.1 ± 1.5
(n=9)

p

0.12
t-test

0.58
t-test

Wild Type

H. Exaggerated group II mGluR-mediated suppression of Fmr1 KO NFS interneuron
EPSCs is layer specific.
After quantifying these basic intrinsic properties, a stimulating electrode was placed into
layer 4 of the visual neocortex to evoke excitatory currents in layer 2/3 NFS interneurons
(Figure 2.1). The baseline area response of wild type NFS interneurons was 4.0 ± 0.4 pC and
APDC application significantly reduced the area to 0.9 ± 0.2 pC (n=16, p<0.001, paired t-test,
Figure 4.9 Ai-iii). The area response after APDC application was 21.2 ± 2.8 % of the control
response (n=16, Figure 4.9C). In Fmr1 KO superficial NFS interneurons, a similar change was
observed (Figure 4.9 Bi-iii). The average pre-drug response measured 3.1 ± 0.5 pC, and after
APDC application the average EPSC charge was reduced to 0.9 ± 0.2 pC (n=9, p<0.001, paired
t-test, Figure 4.9 10Bi-iii). The Fmr1 KO response in the presence of APDC, showed a 67.4 ±
2.8% reduction (n=9, Figure 4.9C). Comparison of the area response between WT and Fmr1
KO animals was not significant (p>0.05, t-test, Figure 4.9 C). Thus, we determined that the
enhanced suppression we see in Fmr1 KO layer 5/6 NFS interneurons in response to group II
mGluR activation is a layer specific dysregulation of the visual neocortex.

	

151	

4.4 Discussion
In these studies, we investigated the effect of group II mGluR activation on the
modulation of excitatory drive onto visual neocortical interneurons in the Fmr1 KO mouse.
Although neocortical hyperexcitability is speculated to be a major contributor of several Fragile
X phenotypes, functional neocortical studies remain underrepresented in the Fragile X literature.
Here we report that group II mGluR activation leads to suppression of glutamatergic
transmission onto NFS interneurons in the layer 5/6 visual neocortex, and this process is
exaggerated in the Fmr1 KO mouse (Figure 4.11). We have also shown previously that this
leads to disinhibition onto layer 5/6 pyramidal neurons in the Fmr1 KO, and presumably creates
a hyperexcitable layer 5/6 network (Figure 4.11).
Perhaps one of the most interesting results of this study, was the observation that group II
mGluR signaling is altered in the visual neocortex in the absence of FMRP. This was highly
unexpected, as only one study to-date has ever implicated group II mGluRs in FXS pathology.
This study found that chronic administration of a group II mGluR antagonist could restore the
exaggerated mGluR-LTD phenotype observed in the hippocampus of Fmr1 KO mice (Choi et
al., 2011). It was difficult for the authors to explain these results as no previous studies have
experimentally demonstrated the presence of FMRP downstream of the group II mGluR
signaling cascade. These authors suggest that the canonical signaling cascades of group I and
group II mGluRs crosstalk due to the fact that they are both capable of binding Gq and Gi gproteins. Thus, the authors suggest that it is actually the loose association of group II mGluRs
with Gq signaling that may be responsible for the restoration of LTD in the presence of a group
II mGluR antagonist. We do not believe this to be the case in our results. It is now known that
FMRP does not interact directly with the canonical Gq signaling pathway (Huber et al., 2001;

	

152	

Gallagher et al., 2004), so in the event that presynaptic group II mGluRs did loosely associate
with Gq binding protein, we would expect this to have no effect when activated in the Fmr1 KO
mouse. Interestingly, two other pathways that are modulated by group II mGluRs include the
ERK/MAPK pathway and the PI3K-mTOR pathway (Niswender and Conn, 2010). Both of
these pathways lie downstream from group I mGluRs. The PI3K-mTOR pathway interacts with
activates FMRP protein to negatively regulate protein translation (Bhakar, Dolen, & Bear.,
2012), so it is possible that altered modulation of this pathway by group II mGluRs in the
absence of FMRP is causing the changes we observe. Additionally, group II mGluRs bind
presynaptic ion channels, the result of which is decreased neurotransmitter release. It has been
shown that the Bγ subunit of mGluRs can interact with both K+ and Ca2+ channels, including P/Q
Ca2+ channels that are abundantly located at the axon terminal (O’Connor et al., 1999). The
mRNA targets of FMRP also include regulate the several different potassium channels and most
calcium channels (Brager and Johnsten, 2014). Therefore, I suggest that some of the same ion
channels modulated by group II mGluRs are upregulated in the absence of FMRP. Finally, Bγ
subunits have also been shown in binding assays to associate directly with exocytotic release
machinery proteins including isolated binding with syntaxin and/or SNAP-25 as well as
combined binding with SNARE complexes comprised of syntaxin, SNAP-25, and VAMP
(Blackmer et al., 2001). These proteins are also targets of FMRP. Thus, it is possible that
activation of group II mGluRs in the Fmr1 KO leads to altered inhibition of release machinery
by GBγ upon dissociation from the g-protein complex.
Another interesting result of our study was the discovery that the deficits we see in group
II mGluR signaling in the Fmr1 KO are mediated presynaptically. It is widely accepted that
group II mGluRs are located abundantly in the extrasynaptic space of presynaptic neurons, where

	

153	

they can be activated in instances of excess glutamate in the synapse (Scanziani, 1997).
However, the axonal or presynaptic role of FMRP remains largely unknown. FMRP targets
mRNAs that comprise one-third of the presynaptic proteome (Darnell et. al 2011) and is
expressed presynaptically in granules (Antar et al., 2006; Christie et al., 2009; Till et al., 2012,
Akins et al., 2009; Akins et al., 2012), indicating that its absence is likely to have drastic
consequences on presynaptic function of the synapse. A recent electrophysiological study using
Fmr1 KO mosaic mice, demonstrated that loss of presynaptic FMRP recapitulates the decrease in
excitatory drive onto FS interneurons in layer 4 of the somatosensory cortex (Patel et al, 2013).
Deletion of postsynaptic FMRP could not recapitulate this deficit. Furthermore, loss of
presynaptic FMRP had no effect on excitatory synapses onto excitatory neurons, indicating that
the presynaptic function of FMRP at excitatory neurons is target cell specific. This is intriguing
when we consider that our data shows defective excitatory drive onto NFS interneurons with no
changes in excitatory drive onto pyramidal neurons, and this effect is mediated by presynaptic
group II mGluRs. Incidence of this has been observed in a human with a unique point mutation
in the Fmr1 gene that led to loss of function of presynaptic FMRP with intact function of
postsynaptic FMRP. The result of this mutation was a child with global developmental delay,
intellectual disability, and intractable seizures (Myrick et al., 2015).
What is the consequence for visual processing in the Fmr1 KO mouse in the presence of
exaggerated group II mGluR signaling onto NFS interneurons? This question requires
understanding the role of NFS interneurons in the neocortical network. Local glutamatergic
inputs from pyramidal neurons onto NFS interneurons are facilitating, unlike those of PV+
interneurons which are depressing. This facilitating capability, allows SST+ interneurons to
increase their feedback inhibition as the rate and duration of presynaptic discharge increases. It

	

154	

is speculated that this unique feature, makes NFS interneurons likely to be preferentially
activated during periods of high network activity. Repetitive stimulation during high network
activity will elevate glutamate levels in the synapse (Scanziani et al., 1997), and this excess
glutamate would likely spillover at the synapse and activate extrasynaptically located group II
mGluRs. This activation, in turn, would help limit further cortical excitation by decreasing
neurotransmitter release. In the case of the Fmr1 KO mouse, however, the preferential
exaggeration of glutamatergic suppression at pyramidal neuron synapses onto somatostatin
interneurons, would be expected to cause an imbalance of the ratio of excitation/inhibition in
layer 5/6 neocortical circuitry tending toward hyperexcitability. The majority of layer 5/6
pyramidal neurons in the visual cortex send feedback projections to layer 2/3 neurons that
interact with higher order visual centers. So it is likely that these centers will also be inundated
with excitatory input, which could cause problems with sensory integration and potentially
epileptic activity. Further, some layer 5/6 pyramidal neurons project to subcortical structures
such as the pulvinar nucleus of the thalamus and the superior colliculus, suggesting that these
structures may also be affected.
One limitation of our study is that we did not investigate the third interneuron subtype
found in the layer 5/6, namely the serotonin 5HT3a receptor expressing interneurons. It is,
therefore, possible that group II mGluRs modulate excitatory transmission of these interneurons,
and that this modulation is altered in the Fmr1 KO mouse. However, we speculate that because
these interneurons comprise no more than 15% of layer 5/6 interneurons (Rudy et al., 2011), it is
likely that any potential changes that exist are relatively small compared to the alterations we see
in excitation of NFS interneurons. Additionally, methodological limitations do not allow us to
precisely identify the source of the glutamatergic inputs onto our NFS interneurons, and so we

	

155	

cannot say with certainty which excitatory axons are affected in the Fmr1 KO. However, we
hypothesize that the majority of excitatory inputs that we were stimulating originate within V1.
In several instances the subcortical area of our slices was removed, and we still observed changes
in group II mGluR signaling onto NFS neurons. Furthermore, thalamic afferents are not known
to project to layer 5, and have very weak connections to NFS neurons in general (Beierlein and
Connors 2000; Beierlein and Connors 2003).
In conclusion, we now add to the body of literature by implicating group II mGluRs in
the pathology of the visual system in FXS.

Human studies indicate that FMRP is particularly

abundant in regions of the magnocellular pathway, and that this pathway may be more
susceptible to loss of FMRP. This anatomical study is supported with evidence showing that
children with FXS have reduced sensitivity for psychostimuli that probe the M pathway but
maintain sensitivity to those stimuli that probe the P pathway (Kogan et al., 2004). Affected
individuals with FXS also show marked visual-motor deficits. V1 of the Fmr1 KO recapitulates
the abnormal spine morphology seen in layer 5/6 visual neocortex of humans, and also exhibits a
hyperconnected profile in magnetic resonance imaging studies (Haberl et al., 2015). Thus it is
our hope, that this discovery leads the way to understanding the circuit mechanisms that give rise
to these visual defects, and that restoration of inhibition may potentially serve as a means to
rescue these sensory impairments.

	

156	

APPENDIX

	

157	

Figure 4.1. Intrinsic excitability is decreased in layer 5/6 Fmr1 KO visual neocortical NFS
interneurons, but not FS interneurons.
Ai. Representative WT trace of a layer 5/6 FS PV+ interneuron obtained from a current-step
injection protocol. ii. Same as Ai but for Fmr1 KO. Bi. Representative WT trace of a layer 5/6
NFS SST+ interneuron obtained from a current-step injection protocol. ii. Same as Bi but for
Fmr1 KO. Ci. Average input resistance of FS PV+ interneurons for both WT (n=22) and Fmr1
KO (n=10). No significant differences in input resistance or membrane potential were observed
in PV+ interneurons between genotypes (p > 0.05, t-test). Di. Average input resistance of NFS
	

158	

Figure 4.1. (cont’d)
SST+ interneurons for both WT (n=39) and Fmr1 KO (n=32). ii. Average membrane potential of
NFS SST+ interneurons for both WT (n=39) and Fmr1 KO (n=22). No significant differences in
input resistance or membrane potential were observed in NFS SST+ interneurons between
genotypes (p > 0.05, t-test).

	

159	

Figure 4.2. Group I mGluRs do not modulate excitatory synaptic transmission onto WT
and Fmr1 KO layer 5/6 FS interneurons.

	

160	

Figure 4.2. (cont’d)
Ai. Representative trace showing the response of a layer 5/6 wild type FS interneuron EPSC to
brief application of group I mGluR agonist DHPG (25µM). Aii. Raw time course of the change
in peak amplitude in a single wild type FS cell before, during, and following application of
DHPG. Aiii. Wild type normalized time course of the population peak amplitude change to
DHPG application (n=11). Bi-iii. Same as Ai-iii, but for Fmr1 KO. C. Average FS interneuron
EPSC amplitude expressed as % of control for all cells tested in both WT and Fmr1 KO
(WT=11, KO=9). No statistical differences were found between genotypes (p>0.05, t-test).

	

161	

Figure 4.3. Group II mGluR-mediated modulation of excitatory transmission onto FS
interneurons is unaltered in the visual neocortex of the Fmr1 KO mouse.
Ai. Representative trace showing a wild type FS interneuron EPSC evoked by electrical
stimulation of layer 2/3. Traces show WT EPSC response before, during, and after brief

	

162	

Figure 4.3. (cont’d)
application of APDC (75 µM). Aii. Raw time course of the change in WT FS interneuron EPSC
amplitude in response to APDC. Aiii. WT population data showing the normalized time course
of the change in peak amplitude in response to APDC (n=11). Bi-iii. Same as Ai-iii, but for
Fmr1 KO (n=7). C. Average FS interneuron EPSC amplitude in response to APDC expressed as
% of control for both WT and Fmr1 KO (WT=11, KO=7). No statistical differences were found
between peak amplitude of WT and Fmr1 KO in the presence of APDC (p > 0.05, t-test).

	

163	

Figure 4.4. Group II mGluR-mediated suppression of excitatory transmission onto NFS
interneurons is exaggerated in the visual neocortex of the Fmr1 KO mouse.
Ai. Representative trace showing a wild type NFS interneuron EPSC evoked by electrical
stimulation of layer 2/3. Traces show WT EPSC response before, during, and after brief
application of APDC (75 µM). Aii. Raw time course of the change in NFS WT EPSC charge
before, during, and after APDC application. Aiii. WT population data showing the normalized

	

164	

Figure 4.4. (cont’d)
time course of the charge in response to APDC (n=13). Bi-iii. Same as Ai-iii, but for the Fmr1
KO. C. Average NFS interneuron population EPSC charge in the presence of APDC expressed
as % of control for both wild type and Fmr1 KO (WT=13, KO=10). Fmr1 KO mice show a
statistically significant increase in the group II mGluR-mediated suppression of excitatory drive
onto SST+ interneurons compared to WT (p = 5.3 x 10-4, t-test).

	

165	

Figure 4.5. Group II mGluR activation does not alter miniature EPSC amplitude or
frequency in WT and Fmr1 KO layer 5/6 visual neocortical NFS interneurons.
Ai. Raw data showing a WT miniature EPSC trace before, during, and after recovery from
application of group II mGluR agonist APDC (75 µM). Aii. WT population data showing
average change in mEPSC amplitude in response to group II mGluR activation (WT=4). Aiii.
Population data showing WT mEPSC frequency change in response to APDC (WT=4). Pooled
population data show no change in WT amplitude or frequency in response to application of
APDC (p > 0.05, t-test). Bi-iii. Same as Ai-iii, but for Fmr1 KO. Statistical analysis reveals no

	

166	

Figure 4.5. (cont’d)
significant differences in mEPSC amplitude or frequency in response to group II mGluR
activation in Fmr1 KO layer 5/6 NFS interneurons.

	

167	

Figure 4.6. Group II mGluR-mediated suppression of EPSCs in layer 5/6 NFS interneurons
leads to a change in paired pulse ratio.
Ai. Representative traces showing wild type paired EPSCs before, during, and after application
of group II mGluR agonist APDC (75 µM). Aii. Raw time course of the change in amplitude of
	

168	

Figure 4.6. (cont’d)
EPSC 1 (black) and EPSC 2 (gray) in response to APDC. Aiii. Raw time course of the change in
paired pulse ratio (PPR) in response to APDC. Bi-iii. Same as Ai-iii, but in the Fmr1 KO. C.
Average population PPR in the presence of APDC expressed as % of control for both WT and
Fmr1 KO (WT=6, KO=5). Pooled data show no statistical difference in PPR between WT and
Fmr1 KO (p>0.05, t-test).

	

169	

Figure 4.7. EPSCs evoked in strontium show a decrease in frequency in response to
application of group II mGluR activation in both WT and Fmr1 KO layer 5/6 NFS
interneurons.
Ai. Raw data showing a WT NFS interneuron EPSC response in standard Ca2+ physiological
saline, following a substitution with Sr2+ saline, during application of group II mGluR agonist
APDC (15 µM), and after drug washout. Bi. Same as Ai, but for Fmr1 KO. C. Population
change in mEPSC amplitude in response to application of APDC. D. Population change in
mEPSC frequency in response to APDC. For both C and D, black bars indicate the average
mEPSC response from the pooled population data. WT and Fmr1 KO cells show a change in
mEPSC frequency without showing a change in mEPSC amplitude (WT=8, KO=5), indicating a
predominantly presynaptic effect of group II mGluRs.
	

170	

Figure 4.8. The intrinsic properties Fmr1 KO layer 2/3 NFS interneurons are unaltered.
Ai. Representative tracing of a layer 2/3 wild type NFS interneuron during a current-step
injection protocol. Aii. Similar to figure Ai but recorded in a layer 2/3 NFS interneuron from
the Fmr1 KO mouse. B. Population data show the average input resistance in wild type and
Fmr1 KO cells (WT=16, KO=9). C. Population data show the average membrane potential of
WT and Fmr1 KO cells (WT=16, KO=9, p > 0.05). No statistically significant differences were
detected between genotypes for either membrane potential or input resistance (p > 0.05, t-test).

	

171	

Figure 4.9. Altered group II mGluR-mediated suppression of NFS interneuron EPSCs in
the Fmr1 KO is layer specific.
Ai. Representative trace showing a layer 2/3 wild type NFS interneuron EPSC evoked by
electrical stimulation of layer 4. Traces show EPSC response before, during, and after brief
application of APDC (75 µM). Aii. Raw time course of the change in WT EPSC charge before,
during, and after APDC application. Aiii. WT population data showing the normalized time
	

172	

Figure 4.9. (cont’d)
course of the change in charge in response to APDC (n=16). Bi-iii. Same as Ai-iii, but for Fmr1
KO (n=9). C. Average population EPSC charge in the presence of APDC expressed as % of
control for both WT and Fmr1 KO (WT=16, KO=9). No statistical differences were found
between the charge of WT and Fmr1 KO in the presence of APDC (p > 0.05, t-test).

	

173	

Figure 4.10. Group II mGluR-mediated suppression of EPSCs on NFS interneurons is
increased in the Fmr1 KO mouse.
	

174	

Figure 4.10. (cont’d)
Top: Circuit diagram for a WT mouse. In chapter 4, excitatory postsynaptic currents (EPSCs)
were evoked using a stimulating electrode placed in layer 2/3 (S). We show that activation of
group II mGluRs leads to suppression of EPSCs at glutamatergic synapses onto layer 5/6 FS
interneurons (red neuron) and NFS interneurons (green neuron) in WT mice. Bottom: Circuit
diagram for an Fmr1 KO mouse. The magnitude of EPSC suppression is similar between WT
and Fmr1 KO mouse on FS interneurons (red neuron, black checkmark). However, the
magnitude of suppression of EPSCs on NFS interneurons (black circle, bottom) is enhanced in
the Fmr1 KO.

	

175	

Figure 4.11. Activation of presynaptic group II mGluRs leads to exaggerated suppression
of excitatory drive onto NFS interneurons and disinhibition of GABAergic signaling in the
visual neocortex of the Fmr1 KO mouse.
A schematic diagram showing changes in the layer 5/6 microcircuit of the visual neocortex in the
Fmr1 KO mouse. Ai. Activation of presynaptic group II mGluRs in WT mice leads to a
decrease in glutamatergic transmission onto NFS interneurons in layer 5/6 of V1. This leads to a
decrease in inhibitory transmission at GABA synapses. Aii. Activation of presynaptic group II
mGluRs in the Fmr1 KO mouse, leads to exaggerated decreases in glutamatergic
neurotransmission onto NFS interneurons. This exaggeration leads to disinhibition of excitatory
pyramidal neurons, and potential hyperexcitability. NFS interneurons predominantly target the
dendritic tufts of layer 5 pyramidal neurons, but may also synapse onto the dendrites of neurons
in layer 2-4. The net effect of these microcircuit changes is expected to be a disinhibited V1
network. This could affect feed forward signaling to higher order cortical areas connected to V1,
as well as feedback signaling to subcortical structures.

	

176	

REFERENCES

	

177	

REFERENCES
Akins, M. R., Berk-Rauch, H. E., & Fallon, J. R. (2009). Presynaptic translation: stepping out of
the postsynaptic shadow. Front Neural Circuits, 3, 17. doi:10.3389/neuro.04.017.2009
Akins, M. R., Leblanc, H. F., Stackpole, E. E., Chyung, E., & Fallon, J. R. (2012). Systematic
mapping of fragile X granules in the mouse brain reveals a potential role for presynaptic
FMRP in sensorimotor functions. J Comp Neurol, 520(16), 3687-3706.
doi:10.1002/cne.23123
Antar, L. N., Afroz, R., Dictenberg, J. B., Carroll, R. C., & Bassell, G. J. (2004). Metabotropic
glutamate receptor activation regulates fragile x mental retardation protein and FMR1
mRNA localization differentially in dendrites and at synapses. J Neurosci, 24(11), 26482655. doi:10.1523/JNEUROSCI.0099-04.2004
Antar, L. N., Li, C., Zhang, H., Carroll, R. C., & Bassell, G. J. (2006). Local functions for FMRP
in axon growth cone motility and activity-dependent regulation of filopodia and spine
synapses. Mol Cell Neurosci, 32(1-2), 37-48. doi:10.1016/j.mcn.2006.02.001
Beierlein, M., Gibson, J. R., & Connors, B. W. (2000). A network of electrically coupled
interneurons drives synchronized inhibition in neocortex. Nat Neurosci, 3(9), 904-910.
doi:10.1038/78809
Beierlein, M., Gibson, J. R., & Connors, B. W. (2003). Two dynamically distinct inhibitory
networks in layer 4 of the neocortex. J Neurophysiol, 90(5), 2987-3000.
doi:10.1152/jn.00283.2003
Bhakar, A. L., Dolen, G., & Bear, M. F. (2012). The pathophysiology of fragile X (and what it
teaches us about synapses). Annu Rev Neurosci, 35, 417-443. doi:10.1146/annurev-neuro060909-153138
Blackmer, T., Larsen, E. C., Takahashi, M., Martin, T. F., Alford, S., & Hamm, H. E. (2001). G
protein betagamma subunit-mediated presynaptic inhibition: regulation of exocytotic
fusion downstream of Ca2+ entry. Science, 292(5515), 293-297.
doi:10.1126/science.1058803
Brager, D. H., & Johnston, D. (2014). Channelopathies and dendritic dysfunction in fragile X
syndrome. Brain Res Bull, 103, 11-17. doi:10.1016/j.brainresbull.2014.01.002
Chang, Y. M., & Luebke, J. I. (2007). Electrophysiological diversity of layer 5 pyramidal cells in
the prefrontal cortex of the rhesus monkey: in vitro slice studies. J Neurophysiol, 98(5),
2622-2632. doi:10.1152/jn.00585.2007
Choi, C. H., Schoenfeld, B. P., Bell, A. J., Hinchey, P., Kollaros, M., Gertner, M. J., . . .
McBride, S. M. (2011). Pharmacological reversal of synaptic plasticity deficits in the

	

178	

mouse model of fragile X syndrome by group II mGluR antagonist or lithium treatment.
Brain Res, 1380, 106-119. doi:10.1016/j.brainres.2010.11.032
Christie, S. B., Akins, M. R., Schwob, J. E., & Fallon, J. R. (2009). The FXG: a presynaptic
fragile X granule expressed in a subset of developing brain circuits. J Neurosci, 29(5),
1514-1524. doi:10.1523/JNEUROSCI.3937-08.2009
Darnell, J. C. (2011). Defects in translational regulation contributing to human cognitive and
behavioral disease. Curr Opin Genet Dev, 21(4), 465-473. doi:10.1016/j.gde.2011.05.002
Gibson, J. R., Bartley, A. F., Hays, S. A., & Huber, K. M. (2008). Imbalance of neocortical
excitation and inhibition and altered UP states reflect network hyperexcitability in the
mouse model of fragile X syndrome. J Neurophysiol, 100(5), 2615-2626.
doi:10.1152/jn.90752.2008
Haberl, M. G., Zerbi, V., Veltien, A., Ginger, M., Heerschap, A., & Frick, A. (2015). Structuralfunctional connectivity deficits of neocortical circuits in the Fmr1 (-/y) mouse model of
autism. Sci Adv, 1(10), e1500775. doi:10.1126/sciadv.1500775
Hagerman, P. J. (2008). The fragile X prevalence paradox. J Med Genet, 45(8), 498-499.
doi:10.1136/jmg.2008.059055
Hagerman, P. J., & Stafstrom, C. E. (2009). Origins of epilepsy in fragile X syndrome. Epilepsy
Curr, 9(4), 108-112. doi:10.1111/j.1535-7511.2009.01309.x
Kawaguchi, Y., & Kubota, Y. (1997). GABAergic cell subtypes and their synaptic connections
in rat frontal cortex. Cereb Cortex, 7(6), 476-486.
Larkman, A., & Mason, A. (1990). Correlations between morphology and electrophysiology of
pyramidal neurons in slices of rat visual cortex. I. Establishment of cell classes. J
Neurosci, 10(5), 1407-1414.
Markram, H., Toledo-Rodriguez, M., Wang, Y., Gupta, A., Silberberg, G., & Wu, C. (2004).
Interneurons of the neocortical inhibitory system. Nat Rev Neurosci, 5(10), 793-807.
doi:10.1038/nrn1519
Myrick, L. K., Deng, P. Y., Hashimoto, H., Oh, Y. M., Cho, Y., Poidevin, M. J., . . . Klyachko,
V. A. (2015). Independent role for presynaptic FMRP revealed by an FMR1 missense
mutation associated with intellectual disability and seizures. Proc Natl Acad Sci U S A,
112(4), 949-956. doi:10.1073/pnas.1423094112
Niswender, C. M., & Conn, P. J. (2010). Metabotropic glutamate receptors: physiology,
pharmacology, and disease. Annu Rev Pharmacol Toxicol, 50, 295-322.
doi:10.1146/annurev.pharmtox.011008.145533

	

179	

Patel, A. B., Hays, S. A., Bureau, I., Huber, K. M., & Gibson, J. R. (2013). A target cell-specific
role for presynaptic Fmr1 in regulating glutamate release onto neocortical fast-spiking
inhibitory neurons. J Neurosci, 33(6), 2593-2604. doi:10.1523/JNEUROSCI.244712.2013
Rudy, B., Fishell, G., Lee, S., & Hjerling-Leffler, J. (2011). Three groups of interneurons
account for nearly 100% of neocortical GABAergic neurons. Dev Neurobiol, 71(1), 4561. doi:10.1002/dneu.20853
Scanziani, M., Salin, P. A., Vogt, K. E., Malenka, R. C., & Nicoll, R. A. (1997). Use-dependent
increases in glutamate concentration activate presynaptic metabotropic glutamate
receptors. Nature, 385(6617), 630-634. doi:10.1038/385630a0
Till, S. M., Wijetunge, L. S., Seidel, V. G., Harlow, E., Wright, A. K., Bagni, C., . . . Kind, P. C.
(2012). Altered maturation of the primary somatosensory cortex in a mouse model of
fragile X syndrome. Hum Mol Genet, 21(10), 2143-2156. doi:10.1093/hmg/dds030

	

180	

CHAPTER V: GENERAL CONCLUSIONS
Fragile X Syndrome affects 1 in 4000 males and 1 in 8000 females (Hagerman, 2008).
While this may give the impression that FXS is relatively uncommon, in fact it is the most
common genetically inherited disease of intellectual disability (Hagerman, 2009). Additionally,
30% of children with FXS exhibit autism (Gibson et al., 2008) and another 20-25% of children
exhibit epilepsy (Stafstrom, 2009), making the pathogenic mechanisms we uncover in the Fmr1
KO mouse invaluable to our insight about FXS as well as autism spectrum disorders, epilepsy
syndromes, and other intellectual disabilities. In my dissertation, I investigated neocortical
pathogenesis in the Fmr1 KO mouse. Electrophysiological studies of neocortical function in the
Fmr1 KO are underrepresented relative to studies that have been done in brain regions such as
the hippocampus and amygdala (Sabanov, 2016; Olmos-Serrano et al., 2010), implicating
requirement for further investigations of the neocortical pathogenesis in FXS. My decision to
investigate the visual neocortex of the Fmr1 KO was based on previous studies implicating V1 in
FXS pathology. Pyramidal neuron dendritic spines in layer 2/3 and 5/6 exhibit immature
morphologies in V1 of the Fmr1 KO mouse (Dolen et al., 2007; Comery et al., 1997),
substantiating what has been reported in human studies (Rudelii., 1985; Hinton et al., 1991;
Irwin, Galvez, & Greenough, 2000). Additionally, synaptic plasticity in V1 is altered in the
Fmr1 KO. Specifically, our lab previously showed that layer 5/6 visual neocortical LTP is
absent in the Fmr1 KO (Wilson & Cox, 2007). This LTP was mediated in part, by group I
mGluRs (Wilson & Cox, 2007). Specifically, it appeared that mGluR5 was the instigating
receptor in this defect. These results, led me to hypothesize that other processes mediated by
mGluR signaling would be altered in the visual neocortex of Fmr1 KO mice, leading to circuit
hyperexcitability. In Chapter 2, I investigated the effect of mGluRs on the membrane

	

181	

excitability of layer 5/6 pyramidal neurons, as these neurons are morphologically different in the
Fmr1 KO and would be expected to directly impact cortical hyperexcitability. I determined that
both group I and group II mGluRs modulate membrane excitability of pyramidal neurons, but
there is no difference between the WT and Fmr1 KO mouse. Thus, it is evident from these
results that not all cellular and synaptic processes modulated by mGluRs will be altered in the
Fmr1 KO mouse. Knowing that group I mGluRs modulate long term changes in synaptic
transmission, I next asked whether mGluRs modulate short term dynamics of synaptic
transmission differently in the Fmr1 KO. Previous studies have indicated that mGluRs can
reversibly depress excitatory synaptic transmission in the visual neocortex (Sladeczek et al.,
1993; Flavin, Jin & Daw, 2000). My studies show that group II mGluRs suppress glutamatergic
transmission in layer 5/6 and 2/3 pyramidal neurons. However, I did not observe any significant
differences in the magnitude of these changes when comparing WT and Fmr1 KO animals. How
can this be the case when mGluR-LTD is significantly reduced in layer 5/6 of the Fmr1 KO? It
is well known that not all consequences of mGluR activation require protein synthesis. For
example, it has been shown that the Bg subunits of mGluRs can directly activate ion channels
independent of protein translation (Niswender & Conn, 2010). Thus, I speculate that the ability
of mGluRs to modulate membrane excitability and suppress synaptic transmission in the short
term may not require protein synthesis and may therein not be affected by the loss of FMRP. A
simple way to test this hypothesis would be to include a protein inhibitor in the bath and/or
intracellular electrode to determine whether blockade of protein synthesis would occlude the
mGluR effects that I observed. Understanding that mGluR-mediated modulation of excitatory
neurons is normal in the Fmr1 KO, I hypothesized that mGluRs may be altering GABAergic
transmission in the Fmr1 KO visual neocortex, as this could also lead to hyperexcitable

	

182	

neocortical circuitry. In Chapter 3, I tested this hypothesis by quantifying the changes in evoked
IPSC amplitude in layer 5/6 pyramidal neurons in response to a group II mGluR agonist.
Intact GABAergic regulation of pyramidal neurons is required to maintain network
excitability within the normal range. Defects in GABAergic transmission are known to cause
hyperexcitability and/or epileptiform activity (Olmos-serrano et al., 2010). Additionally, mGluRs
have been shown to modulate GABAergic transmission (Doi et al., 2002; Liu, Petrof, and
Sherman, 2014). In the visual neocortex, it has specifically been shown that group II mGluRs
can act presynaptically to decrease GABAergic transmission by reducing release probability,
albeit through a currently unknown mechanism (Liu, Petrof, and Sherman, 2014). This suggests
that activation of group II mGluRs in the visual neocortex of the Fmr1 KO mouse, may lead to
alterations in GABAergic transmission onto pyramidal neurons. My results indicate that
activation of group II mGluRs leads to suppression of inhibitory transmission that is significantly
increased in the Fmr1 KO. By blocking glutamatergic transmission in the slice, I was then able
to effectively isolate GABAergic inputs onto my pyramidal neurons. During this condition I
discovered that group II mGluRs continue to suppress evoked IPSCs, but the magnitude of this
suppression is similar between WT and Fmr1 KO pyramidal neurons. I validated these results
with experiments looking at the frequency and amplitude of spontaneous and miniature IPSCs. I
discovered that although group II mGluRs are present at presynaptic terminals of GABAergic
neurons and are capable of decreasing GABAergic transmission, this mechanism of regulation is
intact in the Fmr1 KO. This indicates that the exaggerated suppression of inhibition that I
observe in the Fmr1 KO pyramidal neurons, is due to the effects of group II mGluRs at
excitatory synapses onto interneurons. I subsequently hypothesized that group II mGluRs would

	

183	

suppress excitatory transmission onto inhibitory interneurons, and that this effect would be
exaggerated in the Fmr1 KO mouse. In Chapter 4, I tested that hypothesis.
Interneurons comprise 20-30% of all neocortical subtypes (Markram et al., 2004). The
two most abundant subtypes, PV+ and SST+ interneurons, comprise 70% of the interneuron
population (Rudy et al., 2011). Furthermore, previous FXS studies have implicated both FS
interneurons and SST+ interneurons in the neocortical pathogenesis of the Fmr1 KO mouse
(Gibson et al., 2008; Paluskiewicz et al., 2011). Thus, it seemed likely that these interneuron
subtypes may be responsible for mediating the inhibitory changes I see upon activation of group
II mGluRs. In chapter 4, I evoked excitatory currents from both FS and NFS interneurons in the
layer 5/6 and quantified the change in either charge or amplitude in response to pharmacological
activation of group II mGluRs. Activation of group II mGluRs led to suppression of
glutamatergic transmission onto both layer 5/6 FS and NFS interneurons. Further, statistical
analysis revealed that the suppression of glutamatergic transmission onto SST+ interneurons was
exaggerated in the Fmr1 KO, supporting my hypothesis that decreased excitatory drive onto
layer 5/6 visual neocortical interneurons leads to decreased GABAergic activity onto pyramidal
neurons in the Fmr1 KO. In contrast, this exaggeration of group II mGluR-mediated
suppression of glutamatergic drive was not seen when FS cells were the postsynaptic targets.
Thus, my effect is target-cell specific. To complete my investigation, I asked two final
questions: Where are these group II mGluRs located? Is this pathogenic mechanism layer
specific or does it occur in more superficial lamina? To answer the first question, I performed a
series of experiments that are traditionally used to ascertain whether a synaptic effect is pre- or
post- synaptic. I determined that group II mGluRs are present at both excitatory neuron
terminals and on the postsynaptic interneuron targets; however, the suppression of EPSCs was

	

184	

most likely due to a presynaptic mechanism. This was a novel discovery, as the role of FMRP
presynaptically remains largely unknown. Finally, by repeating my evoked experiments in layer
2/3, I determined that the alterations observed in layer 5/6 are layer specific. While the
specificities of this presynaptic mechanism are unclear, I predict that activation of group II
mGluRs in the absence of FMRP is exaggerating one or more of the following processes:
activation of K+ channels, inhibition of Ca2+ channels, protein translation downstream of PI3KmTOR, ERK/MAPK, or Gi signaling cascades, or inhibition of the exocytotic release machinery
(For details see Chapter 4.4). For a schematic diagram of these cellular mechanisms, see the
appendix (Figure 5.1) is Nevertheless, these results provide evidence in favor of a disinhibited
deep layer neocortical network. This is likely to have several consequences including aberrant
sensory processing in all target neurons of layer 5/6 pyramidal axons. Although my dissertation
studies do not permit definitive identification of these target neurons, I can speculate potential
defective pathways based on previous studies. Two-thirds of layer 5/6 pyramidal neurons target
layer 2/3 (Callaway, 1998), and it is highly likely that some of the targets of these disinhibited
neurons are excitatory cells in these superficial layers. These neurons encode information for the
dorsal and ventral streams of visual processing, and so both of these pathways may be altered in
FXS. Interestingly, individuals with FXS perform poorly on visual-motor tasks compared to
both chronological age and mental age-matched controls, implicating potential defects in the M
pathway of visual processing (Kogan et al., 2004; Kogan et al., 2004b). Thus, it is equally
possible that the disinhibited pyramidal neurons that I recorded from may be targeting layer 2/3
neurons that specifically process M pathway information. Additionally, deep layer pyramidal
neurons also project axons to the pulvinar nucleus of the thalamus and the superior colliculus,

	

185	

both of which are important for visual attention and eye movement. Defects in these pathways
would also be expected to lead to severe defects in processing of visual motion.
The research I have provided here is invaluable not only to our understanding of FXS, but
potentially to our understanding of autism. FMRP targets the mTOR inhibitors Pten, Nf1, and
Tsc2 (Darnell et al., 2011), so it is possible that the mechanisms underlying aberrant behavior in
FXS overlap with the mechanisms underlying autism pathology. Theoretically, this should mean
that any pharmacological interventions designed to treat FXS may also be clinically beneficial to
autism patients. Group II mGluRs are currently being investigated as useful drug targets in a
variety of neurological and psychiatric conditions including schizophrenia, addiction, major
depression, anxiety, parkinon’s, hungtington’s, alzheimer’s, and epilepsy (Hovelso et al., 2012).
Additionally, clinical trials in schizophrenic patients using an oral prodrug for group 2/3 mGluRs
have reported significant improvement in both positive and negative symptoms (Patil et al.,
2007). It appears that group II mGluRs, predominantly through their presynaptic actions on
neurotransmitter release, have the potential to ameliorate a variety of symptoms. However, with
the exception of major depression, the majority of the aforementioned studies show benefits of
group II mGluR agonism rather than antagonism (Hovelso et al., 2012). On the contrary, my
dissertation studies suggest that antagonism of group II mGluRs would be more beneficial in
restoring the defects I observe in V1. However, group II mGluRs have a widespread distribution
in the CNS and have been shown to exert antiepileptic effects in vitro and in vivo, (Anwyl, 1999)
so administration of a group II antagonist could potentially lead to excitotoxicity. Thus, I
suggest that my dissertation data provide additional support for the idea of using GABAergic
agonists to ameliorate FXS symptoms. Agonizing GABAergic transmission to compensate for a
decrease of excitatory drive onto interneurons, may provide a safer alternative to the use of a

	

186	

group II antagonist. Clinical trials are already underway to investigate the therapeutic potential
of GABA agonists in FXS (Davenport et al., 2016). The most promising drug candidate to-date
is acamprosate, a compound that is hypothesized to effect GABAA signaling as well as
metabotropic glutamate receptor signaling. An open-label study showed that 75% of patients
were “very much improved” or “much improved” on several outcome measures for ADHD,
social responsiveness, and other cognitive-behavioral measures (Erickson et al., 2013). As a
result, a current Phase II placebo-controlled study is being conducted on the effects of
acamprosate. It is my hope that my dissertation provides further mechanistic support in favor of
the use of GABAergic agonists for clinical treatment of FXS phenotypes.
While electrophysiology is the gold standard for assessing the functional effects of
pathological processes on the synapse, even within the field different techniques present different
limitations. My dissertation studies took advantage of the whole-cell patch clamp technique in
combination with pharmacology, to begin to determine which neurons and synapses were altered
in layer 5/6 of the visual neocortex. While I determined that the glutamatergic excitation of
somatostatin neurons was specifically altered, there continues to be questions regarding the
synapse under investigation. For example, electrical stimulation is activating several excitatory
inputs onto the postsynaptic cell. With this stimulation method it is impossible for me to
determine the source of these excitatory inputs. Although I hypothesize that the glutamatergic
inputs onto somatostatin interneurons come from pyramidal neurons residing in the neocortex, to
quantitatively answer this question it will be advantageous to conditionally express channel
rhodopsin (CHR2) in excitatory neurons in either the layer 5/6 or the layer 2/3 of the neocortex.
This can be accomplished by injection of CHR2 into transgenic mice selectively expressing CRE
in layer 5/6 (Beltramo et al., 2013), or by using a transgenic mouse that is already engineered to

	

187	

express CHR2 in layer 5/6 pyramidal neurons (Sohal et al., 2009; Arienkiel et al., 2007). Whole
cell recordings can then be obtained from somatostatin neurons in layer 5/6, and blue light can be
used to activate all glutamatergic neurons expressing channel rhodopsin, which in this case
would be layer 5/6 excitatory neurons. This would effectively keep all other excitatory inputs
onto the somatostatin neurons silent, with the exception of spontaneous firing. This selective
targeting would allow me to determine where my glutamatergic inputs are coming from.
Additionally, although not a limitation per say, the experiments I used to localize my
group II mGluR effect presynaptically, could be supported by isolating any potential
postsynaptic group II mGluR effects on my somatostatin interneuron. Although group II
mGluRs have been shown to hyperpolarize membranes when they are localized to the axosomatic compartment of the postsynaptic cell, this localization of group II mGluRs is less
common. Further, while collecting my data I see almost no change in the holding current of my
somatostatin neurons when I apply a group II mGluR agonist to the slice preparation. For these
reasons, I strongly suspect that there are minimal, if any, group II mGluRs located upstream from
the axon terminal of my SST+ interneurons. Nonetheless, to verify this hypothesis, I could
perform glutamate uncaging using a two-photon microscope. Glutamate uncaging in the presence
of TTX, could be used to evoke an excitatory postsynaptic potential in the SST+ neuron. I
would expect application of APDC to have little or no effect on the amplitude of this response,
which would verify my previous results showing that the alterations I see in the Fmr1 KO mouse
are occurring due to presynaptic group II mGluRs.
Finally, it would be advantageous to determine whether restoration of FMRP could
rescue the exaggerated group II mGluR-mediated suppression of excitation that I observe in
layer 5/6 SST+ interneurons. For this, I would have to use a transgenic mouse that conditionally

	

188	

restores FMRP function. The group II mGluR effects that I observe may be a direct consequence
of the loss of FMRP, or they may be a neurological compensation for alterations in glutamatergic
transmission. If restoring FMRP fails to reinstate normal levels of group II mGluR signaling,
then I would suspect that my observation is a potentially irreversible compensatory mechanism
to offset a potential excess of glutamatergic transmission. Regardless of what direction the
research takes from here, my dissertation provides strong evidence in favor of a hyperexcitable
visual neocortical network.

	

189	

APPENDIX

	

190	

Figure 5.1. Altered presynaptic processes in the Fmr1 KO mouse upon activation of group
II mGluRs.
We demonstrated in previous chapters that activation of group II mGluRs at presynaptic
glutamatergic inputs onto NFS interneurons (light green NFS), leads to suppression of evoked
excitatory currents (EPSCs). This process is exaggerated in the Fmr1 KO, leading to

	

191	

Figure 5.1. (cont’d)
suppression of GABA transmission (red dots). The net effect is disinhibition of downstream
excitatory pyramidal neurons. This schematic diagram presents a theoretical model to explain
potential changes in cellular mechanisms occurring downstream of group II mGluRs in the Fmr1
KO mouse. FMRP is a regulator of mRNA translation, and has been shown to modulate the
following mRNAs presynaptically: K+ channels (Deng et al., 2014), Ca2+ channels (Brager and
Johnston, 2014), and presynaptic vesicular release machinery such as syntaxin 1, SNAP-25, and
VAMP (Tang et al., 2015). Activation of group II mGluRs allows a subunit to inhibit adenylyl
cyclase (AC), leading to decreases in cAMP, pKA, protein translation, and synaptic transmission
(Hovelso et al., 2012). The Bg subunit has been shown to directly interact with both K+ and Ca2+
channels at the presynaptic terminal to decrease the excitability of the axon terminal, leading to
decreased transmission (Chavis et al., 1994; O’Connor et al., 1999). By can also directly interact
with vesicular release machinery to decrease synaptic transmission (Hovelso et al., 2012). I
propose that FMRP is either directly downstream of group II mGluR signaling cascades, or it is
regulating the translation of the same proteins that are modulated by group II mGluR activation.
In the absence of FMRP, we would expect altered levels of protein synthesis of the above
mentioned synaptic proteins, which leads to changes in the functional modulation of the synapse
upon activation of group II mGluRs.

	

192	

REFERENCES

	

193	

REFERENCES
Anwyl, R. (1999). Metabotropic glutamate receptors: electrophysiological properties and role in
plasticity. Brain Res Rev, 29(1), 83-120.
Arenkiel, B. R., Peca, J., Davison, I. G., Feliciano, C., Deisseroth, K., Augustine, G. J., . . . Feng,
G. (2007). In vivo light-induced activation of neural circuitry in transgenic mice
expressing channelrhodopsin-2. Neuron, 54(2), 205-218.
doi:10.1016/j.neuron.2007.03.005
Beltramo, R., D'Urso, G., Dal Maschio, M., Farisello, P., Bovetti, S., Clovis, Y., . . . Fellin, T.
(2013). Layer-specific excitatory circuits differentially control recurrent network
dynamics in the neocortex. Nat Neurosci, 16(2), 227-234. doi:10.1038/nn.3306
Brager, D. H., & Johnston, D. (2014). Channelopathies and dendritic dysfunction in fragile X
syndrome. Brain Res Bull, 103, 11-17. doi:10.1016/j.brainresbull.2014.01.002
Chavis, P., Shinozaki, H., Bockaert, J., & Fagni, L. (1994). The metabotropic glutamate receptor
types 2/3 inhibit L-type calcium channels via a pertussis toxin-sensitive G-protein in
cultured cerebellar granule cells. J Neurosci, 14(11 Pt 2), 7067-7076.
Comery, T. A., Harris, J. B., Willems, P. J., Oostra, B. A., Irwin, S. A., Weiler, I. J., &
Greenough, W. T. (1997). Abnormal dendritic spines in fragile X knockout mice:
maturation and pruning deficits. Proc Natl Acad Sci U S A, 94(10), 5401-5404.
Darnell, J. C. (2011). Defects in translational regulation contributing to human cognitive and
behavioral disease. Curr Opin Genet Dev, 21(4), 465-473. doi:10.1016/j.gde.2011.05.002
Davenport, M. H., Schaefer, T. L., Friedmann, K. J., Fitzpatrick, S. E., & Erickson, C. A. (2016).
Pharmacotherapy for Fragile X Syndrome: Progress to Date. Drugs, 76(4), 431-445.
doi:10.1007/s40265-016-0542-y
Deng, P. Y., & Klyachko, V. A. (2016). Genetic upregulation of BK channel activity normalizes
multiple synaptic and circuit defects in a mouse model of fragile X syndrome. J Physiol,
594(1), 83-97. doi:10.1113/JP271031
Doi, A., Ishibashi, H., Jinno, S., Kosaka, T., & Akaike, N. (2002). Presynaptic inhibition of
GABAergic miniature currents by metabotropic glutamate receptor in the rat CNS.
Neuroscience, 109(2), 299-311.
Erickson, C. A., Davenport, M. H., Schaefer, T. L., Wink, L. K., Pedapati, E. V., Sweeney, J. A.,
. . . Berry-Kravis, E. (2017). Fragile X targeted pharmacotherapy: lessons learned and
future directions. J Neurodev Disord, 9, 7. doi:10.1186/s11689-017-9186-9

	

194	

Flavin, H. J., Jin, X. T., & Daw, N. W. (2000). 2R,4R-4-Aminopyrrolidine-2,4-dicarboxylate
(APDC) attenuates cortical EPSPs. Brain Res, 873(2), 212-217.
Gibson, J. R., Bartley, A. F., Hays, S. A., & Huber, K. M. (2008). Imbalance of neocortical
excitation and inhibition and altered UP states reflect network hyperexcitability in the
mouse model of fragile X syndrome. J Neurophysiol, 100(5), 2615-2626.
doi:10.1152/jn.90752.2008
Hagerman, F. C., Walsh, S. J., Staron, R. S., Hikida, R. S., Gilders, R. M., Murray, T. F., . . .
Ragg, K. E. (2000). Effects of high-intensity resistance training on untrained older men.
I. Strength, cardiovascular, and metabolic responses. J Gerontol A Biol Sci Med Sci,
55(7), B336-346.
Hagerman, P. J. (2008). The fragile X prevalence paradox. J Med Genet, 45(8), 498-499.
doi:10.1136/jmg.2008.059055
Hayashi, Y., Momiyama, A., Takahashi, T., Ohishi, H., Ogawa-Meguro, R., Shigemoto, R., . . .
Nakanishi, S. (1993). Role of a metabotropic glutamate receptor in synaptic modulation
in the accessory olfactory bulb. Nature, 366(6456), 687-690. doi:10.1038/366687a0
Hinton, V. J., Brown, W. T., Wisniewski, K., & Rudelli, R. D. (1991). Analysis of neocortex in
three males with the fragile X syndrome. Am J Med Genet, 41(3), 289-294.
doi:10.1002/ajmg.1320410306
Hovelso, N., Sotty, F., Montezinho, L. P., Pinheiro, P. S., Herrik, K. F., & Mork, A. (2012).
Therapeutic potential of metabotropic glutamate receptor modulators. Curr
Neuropharmacol, 10(1), 12-48. doi:10.2174/157015912799362805
Irwin, S. A., Galvez, R., & Greenough, W. T. (2000). Dendritic spine structural anomalies in
fragile-X mental retardation syndrome. Cereb Cortex, 10(10), 1038-1044.
Irwin, S. A., Idupulapati, M., Gilbert, M. E., Harris, J. B., Chakravarti, A. B., Rogers, E. J., . . .
Greenough, W. T. (2002). Dendritic spine and dendritic field characteristics of layer V
pyramidal neurons in the visual cortex of fragile-X knockout mice. Am J Med Genet,
111(2), 140-146. doi:10.1002/ajmg.10500
Liu, B., Li, L., Chen, J., Wang, Z., Li, Z., & Wan, Q. (2013). Regulation of GABAA receptors
by fragile X mental retardation protein. Int J Physiol Pathophysiol Pharmacol, 5(3), 169176.
Niswender, C. M., & Conn, P. J. (2010). Metabotropic glutamate receptors: physiology,
pharmacology, and disease. Annu Rev Pharmacol Toxicol, 50, 295-322.
doi:10.1146/annurev.pharmtox.011008.145533

	

195	

O'Connor, V., El Far, O., Bofill-Cardona, E., Nanoff, C., Freissmuth, M., Karschin, A., . . .
Boehm, S. (1999). Calmodulin dependence of presynaptic metabotropic glutamate
receptor signaling. Science, 286(5442), 1180-1184.
Olmos-Serrano, J. L., Paluszkiewicz, S. M., Martin, B. S., Kaufmann, W. E., Corbin, J. G., &
Huntsman, M. M. (2010). Defective GABAergic neurotransmission and pharmacological
rescue of neuronal hyperexcitability in the amygdala in a mouse model of fragile X
syndrome. J Neurosci, 30(29), 9929-9938. doi:10.1523/JNEUROSCI.1714-10.2010
Paluszkiewicz, S. M., Olmos-Serrano, J. L., Corbin, J. G., & Huntsman, M. M. (2011). Impaired
inhibitory control of cortical synchronization in fragile X syndrome. J Neurophysiol,
106(5), 2264-2272. doi:10.1152/jn.00421.2011
Patel, A. B., Hays, S. A., Bureau, I., Huber, K. M., & Gibson, J. R. (2013). A target cell-specific
role for presynaptic Fmr1 in regulating glutamate release onto neocortical fast-spiking
inhibitory neurons. J Neurosci, 33(6), 2593-2604. doi:10.1523/JNEUROSCI.244712.2013
Rudelli, R. D., Brown, W. T., Wisniewski, K., Jenkins, E. C., Laure-Kamionowska, M., Connell,
F., & Wisniewski, H. M. (1985). Adult fragile X syndrome. Clinico-neuropathologic
findings. Acta Neuropathol, 67(3-4), 289-295.
Rudy, B., Fishell, G., Lee, S., & Hjerling-Leffler, J. (2011). Three groups of interneurons
account for nearly 100% of neocortical GABAergic neurons. Dev Neurobiol, 71(1), 4561. doi:10.1002/dneu.20853
Sladeczek, F., Momiyama, A., & Takahashi, T. (1993). Presynaptic inhibitory action of a
metabotropic glutamate receptor agonist on excitatory transmission in visual cortical
neurons. Proc Biol Sci, 253(1338), 297-303. doi:10.1098/rspb.1993.0117
Sohal, V. S., Zhang, F., Yizhar, O., & Deisseroth, K. (2009). Parvalbumin neurons and gamma
rhythms enhance cortical circuit performance. Nature, 459(7247), 698-702.
doi:10.1038/nature07991
Stafstrom, C. E. (2009). Severe epilepsy syndromes of early childhood: the link between genetics
and pathophysiology with a focus on SCN1A mutations. J Child Neurol, 24(8 Suppl),
15S-23S. doi:10.1177/0883073809338152
Tang, B., Wang, T., Wan, H., Han, L., Qin, X., Zhang, Y., . . . Liao, L. (2015). Fmr1 deficiency
promotes age-dependent alterations in the cortical synaptic proteome. Proc Natl Acad Sci
U S A, 112(34), E4697-4706. doi:10.1073/pnas.1502258112
Wilson, B. M., & Cox, C. L. (2007). Absence of metabotropic glutamate receptor-mediated
plasticity in the neocortex of fragile X mice. Proc Natl Acad Sci U S A, 104(7), 24542459. doi:10.1073/pnas.0610875104

	

196