A novel role for the ER-calcium sensor STIM2 in regulating … · 2018. 1. 8. · A novel role for the ER-calcium sensor STIM2 in regulating dendritic spine morphogenesis and AMPA

i

A novel role for the ER-calcium sensor STIM2 in

regulating dendritic spine morphogenesis and AMPA

receptor trafficking

Wong Loo Chin

(School of Biological Sciences’ Bachelor program (Honours)/ Nanyang

Technological University)

A THESIS SUBMITTED

FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF SCHOOL OF MEDICINE

NATIONAL UNIVERSITY OF SINGAPORE

2015

ii

DECLARATION

I hereby declare that the thesis is my original work and it has been written by me

in its entirety. I have duly acknowledged all the sources of information which

have been used in the thesis.

This thesis has also not been submitted for any degree in any university

previously.

Signature of student: Date:

iii

ACKNOWLEDGEMENTS

I will like to thank my supervisor, Assistant Professor Marc Laurent Fivaz, for his

constant guidance, support, discussion and patience throughout the duration of my

PhD. This work would not have been possible without him.

I am also thankful for my colleagues and the members of my lab, for their

constant assistance and encouragement.

I am very grateful to my friends and family for going through this phase in life

with me and providing me with their unwavering encouragement and support, for

which I would not have made it this far.

And thanks be to God, with whom all things are possible.

iv

TABLE OF CONTENTS

SUMMARY ix

LIST OF TABLES x

LIST OF FIGURES xi

LIST OF

ABBREVIATIONS

xii

LIST OF PUBLICATIONS xiv

INTRODUCTION

1.1 Synapse 1

1.1.1 Synaptic plasticity 3

1.2 Dendritic spines 6

1.2.1 Dendritic spine morphology, formation and

maturation

7

1.2.2 Proposed functions of dendritic spines 9

1.3 AMPARs and synaptic plasticity 10

1.3.1 AMPARs trafficking 14

1.4 Calcium and synaptic plasticity 15

1.5 ER and synaptic plasticity 17

1.6 SOCE 19

1.6.1 SOCE in neurons 21

1.7 STIMs: mechanism of action 23

v

1.7.1 STIM1 versus STIM2 25

1.7.2 Functions of STIMs in neurons 27

1.8 Main hypothesis and research objectives 30

MATERIALS AND METHODS

2.1 DNA constructs 31

2.2 Short hairpin RNA (shRNA) 32

2.3 Antibodies used and their dilutions 32

2.4 Media and Buffers 33

2.5 Tissue Culture 35

2.5.1 Rat primary hippocampal and cortical neuronal

culture

35

2.5.2 Organotypic slice culture 36

2.6 Lentivirus production 36

2.7 Transfection 37

2.7.1 Lipofectamine 37

2.7.2 Electroporation 37

2.7.3 Transduction 38

2.7.4 Biolistic transfection 38

2.8 Electrophysiology 39

2.9 Immunocytochemistry 39

2.10 Organotypic culture fixation and scaling 40

2.11 Imaging 40

vi

2.12 Preparation of protein extracts from cells or tissue 41

2.13 Biochemical isolation of the PSD 42

2.14 Immunoprecipitation (IP) 43

2.15 Western transfer and immunoblotting 43

2.16 Measurement of cAMP levels and PKA activity 44

2.17 GluA1 endocytosis 44

2.18 Image analysis 45

2.18.1 Spine detection and classification 45

2.18.2 Scoring synaptic density 45

2.18.3 Sholl analysis 45

2.18.4 SEP-GluA1 insertion 46

2.18.5 Detection of endocytic GluA1 puncta 46

2.18.6 Granularity 46

2.19 Statistics 46

RESULTS

3.1 Expression of STIM isoforms differs in different parts of

the brain and is developmentally regulated

48

3.2 STIM2 localizes to large dendritic spines and is enriched in

the post-synaptic density

50

3.3 STIM2 regulates dendritic spine morphogenesis 54

3.4 STIM2 shapes spontaneous synaptic transmission 58

3.5 STIM2 reciprocally regulates phosphorylation of GluA1 on 61

vii

Ser831 and Ser845

3.6 STIM2 does not affect adenylate cyclase and PKA activity 64

3.7 STIM2 promotes exocytosis but inhibits endocytosis of

AMPARs

68

3.8 STIM2 forms a complex with GluA1 and AKAP150/PKA

and couples PKA to GluA1

71

3.9 The STIM2 SOAR domain mediates PKA-dependent

phosphorylation of GluA1 and is important for dendritic

spine morphogenesis

74

3.10 SOCE does not seem to be required for dendritic spine

morphogenesis

78

3.11 STIM2 translocates to puncta in response to cLTP and

promotes local assembly of a STIM2-AKAP150/PKA-

GluA1 signaling complex

79

3.12 Other possible impact of STIM2 on post-synaptic

components

81

DISCUSSION

4.1 STIM2 regulates dendritic spine morphogenesis 84

4.2 Formation and/or maturation of dendritic spines 86

4.3 Silent Synapses 87

4.4 Role of Ca2+

in STIM2 dependent dendritic spine

morphogenesis and remodeling

87

viii

4.5 STIM2 versus STIM1 in dendritic spine morphogenesis

and remodeling

89

4.6 STIM2 and Orai in synaptic plasticity 92

4.7 STIM2 activation 93

4.8 cLTP: Forskolin/Rolipram-induced cLTP vs Glycine-

induced cLTP

94

4.9 STIM2 forming a complex with AKAP150, PKA and

GluA1

95

4.10 GluA1 phosphorylation and LTP/LTD 95

4.11 STIM2 signals at ER-PM contact sites and enforces

proximity between GluA1, AKAP150 and PKA

97

4.12 Working model for STIM2-dependent phosphorylation of

GluA1 and dendritic spine remodeling

99

CONCLUSION 103

REFERENCES 104

APPENDICES

ix

SUMMARY

The endoplasmic reticulum (ER) is an organelle that can be found throughout the

axon, dendrites and synapses of neurons and it is known to regulate synaptic

function and plasticity, primarily through Ca2+

release from the ER. However, it is

unknown if the ER can directly communicate with the synapse and instruct

specific synaptic modifications through mechanisms other than ER Ca2+

release.

Stromal Interaction Molecules (STIMs), STIM1 and STIM2, are ER-resident

transmembrane Ca2+

sensors that mediate Store-Operated Ca2+

Entry (SOCE) in

non-excitable cells. Here, STIM2 was identified to be an important organizer of

excitatory synapses in the brain. STIM2 but not STIM1, was shown to regulate

the formation of dendritic spines and shape spontaneous synaptic transmission in

excitatory neurons. Furthermore, STIM2 can translocate into puncta in dendrites

and dendritic spines in response to cyclic adenosine monophosphate (cAMP)

elevation, and mediates protein kinase A (PKA)-dependent phosphorylation of the

GluA1 subunit of the AMPA receptor. This phosphorylation event is mediated by

local assembly of a STIM2-AKAP79/150-PKA complex that promotes AMPA

receptor surface delivery through opposing effects on GluA1 exocytosis and

endocytosis, thus regulating activity-dependent surface delivery of the AMPA

receptor. Collectively, this study suggests a novel mechanism of synaptic

plasticity in which the ER communicates with the synapse through dynamic

assembly of a PKA signaling complex at ER-PM contact sites.

x

LIST OF TABLES

Table 1: The antibodies used in this project, their manufacturers, the

dilutions and usage

32

Table 2: The buffers used in this project, their components and usage. 33

xi

LIST OF FIGURES

Figure 1. Differences between chemical and electrical synapses. 3

Figure 2. Morphology of different types of dendritic spines. 9

Figure 3. Structure of AMPA receptors (AMPARs). 11

Figure 4. Mechanisms in which calcium can enter the dendritic

spines and dendrites after depolarization.

17

Figure 5. Mechanism showing how SOCE is activated. 21

Figure 6. Domain organization of the human (h) and mouse (m)

STIM proteins.

27

Figure 7. STIMs expression differs in different parts of the brain

and is developmentally regulated.

49

Figure 8. STIM2 is localized to dendritic spines. 52

Figure 9. STIM2 is required for dendritic spines

formation/maturation and dendrite branching.

56

Figure 10. STIM2 shapes functional synaptic inputs. 59

Figure 11. Reciprocal regulation of GluA1 Ser831 and Ser845

phosphorylation by STIM2.

63

Figure 12. STIM2 knockdown does not affect cAMP production

and PKA activity.

66

Figure 13. STIM2 mediates cAMP-dependent surface delivery of

GluA1.

69

Figure 14. STIM2 forms a complex with AKAP150, PKA and

GluA1 and is required for GluA1 to interact with AKAP150/PKA.

73

xii

Figure 15. The STIM2 SOAR domain mediates phosphorylation of

and interaction with GluA1.

77

Figure 16. SOCE is not crucial for STIM2 dependent dendritic

spine morphogenesis and remodelling.

79

Figure 17. cAMP can trigger STIM2 puncta formation and

translocation into dendritic spines.

80

Figure 18. Other potential mechanism in which STIM2 may affect

spinogenesis.

83

Figure 19. Sequence alignment between STIM1 and STIM2. 91

Figure 20. Model for STIM2-dependent regulation of AMPAR

phosphorylation and trafficking at the interface between the ER and

the PM.

102

xiii

LIST OF ABBREVIATIONS

AKAP 79/150 A-kinase anchoring protein 79/150

AMPARs α-amino-3-hydroxy-5-methyl-4-isoxazoleproprionate

receptors

CAD CRAC activation domain

CaMKII calmodulin kinase II

cAMP cyclic adenosine monophosphate

CC coiled-coil

cereb cerebellum

CICR Ca2+

-induced Ca2+

release

cLTP chemical LTP

cPKA catalytic protein kinase A

CRACM/Orai Ca2+

release–activated Ca2+

channel pore

CREB cAMP/Ca2+ response element-binding protein

ctx cortex

DIV days in vitro

ER endoplasmic reticulum

GABA γ-aminobutyric acid

hipp hippocampus

ICC Immunocytochemistry

IP3 inositol-1,4,5-trisphosphate

IP3Rs inositol-1,4,5-trisphosphate receptors

xiv

LTP and LTD Long term potentiation and depression

mGluRs metabotropic glutamate receptors

NBQX 6-nitro-7-sulphamoylbenzol[f]quinoxaline-2,3-dione

NMDARs N-Methyl-D-aspartate receptors

OASF Orai1-activating small fragment

PKA protein kinase A

PKC protein kinase C

PM plasma membrane

PMCA plasma membrane Ca2+

ATPase

Pre/SV presynaptic membranes/synaptic vesicles

PSD post-synaptic density

rPKA regulatory protein kinase A

RyRs ryanodine receptors

SAM sterile α-motif

SAP 97 synapse associated protein 97

scr scrambled shRNA

SER smooth endoplasmic reticulum

Shank SH3 and multiple ankyrin repeat domains

shRNA short hairpin Ribonucleic Acid

SIM Structured Illumination Microscopy

SOAR STIM–Orai activating region

SOC store operated Ca2+

channels

SOCE store-operated calcium entry

xv

STIMs stromal interaction molecules

TIRF Total Internal Reflection Fluorescence

VGCCs voltage-gated Ca2+

channels

xvi

List of Publications

Gisela Garcia-Alvarez, Bo Lu*, Kenrick An Fu Yap*, Loo Chin Wong, Jervis

Vermal Thevathasan, Lynette Lim, Fang Ji, Kia Wee Tan, James J. Mancuso,

Willcyn Tang, Shou Yu Poon, George J. Augustine and Marc Fivaz. “STIM2

regulates PKA-dependent phosphorylation and trafficking of AMPARs”.

Lim L, Jackson-Lewis V, Wong LC, Shui GH, Goh AX, Kesavapany S, Jenner

AM, Fivaz M, Przedborski S, Wenk MR. “Lanosterol induces mitochondrial

uncoupling and protects dopaminergic neurons from cell death in a model for

Parkinson’s disease”. Cell Death Differ. 2012 Mar; 19(3):416-27.

Wong L. C, Lu B, Tan K. W, and Fivaz M., “Fully-automated image processing

software to analyze calcium traces in populations of single cells”. Cell Calcium

2010 (48) 270-274

1

Introduction

This thesis explores the interaction of the endoplasmic reticulum (ER) with

excitatory synapses in the mammalian nervous system and focuses, in particular,

on the role of the ER-resident STIM2 protein on synapse formation, structure and

function. The following introduction presents an overview on synaptic plasticity

and its relationship with ER signaling, with emphasis on the biology of the STIM

proteins.

1.1 Synapse

Synapses are highly specialized structures that allow one neuron to pass an

electrical or chemical signal to another neuron (Purves et al., 2001). The human

brain consists of approximately 100 billion neurons with each neuron making an

estimate of about 1000 to 10,000 synaptic connections with other nerve cells

depending on the part of the brain (Brewer et al., 2009).

Synapses can be classified into two types, electrical and chemical. Electrical

synapses are made up of a gap junction at a distance of about 3.5nm of each other

between the pre and postsynaptic neuron, where there is a direct mechanical and

electrical conductive link between the two neurons (Figure 1) (O'Brien, 2014;

Pereda, 2014; Purves et al., 2001). Thus electrical changes in the presynaptic

neuron can directly induce a bidirectional electrical change in the postsynaptic

neuron. When the presynaptic action potential propagates to the postsynaptic

neuron, the resting membrane potential of the postsynaptic neuron is able to

2

simultaneously propagate back to the presynaptic neuron (Pereda, 2014). Fast

responses such as defensive reflexes are often made up of electrical synapses as

they conduct nerve impulses faster than chemical synapses (O'Brien, 2014;

Pereda, 2014; Purves et al., 2001).

In chemical synapses, the distance between the pre and postsynaptic neuron is

known as the synaptic cleft, which is about 20 to 40nm (Figure1) (Pereda, 2014;

Purves et al., 2001). Depolarisation of the presynaptic neuron causes the release

of neurotransmitters from the presynaptic terminal into the synaptic cleft, which

will then diffuse and bind receptors on the postsynaptic terminal. The binding of

the neurotransmitter can trigger a cascade of events and have complex effects on

the postsynaptic neuron. The remaining neurotransmitter is then cleared by either

reuptake by the presynaptic cell, and then repackaged for future release, or broken

down enzymatically (Purves et al., 2001). Chemical synapses are classified based

on the type of neurotransmitters they release and can co-exist with electrical

synapses (Pereda, 2014).

3

1.1.1 Synaptic plasticity

Synaptic plasticity is defined as the ability of synapses to change their strength or

conductance in response to a broad range of stimuli. Both electrical and chemical

synapses have been observed to be constantly changing in response to stimuli,

although the changes at a chemical synapse are more complex and better studied

(O'Brien, 2014; Pereda, 2014). Plasticity at chemical synapses can be achieved by

both pre- and postsynaptic mechanisms. These include activity-dependent

changes in the number and activity of postsynaptic ionotropic receptors,

Figure 1. Differences between chemical and electrical synapses. Electrical

synapse is mediated by clusters of intercellular channels called gap junctions that

connect the interior of two adjacent neurons which directly enable the bidirectional

passage of electrical currents. Chemical synapses are classified according to the

neurotransmitters released at the presynaptic site. Complex molecular machinery is

required at both the presynaptic terminal for the release of neurotransmitter, and at

postsynaptic terminals to detect and translate the presynaptic message

(neurotransmitters) into various postsynaptic events. Electrical synapses are

mediated by intracellular channels called gap junctions, which connect the interior

of two adjacent neurons. This enables direct bidirectional passage of electrical

currents carried by ions. Adapted from Pereda, A.E. Nature Reviews Neuroscience

15, 250-263.

4

modification in synapse structure and size, or changes the amount of

neurotransmitters released pre-synaptically. In addition, neuronal plasticity can

also occur by the addition or removal of synapses. In electrical synapses,

adjustments in conductance can be achieved by changes in the expression levels

of gap junction proteins (O'Brien, 2014).

Synaptic plasticity at chemical synapses has been linked to learning and memory

(Martin et al., 2000; Martin and Morris, 2002; Purves et al., 2001). Donald O.

Hebb first proposed a mechanism in which synaptic connections can become

stronger with repeated and persistent presynaptic stimulation of the postsynaptic

neuron (Hebb, 1949). This is known as Hebbs’ rule and is one of the first

proposed models that link synaptic plasticity to learning and memory formation.

Synapses are strengthened and reinforced when both the pre and postsynaptic

terminals are persistently activated at the same time and the strengthened synapse

is thought to be responsible for learning or storing new memories (Purves et al.,

2001).

The most widely accepted experimental model to support the association between

synaptic plasticity and learning and memory is long term potentiation (LTP) and

depression (LTD). LTP was first discovered in the hippocampus of the adult

rabbit (Lomo, 1966; Lømo, 2003), a structure in the brain that is important for the

formation and storage of memories (Purves et al., 2001). When the presynaptic

fibers of the perforant pathway were exposed to brief high frequency stimulations

5

(∼ 100 Hz), a long-lasting increase in the strength of synaptic transmission is

observed from the postsynaptic cells of the dentate gyrus and this is known as

LTP (Lomo, 1966; Lømo, 2003). On the other hand, when neurons are exposed to

prolonged low frequency (∼ 1 Hz) stimulation, a persistent reduction in synaptic

transmission is observed and this is known as LTD (Bear and Malenka, 1994;

Bliss and Collingridge, 1993; Nicoll and Malenka, 1999).

LTP and LTD have been proposed to be the neural substrate of learning and

memory (Barnes, 1995; McGaugh, 2000; Purves et al., 2001; Stevens, 1998;

Stuchlik, 2014), although causal evidence for a role of LTP/LTD in

learning/memory has been difficult to establish. A recent study has, however,

provided further support for a role of LTP in fear conditioning, a form of

conditioned learning in which a mouse learns to associate a neutral stimulus such

as a tone to an aversive stimulus, in the form of an electric foot shock. In this

recent study, the tone was replaced with optogenetic stimulation of the auditory

neural inputs to the amygdala, a brain region known to be essential for fear

conditioning (Nabavi et al., 2014). Optogenetic stimulation triggering LTP in the

amygdala was sufficient to induce a fear response after the rat was conditioned to

associate the foot shock with optogenetic stimulation of the auditory inputs. On

the other hand, optogenetic stimulation triggering LTD is able to inactivate the

memory (Nabavi et al., 2014). This study suggests that LTP is used to create a

memory, such as fear conditioning and that LTD can be used to inactivate the

6

memory, further strengthening the notion that LTP and LTD is important for

learning and memory.

1.2 Dendritic spines

The postsynaptic terminals of excitatory synapses are structures known as

dendritic spines. Dendritic spines are small membrane protrusions from neuronal

surfaces such as the soma, dendrites or even axon hillock (Bosch and Hayashi,

2012; Nimchinsky et al., 2002). Most excitatory synapses (more than 90%)

terminate on dendritic spines in the brain (Nimchinsky et al., 2002) although

sometimes inhibitory connections can be made onto the same spine head (Higley,

2014; Katona et al., 1999). Dendritic spines can be found on most principal

neurons in the brain, such as the pyramidal neurons of the neocortex, the medium

spiny neurons of the striatum, and the Purkinje cells of the cerebellum

(Nimchinsky et al., 2002). The excitatory pyramidal neurons of the hippocampus,

are one of the main type of neurons in the densely packed hippocampus that are

lined with dendritic spines on the dendrites (Nimchinsky et al., 2002; Purves et

al., 2001). Different neurons have different dendritic spines densities and

development, and the organelles and receptor compositions in a dendritic spines

are also different (Nimchinsky et al., 2002). However, not all neurons have

dendritic spines. For example, γ-aminobutyric acid (GABA) ergic interneurons

are still able to function without dendritic spines (Scheuss and Bonhoeffer, 2013).

7

The importance of synapses are seen in many neuropsychiatric diseases and

neurological disorders such as mental retardation, autism spectrum disorders, and

schizophrenia where synapse structure and function are altered (Van Spronsen

and Hoogenraad, 2010). Importantly, abnormal spine structures and spine loss

have been observed in various neurological disorders, such as fragile X syndrome,

Rett’s syndrome, Down syndrome, schizophrenia and Alzheimer’s disease (Knafo

et al., 2012; Penzes et al., 2011; van Spronsen and Hoogenraad).

1.2.1 Dendritic spine morphology, formation and maturation

A dendritic spine typically consists of a spine head, with volume ranging from

0.01µm3 to 0.8µm

3, connected to the neuron by a thin spine neck, with a diameter

less than 0.1 µm. Based on their shape, dendritic spines can be classified into

three main categories: “thin", "stubby" and "mushroom" (Figure 2). Mushroom

spines have a large spine head with a narrow neck. Stubby spines do not have an

obvious neck and thin spines have a small spine head with a narrow neck

(Nimchinsky et al., 2002). Another category, the filopodium, has been used to

classify dendritic protrusions that have hair like morphology. It is believed that in

a developing neuron, a filopodium is first formed, which dynamically samples the

environment in search for presynaptic targets (Richards et al., 2005). Once a

contact with a presynaptic terminal is formed, the filopodium then retracts and

becomes a dendritic spine (Sorra and Harris, 2000). It has been proposed that in

the absence of activity, there is no formation of new dendritic spines and existing

ones can disappear or be converted into thin filopodia in pyramidal neurons of the

8

hippocampus (Segal, 2010). Dendritic spines on the Purkinje neurons on the other

hand, can form in the absence of presynaptic input (Yuste and Bonhoeffer, 2004),

suggesting that the requirements and the process of forming a dendritic spine are

different on different type of neurons.

In general, dendritic spines with a larger spine head such as the mushroom spine

are considered to be mature and the maturity of a dendritic spine is dependent on

the organelles and receptors present. Most mature dendritic spines in the

pyramidal neurons of the CA1 region in the hippocampus have a well-defined

spine apparatus, which is made up of cisternae of smooth endoplasmic reticulum

(SER) for calcium handling (Korkotian and Segal, 2011; Segal and Korkotian,

2014; Segal et al., 2010). In addition, dendritic spines require α-amino-3-

hydroxy-5-methyl-4-isoxazoleproprionate receptors (AMPARs) to be synaptically

inserted in order for dendritic spines to expand and mature (Kopec et al., 2006;

Korkotian and Segal, 2007). The importance of AMPARs for dendritic spine

maturation is observed when chronic blockage of AMPARs with 6-nitro-7-

sulphamoylbenzol[f]quinoxaline-2,3-dione (NBQX) led to disappearance of

dendritic spines (Mateos et al., 2007). Large dendritic spine heads are also

correlated to strong synapses and changes in spine size can affect synaptic

strength and amplitude of both miniature and evoked synaptic currents

(Nimchinsky et al., 2002; Segal, 2010).

9

1.2.2 Proposed functions of dendritic spines

There are several proposed functions of dendritic spines. First, they are thought to

increase synaptic connectivity with as many axons as possible during

development (Yuste, 2011). For example, CA3 neurons send their axons to the

dendrites of CA1 neurons in the hippocampus. The axons of CA3 are almost sent

in a straight manner over to the extending dendrites of the CA1 neurons (Yuste,

2011). To form as many synaptic connections as possible, dendrites branch out as

wide as possible and having dendritic spines increases the surface area and chance

of forming a synapse. In addition, dendritic spines are highly dynamic and

motile, which can aid in looking for axonal terminals (Konur and Yuste, 2004)

Second, dendritic spines allow the linear integration of inputs without affecting

other dendritic spines. For example, not all postsynaptic dendritic spines are

activated at the same time and/or with the same intensity. The neuron has to

Figure 2. Morphology of different types of dendritic spines. The 3 types of

dendritic spines that can be found on an excitatory neuron; mushroom, stubby and

thin dendritic spines. All 3 morphology differs in the area of the spine head and the

neck diameter. Mushroom spines are thought to be considered matured while thin

spines are considered to be immature. Filopodium are considered to be precursors

of dendritic spines and they lack a dendritic spine head.

10

receive all the inputs from all the dendritic spines and integrate the inputs in a

manner where the signals do not interfere with each other. This can be achieved

with the electrical isolation of a narrow spine neck, which serves as a barrier that

protects the dendrites from the dendritic spine open conductance (Yuste, 2011).

Third, dendritic spines are highly plastic and their number, size and shape can

rapidly change in response to experience or synaptic stimulation (Yuste, 2011).

Spine morphology is intimately linked to synapse function – small spines support

reduced synaptic transmission, while long-term potentiation (LTP), a form of

long-lasting increase in synaptic strength causes enlargement of spine heads

(Matsuzaki et al., 2004a).

1.3 AMPARs and synaptic plasticity

AMPARs are one of the main ionotropic glutamate receptors that mediate the

majority of fast excitatory neurotransmission in the brain (Huganir and Nicoll,

2013a) and they have been linked to the size and maturity of dendritic spines

(Hanley, 2008; Hering and Sheng, 2001; Matsuzaki et al., 2001). AMPARs

consist of 4 subunits (GluA1-4) which assemble in different combinations of

dimer-of-dimers to form tetrameric ion channels with distinct functional

properties (Figure 3) (Greger et al.; Madden, 2002; Mayer, 2005; Shepherd and

Huganir, 2007). They are given the name AMPA receptors because of their

selectivity for AMPA (Honoré et al., 1982). The binding of glutamate or AMPA

to AMPARs results in opening of the channel and influx of Na+, which causes the

11

neuron to depolarize. AMPARs containing the GluA2 subunit are impermeable to

Ca2+

. This is due to the RNA editing of the GluA2 mRNA (Wright and Vissel,

2012; Yamashita and Kwak, 2014), which involves post-transcriptional

modification of a codon originally encoding Glutamine at position 607.

Specifically, RNA editing converts Adenosine to Inosine in the critical codon,

resulting in the replacement of the uncharged amino acid glutamine (Q) with the

positively charged arginine (R) (Wright and Vissel, 2012; Yamashita and Kwak,

2014), making it energetically unfavorable for Ca2+

to pass through the ion

channel containing GluA2 subunit. Ca2+

impermeable AMPARs are thought to

protect the neuron from excitotoxicity.

All 4 subunits of the AMPAR can be phosphorylated on their intracellular

carboxy terminal tails. Phosphorylation in the C-terminal domain can occur on

several different sites and regulates various aspects of AMPAR function,

including trafficking and channel properties (Song and Huganir, 2002). GluA1

Figure 3. Structure of AMPA receptors (AMPARs). Individual subunits are

made up of four transmembrane domains and the AMPAR consists of four

subunits, which is assembled in different combinations of dimer-of-dimers to form

tetrameric ion channels. The dimers are usually two different subunits, such as

GluR1 and -2 or GluR2 and -3. Adapted from Shepherd, J.D., and Huganir, R.L.

Annual Review of Cell and Developmental Biology 23, 613-643.

12

has an unusually long C-terminal cytoplasmic tail, which is primarily

phosphorylated on two Serine residues, pSer831 and pSer845. These two

phosphorylation events have been extensively studied and regulate activity-

dependent changes in synaptic localization and channel properties of GluA1-

containing AMPARs (Banke et al., 2000; Derkach et al., 1999; Esteban et al.,

2003).

Both S831 and S845 residues have been implicated in trafficking of AMPARs,

LTP, LTD and spatial memory (Esteban et al., 2003; Huganir and Nicoll, 2013b;

Kessels and Malinow, 2009; Lee et al., 2010a; Lee et al., 2003; Makino et al.,

2011). In knock-in mice with mutations to both the GluA1 phosphorylation sites

(S831A and S845A mutants), LTP was moderately reduced but still present,

suggesting that both sites are not critical for LTP but for stabilizing LTP (Lee et

al., 2003). In contrast, knock-in mutant mice with either S831A or S845A

mutation still have normal LTP, suggesting that phosphorylation of either site is

sufficient to maintain stable LTP (Lee et al., 2010a). In addition, only LTD is

impaired in S845A mice, but not S831A mice, which suggests that S845 is

important for LTD (Lee et al., 2010a).

The phosphorylation of GluA1 S831 is dependent on protein kinase C (PKC) and

also CaMKII, which are both calcium-dependent kinases. The role of GluA1 S831

in LTP is unclear but persistent increase in S831 phosphorylation has been

observed following LTP (Barria et al., 1997; Lee et al., 2000b). The

13

phosphorylation of Ser831 can also directly affect the channel properties of the

AMPAR (Derkach et al., 1999) and has been thought to mediate the observed

increase in AMPAR conductance following LTP (Benke et al., 1998; Luthi et al.,

2004; Poncer et al., 2002).

In contrast, the phosphorylation of GluA1 S845 is mediated by protein kinase A

(PKA), which is a cAMP dependent kinase. Phosphorylation ofS845 at the

synapse requires A-kinase anchoring protein 79/150 (AKAP79 in humans and

AKAP150 in rodents) (Tavalin et al., 2002), which brings PKA in close proximity

to GluA1 by binding to synapse associated protein 97 (SAP 97) (Colledge et al.,

2000). SAP97 is a PSD scaffold that is known to interact with an ER pool of

GluA1 early in the biosynthetic pathway (Sans et al., 2001) and also co-localize

with GluA1 at postsynaptic sites (Valtschanoff et al., 2000). Phosphorylation of

Ser845 has been shown to increase channel open probability (Banke et al., 2000)

and promotes synaptic incorporation of GluA1 (Esteban et al., 2003), through

mechanisms that are not completely understood. Synaptic recruitment of

AMPARs correlates with increase in synaptic efficacy and enlargement of

dendritic spines (Harris et al., 1992; Matsuzaki et al., 2004a; Nusser et al., 1998).

The signaling mechanisms that coordinate synaptic levels of AMPARs with spine

size are presumably key for the expression of synaptic plasticity, but remain

poorly characterized.

14

1.3.1 AMPARs trafficking

The addition and removal of AMPARs at postsynaptic sites is important for the

expression of LTP and LTD (Huganir and Nicoll, 2013b; Kessels and Malinow,

2009). AMPARs are mostly concentrated at excitatory synapses (Craig et al.,

1993) and there is recruitment of GluA1 to dendritic spines after LTP induction,

which correlates with synaptic strengthening (Hayashi et al., 2000; Shi et al.,

1999) and this activity-dependent strengthening of the synapse is dependent on

GluA1 (Kessels and Malinow, 2009). In contrast, studies have shown that there is

rapid endocytosis of AMPARs from the plasma membrane of synapses when

neurons are subjected to chemically induced LTD using glutamate or NMDA

(Beattie et al., 2000; Carroll et al., 1999; Ehlers, 2000b). The main mechanism to

remove GluA1 containing AMPARs is through dephosphorylation of GluA1 at

S831 and S845 with phosphatases (Kameyama et al., 1998; Lee et al., 2000a;

Mulkey et al., 1993).

Other than trafficking of AMPARs at the synapse, AMPARs have also been

discovered to be mobile at extrasynaptic sites and can enter synapses. (Borgdorff

and Choquet, 2002; Opazo and Choquet, 2011). In addition to trafficking of

AMPARs at extrasynaptic sites, AMPARS can also laterally diffuse in and out of

synapses to regulate the amount of AMPARs at the synapse (Opazo and Choquet,

2011). Using Fluorescent Recovery After Photobleaching to monitor superecliptic

pHluorin-tagged GluA1, a pH sensitive form of GFP which preferentially labels

15

AMPARs expressed on the cell surface, AMPARs have also been shown to be

recruited to synapses through lateral diffusion from extrasynaptic sites during

LTP (Makino and Malinow, 2009).

1.4 Calcium and synaptic plasticity

Calcium is a ubiquitous second messenger that regulates a broad range of

neuronal functions including neurite outgrowth, axon pathfinding (Gomez and

Zheng, 2006; Wen et al., 2004), and gene expression (Tabuchi, 2008). Calcium

has been shown to be an essential trigger for the expression of synaptic plasticity,

LTP and LTD. (Baker et al., 2013; Mizuno et al., 2001; Rose and Konnerth,

2001).

The basal concentration of cytosolic Ca2+

is ~10,000 fold below that of the

endoplasmic reticulum (ER) concentration, while the extracellular environment

can contain 5-10 times more Ca2+

than the ER (Meldolesi and Pozzan). The influx

of Ca2+

into the cytosol can come from both extracellular and intracellular

sources. Upon stimulation with neurotransmitters (Figure 4), the main rise in

postsynaptic cytosolic calcium that is critical for LTP in excitatory neurons, is

through the entry of Ca2+

via the ionotropic glutamate receptor, N-Methyl-D-

aspartate receptors (NMDARs) (Horak et al., 2014). Ca2+

can also enter into the

cytosol and dendritic spines of neurons through (i) voltage-gated Ca2+

channels

(VGCCs) (Nishiyama et al., 2003), (ii) GluA2-lacking AMPARs, (iii) transient

16

receptor potential channels (Wang and Poo, 2005), (iv) as well as Ca2+

mobilization from internal ER stores.

There are two main ways in which Ca2+

can be mobilized from the ER. The first

is via Ca2+

-induced Ca2+

release (CICR) (Emptage et al., 2001) and the second is

through production of inositol-1,4,5-trisphosphate (IP3) acting on inositol-1,4,5-

trisphosphate receptors ( IP3Rs) (Baker et al., 2013; Tang and Kalil, 2005). CICR

is known to be dependent on ryanodine receptors (RyRs) and is activated in

neurons mainly through influx of Ca2+

from VGCCs at the plasma membrane

(Berrout and Isokawa, 2009; Chavis et al., 1996; Tully and Treistman, 2004). In

contrast, ER Ca2+

release through neuronal IP3R is triggered mainly by

metabotropic or cholinergic inputs (de Sevilla et al., 2008; Power and Sah, 2007).

The local intracellular Ca2+

stores from the ER at a synapse can help to amplify

Ca2+

signals from the channels at the plasma membrane via a positive feedback

mechanism (Horne and Meyer, 1997; Taylor, 1998; Verkhratsky and Shmigol,

1996). As a result of this local amplification, a global Ca2+

wave can travel from

the synapse to the dendrite and cell body (Baker et al., 2013).

Another possible mechanism to increase cytosolic levels of Ca2+

in the cell body

and synapse of neurons is through store-operated calcium entry (SOCE) (Baba et

al., 2003; Bouron, 2000; Emptage et al., 2001). SOCE is activated when ER Ca2+

stores are depleted due to CICR and/or IP3Rs receptors. As a result of the ER

depletion, a sustained phase of Ca2+

entry occurs across the plasma membrane

17

(PM) through the opening of specialized store operated Ca2+

channels (SOC)

(Putney, 2007).

1.5 ER and synaptic plasticity

The ER has been shown to be critical for regulating structural and functional

changes in neural circuits in both the developing and adult nervous systems

through ER calcium stores (Bardo et al., 2006; Mattson et al., 2000) and also the

transport of synaptic components such as NMDARs in the biosynthetic pathway

(Horak et al., 2014). The ER has also been linked to pathological synaptic defects

(Salminen et al., 2009) and abnormal ER Ca2+

and protein homeostasis have been

implicated in several major neurodegenerative disorders, including Alzheimer's

and Parkinson's disease (Mattson et al., 2000). Despite the importance of the ER

Figure 4. Mechanisms in which calcium can enter the dendritic spines and

dendrites after depolarization. Synaptic activation depolarizes the synapse

through AMPARs. This enhances the conductance of NMDARs and activates

voltage sensitive calcium channels (VSCCs) or also known as voltage gated

calcium channels (VGCCs), contributing to additional depolarization and increase

in Ca2+

entry. Ca2+

release from internal stores such as the smooth endoplasmic

reticulum (SER) can also increase the Ca2+

levels in the dendritic shaft and

dendritic spines. The membrane potential is regulated by the influx of Na+ and K+

through sodium and potassium channels (not shown). Adapted from Higley, M. J.

& Sabatini, B. L. Neuron 59, 902-913 (2008).

18

in both physiological and pathological synaptic functions, surprisingly little is

known about how and this organelle communicates with synapses.

The ER is a vast continuous dynamic network of sheets and tubular membranes

that connect the soma to the neuron's dendritic and axonal arbors, and protrudes

into large dendritic spines (Bourne and Harris, 2012). Thus, the organization of

the ER is particularly suited to process synaptic inputs locally and to integrate

information over long distances. In addition, a recent study has shown that

increased ER complexity occurs at dendritic branch points and can define sites of

dendritic morphogenesis through confinement of membrane cargo in the ER (Cui-

Wang et al., 2012) and also possibly by regulating trafficking of newly-

synthesized post-synaptic components to dendritic spines (Valenzuela et al.,

2011).

The SER in dendritic spine are unique calcium compartments that can raise Ca2+

levels locally, without affecting the neighboring dendritic compartment or the

dendrite. This allows the neuron to restrict Ca2+

transient to activated synapses

and thus protects the neuron from uncontrolled rise in intracellular Ca2+

. The

ability of the ER to release Ca2+

in response to synaptic or other signaling inputs

is one important mechanism by which this organelle fine-tunes synaptic Ca2+

transients and mediates synaptic plasticity (Finch and Augustine, 1998;

Garaschuk et al., 1997; Lauri et al., 2003).

19

1.6 SOCE

Long-lasting Ca2+

increase mediated by SOCE is critical for several important

cellular functions in non-excitable cells such as growth, motility, secretion and

gene expression (Lewis, 2007). SOCE has been reported in both excitable and

non-excitable cells but and the majority of the work done on SOCE is on non-

excitable cells. SOCE in non-excitable cells is made up of two key players (Figure

5). The first component consists of the stromal interaction molecule (STIM)

proteins (STIM1 and STIM2), which are ER transmembrane proteins that can

sense the reduction of ER Ca2+

concentration and subsequently trigger SOCE. The

second component is the Ca2+

release–activated Ca2+

channel (Orai/CRACM1, 2

and 3), which make up the SOC channels. Both STIM isoforms sense the

decrease in ER Ca2+

through an EF-hand Ca2+

binding motif, which faces the ER

lumen (Liou et al., 2005; Roos et al., 2005). Dissociation of Ca2+

from the EF-

hand motif causes rapid oligomerization (Liou et al., 2007; Luik et al., 2008) and

subsequent migration of STIMs to ER-PM junction sites forming puncta (Liou et

al., 2005; Wu et al., 2006). Once localized to these ER-PM junction sites, STIMs

promotes the accumulation of Orai at apposed sites in the PM (Luik et al., 2006;

Xu et al., 2006), leading to channel activation (Penna et al., 2008). Store depletion

is essential and sufficient to activate SOCE (Putney, 2007).

Severely impaired SOCE has been observed in cells with reduced levels of

STIM1 (Baba et al., 2008; Oh-Hora et al., 2008). Conversely, a point mutant of

20

STIM1 in the EF-Hand motif, which no longer binds Ca2+

, pre-localizes to ER-

PM junction sites (even when stores are full) and leads to constitutive Ca2+

entry

through SOC channels (Liou et al., 2005). A functional Orai1 channel is also

essential for SOCE. A naturally-occurring single point mutation in Orai1 impairs

SOCE and causes severe combined immunodeficiency (SCID) syndrome (Feske

et al., 2006), and a nonsense mutation resulting in STIM1 depletion also leads to a

similar phenotype (Picard et al., 2009). STIM1 and STIM2 knock-out mice die

perinatally and within 4-5 weeks after birth respectively (Baba et al., 2008; Oh-

Hora et al., 2008).

Patients suffering from Orai1- or STIM1-deficiency mainly manifest

immunological diseases such as severe immunodeficiency and autoimmune

disorders (Cahalan, 2009). Other pathologies include anhidrotic ectodermal

dysplasia, congenital myopathy, chronic pulmonary disease and defective dental

enamel calcification, which all suggest the importance of SOCE in humans

(Berna-Erro et al., 2012).

21

1.6.1 SOCE in neurons

Although neuronal SOCE has been reported in neurons, the existence and

relevance of this Ca2+

pathway in the nervous system remain controversial. The

lack of a clear SOCE response in neurons has been observed before (Bouron et

al., 2005; Garcia-Alvarez et al., 2015; Park et al., 2010). In contrast to non-

excitable cells, where SOCE serves as the predominant Ca2+

entry pathway, store

depletion leads to little Ca2+

entry in neurons compared to Ca2+

influx through

VGCCs (Park et al., 2010). Moreover, influx of Ca2+

through Orai1 and the L-

Figure 5. Mechanism showing how SOCE is activated. ER Ca2+

stores are not

depleted, and STIM1 and Orai1 are dispersed throughout the ER and plasma

membrane, respectively (top panel). ER Ca2+

store depletion (middle panel) triggers

STIM1 to oligomerize into puncta and accumulate ER-PM junction sites. At the

same time, Orai1 is recruited and accumulates in regions of the plasma membrane

directly opposite the STIM1 clusters. The co-localization of STIM1 and Orai1

activates Orai1 channels and allows Ca2+ influx into the cell (bottom panel).

Adapted from Lewis, R. S. Nature 446, 284-287.

22

type voltage-gated Ca2+

channel, Cav1.2, are reciprocally regulated by STIM1 and

to a smaller extent by STIM2 (Park et al., 2010; Wang et al., 2010b). STIM1 can

bind to the C terminus of Cav1.2 to inhibit gating of the channel (Park et al., 2010)

and activation of STIM1 strongly suppresses the activity of Cav1.2 while

triggering SOCE through Orai (Wang et al., 2010a). This suggests that in the

presence of high levels of STIM1, Ca2+

influx cannot simultaneously occur

through Cav1.2 and Orai channels.

Other studies report, however, SOCE activity in nerve cells (Berna-Erro et al.,

2009; Gemes et al., 2011; Gruszczynska-Biegala et al., 2011; Sun et al., 2014).

For example, SOCE has been proposed to regulate the spontaneous release of

neurotransmitters in neurons (Emptage et al., 2001). In this study, the authors

show that depletion of ER Ca2+

as a result of CICR can activate SOCE, which

then influences the frequency of spontaneous neurotransmitter release in

presynaptic terminals of pyramidal neurons (Emptage et al., 2001). A recent study

has reported that SOCE is attenuated in depolarized cerebellar granule cells but is

constitutively active in resting cells (Lalonde et al., 2014b), suggesting that SOCE

may strongly depend on neuronal activity. STIM2 and SOCE have also been

implicated in hypoxic neuronal cell death (Berna-Erro et al., 2009), and recently

in dendritic spine maturation (Sun et al., 2014). SOCE has also been proposed to

be perturbed in Huntington’s (Bezprozvanny and Hayden, 2004) and Alzheimer’s

disease (Herms et al., 2003; Leissring et al., 2000; Smith et al., 2002), although

the involvement of SOCE in these pathologies primarily rely on Ca2+

23

measurements performed in non-neuronal cells such as fibroblasts derived from

patients.

Possible reasons for the inconsistencies of neuronal SOCE can be mainly grouped

into two categories. One, the use of various neuron cell types at different

developmental stages (see (Bouron et al., 2005)). Two, variations in the Ca2+

“addback” protocol for measuring SOCE, which involves depleting ER Ca2+

using drugs such as thapsigargin in a Ca2+

free extracellular media. Ca2+

is then

added back to the extracellular media in order for the influx of Ca2+

into the

cytoplasm through SOCE channels (Wong et al., 2010). The amount of

thapsigargin added, how Ca2+

is depleted and added back after store depletion can

affect how SOCE is measured. Even if SOCE is present in neurons, the

contribution of Ca2+

influx will be minimal as compared with Ca2+

influx through

voltage-gated calcium channels (VGCCs), which are the most abundant Ca2+

channels in neurons, during synaptic activities (Hooper et al., 2014). Thus, it has

been proposed that SOCE will most likely occur when neurons are at rest

(Lalonde et al., 2014a). Taken together, although the existence and relevance of

neuronal is debated, SOCE still provides an attractive model for rising

intracellular Ca2+

within a dendritic spine to regulate synaptic plasticity.

1.7 STIMs: mechanism of action.

The luminal N-terminal region of both STIM1 and STIM2 contains an EF-hand

and a sterile α-motif (SAM) domain (Berna-Erro et al., 2012) (Figure 6). Upon

24

ER Ca2+

depletion, Ca2+

dissociates from the EF-hand domain, exposing the

hydrophobic residues in both the EF-hand domain and the SAM domain which

triggers oligomerization of STIMs (Baba et al., 2006; Covington et al., 2010; Li et

al., 2007; Park et al., 2009; Williams et al., 2002). Therefore, the EF-hand-SAM

region is required to sense changes in ER Ca2+

levels and trigger oligomerization

of the STIM proteins.

The main domain in the cytoplasmic C-terminus of both STIM1 and STIM2 is the

STIM–Orai activating region (SOAR), which mediates direct coupling with Orai

channels (Figure 6). This region is similar to the CRAC activation domain (CAD)

or Orai1-activating small fragment (OASF) (Berna-Erro et al., 2012). The SOAR

domain contains a polybasic region (KIKKKR), which is crucial for the

interaction and activation of Orai1 (Calloway et al., 2010; Korzeniowski et al.,

2010; Yang et al., 2012). Another region within the SOAR domain, the coiled-coil

(CC1) domain, has been shown to be crucial for maintaining STIM1 oligomerized

state after store depletion (Covington et al., 2010; Yang et al., 2012) and deletion

of the CC1 region prevents STIM1 oligomerization (Baba et al., 2006). Therefore,

the SOAR region has two different roles. First, it mediates the transition to an

oligomerized state (Covington et al., 2010) and second, it binds to Orai channels

to trigger the activation of Orai channels (Park et al., 2009; Yuan et al., 2009).

Other than interacting with Orai1, STIM1 has been proposed to interact indirectly

with adenylate cyclase (Lefkimmiatis et al., 2009) and the Ca2+

pump, plasma

25

membrane Ca2+

ATPase (PMCA) (Krapivinsky et al., 2011). Other functions of

STIM1 include a heat (Xiao et al., 2011) and oxidative stress sensor (Hawkins et

al., 2010), suggesting that STIM proteins may impact multiple signal transduction

pathways.

1.7.1 STIM1 versus STIM2

The STIM1 and STIM2 genes are evolutionarily conserved and probably descend

from duplication of an ancestral STIM gene about 500 million years ago (Collins

and Meyer, 2011). Interestingly, this gene duplication coincides with the

emergence of dendritic spines and the explosion of brain complexity in higher

vertebrates (Garcia-Lopez et al., 2010).

Although conserved, there are still some minute differences between the two

STIM isoforms. First, the reported Ca2+

affinity (Kd) of STIM1 and STIM2 is

~200 µM and ~400 µM respectively (Soboloff et al., 2012). Thus STIM2 has a

lower affinity for Ca2+

than STIM1 and is consequently more sensitive to smaller

decreases in ER Ca2+

concentration (Brandman et al., 2007). In contrast, STIM1

requires a larger ER Ca2+

depletion, which typically occurs in in response to acute

receptor-mediated signaling. This has led to the notion than STIM2 regulates

basal Ca2+

influx and homeostasis, while STIM1 underlies receptor-mediated

store-operated Ca2+

influx (Brandman et al., 2007).

26

Second, STIM2 does not activate Orai channels as efficiently as STIM1 (Bird et

al., 2009; Parvez et al., 2008; Zhou et al., 2009). This can be explained by the

slower rate at which the EF hand-SAM motif of STIM2 unfolds and aggregates

upon Ca2+

dissociation from the EF hand (Stathopulos et al., 2009; Zheng et al.,

2011). In addition, overexpression of STIM2 has been shown to have a negative

effect on SOCE (Soboloff et al., 2006b). This may help to prevent uncontrolled

activation of SOCE when STIM1 is activated since STIM2 is more sensitized to

ER Ca2+

fluctuations.

Third STIM2 is exclusively expressed in the ER but STIM1 has been reported to

be expressed in the both the ER and PM (~ 5-10%) (Cahalan, 2009;

Hewavitharana et al., 2008; Saitoh et al., 2011; Soboloff et al., 2006a; Soboloff et

al., 2006b). However, the presence of STIM1 in the PM is still debatable because

the presence of STIM1 at the PM is often assessed in STIM1-overexpressing

cells, which can lead to mislocalization of STIM1. The ability of STIM2 to

remain in the ER is probably due to the strong ER-retention sequence in the

C-terminal, which is lacking in STIM1(Soboloff et al., 2012).

27

1.7.2 Functions of STIMs in neurons

STIM1 and STIM2 are differentially expressed in the mouse brain. Skibinska-

Kijek et al., have shown that STIM1 and STIM2 mRNA can be detected in all

major parts of the mouse brain but the level of STIM2 mRNA is about 7 times

higher than that of STIM1 in the hippocampus and 3 times higher in other parts of

the brain (Skibinska-Kijek et al., 2009a). Another interesting observation is that

STIM1 protein levels correlate strongly with mRNA expression in the brain,

which suggests that STIM1 expression is primarily regulated at the transcriptional

level. In contrast, STIM2 protein levels do not follow the mRNA level, which

suggests a more complex regulation. More importantly, the protein expression of

Figure 6. Domain organization of the human (h) and mouse (m) STIM proteins. The human and mouse STIM1 proteins share about 97% in amino acid homology

while human and mouse STIM2 share about 92% in homology. The amino- and the

carboxyl-terminus are represented as N and C, respectively. Different domains are

represented in different colors. Boundaries and amino acid position are indicated by

the numbers above and below the domains. Adapted from Berna-Erro, A., G. E.

Woodard, et al. Journal of Cellular and Molecular Medicine 16(3): 407-424.

28

STIM1 and STIM2 is highest in the Purkinje cells in the cerebellum and in the

hippocampus respectively (Skibinska-Kijek et al., 2009b). The cerebellum plays

an important part in fine motor control and Purkinje cells send inhibitory

projections to the deep cerebellar nuclei to modulate excitation in the cerebellum

(Purves et al., 2001). The hippocampus on the other hand is involved in memory

formation and is mostly made up of excitatory pyramidal neurons (Purves et al.,

2001). A study indicates that thapsigargin-evoked store depletion in cortical

neurons can cause the redistribution of YFP-STIM1 and YFP-STIM2 to puncta,

consistent with how the STIM-Orai machinery operates to regulate SOCE in non-

neuronal cells (Klejman et al., 2009).

Importantly, recent studies have shown that, in addition to Orai, STIM1 can

modulate the activity of VGCCs (Park et al., 2010). STIM1 has been shown to

interact with L-type VGCCs to promote the removal of the VGCCs from the PM

and inhibit the channel activity (Park et al., 2010). In addition, STIM1 may play a

role in synaptic plasticity in both cultured cerebellar granule neurons by

promoting SOCE when the neuron is hyperpolarized (Lalonde et al., 2014a), and

acutely isolated Purkinje neurons by regulating neuronal calcium signaling and

mGluR1-dependent synaptic transmission (Hartmann et al., 2014).

In contrast to STIM1, very little is known about the interacting partners of

STIM2. A recent study has shown that STIM2 is important for mushroom spines

to stabilize through STIM2 dependent SOCE (Sun et al., 2014). In another study,

29

STIM2 deficient mice took more time to learn and remember as compared to the

wild-type mice in the Morris water maze test, indicating a defect in hippocampus-

dependent spatial memory function (Berna-Erro et al., 2009). But the health of the

mice may contribute to the defect in learning and memory as the STIM2 deficient

mice also die a few weeks after birth (Cahalan, 2009; Oh-Hora et al., 2008).

Taken together, both STIM isoforms may play a role in plasticity by regulating

Ca2+

through SOCE in different parts of the brain. However, it remains to be

determined whether STIMs can interact with other partners in the brain and

regulate synaptic plasticity through mechanisms other than SOCE.

30

1.8 Main hypothesis and research objectives

The function of SOCE and STIM proteins are well established in non-excitable

cells but their functions still remains to be elucidated in the brain. The

confinement of ER localized STIM proteins to dendritic spines or postsynaptic

sites may provide a potential mechanism for localized Ca2+

influx and Ca2+

signaling microdomains in neurons to regulate synaptic plasticity. Therefore, the

main aim of this study is to:

(i) Determine the impact of the STIM proteins on synaptic structure and

function.

(ii) Dissect out the mechanisms by which STIM proteins regulate synapse

function and synaptic plasticity.

31

Materials and Methods

This section contains the materials and methods used in the thesis. I have

generated all the constructs used in this thesis unless otherwise stated. I have also

set up the shRNA system, lentivirus system and transduction, organotypic slice

culture and the biolistic transfection of the organotypic culture.

2.1 DNA constructs

The following lentiviral vectors pll3.7 (#11795) and FUGW-GFP (#14883), the

packaging plasmids Delta8.9 (#12263) and VSVG (#8454), and the pCI-SEP-

GluR1 plasmid (#24000) were obtained from Addgene. The PKA activity sensor,

AKAR3EV, was a gift from Dr. Miki Matsuda (Komatsu et al., 2011). The eGFP

of the FUGW plasmid was replaced with mCherry using the BamHI and EcoRI

sites to create a mCherry version of the FUGW. Human pDS-YFP-STIM 1 and 2

constructs were gifts from Dr. Tobias Meyer (Brandman et al., 2007; Liou et al.,

2005). The pDS-YFP-STIM2 constructs used in this thesis are resistant against

the shRNAs generated against the rodent versions of STIM2 and are thus used as

a rescue construct. The human STIM2 mutants, YFP-STIM2Δcyto and YFP-

STIM2ΔSOAR constructs, were made by deleting the cytoplasmic region (aa 329-

841) and SOAR region (aa 438-538) of the human STIM2 protein, YFP-STIM2

(accession number: AAI71766). YFP-STIM2 LQAA was made by mutating

LQ347/348AA (Yuan et al., 2009). All STIM and shRNA constructs used in this

thesis were cloned into and expressed in FUGW vectors.

32

2.2 Short hairpin RNA (shRNA)

To knockdown STIMs, STIM1 and STIM2 shRNA sequences were first cloned

into the pll3.7 lentiviral vector using the HpaI and Xho sites. The shRNA together

with U6 promoter were then transferred from pll3.7 to FUGW using the Pac1 site.

The following 19mer shRNA sequences were used: scramble (scr), 5’-

CGATACTGAACGAATCGAT-3’; STIM1, 5’-TCCAGGCAGGAAGAAGTTT-

3’; STIM2#1, 5’-ACCAAGAGCATGATCTTCA-3’; STIM2#2, 5’-

GGAACGAAAGATGATGGAT-3’. The shRNAs in the FUGW plasmids are

driven by the U6 promoter. Cells that are transfected or transduced will be

expressing eGFP or mCherrry, which are driven by an ubiquitin promoter. All

efficiency of the shRNAs were validated in neurons transduced with lentivirus

expressing the shRNAs using Western Blot.

2.3 Antibodies used and their dilutions

Ab species/type clone application/dil. source -Stim2 rabbit, pAb NA WB 1:500 Cell Signaling, 4917S

-Stim1 rabbit, pAb NA WB 1:500 ProSci, 4119

-GluA1-NT mouse, mAb RH95 WB 1:1000, IP

1:10, ICC 1:500

Millipore, MAB2263

-pGluA1 Ser845 mouse, mAb EPR 2148 WB 1:1000 Millipore, 04-1073

-pGluA1 Ser831 mouse, mAb N453 WB 1:1000 Millipore, 04-823

-GluA2 rabbit, pAb WB 1:1000 Synaptic Systems

182 103

-GluN1 mouse, mAb M68 WB 1:1000 Synaptic Systems

114 011

-CaMKII rabbit, pAb NA WB 1:1000 Cell Signaling, 3362

-syntaxin mouse, mAb HPC1 WB 1:20000 Sigma, S0664

-VAMP2 mouse, mAb SP10 WB 1:1000, ICC Covance, MMS-616R

33

1:1000

-Homer rabbit, pAb NA ICC 1:1000 Synaptic Systems

160003

-PSD95 mouse, mAb 70-18 WB 1:1000 Millipore, MAB1596

-PKA-R1-/ rabbit, pAb NA WB 1:1000 Cell Signaling, 3927

-PKA-C- rabbit, mAb EP2102Y WB 1:1000 Abgent, AJ1612a

-PKA-C- mouse, mAb 5B ICC 1:500 BD Transduction

Laboratories, 610980

-AKAP150 goat, pAb NA IP 1:10 Santa Cruz, sc-6445

-AKAP150 rabbit, pAb NA WB 1:1000 Millipore, 07-210

-calnexin rabbit, pAb NA 1:2000 Abcam, 22595

-actin mouse, mAb AC15 WB 1:20000 Sigma, A1978

-GAPDH mouse, mAb 6C5 WB 1:20000 Millipore, MAB374

R mouse, mAb NA WB 1:1000 Millipore, 05-432

R rabbit, pAb NA WB 1:1000, IP

1:10

Millipore, 07-632

-rIgG HRP goat NA WB 1:10000 Jackson

-mIgG HRP goat NA WB 1:10000 Jackson

Alexa Fluor 488

Anti-mIgG

goat NA ICC 1:1000 Invitrogen , A110011

Alexa Fluor 568

Anti-mIgG

goat NA ICC 1:1000 Invitrogen, A11004

Alexa Fluor 488

Anti-rIgG


Alexa Fluor 568

Anti-mIgG


Alexa Fluor 633

Anti-mIgG


2.4 Media and Buffers

Name Components Purpose

CMF-HBSS Calcium-, magnesium-, and

bicarbonate-free Hank’s balanced salt

solution (HBSS) buffered with 10 mM

HEPES, pH 7.3

For dissection and washing of

primary hippocampal neurons

Glial Media Minimal essential medium (MEM)

supplemented with glucose (0.6%

wt/vol), 100 U/ml penicillin, 100mg/ml

streptomycin, and containing 10%

(vol/vol) horse serum

For culturing and maintaining

glial cells

Neuronal Plating

Medium

MEM supplemented with glucose

(0.6% wt/vol) and containing 10%

(vol/vol) horse serum

For seeding primary

hippocampal neurons

Neurobasal B27

media

Neurobasal Medium with 2mM

GlutaMAX-I and B27 supplement


primary hippocampal neurons

Table 1: The antibodies used in this project, their manufacturers, the dilutions and usage.

34

Slice Dissection

Media

MEM with 2mM GlutaMax-1, 100

U/ml penicillin and 100 mg/ml

streptomycin, and 1X HBSS buffered

with 10mM Tris and 10mM HEPES.

For dissection of organotypic

slice culture

Slice Maintenance

Media

50% MEM, 25% HBSS, 25% Horse

Serum, 2mM GlutaMax-1, 5µg/ml

Insulin, 5µg/ml Transferrin, 5ng/ml

Sodium Selenite, 2.64mg/ml Glucose,

0.8µg/ml Vitamin C and 100 U/ml

penicillin and 100 mg/ml streptomycin


organotypic slices

293T Media DMEM containing 10% fetal calf

serum (vol/vol), 100U/ml penicillin,

100µg/ml streptomycin and 20mM

glutamine


293T media during lentivirus

production

IP Buffer 25mM Tris-HCl (pH=8), 27.5mM

NaCl, 20mM KCl, 25mM Sucrose,

10mM EDTA (pH=8), 10mM EGTA

(pH=8), 1mM DTT, 10% (vol/vol)

glycerol, 0.5% NP-40, complete

protease inhibitors (Roche), and

phosphatases inhibitors (Roche)

Lysis buffer to lyse cells for

immunoprecipitation

Scale A2 4M Urea, 10% Glycerol and 0.1% TX-

100

To clear organotypic slices

after fixation

Electrophysiology

Internal solution

120mM K-Gluconate, 9mM KCl,

10mM KOH, 3.48mM MgCl2, 4mM

NaCl, 10mM HEPES, 4mM Na2ATP,

0.4mM Na3GTP, 19.5mM Sucrose,

5mM EGTA.

Electrophysiology

External Solution

110mM NaCl, 5mM KCl, 2mM CaCl2,

0.8mM MgCl2,10mM HEPES pH 7.4

and 10mM D-glucose

Tyrode’s Imaging

Buffer

119mM NaCl, 2.5mM KCl, 2.0mM

CaCl, 2.0mM MgCl2, 25mM HEPES

and 30mM glucose

Buffer to maintain the pH and

to keep cells alive during live

cell imaging experiments

RIPA buffer 50mM Tris-HCl (pH=8), 150mM

NaCl, 1mM EDTA, 1% NP-40,

complete protease inhibitors (Roche),

and phosphatases inhibitors (Roche)

Lysis buffer to lyse cells for

Western Blot

Sodium dodecyl

sulfate

polyacrylamide gel

electrophoresis (SDS-

PAGE) sample buffer

62.5mM Tris–HCl (pH 6.8), 2% SDS,

10% glycerol and 0.01 % bromophenol

blue

Protein denaturing and elution

buffer

Running buffer 250mM Tris, 1.92M Glycine and 1%

SDS

For running SDS-PAGE

Transfer buffer 25mM Tris-HCl (pH 8.0), 192mM

glycine, 5% methanol and 0.1% SDS

For Western transfer buffer

Tris-Buffered Saline

containing Tween 20

(TBST)

Tris-HCl (pH7.6), 137mM NaCl and

0.05% Tween 20

For Western membrane

washing

Phosphate Buffered

Saline (PBS)

10mM Phosphate buffer (pH 7.4) and

137mM NaCl

For washing cells during

immunocytochemistry

Table 2: The buffers used in this project, their components and usage.

35

2.5 Tissue Culture

2.5.1 Rat primary hippocampal and cortical neuronal culture

Rat hippocampal and cortex neurons were prepared from embryonic (E18 and

E16 respectively) Sprague Dawley rat embryos (InVivos, Singapore) as described

in the Banker protocol (Kaech and Banker, 2006). The hippocampus or cortex

were dissected, digested in 0.25% Trypsin (Invitrogen), triturated to isolate cells

from the tissue, and plated on poly-L-lysine (Sigma) coated 6-well plates or

coverslips. For the glial feeder layer, cortex from post-natal (P0) pups were

dissected and digested in 0.25% Trypsin (Invitrogen) with 0.1% DNAse (Roche),

triturated and filtered through a cell strainer to isolate cells from the cell debris

and then plated on T75 flask with glial media (Gibco). For imaging and

electrophysiology experiments, hippocampal neurons were grown on glass

coverslips with Neurobasal B27 media (Gibco) containing 2mM GlutaMax-1

(Gibco) and 2µg/ml FDU (Sigma) on top of a glial feeder layer, strictly adhering

to the Banker protocol. For biochemical experiments, hippocampal or cortical

neurons were grown at high density on poly-L-Lysine coated dishes and

maintained in Neurobasal B27 with 2mM GlutaMax. All animal experiments in

this study were conducted under the regulations of the Institutional Animal Care

and Use Committee (IACUC) in Singapore.

36

2.5.2 Organotypic slice culture

Hippocampal organotypic slices were prepared from post-natal (P5–6) Sprague

Dawley rats according to the following protocol (Gogolla et al., 2006). The

hippocampus were isolated from the P0 pups and hippocampal slices (350 µm)

were prepared using a tissue chopper and transferred onto cell culture inserts

(Millicell, 0.4µM pore size, Millipore). The hippocampal slices were maintained

in slice maintenance media (50% MEM, 25% HBSS, 25% Horse Serum, 1X

GlutaMax, 5µg/ml Insulin, 5µg/ml Transferrin, 5ng/ml Sodium Selenite,

2.64mg/ml Glucose, 0.8µg/ml Vitamin C and 1X Penicillin and Streptomycin).

Slice media was changed every 2 days.

2.6 Lentivirus production

FUGW lentiviral particles were produced and purified according to (Lois et al.,

2002; Tiscornia et al., 2006). Human embryonic kidney cells (HEK293T) were

first grown to 70 – 80% confluency in a 10 cm tissue culture dish with 293T

media supplemented with Geneticin (G418). The cells were then transfected with

the 10µg of FUGW plasmid containing the gene or shRNA of interest (transfer

plasmid), 10µg of Delta8.9 (packaging plasmid) and 10µg of VSVG (envelop

plasmid) and cultured in 293T media (DMEM containing 10% Fetal Calf Serum

(Gibco), 100U/ml penicillin, 100µg/ml streptomycin and 20mM glutamine

(Gibco)) for 48 hours. The viruses were isolated from the cell media by

centrifugation (500g for 10 min) and filtration through a 0.45µm filter (Sartorius).

37

The lentivirus titer was measured by infecting 1 x 105 neurons using different

serial dilutions of the same virus. The lentivirus titer can be calculated by the

following equation:

𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑖𝑛𝑓𝑒𝑐𝑡𝑖𝑛𝑔 𝑢𝑛𝑖𝑡𝑠/µ𝑙 =% 𝑜𝑓 𝑖𝑛𝑓𝑒𝑐𝑡𝑒𝑑 𝑐𝑒𝑙𝑙 𝑋 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓𝑐𝑒𝑙𝑙𝑠 𝑠𝑒𝑒𝑑𝑒𝑑 𝑋 𝑠𝑒𝑟𝑖𝑎𝑙 𝑑𝑖𝑙𝑢𝑡𝑖𝑜𝑛𝑠

𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑣𝑖𝑟𝑢𝑠 𝑎𝑑𝑑𝑒𝑑

The use of the recombinant lentivirus has been approved by Genetic Modification

Advisory Committee and Ministry of Health Biosafety Branch in Singapore.

2.7 Transfection

2.7.1 Lipofectamine

For dendritic spine analysis, neurons seeded on 18mm diameter coverslips were

transfected by Lipofectamine-2000® (Invitrogen). 2ug of DNA and 4µl of

Lipofectamine-2000 were diluted separated into 150µl of Optimem® (Gibco).

The procedure for transfection of the neurons is similar to that as mentioned in the

manufacturer’s protocol except that the neurons were only incubated with the

DNA and Lipofectamine-2000® mixture in Neuronal Plating Medium for 40 min.

The neurons were then transferred to Neurobasal B27 media (Gibco) containing

2mM GlutaMax-1 (Gibco) and 2µg/ml FDU (Sigma) on top of a glial feeder

layer.

2.7.2 Electroporation

DNA and shRNA constructs were introduced in neurons by electroporation using

the Rat Neuron Nucleofector kit II (Amaxa Biosystems, Lonza). Freshly

dissociated neurons were mixed with the appropriate amount of plasmids in the

38

electroporation solution as provided and then transferred to an electroporation

cuvette where they are electroporated (Rat hippocampal neurons G-003 program).

Neuronal plating media is then added to the neurons in the cuvette and neurons

are seeded either on coverslips or poly-L-Lysine coated dishes. The neuronal

plating media is replaced with Neurobasal B27 media (Gibco) containing 2mM

GlutaMax-1 (Gibco) and 2µg/ml FDU (Sigma) after 3 hours.

2.7.3 Transduction

All neurons were transduced on the day of seeding or after by adding virus to the

neurons at a multiplicity of infection (MOI) between 2 and 3, to get an 86-95%

efficiency of transduced cells. The media will then be replaced with Neurobasal

B27 media (Gibco) containing 2mM GlutaMax-1 (Gibco) or transferred onto glial

with Neurobasal B27 media 24 hours after incubation with the virus.

2.7.4 Biolistic transfection

To allow the plasmid DNA to attach to the gold, 3µg of plasmid DNA was mixed

with 1 µg of gold microcarrier (1.0µm diameter) and 20µl of 0.05M spermidine.

The mixture was then vortexed while 20µl of CaCl2 is added. The mixture was

allowed to precipitate before the supernatant was separated from the DNA-gold

pellet by centrifugation. To remove any remaining water from the DNA-gold

pellet, 100% ethanol was added to the gold pellet, vortexed and sonicated before

centrifugation to separate the DNA-gold pellet again. To adhere the gold particles

to the plastic tubing cartridge, the gold pellet was resuspended in 0.05g/ml of

polyvinylpyrrolidone (PVP) dissolved in ethanol. The DNA-gold was then

39

vortexed and sonicated again before allowing it to settle in the plastic tubing

cartridge. The ethanol supernatant was removed and the tubing was allowed to

dry with continuous nitrogen passing through it. The cartridges are loaded into the

Helios Gun Gene (BioRad) and the plasmid DNA enters into neurons by helium

discharge at a high velocity of 200psi.

2.8 Electrophysiology

Whole-cell patch-clamp recordings were performed in DIV15-18 hippocampal

neurons expressing the shRNAs and fluorescent protein to measure the miniature

excitatory postsynaptic currents (mEPSCs). The neurons were held at -70mV

using a multiClamp 700B amplifier (Axon Instruments, CA) driven by pClamp

(Axon Instruments). Recording pipettes, with resistances of 3-5 MΩ were filled

with an internal solution containing (in mM) 120 K-Gluconate, 9 KCl, 10 KOH,

3.48 MgCl2, 4 NaCl, 10 HEPES, 4 Na2ATP, 0.4 Na3GTP, 19.5 Sucrose, 5 EGTA.

Neurons were bathed in external solution containing (in mM) 110 NaCl, 5 KCl, 2

CaCl2, 0.8 MgCl2,10 HEPES pH 7.4 and 10 D-glucose and supplemented with 0.5

μM TTX, to prevent action potential-evoked EPSCs, and 10 μM BMI to block

GABAergic inhibitory postsynaptic potentials. Data were analyzed using pClamp.

Only recording epochs in which series and input resistances varied by <10% were

included in the analysis.

2.9 Immunocytochemistry

21 DIV primary cultures of hippocampal neurons that were grown on coverslips

were fixed in 0.1 M PBS (pH 7.4) containing 4% PFA and 4% sucrose for 20 min

40

at room temperature. The membranes were then permeabilized with PBS

containing 0.25% Triton X-100 (Sigma) for 5 min and then blocked in 5% normal

goat serum diluted in PBS for 1 hr to prevent non-specific binding. The neurons

were then incubated with the appropriate dilution of primary antibody first and

then subsequently the secondary antibody for 1 hr each at room temperature.

Double or triple stains were done using 488, 568 and 633 Alexa-conjugated

secondary Abs. The coverslips were then mounted onto a glass slide with

mounting medium (Flurosave, Merck) and allowed to dry.

2.10 Organotypic culture fixation and scaling

14 DIV organotypic hippocampal slices (350 µm) were first fixed in 0.1 M PBS

(pH 7.4) containing 4% PFA and 4% sucrose for 20 min at room temperature. The

slices were then washed with PBS for 3 times before scaling with Scale A2

solution (4M Urea, 10% Glycerol and 0.1% TX-100) (Hama et al., 2011) to clear

the slices. The slices were then washed with water to remove any traces of the

scaling solution. The slices are then dried overnight at room temperature before

they were mounted on a glass slide with mounting media.

2.11 Imaging

For fixed neuronal samples, immunofluorescence images were imaged with an

upright laser scanning confocal microscope (LSM710, Zeiss). The images were

processed with Zen 2012 (Zeiss) and ImageJ (Rasband, 1997-2014) . Super

resolution microscopy was also performed on fixed images with a 63x 1.4NA

41

objective on a structured illumination microscope (ELYRA, Zeiss). 3D stack were

maximally projected on a single plane, or subjected to 3D surface rendering using

the Imaris software.

For all live cell imaging experiments, the neurons were imaged in Tyrode’s

imaging buffer (119mM NaCl, 2.5mM KCl, 2.0mM CaCl, 2.0mM MgCl2, 25mM

HEPES and 30mM glucose) a heated stage (36.5oC) with a 40x (NA 1.3) or 63x

(NA 1.4) objective mounted on an inverted Nikon Eclipse TE2000-E microscope

equipped with a spinning-disc confocal scanhead (CSU-10, Yokogawa) and an

auto-focusing system (PFS, Nikon). During time-lapse imaging, cells were

maintained in an on-stage cell incubation chamber set at 36.5oC. Images were

acquired with a Cool SNAP HQ2 CCD camera (Photometrics) driven by

Metamorph 7.6 (Universal Imaging). For FRET imaging, FRET between eCFP

and eYFP is displayed as the intensity in the FRET channel (corrected for bleed

through) divided by the donor (eCFP) intensity (Thevathasan et al., 2013).

2.12 Preparation of protein extracts from cells or tissue

The rat cortex, hippocampus, striatum and cerebellum were lysed and

homogenized in 1ml ice-cold IP buffer For tissue culture, cells were washed

twice with ice-cold phosphate-buffered saline (PBS) and lysed with IP buffer

(25mM Tris-HCl [pH=8], 27.5mM NaCl, 20mM KCl, 25 mM Sucrose, 10 mM

EDTA [pH=8], 10 mM EGTA [pH=8], 1mM DTT, 10% (v/v) glycerol, 0.5%

NP-40, complete protease inhibitors (Roche), and phosphatases inhibitors

42

(Roche)) or RIPA buffer (50 mM Tris-HCl [pH= 8], 150 mM NaCl, 1 mM

EDTA, 1% NP-40, complete protease inhibitors (Roche), and phosphatases

inhibitors (Roche)). The cell lysate was incubated at 4°C for 10 min and then

clarified by centrifugation at 12,000 rpm for 5 min at 4°C to collect the

supernatant.

2.13 Biochemical isolation of the PSD

Post-synaptic densities (PSDs) isolation was adapted from procedures as

described in (Cotman et al., 1974; Cotman and Taylor, 1972; Hahn et al., 2009).

All steps were done on ice or at 4oC. Adult rat forebrains were first dissected and

then homogenized using a Glass/Teflon dounce homogenizer. The cell debris was

isolated from the homogenate by centrifugation at 1000g for 10 min. The

supernatant was then centrifuged at 16,000g for 20 min to yield a membrane

(Memb) and cytoplasmic (Cyto) fractions. The membrane fraction was then

transferred onto a sucrose gradient (1.2 M, 1 M, 0.85 M Sucrose, 100,000g for 2

hrs) to isolate synaptosomes (SPMs). The SPMs were treated with 1% TX-100

and then centrifuged at 35,000g for 20 min. The supernatant contains the synaptic

vesicles (SV). The pellet was further extracted in 1.5% TX-100 and spun at

140,000g for 30 min. The supernatant contains pre-synaptic membranes and was

pooled with the SV fraction (Pre/SV). The pellet is the final PSD fraction. All the

different fractions (equal amount of proteins) were then immunoblotted for

several markers and the STIM2 proteins.

43

2.14 Immunoprecipitation (IP)

Appropriate dilutions of the corresponding antibody (2mg or 4 mg) or mouse IgG

(control) was added to the total cell lysate (200 µg for immunoprecipitation or

1mg for co-immunoprecipitation) and incubated at 4°C for 1 hour before the

addition of protein A/G-sepharose (Pierce). The mixture was then incubated

overnight at 4°C before being washed three times with IP buffer. Bound proteins

were boiled and eluted at 95°C for 5 min in of SDS-PAGE sample buffer

containing 100 mM dithiothreitol (DTT) (Sigma) before being collected by

centrifugation at 12,000 rpm for 5 min at room temperature.

2.15 Western transfer and immunoblotting

Protein samples were separated in either one-dimensional 8%, 10% or 12% SDS-

PAGE according to standard protocols (Sambrook et al., 1989) and

electrophoretically transferred onto a Hybond nitrocellulose membrane (BioRad)

in transfer buffer. Membranes were subsequently blocked for 1 hr with 5% skim

milk in TBST, followed by overnight incubation at 4°C with the primary

antibodies at appropriate dilutions. Membranes were washed in TBST for 15 min

for three times and incubated with the appropriate dilution of secondary

antibodies that are conjugated with horseradish peroxidase-conjugated antibodies

(Jackson) for 1 hr and followed by detection with enhanced chemiluminescence

(ECL-PIERCE). The protein sizes were estimated using prestained SDS-PAGE

Standards Broad Range molecular weight markers (Bio-Rad, USA).

44

2.16 Measurement of cAMP levels and PKA activity

Equal amount of primary neuronal cultures were seeded onto poly lysine coated 6

well plates and grown for 21 DIV. The neurons were either treated with 50 m

Forskolin (F6886, Sigma) and 0.1 m Rolipram (R6520, Sigma), or DMSO as a

control (Sigma), for 30 minutes before harvesting. To measure cAMP levels, the

neurons were lysed with ice cold HCl and the cAMP levels were measured using

an ELISA kit (Calbiochem). To measure PKA activity, the cells were lysed

according to the manufacturer protocol and measured with an ELISA kit (ENZO

Lifescience). The amount of PKA activity was then normalized against the protein

concentration. Only samples whose levels of cAMP or PKA activity that falls

within the linear range of the standard curves are accepted.

2.17 GluA1 endocytosis

GluA1 antibodies (1:100) were incubated with 21 DIV neurons in Tyrode’s

imaging buffer at 4oC for 30 minutes to allow the antibody to bind to the surface

GluA1 on the neuron. 50 m Forskolin and 0.1 m Rolipram were added and the

neurons are brought to 37oC for 30 minutes to allow for endocytosis of the GluA1

that are coated with the antibody. Any remaining surface GluA1 antibodies were

stripped off by washing the neurons with an ice cold antibody stripping buffer

(0.5M NaCl and 0.2M Acetic Acid) for 2 minutes. The cells were then fixed,

permeabilized and internalized GluA1 antibodies were stained with anti-mouse

Alexa 488.

45

2.18 Image analysis

2.18.1 Spine detection and classification

z-stacks of dendritic segments (0.64µm/slice at 63X magnification) imaged by

confocal microscopy that covered the entire dendritic segment were used for

analysis. Images were processed using ImageJ and Metamoprh (Universal

Imaging) to get the maximum intensity projection of the dendritic segments.

Dendritic spines analysis were performed using NeuronStudio (Rodriguez et al.,

2008) and dendritic spines are classified and expressed as number of spines per

unit length.

2.18.2 Scoring synaptic density

A Matlab script was written to analyze three-color images consisting of mCherry

(shRNA), VAMP2 (633 nm, presynaptic marker) and Homer-1 (488 nm, post-

synaptic). Synapses overlapping mCherry-expressing dendrites that are positive

for both VAMP2 and Homer-1 are scored by the script. The same intensity

thresholds were used to segment VAMP2 and Homer-1 puncta in control and

STIM2 silenced neurons. Synaptic density was obtained by computing the

number of detected synapses per unit area of mCherry-expressing dendrites.

2.18.3 Sholl analysis

Apical dendrites from CA1 pyramidal neurons in organotypic slice cultures were

used for Sholl analysis. Neurite traces were traced out automatically using

46

NeuroStudio and the dendrite tracings were mapped onto the neuron. Sholl

analysis was done using Simple Neurite Tracer Plugin in Fiji (Longair et al.,

2011) with the start of the analysis originating from the cell body.

2.18.4 SEP-GluA1 insertion

SEP-GluA1 insertion was quantified by measuring the fold-change in SEP-GluA1

intensity in dendritic segments.

2.18.5 Detection of endocytic GluA1 puncta

A MATLAB script was written to detect and score GluA1 endocytic puncta.

GluA1 puncta were segmented based on intensity and inclusion in mCherry-

expressing dendritic segments. The same intensity threshold was used for both

control and STIM2 silenced cells. Puncta density was obtained by computing the

number of detected puncta per unit area of mCherry-expressing dendrites.

2.18.6 Granularity

Redistribution of YFP-STIM2 to puncta after forsk/rolipr treatment was

quantified using a custom-written MATLAB script that measures the texture (i.e.

granularity) of an image. The script is based on the build-in function rangefilt.m.

2.19 Statistics

All average data are represented as means +/- SEM, unless indicated otherwise.

All statistical significance was determined using two-tailed unpaired or paired t-

tests on paired data sets obtained from cell populations or individual cells

47

respectively. One way ANOVA was used when simultaneously comparing three

or more data sets. In this case, p values were derived from a post-hoc Bonferroni

test. ANOVA and post-hoc Bonferroni tests were done using the ANOVA1 and

multcompare functions in MATLAB.

48

RESULTS

This project is collaboration between several lab members and me. I was the one

that started the concept and the hypothesis behind this project. All the imaging

experiments and results in this thesis were performed and generated by me unless

otherwise stated. All Western Blot data were generated by Dr. Gisela Garcia-

Alvarez unless otherwise stated while the electrophysiology data were generated

by Ji Fang. Permission has been granted by the lab members to reproduce the

figures in this thesis.

3.1 Expression of STIM isoforms differs in different parts of the brain and is

developmentally regulated

Both STIM isoforms are expressed in the brain, with different transcript levels in

different parts of the brain (Dziadek and Johnstone, 2007; Klejman et al., 2009;

Skibinska-Kijek et al., 2009a). The cortex, hippocampus and cerebellum were

isolated from the adult rat brain (more than 8wks old) to check for the protein

levels of both STIM isoforms. STIM2 was particularly high in the rat

hippocampus (Figure 7A). STIM1 was present in all three parts of the brain, with

higher abundance in the cerebellum (Figure 7B), similar to previous reports

(Dziadek and Johnstone, 2007; Klejman et al., 2009; Skibinska-Kijek et al.,

2009a). It is interesting to note that there is a STIM1 upper band (more than

100kDa) in the cortex and the hippocampus but not the cerebellum. One possible

explanation for the upper band could be the dimeric for of STIM1.However, this

remains to be tested. Since both STIM isoforms are expressed in the

49

hippocampus, the hippocampus and pyramidal neurons were chosen to be the

model for most of the experiments.

Both the expression of STIM1 and STIM2 in primary hippocampal neurons were

also developmentally regulated, with STIM2 (100kDa) levels reaching a plateau

peaking at 14 days in vitro (DIVs) (Figure 7C), which is around the time of

synaptogenesis, while STIM1 (85kDa) levels plateaued at 21 DIVs (Figure 7D),

similar to previous report (Keil et al., 2010). This suggests that the STIM isoforms

may serve different functions in different parts of the brain and the expression of

STIMs may be dependent on the development of the neurons or brain.

50

3.2 STIM2 localizes to large dendritic spines and is enriched in the post-

synaptic density

STIMs are ER proteins that translocate into puncta upon sensing ER Ca2+

depletion (Brandman et al., 2007) and STIM1 has been shown to localize in

puncta in both the axon and dendrites, (Dziadek and Johnstone, 2007; Keil et al.,

2010) and also growth cones (Gasperini et al., 2009) in neurons. However, very

little is known about how STIM2 is localized in neurons. As there is no good

antibody for examining the localization of endogenous STIM in neurons, YFP-

STIM1 and YFP-STIM2 were over-expressed in primary hippocampal neurons to

examine the localization of STIM1 and STIM2 in neurons.

First, confocal images of young primary hippocampal neurons (3-5DIV) showed

that both YFP-STIM1 and YFP-STIM2 are localized in puncta in the cell body

and axon of neurons, although STIM2 appeared to be more punctate in the cell

body (arrows in Figure 8A).

Second, super resolution microscopy in matured hippocampal neurons (21DIV)

using structured illumination microscopy (SIM) revealed that YFP-STIM1 was

Figure 7. STIMs expression differs in different parts of the brain and is

developmentally regulated. (A) Immunoblots of STIM2 from adult rat cortex (more

than 8wks old) (ctx), hippocampus (hipp) and cerebellum (cereb). (B) Immunoblots

of STIM1 from adult rat ctx, hipp and cereb. (C) Developmental expression pattern

of STIM2 in dissociated hippocampal neurons. (DIV, days in vitro). (D)

Developmental expression pattern of STIM1 in dissociated hippocampal neurons.

Data generated by and reproduced with permission of Gisela Garcia-Alvarez.

51

homogenously distributed in the dendrites while STIM2 was localized to puncta

in the dendrites and occasionally in dendritic spines (Figure 8B), suggesting that

the expression of STIM1 and STIM2 are different in hippocampal neurons. An

enlargement of another dendritic segment also showed that YFP-STIM2 can be

found inside dendritic spines (arrows in Figure 8C) and dendrites as puncta.

To confirm that STIM2 is found inside dendritic spines, neurons were transfected

with YFP-STIM2 and stained for Homer-1, a postsynaptic marker. Using confocal

microscopy, STIM2 is also found to co-localize with Homer-1 in dendritic spines

(arrows in Figure 8D). A quantification of dendritic spines from confocal images

showed that about 45% of dendritic spines have YFP-STIM2 puncta in them

(Figure 8E) and dendritic spines with YFP-STIM2 tend to have a larger spine area

(Figure 8F). It is possible that there is a gain of function in dendritic spine area

when STIM2 is overexpressed but larger dendritic spines have been observed to

have the ER in them (Nimchinsky et al., 2002), which might explain the presence

of STIM2 in larger dendritic spines.

The tip of a dendritic spine is known as the post-synaptic density (PSD), a

protein-dense compartment containing receptors for neurotransmitters (Kim and

Sheng, 2009; Okabe, 2007). To determine if STIM2 associates with the PSD,

synaptosomes containing the presynaptic membranes/synaptic vesicles (Pre/SV)

and post-synaptic membranes were first isolated from the adult rat brain. The PSD

was then separated from Pre/SV using fractionation on a sucrose gradient and

52

detergent extraction. This procedure led to a clean separation of pre-synaptic

markers (Syntaxin and Vamp2) and post-synaptic markers (PSD95, CamKII and

GluA1) (Figure 8G). STIM2 was clearly enriched in the PSD fraction but not in

the Pre/SV fraction. In contrast, STIM1 was enriched in the Pre/SV fraction

(Figure 8G). STIM2 has also been observed to be in dendritic spines in a recent

study (Sun et al., 2014), which is consistent with the results obtained. Dendritic

spines with a larger spine area have been correlated with a larger synaptic strength

(Matsuzaki et al., 2004a) and the data suggest that STIM2 but not STIM1, may be

important in dendritic spine morphogenesis.

53

54

3.3 STIM2 regulates dendritic spine morphogenesis

Based on results in the previous section, STIM2 localizes to puncta that are found

in dendrites and large dendritic spines, suggesting that STIM2 may play a role

spine morphogenesis. First, the impact of STIMs on dendritic spine

morphogenesis was examined by silencing STIM1 or STIM2 in primary

hippocampal neurons using validated shRNA sequences (Wong et al., 2010)

(Figure 9A and Figure 11B). To ensure that any phenotype observed can be

attributed to the effects of the shRNA post-synaptically, the primary neurons were

transfected sparsely (using Lipofectamine-2000®) so that the majority of

presynaptic inputs received by a transfected cell originated from neurons

expressing normal levels of STIM2. STIM2 knockdown by two different shRNA

sequences led to a marked decrease in both dendritic spine density (Figure 9B and

C) and dendritic spine area (Figure 9D). Both the decrease in spine density and

Figure 8. STIM2 is localized to dendritic spines. (A) Confocal microscopy images

of young primary hippocampal neurons (3-5 DIV) transfected with either YFP-

STIM1 or YFP-STIM2 (arrows showing YFP-STIM1 or YFP-STIM2 puncta). (B)

Super resolution microscopy (SIM) of primary hippocampal neurons (DIV 21) co-

transfected with mCherry (red) and either YFP-STIM2 or YFP-STIM1 (green). (C)

Enlargement of another dendritic segment transfected with YFP-STIM2 (green) and

mCherry (magenta). Scale bar: 5 μm. Arrows point to YFP-STIM2 puncta inside

spine heads. Note that spine necks connecting spine heads to the dendritic shaft are

not always visible on these high resolution images. (D) Mature primary hippocampal

neuron transfected with YFP-STIM2 and stained for Homer. Arrows point to YFP-

STIM2 puncta inside dendritic spine heads labelled with Homer. (E) Percentage of

dendritic spines with STIM2 puncta. n = 711 spines from 2 independent experiments.

(F) Cumulative distribution of spine size (area µm2) with (n = 282 spines) or without

(n = 415 spines) YFP-STIM2. (G) Synaptosomes fractionation and immunoblot

analysis of adult rat brains. WBL, whole-brain lysate; Cyto, cytosol; Memb, total

membranes; SPM, synaptosomes; Pre/SV, presynaptic membranes and synaptic

vesicles; PSD, post-synaptic density. Equal amounts of protein were loaded in each

lane. Data in Figure 8A generated and reproduced with permission by Lu Bo. Data in

Figure 8G generated by and reproduced with permission of Lynette Lim.

55

area could be partially rescued by introducing a shRNA-resistant form of STIM2

(human YFP-STIM2) (Figure 9B, C and D). Dendritic spines can be classified

into mushroom, stubby and thin depending on their morphology. Spine shape also

correlates with spine maturity; spines with large spine heads (i.e. mushroom

spines) are associated with increased synaptic strength (Nimchinsky et al., 2002).

The dendritic spine area for STIM2 knockdown was significantly smaller (Figure

9D) but there was no significant difference in the distribution of spine type

(Figure 9C). In contrast, STIM1 silencing had no detectable effect on both spine

density and shape (Figure 9B and C). The results are consistent with the effect of

STIM2 on dendritic spines size and maturity published recently (Sun et al., 2014)

but more data is required to determine if STIM2 is also important for spine type

morphology.

To determine whether STIM2 is also able to regulate spinogenesis in hippocampal

circuits, pyramidal neurons in the CA1 region of organotypic hippocampal slices

were transfected with STIM2 and control shRNAs using biolistic transfection

(Figure 9E). Biolistic transfection was chosen because of the low transfection

efficiency and pyramidal neurons in the CA1 were chosen because the CA3-CA1

synapse has been well studied and that LTP at that synapse is associated with

spine growth (Segal, 2010). The low transfection efficiency also ensures that there

is very little pre-synaptic contribution from the CA3 to any spine phenotype in the

CA1 neurons. Expression of two independent STIM2 shRNAs in mature CA1

pyramidal neurons (DIV 14-16) resulted in a marked reduction in spine density in

56

apical dendrites (Figure 9F and G) and the decrease in spine density could be

partially rescued by expression of YFP-STIM2. Reduced levels of STIM2 had no

obvious impact on spine type (Figure 9G) but apical dendrite complexity was

slightly reduced when STIM2 was silenced (Figure 9H). Together, these results

indicate that STIM2 but not STIM1, is required for the formation and/or

maintenance of dendritic spines and may also regulate dendritic branching.

57

58

3.4 STIM2 shapes spontaneous synaptic transmission

Dendritic spines are the postsynaptic compartment of an excitatory synapse and

are important for synaptic transmission (Yuste, 2011). Excitatory synaptic

strength is correlated with spine morphology (Matsuzaki et al., 2004b) and

changes in dendritic spines (density or size) have been shown to affect synaptic

transmission (Chen et al., 2014). To address the impact of STIM2 knockdown on

synaptic transmission, miniature excitatory post synaptic currents (mEPSCs) were

recorded from STIM2 knockdown neurons using whole-cell patch clamp

Figure 9. STIM2 is required for dendritic spines formation/maturation and

dendrite branching. (A) Western blot analysis showing the efficiency of STIM1 and

STIM2 shRNA in primary hippocampal neurons (21DIV). (B) Maximal projection

confocal images of hippocampal neurons (DIV 21) co-expressing mCherry and the

indicated shRNAs. For rescue experiments, STIM2 shRNA#1 was co-expressed with

YFPSTIM2. (C) Quantification of spine density and spine type for conditions shown

in (B) using NeuronStudio software. At least 50 dendritic segments comprising > 850

spines from three independent experiments were scored for each condition. (D)

Quantification of dendritic spine area for the conditions shown using Metamorph

software. At least 900 dendritic spines from three independent experiments were

scored for each condition. (E) Low magnification (10X) confocal images of CA1

pyramidal neurons in organotypic slices (DIV 14-16) biolistically transfected with

scr. or STIM2 shRNAs vectors co-expressing GFP (or mCherry, not shown). Primary

and secondary apical dendrites (white box) were selected for high-resolution

dendritic spine analysis. Nuclei are stained with Hoechst (blue). Scale bar: 50 μm. (F)

CA1 neurons were biolistically transfected with the indicated shRNAs, or the STIM2

shRNA#1 together with YFP-STIM2 for rescue experiments. Spines were imaged in

distal apical primary and secondary dendrites. (G) Quantification of spine size and

type. At least 45 dendritic segments comprising > 1000 spines in three independent

experiments were analyzed for each condition. STIM2 silencing decreased spine

density in both dissociated cultures and slices. ***, p < 0.001, ANOVA. The

percentage of thin, stubby and mushroom spines was not significantly affected by

STIM2 shRNA#1 or shRNA#2, p > 0.05, ANOVA. Scale bar: 5 μm. (H) Sholl

analysis of the apical dendrites of CA1 pyramidal neurons expressing scr. shRNA (n

= 30), STIM2 shRNA#1 (n = 22) or STIM2 shRNA#1 with YFP-STIM2 (n = 20).

The Y axis corresponds to the number of apical dendrite intersections with concentric

circles centred at the centroid of the cell body. STIM2 silenced cells show a

significant decrease in the number of intersections between 50 and 100 μm from the

soma. *, p < 0.05, unpaired t-test. Data in Figure 9A generated by and reproduced

with permission of Lynette Lim.

59

recordings (voltage clamp). There was a clear reduction in both the frequency

and amplitude of mEPSCs in STIM2 silenced hippocampal neurons. These

defects were again rescued by the introduction of YFP-STIM2 (Figure 10A-D).

A reduction in mEPSCs frequency can be attributed to either a reduction in

synaptic vesicles release pre-synaptically or a reduction in synapse numbers (Del

Castillo J, 1954; Redman, 1990; Stevens, 1993). The reduction in mEPSCs is

unlikely due to a presynaptic defect because the mEPSCs were measured in

sparsely-transfected neuron cultures with presynaptic inputs originating primarily

from non-transfected cells, which is similar to the experimental design in Figure

9B and F. Therefore, the reduction in mEPSC frequency is most likely to result

from a reduction of synapses, consistent with the observed loss of dendritic

spines. To test whether the mEPSC frequency reduction is due to a reduction in

the number of synapses, neurons were stained for pre- (VAMP2) and post-

synaptic (Homer-1) markers. A synapse is scored by a puncta that is co-stained by

both VAMP2 and Homer-1. Fewer synapses were observed in STIM2 silenced

neurons than control cells (Figure 10E, F), which could explain the observed

reduction in mEPSCs frequency.

In contrast, a reduction in mEPSC amplitude is most likely to originate post-

synaptically through defects in glutamate receptors localization at the synapse

and/or in their channel properties (Redman, 1990). During whole-cell patch clamp

recordings, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors

60

(AMPARs) are the main glutamate receptors that cause the neuron to depolarize

as the resting potential is held at -70mV and thus a defect in AMPAR function

and/or localization is likely the cause of the reduction in mEPSC amplitude.

61

3.5 STIM2 reciprocally regulates phosphorylation of GluA1 on Ser831 and

Ser845

The expression of GluA1 on the PM during LTP is dependent on the

phosphorylation of GluA1. Two widely studied phosphorylation sites on GluA1

are the Ser831 and Ser845. Phosphorylation of Ser831 by CaMKII and PKC

regulates single-channel conductance of the AMPAR (Derkach et al., 1999) and

also delivery of the AMPAR to the synapse (Hayashi et al., 2000).

Phosphorylation of Ser845 by cAMP-dependent PKA increases channel open

probability (Banke et al., 2000) and promotes surface expression of AMPARs

(Esteban et al., 2003; Oh et al., 2006). Both pSer845 and pSer831 residues on

GluA1 have been implicated in long-term potentiation (LTP) and long-term

depression (LTD) in the hippocampus, two forms of synaptic plasticity which

have been associated with learning and memory (Esteban et al., 2003; Lee et al.,

2010a; Lee et al., 2003; Makino et al., 2011).

Figure 10. STIM2 shapes functional synaptic inputs. (A-D) whole-cell patch

clamp recordings of hippocampal neurons (DIV 15-18) sparsely transfected with the

indicated shRNA constructs. (A) Individual mEPSC traces of cells co-expressing the

indicated shRNAs. STIM2 shRNA#1 was co-expressed with YFP-STIM2 for rescue

experiments. (B-C) Quantification of mEPSCs frequency and amplitude (Scr, 23.2±2

pA, 6.7 ±1.0 Hz, n=11; STIM2 shRNA#1, 15.0±0.6 pA, 2.9±0.3 Hz, n = 9; STIM2

shRNA#1 + YFP-STIM2, 24.7±1.6pA, 6.4±0.5 Hz, n = 11; ** p < 0.01, ANOVA).

(D) Cumulative distribution of mEPSCs amplitude. (E) Synaptic density measured in

control and STIM2 silenced hippocampal neurons (DIV 21) by co-staining with a

pre- (VAMP2) and post-synaptic (Homer1) markers. Scale bar: 5 μm. Overlapping

VAMP2/Homer1 puncta are detected and scored (F) by a MATLAB script (see

methods). Synaptic density per 10μm: scr. shRNA, 6.1±0.27, n = 15; STIM2 shRNA

#1, 2.4±0.3, n = 17, p < 0.01, t-test. Data generated by and reproduced with

permission of Ji Fang.

62

To check the phosphorylation state of the AMPARs in neurons with STIM2

silenced, primary hippocampal neurons were transduced with shRNA-expressing

lentiviruses to knockdown STIM2. The Ser845 phosphorylation of GluA1 at

steady state was not decreased at 14DIV old neurons but decreased dramatically

in 21DIV primary hippocampal neurons (Figure 11A), an age where dendritic

spines are already present. Together with results from 7C, where STIM2 levels

peaks at 14DIV, it may be possible that STIM2 only plays a role in maintaining

GluA1 S845 phosphorylation after a synapse has been formed or it may also be

possible that STIM2 has a different role in GluA1 S845 phosphorylation at

different developmental stages. However, another plausible explanation is there is

less STIM2 knock out at 14DIV as compared to 21DIV and thus a weaker

phenotype.

The experiment was repeated with two different STIM2 shRNAs in 21 DIV old

neurons (Figure 11B). Consistent with Figure 11A, S845 phosphorylation levels

were dramatically reduced with two different STIM2 shRNA (Figure 11B).

However, STIM2 silencing resulted in an increase in Ser831 phosphorylation with

no change in total GluA1 levels (Figure 11B). Both the Ser845 and Ser831

phosphorylation phenotypes were rescued by co-expression of YFP-STIM2 at a

level comparable to endogenous STIM2 (Figure 11B).

Ser845 phosphorylation is mediated by PKA, which is important for synaptic

plasticity (Waltereit and Weller, 2003), and PKA activity is dependent on cyclic

63

adenosine monophosphate (cAMP)-mediated dissociation of the regulatory

subunit (rPKA) from the catalytic subunit (cPKA) (Skroblin et al., 2010). cAMP

levels can be artificially raised to acutely activate PKA by treating neurons with

forskolin (an adenylate cyclase agonist), and rolipram (a phosphodiesterase

inhibitor), to inhibit the breakdown of cAMP. This is one of the way to trigger a

process known as chemical LTP (cLTP) (Molnár, 2011), which has been reported

to trigger AMPAR insertion, spine enlargement and LTP in slices (Kopec et al.,

2006; Otmakhov et al., 2004b). Under acute PKA activation using

forskolin/rolipram cLTP, GluA1 Ser845 phosphorylation was induced robustly in

control cells but was strongly inhibited (up to 80%) by two shRNAs against

STIM2 (Figure 11B and C). The defect in phosphorylation can be rescued again

by co-transduction with YFP-STIM2 (Figure 11B, C and D). One possible

explanation for the decrease in the mEPSC amplitude may be the loss of GluA1 at

the synapse and/or loss of channel properties as a result of decrease in GluA1

Ser845 phosphorylation when STIM2 is knocked down. Taken together, these

data shown that STIM2 is essential for GluA1 Ser845 phosphorylation and

suggests that STIM2 functions downstream of adenylate cyclase.

64

3.6 STIM2 does not affect adenylate cyclase and PKA activity

STIM1 and SOCE have been shown to regulate cAMP levels through activation

of adenylate cyclase in non-excitable cells (Lefkimmiatis et al., 2009). To

Figure 11. Reciprocal regulation of GluA1 Ser831 and Ser845 phosphorylation

by STIM2. (A) Immunoblot analysis of GluA1 S845 in hippocampal 14 DIV or 21

DIV) transduced with either scrambled shRNA (scr) or STIM2 shRNA #1. (B-D)

Immunoblot analysis of hippocampal neurons (DIV 21) transduced with the indicated

shRNAs and YFP-STIM2 for rescue experiments. (A) Decreased pSer845 and

increased pSer831 in neurons expressing STIM2 shRNA#1 and #2. Both

phosphorylation phenotypes are rescued by co-expression of YFP-STIM2. (B)

Immunoblot analysis of GluA1 pSer845 in cells treated with DMSO (vehicle) or 50

μm forskolin / 0.1 μm rolipram for 30 min (same blot exposure for all conditions).

(C) Quantification of GluA1 phospho-Ser845 signals by densitometry after

forsk/rolipr treatment (n = 4). Figure 11B-C were repeated and reproduced by Gisela

Garcia-Alvarez.

65

examine if STIM2 can affect GluA1 Ser845 through adenylate cyclase, bulk

cAMP levels and PKA activity were measured in hippocampal neurons by an

enzyme-linked immunosorbent assay (ELISA) assay. There were no significant

differences between control and STIM2 silenced neurons in resting cells, or after

cLTP treatment (Figure 12A). Similar to cAMP levels, there was also no

significant difference in PKA activity between control and STIM2 silenced

neurons in neurons at steady state, or after cLTP treatment (Figure 12B).

Together, these results suggest that STIM2 promotes GluA1 Ser845

phosphorylation downstream of adenylate cyclase and PKA activation.

To measure PKA activity and confirm the PKA ELISA data, an optimized

fluorescence resonance energy transfer (FRET)-based sensor of PKA activity,

AKAR3EV (Komatsu et al., 2011), was introduced in hippocampal neurons,

together with scramble or STIM2 shRNA expressing vectors (Figure 12C). The

FRET sensor works as a distance-dependent intramolecular FRET biosensor that

consists of a cyan fluorescent protein (CFP) tagged with a peptide that can be

phosphorylated by PKA (Zhang et al., 2001), and yellow fluorescent protein

(YFP) tagged with the phosphate binding domain of FHA1 (Komatsu et al.,

2011). Upon phosphorylation by PKA, the FHA1 domain will be able to bind to

the phosphorylated peptide, thus bringing CFP close to YFP and allowing FRET

to occur. Both the control and STIM2 silenced cells have similar starting

AKAR3EV FRET signals and there was no significant difference in AKAR3EV

FRET signals between control and STIM2 silenced cells upon cLTP (Figure

66

12D). The kinetics of PKA activation upon cLTP were also unaltered by STIM2

knockdown and PKA activity even reached slightly higher levels in STIM2

silenced cells, although this difference did not reach statistical significance

(Figure 12E), which is similar to PKA activity measured using ELISA (Figure

12B) .

67

Figure 12. STIM2 knockdown does not affect cAMP production and PKA

activity. (A, B) cAMP levels and PKA activity measured in DMSO or forsk/rolipr-

treated neurons (DIV 20-21) transduced with the indicated shRNAs (n = 5 for each

condition). (C-E) AKAR3EV FRET measurements in neurons (DIV21) expressing

mCherry and the indicated shRNAs. (C) AKAR3EV FRET is displayed as the

FRET/CFP ratio and is shown in control and STIM2 silenced cells before and 3 min

after forsk/rolipr addition. (D) Pair-wise analysis of AKAR3EV FRET in individual

neurons before and 3 min after forsk/rolipr addition in control (n = 17) and STIM2

silenced cells (n = 15). Cells were analyzed from three independent experiments. ** p

< 0.01, paired t-tests. ns, non-significant ( p > 0.05). (E) fold-change in AKAR3EV

FRET induced by forks/rolipr. The shaded areas represent SEM. The average fold

increase in FRET in control and STIM2 silenced cells is not statistically different, p >

0.05, t-test.

68

3.7 STIM2 promotes exocytosis but inhibits endocytosis of AMPARs

Phosphorylation of GluA1 on Ser845 can promote activity-dependent surface

expression of AMPARs (Esteban et al., 2003; Oh et al., 2006) through promoting

exocytosis and inhibiting endocytosis of the GluA1 (Man et al., 2007; Otmakhov

et al., 2004a). To determine if STIM2 is involved in the exocytosis of GluA1, a

variant of GluA1 tagged with a pH-sensitive form of eGFP (Super Ecliptic

pHluorin, SEP) (Miesenbock et al., 1998) at the N terminus was used to examine

GluA1 exocytosis. The probe allows monitoring of AMPAR surface delivery, by

imaging the increase of SEP-GluA1 fluorescence intensity upon exposure of SEP

to the neutral pH of the extracellular environment (Kopec et al., 2006).

To look at activity-dependent GluA1 exocytosis or insertion to the plasma

membrane, cLTP using forskolin/rolipram was used to trigger GluA1 exocytosis.

First, a clear insertion of GluA1 to the dendritic plasma membrane was observed

in hippocampal neurons (DIV 18) co-electroporated with SEP-GluA1 and a

control shRNA (mCherry) using time-lapse confocal imaging during cLTP

(Figure 13A and B). The insertion of SEP-GluA1 mainly occurs in the dendritic

shaft and rarely in dendritic spines, which has been reported in earlier work

showing exocytosis of GluA1 to extrasynaptic sites following synaptic

potentiation (Makino and Malinow, 2009). Second, there was a 25% increase in

dendritic spine area after cLTP using in the same cells based on the mCherry

fluorescence (Figure 13A and C), which is consistent with cLTP inducing spine

enlargement (Kopec et al., 2006; Kopec et al., 2007). In contrast, SEP-GluA1

69

insertion in STIM2 silenced neurons was abolished with cLTP treatment (Figure

13A and B), which can be explained with the loss of GluA1 Ser845

phosphorylation when STIM2 is silenced (Figure 11C). cLTP-mediated spine

enlargement in these cells was also strongly reduced (Figure 13A and C),

suggesting STIM2 is also required for PKA-dependent structural plasticity.

To study GluA1 endocytosis, an antibody reuptake assay was used to directly

determine if STIM2 is necessary for activity dependent GluA1 endocytosis.

Endogenous surface GluA1 with the N terminus facing the extracellular space

was first bound to a GluA1 antibody, which recognizes the N terminus of GluA1,

at 4oC. The endocytosis of the antibody-receptor complex was allowed to proceed

for 37oC for 30 minutes in presence of forskolin/rolipram. Any excess antibodies

that are still bound to surface GluA1 are washed away using an acid wash.

Analysis of the endocytosed GluA1 puncta showed a ~60% increase in GluA1

uptake in STIM2 silenced cells as compared to the control cells (Figure 13D and

E). Together, these data show that STIM2 promotes activity dependent PKA-

dependent AMPAR surface expression by both enhancing exocytosis and

inhibiting endocytosis of GluA1.

70

71

3.8 STIM2 forms a complex with GluA1 and AKAP150/PKA and couples

PKA to GluA1

The data so far indicate that STIM2 is essential for PKA-dependent

phosphorylation and surface expression of GluA1. Our results further suggest that

STIM2 operates downstream of cAMP in regulating GluA1 Ser845

phosphorylation, through a mechanism that remains to be elucidated. In neurons,

PKA is brought in close proximity to GluA1 by the synaptic scaffold AKAP150

(AKAP79 in humans) (Colledge et al., 2000; Hoshi et al., 2005).

To probe the interaction of STIM2 with GluA1, GluA1 was immunoprecipitated

from adult rat brain lysates using GluA1 antibodies. Endogenous STIM2 was

efficiently pulled down by IP of GluA1 from adult rat brain lysates (Figure 14A).

As expected, the interacting partners of GluA1 such as AKAP150, and both rPKA

Figure 13. STIM2 mediates cAMP-dependent surface delivery of GluA1. (A-C)

Hippocampal neurons (DIV 17-19) co-electroporated with scramble or STIM2

shRNA#1 (mCherry) and SEP-GluA1 were imaged by dualcolor confocal

microscopy during stimulation with forsk/rolipr. (A) SEP-GluA1 and mCherry

fluorescence shown before and 30 min after forsk/rolipr treatment in control (left)

and STIM2 silenced cells (right). SEPGluA1 intensity is color-coded. The arrows

show examples of spines that have undergone forsk/rolipr-induced enlargement.

Scale bar: 5 μm. (B) Average change in SEP-GluA1 intensity in control and STIM2

silenced cells measured in 16 dendritic segments from 4 independent experiments for

each condition. Error bars, SEM. (C) Pair-wise analysis of individual spine surface

area before and after forks/rolipr in control and STIM2 silenced neurons. *** p <

0.001, ** p < 0.01, paired t-tests. (D) GluA1 endocytosis assay. Surface GluA1 in

control or STIM2 silenced neurons (DIV20) was bound to an Ab at 4oC and the Ab-

receptor complex allowed to internalize for 30 min at 37oC in presence of

forks/rolipr. Surface Abs were stripped after incubation at 4oC (control) or 37

oC.

Endocytic GluA1 puncta located in mCherry-expressing dendrites were then

segmented and scored using a MATLAB-based script. Scale bar: 10 μm. (E) Average

density of GluA1 endocytic puncta. More than 50 dendritic segments from three

independent experiments were quantified for each condition. p < 0.01, t-test.

72

and cPKA were also co-immunopurified. Likewise, YFP-STIM2, AKAP150,

rPKA and cPKA were co-purified in GluA1 IPs prepared from hippocampal

neurons transduced with YFP-STIM2 (Figure 14B). In the converse IP

experiment, YFP-STIM2 efficiently pulled down GluA1, AKAP and the two PKA

subunits (Figure 14C).

To determine whether STIM2 is required for AKAP-dependent coupling of PKA

to GluA1, GluA1 was immunoprecipated from neurons transduced with scramble

or STIM2 shRNAs. STIM2 depletion strongly reduced co-IP of AKAP150 and

rPKA, and completely disrupted binding of cPKA to GluA1 (Figure 14D).

Interaction of both PKA subunits and AKAP150 with GluA1 was partially

restored with the expression of YFP-STIM2 as a rescue (Figure 14D). The results

show that STIM2 is part of a complex that consists of the AMPAR, AKAP150

and PKA and mediates GluA1 Ser845 phosphorylation by enforcing proximity

between GluA1 and AKAP150/PKA.

In STIM2 silenced neurons, less AKAP150 was co-immunoprecipitated from

GluA1 (Figure 14D). One possible explanation for this could be mislocalization

of AKAP150 in STIM2 silenced cells. To test this hypothesis, neurons were

transfected with either the control scrambled shRNA or STIM2 shRNA and

stained for AKAP150 and GluA1. Preliminary data does not suggest any major

difference in the localization of AKAP150 in control or STIM2 silenced neurons.

AKAP150 is mostly localized to the plasma membrane in neurons transfected

73

with the scr shRNA but AKAP150 localization appears to be more punctate in the

dendritic shaft of STIM2 silenced neurons (Figure 14E). This possible

relocalization to the dendritic shaft could reflect the fact that STIM2 silenced

neurons display fewer dendritic spines. To confirm if localization of AKAP150 is

dependent on STIM2, the localization of endogenous or overexpressed AKAP150

should be studied at different neurodevelopmental stages before and after the

formation of synapses.

Figure 14. STIM2 forms a complex with AKAP150, PKA and GluA1 and is

required for GluA1 to interact with AKAP150/PKA. (A) Co-IPs from adult rat

brains (B) Co-IPs from GluA1 in primary hippocampal neurons (21DIV) transduced

with YFP-STIM2 (C) Reverse Co-IPs from GFP in primary hippocampal neurons

transduced with YFP-STIM2 adult rat brains (D) Co-IP fromGluA1 in hippocampal

neurons (21DIV) transduced with the indicated shRNAs and rescue. Fractions were

immunoblotted with the indicated Abs. The ratio indicated in the input lane reflects

the fraction of input loaded relative to the IP fraction. (E) Immunocytochemistry of

primary hippocampal neurons (21DIV) transfected with either scrambled or STIM2

shRNA and stained for AKAP150 and GluA1. Data in Figure 14A-D generated by

and reproduced with permission of Gisela Garcia-Alvarez.

74

3.9 The STIM2 SOAR domain mediates PKA-dependent phosphorylation of

GluA1 and is important for dendritic spine morphogenesis

The STIM-Orai-Activating-Region (SOAR) domain in the cytoplasmic region has

been implicated in SOCE activation (Wang et al., 2014) and Cav1.2 inhibition

(Park et al., 2010; Wang et al., 2010b). Hence, the cytoplasmic region of STIM2

is a potential candidate for interacting with GluA1 and AKAP150/PKA complex.

Two STIM2 mutants, lacking the entire cytoplasmic moiety (YFP-STIM2Δcyto),

or the SOAR domain (YFP-STIM2ΔSOAR) were created (Figure 15A) and

transfected into primary hippocampal neurons to address if the cytoplasmic region

of STIM2 is important for the interaction of GluA1 with AKAP150/PKA.

Both YFP-STIM2Δcyto and YFP-STIM2ΔSOAR retain their ER localization as

compared toYFP-STIM2 and can also be found in dendritic spines (Figure 15B).

This suggests that ER and transmembrane domains are sufficient to retain STIM2

in the ER and that STIM2 is localized to dendritic spines because of ER presence

in the dendritic spines. However, one difference is that YFP-STIM2ΔSOAR is

localized into puncta, while YFP-STIM2Δcyto is more evenly distributed (Figure

15B), which is consistent with the literature suggesting that the C terminus of

STIM2 is important for puncta formation (Baba et al., 2006; Deng et al., 2009; Li

et al., 2007; Park et al., 2009; Williams et al., 2001)

To determine which region of STIM2 is important for regulating the S845

phosphorylation of GluA1, hippocampal neurons were transduced with YFP-

75

STIM2Δcyto, YFP-STIM2ΔSOAR and YFP-STIM2. The overexpression of

YFP-STIM2 led to an increase in Ser845 phosphorylation compared to non-

transfected cells, suggesting a gain of function (Figure 15C). On the other hand,

YFP-STIM2Δcyto and YFP-STIM2ΔSOAR led to a decrease in GluA1 Ser845

phosphorylation in either resting cells or after cLTP treatment (Figure 15C and

D), suggesting that they act in a dominant-negative manner. Furthermore, we

observed a marked reduction in the interaction of AKAP150 and cPKA and

endogenous STIM2 with GluA1 in neurons transduced with YFP-STIM2 WT,

Δcyto or ΔSOAR mutants (Figure 15E). This suggests that the SOAR region of

STIM2 is crucial for the interaction of STIM2, AKAP150 and cPKA with GluA1,

which drives GluA1 S845 phosphorylation.

To determine the impact of the STIM2 mutants on dendritic spines, neurons were

transfected with YFP-STIM2Δcyto and YFP-STIM2ΔSOAR. Overexpression of

YFP-STIM2 led to an increase in dendritic spine size area (Figure 15B and F). In

contrast, the overexpression of the YFP-STIM2Δcyto and YFP-STIM2ΔSOAR

mutants led to a reduction of spine head size and spine density (Figure 15B, F and

G). The size and maturity of a dendritic spine has been linked to the expression of

AMPAR on the dendritic spine (Hanley, 2008; Hering and Sheng, 2001;

Matsuzaki et al., 2001). One possible explanation for the reduction in dendritic

spine head size is the decrease of GluA1 in synapses as a result of reduced GluA1

S845 phosphorylation when either of the STIM2 mutants is expressed. However,

it is also possible that the dendritic spine phenotype may not be related to

76

AMPAR. Based on the results, the SOAR domain is important for mediating the

STIM2-dependent phosphorylation of GluA1 and also dendritic spine

morphogenesis.

77

Figure 15. The STIM2 SOAR domain mediates phosphorylation of and

interaction with GluA1. (A) Cartoon showing the primary sequence of YFP-STIM2

WT, ΔSOAR and Δcyto. (B) Hippocampal neurons (21 DIV) coexpressing mCherry

and the indicated YFP-STIM2 variants. Scale bar: 10 μm. Both YFP-STIM2 and

YFPSTIM2ΔSOAR are localized to puncta, while YFP-STIM2Δcyto is more evenly

distributed. Spine size and density is reduced in neurons expressing YFP-

STIM2Δcyto and YFP-STIM2ΔSOAR. (C-E) Cortical neurons (21DIVs) transduced

with YFP-STIM2 WT and mutants. (C) Immunoblot analysis of GluA1 pSer845 in

cells expressing the indicated constructs and treated with DMSO (vehicle) or

forsk/rolipr for 30 min. NT: nontransduced cells. (D) Densitometry analysis of

pSer845 after forsk/rolipr treatment from four independent experiments. (E) IPs from

cells overexpressing YFP-STIM2 WT, Δcyto or ΔSOAR with GluA1 Ab or control

IgG. Lysates are the same used in (C, DMSO). Note the marked decrease in

endogenous STIM2, AKAP and cPKA pulled down from cells expressing the STIM2

mutants. (F, G) Distribution of spine size (area) (F) and spine density (G) scored in

neurons transfected with the indicated constructs or mock-transduced. More than 400

spines from three independent experiments were analyzed for each condition. *** p <

0.001, ANOVA. Data in Figure 15C-E were generated by and reproduced with

permission of Gisela Garcia-Alvarez

78

3.10 SOCE does not seem to be required for dendritic spine morphogenesis

Neuronal SOCE has been detected in recent studies and STIM2 knockout results

in lower ER calcium levels and SOCE in neurons (Berna-Erro et al., 2009; Sun et

al., 2014). To address if SOCE is necessary for dendritic spine morphogenesis, a

SOCE deficient STIM2 mutant was created by mutating the first two amino acids

of the SOAR domain (LQ347/348AA) (Yuan et al., 2009). The STIM1 version of

this mutant is still able to sense ER Ca2+

depletion and also interact with Orai1,

but is incapable of activating Orai channels and activate SOCE (Yuan et al.,

2009). When expressed in neurons, STIM2 LQAA is still localized to the

dendritic spines (arrows in Figure 16A) and the expression is similar to YFP-

STIM2.

The preliminary data also suggests that neither dendritic spine density nor the

dendritic spine area were affected when STIM2 LQAA was overexpressed in

primary hippocampal neurons in comparison to overexpression of YFP-STIM2

(Figure 16B and C), suggesting SOCE is not necessary for dendritic spine

formation/maturation. However, neuronal SOCE has to be measured in neurons

overexpressing the STIM2 LQAA mutant to determine if SOCE is necessary for

dendritic spine morphogenesis as the construct has only been tested in non-

neuronal cells to look at SOCE activity (Yuan et al., 2009).

79

3.11 STIM2 translocates to puncta in response to cLTP and promotes local

assembly of a STIM2-AKAP150/PKA-GluA1 signaling complex

STIM2 is required for GluA1 S845 phosphorylation in response to cLTP (Figure

11C). To address where STIM2 mediates GluA1 S845 phosphorylation, YFP-

STIM2 was imaged in hippocampal neurons during cLTP using live-cell confocal

imaging. YFP-STIM2 is redistributed into puncta in dendritic shafts (Figure 17A

Figure 16. SOCE is not crucial for STIM2 dependent dendritic spine

morphogenesis and remodelling. (A) Confocal image of a hippocampal neuron co-

transfected with either YFP-STIM2 or YFP-STIM2 LQAA (green) and mCherry

(red), and stained for GluA1 (blue). Localisation of YFP-STIM2 LQAA is similar to

YFP-STIM2 and can be found in dendritic spines. (B, C) Distribution of spine

density (B) and spine area (C) in neurons co-transfected with the either YFP-STIM2

or YFP-STIM2 LQAA and mCherry. At least 50 dendritic segments comprising >

500 spines from two independent experiments were scored for each condition.

Unpaired t-test.

80

and B) and is able to translocate in dendritic spines in response to cAMP elevation

(Figure 17C). In addition, more STIM2 is co-immunopurified with GluA1 after

cLTP (Figure 17D). Therefore, cAMP is able to promote STIM2 puncta

formation, which may be a possible site for the local assembly of a STIM2-

AKAP/PKA-GluA1 signaling complex.

Figure 17. cAMP can trigger STIM2 puncta formation and translocation into

dendritic spines. (A-C) Hippocampal neurons (DIV 21) electroporated with YFP-

STIM2 and mCherry and imaged by live-cell confocal microscopy during forsk/rolipr

treatment. (A) Confocal image of a dendritic segment before and 30 min after

forsk/rolipr addition. Note the redistribution of YFP-STIM2 in puncta along the

dendritic shaft. (B) Quantification of YFP-STIM2 re-localization to puncta by a

custom-written granularity script (see methods). n = 9. Shaded error bars indicate

SEM. (C) Dual-color imaging showing cAMP-induced redistribution of YFPSTIM2

(green) in dendritic spines labelled with mCherry (magenta). Snapshots are shown

before and 30 min after forsk/rolipr addition. Arrows indicate spines with increased

YFP-STIM2 intensity after cAMP elevation. (D) Co-IPs of cortical neurons (DIV 21)

treated with DMSO (vehicle) or forsk/rolipr for 30 min with a GluA1 Ab or control

IgGs. Data in Figure 17D were generated by and reproduced with permission of

Gisela Garcia-Alvarez.

81

3.12 Other possible impact of STIM2 on post-synaptic components

Dendritic spines contain a specialized cytoskeleton of dynamic filamentous actin

(F-actin) and this allows dendritic spines to be highly dynamic in both shape and

size (Hotulainen and Hoogenraad, 2010; Oertner and Matus, 2005). F-actin is

regulated by Ca2+

(Oertner and Matus, 2005) and it has been shown that dendritic

spines are lost when actin polymerization is disrupted (Kim et al., 2013a). To test

the possibility that STIM2 regulates actin dynamics, neurons were stained with

Phalloidin Rhodamine, which binds F-actin. The F-actin staining is similar in the

control and STIM2-knockdown neurons, suggesting that F-actin distribution is not

affected (Figure 18A). However, it may still be possible that actin polymerization

and turnover are affected and this can be addressed by using live imaging and

techniques such as fluorescent recovery after photobleaching to look at actin

dynamics.

Homer-1 is a scaffold protein that is enriched in dendritic spines and is important

for dendritic spines formation (Jaubert et al., 2007). Homer binds to proteins such

as group 1 metabotropic glutamate receptors, IP3R and ryanodine receptors and

has also been shown to modulate intracellular signaling in neurons (Fagni et al.,

2002; Xiao et al., 2000). To understand if the synaptic scaffold proteins are

affected in STIM2 silenced neurons since STIM2 might be a scaffold protein at

the plasma membrane, Homer1 was stained in STIM2 silenced neurons. A

staining of Homer in STIM2 silenced neurons showed that Homer can still be

found in immature looking dendritic spines (arrows in Figure 18B) but the

82

majority of Homer is distributed to the dendritic shaft as compared to the control

and STIM1 silenced neurons (Figure 10E and 18B). A possible explanation for

this is the loss of dendritic spines when STIM2 is silenced and thus Homer is

concentrated to the dendritic shaft. But it is still possible that the synaptic scaffold

is disrupted as a result of STIM2 knockdown. This can be addressed by looking at

the localization of Homer before the formation of synapses in STIM2 silenced

neurons and also if STIM2 silencing can disrupt the interaction of Homer1 with

its interacting partners.

Another possible mechanism in which dendritic spine morphogenesis is affected

when STIM2 is silenced is through cAMP/Ca2+

response element-binding protein

(CREB). CREB is a nuclear factor that is activated and binds to DNA to regulate

transcription when it gets phosphorylated by PKA (Benito and Barco, 2010;

Kandel, 2012; Kim et al., 2013b; Sakamoto et al., 2011). CREB is involved in late

phase LTP and CREB has been shown to be important for dendritic spine

formation (Benito and Barco, 2010; Kandel, 2012; Kim et al., 2013b; Sakamoto et

al., 2011). Therefore, it may be possible that STIM2 can affect the activity of

CREB through PKA dependent signaling pathways, which will in turn lead to less

dendritic spines formation. To determine if CREB is affected when STIM2 is

silenced, both phospho-CREB (S133) and CREB were immunoblotted.

Preliminary data showed that pCREB levels reduced at 21DIV when synapses are

formed (Figure 18C). This is similar to S845 phosphorylation trend (Figure 11A),

suggesting that STIM2 only affects synaptic activity when synapses are formed.

83

Therefore, it is possible that STIM2 may affect dendritic spine morphogenesis

through CREB but it is unknown how STIM2 can regulate the activity of CREB,

which may be through PKA or calcium.

Figure 18. Other potential mechanism in which STIM2 may affect spinogenesis.

(A) Hippocampal neurons (DIV 21) transfected with either scrambled or STIM2

shRNA and stained with phalloidin rhodamine to check for F-actin distribution. (B)

Hippocampal neurons (DIV 21) transfected with either scrambled, STIM1 shRNA or

STIM2 shRNA and stained with Homer. (C) Immunoblot analysis of phosphor-

CREB (pCREB), CREB and GluA1 in hippocampal neurons (21 DIV) transduced

with either scrambled shRNA (scr) or STIM2 shRNA #1.

84

DISCUSSION

This study has clearly demonstrated that the ER Ca2+

sensor, STIM2, is important

for dendritic spine morphogenesis and can form a complex with AKAP150/PKA-

GluA1 to regulate GluA1 S845 phosphorylation, although it is unknown if these

two events are related. This is a novel function for STIM2, a protein that resides

exclusively in the ER (Ercan et al., 2012), and provides a mechanism by which

the ER may specifically and locally regulate synaptic structure and function. In

this section, the implications of these findings and the shortfalls will be addressed.

4.1 STIM2 regulates dendritic spine morphogenesis

A recent study has shown the correlation between reduced levels of STIM2 and

reduction of mature spines phenotype, which can be rescued by the ectopic

expression of STIM2, in a presenilin-1 (PS1) mouse model (M146V) of familial

Alzheimer’s disease (Sun et al., 2014). In addition, cre-mediated excision of

STIM2 in hippocampal neurons of floxed STIM2 mice also resulted in a decrease

in the fraction of mushroom spines (Sun et al., 2014), thus providing independent

evidence for a role of STIM2 in dendritic spine maturation. The defect in

dendritic spine maturation was attributed to a reduction in synaptic SOCE and

CaMKII signaling, which is important for synaptic plasticity (Sun et al., 2014).

Together with the results in this thesis, it is also possible that STIM2 may regulate

hippocampal dendritic spine remodeling through a PKA dependent pathway,

which has been implicated in spine growth and remodeling (Robertson et al.,

2009; Smith et al., 2009).

85

In addition, the size and maturity of a dendritic spine has a tight correlation with

the expression of AMPARs on the dendritic spine (Hanley, 2008; Hering and

Sheng, 2001; Matsuzaki et al., 2001). AMPARs have been shown to be assembled

in the ER and then exported to the Golgi (Ziff, 2007) and it has been proposed

that AMPARs can directly (Passafaro et al., 2003) or indirectly (McKinney, 2010)

induce spine morphogenesis and remodeling. It is thus possible that STIM2 may

regulate dendritic spine size and maturation by controlling the synaptic

recruitment of AMPAR through a PKA dependent pathway in response to

synaptic activity.

PKA is localized and sequestered by AKAP150 to neuronal postsynaptic

membrane such as dendritic spines (Carr et al., 1992; Colledge and Scott, 1999;

Smith et al., 2006a). The expression of fluorescent-tagged cPKA and/or rPKA in

neurons can be used to address if the localization of PKA in neurons is dependent

on STIM2. To address if PKA activity is really involved in the spine phenotype

when STIM2 is silenced, a constitutively active cPKA subunit can be used to try

to rescue the dendritic spine phenotype. In addition, to determine if the decrease

in GluA1 S845 phosphorylation is also responsible for the dendritic spine

phenotype, a constitutively active GluA1 mutant (S845D) can be used to rescue

the spine phenotype when STIM2 is silenced.

86

4.2 Formation and/or maturation of dendritic spines

Dendritic spines have been proposed to first form as dendritic filopodia before

maturing into a dendritic spine, although spines may also be formed without a

filopodium-intermediate state (Kwon and Sabatini, 2011). There are currently two

models for the formation of dendritic spines from filopodia. One, dendritic

filopodia first grow and then search and recruit for a presynaptic partner. Two, a

presynaptic connection is made on the dendritic shaft, forming a synapse and this

is followed by extension and formation of spines (Segal and Korkotian, 2014;

Yoshihara et al., 2009). However, in both scenarios, the newly formed dendritic

spines require maturation after formation. Immature spines usually have a smaller

dendritic spine head and lack signaling capabilities as compared to mature

dendritic spines (Yoshihara et al., 2009). Similar phenomenon were observed in

the current study (Figure 9C, D and G) which is also in agreement with the

phenotype observed in the recent study on STIM2 (Sun et al., 2014). In addition,

there is less GluA1 S845 phosphorylation with cLTP treatment when STIM2 is

silenced (Figure 11C), suggesting that there may be less AMPAR that are inserted

into the dendritic spines (Figure 13A) (Esteban et al., 2003; Oh et al., 2006). Less

AMPARs in dendritic spines have been linked to the immaturity of dendritic

spines (Hanley, 2008; Hering and Sheng, 2001; Matsuzaki et al., 2001).

To address if STIM2 is required for the maturation of dendritic spines, STIM2 can

be silenced when dendritic spines are already formed at 21DIV. The number and

87

type of dendritic spines can then be checked at a later stage (28DIV) and

compared to the numbers on 21DIV.

4.3 Silent Synapses

Although AMPARs have been implicated in synapse formation and maturation, a

subset of dendritic spines, known as silent synapses can still form without

AMPARs (Hanse et al., 2013). Silent synapses are excitatory glutamatergic

synapses where the postsynaptic membrane contains only NMDARs and no

AMPARs. Since AMPARs are required for depolarization of the synapse and

removal of the Mg2+

block in NMDARs, silent synapses are unable to respond to

any synaptic vesicles released from the presynaptic compartment. Thus, with less

GluA1 inserted into the synapse as a result of STIM2 silencing, it may be possible

that the remaining synapses in STIM2 silenced neurons are silent to some extent

as mEPSCs are still measured. This may also explain the reduction in mEPSCs

amplitude, although it may also be possible that the conductance of AMPARs is

also affected. The presence of silent synapses can be checked by measuring

mEPSCs at a resting membrane potential of -40mV, where the NMDARs can be

activated at this resting potential.

4.4 Role of Ca2+

in STIM2-dependent dendritic spine morphogenesis and

remodeling

The main function of STIM2 is to sense a depletion of ER Ca2+

and regulate ER

Ca2+

homeostasis through SOCE in non-neuronal cells (Brandman et al., 2007).

88

However, the presence of SOCE in neurons has been widely debated. The lack of

a reliable neuronal SOCE response has been reported in previous reports (Bouron

et al., 2005; Garcia-Alvarez et al., 2015; Park et al., 2010), though neuronal

SOCE has been successfully detected in other recent studies (Berna-Erro et al.,

2009; Sun et al., 2014). However, the presence of neuronal SOCE cannot be

dismissed until proven, especially since it provides an attractive model for rising

intracellular Ca2+

within a dendritic spine to regulate synaptic plasticity.

A recent study has shown that Ca2+

entry through STIM2-dependent SOCE can

regulate CaMKII to affect dendritic spine maturation and thus regulating synaptic

plasticity (Sun et al., 2014). On the other hand, the use of the SOCE defective

mutant, YFP-STIM2 LQAA, seemed to suggest that SOCE is not essential for

dendritic spine morphogenesis and remodeling (Figure 16A). Overall cAMP

levels were also not affected (Figure 12A), which is dependent on Ca2+

levels and

CaMKII (Ferguson and Storm, 2004; Potter et al., 1980). In addition, activation of

STIM2 has been shown to mediate slower Orai1 activation than STIM1 (Parvez et

al., 2008; Zhou et al., 2009). Therefore, it is unlikely that overall Ca2+

levels in

the neurons are drastically affected by STIM2 silencing. STIM2 may have

evolved a new function in dendritic spine morphogenesis and remodeling that is

SOCE independent by forming a signaling complex with AKAP150/PKA-GluA1

to control PKA-dependent phosphorylation of GluA1 on Ser845.

89

The impact of Ca2+

influx, ER Ca2+

depletion and cytosolic Ca2+

changes on how

STIM2 can mediate dendritic spine morphogenesis and PKA-dependent GluA1

S845 phosphorylation should also be determined. For example, can ER Ca2+

depletion activate STIM2 to trigger dendritic spine morphogenesis and

remodeling and also PKA-dependent GluA1 S845 phosphorylation? This can be

addressed by using an EF-hand dominant positive mutant of STIM (D76A), which

mimics a constitutive active STIM that transmits ‘‘empty store’’ signal regardless

of actual Ca2+

level in ER (Klejman et al., 2009). The use of drugs such as

thapsigargin or cyclopiazonic acid can also be used to deplete ER Ca2+

(Emptage

et al., 2001).

4.5 STIM2 versus STIM1 in dendritic spine morphogenesis and remodeling

Both STIM1 and STIM2 are highly conserved, except for variations in their

amino-terminal, and carboxy-terminal segments (Berna-Erro et al., 2012; Deng et

al., 2009; Soboloff et al., 2012). The presence of STIM2 in dendritic spines can

be attributed to the presence of the ER in the dendritic spines, which can explain

why STIM1 can be also found in dendritic spines (Segal and Korkotian, 2014).

In agreement with the recent study (Sun et al., 2014) , STIM2 is the only STIM

isoform that can affect dendritic spine morphogenesis and remodeling. The

cytoplasmic and SOAR domains of STIM2 were shown to be important for

dendritic spine morphogenesis and remodeling. However, it is unknown whether

the dendritic spine phenotype is caused by the failure of STIM2 to form a

90

complex with AKAP150/PKA-GluA1 to regulate PKA-dependent

phosphorylation of GluA1 on Ser845 (Figure 15C, E, F and G). The SOAR

domain between STIM1 and STIM2 shows very little variation (Figure 19) and

thus it is possible that STIM1 may be able to rescue the phenotype observed when

STIM2 is silenced. However, when STIM1 is silenced in hippocampal neurons,

no phenotype on dendritic spines was observed (Figure 9B and C). Hence, it is

unlikely that STIM1 plays a role in dendritic spine morphogenesis and

remodeling.

A possible explanation for this is the differences in the cytoplasmic domain

between STIM2 and STIM1 (Figure 19). (i), STIM2 has a proline/histidine rich

motif while STIM1 has a proline/serine rich motif at the similar position

(Soboloff et al., 2012). The function of the proline/histidine rich region is

currently unknown. (ii), the C-terminal of both STIM isoforms differs

significantly. The Lys-rich C-terminal end of STIM1 has been shown to be

important for puncta formation, by interacting with acidic phospholipids at the

plasma membrane, and allowing the activation of Orai for SOCE channels (Liou

et al., 2007; Park et al., 2009; Yuan et al., 2009). The extreme C-terminus of

STIM2 (~200 residues) has also been shown to maintain STIM1 in an inactivate

state when ER Ca2+

stores are full and thus is thought to function sterically as

intramolecular shields to occlude the SOAR region (Yu et al., 2011). This region

is absent in D. melanogaster, suggesting that STIMs may have evolved new

functions in higher organisms (Soboloff et al., 2012).

91

To pinpoint the region of STIM2 that is important for the phenotypes observed in

STIM2 silenced neurons, the proline/histidine rich motif or the extreme C-

terminus region of STIM2 can be deleted to check for phenotypes similar to

STIM2 silenced neurons. In addition, the phenotypes observed in STIM2 silenced

cells can also be rescued with STIM1 chimera containing STIM2 motifs. For

example, the proline/serine rich motif of STIM1 can be substituted with the

proline/histidine rich motif of STIM2. The C-terminus replaced of STM1 can also

be replaced with that of STIM2, which is lysine rich.

Figure 19. Sequence alignment between STIM1 and STIM2. Similar regions are

highlighted in blue while differences are in black. The major domains have been

described and shown in Figure 5. Adapted from Soboloff, J., B. S. Rothberg, et al.

Nat Rev Mol Cell Biol 13(9): 549-565.

92

4.6 STIM2 and Orai in synaptic plasticity

The plasma membrane calcium channel, Orai, can be found be in different parts

of the brain (Gross et al., 2007; Gwack et al., 2007) and has been shown to be also

important for synaptic plasticity through regulating ER Ca2+

stores (Korkotian and

Segal, 2011) . Recently, Orai has been shown to interact with AKAP79, the

human analog of AKAP150, to recruit the phosphatase, calcineurin, into a

membrane signaling complex that is ER Ca2+

store dependent in HEK293 cells

(Kar et al., 2014).

Although Orai was not investigated in this study, it is possible that STIM2

functions with or through Orai to regulate the dendritic spine phenotype by

forming a signaling complex with AKAP150/PKA and GluA1 to regulate GluA1

S845 phosphorylation. The SOAR domain is important for the interaction and

activity of Orai channels (Yuan et al., 2009) and YFP-STIM2ΔSOAR mutant was

unable to interact with AKAP150, PKA and GluA1 (Figure 15E) and also had a

reduction in dendritic spines (Figure 15G). In addition, as compared to the YFP-

STIM2ΔSOAR mutant, the loss in dendritic spine phenotype was not manifested

in the YFP-STIM2 LQAA, a mutant which can still interact with Orai but not

activate Orai in HEK cells (Yuan et al., 2009). Thus it is still possible that Orai

interaction with STIM2 is necessary for the signaling complex to form. To

address if Orai is important, GluA1 or STIM2 can be pulled down to check for the

presence of Orai together with AKAP150 and PKA. GluA1 can also be pulled

down in STIM2 silenced cells to check for the presence of Orai. If Orai is present

93

in the STIM2/AKAP150/PKA/GluA1 complex, a knockdown of Orai will be

necessary to show that Orai is necessary for the signaling complex to form.

4.7 STIM2 activation

STIM proteins were first discovered to be activated in response to ER Ca2+

store

depletion to trigger SOCE (Liou et al., 2005). A recent study in pancreatic β cells

showed that cAMP can trigger STIM1 translocation close to the plasma

membrane without co-recruitment of Orai1 nor the activation of SOCE (Tian et

al., 2012). Similarly, cAMP was also shown to be able to trigger STIM2 puncta

formation and translocation into dendritic spines (Figure 17A and C), suggesting

that Ca2+

dissociation from the STIM sensors is not an absolute requirement for

STIM2 puncta formation and migration to the plasma membrane. cAMP is a

secondary messenger that can regulate synaptic plasticity and memory formation

through gene transcription and PKA-dependent phosphorylation of synaptic

targets (Waltereit and Weller, 2003). Thus based on the results, this is a novel

way in which cAMP can fine tune synaptic plasticity directly through STIM2, but

the exact mechanisms by which cAMP can trigger redistribution of STIM2

remains unknown.

However, it has been shown that cAMP can increase the sensitivity of IP3R to IP3

(Rossi et al., 2012). STIM2 puncta formation and translocation to the plasma

membrane may be simply due to more ER Ca2+

store depletion with increased

sensitivity of the IP3R. This may prime the activation of STIM2 with synaptic

94

activation or cLTP where both the cAMP and Ca2+

pathways cooperate to

increase the sensitivity of STIM2 to synaptic activity. To address if IP3R is

necessary for STIM2 activation under cLTP, the IP3R inhibitor, Xestospongin C,

can be used to block ER Ca2+

depletion through IP3R to check if STIM2 is

activated or not under cLTP. If the activation of STIM2 by cAMP is through

IP3R, STIM2 may regulate synaptic plasticity through both glutamate receptors

(cAMP production) and metabotropic glutamate receptors (IP3R activation and

also cAMP production).

4.8 cLTP: Forskolin/Rolipram-induced cLTP vs Glycine-induced cLTP

The cLTP protocol used in this study, with the adenylate cyclase activator,

forskolin, and the phosphodiestarase inhibitor, rolipram, has been widely used in

slice cultures and also dissociated hippocampal neuronal cultures (Molnár, 2011).

The application of the two drugs bypasses the need for synaptic activation and

raises the level of cAMP, which activates PKA. PKA dependent signaling

pathways that underlies synaptic plasticity are then also activated (Otmakhov et

al., 2004b). Thus, forskolin/rolipram induced cLTP is ideal for studying the

connection on STIM2 and GluA1 S845 phosphorylation since the effect is direct.

To determine if STIM2 can affect GluA1 S845 phosphorylation under synaptic

activation, a glycine induced cTLP can be used. This protocol makes use of

bicuculline, glycine and strychnine in an Mg2+

free environment (Molnár, 2011).

Bicuculline is a GABA receptor antagonist, while glycine is a NMDAR agonist

95

and strychnine prevents the inhibitory effects of glycine receptors, which allows

the influx of chloride into the neuron. Hence this protocol makes use of the

activation of NMDARs at the synapse upon synaptic activation from spontaneous

release of glutamate pre-synaptically.

The phosphorylation state of GluA1 can be checked with the glycine-induced

cLTP protocol to confirm that STIM2 is also necessary for GluA1 S845

phosphorylation under synaptic activation.

4.9 STIM2 forming a complex with AKAP150, PKA and GluA1

STIM2 is part of a protein complex comprising GluA1, PKA and AKAP150 and

is required for GluA1 to interact with AKAP150 and PKA (Figure 14D). STIM2

may form a protein complex indirectly with AKAP150, GluA1 and other binding

partners. The assembly of this complex can take place in large dendritic spines,

where STIM2, AMPARs and AKAP150 are known to reside, or at ER-PM

contact sites in dendritic shafts or in the cell body, which contain the bulk of the

ER.

4.10 GluA1 phosphorylation and LTP/LTD

LTP and LTD are two models of synaptic plasticity that have been associated

with learning and memory, and both pSer845 and pSer831 residues play a

regulatory role in LTP and LTD (Esteban et al., 2003; Lee et al., 2010a; Lee et al.,

2003; Makino et al., 2011). GluA1 pSer845 has been implicated in activity-

96

dependent AMPAR surface delivery and plasticity and is dependent on

AKAP79/150 and PKA. But the location of that phosphorylation event and the

involvement of STIM2 with the known players are unclear.

The phosphorylation of S845 residue has been implicated in LTP and is necessary

for the insertion of GluA1 into the synapse. This is achieved by exocytosis of

AMPARs to extra- or peri-synaptic sites, lateral recruitment to the synapse, and

decreased AMPAR endocytosis (Ehlers, 2000a; Esteban et al., 2003; Makino and

Malinow, 2009; Oh et al., 2006). STIM2 mediates at least 80% of GluA1 Ser845

phosphorylation (Figure 11D) and is functioning downstream of adenylate cyclase

and PKA (Figure 12A and B respectively) through the assembly of a protein

complex consisting of GluA1, PKA and AKAP150. AKAP150 anchors PKA to

dendritic spines (Carr et al., 1992; Colledge and Scott, 1999; Smith et al., 2006a)

and is responsible for PKA regulation of AMPAR activity (Lu et al., 2008). It

may be possible that STIM2 is required for the localization of AKAP/PKA to

postsynaptic membrane (Figure 12C and 14F) and uncoupling of AKAP/PKA

from GluA1 may explain the phosphorylation deficit detected in STIM2 silenced

neurons. Thus, ER-PM junction sites between STIM2 and AKAP/PKA/GluA1

could serve as sites for activity dependent insertion of AMPARs into the synapse

and it is through such sites that the synapse may form, mature, and become

plastic.

97

The dephosphorylation of the S845 is also critical for LTD and it has been shown

that LTD dissociates AKAP150 from the AMPAR complex (Smith et al., 2006b).

We found that STIM2 is required for AKAP150 to interact with GluA1 but it is

unknown if STIM2 also plays a part in dissociating AKAP150 from GluA1.

Interestingly, in contrast to GluA1 S845, GluA1 S831 phosphorylation increases

when STIM2 is silenced (5A). It has been shown that CamKII activity is

decreased when STIM2 is silenced in neurons (Sun et al., 2014) and CamKII

activity is required for GluA1 S831 phosphorylation (Lee et al., 2003). Thus, the

increased in GluA1 S831 phosphorylation in STIM2 silenced neurons appears to

be CamKII independent and of unknown mechanism. The S831 has been shown

to be able to substitute S845 in mediating LTP in the hippocampus of S845A

knockin mutant mouse (Lee et al., 2010b) and is also involved in homeostatic

plasticity (Goel et al., 2006), which is the ability of neurons to regulate their own

excitability relative to network activity. Therefore, a possible explanation for the

increase in S831 phosphorylation could be a compensatory mechanism for the

decrease in S845 phosphorylation.

4.11 STIM2 signals at ER-PM contact sites and enforces proximity between

GluA1, AKAP150 and PKA

The presence of ER-PM junction sites was first reported in muscles cells (Porter

and Palade, 1957) using ultrastructural studies in the late 1950's. Subsequently, it

was also discovered in neurons (Cui-Wang et al., 2012; Hayashi et al., 2008;

98

Rosenbluth, 1962) but the components of ER-PM junction in neurons still remains

unknown. There are several criteria to classify an ER-PM junction site. (i) the ER

and PM are in close proximity to create a restricted protein-rich cytoplasmic

space. (ii) both the ER and the PM must still retain their identity and not fuse. (iii)

the stability of the ER-PM junction sites can range from minutes to days

(Carrasco and Meyer, 2011). In neurons, ER–PM contact sites can appear as

discrete punctate regions at the presynaptic side (Hayashi et al., 2008) but appear

more extensive (>1 μm) in the cell body, at bases of axons and dendrites, and at

the PM of cells neighboring synaptic boutons (Rosenbluth, 1962).

There are several evidences to suggest that STIM2 may mediate PKA dependent

GluA1 phosphorylation at ER-PM contact sites. (i) STIM2 puncta in the primary

hippocampal neurons (Figure 8B and C) are most likely to correspond to ER-PM

contact sites because cLTP is able to induce STIM2 puncta formation at the PM

of neurons using Total Internal Reflection Fluorescence (TIRF) imaging (Garcia-

Alvarez et al., 2015) (ii) increasing the amount of cAMP triggers rapid

translocation of YFP-STIM2 into puncta in the dendrites and dendritic spines

(Figure 17A and C) and this also leads to more STIM2 co-immunopurified with

GluA1 (Figure 17D). Localization at ER-PM junctions would therefore position

STIM2 in close proximity to the plasma membrane and allow direct interaction

with cytoplasmic domains of surface receptors to facilitate the phosphorylation of

GluA1 S845 by PKA. The use of transmission electron microscope will be able to

address if STIM2 is really present at such sites or not.

99

Another possibility is that STIM2 can interact with GluA1 at locations other than

the PM. The assembly of AMPAR subunits occurs in the ER (Greger et al., 2007)

and thus, it is tempting to speculate that STIM2 can also regulate the assembly or

the exocytosis of the AMPARs at the ER. Local exocytosis of AMPARs at distal

ER exit sites, in close proximity to ER-PM contacts, would provide tight spatial

and temporal control for activity-dependent synaptic delivery of AMPARs. To

address if STIM2 is able to interact with GluA1 in the ER, STIM2 can be pulled

down from a brain ER fractionation to check for the presence of GluA1 and also

the phosphorylation state of GluA1.

In addition, recycling endosomes can also provide a mobilizeable pool of GluA1

for plasticity. Indeed, phosphorylation of the S845 residue is critical for surface

delivery of GluA1 from recycling endosomes (Ehlers, 2000a). To check if

STIM2 interacts with GluA1 in recycling endosomes, the use of transmission

electron microscope will be able to address if STIM2 is in recycling endosomes

with GluA1. Another alternative is to make use of the antibody reuptake

experiment (Figure 13D and E) to determine if endocytosed GluA1 colocalizes

withYFP-STIM2 in early/recycling endosomes.

4.12 Working model for STIM2-dependent phosphorylation of GluA1 and

dendritic spine remodeling

100

With the majority of STIM2 puncta in the dendritic shaft, which are extrasynaptic,

ER-PM junction sites in the dendritic shafts can be potential “hot spots” for

STIM2-dependent AMPAR phosphorylation, to coordinate activity-dependent

delivery and maintenance of AMPARs at synapses (Figure 20).

The release of synaptic vesicles containing glutamate can trigger the following

events: (i), glutamate can activate AMPARs and trigger the depolarization of the

neuron. This allows for the activation of the NMDARs, which can allow the

influx of Ca2+

into the neuron. VGCCs are also activated upon depolarization,

which also allows the influx of Ca2+

into the neuron. The influx of Ca2+

will

trigger the activation of Ca2+

-dependentadenylate cyclase and thus increase the

levels of cAMP (Wang and Zhang, 2012), which can directly or indirectly activate

STIM2.

(ii), glutamate can also activate metabotropic glutamate receptors (mGluRs),

which can also be found on dendrites (Luján et al., 1997). The activation of

mGluRs is able to activate IP3Rs on the ER, which are primed by cAMP (Rossi et

al., 2012), and thus depleting the ER Ca2+

store. This depletion of ER Ca2+

stores

is also able to activate STIM2.

Upon the activation of STIM2, STIM2 will translocate to ER-PM contact sites,

where AKAP150/PKA is recruited to phosphorylate GluA1 at S845. The

phosphorylation of GluA1 at S845 allows more AMPARs to be inserted into the

101

PM at extrasynaptic sites. Lateral recruitment of GluA1 to the synapse (Esteban et

al., 2003; Makino and Malinow, 2009; Oh et al., 2006) then occurs during LTP.

(iii), As a result of glutamate receptor activation, STIM2 regulates

phosphorylation of CREB, which turns on a transcriptional program that

ultimately leads to dendritic spine formation (Benito and Barco, 2010; Kandel,

2012; Kim et al., 2013b; Purves et al., 2001; Sakamoto et al., 2011).

This working model suggests that STIM2 integrates both Ca2+

and cAMP signals

to regulate activity-dependent synaptic insertion of the AMPAR and spine

remodeling via both local and long-range (i.e. synapse to nucleus) signaling

mechanisms.

102

Figure 20. Model for STIM2-dependent regulation of AMPAR phosphorylation

and trafficking at the interface between the ER and the PM. cAMP elevation and

decrease in ER Ca2+

store resulting from synaptic or other signaling inputs can

triggers translocation of STIM2 to ER-PM contact sites and contribute to the

dynamic assembly of a PKA signaling complex that drives phosphorylation of GluA1

Ser845 and surface delivery of the AMPAR.

103

Conclusion

This study has provided a novel mechanism of synaptic remodeling which

involves functional coupling between the ER and the dendritic PM through

STIM2 to regulate synapse plasticity. STIM2 has been shown to be crucial for

forming a STIM2 dependent complex with AKAP150/PKA/GluA1 and this

complex is required for PKA dependent phosphorylation of GluA1 S845.

Through the phosphorylation of GluA1 S845, AMPARs can be inserted into the

synapse for the maturation of the synapse. Therefore, this work suggests ER-PM

coupling can be a fundamental feature of cell and synapse physiology.

104

References

Baba, A., Yasui, T., Fujisawa, S., Yamada, R.X., Yamada, M.K., Nishiyama, N.,

Matsuki, N., and Ikegaya, Y. (2003). Activity-evoked capacitative Ca2+ entry:

Implications in synaptic plasticity. Journal of Neuroscience 23, 7737-7741.

Baba, Y., Hayashi, K., Fujii, Y., Mizushima, A., Watarai, H., Wakamori, M.,

Numaga, T., Mori, Y., Iino, M., Hikida, M., et al. (2006). Coupling of STIM1 to

store-operated Ca2+ entry through its constitutive and inducible movement in the

endoplasmic reticulum. Proceedings of the National Academy of Sciences 103,

16704-16709.

Baba, Y., Nishida, K., Fujii, Y., Hirano, T., Hikida, M., and Kurosaki, T. (2008).

Essential function for the calcium sensor STIM1 in mast cell activation and

anaphylactic responses. Nat Immunol 9, 81-88.

Baker, K.D., Edwards, T.M., and Rickard, N.S. (2013). The role of intracellular

calcium stores in synaptic plasticity and memory consolidation. Neuroscience &

Biobehavioral Reviews 37, 1211-1239.

Banke, T.G., Bowie, D., Lee, H., Huganir, R.L., Schousboe, A., and Traynelis,

S.F. (2000). Control of GluR1 AMPA receptor function by cAMP-dependent

protein kinase. J Neurosci 20, 89-102.

Bardo, S., Cavazzini, M.G., and Emptage, N. (2006). The role of the endoplasmic

reticulum Ca2+ store in the plasticity of central neurons. Trends Pharmacol Sci

27, 78-84.

Barnes, C.A. (1995). Involvement of LTP in memory: Are we “searching under

the street light”? Neuron 15, 751-754.

Barria, A., Muller, D., Derkach, V., Griffith, L.C., and Soderling, T.R. (1997).

Regulatory Phosphorylation of AMPA-Type Glutamate Receptors by CaM-KII

During Long-Term Potentiation. Science 276, 2042-2045.

Bear, M.F., and Malenka, R.C. (1994). Synaptic plasticity: LTP and LTD. Current

Opinion in Neurobiology 4, 389-399.

Beattie, E.C., Carroll, R.C., Yu, X., Morishita, W., Yasuda, H., von Zastrow, M.,

and Malenka, R.C. (2000). Regulation of AMPA receptor endocytosis by a

signaling mechanism shared with LTD. Nat Neurosci 3, 1291-1300.

Benito, E., and Barco, A. (2010). CREB's control of intrinsic and synaptic

plasticity: implications for CREB-dependent memory models. Trends in

Neurosciences 33, 230-240.

Benke, T.A., Luthi, A., Isaac, J.T.R., and Collingridge, G.L. (1998). Modulation

of AMPA receptor unitary conductance by synaptic activity. Nature 393, 793-797.

Berna-Erro, A., Braun, A., Kraft, R., Kleinschnitz, C., Schuhmann, M.K.,

Stegner, D., Wultsch, T., Eilers, J., Meuth, S.G., Stoll, G., et al. (2009). STIM2

regulates capacitive Ca2+ entry in neurons and plays a key role in hypoxic

neuronal cell death. Sci Signal 2, ra67.

Berna-Erro, A., Woodard, G.E., and Rosado, J.A. (2012). Orais and STIMs:

physiological mechanisms and disease. Journal of Cellular and Molecular

Medicine 16, 407-424.

105

Berrout, J., and Isokawa, M. (2009). Homeostatic and stimulus-induced coupling

of the L-type Ca(2+) channel to the ryanodine receptor in the hippocampal neuron

in slices. Cell calcium 46, 30-38.

Bezprozvanny, I., and Hayden, M.R. (2004). Deranged neuronal calcium

signaling and Huntington disease. Biochemical and Biophysical Research

Communications 322, 1310-1317.

Bird, G.S., Hwang, S.-Y., Smyth, J.T., Fukushima, M., Boyles, R.R., and Putney

Jr, J.W. (2009). STIM1 Is a Calcium Sensor Specialized for Digital Signaling.

Current Biology 19, 1724-1729.

Bliss, T.V.P., and Collingridge, G.L. (1993). A synaptic model of memory: Long-

term potentiation in the hippocampus. Nature 361, 31-39.

Borgdorff, A.J., and Choquet, D. (2002). Regulation of AMPA receptor lateral

movements. Nature 417, 649-653.

Bosch, M., and Hayashi, Y. (2012). Structural plasticity of dendritic spines.

Current Opinion in Neurobiology 22, 383-388.

Bourne, J.N., and Harris, K.M. (2012). Nanoscale analysis of structural synaptic

plasticity. Curr Opin Neurobiol 22, 372-382.

Bouron, A. (2000). Activation of a capacitative Ca<sup>2+</sup> entry pathway

by store depletion in cultured hippocampal neurones. FEBS Letters 470, 269-272.

Bouron, A., Altafaj, X., Boisseau, S., and De Waard, M. (2005). A store-operated

Ca2+ influx activated in response to the depletion of thapsigargin-sensitive Ca2+

stores is developmentally regulated in embryonic cortical neurons from mice.

Brain Res Dev Brain Res 159, 64-71.

Brandman, O., Liou, J., Park, W.S., and Meyer, T. (2007). STIM2 is a feedback

regulator that stabilizes basal cytosolic and endoplasmic reticulum Ca2+ levels.

Cell 131, 1327-1339.

Brewer, G.J., Boehler, M.D., Pearson, R.A., DeMaris, A.A., Ide, A.N., and

Wheeler, B.C. (2009). Neuron network activity scales exponentially with synapse

density. Journal of neural engineering 6, 014001-014001.

Cahalan, M.D. (2009). STIMulating store-operated Ca2+ entry. Nat Cell Biol 11,

669-677.

Calloway, N., Holowka, D., and Baird, B. (2010). A Basic Sequence in STIM1

Promotes Ca2+ Influx by Interacting with the C-Terminal Acidic Coiled Coil of

Orai1. Biochemistry 49, 1067-1071.

Carr, D.W., Stofko-Hahn, R.E., Fraser, I.D., Cone, R.D., and Scott, J.D. (1992).

Localization of the cAMP-dependent protein kinase to the postsynaptic densities

by A-kinase anchoring proteins. Characterization of AKAP 79. J Biol Chem 267,

16816-16823.

Carrasco, S., and Meyer, T. (2011). STIM proteins and the endoplasmic

reticulum-plasma membrane junctions. In Annual Review of Biochemistry, pp.

973-1000.

Carroll, R.C., Lissin, D.V., Zastrow, M.v., Nicoll, R.A., and Malenka, R.C.

(1999). Rapid redistribution of glutamate receptors contributes to long-term

depression in hippocampal cultures. Nat Neurosci 2, 454-460.

106

Chavis, P., Fagni, L., Lansman, J.B., and Bockaert, J. (1996). Functional coupling

between ryanodine receptors and L-type calcium channels in neurons. Nature 382,

719-722.

Chen, Y., Wang, Y., Ertürk, A., Kallop, D., Jiang, Z., Weimer, Robby M.,

Kaminker, J., and Sheng, M. (2014). Activity-Induced Nr4a1 Regulates Spine

Density and Distribution Pattern of Excitatory Synapses in Pyramidal Neurons.

Neuron 83, 431-443.

Colledge, M., Dean, R.A., Scott, G.K., Langeberg, L.K., Huganir, R.L., and Scott,

J.D. (2000). Targeting of PKA to Glutamate Receptors through a MAGUK-

AKAP Complex. Neuron 27, 107-119.

Colledge, M., and Scott, J.D. (1999). AKAPs: from structure to function. Trends

in Cell Biology 9, 216-221.

Collins, S.R., and Meyer, T. (2011). Evolutionary origins of STIM1 and STIM2

within ancient Ca2+ signaling systems. Trends Cell Biol 21, 202-211.

Cotman, C.W., Banker, G., Churchill, L., and Taylor, D. (1974). Isolation of

postsynaptic densities from rat brain. J Cell Biol 63, 441-455.

Cotman, C.W., and Taylor, D. (1972). Isolation and structural studies on synaptic

complexes from rat brain. J Cell Biol 55, 696-711.

Covington, E.D., Wu, M.M., and Lewis, R.S. (2010). Essential Role for the

CRAC Activation Domain in Store-dependent Oligomerization of STIM1.

Molecular Biology of the Cell 21, 1897-1907.

Craig, A.M., Blackstone, C.D., Huganir, R.L., and Banker, G. (1993). The

distribution of glutamate receptors in cultured rat hippocampal neurons:

Postsynaptic clustering of AMPA selective subunits. Neuron 10, 1055-1068.

Cui-Wang, T., Hanus, C., Cui, T., Helton, T., Bourne, J., Watson, D., Harris,

K.M., and Ehlers, M.D. (2012). Local zones of endoplasmic reticulum complexity

confine cargo in neuronal dendrites. Cell 148, 309-321.

de Sevilla, D.F., Núñez, A., Borde, M., Malinow, R., and Buño, W. (2008).

Cholinergic-Mediated IP3-Receptor Activation Induces Long-Lasting Synaptic

Enhancement in CA1 Pyramidal Neurons. The Journal of Neuroscience 28, 1469-

1478.

Del Castillo J, K.B. (1954). Quantal components of the end-plate potential. J

Physiol (Lond), 560–573.

Deng, X., Wang, Y., Zhou, Y., Soboloff, J., and Gill, D.L. (2009). STIM and

Orai: Dynamic Intermembrane Coupling to Control Cellular Calcium Signals.

Journal of Biological Chemistry 284, 22501-22505.

Derkach, V., Barria, A., and Soderling, T.R. (1999). Ca2+/calmodulin-kinase II

enhances channel conductance of alpha-amino-3-hydroxy-5-methyl-4-

isoxazolepropionate type glutamate receptors. Proc Natl Acad Sci U S A 96,

3269-3274.

Dziadek, M.A., and Johnstone, L.S. (2007). Biochemical properties and cellular

localisation of STIM proteins. Cell Calcium 42, 123-132.

Ehlers, M.D. (2000a). Reinsertion or degradation of AMPA receptors determined

by activity-dependent endocytic sorting. Neuron 28, 511-525.

Ehlers, M.D. (2000b). Reinsertion or Degradation of AMPA Receptors

Determined by Activity-Dependent Endocytic Sorting. Neuron 28, 511-525.

107

Emptage, N.J., Reid, C.A., and Fine, A. (2001). Calcium stores in hippocampal

synaptic boutons mediate short-term plasticity, store-operated Ca2+ entry, and

spontaneous transmitter release. Neuron 29, 197-208.

Esteban, J.A., Shi, S.H., Wilson, C., Nuriya, M., Huganir, R.L., and Malinow, R.

(2003). PKA phosphorylation of AMPA receptor subunits controls synaptic

trafficking underlying plasticity. Nature Neuroscience 6, 136-143.

Fagni, L., Worley, P.F., and Ango, F. (2002). Homer as Both a Scaffold and

Transduction Molecule, Vol 2002.

Ferguson, G.D., and Storm, D.R. (2004). Why Calcium-Stimulated Adenylyl

Cyclases?, Vol 19.

Feske, S., Gwack, Y., Prakriya, M., Srikanth, S., Puppel, S.H., Tanasa, B., Hogan,

P.G., Lewis, R.S., Daly, M., and Rao, A. (2006). A mutation in Orai1 causes

immune deficiency by abrogating CRAC channel function. Nature 441, 179-185.

Finch, E.A., and Augustine, G.J. (1998). Local calcium signalling by inositol-

1,4,5-trisphosphate in Purkinje cell dendrites. Nature 396, 753-756.

Garaschuk, O., Yaari, Y., and Konnerth, A. (1997). Release and sequestration of

calcium by ryanodine-sensitive stores in rat hippocampal neurones. J Physiol 502

( Pt 1), 13-30.

Garcia-Alvarez, G., Lu, B., Yap, K.A.F., Wong, L.C., Thevathasan, J.V., Lim, L.,

Ji, F., Tan, K.W., Mancuso, J.J., Tang, W., et al. (2015). STIM2 regulates PKA-

dependent phosphorylation and trafficking of AMPARs. Molecular Biology of the

Cell 26, 1141-1159.

Garcia-Lopez, P., Garcia-Marin, V., and Freire, M. (2010). Dendritic spines and

development: towards a unifying model of spinogenesis--a present day review of

Cajal's histological slides and drawings. Neural Plast 2010, 769207.

Gasperini, R., Choi-Lundberg, D., Thompson, M., Mitchell, C., and Foa, L.

(2009). Homer regulates calcium signalling in growth cone turning. Neural

Development 4, 29.

Gemes, G., Bangaru, M.L., Wu, H.E., Tang, Q., Weihrauch, D., Koopmeiners,

A.S., Cruikshank, J.M., Kwok, W.M., and Hogan, Q.H. (2011). Store-operated

Ca2+ entry in sensory neurons: functional role and the effect of painful nerve

injury. J Neurosci 31, 3536-3549.

Goel, A., Jiang, B., Xu, L.W., Song, L., Kirkwood, A., and Lee, H.-K. (2006).

Cross-modal regulation of synaptic AMPA receptors in primary sensory cortices

by visual experience. Nat Neurosci 9, 1001-1003.

Gogolla, N., Galimberti, I., DePaola, V., and Caroni, P. (2006). Preparation of

organotypic hippocampal slice cultures for long-term live imaging. Nat Protocols

1, 1165-1171.

Gomez, T.M., and Zheng, J.Q. (2006). The molecular basis for calcium-dependent

axon pathfinding. Nat Rev Neurosci 7, 115-125.

Greger, I.H., Ziff, E.B., and Penn, A.C. Molecular determinants of AMPA

receptor subunit assembly. Trends in Neurosciences 30, 407-416.

Greger, I.H., Ziff, E.B., and Penn, A.C. (2007). Molecular determinants of AMPA

receptor subunit assembly. Trends in Neurosciences 30, 407-416.

Gross, S.A., Wissenbach, U., Philipp, S.E., Freichel, M., Cavalié, A., and

Flockerzi, V. (2007). Murine ORAI2 Splice Variants Form Functional Ca2+

108

Release-activated Ca2+ (CRAC) Channels. Journal of Biological Chemistry 282,

19375-19384.

Gruszczynska-Biegala, J., Pomorski, P., Wisniewska, M.B., and Kuznicki, J.

(2011). Differential roles for STIM1 and STIM2 in store-operated calcium entry

in rat neurons. PLoS One 6, e19285.

Gwack, Y., Srikanth, S., Feske, S., Cruz-Guilloty, F., Oh-hora, M., Neems, D.S.,

Hogan, P.G., and Rao, A. (2007). Biochemical and Functional Characterization of

Orai Proteins. Journal of Biological Chemistry 282, 16232-16243.

Hahn, C.G., Banerjee, A., Macdonald, M.L., Cho, D.S., Kamins, J., Nie, Z.,

Borgmann-Winter, K.E., Grosser, T., Pizarro, A., Ciccimaro, E., et al. (2009). The

post-synaptic density of human postmortem brain tissues: an experimental study

paradigm for neuropsychiatric illnesses. PLoS One 4, e5251.

Hama, H., Kurokawa, H., Kawano, H., Ando, R., Shimogori, T., Noda, H.,

Fukami, K., Sakaue-Sawano, A., and Miyawaki, A. (2011). Scale: a chemical

approach for fluorescence imaging and reconstruction of transparent mouse brain.

Nat Neurosci 14, 1481-1488.

Hanley, J.G. (2008). AMPA receptor trafficking pathways and links to dendritic

spine morphogenesis. Cell Adhesion and Migration 2, 276-282.

Hanse, E., Seth, H., and Riebe, I. (2013). AMPA-silent synapses in brain

development and pathology. Nat Rev Neurosci 14, 839-850.

Harris, K.M., Jensen, F.E., and Tsao, B. (1992). Three-dimensional structure of

dendritic spines and synapses in rat hippocampus (CA1) at postnatal day 15 and

adult ages: implications for the maturation of synaptic physiology and long-term

potentiation. J Neurosci 12, 2685-2705.

Hartmann, J., Karl, R.M., Alexander, R.P., Adelsberger, H., Brill, M.S.,

Ruhlmann, C., Ansel, A., Sakimura, K., Baba, Y., Kurosaki, T., et al. (2014).

STIM1 Controls Neuronal Ca(2+) Signaling, mGluR1-Dependent Synaptic

Transmission, and Cerebellar Motor Behavior. Neuron 82, 635-644.

Hawkins, B.J., Irrinki, K.M., Mallilankaraman, K., Lien, Y.C., Wang, Y.,

Bhanumathy, C.D., Subbiah, R., Ritchie, M.F., Soboloff, J., Baba, Y., et al.

(2010). S-glutathionylation activates STIM1 and alters mitochondrial

homeostasis. J Cell Biol 190, 391-405.

Hayashi, M., Raimondi, A., O'Toole, E., Paradise, S., Collesi, C., Cremona, O.,

Ferguson, S.M., and De Camilli, P. (2008). Cell- and stimulus-dependent

heterogeneity of synaptic vesicle endocytic recycling mechanisms revealed by

studies of dynamin 1-null neurons. Proceedings of the National Academy of

Sciences of the United States of America 105, 2175-2180.

Hayashi, Y., Shi, S.-H., Esteban, J.A., Piccini, A., Poncer, J.-C., and Malinow, R.

(2000). Driving AMPA Receptors into Synapses by LTP and CaMKII:

Requirement for GluR1 and PDZ Domain Interaction. Science 287, 2262-2267.

Hebb, D.O. (1949). Organization of Behavior (New York: John Wiley and Sons).

Hering, H., and Sheng, M. (2001). Dentritic spines : structure, dynamics and

regulation. Nat Rev Neurosci 2, 880-888.

Herms, J., Schneider, I., Dewachter, I., Caluwaerts, N., Kretzschmar, H., and Van

Leuven, F. (2003). Capacitive calcium entry is directly attenuated by mutant

109

presenilin-1, independent of the expression of the amyloid precursor protein.

Journal of Biological Chemistry 278, 2484-2489.

Hewavitharana, T., Deng, X., Wang, Y., Ritchie, M.F., Girish, G.V., Soboloff, J.,

and Gill, D.L. (2008). Location and Function of STIM1 in the Activation of Ca2+

Entry Signals. Journal of Biological Chemistry 283, 26252-26262.

Higley, M.J. (2014). Localized GABAergic inhibition of dendritic Ca2+

signalling. Nat Rev Neurosci 15, 567-572.

Honoré, T., Lauridsen, J., and Krogsgaard-Larsen, P. (1982). The Binding of

[3H]AMPA, a Structural Analogue of Glutamic Acid, to Rat Brain Membranes.

Journal of Neurochemistry 38, 173-178.

Hooper, R., Rothberg, B.S., and Soboloff, J. (2014). Neuronal STIMulation at

Rest, Vol 7.

Horak, M., Petralia, R.S., Kaniakova, M., and Sans, N. (2014). ER to synapse

trafficking of NMDA receptors. Frontiers in Cellular Neuroscience 8, 394.

Horne, J.H., and Meyer, T. (1997). Elementary Calcium-Release Units Induced

by Inositol Trisphosphate. Science 276, 1690-1693.

Hoshi, N., Langeberg, L.K., and Scott, J.D. (2005). Distinct enzyme combinations

in AKAP signalling complexes permit functional diversity. Nat Cell Biol 7, 1066-

1073.

Hotulainen, P., and Hoogenraad, C.C. (2010). Actin in dendritic spines:

connecting dynamics to function. The Journal of Cell Biology 189, 619-629.

Huganir, R.L., and Nicoll, R.A. (2013a). AMPARs and synaptic plasticity: the

last 25 years. Neuron 80, 704-717.

Huganir, Richard L., and Nicoll, Roger A. (2013b). AMPARs and Synaptic

Plasticity: The Last 25 Years. Neuron 80, 704-717.

Jaubert, P.J., Golub, M.S., Lo, Y.Y., Germann, S.L., Dehoff, M.H., Worley, P.F.,

Kang, S.H., Schwarz, M.K., Seeburg, P.H., and Berman, R.F. (2007). Complex,

multimodal behavioral profile of the Homer1 knockout mouse. Genes, Brain and

Behavior 6, 141-154.

Kaech, S., and Banker, G. (2006). Culturing hippocampal neurons. Nat Protocols

1, 2406-2415.

Kameyama, K., Lee, H.-K., Bear, M.F., and Huganir, R.L. (1998). Involvement of

a Postsynaptic Protein Kinase A Substrate in the Expression of Homosynaptic

Long-Term Depression. Neuron 21, 1163-1175.

Kandel, E. (2012). The molecular biology of memory: cAMP, PKA, CRE, CREB-

1, CREB-2, and CPEB. Molecular Brain 5, 14.

Kar, P., Samanta, K., Kramer, H., Morris, O., Bakowski, D., and Parekh, Anant B.

(2014). Dynamic Assembly of a Membrane Signaling Complex Enables Selective

Activation of NFAT by Orai1. Current Biology 24, 1361-1368.

Katona, I., Acsády, L., and Freund, T.F. (1999). Postsynaptic targets of

somatostatin-immunoreactive interneurons in the rat hippocampus. Neuroscience

88, 37-55.

Keil, J.M., Shen, Z., Briggs, S.P., and Patrick, G.N. (2010). Regulation of STIM1

and SOCE by the Ubiquitin-Proteasome System (UPS). PLoS ONE 5, e13465.

Kessels, H.W., and Malinow, R. (2009). Synaptic AMPA Receptor Plasticity and

Behavior. Neuron 61, 340-350.

110

Kim, E., and Sheng, M. (2009). The postsynaptic density. Current Biology 19,

R723-R724.

Kim, I.H., Racz, B., Wang, H., Burianek, L., Weinberg, R., Yasuda, R., Wetsel,

W.C., and Soderling, S.H. (2013a). Disruption of Arp2/3 Results in Asymmetric

Structural Plasticity of Dendritic Spines and Progressive Synaptic and Behavioral

Abnormalities. The Journal of Neuroscience 33, 6081-6092.

Kim, J., Kwon, J.T., Kim, H.S., and Han, J.H. (2013b). CREB and neuronal

selection for memory trace. Frontiers in neural circuits 7, 44.

Klejman, M.E., Gruszczynska-Biegala, J., Skibinska-Kijek, A., Wisniewska,

M.B., Misztal, K., Blazejczyk, M., Bojarski, L., and Kuznicki, J. (2009).

Expression of STIM1 in brain and puncta-like co-localization of STIM1 and

ORAI1 upon depletion of Ca(2+) store in neurons. Neurochem Int 54, 49-55.

Knafo, S., Gouras, G.K., Yan, X.X., and Spires-Jones, T. (2012). Pathology of

synapses and dendritic spines. Neural Plasticity 2012.

Komatsu, N., Aoki, K., Yamada, M., Yukinaga, H., Fujita, Y., Kamioka, Y., and

Matsuda, M. (2011). Development of an optimized backbone of FRET biosensors

for kinases and GTPases. Mol Biol Cell 22, 4647-4656.

Konur, S., and Yuste, R. (2004). Imaging the motility of dendritic protrusions and

axon terminals: roles in axon sampling and synaptic competition. Molecular and

Cellular Neuroscience 27, 427-440.

Kopec, C.D., Li, B., Wei, W., Boehm, J., and Malinow, R. (2006). Glutamate

receptor exocytosis and spine enlargement during chemically induced long-term

potentiation. Journal of Neuroscience 26, 2000-2009.

Kopec, C.D., Real, E., Kessels, H.W., and Malinow, R. (2007). GluR1 Links

Structural and Functional Plasticity at Excitatory Synapses. The Journal of

Neuroscience 27, 13706-13718.

Korkotian, E., and Segal, M. (2007). Morphological constraints on calcium

dependent glutamate receptor trafficking into individual dendritic spine. Cell

Calcium 42, 41-57.

Korkotian, E., and Segal, M. (2011). Synaptopodin regulates release of calcium

from stores in dendritic spines of cultured hippocampal neurons. The Journal of

Physiology 589, 5987-5995.

Korzeniowski, M.K., Manjarrés, I.M., Varnai, P., and Balla, T. (2010). Activation

of STIM1-Orai1 Involves an Intramolecular Switching Mechanism, Vol 3.

Krapivinsky, G., Krapivinsky, L., Stotz, S.C., Manasian, Y., and Clapham, D.E.

(2011). POST, partner of stromal interaction molecule 1 (STIM1), targets STIM1

to multiple transporters. Proc Natl Acad Sci U S A 108, 19234-19239.

Kwon, H.B., and Sabatini, B.L. (2011). Glutamate induces de novo growth of

functional spines in developing cortex. Nature.

Lalonde, J., Saia, G., and Gill, G. (2014a). Store-Operated Calcium Entry

Promotes the Degradation of the Transcription Factor Sp4 in Resting Neurons,

Vol 7.

Lalonde, J., Saia, G., and Gill, G. (2014b). Store-operated calcium entry promotes

the degradation of the transcription factor Sp4 in resting neurons. Sci Signal 7,

ra51.

111

Lauri, S.E., Bortolotto, Z.A., Nistico, R., Bleakman, D., Ornstein, P.L., Lodge, D.,

Isaac, J.T., and Collingridge, G.L. (2003). A role for Ca2+ stores in kainate

receptor-dependent synaptic facilitation and LTP at mossy fiber synapses in the

hippocampus. Neuron 39, 327-341.

Lee, H.-K., Barbarosie, M., Kameyama, K., Bear, M.F., and Huganir, R.L.

(2000a). Regulation of distinct AMPA receptor phosphorylation sites during

bidirectional synaptic plasticity. Nature 405, 955-959.

Lee, H.-K., Takamiya, K., He, K., Song, L., and Huganir, R.L. (2010a). Specific

roles of AMPA receptor subunit GluR1 (GluA1) phosphorylation sites in

regulating synaptic plasticity in the CA1 region of hippocampus. J Neurophysiol

103, 479-489.

Lee, H.-K., Takamiya, K., He, K., Song, L., and Huganir, R.L. (2010b). Specific

Roles of AMPA Receptor Subunit GluR1 (GluA1) Phosphorylation Sites in

Regulating Synaptic Plasticity in the CA1 Region of Hippocampus, Vol 103.

Lee, H.K., Barbarosie, M., Kameyama, K., Bear, M.F., and Huganir, R.L.

(2000b). Regulation of distinct AMPA receptor phosphorylation sites during

bidirectional synaptic plasticity. Nature 405, 955-959.

Lee, H.K., Takamiya, K., Han, J.S., Man, H., Kim, C.H., Rumbaugh, G., Yu, S.,

Ding, L., He, C., Petralia, R.S., et al. (2003). Phosphorylation of the AMPA

receptor GluR1 subunit is required for synaptic plasticity and retention of spatial

memory. Cell 112, 631-643.

Lefkimmiatis, K., Srikanthan, M., Maiellaro, I., Moyer, M.P., Curci, S., and

Hofer, A.M. (2009). Store-operated cyclic AMP signalling mediated by STIM1.

Nat Cell Biol 11, 433-442.

Leissring, M.A., Akbari, Y., Fanger, C.M., Cahalan, M.D., Mattson, M.P., and

LaFerla, F.M. (2000). Capacitative calcium entry deficits and elevated luminal

calcium content in mutant presenilin-1 knockin mice. Journal of Cell Biology

149, 793-797.

Lewis, R.S. (2007). The molecular choreography of a store-operated calcium

channel. Nature 446, 284-287.

Li, Z., Lu, J., Xu, P., Xie, X., Chen, L., and Xu, T. (2007). Mapping the

Interacting Domains of STIM1 and Orai1 in Ca2+ Release-activated Ca2+

Channel Activation. Journal of Biological Chemistry 282, 29448-29456.

Liou, J., Fivaz, M., Inoue, T., and Meyer, T. (2007). Live-cell imaging reveals

sequential oligomerization and local plasma membrane targeting of stromal

interaction molecule 1 after Ca2+ store depletion. Proc Natl Acad Sci U S A 104,

9301-9306.

Liou, J., Kim, M.L., Heo, W.D., Jones, J.T., Myers, J.W., Ferrell, J.E., Jr., and

Meyer, T. (2005). STIM is a Ca2+ sensor essential for Ca2+-store-depletion-

triggered Ca2+ influx. Curr Biol 15, 1235-1241.

Lois, C., Hong, E.J., Pease, S., Brown, E.J., and Baltimore, D. (2002). Germline

Transmission and Tissue-Specific Expression of Transgenes Delivered by

Lentiviral Vectors. Science 295, 868-872.

Lomo, T. (1966). Frequency potentiation of excitatory synaptic activity in the

dentate area of the hippocampal formation. Acta Physiol Scand 68, 28.

112

Lømo, T. (2003). The discovery of long-term potentiation. Philosophical

Transactions of the Royal Society B: Biological Sciences 358, 617-620.

Longair, M.H., Baker, D.A., and Armstrong, J.D. (2011). Simple neurite tracer:

Open source software for reconstruction, visualization and analysis of neuronal

processes. Bioinformatics 27, 2453-2454.

Lu, Y., Zhang, M., Lim, I.A., Hall, D.D., Allen, M., Medvedeva, Y., McKnight,

G.S., Usachev, Y.M., and Hell, J.W. (2008). AKAP150-anchored PKA activity is

important for LTD during its induction phase. J Physiol 586, 4155-4164.

Luik, R.M., Wang, B., Prakriya, M., Wu, M.M., and Lewis, R.S. (2008).

Oligomerization of STIM1 couples ER calcium depletion to CRAC channel

activation. Nature 454, 538-542.

Luik, R.M., Wu, M.M., Buchanan, J., and Lewis, R.S. (2006). The elementary

unit of store-operated Ca2+ entry: local activation of CRAC channels by STIM1

at ER-plasma membrane junctions. J Cell Biol 174, 815-825.

Luján, R., Roberts, J.D.B., Shigemoto, R., Ohishi, H., and Somogyi, P. (1997).

Differential plasma membrane distribution of metabotropic glutamate receptors

mGluR1α, mGluR2 and mGluR5, relative to neurotransmitter release sites.

Journal of Chemical Neuroanatomy 13, 219-241.

Luthi, A., Wikstrom, M., Palmer, M., Matthews, P., Benke, T., Isaac, J., and

Collingridge, G. (2004). Bi-directional modulation of AMPA receptor unitary

conductance by synaptic activity. BMC Neuroscience 5, 44.

Madden, D.R. (2002). The structure and function of glutamate receptor ion

channels. Nat Rev Neurosci 3, 91-101.

Makino, H., and Malinow, R. (2009). AMPA receptor incorporation into synapses

during LTP: the role of lateral movement and exocytosis. Neuron 64, 381-390.

Makino, Y., Johnson, R.C., Yu, Y., Takamiya, K., and Huganir, R.L. (2011).

Enhanced synaptic plasticity in mice with phosphomimetic mutation of the GluA1

AMPA receptor. Proc Natl Acad Sci U S A 108, 8450-8455.

Man, H.Y., Sekine-Aizawa, Y., and Huganir, R.L. (2007). Regulation of alpha-

amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor trafficking through

PKA phosphorylation of the Glu receptor 1 subunit. Proc Natl Acad Sci U S A

104, 3579-3584.

Martin, S.J., Grimwood, P.D., and Morris, R.G.M. (2000). Synaptic Plasticity and

Memory: An Evaluation of the Hypothesis. Annual Review of Neuroscience 23,

649-711.

Martin, S.J., and Morris, R.G.M. (2002). New life in an old idea: The synaptic

plasticity and memory hypothesis revisited. Hippocampus 12, 609-636.

Mateos, J.M., Lüthi, A., Savic, N., Stierli, B., Streit, P., Gähwiler, B.H., and

McKinney, R.A. (2007). Synaptic modifications at the CA3–CA1 synapse after

chronic AMPA receptor blockade in rat hippocampal slices. The Journal of

Physiology 581, 129-138.

Matsuzaki, M., Ellis-Davies, G.C.R., Nemoto, T., Miyashita, Y., Iino, M., and

Kasai, H. (2001). Dendritic spine geometry is critical for AMPA receptor

expression in hippocampal CA1 pyramidal neurons. Nat Neurosci 4, 1086-1092.

Matsuzaki, M., Honkura, N., Ellis-Davies, G.C., and Kasai, H. (2004a). Structural

basis of long-term potentiation in single dendritic spines. Nature 429, 761-766.

113

Matsuzaki, M., Honkura, N., Ellis-Davies, G.C.R., and Kasai, H. (2004b).

Structural basis of long-term potentiation in single dendritic spines. Nature 429,

761-766.

Mattson, M.P., LaFerla, F.M., Chan, S.L., Leissring, M.A., Shepel, P.N., and

Geiger, J.D. (2000). Calcium signaling in the ER: its role in neuronal plasticity

and neurodegenerative disorders. Trends Neurosci 23, 222-229.

Mayer, M.L. (2005). Glutamate receptor ion channels. Current Opinion in

Neurobiology 15, 282-288.

McGaugh, J.L. (2000). Memory--a Century of Consolidation. Science 287, 248-

251.

McKinney, R.A. (2010). Excitatory amino acid involvement in dendritic spine

formation, maintenance and remodelling. J Physiol 588, 107-116.

Meldolesi, J., and Pozzan, T. The endoplasmic reticulum Ca2+ store: a view from

the lumen. Trends in Biochemical Sciences 23, 10-14.

Miesenbock, G., De Angelis, D.A., and Rothman, J.E. (1998). Visualizing

secretion and synaptic transmission with pH-sensitive green fluorescent proteins.

Nature 394, 192-195.

Mizuno, T., Kanazawa, I., and Sakurai, M. (2001). Differential induction of LTP

and LTD is not determined solely by instantaneous calcium concentration: an

essential involvement of a temporal factor. European Journal of Neuroscience 14,

701-708.

Molnár, E. (2011). Long-term potentiation in cultured hippocampal neurons.

Seminars in Cell & Developmental Biology 22, 506-513.

Mulkey, R., Herron, C., and Malenka, R. (1993). An essential role for protein

phosphatases in hippocampal long-term depression. Science 261, 1051-1055.

Nabavi, S., Fox, R., Proulx, C.D., Lin, J.Y., Tsien, R.Y., and Malinow, R. (2014).

Engineering a memory with LTD and LTP. Nature advance online publication.

Nicoll, R.A., and Malenka, R.C. (1999). Expression mechanisms underlying

NMDA receptor-dependent long-term potentiation. In Annals of the New York

Academy of Sciences, pp. 515-525.

Nimchinsky, E.A., Sabatini, B.L., and Svoboda, K. (2002). Structure and function

of dendritic spines, pp. 313-353.

Nishiyama, M., Hoshino, A., Tsai, L., Henley, J.R., Goshima, Y., Tessier-

Lavigne, M., Poo, M.M., and Hong, K. (2003). Cyclic AMP/GMP-dependent

modulation of Ca2+ channels sets the polarity of nerve growth-cone turning.

Nature 423, 990-995.

Nusser, Z., Lujan, R., Laube, G., Roberts, J.D., Molnar, E., and Somogyi, P.

(1998). Cell type and pathway dependence of synaptic AMPA receptor number

and variability in the hippocampus. Neuron 21, 545-559.

O'Brien, J. (2014). The ever-changing electrical synapse. Current Opinion in

Neurobiology 29, 64-72.

Oertner, T.G., and Matus, A. (2005). Calcium regulation of actin dynamics in

dendritic spines. Cell Calcium 37, 477-482.

Oh-Hora, M., Yamashita, M., Hogan, P.G., Sharma, S., Lamperti, E., Chung, W.,

Prakriya, M., Feske, S., and Rao, A. (2008). Dual functions for the endoplasmic

114

reticulum calcium sensors STIM1 and STIM2 in T cell activation and tolerance.

Nat Immunol 9, 432-443.

Oh, M.C., Derkach, V.A., Guire, E.S., and Soderling, T.R. (2006). Extrasynaptic

membrane trafficking regulated by GluR1 serine 845 phosphorylation primes

AMPA receptors for long-term potentiation. J Biol Chem 281, 752-758.

Okabe, S. (2007). Molecular anatomy of the postsynaptic density. Molecular and

Cellular Neuroscience 34, 503-518.

Opazo, P., and Choquet, D. (2011). A three-step model for the synaptic

recruitment of AMPA receptors. Molecular and Cellular Neuroscience 46, 1-8.

Otmakhov, N., Khibnik, L., Otmakhova, N., Carpenter, S., Riahi, S., Asrican, B.,

and Lisman, J. (2004a). Forskolin-induced LTP in the CA1 hippocampal region is

NMDA receptor dependent. J Neurophysiol 91, 1955-1962.

Otmakhov, N., Khibnik, L., Otmakhova, N., Carpenter, S., Riahi, S., Asrican, B.,

and Lisman, J. (2004b). Forskolin-Induced LTP in the CA1 Hippocampal Region

Is NMDA Receptor Dependent, Vol 91.

Park, C.Y., Hoover, P.J., Mullins, F.M., Bachhawat, P., Covington, E.D.,

Raunser, S., Walz, T., Garcia, K.C., Dolmetsch, R.E., and Lewis, R.S. (2009).

STIM1 Clusters and Activates CRAC Channels via Direct Binding of a Cytosolic

Domain to Orai1. Cell 136, 876-890.

Park, C.Y., Shcheglovitov, A., and Dolmetsch, R. (2010). The CRAC channel

activator STIM1 binds and inhibits L-type voltage-gated calcium channels.

Science 330, 101-105.

Parvez, S., Beck, A., Peinelt, C., Soboloff, J., Lis, A., Monteilh-Zoller, M., Gill,

D.L., Fleig, A., and Penner, R. (2008). STIM2 protein mediates distinct store-

dependent and store-independent modes of CRAC channel activation. The

FASEB Journal 22, 752-761.

Passafaro, M., Nakagawa, T., Sala, C., and Sheng, M. (2003). Induction of

dendritic spines by an extracellular domain of AMPA receptor subunit GluR2.

Nature 424, 677-681.

Penna, A., Demuro, A., Yeromin, A.V., Zhang, S.L., Safrina, O., Parker, I., and

Cahalan, M.D. (2008). The CRAC channel consists of a tetramer formed by Stim-

induced dimerization of Orai dimers. Nature 456, 116-120.

Penzes, P., Cahill, M.E., Jones, K.A., Vanleeuwen, J.E., and Woolfrey, K.M.

(2011). Dendritic spine pathology in neuropsychiatric disorders. Nature


Pereda, A.E. (2014). Electrical synapses and their functional interactions with

chemical synapses. Nature Reviews Neuroscience 15, 250-263.

Picard, C., McCarl, C.A., Papolos, A., Khalil, S., LÃ¼thy, K., Hivroz, C.,

LeDeist, F., Rieux-Laucat, F., Rechavi, G., Rao, A., et al. (2009). STIM1

mutation associated with a syndrome of immunodeficiency and autoimmunity.

New England Journal of Medicine 360, 1971-1980.

Poncer, J.C., Esteban, J.A., and Malinow, R. (2002). Multiple Mechanisms for the

Potentiation of AMPA Receptor-Mediated Transmission by α-Ca2+/Calmodulin-

Dependent Protein Kinase II. The Journal of Neuroscience 22, 4406-4411.

115

Porter, K.R., and Palade, G.E. (1957). Studies on the endoplasmic reticulum. III.

Its form and distribution in striated muscle cells. J Biophys Biochem Cytol 3,

269-300.

Potter, J.D., Piascik, M.T., Wisler, P.L., Robertson, S.P., and Johnson, C.L.

(1980). CALCIUM DEPENDENT REGULATION OF BRAIN AND CARDIAC

MUSCLE ADENYLATE CYCLASE*. Annals of the New York Academy of

Sciences 356, 220-231.

Power, J.M., and Sah, P. (2007). Distribution of IP(3)-mediated calcium

responses and their role in nuclear signalling in rat basolateral amygdala neurons.

The Journal of Physiology 580, 835-857.

Purves, D., Augustine, G.J., Fitzpatrick, D., Katz, L.C., LaMantia, A.-S.,

McNamara, J.O., and Williams, S.M. (2001). Neuroscience (2nd Edition)

(Sinauer Associates; 2001

).

Putney, J.W., Jr. (2007). Recent breakthroughs in the molecular mechanism of

capacitative calcium entry (with thoughts on how we got here). Cell Calcium 42,

103-110.

Rasband, W.S. (1997-2014) (U. S. National Institutes of Health, Bethesda,

Maryland, USA).

Redman, S. (1990). Quantal analysis of synaptic potentials in neurons of the

central nervous system, Vol 70.

Richards, D.A., Mateos, J.M., Hugel, S., de Paola, V., Caroni, P., Gähwiler, B.H.,

and McKinney, R.A. (2005). Glutamate induces the rapid formation of spine head

protrusions in hippocampal slice cultures. Proceedings of the National Academy

of Sciences 102, 6166-6171.

Robertson, H.R., Gibson, E.S., Benke, T.A., and Dell'Acqua, M.L. (2009).

Regulation of postsynaptic structure and function by an A-kinase anchoring

protein-membrane-associated guanylate kinase scaffolding complex. J Neurosci

29, 7929-7943.

Rodriguez, A., Ehlenberger, D.B., Dickstein, D.L., Hof, P.R., and Wearne, S.L.

(2008). Automated three-dimensional detection and shape classification of

dendritic spines from fluorescence microscopy images. PLoS ONE 3.

Roos, J., DiGregorio, P.J., Yeromin, A.V., Ohlsen, K., Lioudyno, M., Zhang, S.,

Safrina, O., Kozak, J.A., Wagner, S.L., Cahalan, M.D., et al. (2005). STIM1, an

essential and conserved component of store-operated Ca2+ channel function. J

Cell Biol 169, 435-445.

Rose, C.R., and Konnerth, A. (2001). Stores Not Just for Storage: Intracellular

Calcium Release and Synaptic Plasticity. Neuron 31, 519-522.

Rosenbluth, J. (1962). Subsurface cisterns and their relationship to the neuronal

plasma membrane. J Cell Biol 13, 405-421.

Rossi, A.M., Tovey, S.C., Rahman, T., Prole, D.L., and Taylor, C.W. (2012).

Analysis of IP3 receptors in and out of cells. Biochimica et Biophysica Acta

(BBA) - General Subjects 1820, 1214-1227.

Saitoh, N., Oritani, K., Saito, K., Yokota, T., Ichii, M., Sudo, T., Fujita, N.,

Nakajima, K., Okada, M., and Kanakura, Y. (2011). Identification of functional

116

domains and novel binding partners of STIM proteins. Journal of Cellular

Biochemistry 112, 147-156.

Sakamoto, K., Karelina, K., and Obrietan, K. (2011). CREB: A multifaceted

regulator of neuronal plasticity and protection. Journal of Neurochemistry 116, 1-

9.

Salminen, A., Kauppinen, A., Suuronen, T., Kaarniranta, K., and Ojala, J. (2009).

ER stress in Alzheimer's disease: a novel neuronal trigger for inflammation and

Alzheimer's pathology. Journal of Neuroinflammation 6, 41-41.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A

Laboratory Manual. (New York: Cold Spring Harbor Laboratory Press).

Sans, N., Racca, C., Petralia, R.S., Wang, Y.-X., McCallum, J., and Wenthold,

R.J. (2001). Synapse-Associated Protein 97 Selectively Associates with a Subset

of AMPA Receptors Early in their Biosynthetic Pathway. The Journal of


Scheuss, V., and Bonhoeffer, T. (2013). Function of Dendritic Spines on

Hippocampal Inhibitory Neurons. Cerebral Cortex.

Segal, M. (2010). Dendritic spines, synaptic plasticity and neuronal survival:

activity shapes dendritic spines to enhance neuronal viability. European Journal of


Segal, M., and Korkotian, E. (2014). Endoplasmic reticulum calcium stores in

dendritic spines. Frontiers in Neuroanatomy 8.

Segal, M., Vlachos, A., and Korkotian, E. (2010). The spine apparatus,

synaptopodin, and dendritic spine plasticity. Neuroscientist 16, 125-131.

Shepherd, J.D., and Huganir, R.L. (2007). The Cell Biology of Synaptic

Plasticity: AMPA Receptor Trafficking. Annual Review of Cell and

Developmental Biology 23, 613-643.

Shi, S.-H., Hayashi, Y., Petralia, R.S., Zaman, S.H., Wenthold, R.J., Svoboda, K.,

and Malinow, R. (1999). Rapid Spine Delivery and Redistribution of AMPA

Receptors After Synaptic NMDA Receptor Activation. Science 284, 1811-1816.

Skibinska-Kijek, A., Wisniewska, M.B., Gruszczynska-Biegala, J., Methner, A.,

and Kuznicki, J. (2009a). Immunolocalization of STIM1 in the mouse brain. Acta

Neurobiol Exp (Wars) 69, 413-428.

Skibinska-Kijek, A., Wisniewska, M.B., Gruszczynska-Biegala, J., Methner, A.,

and Kuznicki, J. (2009b). Immunolocalization of STIM1 in the mouse brain. Acta

Neurobiologiae Experimentalis 69, 413-428.

Skroblin, P., Grossmann, S., Schafer, G., Rosenthal, W., and Klussmann, E.

(2010). Mechanisms of protein kinase A anchoring. Int Rev Cell Mol Biol 283,

235-330.

Smith, D.L., Pozueta, J., Gong, B., Arancio, O., and Shelanski, M. (2009).

Reversal of long-term dendritic spine alterations in Alzheimer disease models.

Proc Natl Acad Sci U S A 106, 16877-16882.

Smith, I.F., Boyle, J.P., Vaughan, P.F.T., Pearson, H.A., Cowburn, R.F., and

Peers, C.S. (2002). Ca2+ stores and capacitative Ca2+ entry in human

neuroblastoma (SH-SY5Y) cells expressing a familial Alzheimer's disease

presenilin-1 mutation. Brain Research 949, 105-111.

117

Smith, K.E., Gibson, E.S., and Dell'Acqua, M.L. (2006a). cAMP-dependent

protein kinase postsynaptic localization regulated by NMDA receptor activation

through translocation of an A-kinase anchoring protein scaffold protein. J

Neurosci 26, 2391-2402.

Smith, K.E., Gibson, E.S., and Dell’Acqua, M.L. (2006b). cAMP-Dependent

Protein Kinase Postsynaptic Localization Regulated by NMDA Receptor

Activation through Translocation of an A-Kinase Anchoring Protein Scaffold

Protein. The Journal of Neuroscience 26, 2391-2402.

Soboloff, J., Rothberg, B.S., Madesh, M., and Gill, D.L. (2012). STIM proteins:

dynamic calcium signal transducers. Nat Rev Mol Cell Biol 13, 549-565.

Soboloff, J., Spassova, M., Dziadek, M., and Gill, D. (2006a). Calcium signals

mediated by STIM and Orai proteins - a new paradigm in inter-organelle

communication. Biochim Biophys Acta 1763, 1161 - 1168.

Soboloff, J., Spassova, M.A., Hewavitharana, T., He, L.-P., Xu, W., Johnstone,

L.S., Dziadek, M.A., and Gill, D.L. (2006b). STIM2 Is an Inhibitor of STIM1-

Mediated Store-Operated Ca2+ Entry. Current Biology 16, 1465-1470.

Song, I., and Huganir, R.L. (2002). Regulation of AMPA receptors during

synaptic plasticity. Trends in Neurosciences 25, 578-588.

Sorra, K.E., and Harris, K.M. (2000). Overview on the structure, composition,

function, development, and plasticity of hippocampal dendritic spines.

Hippocampus 10, 501-511.

Stathopulos, P.B., Zheng, L., and Ikura, M. (2009). Stromal Interaction Molecule

(STIM) 1 and STIM2 Calcium Sensing Regions Exhibit Distinct Unfolding and

Oligomerization Kinetics. Journal of Biological Chemistry 284, 728-732.

Stevens, C.F. (1993). Quantal release of neurotransmitter and long-term

potentiation. Cell 72, 55-63.

Stevens, C.F. (1998). A Million Dollar Question: Does LTP = Memory? Neuron

20, 1-2.

Stuchlik, A. (2014). Dynamic learning and memory, synaptic plasticity and

neurogenesis: An update. Frontiers in Behavioral Neuroscience 8.

Sun, S., Zhang, H., Liu, J., Popugaeva, E., Xu, N.J., Feske, S., White, C.L., 3rd,

and Bezprozvanny, I. (2014). Reduced Synaptic STIM2 Expression and Impaired

Store-Operated Calcium Entry Cause Destabilization of Mature Spines in Mutant

Presenilin Mice. Neuron 82, 79-93.

Tabuchi, A. (2008). Synaptic Plasticity-Regulated Gene Expression: a Key Event

in the Long-Lasting Changes of Neuronal Function. Biological and

Pharmaceutical Bulletin 31, 327-335.

Tang, F., and Kalil, K. (2005). Netrin-1 induces axon branching in developing

cortical neurons by frequency-dependent calcium signaling pathways. J Neurosci

25, 6702-6715.

Tavalin, S.J., Colledge, M., Hell, J.W., Langeberg, L.K., Huganir, R.L., and Scott,

J.D. (2002). Regulation of GluR1 by the A-Kinase Anchoring Protein 79

(AKAP79) Signaling Complex Shares Properties with Long-Term Depression.

The Journal of Neuroscience 22, 3044-3051.

118

Taylor, C.W. (1998). Inositol trisphosphate receptors: Ca2+-modulated

intracellular Ca2+ channels. Biochimica et Biophysica Acta (BBA) - Molecular

and Cell Biology of Lipids 1436, 19-33.

Thevathasan, J.V., Tan, E., Zheng, H., Lin, Y.C., Li, Y., Inoue, T., and Fivaz, M.

(2013). The small GTPase HRas shapes local PI3K signals through positive

feedback and regulates persistent membrane extension in migrating fibroblasts.

Mol Biol Cell 24, 2228-2237.

Tian, G., Tepikin, A.V., Tengholm, A., and Gylfe, E. (2012). cAMP induces

stromal interaction molecule 1 (STIM1) puncta but neither Orai1 protein

clustering nor store-operated Ca2+ entry (SOCE) in islet cells. J Biol Chem 287,

9862-9872.

Tiscornia, G., Singer, O., and Verma, I.M. (2006). Production and purification of

lentiviral vectors. Nat Protoc 1, 241-245.

Tully, K., and Treistman, S.N. (2004). Distinct Intracellular Calcium Profiles

Following Influx Through N- Versus L-Type Calcium Channels: Role of Ca2+-

Induced Ca2+ Release, Vol 92.

Valenzuela, J.I., Jaureguiberry-Bravo, M., and Couve, A. (2011). Neuronal

protein trafficking: Emerging consequences of endoplasmic reticulum dynamics.

Molecular and Cellular Neuroscience 48, 269-277.

Valtschanoff, J.G., Burette, A., Davare, M.A., Soren Leonard, A., Hell, J.W., and

Weinberg, R.J. (2000). SAP97 concentrates at the postsynaptic density in cerebral

cortex. European Journal of Neuroscience 12, 3605-3614.

van Spronsen, M., and Hoogenraad, C.C. Synapse pathology in psychiatric and

neurologic disease. Curr Neurol Neurosci Rep 10, 207-214.

Van Spronsen, M., and Hoogenraad, C.C. (2010). Synapse pathology in

psychiatric and neurologic disease. Current Neurology and Neuroscience Reports

10, 207-214.

Verkhratsky, A., and Shmigol, A. (1996). Calcium-induced calcium release in

neurones. Cell Calcium 19, 1-14.

Waltereit, R., and Weller, M. (2003). Signaling from cAMP/PKA to MAPK and

synaptic plasticity. Mol Neurobiol 27, 99-106.

Wang, G.X., and Poo, M.M. (2005). Requirement of TRPC channels in netrin-1-

induced chemotropic turning of nerve growth cones. Nature 434, 898-904.

Wang, H., and Zhang, M. (2012). The role of Ca 2+-stimulated adenylyl cyclases

in bidirectional synaptic plasticity and brain function. Reviews in the

Neurosciences 23, 67-78.

Wang, X., Wang, Y., Zhou, Y., Hendron, E., Mancarella, S., Andrake, M.D.,

Rothberg, B.S., Soboloff, J., and Gill, D.L. (2014). Distinct Orai-coupling

domains in STIM1 and STIM2 define the Orai-activating site. Nature

communications 5, 3183.

Wang, Y., Deng, X., Mancarella, S., Hendron, E., Eguchi, S., Soboloff, J., Tang,

X.D., and Gill, D.L. (2010a). The calcium store sensor, STIM1, reciprocally

controls Orai and Ca V1.2 channels. Science 330, 105-109.

Wang, Y., Deng, X., Mancarella, S., Hendron, E., Eguchi, S., Soboloff, J., Tang,

X.D., and Gill, D.L. (2010b). The calcium store sensor, STIM1, reciprocally

controls Orai and CaV1.2 channels. Science 330, 105-109.

119

Wen, Z., Guirland, C., Ming, G.L., and Zheng, J.Q. (2004). A

CaMKII/calcineurin switch controls the direction of Ca(2+)-dependent growth

cone guidance. Neuron 43, 835-846.

Williams, R.T., Manji, S.S., Parker, N.J., Hancock, M.S., Van Stekelenburg, L.,

Eid, J.P., Senior, P.V., Kazenwadel, J.S., Shandala, T., Saint, R., et al. (2001).

Identification and characterization of the STIM (stromal interaction molecule)

gene family: coding for a novel class of transmembrane proteins. Biochem J 357,

673-685.

Williams, R.T., Senior, P.V., Van Stekelenburg, L., Layton, J.E., Smith, P.J., and

Dziadek, M.A. (2002). Stromal interaction molecule 1 (STIM1), a transmembrane

protein with growth suppressor activity, contains an extracellular SAM domain

modified by N-linked glycosylation. Biochimica et Biophysica Acta (BBA) -

Protein Structure and Molecular Enzymology 1596, 131-137.

Wong, L.C., Lu, B., Tan, K.W., and Fivaz, M. (2010). Fully-automated image

processing software to analyze calcium traces in populations of single cells. Cell

Calcium 48, 270-274.

Wright, A.L., and Vissel, B. (2012). The essential role of AMPA receptor GluR2

subunit RNA editing in the normal and diseased brain. Frontiers in Molecular

Neuroscience 5.

Wu, M.M., Buchanan, J., Luik, R.M., and Lewis, R.S. (2006). Ca2+ store

depletion causes STIM1 to accumulate in ER regions closely associated with the

plasma membrane. J Cell Biol 174, 803-813.

Xiao, B., Cheng Tu, J., and Worley, P.F. (2000). Homer: a link between neural

activity and glutamate receptor function. Current Opinion in Neurobiology 10,

370-374.

Xiao, B., Coste, B., Mathur, J., and Patapoutian, A. (2011). Temperature-

dependent STIM1 activation induces Ca(2)+ influx and modulates gene

expression. Nat Chem Biol 7, 351-358.

Xu, P., Lu, J., Li, Z., Yu, X., Chen, L., and Xu, T. (2006). Aggregation of STIM1

underneath the plasma membrane induces clustering of Orai1. Biochem Biophys

Res Commun 350, 969-976.

Yamashita, T., and Kwak, S. (2014). The molecular link between inefficient

GluA2 Q/R site-RNA editing and TDP-43 pathology in motor neurons of sporadic

amyotrophic lateral sclerosis patients. Brain Research 1584, 28-38.

Yang, X., Jin, H., Cai, X., Li, S., and Shen, Y. (2012). Structural and mechanistic

insights into the activation of Stromal interaction molecule 1 (STIM1).

Proceedings of the National Academy of Sciences 109, 5657-5662.

Yoshihara, Y., De Roo, M., and Muller, D. (2009). Dendritic spine formation and

stabilization. Current Opinion in Neurobiology 19, 146-153.

Yu, F., Sun, L., Courjaret, R., and Machaca, K. (2011). Role of the STIM1 C-

terminal domain in STIM1 clustering. Journal of Biological Chemistry.

Yuan, J.P., Zeng, W., Dorwart, M.R., Choi, Y.-J., Worley, P.F., and Muallem, S.

(2009). SOAR and the polybasic STIM1 domains gate and regulate Orai channels.

Nat Cell Biol 11, 337-343.

Yuste, R. (2011). Dendritic spines and distributed circuits. Neuron 71, 772-781.

120

Yuste, R., and Bonhoeffer, T. (2004). Genesis of dendritic spines: insights from

ultrastructural and imaging studies. Nat Rev Neurosci 5, 24-34.

Zhang, J., Ma, Y., Taylor, S.S., and Tsien, R.Y. (2001). Genetically encoded

reporters of protein kinase A activity reveal impact of substrate tethering.

Proceedings of the National Academy of Sciences 98, 14997-15002.

Zheng, L., Stathopulos, P.B., Schindl, R., Li, G.-Y., Romanin, C., and Ikura, M.

(2011). Auto-inhibitory role of the EF-SAM domain of STIM proteins in store-

operated calcium entry. Proceedings of the National Academy of Sciences 108,

1337-1342.

Zhou, Y., Mancarella, S., Wang, Y., Yue, C., Ritchie, M., Gill, D.L., and

Soboloff, J. (2009). The Short N-terminal Domains of STIM1 and STIM2 Control

the Activation Kinetics of Orai1 Channels. Journal of Biological Chemistry 284,

19164-19168.

Ziff, E.B. (2007). TARPs and the AMPA Receptor Trafficking Paradox. Neuron

53, 627-633.

Documents

A novel role for the ER-calcium sensor STIM2 in regulating … · 2018. 1. 8. · A novel role for the ER-calcium sensor STIM2 in regulating dendritic spine morphogenesis and AMPA