The Characterization of G-protein Coupled Receptors in
Isolated Rat Dorsal Root Ganglion Cells
YEUNG, Barry Ho Sing
A Thesis Submitted in Partial Fulfillment
of the Requirements for the Degree of
Master of Philosophy
in
Pharmacology
I The ChineseSeptembe Universitr y201 of1 Hon g Kong |1 I e SEP 2ot^ij WrwiiiTY 一"/M Thesis/Assessment Committee
Professor John A. RUDD (Chair)
Professor Helen WISE (Thesis Supervisor)
Professor Wing Tai CHEUNG (Committee Member)
Professor Daisy K.Y. SHUM (External Examiner)
I Abstract
Abstract of thesis entitled:
The Characterization of G-protein Coupled Receptors in Isolated Rat Dorsal Root Ganglion Cells
Submitted by Barry Ho Sing YEUNG
For the degree of Master of Philosophy in Pharmacology
At The Chinese University of Hong Kong in September 2011
Sensory stimuli are detected by primary sensory neurons whose cell bodies reside in the dorsal root ganglia (DRG) of the peripheral nervous system (PNS). In the DRG， glial cells, including satellite glial cells (SGCs) and Schwann cells, are closely associated with the primary sensory neurons. The primary functions of glial cells are to regulate the trafficking of extracellular ions and neurotransmitters for proper neuronal function and survival. In the early study of pain, DRG were examined and
sensory neurons were presumed to be the sole players leading to the development of
pain while the associated glial cells were thought not to be directly involved.
Recently, however, glial cells have emerged as important contributors to the
maintenance and amplification of pain by bidirectional neuron-glia interactions.'
i In isolated DRG cell cultures, neurons and glial cells in a mixed cell population can be enriched and studied separately. We have recently characterized the functional expression of pain-related prostanoid receptors (EP4 and IP receptors) in isolated cultures of DRG neurons and glial cells using pharmacological methods. Besides, the prostanoid receptor agonist-stimulated responses in glial cells were curiously
inhibited by co-culture with neurons revealing a novel neuron-glia interaction. To
investigate the effect of neurons on glial cells in isolated cultures, the phenotypic properties of different glial cell subtypes were first studied. Antibodies against glial
fibrillary acidic protein (GFAP) identified two major types of glial cells, one
population had a round nucleus with spreading cytoplasm, and the other population
was bipolar or multipolar and spindle shaped with an elongated nucleus. When DRG
cells were cultured over time, glial cells proliferated and the number of GFAP-
immunoreactive glial cells increased. These characteristics have also been reported in
neuropathic pain conditions in vivo which may suggest that isolated glial cells
behaved similarly to those glial cells in neuropathic pain models. The identification
of glial cell subtypes and fibroblasts using glial cell-specific antibodies was
inconclusive and adaptation of other immunocytochemical methods did not facilitate
their identification.
The second part of the study was to characterize the expression of other pain-related
ii Gs- and Gj/o-coupled G-protein coupled receptors (GPCRs) on neurons and glial cells, and to investigate if the receptor-specific activities of glial cells were also inhibited by co-culture with neurons. Results revealed the presence of P2-adrenoceptors (P2-
ARs) and calcitonin gene-related peptide (CGRP) receptors, but not pi-adrenoceptors
(Pi-ARs), on neurons and glial cells. Consistent with the EP4 and IP receptor agonist- stimulated responses, the CGRP-stimulated responses in mixed DRG cells were smaller than those in the glial cells. This inhibitory pattern was not seen with P2-AR agonist stimulation. Characterization of Gj/o-coupled GPCRs was prevented due to the failure of Gj/�agonist tso inhibit forskolin-stimulated responses. With all agonist and forskolin stimulations, neuron-enriched cells produced smaller responses than mixed DRG cells and glial cells. Addition of nerve growth factor (NGF) did not enhance neuronal responses, despite facilitating neurite outgrowth of neurons. With strong evidence that the Gi/o-coupled GPCR should be expressed on DRG cells, other methods need to be identified in order to characterize the presence of these receptors in DRG neurons and glial cells.
In conclusion, this thesis suggested that DRG glial cells in isolated cultures may mimic neuropathic pain conditions after nerve injury in vivo, and revealed the necessity to consider the roles played by the associated glial cells and the neuron-glia interactions during investigations using DRG cells.
iii 論文摘要
外周神經糸統中（PNS)，背根神經節(DRG)包含主感覺細胞體，負責偵測感官
刺激。而DRG內的神經膠質細胞，包括衛星膠質細胞（SGCs)和施旺細胞，則
緊密地聯繫著主感覺神經元°神經膠質細胞的主要作用，是調節神經元的外離
子和神經傳送素的流量，使神經元得以正常運作和生存。早期關於疼痛的
DRG硏究指出，感覺神經元是產生疼痛感覺的唯一原素，而聯繫著感覺神經
元的神經膠質細胞，則對疼痛感覺的產生過程沒有直接影響。直'至最近，神經
膠質細胞被視為產生疼痛感覺的重要原素，因為神經元和神經膠質細胞相互作
用可維持及增加疼痛的感覺。
在已分離的DRG細胞培養內，DRG混合細胞群中的感覺神經元和神經膠質細
胞可以分別被豐富和硏究。最近，我們透過藥理學方法，揭示了在已分離的
DRG感覺神經元和神經膠質細胞培養中，與疼痛相關並具功能的前列腺素受
體(EP4和IP受體）。另外，當神經膠質細胞與神經元共同培養的時候，神經元
奇異地抑制了神經膠質細胞被前列腺素受體激動劑刺激後的反應，這揭示了新
的神經元和神經膠質細胞相互作用。為了探究在分離培養內神經元對神經膠質
細胞的影響，我們得先探究不同神經膠質細胞的子類型的生理特性。利用能夠
辨識神經膠質原纖維酸性蛋白（GFAP)的抗體，我們在細胞培養內識別出兩類
型的神經膠質細胞：第一類型的神經膠質細胞擁有圓形的細胞核和展開的_胞
質；第二類型的神經膠質細胞擁有瘦長的細胞核，以及屬兩極或多極的妨錘形
iv 細胞質。隨著培養DRG細胞的時間增加，神經膠質細胞的數目，以及GFAP
螢光訊號的強度皆有增長。以上特徵亦見於神經病情況，表示在分離狀態的神
經膠質細胞和在活體內神經病狀態的神經膠質細胞，兩者表現相似。至現階段，
神經膠質細胞抗體仍未能成功識別神經膠質細胞的子類型和成纖維細胞，至於
採用其他免疫細胞學的方法，亦未見對細胞識別有任何幫助。
此硏究的第二部分，是要在神經元和神經膠質細胞上，揭示其他與疼痛相關的
Gs偶聯和Gi/�偶聯的G蛋白偶聯受體（GPCRs)，再探究神經元的存在，會否
抑制神經膠質細胞的受體特異性活動。結果顯示，在神經元和神經膠質細胞上，
可找到卩2臂上腺素受體（P2-ARS)和降血韩素基因相關縮氣酸（CGRP)受體，
但沒有Pi臂上腺素受體(pi-ARs)。另外，比較發現，細胞被CGRP剌激後，
混合DRG細胞的神經膠質細胞的反應比單一神經膠質細胞的反應大，這顯示
與EP4和IP受體激動劑的刺激反應結果一致。相反，當p2-AR激動劑分別刺
激混合DRG細胞和神經膠質細胞時，卻沒有發現相近的抑制效果。綜合各Gs-
GPCR激動劑的剌激測試結果，我們發現神經元富集細胞受刺激後的反應，較
混合DRG細胞和神經膠質細胞受剌激後的反應為細。另外，我們無法揭示Gi/�
偶聯的GPCRs，因為激動劑的刺激測試結果未能抑制細胞受福斯高林
(forskolin)剌激後的反應。而在細胞中加入神經生長因素（NGF)之後，結果
雖然未能增加神經元富集細胞受剌激後的反應，但卻促進了神經元突軸的$長。
雖然現在存有理據證明在DRG細胞上可找到Gi/�偶聯的GPCRs，但我們還需
X 使用其他測試方法來確認GI/�偶聯的GPCRS在DRG神經元和神經膠質細胞上
的存在。
‘ 總括而言，此論文提出在分離培養中的DRG神經膠質細胞，能模擬在活體內
神經受傷後的神經病情況。另外，也揭示了在硏究DRG的同時，應考慮關聯
神經膠質細胞，以及神經元和神經膠質細胞相互作用的角色。
I
vi Acknowledgements
I consider my thesis incomplete without this acknowledgement section. It is my
pleasure to give thanks to a few people who made this thesis possible. First and
” foremost, I would like to express my deepest gratitude and appreciation to my
supervisor, Professor Helen Wise, for her continuous guidance and support of my
study. I appreciate her patience, enthusiasm and immense knowledge, which made
my MPhil experience inspiring and joyful. She is such a perfect listener and advisor
that I just happened to share all my ups and downs with her. Not only has she given
me professional guidance on scientific research, but she has also given me personal
guidance on life in general.
Besides, I would like to thank the rest of my thesis committee members: Professor
John Rudd, Professor WT Cheung and Professor Daisy Shum for reading my thesis
and giving insightful suggestions, comments and questions on it.
I would also like to say a big "Thank You" to all my laboratory colleagues, including
Mr. Kevin Chow, Dr. Johnson Ng and Mr. Terence Chu, for all the ideas and
suggestions that came throughout our work.
My time at CUHK was made enjoyable in large part due to my buddies at the former
Department of Pharmacology: Ms. Stella Chai, Ms. Janina Efpunkt, Dr. Dave Ho, Dr.
Winnie Kan, Ms. Stephanie Knapp, Dr. Jessica Law, Ms. Joyce Lee, Mr. Thomas
Pffeifer, Ms. Christina Poon, Mr. Dirk Saalfrank... Thanks for all the fun that you all
have brought me in the last two years.
I gratefully acknowledge the research funds from the Graduate School of CUHK, the
overseas conference grants from the School of Biomedical Sciences of CUHK, and
the Hong Kong Pharmacology Society. Special thanks to ONO Pharmaceuticals
vii (Japan) for providing EP-specific ligands and Schering AG (Germany) for providing cicaprost for this study.
Last but not least，I would like to thank my family, especially my parents, for all their unflagging love, support and encouragement throughout my life. Thanks also to Ms.
Mimi Chan and Mr. Paul Wong for their advice and assistance during the thesis process. Finally，I express my profound appreciation to my girlfriend, Ms. Joey
Wong, for her professional English-to-Chinese abstract translation and thesis proofreading effort. Her love, care, patience, understanding, encouragement and, above all, super delicious food have made my day.
Barry HS YEUNG (MPhil) The Chinese University of Hong Kong Sept 2011
I
viii Publications based on work in this thesis
Article
Ng KY, Yeung B, Wong YH，Wise H (2011) • Dorsal root ganglion neurons inhibit adenylyl cyclase activity in associated glial cells (Manuscript in preparation)
Oral presentations
Ng KY, Yeung B, Wong YH, Wise H (2010) Neurons regulate Gg-coupled GPCRs on glial cells in adult rat dorsal root ganglion cell cultures The International Conference of Pharmacology 一 The 3��Mainland Taiwa, n and Hong Kong Symposium of Pharmacology, Shenyang, China
Yeung B, Ng KY, Wong YH, Wise H (2010) Characterization of Gs-coupled GPCRs in cultures of adult rat dorsal root ganglia cells in relation to neuron-glia interactions The 13th Scientific Meeting of the Hong Kong Pharmacology Society, Hong Kong SAR
Poster presentations
Yeung B, Ng KY, Wong YH, Wise H (2010) Neuron-glia interactions for Gs-coupled GPCRs in cultures of adult rat dorsal root ganglia cells The 28th Scientific Meeting of The Hong Kong Society of Neurosciences and The Annual Biennial Meeting of The Biophysical Society of Hong Kong, Hong Kong SAR
Yeung B, Chow BS, Chu CH, Ng KY, Wise H (2011) Interactions between dorsal root ganglion neurons and glial cells in vitro The IQth World Congress on Inflammation, Paris, France
ix List of abbreviations
[^H]AXP [^H]adenine nucleotides
‘ A9-THC1 A9-tetrahydrocannabinol
3X PBS 3X Dulbecco's phosphate buffered saline
5-HTIA Serotonin lA
7TM Seven transmembrane
8-OH-DPAT 8-Hydroxy-2-dipropylaminotetralin
AC Adenylyl cyclase
AM Adrenomedullin receptor
AR Adrenoceptor
BSA Bovine serum albumin
cAMP Cyclic adenosine monophosphate
CB Cannabinoid
CGRP Calcitonin gene-related peptide
CLR Calcitonin receptor-like receptor
CX43 Connexin 43
DAG Diacyl glycerol
ddHaO Double deionized water
DMSO Dimethyl sulfoxide
DNase I Deoxyribonuclease I
DRG Dorsal root ganglion
X E/DRY Glutamic acid/aspartic acid-arginine-tyrosine
EAATl Excitatory amino acid transporters 1
FITC Fluorescein isothiocyanate
G protein Guanine-nucleotide regulatory protein
GABA Gamma aminobutyric acid
GalC Galactocerebroside
GDP Guanine diphosphate
GFAP Glial fibrillary acidic protein
GLAST Glutamate aspartate transporter
GPCR G-protein coupled receptor
HBS HEPES-buffered saline
HCl Hydrochloric acid
IBMX 3-isobutyl-1 -methyl-xanthine
IB4 Isolectin B4
IL-ip Interleukin-ip
IL-6 Interleukin-6
IP3 1,4,5-triphosphate
Kir 4.1 channel Inwardly rectifying K+ channel
MAP kinase Mitogen-activated protein kinase
NGF Nerve growth factor
NO Nitric oxide
N/OFQ Nociceptin ‘
OCT Optimal cutting temperature
OR Opioid receptor
xi ORLl receptor Opioid receptor like 1 receptor
PAG Phosphate-activated glutaminase
PBS Phosphate buffered saline
PEG Polyethylene glycol
PFA Paraformaldehyde
PGE2 Prostaglandin E2
PGI2 Prostacyclin I2
PIP2 Phosphatidylinositol 4,5-bisphosphate
PKA Protein kinase A
PKC Protein kinase C
PLC-P Phospholipase C-p
PNS Peripheral nervous system
RAMP Receptor activity-modifying protein
RCP Receptor component protein
SEM Standard error mean
SGC Satellite glial cell
TCA Trichloroacetic acid
TNF-a Tumor necrosis factor-a
TTX-R Tetrodotoxin-resistant
TUJ-1 Class m p-tubulin
USG Ultroser G
I
xii Table of contents
Abstract
��^�.
Acknowledgements v//
Publications based on work in this thesis, ix
List of abbreviations
Chapter 1 Introduction
1.1 Dorsal root ganglion cells
1.1.1 Primary sensory neurons \
1.1.2 Non-neuronal cells
1.1.2.1 Satellite glial cells
1.1.2.2 Schwann cells
1.2 Peripheral sensitization
1.3 Neuron-glia interactions 9
1.4 Aim of Thesis
Chapter 2 Materials，media，buffers and solutions 13
2.1 Materials
2.2 Culture media, buffer and solutions 19
2.2.1 Culture media
2.2.2 General culture buffers and culture plate coating reagents 19
2.3 Antibodies used for identifying DRG cells 23
2.3.1 Primary antibodies 23
2.3.2 Secondary antibodies 23
Chapter 3 Methods 24
3.1 Preparation of DRG cell cultures 24
3.2 Preparation of neuron-enriched and glial cell cultures '25
3.3 Immunocytochemistry 26
xiii 3.4 Immunohistochemistry 27
3.4 Determination of [^H]cAMP production in DRG cells 28
3.4.1 Principle of assay 28
3.4.2 Loading DRG cells with [^H]adenine 28
• - 3.4.3 Column preparation 28
3.4.4 Measurement of fH]cAMP production in DRG cells 29
3.4.5 Data analysis 30
Chapter 4 Identification of DRG cells in dissociated cultures 31
4.1 Introduction
4.2 Aim of study
4.3 Results 35 4.3.1 Identification of DRG cells in isolated cultures 35
4.3.2 Activation and proliferation of glial cells in isolated cell cultures 36
4.3.3 Identification of glial cells in cultures 38
4.3.4 Modification of staining methods 40
4.3.5 Immunohistochemistry to identify DRG cells in DRG slices 42
4.3.6 Comparison of antibody staining in whole DRG and isolated DRG cells 44
4.4 Discussion 44
4.5 Summary 53
Chapter 5 Characterization of GPCRs in isolated DRG cultures 69
5.1 Introduction
5.1.1 G-protein coupled receptors
5.1.2 Pharmacological characterization of prostanoid receptors on DRG cells 73
5.1.3 Gs- and Gi/o-coupled GPCRs in DRG cells 75
5.1.3.1 Gs-coupled GPCR: ^-adrenoceptors 76
5.1.3.2 Gs-coupled GPCR: CGRP receptors 79
5.1.3.3 Gi/o-coupled GPCR: a2-adrenoceptors 82
5.1.3.4 Gi/o-coupled GPCR: Cannabinoid receptors 85
5.1.3.5 Gi/o-coupled GPCR: 5-HT,a receptors gg
5.1.3.6 Gj/o-coupled GPCR: opioid and opioid-receptor-like 1 receptors 90
5.2 Aims of study
xiv 5.3 Results 94
5.3.1 Characterization of prostanoid receptors in isolated DRG cells 94
5.3.2 Characterization of CGRP receptors in isolated DRG cells 96
5.3.3 Investigation of the effect of CGRP 8.37 on CGRP responses 97
5.3.4 Characterization of Pi-adrenoceptors in isolated DRG cells 97
5.3.5 Characterization of P2-adrenoceptors in isolated DRG cells 98
5.3.6 Identification of J5-adrenoceptor subtype mediating isoprenaline-stimulated responses..
99
5.3.7 Characterization of (X2-adrenceptors in isolated DRG cells 100
5.3.8 Characterization of cannabinoid 1 receptors in isolated DRG cells 100
5.3.9 Characterization of cannabinoid 2 receptors in isolated DRG cells 101
5.3.10 Characterization of 5-HTia receptors in isolated DRG cells 101
5.3.11 Characterization of ^-opioid receptors in isolated DRG cells 102
5.3.12 Characterization of opioid-receptor-like 1 receptors in isolated DRG cells 102
5.3.13 Effect of nerve growth factor on DRG cells 103
5.4 Discussion 106
5.5 Summary
Chapter 6 Conclusion and further studies 134
References. 237
XV Chapter 1
Introduction
1.1 Dorsal root ganglion cells
1.1.1 Primary sensory neurons
The dorsal root ganglia (DRG) are located in the intervertebral spaces in the peripheral nervous system (PNS) adjacent to the spinal cord (Tang et al, 2007), and they consist of neurons and non-neuronal cells (Hanani, 2005). The neurons in the
DRG are called primary sensory neurons, which are responsible for converting sensory stimuli into sensory signals. The primary sensory neurons are pseudounipolar where one process extends to the periphery such as skin, muscle and internal organs, while the other central process extends and terminates at the dorsal horn of the spinal cord (Hanani, 2005). Upon various sensory stimulations, including mechanical, thermal and chemical stimuli, the primary sensory neurons are responsible for detecting and conveying sensory information from the periphery, and transmitting the information in terms of nerve impulses to the brain for perception of these sensations.
‘
The primary sensory neurons are classified based on their anatomical and functional characteristics. The large diameter neurons are heavily myelinated and they give rise
1 to Aa and Ap fibers (Julius et al., 2001; Milligan et al, 2009). The myelin sheaths on these fibers help maintain the electric current in the axon and facilitate propagation of impulses along the fiber from node to node (see Section 1.1.2.2). This feature of large diameter neurons leads to the fast-conducting velocity for signal transmission.
The large diameter neurons consist of low threshold mechanoreceptors at the terminal and account for detection of innocuous stimulations such as light touch or proprioception (Julius et aL, 2001; Milligan et al； 2009; Zhou et al, 2010). The medium diameter neurons are lightly myelinated and give rise to the A5 fibers. Being lightly myelinated, nerve impulses cannot be efficiently propagated from node to node, and therefore, the transmission velocity of the medium diameter neurons is not as fast as that of the large diameter neurons. The medium diameter neurons are responsible for transmitting harmful stimuli into nociception, a term to refer to perception of pain. Lastly, the smaller diameter neurons are non-myelinated, which give rise to the C fibers. They have high threshold for activation and they are responsible for transmitting nociceptive signals slowly. The difference in the transmission velocities between the A6 and C fibers result in the generation of the first pain, mediated by the A6 fiber, being sharp, acute and rapid, and the second pain,
i mediated by the C fiber, being more dull, diffuse and delayed (Julius et al, 2001).
The small diameter neurons are further divided into peptidergic and non-peptidergic
2 neurons. The peptidergic, small-diameter neurons express peptide neurotransmitters such as substance P and calcitonin gene-related peptide (CGRP). They are characterized by the expression of high affinity tyrosine kinase receptors, TrkA receptors, for nerve growth factor (NGF). In contrast, the non-peptidergic, small- diameter neurons do not express substance P, CGRP and TrkA receptors, but they can be identified by the binding of lectins such as Griffonia simplicifolia isolectin B4
(IB4), and the expression of P2X3 receptors, which are a subtype of ATP-gated ion channel (Julius et al., 2001).
1.1.2 Non-neuronal cells
In the DRG, other than sensory neurons, there are also non-neuronal cells. The
predominant non-neuronal cells include satellite glial cells (SGCs), Schwann cells
and fibroblasts. The glial cells, consisting of SGCs and Schwann cells, have close
associations with the sensory neurons providing not only passive structural supports,
but also active contribution during normal and pathological states by interactions
with neurons.
1.1.2.1 Satellite glial cells
The SGCs are a type of glial cell in the PNS that closely enwrap the surface of a
neuronal cell body (Pannese, 1981). They are precisely described as to surround the
3 cell body and the initial segment of the axon of primary sensory neurons, leaving a gap of 20 run between themselves and the neuronal cell surface (Ohara et al, 2009;
Pannese, 1981; Vit et al., 2006). The complete wrapping of the neuronal cell body
separates adjacent neurons and forms a discrete neuron-SGC structural unit, which
consists of a single neuronal cell body and an envelope of the SGCs. However,
clusters of two or three neuronal cell bodies enclosed by the same SGCs can
occasionally be found (Hanani, 2005; Pannese, 2010). The neuronal ensheathing
feature is unique to SGCs and is not found in other types of glial cells in the central
nervous system (CNS) such as astrocytes (Hanani, 2005). Each neuronal cell body is
normally surrounded by approximately eight SGCs in rat in vivo while the ratio of
neurons to SGCs varies in different species (Ledda et al., 2004). The principle
function of SGCs enveloping neurons is to regulate the external chemical
microenvironment of neurons by physically preventing excessive access of
extracellular ions, such as K+, into neurons (Vit et ah, 2006).
The SGCs are connected among themselves by gap junctions (Ohara et al., 2008) and
they have been shown to communicate with each other by intracellular Ca^^ waves
(Suadicani et al, 2010). They also possess mechanisms for the release of cytokines
and adenosine triphosphate (ATP) for signaling (Suadicani et al., 2010; Takeda et al,
2009). Besides, the various functions of SGCs can be revealed by the expression of
4 numerous ion channels, transporters and surface receptors including GPCRs, such as
P2Y4 receptors for ATP. For instance, the SGCs express inwardly rectifying K+ channels (Kir 4.1 channels) that are responsible for the uptake of extracellular K+ into their cell body and the dissipation of the K+ ions through the Ca^"^-activated K+ channels and the gap junction proteins, such as connexins (Vit et al, 2006).
Providing this function, the SGCs help maintain a low extracellular K+ concentration for controlling neuronal resting membrane potential and neuronal excitability (Vit et aL, 2006). Besides, the SGCs also possess transporters to take up neurotransmitters such as glutamate and gamma aminobutyric acid (GABA) (Jasmin et aL, 2010;
Schon et al, 1974). For example, in human, SGCs express excitatory amino acid transporters 1 (EAATl), which is equivalent to the glutamate aspartate transporter
(GLAST) in rodent, to transport glutamate into their cell body (Ohara et al., 2009;
Pannese, 2010). Once in the cytoplasm of SGCs, glutamate will be amidated to
glutamine by another enzyme glutamine synthetase (GS). Glutamine will then be
subsequently released to the extracellular space. The glutamine will then be taken up
by the nearby neurons into the cytoplasm and be converted back to glutamate by the
enzyme phosphate-activated glutaminase (PAG). The SGCs, in this way, provide
f
very important protective roles to neurons by maintaining a low extracellular
concentration of glutamate to prevent interminable activation of neurons, which
5 might lead to excitotoxicity, a condition when neurons are damaged due to high concentrations of excitatory neurotransmitters. It has therefore been accepted that the primary function of SGCs is to provide tight regulation on the trafficking of extracellular ions and neurotransmitters and thereby to prevent the accumulation of any of these substances that could potentially harm neurons.
It has long been assumed that SGCs are passive and protective, but recently it has been suggested that the roles of SGCs were being overlooked and underestimated
(Ledda et al, 2004). The functions of SGCs are more than being passive and protective. There is now a great deal of evidence that SGCs actually play very
important roles in modulating neuronal activities and the development and the
maintenance of pain (see Section 1.3).
1.1.2.2 Schwann cells
Schwann cells are the other type of glial cells in the DRG. They are classified as
myelinating Schwann cells and non-myelinating Schwann cells (Campana, 2007).
The myelinating Schwann cells produce myelin, a lipid rich membrane, and wrap
their plasma membrane around the axon of the medium and large diameter neurons
I
to form a segment of myelin sheath (Campana, 2007). The myelin sheaths are not
continuous, leaving gaps in between called nodes of Ranvier (Bhatheja et al., 2006).
6 The sheathed regions of the axon membrane are highly insulated, in which only little current can leak across it. At the node of Ranvier, voltage-gated Na+ channels are highly concentrated and therefore, depolarization of the axon membrane is actively confined at the nodes of Ranvier. Thus, action potential generated at one node almost immediately spreads to the next node and continues. This allows action potential along the axon to jump from node to node, a process called saltatory conduction.
As a result of saltatory conduction, the speed and the efficiency of action potential propagation along a myelinated axon is greatly enhanced when compared with that along a non-myelinated axon. The non-myelinating Schwann cells do not produce myelin and they ensheath the small diameter neurons (Campana, 2007). The functions of non-myelinating Schwann cells are thought to provide supportive roles for and communicate with the small diameter neurons for proper neuronal functions
(Chen et al., 2003). It is suggested that Schwann cells are sensitive to nerve injury by
modulating their phenotype, proliferation, migration and secretion of pro-
inflammatory cytokines, such as tumor necrosis factor-a (TNF-a), interleukin ip (IL-
IP) and interleukin 6 (IL-6) (Campana, 2007). In addition, the activated Schwann
cells are also recruited to form a scaffold for regenerating axons to grow along,
which ultimately lead to the re-innervation of the target and functional recovery of
the axon following nerve injury (Vroemen et al, 2003).
7 1.2 Peripheral sensitization
Under normal physiological conditions, pain signals are not generated due to the high threshold of nociceptive neurons for activation (Milligan et al., 2009). Nevertheless, after nerve injury, primary sensory neurons are subject to the change of their plasticity, which refers to the change of neuronal functions and chemical profiles
(Woolf et al., 2000). These changes alter the normal functions of neurons. For example, upon repeated noxious stimulation, nociceptive neurons increase their responsiveness by lowering the threshold to generate pain signals. This results in increased and more frequent responses than that of a given normal afferent input.
This situation is often referred as hyperalgesia, a state in which the sensitivity to pain is augmented. In other words, pain is elicited by even gentle stimuli when hyperalgesia has developed.
Neuropathic pain refers to a type pathological pain resulting from nerve injuries in the PNS (Vit et al., 2008). It leads to hyperalgesia that is resistant to treatment by common analgesics. The major factor leading to hyperalgesia is the change of plasticity of the DRG neurons from the exclusive detection of noxious stimulus to the detection of innocuous stimuli, a process called peripheral sensitization (Woolfer al.,
2007). The development and the maintenance of neuropathic pain not only involves primary sensory neurons, but also immune and glial cells (Scholz et al., 2007).
8 Neurons are sensitized through several different pathways. One of the important ways is to enhance the production of intracellular cyclic adenosine monophosphate
(cAMP). In response to physical injury or inflammatory insult, prostanoids are released by the injured cells and immune cells (Julius et al, 2001). Among the prostanoids released, prostaglandin Ei (PGE2) is one of the key contributors to the development of pain. PGE2 binds to EP receptors such as EP4, a GPCR which couples with Gs protein and is expressed on primary sensory neurons (Bley et ah,
1998). This results in the increase of the intracellular level of cAMP at the nociceptor by enhancing the activity of adenylate cyclase (Bley et al, 1998). The increase in
cAMP level activates protein kinase A (PKA) and its downstream pathways, causing
the phosphorylation of voltage-gated channels, such as the tetrodotoxin-resistant
(TTX-R) Na+ channel, and the ligand-gated ion channel, TRPVl (Bhave et al.，2002;
Hucho et al., 2007). For example, phosphorylation of Na+ channels potentiates the
influx of Na+, depolarizes the nociceptor, and lowers the threshold for generating an
action potential, which favors repetitive firings of pain signals by nociceptive
neurons (Julius et al, 2001).
1.3 Neuron-glia interactions
The above descriptions of SGCs and Schwann cells demonstrated their essential
protective roles to DRG neurons (see Sections 1.1.2.1 and 1.1.2.2). The relationship
9 of glial cells to neurons has long been considered as passive. Moreover, it was thought that neurons were the only cells that mediated pain sensations because of the characteristics to transduce noxious signals. However, it has recently been suggested that interactions between neurons and glial cells are bidirectional and glial cells play
important roles in modulating pain development and processing. The evidence for
interactions between neurons and glial cells to modulate pain processing includes the
demonstration that silencing Kir 4.1 channel subunit in SGCs of the rat trigeminal
ganglion (TG) results in pain-like behavior in the absence of nerve injury (Vit et al.,
2008). Besides, we previously demonstrated that glial cells induced neurite retraction
of IB4-negative large diameter neurons when Gj/�protein wers e inhibited. This
indicates that glial cells constitutively release factors that influence neurite outgrowth
(Ng et al, 2010). Moreover, increase in expression of glial cell specific gap junction
protein subunit, connexin 43 (CX43), was reported in a rat model with chronic
constriction injury (Ohara et al., 2008). The role of CX43 in pain was revealed by the
reduced pain-like behavior after down-regulation of CX43 using siRNA (Ohara et al，
2008).
Recently we showed that pain-related prostanoid receptors (EP4 and IP receptors)
were surprisingly expressed on glial cells in vitro, in addition to neurons (Ng et al.,
2011). The majority of prostanoid-stimulated cAMP production was seen in glial
10 cells although production of cAMP in neurons is proposed to facilitate pain.
Moreover, the presence of neurons inhibited prostanoid-stimulated cAMP production in glial cells (see Section 5.1.2). The expression of pain-related receptors on glial
cells and the inhibitory effect of neurons on glial cell responses suggested that it was
important to consider the roles played by glial cells and the effect of neuron-glial
interactions.
Together with evidence to support bidirectional neuron-glial interactions, it is now
clear and accepted that the glial cells play very active roles in modulating the
neuronal activity. The glial cells modulate the process of pain development and
therefore, the functions of glial cells should not be overlooked.
1.4 Aim of Thesis
It is now clear that neurons and glial cells undergo bidirectional interactions to
influence the activities of each other. Besides, we have recently identified the
presence of pain-related GPCRs (EP4 and IP receptors) on both DRG neurons and
glial cells in vitro; it is therefore appropriate to reinterpret the expressions of GPCRs
that mediate pain on DRG neurons and glial cells. The full understanding of the
neuron-glia interaction requires the identification of neurons and the types of glial
cells in isolated DRG cell preparations.
11 The aim of first part of the thesis was to identify different types of glial cells and any fibroblasts in isolated DRG cultures with the use of various antibodies specific for different non-neuronal cells. Then, in response to our recent finding that shows (1) pain-related prostanoid receptors (EP4 and IP receptors) were expressed on both neurons and glial cells in cultures of rat DRG and (2) prostanoid receptor agonist-
stimulated responses in glial cells were inhibited by co-culture with neurons, the aim
of the second part of the thesis was to characterize the expression of other pain-
related Gs- and Gj/o-coupled GPCRs on neurons and glial cells in cultures with the
use of receptor-specific agonists and antagonists. Besides, we aimed to identify if
any other receptor-specific activity of glial cells would also be inhibited by co-
culture with DRG neurons in the same way as seen with prostanoid receptor agonist-
stimulated responses.
12 Chapter 2
Materials, media, buffers and solutions
2.1 Materials
[8-^H]adenine (specific activity 29.0 Ci/mmol, stored at 4 Amersham)
8-hydroxy-DPAT hydrobromide (8-OH-DPAT; Tocris Bioscience) was dissolved in
ddH20 at 20 mM and aliquots were stored at -20 °C.
Acetone (Lab Scan)
Adenosine 5'-triphosphate (ATP; Sigma)
Alumina column (Aluminum oxide, neutral type WN-3; Sigma)
AM251 (Tocris Bioscience) was dissolved in absolute ethanol at 20 mM and aliquots
were stored at -20
AM630 (Tocris Bioscience) was dissolved in DMSO at 50 mM and aliquots were
stored at -20
S-(-)-atenolol (Tocris Bioscience) was dissolved in ddHbO at 25 mM and aliquots
were stored at -20
4
Bovine serum albumin (BSA; Sigma)
13 CGRP (species: rat; Tocris Bioscience) was dissolved in ddHaO at 1 mM and
aliquots were stored at -20 °C.
CGRPg-s? (species: rat; Tocris Bioscience) was dissolved in ddHiO at 10 mM and
aliquots were stored at -20 °C.
Cicaprost (Gift from Schering AG, Germany) was dissolved in ddHiO at 1 mM and
aliquots were stored at -20
Collagenase B (Roche) was prepared as a 1.25% stock in Ham's F14 medium and
aliquots were stored at -20 °C.
DAMGO (Tocris Bioscience) was dissolved in ddHzO at 10 mM and aliquots were
stored at -20 °C.
Deoxyribonuclease I (DNase I; Sigma)
Dimethyl sulfoxide (DMSO; Sigma)
Dorminal (20% pentobarbital solution; Alfasan) was diluted to a 5% solution with
ddH20 before use.
Dowex column (AG SOW X4 resin, 200 - 400 mesh hydrogen form; Bio Rad)
Ethanol (Absolute; Merck)
14 Formoterol hemifumarate (Tocris Bioscience) was dissolved in DMSO at 50 mM
and aliquots were stored at -20
Forskolin (Sigma) was dissolved in DMSO as a 10 mM stock and aliquots were
stored at -20
Glass coverslips (12 mm and 25 mm; Menzel Glaser)
Glass slide (Marienfeld Laboratory Glassware)
Glucose (Sigma)
L-Glutamine (Life Technologies) was prepared as a 200 mM solution in ddHzO,
filter sterilized, and aliquots were stored at -20
Glycerol (Sigma)
HEPES (Sigma)
Hoechst 33342 (Invitrogen) was prepared as a 10 mg/ml stock in ddHzO，aliquots
were stored at -20
Hydrochloric acid (HCl, 12.1 M; Merck)
IBMX (3-isobutyl-1 -methyl-xanthine; Sigma) was prepared as a 100 mM stock in
DMSO and aliquots were stored at -20
15 ICI118551 hydrochloride (Tocris Bioscience) was dissolved in DMSO at 10 mM
and aliquots were stored at -20
Imidazole hydrochloride (Sigma)
Isoproterenol hydrochloride (Isoprenaline; Tocris Bioscience) was dissolved in
ddHzO at 100 mM and aliquots were stored at -20 °C.
(±)-Jl 13397 (Tocris Bioscience) was dissolved in absolute ethanol at 50 mM and
aliquots were stored at -20°C.
L-759656 (Tocris Bioscience) was dissolved in DMSO at 100 mM and aliquots were
stored at -20�C.
Laminin (Roche) was prepared as a 0.25 mg/ml stock in sterile PBS and aliquots
were stored at -20
R-(+)-methanandamide (Tocris Bioscience) was dissolved in absolute ethanol at 13.8
mM and aliquots were stored at -20
Methanol (Lab Scan)
Naloxone hydrochloride (Tocris Bioscience) was dissolved in ddHaO at 100 mM and
aliquots were stored at -20
16 Nerve growth factor (NGF mouse 2.5s; Alomone Laboratories) was dissolved in F14
medium at 10 jig/ml and aliquots were stored at -80
Nociceptin (N/OFQ; Tocris Bioscience) was dissolved in ddHzO at 10 mM and
aliquots were stored at -20
Normal Donkey serum (Jackson Immunoresearch)
ONO-AE1-329 (Gift from ONO Pharmaceuticals, Japan) was dissolved in ddHaO at
100 mM and aliquots were stored at -80
OptiPhase 'Hi-safe'3 (Perkin Elmer)
Paraformaldehyde (Sigma)
Penicillin/Streptomycin (10000 units/ml and 10000 |xg/ml; Gibco) was reconstituted
with 20 ml ddHzO and stored at 4
Poly-DL-omithine (Sigma) was prepared as a 25 mg/ml stock solution in ddHaO,
filter sterilized, aliquots were stored at -20
Polyethylene glycol (PEG; Sigma)
Prostaglandin E2 (PGE2； Sigma) was dissolved in absolute ethanol at 10 mM and 4
aliquots were stored at -20
17 Rauwolscine hydrochloride (yohimbine; Tocris Bioscience) was dissolved in DMSO
at 20 mM and aliquots were stored at -20
Soybean trypsin inhibitor (Sigma)
Trichloroacetic acid (TCA; Sigma) was prepared as a 10% stock (w/v) in ddHaO,
stored at 4 °C.
Triton X-100 (Sigma)
Trypan blue was prepared as a 0.4% stock in PBS, stored at room temperature.
Trypsin (Sigma) was prepared as a 25 mg/ml stock in sterile PBS, aliquots were
stored at -20
Trypsin inhibitor (Type I-S, From Soybean; Sigma)
UK14304 (Tocris Bioscience) was dissolved in DMSO at 100 mM and aliquots were
stored at -20
Ultroser G (USG; Pall Life Science) was reconstituted with 20 ml sterile ddHzO and
stored at 4
S-WAY100135 dihydrochloride (Tocris Bioscience) was dissolved in DMSO at 20
mM and aliquots were stored at -20
Common laboratory reagents not listed (ACS grade) were purchased from Sigma.
18 All disposable culture plates were purchased from Nunc.
2.2 Culture media, buffer and solutions
2.2.1 Culture media
Ham's Nutrient Mixture F14 (F14 medium; SAFC Biosciences). One packet of
F14 powder was reconstituted with 1 L ddHzO, and supplemented with 1.176 g
NaHCOs. The medium was adjusted to pH 7.2 with 1 M HCl and filter sterilized to
give a final pH 7.4. Penicillin and streptomycin was added to the medium to give a
final concentration of 100 units/ml and 100 |ig/ml, respectively. Medium was stored
at 40c.
Complete F14 medium. 191 ml of F14 medium was supplemented with 8 ml of
serum substitute USG (4%, v/v) and 1 ml of L-glutamine (1 mM) immediately prior
to use.
2.2.2 General culture buffers and culture plate coating reagents
Dulbecco's PBS buffer (PBS: Ca2+/Mg2+ free) was prepared by dissolving NaCl
(137 mM), Na2HP04 (10 mM), KCl (2.6 mM) and KH2PO4 (1.8 mM) with ddHzO.
The solution was adjusted to pH 7.5 with 1 M NaOH, filter sterilized and stored at
40c.
19 Laminin was freshly prepared by diluting stock solution from 0.25 mg/ml to 5
j^g/ml with sterile PBS and added into poly-DL-omithine coated wells for at least 1 h
at room temperature.
Poly-DL-ornithine was prepared the day before cell dissociation by diluting stock
solution from 25 mg/ml to 500 |ig/ml with ddHsO and added into wells of cell
culture plates overnight at 4
2.2.3.1 Buffers and solutions for immuno cy to chemistry
Triton X-100 (0.3%, v/v) was prepared by adding 0.6 ml of Triton X-100 to
199.4 ml of PBS and stored at
Triton X-100 (0.01%, v/v) was prepared by adding 0.3 ml of Triton X-100 (0.3%)
to 8.7 ml of PBS immediately prior to experiment.
Blocking solution (3%, v/v) was prepared by adding 0.3 ml of normal donkey
serum to 9.7 ml of PBS immediately prior to experiment.
Kryofix solution was prepared by dissolving 10 g of PEG into 79 ml of absolute
ethanol and 74 ml of ddHzO (absolute ethanol:H20:PEG300 at a ratio of 7.9:7.4:1)，
and stored at 4 ‘
Acetone and methanol (1:1) solution was prepared by adding 10 ml of acetone to
10 ml of methanol and stored at 4
20 Paraformaldehyde (4%, w/v) was prepared by dissolving 0.48 g of paraformaldehyde in 8 ml of ddUjO with 80 of 1 M NaOH at 60 ''C. Then, the solution was removed from heat and was supplemented with 4 ml of 3X PBS.
3X Dulbecco，s PBS (3X PBS: Ca^^/Mg^'' free) was prepared by dissolving NaCl
(411 mM), Na2HP04 (30 mM), KCl (7.8 mM) and KH2PO4 (5.4 mM) with ddHzO.
The solution was adjusted to pH 7.5 with 1 M NaOH, filter sterilized and stored at
Mounting medium (90% glycerol) was prepared by mixing 18 ml of glycerol with
2 ml of ddHzO and stored at room temperature.
2.2.3.2 Buffers and solutions for assay of [^H]cAMP production
HEPES-buffered saline (HBS) buffer was prepared by dissolving HEPES (15 mM),
NaCl (140 mM), KCl (4.7 mM), CaCl2.2H20 (2.2 mM), MgCl2-6H20 (1.2 mM) and
KH2PO4 (1.2 mM) in ddUjO. The buffer was adjusted to pH 7.4 with 1 M NaOH, filter sterilized and stored at 4 °C.
Washing Buffer was prepared by adding glucose (3.3 mM) to HBS immediately before cAMP assay.
•
Assay Buffer was prepared by adding IBMX (1 mM) into washing buffer immediately prior to experiment.
21 Imidazole solution (0.5 M) was prepared by dissolving 104.5 g of imidazole powder and 29 g of NaOH in ddHsO. The solution was adjusted to pH 7.5 with 1 M
NaOH, made up to 2 L and stored at room temperature.
Imidazole solution (0.1 M) was prepared by diluting 400 ml of 0.5 M Imidazole
solution with 1600 ml of ddHsO and stored at room temperature.
Hydrochloric acid (4 M) was prepared by diluting 330 ml of 12.1 M HCl to 670 ml
of ddHzO and stored at room temperature.
Hydrochloric acid (1 M) was prepared by diluting 500 ml of 4 M HCl to 1500 ml
of ddH20 and stored at room temperature.
Stop solution (10% TCA with 2 mM ATP) was prepared by dissolving ATP (2
mM) with 10% TCA. The stop solution was freshly prepared and temporary kept at
4 until use.
22 2.3 Antibodies used for identifying DRG cells
2.3.1 Primary antibodies
Antibody Host Type Supplier Catalog number
CNPase Mouse Monoclonal IgG Millipore MAB326
GalC Rabbit Polyclonal IgG Sigma G9152
GFAP Mouse Monoclonal IgG 1 Sigma G6171
GLAST Guinea Pig Polyclonal IgG Chemicon AB1783
GS Mouse Monoclonal IgG2a Chemicon P20749
Sioop Mouse Monoclonal IgG 1 Sigma S2532
SK3 Rabbit Polyclonal IgG Alomore APC025
Thyl.l Mouse Monoclonal IgM Sigma M7898
TUJ-1 Rabbit Polyclonal IgG Sigma T2200
2.3.2 Secondary antibodies
Antibody Supplier Catalog number
Alexa Fluor 647 anti-mouse IgG Invitrogen A21463
Alexa Fluor 647 anti-rabbit IgG Invitrogen A31573
FITC anti-guinea pig IgG Jackson 706-095-148 Immunoresearch
FITC anti-mouse IgG Jackson 715-095-150 Immunoresearch FITC anti-mouse IgG subtype Fey Jackson 115 095 206 subclass 2a specific Immunoresearch
T A/r u • Jackson 715-095-140 FITC anti-mouse IgM u chain specific , , Immunoresearch
FITC anti-rabbit IgG �ackso n 711-095-152 Immunoresearch
23 Chapter 3
Methods
3.1 Preparation of DRG cell cultures
All experiments were performed under license from the Government of the Hong
Kong SAR and approval by the Animal Experimentation Ethics Committee of the
Chinese University of Hong Kong. The dorsal root ganglia were removed from all levels of the spinal cord of male Sprague—Dawley rats (150-200 g) and cultures were prepared as described previously (Ng et al., 2010). Briefly, rats were deeply anaesthetized with pentobarbitone (135 mg/kg, i.p.). The whole spinal column was dissected out from rostral end of the hind legs to the base of the skull. Excess tissue was removed with the use of small scissors and rongeurs to expose the dorsal side of the vertebra. The spinal cord was exposed by cutting a strip of bone from the dorsal roof of the vertebral column using Malleus Nippers. Using Vannas curved spring scissors and fine forceps, dorsal root ganglia were dissected out and placed immediately in complete Ham's F14 medium, followed by incubation with 0.125% collagenase for 3 h and with 0.25% trypsin for 30 min in serum-free Ham's F14 medium. The dorsal root ganglia were then resuspended in DNase I (0.44 mg/ml)
< and soybean trypsin inhibitor (0.55 mg/ml), and dispersed by series of triturations using a flame-polished salinized Pasteur pipette. The cell suspension was centrifiiged through a cushion of BSA (15%) to eliminate much of the myelin and cellular debris.
24 After that, the supernatant was aspirated and the cell pellet was resuspended in complete Ham's F14 medium. Morphologically distinct phase-bright neurons and phase-dark glial cells were counted using a hemocytometer slide. Cells were typically seeded at 5000 neurons per well onto 24-well tissue culture plates pre- coated with poly-DL-omithine (500 |a,g/ml) and laminin (5 |j.g/ml) and assayed after two days in culture in an atmosphere of 5% CO2 at 37 for immunocytochemical and cAMP studies.
3.2 Preparation of neuron-enriched and glial cell cultures
To prepare neuron-enriched cell cultures, mixed DRG cells were plated onto poly-
DL-omithine coated 10-cm tissue culture dishes (with a maximum of cells from two rats per dish), as described in Lindsay (1988). After an overnight incubation, the loosely attached neuronal cells were gently removed from the more firmly attached
glial cells by a repetitive stream of F14 medium (4 ml) delivered using a flame-
polished salinized Pasture pipette.
The remaining glial cells on the 10-cm tissue culture dishes were harvested after
incubation with 1 ml of trypsin (0.05% in PBS) for 1 min at 37� Cwith 5% CO2.
DRG neurons and glial cells (small phase-dark cells) were counted using a
hemocytometer slide. Neurons and glial cells were seeded onto poly-DL-
omithine/laminin-coated 24-well tissue culture plates at a density of 5000 neurons
25 per well and 10000 glial cells per well, respectively. Neuron-enriched cells and glial cells were assayed after one day in culture for immunocytochemical and cAMP studies. Cell numbers in neuron-enriched and glial cell cultures were similar to those in the mixed DRG cell cultures on the day of assay.
3.3 Immunocytochemistry
Cells cultured on coated glass coverslips were washed twice with PBS and then preserved in kryofix solution (Raye et al, 2007) at 4 if immunocytochemistry was not performed immediately. Otherwise, coverslips with cells were washed twice with PBS (5 min interval between each wash) and then fixed with ice-cold
acetone:methanol (1:1) for 20 min or with 4% paraformaldehyde (PFA) for 10 min
at room termperature. Cells were then washed three times with PBS and incubated
with 0.3% Triton X-100 with 3% donkey serum for 30 min to permeabilize cells and
block non-specific antibody binding. Cells were then incubated with primary
antibodies at 4 °C overnight, followed by secondary antibody incubation for 1 h at
room temperature. The primary and secondary antibodies employed and sources are
listed in Sections 2.3.1 and 2.3.2. All samples were incubated for 4 min with
Hoechst 33342 stain (0.2 (xg/ml) to identify cell nuclei. The coverslips were mounted
4
with glycerol (90%) on glass slides and imaged using an Olympus 1X51 inverted
microscope (Center Valley, PA, USA) with Qimaging Retiga-2000R CCD camera
26 (Surrey, BC, Canada). Fluorescent images were merged using Adobe Photoshop
CS4 (San Jose, CA, USA).
3.4 Immunohistochemistry
Whole dorsal root ganglia were dissected from all spinal levels of Sprague-Dawley rats and fixed with 4% paraformaldehyde in PBS at 4 for 24 h. The fixed DRG tissues were stored in sucrose solution (30% sucrose, 10% glycerol) for at least 48 h at 4 °C. Whole DRG were then placed in embedding medium, optimal cutting temperature (OCT) and frozen at -80 °C. DRG were cut into thin sections (15 |Lim) using a cryostat, thaw-mounted on slides and stored at -80 DRG sections were incubated in 0.01% Triton X-100 for 15 min and then with blocking solution for 30 min at room temperature. DRG sections were then incubated with primary
antibodies at 4 °C overnight, followed by secondary antibody incubation for 1 h at
room temperature. The primary and secondary antibodies employed and sources are
listed in Sections 2.3.1 and 2.3.2. All samples were incubated for 4 min with
Hoechst 33342 stain (0.2 jig/ml) to identify cell nuclei. The glass slides with whole
DRG sections were mounted with glycerol (90%) and imaged using an Olympus
1X51 inverted microscope (Center Valley, PA, USA) with Qimaging Retiga-2000R
I
CCD camera (Surrey, BC, Canada). Fluorescent images were merged using Adobe
Photoshop CS4 (San Jose, CA, USA).
27 3.4 Determination of [^H]cAMP production in DRG cells
3.4.1 Principle of assay
The adenylyl cyclase activity was determined by loading DRG cells with
[^H]adenine which is converted to [^H]ATP inside the cells. When adenylyl cyclase is stimulated, [^H]ATP is converted to [^H]cAMP. Breakdown of [^H]cAMP was prevented by incubating cells with IBMX (a phosphodiesterase inhibitor) so that the accumulation of [^H]cAMP was measured. [^H]cAMP was then separated from other tritium-labeled adenine derivatives ([^H]AXP), including [^H]ATP and [^H]ADP, using Dowex-Alumina column chromatography. The [^H]cAMP production in the presence of IBMX reflects the activity of adenylyl cyclase (Barber et al, 1980).
3.4.2 Loading DRG cells with [^H]adenine
To determine the total cAMP production in DRG cell cultures, cells were washed with 1 ml of F14 medium and incubated with 0.5 ml of complete F14 medium
containing [^H]adenine (2 |LiCi/ml; 1 |iCi/well) for 16 h before the [^H]cAMP assay.
3.4.3 Column preparation
Dowex AG50W-X4 resin columns were washed with 10 ml of distilled water before
use. After use, these columns were washed with three washing cycles with 4 ml of 1
M HCl followed by 10 ml distilled water. The alumina columns were washed with 8
28 ml of 0.1 M imidazole solution before use. After use, the columns were washed four times with 8 ml of 0.1 M imidazole solution.
3.4.4 Measurement of [^H]cAMP production in DRG cells
On the assay day, cells were washed twice with 1 ml of washing buffer and incubated with 0.5 ml assay buffer (HBS containing 1 mM IBMX) in the presence or absence of antagonists for 15 min followed by 30 min incubation with agonist at
37 in assay buffer. Treatment was stopped by addition of 0.5 ml of ice-cold stop solution (10% TCA with 2 mM ATP). Then, the samples were left on ice for at least
1 h for the extraction of [^H]cAMP and other tritium-labeled adenine derivatives.
This method measures both intracellular and extracellular [ H]cAMP.
[3h]cAMP was isolated from other [^H]adenine nucleotides using column chromatography. Firstly, the Dowex columns were placed over scintillation vials.
Then samples (in 1 ml 10% TCA extracts) were loaded on to the columns followed
by 3 ml of distilled water to elute the [^H]AXP into the scintillation vials. The eluate
was mixed thoroughly with 7 ml of scintillant (OptiPhase ’Hi-safe’ 3).
The Dowex columns were then placed over the alumina columns. The remaining
[^H]cAMP fraction was eluted with 10 ml of distilled water. When water had
completely run through both set of columns, the alumina columns were already
loaded with [^H]cAMP fraction. Then, the alumina columns were placed over
29 another set of scintillation vials and fH]cAMP was eluted by 6 ml of 0.1 M imidazole solution. At last, the eluate was mixed thoroughly with 10 ml of scintillant
(OptiPhase 'Hi-safe' 3).
In parallel to test samples, blank samples (a mixture of 0.5 ml assay buffer and
0.5 ml stop solution) were also added to the columns and processed the same way as for samples for the determination of the background count of the columns. The value obtained was subtracted from all the test values. The radioactivity of samples was measured by liquid scintillation counting (Packard LS2900TR).
3.4.5 Data analysis
The production of [ HJcAMP from cellular [ H]ATP was estimated as the ratio of radiolabelled cAMP to total AXP, and was expressed as x 100% (i.e. % conversion) (Salomon, 1991). For the characterization of Gg-GPCRs, the percent inhibition of the agonist-stimulated response by the antagonist was calculated as:
1 (response of agonist in the presence of antagonist - basal) � (response of agonist alone - basal)
All assays were performed in duplicate. Comparison between groups were made using ANOVA (analysis of variance) with Bonferroni's post hoc test. Statistical
f significance was taken asp < 0.05.
30 Chapter 4
Identification of DRG cells in dissociated cultures
4.1 Introduction
There is considerable interest in the use of dissociated cultures of DRG cells for the study of the properties of individual neurons and the underlying mechanism of increased excitability in response to nerve injuries. Much about the .process of pain development has been learnt from in vitro studies using DRG cells. For example, the isolated small diameter neurons have been used as a model for electrophysiology studies to investigate pain processes in cell cultures (Gold et al, 1996; Harper et al.,
1985). Moreover, recent evidence shows that the glial cells, which are in close association with neurons, modulate pain development and processing directly by
sensitizing neurons following nerve injury (Capuano et al., 2009). An advantage of
studying DRG cells in culture is that DRG cells can be studied separately in cultures
with the aid of cell separation techniques. For instance, neurons and non-neuronal
cells in a mixed DRG cell population can be isolated in cultures, in which the
investigation on the activity of a particular type of cell with another type of cell
t becomes readily available.
Morphologically, DRG neurons can easily be distinguished from the non-neuronal
31 cells based on the cell sizes under the view of phase-contrast microscope at a low magnification. However, whether a certain non-neuronal cell is a SGC, Schwann cell or fibroblast cannot be judged simply by its morphology because there are currently no well-defined and consistent descriptions of different non-neuronal cells in both live and fixed cell cultures. In living cultures, TG SGCs were defined as small, dark cells with an elongated shape (Capuano et al., 2009). While in fix and stained cultures, TG SGCs were described as having a fusiform shape whereas fibroblasts were flat cells with an large elongated nucleus (Ceruti et al, 2011). In another study,
DRG SGCs and Schwann cells were distinguished by recognizing SGCs as being multipolar while Schwann cells as being predominately spindle shaped, small bipolar or tripolar (Ahmed et al, 2009). As the study of neuron-glia interactions requires the ability to identify the subtypes of non-neuronal cells, the use of immunocytochemical methods to identify DRG cells using cell type-specific antibodies is useful.
DRG neurons in vitro are commonly identified using antibodies against class III p- tubulin (TUJ-1) (Backstrom et al., 2000; Ng et al, 2010)，while non-neuronal cells, including SGCs, Schwann cells and fibroblasts, are identified by many different cell type-specific antibodies. However, a majority of these antibodies were used in
32 immunohistochemical studies (Jasmin et al, 2010; Zhang et al, 2009) but rarely in in vitro cell-based studies (Capuano et al, 2009; Ceruti et al, 2011).
In the study of DRG cells in vitro, by the use immunocytochemical method, we looked for antibodies that recognize antigenic markers which are expressed on a specific type of DRG cells. Glial fibrillary acidic protein (GFAP) is an antigenic marker for DRG glial cells. A low level of GFAP expression on DRG glial cells was reported in vivo during normal state and the expression of GFAP was found to be increased upon activation such as following nerve injury (Vit et al.’ 2006). GFAP is therefore frequently used as a marker of glial cell activation (Xie et al., 2009; Zhang et al, 2009). To specifically identify SGCs, many different antibodies have been used, including antibodies that recognize SGC-specific glutamate transporter,
GLAST (Ohara et aL, 2008), small-conductance Ca^'^-activated K"^ channel, SK3
(Vit et aL, 2006), and GS (Hanani, 2005). For Schwann cells, antigenic markers such as galactocerebroside (GalC) (Sobue et al, 1986)，SIOOP (Campana, 2007), and
CNPase (Toma et al, 2007) have been used. For fibroblasts, anti-Thy-1 antibody
(Fields, 1980) has been used.
It is common to use immunocytochemical methods to characterize cultured cells.
The uses of preservative and different fixatives have been adapted to work, with the
immunocytochemical methods. Preservatives, such as kryofix preserve cells (Khaira
33 et al., 2009; Raye et aL, 2007), were used in order to add flexibility on when immunocytochemistry is performed. The roles of fixation are to maintain cellular structure, retain antigenicity and prevent autolysis, and thus aid antibody binding
(St-Laurent et al, 2006). Two commonly used fixatives are acetone methanol (1:1) solution and 4% PFA solution. The two fixatives have different fixation mechanisms to aid antibody binding. The acetone methanol (1:1) solution fixes cells by denaturing and modifying the tertiary structure of proteins while preserving antigenic sites for antibody binding (St-Laurent et al., 2006). PFA solution, on the other hand, fixes cells by insolubilizing intracellular substances such as carbohydrates, lipids and nucleic acids and forming cross-link protein molecules by forming covalent bonds. Different immunochemical methods may give different fluorescence patterns and intensities. These may affect subsequent interpretation of immunoreactive cells and hence, an optimal method of fixation often needs to be determined for each cellular marker (St-Laurent et al.，2006). Using adequate preservative and fixation methods are therefore important for immunocytochemical analysis.
4.2 Aim of study
Isolated DRG cell cultures contain a mixture of neurons and non-neuronal cells
including SGCs, Schwann cells and fibroblasts. To fully study the properties and
34 interactions among DRG cells, it is necessary to distinguish the neurons and different glial cells from one another. With the different antigenic markers available for the identification of various DRG cells, the aim of this study was to identify the types of non-neuronal cells including glial cells (SGCs and Schwann cells) and fibroblasts present.
4.3 Results
4.3.1 Identification of DRG cells in isolated cultures
A typical dissociated DRG cell culture contains a mixture of neurons and non- neuronal cells. This mixture of cells could be identified under the view of phase- contrast microscope (Fig. 4.1 A). DRG neurons were identified in cultures with larger,
spherical and phase-bright appearance (arrows) while non-neuronal cells were
identified with smaller, elongated and phase-dark appearance (arrow heads). The preparation of neuron-enriched cell cultures eliminated over 70% of the proliferating
non-neuronal cells, while the preparation of glial cell cultures eliminated over 99%
of neurons. On day 2, the estimated purity of neurons was 20% in mixed DRG cells
and was around 40% in neuron-enriched cells (Fig. 4.1 A and B). The estimated
purity of glial cells was over 99% in glial cell cultures (Fig. 4.1C). To control the
proliferation of contaminating glial cells, we have previously incubated neuron-
containing cells with arabinoside C (10 pM), 5-fluro-2‘-deoxyuridine (50 ^iM) and
35 uridine (150 |LIM) but these drugs failed to produce marked effect on limiting glial cell numbers.
To characterize different DRG cells in cultures, various antibodies that recognize antigenic markers on neurons, glial cells and fibroblast were used. Anti-TUJ-1 antibodies were used for the identification of DRG neurons, anti-GFAP antibodies were used to identify activated glial cells, and Hoechst 33342 stains were used to label cell nuclei (Fig. 4.2). All neurons and the associated neurites were immunoreactive for TUJ-1 (red) while activated glial cells were immunoreactive for
GFAP (green). GFAP-immunoreactive glial cells were grouped into two broad morphological categories. The first group had a round nucleus and spreading cytoplasm (Fig. 4.2F; arrows) and the other group had an elongated nucleus with a spindle shape and was bipolar or multipolar with fine projections (arrow heads).
TUJ-1- and GFAP-negative cells were identified with Hoechst 33342 stain, in which only cell nuclei were stained blue (asterisks).
4.3.2 Activation and proliferation of glial cells in isolated cell cultures
GFAP is frequently used as a marker of glial cell activation. With this characteristic,
I the activation and proliferation of glial cells in the three DRG cell cultures were
assessed over five days in culture. As neuron-enriched cells and glial cells were
36 prepared one day after the mixed DRG cell cultures, representative images of neuron-enriched and glial cells were shown from day 2. Glial cells proliferated over time in all three cultures as indicated by the increase of blue cell nuclei (Fig. 4.3).
Comparing the representative fluorescence images of mixed DRG cells on day 1 and day 5 (Fig. 4.4A-B), it suggested an increase in glial cell number, GFAP- immunoreactive glial cells and fluorescence signal intensity of GFAP. For cell quantification analyses, two fluorescence images from random fields were taken at
20 X magnification and over 100 cells per image were counted in a double-blinded manner. This preliminary study (n = 1) showed the increase in glial cells number over the time in all the cell cultures (Fig. 4.4C). Quantifying the GFAP- immunoreactive glial cells, the percent of GFAP-immunoreactive glial cells was 61
士 50/0 in mixed cell, 33 士 6o/o in neuron-enriched cells and 44 士 4o/o in glial cells on day 2 (Fig. 4.4D). The percent of GFAP-immunoreactive glial cells increased over time, showing 80 士 3% in mixed cells, 74 士 1% in neuron-enriched cells and 70 士
2% in glial cells on day 5. Besides, glial cells with absent, low or high GFAP fluorescence intensity in the mixed DRG cells were assessed by scoring the GFAP fluorescence intensities from 0 to 9, where 0 represented absence of GFAP characteristics, 1 - 4 represented low GFAP fluorescence intensity and 5-9 represented high GFAP fluorescence intensity. Two representative images of day 1
37 and day 5 (Fig. 4.4E) were assessed with naked eye in a double-blinded manner.
Results indicated that not only did the percent of GFAP-immunoreactive glial cells increase, the GFAP fluorescence intensity of activated glial cells was also shown to increase over time. In mixed cells on day 1, the majority of glial cells (53 士 6o/o) were GFAP-negative, with only 4 士 P/o of glial cells showing high GFAP fluorescence intensity. The GFAP fluorescence intensity increased over time, in which 13 ± 3% of glial cells were GFAP-negative and 58 士 6o/o of glial cells showed high GFAP fluorescence intensity on day 5. Neuron-enriched and glial cell cultures demonstrated the similar pattern of increasing GFAP fluorescence intensity of glial cells over time.
4.3.3 Identification of glial cells in cultures
The isolated DRG cell cultures contained mixtures of neurons and glial cells. The use of GFAP antibodies identified only a portion but not all glial cells, GFAP- negative cells could be SGCs, Schwann cells or fibroblasts. To identify the different non-neuronal cells in isolated DRG cultures, different non-neuronal cell-specific antibodies were used. For the identification of SGCs in dissociated DRG cell cultures, antibodies against antigenic markers expressed on SGCs were used. In addition to antibodies against GFAP (Fig. 4.5), SGC-specific antibodies against
38 GLAST (Fig. 4.6), GS (Fig. 4.7) and SK3 were used (Fig. 4.8). For the identification of fibroblasts, antibodies against Thyl.l were used (Fig. 4.9). For the identification of Schwann cells, antibodies against Schwann cell-specific antigenic markers such as
CNPase (Fig. 4.10), SlOOp (Fig. 4.11) and GalC (Fig. 4.12) were used. Despite the wide use of these antibodies on DRG or TG sections, many of these antibodies have rarely been used in DRG cell cultures. The dilutions of antibodies used in this study
were referenced from previous studies that identified different non-neuronal cells in
sections or cell cultures of DRG or TG. In this study, except for GFAP antibodies
that produced specific staining pattern (Fig. 4.5D) in vitro, all other antibodies
against non-neuronal cell-specific antigenic markers tested on mixed DRG cells
were detected on both neurons and non-neuronal cells with no specific binding
patterns when fixed with acetone methanol (1:1) solution. For example, although
GLAST is antigenic protein that had been shown to be expressed in vivo on SGCs
only, antibodies against GLAST were detected on both neurons (Fig. 4.6D; arrows)
and non-neuronal cells (arrow heads) generating no specific and sharp staining
pattern in vitro. Either increasing or decreasing the antibody dilutions relative to the
suggested dilution did not enhance the specificity of the binding of antibodies (data
not shown). Incubations with other non-neuronal cell-specific antibodies showed
similar non-specific staining patterns. The binding of non-neuronal cell-specific
39 antibodies on DRG neurons and non-neuronal cells prevented the identification of glial cell subtypes and fibroblasts in dissociated cultures. Pure fibroblasts isolated from adult rat cardiomyocytes were kindly provided by Ms Jessica Law (CUHK).
These fibroblasts were stained with anti-Thyl.l antibodies (Fig. 4.9F). Fibroblasts were fixed with acetone methanol (1:1) solution without kryofix. All fibroblast were labelled with anti-Thyl.l antibodies revealing its irregular cell shape, large and spread cytoplasm. Assuming that any fibroblasts in isolated DRG cell cultures had similar morphology as what was seen in Fig. 4.9F, it was suggested that there were no fibroblasts in our isolated DRG cell cultures.
4.3.4 Modification of staining methods
In an attempt to produce specific staining patterns for the non-neuronal cell-specific antibodies for the identification of glial cell subtypes and fibroblasts in isolated cultures, the effect of using cell preservative and different fixatives was studied. The use of a cell preservative, such as kryofix, preserves cells and adds flexibility to when immunocytochemistry is performed. Our previous studies used kryofix solution to preserve cells and stain cells with anti-TUJ-1, -GFAP, -CGRP,- IB4, -IP
I and EP4 and -Trk A antibodies (Ng et al, 2011; Ng et al, 2010) without suggesting any problem with the use of kryofix solution. However, to our knowledge, the effect
40 ofkryofix solution on specific antibody binding in DRG cells has not been evaluated.
As a result of that, we studied the effect of kryofix on specific antibody binding by
incubating DRG cells with kryofix for 4 h (kryofix-treated group) before fixing cells
with acetone methanol (1:1) solution. The staining patterns and fluorescence
intensities of non-neuronal cell-specific antibodies were compared with those of the
DRG cells without incubation with kryofix (control group). Results indicated that
among all the non-neuronal cell-specific antibodies tested, neither kryofix-treated
group nor control group produced specific antibody binding patterns (Fig. 4.5-4.12,
D-E). Non-neuronal cell-specific antibodies were detected on both neurons and non-
neuronal cells in both control and kryofix-treated groups. Comparing the
fluorescence intensities of GFAP antibodies between the two groups, GFAP-
immunoreactive glial cells in the kryofix-treated group were not as sharp as and as
bright as those in the control group (Fig. 4.5D-E). This indicated that in general,
kryofix treatment reduced antibody binding and hence reduced the ability to identify
immunoreactive cells and their shape for glial cell subtype identification. As a result,
the use ofkryofix to preserve DRG cells is not recommended.
We have previously showed the presence of EP4 and IP receptors on DRG neurons
and glial cells using 4% PFA solution (Ng et al., 2010). The effect on the use of
different fixatives including acetone methanol (1:1) solution and 4% PFA solution,
41 effects were therefore studied and compared. Both groups of cells that were fixed with acetone methanol (1:1) solution and 4% PFA solution gave non-specific binding patterns (Fig. 4.5-4.12, C-D). Except for anti-GFAP antibodies, all other antibodies labelled both neurons and glial cells that prevented the identification of different glial cells and fibroblasts. Comparing the two fixatives, the fluorescence signals of
GFAP-immunoreactive glial cells were weaker and less distinct when fixed with 4%
PFA solution when compared with acetone 1:1 solution (Fig. 4.5C-D). The results
suggested that fixing DRG cells with acetone methanol (1:1) solution produced
stronger and more distinct fluorescence signal, therefore the use of acetone methanol
(1:1) solution is recommended over 4% PFA solution as DRG cell fixative.
4.3.5 Immunohistochemistry to identify DRG cells in DRG slices
Results indicated that the use of immunocytochemical methods could not lead to the
identification of the different types of glial cells and fibroblasts in isolated cultures.
With the wide use of non-neuronal cell-specific antibodies in DRG sections, these
antibodies were tested in DRG cell sections to identify different glial cells and
fibroblasts (Fig. 4.5-4.12, A-B). As the different types of glial cells and fibroblasts
I
are confined in distinct locations, the specificities of different antibodies against
different types of non-neuronal cells could be evaluated. As expected, GFAP
42 immunoreactivity (Fig. 4.5B) was not detected on whole DRG acutely isolated from uninjured rat. All the antibodies against SGC-specific markers, including GLAST,
GS and SK3 were detected on SGCs but not on neurons and Schwann cells of DRG
sections as expected (Fig. 4.6-4.8，B). As indicated by the fluorescence images,
SGCs could be identified by the immunoreactive cells that were around the cell body
of neurons (Fig. 4.6B; arrow heads). Neurons were not labelled with antibodies
against SGC-specific markers (arrows). Schwann cells were confined in regions
enriched with nerve fibres (asterisk) and these cells were not labelled with the
antibodies against SGCs. Fibroblasts, on the other hand, made up the capsule of the
DRG and were found on the outer region. However, all antibodies specific for SGCs
were also detected on the outer region of whole DRG. The antibodies against
fibroblast-specific marker, Thy 1.1, were detected on both neurons (Fig. 4.9B; arrows)
with weak fluorescence intensity. Antibodies against Schwann cell-specific markers
including CNPase, SlOOp and GalC showed different binding patterns on DRG
sections (Fig. 4.10-4.12，B). Anti-SlOOp antibodies were detected on SGCs and the
outer region of DRG but were not detected on neurons and Schwann cells (Fig. 4.11).
Anti-CNPase (Fig. 4.1 OB) and anti-GalC (Fig. 4.12B) antibodies were detected on
(
neurons, SGCs, Schwann cells and fibroblasts in DRG sections.
43 4.3.6 Comparison of antibody staining in whole DRG and isolated DRG cells
The staining pattern of different non-neuronal cell-specific antibodies varied in isolated cultures and DRG sections. Anti-GFAP antibodies were detected on activated glial cells in isolated cultures but were not detected on whole DRG sections, as expected (Fig. 4.5). Besides, as shown in Fig. 4.6-4.8，antibodies against SGCs failed to identify SGCs in isolated cultures but produced a specific staining pattern in tissue sections. For example, while anti-GLAST antibodies were detected on neurons and non-neuronal cells in isolated cultures (Fig. 4.6C-E), anti-GLAST antibodies were detected in SGCs but not on neurons and Schwann cells in DRG sections (Fig.
4.6B). Such staining patterns were also seen with the other antibodies against SGCs including anti-GS antibodies (Fig. 4.7) and anti-SK3 antibodies (Fig. 4.8). These antibodies allowed identification of different glial cells in DRG sections but not isolated cultures. Antibodies against Schwann cell-specific and fibroblast-specific markers did not help identifying the different glial cells and fibroblasts in either isolated cultures or sections (Fig. 4.9-4.12).
4.4 Discussion
I
The DRG neurons and non-neuronal cells were readily distinguished in isolated live cell cultures using phase-contrast microscopy and were distinguished in fixed cell
44 cultures using immunocytochemical methods. When viewed under phase-contrast microscope, neurons were the large phase bright cells while non-neuronal cells were the small phase dark cells. Using immunocytochemical methods, TUJ-1 antibody recognized all DRG neurons of different sizes and their associated neurites with high
specificity. The use of GFAP antibodies revealed the different morphologies of
activated glial cells in cultures. A majority of activated glial cells could be broadly
classified into two morphological classes including the first class of activated glial
cells baring a round nucleus and spread cytoplasm and the other class had an
elongated nucleus and spindle shape, and was bipolar or multipolar with fine
projections. With a longer incubation period (> four days in culture), it was observed
from the fluorescence images that more glial cells were classified as cells baring
flattened nucleus, spindle shape, bipolar or multipolar with fine projections.
However, after four days, the density of glial cells was too intense to study clearly
the different morphologies of individual glial cells and therefore, quantification of
the morphologies cannot be easily achieved. In order to investigate this phenomenon
further, live cell microscopy is needed to visualize and trace the proliferation and
changes of glial cell morphologies over time in cultures at a lower cell density. By
I
doing this, it helps clarify whether (1) glial cells with a particular type of
morphology proliferate faster than the other type or (2) the morphology of a
45 particular type of glial cells changes to the other type over time in culture.
Nevertheless, the glial cells in all three DRG cell cultures proliferated over five days in culture. The increased glial cell proliferation in cultures over time is another piece of evidence to support the hypothesis that isolated DRG cell cultures mimics neuropathic pain state (LaMotte, 2007), as increased glial cell proliferation has also been reported in whole DRG after nerve injury (Elson et al, 2004).
Other than proliferation of glial cells in isolated cultures, glial cells in the three cell
cultures expressed GFAP, a marker for glial cell activation. Increased GFAP
expression has been reported after nerve injury of whole DRG (Xie et al., 2009).
With our results that indicated no GFAP expression in whole DRG section acutely
dissociated from uninjured rat but an increased expression over time in isolated
cultures, these ftirther suggested that isolated DRG cultures may mimic neuropathic
pain states (Xie et al, 2009; Zhang et al., 2009). The proportions of GFAP-
immunoreactive glial cells increased over time and over 70% of total glial cell
population was GFAP-positive in all the three cell cultures after five days. The
proportions of GFAP-immunoreactive cells were larger in neuron-containing cultures
i
than in pure glial cell cultures from day 2 to day 5. These results suggested that glial
cell activation in cultures over time was likely to depend on the presence of neurons.
Furthermore, the fluorescence intensities of GFAP-immunoreactive glial cells 46 increased over time in mixed DRG cells and in neuron-enriched cells and pure glial cells. The fluorescence intensity of GFAP-immunoreactivity is reported to reflect the amount of GFAP fairly well (Norton et al, 1992) therefore our result suggested that the amount of GFAP being produced in glial cells increased over time in cultures.
We demonstrated that not only did the number of GFAP-immunoreactive glial cells
increase, but the amount of GFAP being produced also increased over time in culture.
Taken together, the physiological characteristics of glial cells in isolated cultures
were similar to that of glial cells in neuropathic pain conditions. For example, our
dissociated DRG glial cells proliferated in cultures while this observation has also
been reported in vivo after nerve injury (Elson et al., 2004). Besides, dissociated
DRG glial cells expressed GFAP over time in cultures and this characteristic has also
been suggested in vivo after nerve injury (Xie et al, 2009). It is therefore reasonable
to suggest that DRG cells in isolated cultures may mimic neuropathic pain models in
vivo. In isolated cultures, neurons are axotomized and this condition is similar to
what happens in neuropathic pain models in vivo. Injured neurons may trigger or
prevent the release of factors leading to activation and proliferation of glial cells that
leads to neuropathic pain (Hanani, 2005). :
Antibodies against GFAP were used in the identification of glial cells in this study.
However, there are some drawbacks with the use of GFAP. First and foremost is that
47 GFAP cannot serve as a cell marker for distinguishing SGCs from Schwann cells in
DRG cell cultures. We have previously obtained nestin-positive Schwann cells (a gift from Prof. Daisy Shum, HKU) and demonstrated that Schwann cells were immunoreactive for GFAP (unpublished data). Besides, there is evidence that, non- myelinating Schwann cells also express GFAP (Jessen et al, 1984; Mosahebi et al,
2001; Norton et al.’ 1992). In in vivo study, SGCs and Schwann cells can be distinguished based on their locations relative to neurons in intact DRG, so there is normally no trouble with identifying SGCs and Schwann cells. However, in isolated cultures, there is still no clear way for the identification of GFAP-immunoreactive
SGCs and Schwann cells. Besides, GFAP is primarily used as a glial cell activation
marker, and it is undetectable in normal DRG glial cells (Norton et al., 1992) and the
expression is increased only upon activation (Xie et al, 2009). As a result of that, it
should be noted that morphologies assessed in this study were based on the
quantification of GFAP-immunoreactive (or as defined as activated) glial cells
including both GFAP-immunoreactive SGCs and Schwann cells. The identity of
GFAP-negative cells remains to be confirmed.
To distinguish SGCs, Schwann cells and fibroblasts in isolated DRG cell cultures,
various antibodies were used in this study. However, the antibodies against non-
neuronal cells used were not specifically detected on a subtype of glial cells or
48 fibroblasts. Instead, these antibodies bound to both neurons and other glial cells. For example, GLAST and GS are the two proteins responsible for transporting glutamate from extracellular space into glial cell body for the conversion into glutamine. These proteins were found exclusively in SGCs in whole DRG (Hanani, 2010; Ohara et al,
2008) and they were not known to be present in neurons. However, anti-GLAST and anti-GS antibodies were also detected on both neurons and non-neuronal cells in vitro. These results suggested that the antibodies tested were either" non-specific or that the antigens (GLAST and GS proteins) were expressed on neurons in isolated cultures.
When compared with the specific and sharp staining pattern produced by anti-GFAP
antibodies that allowed the recognition of the shapes of activated glial cells, the other
non-neuronal cell-specific antibodies labelled both neurons and glial cells with very
weak and faint signals, preventing the identification of immunoreactive non-
neuronal cells and their cell morphologies. In an attempt to improve the antibody
binding for the identification of glial cells, we adapted different immunochemical
methods that are commonly used. Firstly, we investigated if the preservation of cells
in kryofix for 4 h would affect the specificity of antibody binding and the
corresponding signal intensity. The use of kryofix to preserve cells has been used in
immunocytochemistry protocols (Khaira et al, 2009; Ng et al, 2011; Ng et al.’ 2010;
49 Raye et al., 2007). However, to our knowledge, there are currently no reports on the effect of kryofix on specific antibody binding. The use of kryofix did not enhance the specific binding of antibodies but reduced the fluorescence signal intensity of the anti-GFAP and anti-CNPase antibodies. Generally speaking, the use of kryofix to preserve cells is not recommended. Other than the use of preservative, the method of fixative was also tested and compared. Results suggested that neither of the fixatives allowed any of the non-neuronal cell-specific antibodies to produce specific antibody binding patterns for the identification of subtypes of glial cells and fibroblasts in cultures. Comparing the two fixatives, the use of acetone methanol (1:1)
solution generally gave sharper and more distinct fluorescence signals that the use of
4% PFA solution. Therefore, the use of acetone methanol (1:1) solution as DRG cell
fixative is recommended over 4% PFA solutions. Having revealed that the use of
different fixatives and the use of kryofix had no effect in enhancing the antibody
binding specificity, it is suggested to optimize the concentration of the blocking
buffer to further the adaption of immunocytochemical protocol. Other studies had
use 15% goat serum (Capuano et al., 2009) or 0.5% BSA (Capuano et al, 2009) to
reduce non-specific binding. In this study, 0.3% donkey serum was used to reduce
f
non-specific binding and increasing the concentration of blocking buffer should be
considered.
50 While we could not optimize the immunocytochemical methods to produce specific staining of non-neuronal cell-specific antibodies for the identification of glial cell subtypes and fibroblasts in isolated cultures, some of these antibodies have been used for identification of glial cell subtypes and fibroblasts by others (Ceruti et al.,
2008; Gladman et al., 2010; Vroemen et al., 2003). However, studies did not always provide the complete evidence on the binding patterns of the antibodies. For example, anti-SlOOp antibodies were used to identify Schwann cells in isolated DRG cell
cultures, but no representative image of S1 OOp-immunoreactive cells was shown,
and even without that, the author suggested that approximately 20% of DRG cells
were Schwann cells (Gladman et al., 2010). Moreover, anti-SlOOp antibodies had
been suggested not be an ideal marker (Hanani, 2005) because this antigen was
shown to be present on SGCs and injured neurons of whole DRG (Vega et al., 1991).
The same also applies to anti-CNPase and anti-Thyl.l antibodies. Anti-CNPase
antibodies had been used to identify SGCs (Ceruti et al, 2008) and anti-Thyl.l
antibodies have been shown to recognize both neurons and fibroblasts in cultures
(Fields et al, 1978; Yang et al., 2008). Although the identification of glial cell
subtypes and fibroblasts in isolated cultures would be useful to study the
composition of different glial cells and fibroblasts in isolated DRG cell cultures,
none of the non-neuronal cell-specific antibodies produced specific antibody binding
51 patterns for identification. This study herein suggested not to rely solely on the results obtained from immunocytochemical studies for cell identification purpose. It is also essential to optimize the staining condition for each antibody, to provide proper positive and negative controls for the immunostaining and to support immunostaining results using other methods.
The use of non-neuronal specific antibodies in whole DRG sections produced more specific antibody binding patterns to identify SGCs in whole DRG. These results revealed that in our study, the failure of identification of isolated non-neuronal cells using SGC-specific antibodies was not due to the non-specificities of the antibodies, but may due the change of antigenic protein expression in dissociated cells cultures over time. This is supported by the evidence that reduced expression of protein mRNA of DRG cells occurred quickly after cells were isolated and plated (Franklin et al, 2009). Besides, the mRNA expression of glial cell-specific GS has recently been tested on glial cell cultures in our lab. The results that GS mRNA expression
dropped from day 1 to day 4 revealed the decrease of the expression of this antigenic
marker over time in cultures.
In order to investigate the antigenic marker mRNA expression in non-neuronal cells
with a particular morphology, one could test the presence of SGC-specific, Schwann
cells-specific, and fibroblast-specific marker mRNA using RT-PCR that allows
52 amplification and detection of mRNA from a single cell. Then one would be certain whether antigenic markers of glial cells subtypes and fibroblasts were expressed in
cultures. The use of antibodies that recognize antigenic markers specifically
expressed on different DRG cells enables the recognition of non-neuronal cells,
including SGCs, Schwann cells and fibroblasts, in isolated DRG cell cultures. This
makes the identification of different DRG cells independent of cellular morphology
in isolated cultures. The identification of different glial cells in isolated cultures is
remained to be investigated in order to reveal the properties of different glial cells
and any possible neuron-glia interactions.
4.5 Summary
In this study, the anti-TUJ-1 and anti-GFAP antibodies were used to identify neurons
and activated glial cells in isolated cultures, respectively. The use of GFAP
antibodies revealed the different morphologies of activated glial cells in culture. The
majority of GFAP-immunoreactive glial cells could be broadly classified into two
morphological classes. The first class of activated glial cells had round nucleus and
spread cytoplasm. The second class of activated glial cells had elongated nucleus
with spindle shape and was bipolar or multipolar with fine projection. It was also
evident that the proliferation of activated glial cells was faster in the presence of
neurons, revealing neuron-glia interactions where neurons influence the proliferation
53 of glial cells. Moreover, the fluorescence intensity of GFAP-immunoreactive glial cells increased over time in cultures, indicating that more GFAP are expressed.
Taken together, these characteristics of glial cells in isolated cultures were similar to those as expressed during nerve injury and therefore, it was suggested that isolated cell cultures mimicked neuropathic pain models.
Other non-neuronal cell-specific antibodies used in this study failed to produce specific staining patterns for the identification of glial cell subtypes and fibroblasts.
Adapting different immunocytochemical methods did not help improve the specific staining patterns. However, these antibodies produced specific antibody staining patterns when used in whole DRG sections. These results suggested that the failure of giving specific staining patterns in isolated cultures was not due to the non- specificity of the antibodies but might be due to the change of expression of antigenic markers of DRG cells in isolated cultures.
54 iMHynatanPEB
Hiii^npH
•S^^ESmBBSnEl^^HI
Figure 4.1 Representative phase contrast microscopic field of typical dissociated
DRG cell cultures. Images of (A) mixed DRG cells, (B) neuron-enriched cells and
(C) glial cells were taken after two days. Neurons (phase-bright cells; arrows) and ( non-neuronal cells (phase-dark cells; arrow heads) were identified in (A). Scale bar represents 25 |im.
55 •• •• Figure 4.2 Identification of neurons and glial cells in dissociated DRG cell cultures. Representative fluorescence microscopic images of (A-B) mixed DRG cells, (C-D) neuron-enriched cells and (E-F) glial cells of isolated DRG cells after two days in cultures taken at two mangifications. Neurons were identified with anti-
TUJ-1 antibodies (1:1000; Alexa fluor 647 secondary antibodies, 1:1000; red). Glial cells were identified with anti-GFAP antibodies (1:500; FITC anti-mouse secondary antibodies, 1:500; green), illustrating spindle shaped cells with elongated nucleus (F, arrow heads), and cells with round and spreading cytoplasm (F, arrows). Cell nuclei were identified with Hoechst 33342 stain (blue, asterisk). Scale bars represent 25 ^im.
56 们 ^^Q ^ 寸 •••••••• •
s
3
s
I
s
Figure 4.3 Fluorescence images of glial cells in isolated DRG cultures over time. Representative fluorescence images of (A-E) mixed DRG cells,
(F-I) neuron-enriched cells and (J-M) glial cells over five days. Cells were fixed with acetone methanol (1:1) solution. Glial cells were identified
with anti-GFAP antibodies (1:500; FITC secondary antibodies, 1:500; green). Cell nuclei were identified with Hoechst 33342 stain (blue) indicating
the increase in glial cell numbers of the three cell cultures over five days. Scale bar represents 100 jim.
-： cn 00 69 • I
D O GFAP-immunoreactive Number of glial cells per glial cell population (%) standard field � -A -A lo lo o ^ fe S g g ?????? � \ iil 1 T Szs ro. HHO ^ t pV*
oiJ I-CHIHI^
Percent of total glial cell population (%) m to w Ol o> N oooooooo
芸 H •• IH II
G)
‘- H I ' Figure 4.4 Proliferation and GFAP-immunoreactivity of glial cells over time in isolated cultures. Representative fluorescence images of
mixed DRG cells on (A) day 1 and (B) day 5 visualizing increase in glial cell numbers (blue cell nuclei) and GFAP-immunoreactive glial cells
(green). Scale bar represents 25 ^m. (C-D) Quantification of the number of glial cells and the percentage of GFAP-immunoreactive glial cells
from two representative fluorescence images (counting over 100 cells per image) of mixed DRG cells, neuron-enriched cells and glial cell
cultures over five days. (E) Quantification of glial cells on day 1 (open bar) and day 5 (closed bar) in terms of GFAP fluorescence intensities
with the absence, low or high GFAP fluorescence intensity from two representative fluorescence images (counting over 100 cells per image) of
mixed DRG cell cultures. Data represent means 士 SD from one experiment.
g Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1) -mm‘'.’-— ••• Fig. 4.5. GFAP-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images of(A-B)
DRG sections and (C-E) mixed DRG cells incubated with anti-GFAP antibodies (1:500; Alexa fluor 647 secondary antibodies, 1:1000; red). DRG 2 withousectionts anwerd e(E fixe) witdh wit kryofixh 4% .PF CelAl solutionuclei nwer aned identifieisolatedd mixe withd Hoechs DRG cellt 3334s wer2 staie fixen (blue)d with. Scal(C) e4 %bar PFs represenA solutiot n2 5o r\im. aceton e methanol (1:1) solution (D) Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1) ^^^^ ， ….^^^^^^^^^^^^^^^^^^ (1:1) solution solution with kryofix EB ••mBm •
Fig. 4.6. GLAST-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images of(A-B)
DRG sections and (C-E) mixed DRG cells incubated with anti-GLAST antibodies (1:1000; FITC secondary antibodies, 1:100; green). DRG sections
were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone methanol (1:1) solution (D) without
and (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars represent 25 ^im. Arrows and arrow heads represent
S neurons and glial cells respectively in (B) DRG sections and (D) isolated DRG cell cultures. Asterisk represents region enriched with Schwann cells. Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1) ^^^^ ， ….^^^^^^^^^^^^^^^^^^ (1:1) solution solution with kryofix 一國•
Fig. 4.7. GS-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images of
(A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-GS antibodies (1:1000; FITC secondary antibodies, 1:100; green). DRG
sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone methanol (1:1)
solution (D) without-如d (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars represent 25 ^im. o> CO Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1)
(1:1) solution solution with kryofix • • —••• Fig. 4.8. SK3-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images of
(A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-SK3 antibodies (1:1000; FITC secondary antibodies, 1:200; green). DRG
sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone methanol (1:1)
solution (D) without and (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars represent 25 ^im. 2 Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1) ^^^^ ， ….^^^^^^^^^^^^^^^^^^ (1:1) solution solution with kryofix
-••• —•••
Fig. 4.9. Thyl.l-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images
of (A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-Thyl.l antibodies (1:2000; FITC secondary antibodies, 1:100; green).
DRG sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone methanol (1:1)
solution (D) without and (E) with kryofix. (F) Fibroblasts enriched from cardiomyocytes fixed with acetone methanol (1:1) solution without
g kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Arrows represent Thy 1.1 immunoreactive neurons in (B) DRG sections and (D) isolated DRG cell cultures. Scale bars represent 25 ^un. Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1)
(1:1) solution solution with kryofix • • —••• Fig. 4.10. CNPase-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence
images of (A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-CNPase antibodies (1:1000; Alexa fluor 647 secondary
antibodies, 1:1000; red). DRG sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution
or acetone methanol (1:1) solution (D) without and (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars o> represent 25 |im. Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1)
(1:1) solution solution with kryofix • • -••• Fig. 4.11. SlOOp-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images
of (A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-SlOOp antibodies (1:1,000; Alexa fluor 647; secondary antibodies,
1:1000; red). DRG sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone
methanol (1:1) solution (D) without and (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars represent 25 o> Phase contrast 4% PFA solution Acetone methanol Acetone methanol (1:1) ^^^^ ， ….^^^^^^^^^^^^^^^^^^ (1:1) solution solution with kryofix
—•••
Fig. 4.12. GalC-immunoreactivity in DRG sections and isolated mixed DRG cells. Representative phase contrast and fluorescence images
of (A-B) DRG sections and (C-E) mixed DRG cells incubated with anti-GalC antibodies (1:1000; Alexa fluor 647 secondary antibodies,
1:1000; red). DRG sections were fixed with 4% PFA solution and isolated mixed DRG cells were fixed with (C) 4% PFA solution or acetone
methanol (1:1) solution (D) without and (E) with kryofix. Cell nuclei were identified with Hoechst 33342 stain (blue). Scale bars represent 25
00 阿. Chapter 5
Characterization of GPCRs in isolated DRG cultures
5.1 Introduction
5.1.1 G-protein coupled receptors
The G-protein coupled receptors (GPCRs) represent a large family of surface receptors which are expressed on all kinds of cells. They are essential for transducing external signals, such as hormones, neurotransmitters or sensory stimuli, into intracellular signals in response to external changes (Pierce et al., 2002). The GPCRs represent important therapeutic targets in drug design. The binding of a GPCR by an agonist or antagonist triggers or blocks the activation of the receptor that ultimately alters downstream signal-transduction pathways to modify intracellular systems, such as to regulate enzyme or ion channel activities. A GPCR classically consists of a seven transmembrane (7TM) domain and is associated with a heterotrimeric guanine- nucleotide regulatory protein (G protein) (Wettschureck et al, 2005). The 7TM protein is composed of amino acids, which starts with an extracellular N-terminus, spanning across the plasma membrane seven times connected by intracellular and extracellular loops, and ends with an intracellular C-terminus. Although all GPCRs share the common feature of possessing the 7TM domain, the amino acids sequence of different GPCRs share almost no overall homologies (Probst et al, 1992).
69 However, significant amino acid sequence homology can be found within some
GPCRs, which allows the classification of three major classes of GPCRs (Foord et al, 2005). The largest class among the three is Class A (or 1) GPCR, which comprises over 80% of the total GPCRs. This class of GPCRs includes hormone receptors, neurotransmitter receptors and light receptors, which share highly conserved amino acids, namely a glutamic acid/aspartic acid-arginine—tyrosine
(E/DRY) motif, at the cytoplasmic end of the third transmembrane domain (Probst et al, 1992). Class B (or 2) GPCRs consist of around 25 members, which includes a variety of hormone receptors and neuropeptide receptors. They are mostly Gg- coupled (discussed below) and are characterized by a large (more than 100 amino acid residues) extracellular domain containing several cysteines (Ulrich et al., 1998).
Class C (or 3) GPCRs include metabotropic glutamate receptors, GABAB receptors, and calcium-sensing receptors. All the receptors in class C possess a very large extracellular domain at the amino terminus (around 500 to 600 amino acid residues)
(Gether，2000) and this structure seems to be important for ligand binding and activation (Pierce et al., 2002).
Extracellular signals are transduced by the G proteins to modulate the internal activity of a cell. G proteins are composed of three protein subunits: a subunit, p subunit and y subunit. The a subunit is associated with the 7TM receptor by covalent
70 linkage and has a GDP/GTP binding site, while P subunit and y subunit are associated with each other, forming an undissociable complex (Wettschureck et al,
2005). In the unstimulated state, the a subunit of the G protein is guanine diphosphate (GDP) bound and is associated with the Py subunit (Gilman, 1987).
When the 7TM receptor is bound by an agonist, the agonist induces a conformational change in the 7TM receptor, allowing the 7TM receptor to function as a guanine nucleotide exchange factor that exchanges the GDP on the a subunit with GTP. The
GTP-bound a subunit then leads to the breaking of its linkages with the py subunit and the 7TM receptor. The dissociation of the a subunit and py subunit from the
7TM receptor then allows regulation of intracellular signaling pathways (Gilman,
1987). The 7TM receptor stays active while the agonist binds to it and therefore, it can catalyze the activation of many G proteins.
Both activated a subunit and py subunits regulate the activity of the target proteins, depending on the types of G protein they belong to. There are many different kinds of Ga protein subtypes and those that are related to the generation of pain signals include Gs, Gi/�an dGq/n proteins. Gs protein activates adenylyl cyclase, which increases the conversion of ATP to cAMP (Simon et al., 1980)，whereas Gi/�protein inhibits adenylyl cyclase activity to decrease the conversion of ATP to cAMP
(Wettschureck et al, 2005). Gq/n protein activates phospholipase C-beta (PLC-P)
71 which hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) into diacyl glycerol
(DAG) and inositol 1,4,5-triphosphate (IP3). IP3 triggers the release of intracellular calcium ions that activates protein kinase C (PKC) (Julius et al., 2001). Activation of
PKC is involved in the pain signaling pathway (Hucho et al, 2005) as PKC- knockout mice exhibit reduced thermal and mechanical hypersensitivity after treatment with adrenaline or acetic acid (Khasar et al, 1999a).
I
5.1.2 cAMP/PKA-mediated hyperalgesia
In the DRG, agents that stimulate the production of cAMP tend to facilitate pain through a process called peripheral sensitization (see Section 1.2). Hyperalgesic agents such as prostanoids, including PGE2 and PGI2, bind to Gs-coupled GPCRs, leading to the dissociation of the Gs protein, which then activates the adenylyl cyclase for the conversion of ATP to cAMP (Simon et al, 1980). The cAMP is the secondary messenger that initiates peripheral hyperalgesia by activating the protein kinase A (PKA) pathway (Aley et al, 1999). The exact mechanism of cAMP/PKA- mediated hyperalgesia is not fully understood, but studies revealed that it involves the modulation of voltage-gated Ca^^ channels and ligand-gated ion channels (Bhave et al, 2002; Fitzgerald et al., 1999). Alteration of these channels induces depolarization of sensory neurons and lowers the threshold for generating an action potential (Bley et al., 1998). This sensitizes the neurons and favors repetitive firing
72 of nociceptive signals by sensory neurons. Increased sensitivity to pain then occurs
(Julius et a/., 2001).
In contrast, analgesic agents such as opiates and cannabinoids alleviate pain by counteracting the changes resulting from the action of hyperalgesic agents. These analgesic agents couple to Gj/o-GPCRs, leading to the dissociation of the Gi/o protein that inhibits adenylyl cyclase. The decrease in cAMP production results in the restoration of the decreased threshold to generate nociceptive signals and therefore decreases pain signaling.
5.1.2 Pharmacological characterization of prostanoid receptors on DRG cells
Pharmacological characterization of functional expression of GPCRs involves the use of receptor-specific agonists and antagonists. The agonist should possess high affinity for the target receptors (Hoyer et al., 1994) and produce measurable responses such as the production of cAMP. To confirm that the agonist-stimulated responses are mediated through the target receptor, the antagonist, on the other hand, should demonstrate the ability to abolish the agonist-stimulated responses by blocking the target receptors (Hoyer et al., 1994). The concentrations of agonist and antagonist should be precisely chosen so that the agonist concentration will be adequate to activate the target receptors but not too high for the agonist to lose
73 selectivity for the target receptors, while the antagonist concentration will just be sufficient enough to block all agonist binding to abolish the agonist-stimulated responses.
Our previous study provided evidence for the presence of pain-related prostanoid receptors (EP4 and IP receptors) on both DRG neurons and glial cells in vitro by using pharmacological tools (Ng et al” 2011). PGE2 (EP receptor agonist), ONO-
AEl-329 (EP4 receptor agonist) and cicaprost (IP receptor agonist) stimulated cAMP production in DRG cells. The PGE2- and ONO-AEl -329-stimulated responses were only inhibited by EP4 receptor antagonist ONO-AE3-208 (100 nM), while cicaprost- stimulated responses were only inhibited by the IP receptor antagonist RO1138452
(100 nM). These results demonstrated that PGE2 and ONO-AEl-329 mediated their responses via EP4 receptors, while cicaprost mediated its responses via IP receptors.
Interestingly, glial cells responded well to both cicaprost and PGE2, while only small cicaprost- and PGEi-stimulated responses could be seen in neuron-enriched cells.
Moreover, with approximately the same number of glial cells in mixed DRG cell cultures and glial cell cultures, the cicaprost-stimulated responses in glial cultures were significantly larger than those in the mixed DRG cultures; PGE2-stimulated responses also showed a similar pattern.
74 The smaller agonist-stimulated responses in mixed DRG cells suggested that the presence of neurons in the mixed DRG cells inhibited the cicaprost- and PGE2- stimulated responses of glial cells. Further investigations were carried out to selectively label glial cells with [^H] adenine and then to measure the responses of glial cells in the presence of unlabeled neurons that mimic the environment of mixed
DRG cells. It was seen that the glial cells alone produced most of the cicaprost- and
PGE2-stimulated responses compared to the glial cells with the addition of neurons.
The result clearly demonstrated that the presence of neurons in mixed DRG cell cultures inhibited the cicaprost- and PGEi-stimulated responses in glial cells (Ng et al, manuscript in preparation). As a consequence of this data, we wanted to determine if similar relationships between DRG neurons and glial cells were seen with other GPCRs implicated in the regulation of pain.
5.1.3 Gs- and Gi/o-coupled GPCRs in DRG cells
Many GPCRs are found on the surface of DRG cells, including neurons and non- neuronal cells. These GPCRs are responsible for mediating the release of neurotransmitters, neuropeptides and sensitizing neurons to facilitate pain, and desensitizing neurons to alleviate pain. Among all the GPCRs expressed on the DRG cells, we selected some Gs-coupled and Gj/o-coupled GPCRs that are clearly associated with pain for characterization. Discussed in the following sections are the
75 pain-related Gg- and Gj/o-coupled GPCRs chosen for this study. The agonists and antagonists used for GPCR characterization are listed in Table 5.1.
5.1.3.1 Gs-coupled GPCR: p-adrenoceptors
The adrenoceptors (ARs) are classified into two groups, a-ARs and p-ARs, based on physiological, pharmacological and molecular cloning experiments (Bond et al,
1988; Bylund, 1985; Bylund, 1988). The adrenoceptors mediate various central and peripheral actions through the binding of endogenous sympathetic catecholamines, including adrenaline and noradrenaline. Adrenaline acts equally on both a-ARs and
P-ARs, while noradrenaline acts preferentially on a-ARs (Bylund et al, 1994).
There are three p-ARs subtypes, namely Pi-, p2- and Ps-ARs. They are predominantly expressed in different tissues for diverse functions, pi-ARs are the dominant receptors in heart and adipose tissue; P2-ARS are expressed on, but not limited to, neurons, vascular, uterine, and bronchial smooth muscle cells (Bylund et al., 1994;
Nagatomo et aL, 2000). Ps-ARs are expressed on adipose tissue, stomach, liver, gastrointestinal tract, smooth and skeletal muscles (Bylund et al., 1994; Evans et al,
1999; Leon et al., 2008). Regardless of the p-AR subtypes, P-ARs, which belong to
< class A GPCR, are primarily coupled to Gs proteins, which stimulate adenylyl cyclase activity for the increase of intracellular cAMP level and leads to different
76 physiological responses (Bylund et al., 1994). Activation of Pi-ARs primarily mediates an increase in heart rate and contractile force, the relaxation of coronary arteries and gastrointestinal smooth muscles (Bylund et al, 1994); activation of p2- and p3-ARs primarily mediate the relaxation of smooth muscle cells of peripheral tissues (Bylund et al, 1994; Evans et al., 1999).
The presence of Pi- and Pz-AR mRNA and the absence of P3-AR mRNA has been reported in cultures of rat DRG cells (Leon et al, 2008). Without identifying the specific subtypes, P-AR-mediated responses have been reported in DRG neurons in vitro (Khasar et al., 1999b; Pluteanu et al, 2002) and in vivo (Khasar et al, 1999b).
The direct evidence for the presence of p-ARs on DRG SGCs is not available in vivo, but the presence of P-ARs has been suggested (Hanani, 2010) to mediate the increased cAMP stimulated by isoprenaline (P1/P2-AR agonist) in SGCs of superior cervical sympathetic ganglia in vitro (Ariano et al., 1982). We have previously shown that isolated DRG cells from adult rats responded to isoprenaline and salbutamol (pi/p2-AR and P2-AR agonist) with an increase of cAMP production
(Rowlands et al, 2001).
The major distinction between the subtypes of p-ARs is their relative affinity to their endogenous ligands, adrenaline and noradrenaline (Bylund et al., 1994; Nagatomo et al., 2000). The pi-ARs have equal affinity for noradrenaline and adrenaline, whereas
77 P2-ARs have a 100-fold higher affinity for adrenaline than noradrenaline (Bylund et al, 1994; Nagatomo et al, 2000). In contrast, p3-ARs have a higher affinity to noradrenaline than adrenaline (Bylund et al, 1994). While there are many potent and selective p2-AR agonists that allow the characterization of the P2-ARS, the Pi-AR specific agonists are currently scarce. Among the few Pi-AR agonists, T-0509 is shown to be ppAR specific (Longhurst et al., 1999; Sato et al., 1996), but it is currently unavailable commercially. Pharmacological characterizations of the Pi-ARs are based on the studies that have used non-selective P1/P2-AR agonist, isoprenaline, to increase cAMP production (Jung et al., 1997; Pluteanu et al, 2002), and those that have used highly selective Pi-AR antagonists, such as atenolol at 10 |aM, to block the isoprenaline-stimulated responses (Morioka et al., 2009). Isoprenaline has a Ki value of 14 nM and 21 nM for rat Pi-ARs for P2-AR, respectively (Naito et al, 1985), while atenolol displays selectivity for Pi-ARs over p2-ARs with Kj values 125 nM and 1250 nM, respectively (Bylund et al, 1994). Specific agonists for P2-ARS include formoterol (Yalcin et al, 2010) while ICI 118551 is commonly used as the p2-AR specific antagonist (Bylund et al., 1994; Morioka et al, 2009; Yalcin et al.,
2010). Formoterol has a Ki value of 1:6 nM for P2-AR (Yalcin et al, 2010) and another study reported that formoterol displays 89-fold more selectivity for p2-ARs
78 than Pi-ARs (Kurose et al., 1998). The antagonist for p2-ARs，ICI 118551, has a Ki value of 4.6 nM for Pz-ARs and 710 nM for Pi-ARs (Bylund et al., 1994).
In this study, the concentrations of isoprenaline (pi/p2-AR agonist) and formoterol
(p2-AR agonist) were both 1 jjM, so that isoprenaline would be sufficient to activate both the p-ARs, while formoterol would only be selective for p2-ARs. Atenolol (pi-
AR antagonist) was chosen at 1 liM and ICI 118551 (p2-AR antagonist) was chosen at 100 nM, so that the antagonists would be sufficient to block all the Pi-AR and P2-
AR responses, respectively.
5.1.3.2 Gs-coupled GPCR: CGRP receptors
CGRP, a well-known hyperalgesic agent, is a 37-amino acid vasodilatory neuropeptide that is strongly associated with migraine (Doods et al., 2007; Walker et al, 2010). A direct role of CGRP in migraine can be simply revealed by the elevation of migraine-like headache after the peripheral injection of CGRP (Lassen et al, 2002; Zhang et al, 2007b). CGRP is released from the activated trigeminal ganglia (TG) nerves, and causes vasodilation of cerebral and dural vessels, which then leads to the transmission of pain signals from intracranial vessels to the CNS
(Durham, 2004). Other than migraine, CGRP has also been found to relate to inflammation, allodynia and other diseases, such as rhinosinusitis (Vause et al, 2010)
79 and cardiovascular disease (Walker et al, 2010). CGRP is widely distributed in the
CNS and PNS and is a very abundant peptide in nervous tissue (Li et al., 2008).
Around 50% of the small- and medium-diameter peptidergic neurons in the DRG synthesize and release CGRP from nerve endings (Segond von Banchet et al.，2002).
Moreover, CGRP has also been found to be released from the cell bodies of trigeminal neurons in the TG, which affects (1) neighboring neurons in an autocrine manner and (2) the associated glial cells in a paracrine manner (Thalakoti et al., 2007;
Vause et a/., 2010).
CGRP stimulates CGRP receptors, which are heterodimers formed between a class B
GPCR, named calcitonin receptor-like receptor (CLR)，and a transmembrane protein, named receptor activity-modifying protein (RAMP) 1 (Lennerz et al, 2008; Li et al.,
2008; Vause et al., 2010; Zhang et al., 2007a). For proper receptor function, an additional protein called receptor component protein (RCP) is also required (Walker et al., 2010). CGRP receptors couple mainly to Gs proteins, but they can also couple to Gi/o proteins (Disa et al, 2000; Walker et al, 2010) and Gq/n proteins (Drissi et al,
1998; Morara et a!., 2008; Walker et al, 2010). Historically, two isoforms of CGRP receptor, CGRPi and CGRP2, have been described (Moreno et al., 2002; Segond von
Banchet et al, 2002). However, whether CGRP2 exists is still in doubt. It is suggested that CGRP2 receptor might actually be the result of CGRP binding to
80 human adrenomedullin (AMi and AM2) receptors, which are both structurally similar to CGRP receptors but are composed of different RAMP isoforms (Hay et al., 2004;
Walker et «/.,2010).
Although CGRP receptors have been studied extensively in TG for their strong association with headache, CGRP receptors are also shown to be expressed on DRG neurons in vivo and in vitro (Ceruti et al, 2011; Segond von Banchet et al., 2002; Ye et al, 1999). It has also been suggested that the activation of CGRP receptors
sensitizes neighboring neurons for the release of mediator (Segond von Banchet et al.,
2002). CGRP receptors are also reported to be present on SGCs in TG both in vivo
and in vitro (Ceruti et al, 2011; Eftekhari et al, 2010), suggesting a role of glial cells
in migraine pathology (Ceruti et al, 2011).
CGRP is a potent agonist for CGRP receptors. A study characterizing CGRP receptor
subtypes in COS-7 cells transfected with CLR and RAMPs indicated that rat CGRP
had affinity for CGRP receptors (EC50： 0.86 nM) as well as AMi (EC50： 147 nM) and
AM2 (EC50： 59 nM) receptors (Husmann et al, 2000). CGRP antagonists are
relatively scarce. CGRPg-s? and BIBN4096BS are the two antagonists for CGRP
(Doods et al, 2000), but BIBN4096BS is currently commercial unavailable. CGRPg.
37，a truncated form of CGRP, is an antagonist (Chiba et al., 1989), with CGRPs-s?
(10 ^M) inhibiting almost 100% of the CGRP-stimulated cAMP production in
81 human neuroblastoma cell lines (Doods et al., 2000). In our previous study, CGRP was used at 1 |aM in order to produce responses which were similar to that stimulated by cicaprost and PGE2 in mixed DRG cells (Rowlands et al, 2001). Therefore, 1 |iM
CGRP was also used in this current study, and 10 pM CGRPg-s? was used to inhibit the CGRP-stimulated cAMP production.
5.1.3.3 Gi/o-coupled GPCR: a2-adrenoceptors
Pharmacological and biochemical studies on the a-ARs have accumulated that indicate the heterogeneities of distinct but related a-ARs, which are the ai- and 012-
ARs (Bylund et al., 1994; Graham et al, 1996; Minneman et al., 1994). The ai-ARs predominantly couple to Gq/n proteins while the az-ARs couple to Gi proteins
(Summers et al, 1993). While activation of both ai-ARs and 0x2-ARs lead to the
contraction of smooth muscles, activation of neuronal a2-ARs has been suggested to
relate to pain (Hein, 2006).
The a2-ARs are found in brain, blood platelets, vascular smooth muscles, skeletal
muscles and nerve endings (Bylund et al, 1994; Lorenz et al, 1990). The subtypes
of ai-ARs (a2A-AR, azs-AR and azc-AR) were characterized by the different affinity
I
for prazosin, an ai/a2-AR antagonist (Lorenz et al, 1990). The a2A-AR had low
affinity for prazosin and was found in the platelet (Bylund et al, 1994; Lorenz et al.,
82 1990). The a2B-AR, showing at least 30-fold higher affinity for prazosin than aiA-AR, was characterized in neonatal rat lung. The azc-AR showed similar affinity for prazosin and was found in opossum kidney tissue (Bylund et al., 1994). Activation of the 012-ARs have been found to mediate functions such as reduction in blood pressure, and sedation and inhibition of fluid secretion, and produce an analgesic effect (Maze et aL, 1991).
The presence of a2-ARs on DRG neurons has been reported in vivo (Shi et al., 2000) in both normal and nerve injured rat model (Cho et al, 1997; Gold et al, 1997;
Supowit et al, 1998). In some in situ hybridization studies, a2c-AR mRNAs were shown to be present in the DRG neurons (Ma et al, 2005; Nicholson et al., 2005).
The a2A-AR expression was found to be up-regulated in DRG neurons following neuropathy and responded to noradrenaline that was released by the nearby
sympathetic efferent nerve fibers (Shi et al, 2000). This suggested the role of a2-ARs in influencing nociception and pain processing (Yalcin et al, 2010). However, the expression of a2-ARs has not been investigated in DRG glial cells.
The activation of az-ARs mediates signaling pathways via both Gu and Gpy subunits.
The a2-ARs couple to Gaj and the activation of the inhibitory Gai protein inhibits the
adenylyl cyclase activity, thus resulting in decreased cellular cAMP levels. Moreover, the Gpy subunit regulates neuronal function by inhibiting neuronal Ca^^ channels and
83 activating GIRK K+ channels and mitogen-activated (MAP) kinases ERK1/2 (Hein,
2006). The consequences of the activation of the a2-ARs include suppression of the entry of Ca^^, hyperpolarization of the sensory neurons, inhibition of nerve impulse firings, a decrease in neurotransmitter release and hence, resulting in the production of an analgesic effect (Maze et al, 1991). However, there has been opposite evidence that os-ARs play an important role in generating sympathetically maintained neuropathic pain (Cho et al, 1997).
Selective agonists for a2-ARs include clonidine and UK14304, while selective antagonists include yohimbine (Bylund et al., 1994). To characterize the ai-ARs,
UK14304 is preferred over clonidine because UK14304 acts as a full agonist while clonidine acts as a partial agonist for the a2-ARs (Maze et al., 1991). UK14304 has a
Ki value of 2.4 nM for 012-ARs and 1900 nM for ai-ARs (Munk et al, 1995). In this study, the concentration of UK14304 was chosen at 1 |iM so that the concentration is sufficient to activate az-ARs but is not too high to lose the selectivity. The antagonist, yohimbine, has a KD value of 100 nM for 012-ARs (Pushpendran et al., 1984) and 1
|iM was chosen so that the concentration of the antagonist would be high enough to block any a2-AR responses.
84 5.1.3.4 Gi/o-coupled GPCR: Cannabinoid receptors
Cannabinoid (CB) receptors are known for their roles in mediating the actions of the cannabinoid drugs that were isolated and identified from a plant species Cannabis sativa, including marihuana, hashish, and bhang (Gaoni et al., 1964). Currently, there are two CB receptors being described, which are the cannabinoid receptor 1 and 2
(CBl and CB2). CBl and CB2 share 44% of the amino acid sequence and differ in their tissue distributions (Howlett et al, 2002). One of the most well-known effect of cannabinoid receptor-mediated action is analgesia (Howlett, 1995). The CB receptors are classified as Class A GPCRs that couple to Gi/�protein tso inhibit adenylyl cyclase activity (Foord et al” 2005; Pertwee, 1997). CBl receptors are found predominantly at central and peripheral nerve terminals mediating inhibition of neurotransmitter release, while CB2 receptors are found predominantly on immune cells modulating cytokine releases (Pertwee, 2005). Having responsibility for producing analgesic effects, it is not a surprise that CB receptors are found in DRG.
The CBl receptors have been well-characterized for their expression in DRG
neurons in situ, and in vitro (Ahluwalia et al, 2000; Bridges et al., 2003; Fischbach
et al., 2007), however, expression of CBl receptors on glial cells is not known.
4
Other than immune cells, the expression of CB2 receptors has recently been
suggested on DRG neurons both in vivo and in vitro (Hsieh et al, 2011; Sagar et al.’
85 2005). Other study has shown that CB2 receptors were found on activated microglial cells, which are the immune-like cells in the CNS (Zhang et al., 2003) but expression of CB2 receptors on DRG glial cells is not known.
The main psychoactive component of cannabis is A9-tetrahydrocannabinol (A9-
THCl). It is responsible for the CNS effects, such as sedation, and PNS effect such as analgesia (Hewlett et al, 2002). There are endogenous CB receptor ligands which are referred as endocannabinoids, including eicosanoids such as arachidonylethanolamide, also known as anandamide, and 2-arachidonoylglycerol, also known as noladin ether (Howlett et al, 2002; Smith et al, 1994; Sugiura et al.,
1995). Synthetic CB receptor agonists include CP55940, R-(+)-WIN55212 and the
A9-THC1 analog, HU-210. Although these agonists have different potencies towards the CB receptors, they display no remarkable selectivity for the subtypes of the CB receptors (Howlett et al, 2002). The development of CB receptor-specific agonists was based on the structural modification of anandamide. Methanandamide is the first
generation of CBl selective agonist modified from anandamide (Howlett et al, 2002;
Khanolkar et al., 1996). Being a molecule that contains a chiral center, the isomers of
methanandamide display different affinities for the CBl receptor. It is reported that
the R-(+)-methanandamide has nine-fold higher affinity than the R-(-)-
methanandamide for the CBl receptors (Abadji et al, 1994). R-(+)-methanandamide
86 has a Ki value of 20 nM and 815 nM for CBl (rat brain) and CB2 (mouse spleen) receptors, respectively (Khanolkar et ah, 1996). To selectively activate CBl receptors, the agonist R-(+)-methanandamide was chosen at 100 nM so that at this concentration, only CBl but not CB2 will be activated. The CBl antagonist, AM251, has a Ki value of 7.49 nM and 2290 nM for CBl (rat brain) and CB2 (mouse spleen) receptors (Lan et al., 1999). Being a 306-fold more selective for the blockade of CBl than CB2, AM251 was initially chosen at 1 fiM so that it was sufficient but selective to inhibit CBl-induced inhibition of adenylyl cyclase activity. The concentration was later increased to 10 |aM ensuing sufficient blocking of CBl receptor by the antagonist.
For CB2 receptors, L759656 was shown to be highly selective, displaying Ki values of 11.8 nM and 4888 nM for human CB2 and CBl receptors in transfected CHO cells, respectively (Huffman, 2000). The concentration chosen for L759656, 1 \iM, would selectively activate CB2 receptors. The antagonist AM630 was shown to specifically inhibit CB2 receptors, displaying Ki values of 31.2 nM and 5152 nM for
human CB2 and CBl receptors, as determined from CBl and CB2 receptor- transfected CHO cells (Ross et al., 1999). The concentration chosen initially for
AM630 was 1 ^iM that was sufficient to selectively abolish CB2-induced inhibition
of adenylyl cyclase activity. This concentration was later increased to 10 |iM to
87 ensure sufficient blocking of the CB2 receptor by the antagonist. The use of AM630 at 10 |LIM has also been shown to reverse anandamide-induced inhibition on neurotransmitter releases and Ca^^ mobilization of dissociated rat DRG cells
(Tognetto et a/., 2001).
5.1.3.5 Gi/o-coupled GPCR: 5-HTIA receptors
The serotonin 1A (5-HTIA) receptor belongs to a subtype of the 5-HT receptor family.
This receptor subtype is classified as a Class A GPCR that couples mainly to Gi/� protein which, when activated, inhibits the activity of the adenylyl cyclase and leads to the reduction of cAMP production (Foord et al., 2005; Hoyer et al, 2002). The 5-
HTIA receptors are found within both the central and peripheral tissues including
DRG neurons (Nicholson et al., 2003).
In the CNS, 5-HTIA receptors are expressed at high density in rat hippocampus. 5-
HT is a neurotransmitter released by the brainstem structures that play an important
role in sensory information processing, which is found to be 5-HTIA receptor-
mediated (Liu et al, 2002; Nicholson et al., 2003). In the PNS, a majority of 5-HTIA
receptors can be found in mast cells and platelets. The endogenous ligand, 5-HT, is
released as an inflammatory agent and it activates the 5-HTIA receptors to influence
physiological effects. One of the consequences of 5-HTIA receptor activation is
88 analgesia. 5-HTIA receptor mRNA has been shown in whole DRG and the mRNA expression of this receptor increased after during inflammatory pain condition (Liu et al, 2005). The presence of 5-HTIA receptors on DRG neurons is supported by electrophysiological study that revealed expression of 5-HTIA receptors on C-fiber of
DRG neurons and activation of 5-HTIA mediated hyperpolarization of the capsaicin-
sensitive C-fiber (Todorovic et al, 1992). Hyperpolarization of C-fiber reduces pain
signals and hence alleviates pain sensation. However, expression of 5-HTIA in DRG
glial cells has not been reported.
To characterize the 5-HTIA receptor, 8-OH-DPAT has been commonly used as a
receptor-specific agonist for 5-HTIA receptors (Newman-Tancredi et al., 1998a;
Todorovic et al., 1992). The KI value of 8-OH-DPAT is 0.58 nM for human 5-HTIA
receptors transfected in CHO cells (Newman-Tancredi et al., 1998a). The
concentration of 8-OH-DPAT used in this study was 1 |LIM, which should activate
majority 5-HTIA receptors. S-WAY100135 is a potent and selective 5-HTIA receptor
antagonist which has a KI value of 10.3 nM for human 5HTIA receptors transfected in
CHO cells (Newman-Tancredi et al., 1998b). The concentration of S-WAY 1001135
was used at 500 nM in this study.
89 5.1.3.6 Gi/o-coupled GPCR: opioid and opioid-receptor-like 1 receptors
The opioid receptors (ORs) are another family of receptors well-known for mediating analgesic effects. There are currently three ORs, namely |I-OR, 5-OR, and K-OR, and a opioid-receptor-like receptor, namely opioid-receptor-like 1 (ORLl) receptor
(Fichna et al., 2007). They all belong to Class A GPCRs and couple to Gi/�proteins.
These receptors are found in many tissues in the CNS and PNS for the mediation of analgesic effects. In the brain, opioid receptors are found to be expressed at high densities in the regions that integrate information about pain, such as the brainstem, the medial thalamus and the hypothalamus (Fichna et al., 2007). In the spinal cord, opioid receptors are located at both presynaptic terminals of the primary sensory neurons and post-synaptic nerve terminals of the dorsal horn of the spinal cord. In the
PNS, the opioid receptors are also expressed in the DRG, immune cells and
subcutaneous tissues (Stein et al, 2009). Currently, the presence of |i-ORs on glial
cells both in vivo and in vitro has not been reported. In the DRG, the |x-ORs have been found to be the most predominant opioid receptor subtype to be expressed
under normal conditions (Ji et al, 1995) and to be up-regulated during inflammation
(Ji et al., 1995; Puehler et al, 2004). Within all the DRG neurons, ^i-ORs were
i
shown to be expressed on small diameter neurons that are nociceptive neurons
containing hyperalgesic neuropeptides, such as CGRP and substance P (Levine et al.,
90 1993; Li et al., 1998). These studies revealed a more important role of [x-OR than 5-
OR and K-OR in modulating nociception. With a vast amount of studies on |i-OR in modulating nociception, we begin studying the expression of opioid receptors on
DRG cells by characterizing the |x-OR.
The signaling pathway of |x-OR activation is well-documented. The activated Gui/� protein dissociated from activated |i-OR inhibits adenylyl cyclase activities that abolish the activation of PKA signaling pathway leading to hyperpolarization of the small diameter neurons. The activated Py subunit, on the other hand, inhibit the voltage-gated Ca^"^ channels and the activation of the K+ channels that hyperpolarize the neurons to reduce the generation of pain signals (Cunha et al, 2010). As a consequence, the excitability of neurons is decreased, attenuating the generation of pain signals and the release of hyperalgesic agents such as substance P and CGRP and therefore, an anti-nociceptive effect occurs (Stein et al., 2009). In addition to the
Gai/o, the Gpy subunit has been suggested to alleviate hypemociception by activating
PI3Ky, a member of the PI3K enzyme family. This leads to the activation of a
PI3K/AKT signaling pathway that stimulates production of nitric oxide (NO) and consequently, NO promotes hyperpolarization of nociceptive neurons by indirectly activating KATP channel to reduce pain signal generation (Cunha et al., 2010). Taken together, opioid receptors have important roles in the controlling pain.
91 Potent agonists for the ji-OR includes morphine and DAMGO, but DAMGO displays higher selective for the ^i-OR over the other receptor subtypes. DAMGO has a Kj value of 2.0 nM and 1200 nM for rat ^I-ORs and K-ORS, respectively (Meng et al.,
1993; Raynor et al., 1994). For |i-OR antagonist, naloxone is preferably used displaying pA! value of 8.6 for |x-ORs (Mulder et al, 1991). For the characterization of |i-ORs, 1 \xM of DAMGO and naloxone were used to specifically activate and
inhibit |i-ORs, respectively, in the DRG cells.
The ORLl receptors share many characteristics with the opioid receptors, including
amino sequence homology, G-protein coupling and the association with pain
processing (Henderson et al., 1997). It has been shown that the mRNA expression of
ORLl receptors was mainly on small diameter and medium diameter neurons which
are the DRG neurons that are responsible for nociception (Chen et al., 2006).
However, expression of ORLl receptors on DRG glial cells is not reported.
Moreover, the mRNA expression of ORLl receptors has been shown to be increased
during inflammation and nerve injury (Chen et al., 2006). ORLl receptors are
coupled to Gj/o protein to inhibit adenylyl cyclase activity, and to the inhibition of
voltage-gated Ca^^ channels and the activation of the K+ channels that hyperpolarize
the neurons to reduce the generation of pain signals and inhibit the release of
hyperalgesic agents such as substance P and CGRP (Henderson et al., 1997; Mika et
92 al, 2003). Nociceptin is used as a potent ORLl receptor agonist with Kj value of 0.4 nM, which has no significant activity at the opioid receptor subtypes (Dooley et al.,
1997; Henderson et al, 1997). Antagonists for ORLl receptors include J11397
which has Ki values of 1.8 nM, 1000 nM, >10000 nM, and 640 nM for human ORLl
receptors ^i-OR, 5-OR and K-OR, respectively (Pol et al, 2004). For the
characterization of ORLl receptors, 1 joM of N/OFQ was chosen to activate ORLl
receptors, while 300 nM J11397 was chosen to inhibit the receptor activation by the
agonist.
5.2 Aims of study
We have previously shown that prostanoid receptors (EP4 and IP receptors) were
expressed on both DRG neurons and glial cells in dissociated cells cultures. These
prostanoid-stimulated responses in glial cells were inhibited by the presence of
neurons. The expression of pain-related receptors on glial cells suggests that glial
cells possess the ability to respond to hyperalgesic agents. The characterization of
pain-related receptors on glial cells may reveal a novel role of glial cells in the
development and maintenance of hyperalgesia and the interaction with neurons. As a
preliminary investigation, this study aimed to investigate if the pronociceptive-
related Gs-coupled and anti-nociceptive Gi/�-couple dGPCRs were expressed on
DRG neurons and glial cells in cultures. By using pharmacological tools, we
93 measured cell responses, i.e. the production of [^H]cAMP in mixed DRG cells, neuron-enriched cells and glial cells upon agonist and antagonist incubation. Besides, we also investigated the effect of neurons in influencing the agonist-stimulated responses in glial cells.
5.3 Results
5.3.1 Characterization of prostanoid receptors in isolated DRG cells
The adenylyl cyclase activity was determined in mixed DRG, neuron-enriched, and glial cells by measuring the percent conversion of [^H]ATP to [^H]cAMP following agonist and/or antagonist treatments. Cicaprost and PGE2 were the agonists used to stimulate the prostanoid receptors in the three DRG cell cultures. We have previously demonstrated that cicaprost and PGE2 stimulated the production of [^H]cAMP in the three DRG cell cultures. Besides, the cicaprost- and PGE2-stimulated responses could only be antagonized by the IP receptor-specific antagonist, ROl 138452, and
EP4 receptor-specific antagonist, ONO-AE3-208, respectively. The results indicated that cicaprost mediated its response via IP receptors and PGE2 mediated its response via EP4 receptors. The pECso values for cicaprost and PGE2 were 7.38 土 0.06 and
I
6.86 土 0.19，respectively, in mixed DRG cells (Ng et al, 2011). As a result, the concentration used in this study to stimulate [^H]cAMP production was chosen at 40
94 nM for cicaprost, which was the EC50 value determined in DRG cell cultures, whereas the concentration for PGE2 was chosen at 1 jxM, which was the
concentration that would produce nearly the maximum responses so that the
responses were comparable to the cicaprost-stimulated responses. Forskolin (1 |iM),
a receptor-independent compound, was used to stimulate adenylyl cyclase activity
directly to increase the production of [^H]cAMP without acting through a receptor. In
this current study, cicaprost and forskolin stimulated adenylyl cyclase activity
leading to the increase [^HjcAMP production in all three DRG cell cultures (Fig. 5.1).
Cicaprost increased the fH]cAMP in mixed DRG cells {p < 0.001) and glial cells {p
< 0.001) while the increase in ^HJcAMP was not shown to be significant in neuron-
enriched cell cultures (p > 0.05). Relative to the basal activities, the increase in
response to cicaprost was 4.6-fold in mixed DRG cells, 1.9-fold in neuron-enriched
cells, and 16.4-fold in glial cells. Forskolin significantly increased the [^H]cAMP in
mixed DRG cells {p < 0.001), neuron-enriched cells {p < 0.001), and glial cells (p <
0.01). Relative to the basal activities, the increase in response to forskolin was 9.5-
fold in mixed DRG cells, 5.4-fold in neuron-enriched cells, and 10.4-fold in glial
cells. Regardless of the cicaprost and forskolin treatments, the stimulated responses
t
of neuron-enriched cells were smaller than that of the mixed DRG cells (p < 0.001).
Comparing the cicaprost-simulated responses between mixed DRG cells and glial
95 cells, the responses produced in the mixed DRG cells were significantly smaller than that in the glial cells (p < 0.001). However, this inhibitory pattern was not seen in forskolin-stimulated responses. The mixed DRG cells produced larger forskolin- stimulated responses when compared with the responses in glial cells (p < 0.001).
5.3.2 Characterization of CGRP receptors in isolated DRG cells
The CGRP receptors in the three DRG cell cultures were characterized by stimulation with the endogenous CGRP receptor agonist, CGRP (1 ^iM), and by the blockade of the receptors with CGRP antagonist, CGRPg-s? (10 ^iM) (Fig. 5.2).
CGRP significantly increased the production of [^H]cAMP in mixed DRG cells (p <
0.01) and glial cells (p < 0.001), while the increase in fH]cAMP was not shown to be significant in neuron-enriched cell cultures {p > 0.05). Relative to the basal activities, the increase in ^HJcAMP stimulated by CGRP was 5.1 士 1.0-fold in mixed DRG cells, 2.1 士 0.5-fold in neuron-enriched cells and 20.1 士 2.8-fold in glial cells. Pretreatment with CGRPg-s? alone had no effect on [^H]cAMP production {p >
0.05) but inhibited CGRP-stimulated fH]cAMP production by 60 土 6o/o (p < 0.001) in mixed DRG cells, 40 土 15o/o in neuron-enriched cells (p < 0.01) and 39 士 3o/o (p <
0.001) in glial cells. CGRP produced a similar response pattern compared with the responses produced by cicaprost in the three DRG cell cultures (Fig 5.3). Both
CGRP-stimulated and cicaprost-stimulated responses in the mixed DRG cells were
96 significantly smaller than those in the glial cells {p < 0.001 and p < 0.01, respectively), but were significantly larger than those in the neuron-enriched cell (p <
0.05 and p < 0.01, respectively).
5.3.3 Investigation of the effect of CGRPs-a? on CGRP responses
In Fig. 5.2, the CGRP-stimulated responses could only be partially inhibited by the antagonist, CGRPg-s? (10 |iM), suggesting that the concentration of agonist or antagonist might not be optimal. Therefore, log agonist concentration-response curves for CGRP in the presence of the CGRPg-s? were generated using mixed DRG cells (Fig. 5.4). The pECso values for CGRP alone and in the presence of CGRPg-s? were 7.40 士 0.25 and of 6.56 士 0.19，respectively (p < 0.05). In the presence of the
CGRP8-37, the dose-response curve shifted to the right with a lower pECso value.
CGRP8-37 significantly decreased both the basal activity and CGRP-stimulated responses at all the concentrations that were tested. The inhibition of CGRP- stimulated responses by the CGRPg-av was clearly insurmountable.
5.3.4 Characterization of pi-adrenoceptors in isolated DRG cells
The Pi-ARs in the three DRG cell cultures were characterized by stimulation with a
Pi/PZ-AR agonist, isoprenaline (1 |IM), and by the blockade with Pi-ARs specific antagonist, S-(-)-atenolol (1 |LIM) (Fig. 5.5). Isoprenaline significantly increased the
97 production of [^H]cAMP in mixed DRG cells, neuron-enriched cells and glial cells {p
<0.001). Relative to the basal activities, the increase in [^H]cAMP was 3.5 土 0.5-fold in mixed DRG cells; 2.9 土 0.2-fold in neuron-enriched cells, and 3.3 土 0.2-fold in glial cells. Pretreatment with S-(-)-atenolol alone had no effect on [^H]cAMP production {p > 0.05) and showed no inhibition on the isoprenaline-stimulated
[^H]cAMP production in all three DRG cell cultures (p > 0.05). Comparing the isoprenaline-simulated responses between mixed DRG and glial cell cultures, the responses in the mixed DRG cells were significantly larger than the responses in the neuron-enriched cells (p < 0.001) and glial cells (p < 0.01).
5.3.5 Characterization of P2-adrenoceptors in isolated DRG cells
The p2-ARs in the three DRG cell cultures were characterized by stimulation with a
P2-AR specific agonist, formoterol (1 |iM), and by the blockade with P2-AR specific antagonist, ICI118551 (1 |LIM) (Fig. 5.6). Formoterol significantly increased the production of [^H]cAMP in mixed DRG cells, neuron-enriched cells and glial cells (p
< 0.001). Relative to the basal activities, the increase in [^H]cAMP was 3.4 土 0.5- fold in mixed DRG cells, 2.7 士 0.3-fold in neuron-enriched cells and 3.4 士 0.2-fold in glial cells. Pretreatment with ICI118551 alone had no effect on [^H]cAMP production (p > 0.05) but inhibited formoterol-stimulated [^H]cAMP production by
77 士 2o/o (p < 0.001) in mixed DRG cells, 78 士 5o/o in neuron-enriched cells (p <
98 0.001) and 76 士 9o/o in glial cells {p < 0.001). Formoterol and isoprenaline produced a different response pattern compared with the responses produced by cicaprost in the three DRG cell cultures (Fig. 5.7) The formoterol and isoprenaline-simulated responses in the mixed DRG cells were similar to those in glial cells. In contrast, for cicaprost stimulations, mixed DRG cell responses were smaller than glial cells responses {p < 0.01) but larger than neuron-enriched cells (p < 0.05).
5.3.6 Identification of p-adrenoceptor subtype mediating isoprenaline- stimulated responses
As shown in Fig. 5.5, there was a lack of inhibition on the isoprenaline-stimulated responses by the pi-AR specific antagonist, S-(-)-atenolol. Experiments were then performed to investigate if isoprenaline-stimulated responses were mediated through p2-ARs. To study this, the three DRG cells cultures were pretreated with Pi-AR specific antagonist, S-(-)-atenolol (1 |LIM), or P2-AR specific antagonist, ICI118551
(100 nM) for 15 min before addition of pi/ P2-AR agonist, isoprenaline (1 |_iM) (Fig.
5.8). Consistent with previous result, isoprenaline stimulated the increase in
[^H]cAMP production in mixed DRG cells {p > 0.05)，neuron-enriched cells (p >
0.01) and glial cells (p > 0.001). S-(-)-atenolol could not inhibit the isoprenaline- stimulated [^H]cAMP production in any of the three cell cultures {p > 0.05), whereas
ICIl 18551 inhibited isoprenaline-stimulated [^H]cAMP production by 100 士 10% in
99 mixed DRG cells (p < 0.001), 58 士 24o/o in neuron-enriched cells (p < 0.01) and 93 士 l%in glial cells (p< 0.001).
5.3.7 Characterization of a2-adrenceptors in isolated DRG cells
The presence of Gj/o-coupled GPCRs is demonstrated when the presence of receptor- specific agonist inhibits the forskolin-stimulated fHJcAMP production, while the presence of receptor-specific antagonist restores the agonist-inhibited forskolin responses. To characterize 012-AR, UK 14304 (1 [oM), an ai-AR specific agonist, was used to inhibit the forskolin-stimulated responses in the three DRG cell cultures (Fig.
5.9), while yohimbine (1 |aM), an az-AR specific antagonist, was added 15 min before the addition of forskolin and UK14304. Treatment with UK143 04 had no effect on [^H]cAMP when treated alone and failed to inhibit on the forskolin- stimulated [^H]cAMP production in all three DRG cell cultures. Pretreatment with yohimbine showed no effect on [^HJcAMP when pretreated alone in all three DRG cell cultures. Results indicated the absence of Gi/o/AC-coupled a2-AR in DRG cell cultures.
5.3.8 Characterization of cannabinoid 1 receptors in isolated DRG cells (
The CBl receptor agonist, R-(+)-methanandamide (100 nM), was used to inhibit the forskolin-stimulated responses in the three DRG cell cultures (Fig. 5.10), while
100 AM251 (10 laM), was added 15 min before the addition of forskolin and AM251.
Treatment with R-(+)-methanandamide had no effect on [^H]cAMP when treated
alone and failed to the forskolin-stimulated [^H]cAMP production in all three DRG
cell cultures. Pretreatment with AM251 showed no effect on [^H]cAMP when
pretreated alone in all three DRG cell cultures. Results indicated the absence of
Gi/o/AC-coupled CBl receptor in DRG cell cultures.
5.3.9 Characterization of cannabinoid 2 receptors in isolated DRG cells
The CB2 receptor agonist, L759656 (1 |JM), was used to inhibit the forskolin-
stimulated responses in the three DRG cell cultures (Fig. 5.11), while AM630 (10
|iM), a CB2 antagonist, was added 15 min before the addition of forskolin in and
L759656. Treatment with L759656 had no effect on [^H]cAMP when treated alone
and failed to inhibit the forskolin-stimulated [^H]cAMP production in all three DRG
cell cultures. Pretreatment with AM640 showed no effect on [^H]cAMP when
pretreated alone in all three DRG cell cultures. Results indicated the absence of
Gi/o/AC-coupled CB2 receptors in DRG cell cultures.
5.3.10 Characterization of 5-HTIA receptors in isolated DRG cells
The 5-HTIA receptor agonist, 8-OH-DPAT (1 pM), was used to inhibit the forskolin-
stimulated responses in the three DRG cell cultures (Fig. 5.12), while S-WAY100135
101 (500 |iM), was added 15 min before the addition of forskolin and 8-OH-DPAT.
Treatment with 8-OH-DPAT had no effect on [^H]cAMP when treated alone and failed to inhibit the forskolin-stimulated fH]cAMP production in all three DRG cell
cultures. Pretreatment with S-WAY100135 showed no effect on [^H]cAMP when
pretreated alone in all three DRG cell cultures. Results indicated the absence of
Gi/o/AC-coupled 5-HTIA receptors in DRG cell cultures.
5.3.11 Characterization of 抖-opioid receptors in isolated DRG cells
The \x-OR receptor agonist, DAMGO (1 |JM), was used to inhibit the forskolin-
stimulated responses in the three DRG cell cultures (Fig. 5.13), while naloxone (1
jiM), a |i-OR receptor antagonist, was added 15 min before the addition of forskolin
and DAMGO. Treatment with DAMGO had no effect on [^H]cAMP when treated
alone and failed to inhibit the forskolin-stimulated fH]cAMP production in all three
DRG cell cultures. Pretreatment with naloxone showed no effect on [^H]cAMP when
pretreated alone in all three DRG cell cultures. Results indicated the absence of
Gi/o/AC-coupled |a-OR receptors in DRG cell cultures.
5.3.12 Characterization of opioid-receptor-like 1 receptors in isolated DRG
cells
The ORLl receptor agonist, N/OFQ (1 jjM), was used to inhibit the forskolin-
102 stimulated responses in the three DRG cell cultures (Fig. 5.14), while J11397 (300 nM), an ORLl receptor antagonist, was added 15 min before the addition of
forskolin and N/OFQ. Treatment with N/OFQ had no effect on [^H]cAMP when
treated alone and failed to inhibit the forskolin-stimulated [^H]cAMP production in
all three DRG cell cultures. Pretreatment with J11397 showed no effect on
[^H]cAMP when pretreated alone all three DRG cell cultures. Results indicated the
absence of Gj/o/AC-coupled ORLl receptors in DRG cell cultures..
5.3.13 Effect of nerve growth factor on DRG cells
For all the agonists that coupled to Gs proteins being tested in this study, neuron-
enriched cells produced responses that were significantly smaller than those of the
mixed DRG cells which was unexpected. This small response could be due to the
decrease of receptor expression in the first two days in cultures and the lack of
neurotropic factors (Franklin et al., 2009)，which could be provided by glial cells
(Zhou et al., 1999). Neurotropic factors such as NGF are essential for survival of
DRG neurons during embryonic development but not in adult DRG neurons (Lindsay,
1988). However, it has been suggested that addition of NGF could facilitate the re-
expression of receptors of adult nociceptive DRG neurons in isolated cultures
(Franklin et al., 2009). We therefore hypothesized that addition of NGF would help
enhance the neuron-enriched cell responses to Gs agonists. We have previously
103 demonstrated that NGF (50 ng/ml) had no effect on the Gs agonist-stimulated responses in mixed DRG cells (Wise, 2006) but the same has not been tested on neuron-enriched cells and glial cells. As a result, the effect of NGF on Gs-stimulated responses was tested in mixed DRG cells, neuron-enriched cells and glial cells (Fig
5.15A-C). Mixed DRG cells were incubated with NGF (50 ng/ml) for 48 h, while
neuron-enriched cells and glial cells were incubated with NGF for 24 h before
assaying the adenylyl cyclase activity. On the assay day, cells were incubated with
control solution (basal), receptor-independent adenylyl cyclase activator, forskolin (1
|LIM), and GS agonists including IP receptor agonist (cicaprost, 40 nM), EP4 agonists
(ONO-AEl-329, 1 ^M; PGE2, 1 |iM) an d P2-AR agonist (formoterol, 1 \M).
Production of [^H]cAMP in response to Gs agonist-stimulations in control and NGF-
treated groups were determined in the three cell cultures. Results indicated that
incubation with NGF had no effect on the Gs agonist-stimulated responses in mixed
DRG cells, neuron-enriched cells and glial cells (p > 0.05, two-way ANOVA). This
indicated that NGF failed to produce larger Gs agonist-stimulated responses in
neuron-enriched cell preparations.
Other than testing on Gs-agonist stimulated responses, the effect of NGF were also I
tested on Gi/o-agonist-stimulated responses in isolated DRG cells. This was because
in our preliminary experiments that was done some time ago, more than 20%
104 inhibition of the forskolin-stimulated adenylyl cyclase activity by A2-AR, 5HTIA and ji-OR receptor agonists (Gi/o agonists) had been observed in mixed DRG cells cultured after 5 days in the presence of NGF (unpublished data). However, this current study failed to show inhibition of the forskolin-stimulated responses by all
Gi/o agonists after 2 days in culture in the absence of NGF. Therefore to mimic our preliminary studies, Gi/o agonist-stimulated responses on mixed DRG cells in the presence of NGF were assayed after 5 days (Fig. 5.15D-E). This preliminary result (n
=1) showed that in the absence of NGF (50 ng/ml)，all Gi/o agonists tested could not inhibit the forskolin-stimulated responses after 5 days in mixed DRG cells. In the presence of NGF, the same was seen in majority of the Gj/o agonists except for 5-
HTia and a2-AR agonists. 8-OH-DPAT (1 |iM), inhibited the forskolin-stimulated responses by 76% while UK14304 (1 \iM) inhibited the basal activity by 42%.
Besides, NGF has been shown to facilitated neurite outgrowth of adult DRG neurons
(Lindsay, 1988). Therefore, mixed DRG cells, neuron-enriched cells and glial cells in the absence or present of NGF were fixed with acetone methanol (1:1) solution after
2 days in cultures and stained with anti-TUJ-1 and anti-GFAP antibodies to identify
DRG neurons and glial cells (Fig. 5.16). Preliminary results (n= 1) indicated that
I
incubation with NGF resulted in neurons with more extensive neurite outgrowth in
neuron-containing cultures (Fig. 5.16 A-D). This observation was consistent with
105 previous findings (Lindsay, 1988). On the other hand, NGF has no effect on the expression of GFAP in glial cells (Fig. 5.16 E-F).
5.4 Discussion
Pharmacological evidence of functional expression of GPCRs is demonstrated by the receptor-specific agonist-stimulated responses being abolished by the receptor- specific antagonist. In the first part of this study, the characterization of pain-related
Gs-GPCRs on DRG cells reconfirmed the presence of EP4 and IP- receptors on both
DRG neurons and glial cells. Besides, this study also revealed the functional expression of Pz-ARs and CGRP receptors, but not Pi-ARs, on DRG neurons and glial cells. Isoprenaline- and formoterol-stimulated responses in glial cells were smaller than those in mixed DRG cells. In contrast, consistent with cicaprost- and
PGE2-stimulated responses, the CGRP-stimulated responses in glial cells were larger than those in the mixed DRG cells.
The expression of p-ARs was reported in DRG neurons (Pluteanu et al., 2002) and was suspected in SGCs (Hanani, 2010) to mediate isoprenaline-stimulated responses.
In this study, we clearly demonstrated that there is an increase of [^H]cAMP in neuron-enriched cells and glial cells upon incubation with isoprenaline, and the isoprenaline-stimulated response was blocked by the P2-AR antagonist, ICIl 18551.
We provided the first evidence for isoprenaline-stimulated [^H]cAMP production in
106 DRG neurons and glial cells via activation of P2-ARS. It is suggested that after nerve injury, the DRG neurons become sensitive to catecholamines which are released by postganglionic neurons of the sympathetic division (Khanolkar et al., 1996), which leads to hyperalgesia (Ferreira, 1980; Khasar et al, 1999b). The expression of P2-
ARs in neurons and glial cells suggested that DRG cell cultures mimic neuropathic pain models.
The isoprenaline- and formoterol-stimulated responses in mixed DRG cells were larger than those in the glial cells. This response pattern was different from the cicaprost- and PGEz-stimulated responses in mixed DRG cells and glial cells. It may suggest that the p2-AR responses of glial cells responses were not inhibited by the presence of DRG neurons. However, due to the larger responses of neuron-enriched cells, any inhibitory effect of neurons might be masked. Our preliminary data may provide evidence that neurons inhibit all Gs- and forskolin-stimulated responses of glial cells.
CGRP stimulated the production of [^H]cAMP in both DRG neurons and glial cells.
Although the presence of CGRP receptors has been shown in neurons in vivo and in
vitro (Ceruti et al, 2011; Segond von Banchet et aL, 2002; Ye et al, 1999) and in
glial cells in vitro (Ceruti et al., 2011), this is the first study to our knowledge that
measured the CGRP-stimulated production of cAMP in glial cells. Although the
107 release of CGRP in the nerve ending is often linked to hyperalgesia, CGRP is also released by neuronal cell body to affect other neurons and their associated glial cells
(Segond von Banchet et al.’ 2002). The expression of CGRP receptors on DRG neurons suggests the possibility that CGRP has a role in communication among neurons (Segond von Banchet et al, 2002) for the release of more CGRP (Thalakoti et al., 2007). The expression of CGRP receptors on glial cells revealed the possible involvement in pain processing in response to the release of CGRP by neurons.
In our DRG cell cultures, neurons and glial cells were seeded at a density such that neuronal cell numbers in neuron-enriched and glial cell numbers in pure glial cell cuhures were similar to those in the mixed DRG cells on the day of assay. This allows for direct comparison among the three cell cultures. Comparing the CGRP- stimulated responses in mixed DRG cells with those of glial cells, the responses in mixed DRG cell were smaller than in the pure glial cells. This inhibitory pattern was consistent with cicaprost- and PGE2-stimulated responses. It suggests that the CGRP-
stimulated responses of glial cells in the mixed DRG cell cultures were also inhibited
by the presence of DRG neurons. However, to prove definitely whether DRG
neurons inhibit the CGRP responses in glial cells will require individual labeling of
<
the glial cells, stimulating with CGRP and measuring the effect of CGRP-stimulated
glial cell responses in the presence of DRG neurons.
108 Although the presence of CGRP receptors on neurons and glial cells was shown, the
CGRP (1 |jM)-stimulated responses in the mixed DRG cells could not completely be antagonized by CGRPg-s? at 10 \iM. The log agonist concentration-response curves indicated that the responses at EC50 concentration of CGRP (40 nM) was completely inhibited by 10 |iM CGRP8-37. Nevertheless, the effect of CGRPg-s? on CGRP- stimulated response was insurmountable, meaning that the effect of CGRPg.s? could not be overcome by increasing the concentration of CGRP. It is not likely that
CGRP8-37 would act as a non-competitive antagonist, because CGRP8-37 reduced the binding of CGRP by competition with the same binding site as CGRP in a concentration-dependent manner (Segond von Banchet et al, 2002). It is controversial whether CGRPg-s? could completely antagonize the effect of CGRP.
There are studies that showed CGRPg-s? acts as a competitive antagonist for CGRP receptors (Chiba et al., 1989; Doods et al., 2000), while some showed the failure of the blocking action by CGRPg-s? (Segond von Banchet et al, 2002; Xu et al.’ 1996).
CGRP8-37 is currently the only antagonist for CGRP receptors that is commercially
available; better CGRP antagonists are yet to be released commercially.
The current study showed that CGRP-stimulated responses in glial cells, as well as <
EP4- and IP-stimulated responses as previously identified (Ng et al.’ 2011), were
inhibited by co-culture with neurons. This led to the investigation of the mode of
109 inhibition by neurons on glial cells. The possible modes of inhibition of glial cells by neurons can mediate directly through cell-cell contact or indirectly by the release of soluble factor(s). To test the modes of inhibition by neurons, we used transwell culturing system to assess the roles of cell-cell contact and the release of soluble factors on glial cell responses to cicaprost. In this study, neuron-enriched cells were cultured in the inserts and glial cells were cultured in the wells of transwell plates.
When [^H]cAMP production was determined separately in neuron-enriched and glial cell cultures, the glial cells in such group (with access of soluble factors being released by neurons in the inserts) demonstrated smaller responses to cicaprost when compared with glial cell alone group (with no neurons in the inserts). This suggested that soluble factors released by neurons were involved in inhibiting glial cell responses. Besides, when neuron-enriched cells and glial cells were being cultured in the wells, responses to cicaprost were the smallest among all other groups. This suggested that cell-cell contact and soluble factors were important for inhibiting glial cells responses (manuscript in preparation).
To account for the larger cicaprost-, PGE2-, and CGRP-stimulated [^H]cAMP productions in glial cells relative to neuron-enriched cells in dissociated cultures, it is
worth investigating the expression of adenylyl cyclase isoforms in DRG cells. There
is currently no report on the expression profile of adenylyl cyclase isoforms in DRG
110 cells. The use of adenylyl cyclase isoform-specific antibodies could be a solution to identify the isoforms, but it is currently unavailable. In pure glial cell cultures, RT_
PGR can be used to detect the mRNA expression of different adenylyl cyclase isoforms. However, this method cannot be used to detect the presence of adenylyl cyclase isoforms in neuron-enriched cells because this culture contains contaminating glial cells which interfere the results interpretation.
The second part of the study was to characterize the presence of Gi/�-couple clasd s of
GPCRs on neurons and glial cells. The presence of the Gi/�-couple GPCd R can be demonstrated when the Gi agonist inhibited the forskolin-stimulated responses and the Gi/o antagonist could restore the agonist-induced inhibition of the forskolin- stimulated responses. The results failed to demonstrate the inhibition of forskolin- stimulated responses by Gi/o agonists. Our previous unpublished data had shown significant inhibition of forskolin-stimulated responses in the mixed DRG cells after five days in culture in the presence of NGF (50 ng/ml). With reports showing that
NGF facilitated the mRNA expression of (i-OR of DRG neurons after three days in
culture (Franklin et al., 2009)，one would suspect that NGF would facilitate the
expression of Gi/�-GPCRs Th. e presence of NGF showed no effect on the increase of
I
cicaprost-stimulated responses in mixed DRG cell cultures over five days (Wise,
2006). However, the effect of NGF cannot be studied on neuron-enriched cells after
111 five days as glial cells in neuron-enriched cell cultures proliferate over time and decrease the purity of the neuron-enriched cell preparation. Studying this
"contaminated" culture therefore becomes inadequate after three days. However, as demonstrated the lack of effect by NGF over five days on Gs-stimulated responses in the mixed DRG cells, it is also suspected that NGF will not produce significant effect on Gi/o-stimulated responses over time in cultures.
Despite strong evidence showing the presence of pain-related Gj-coupled GPCRs on
DRG cells, our system of measuring the inhibition of forskolin-stimulated [^H]cAMP production by Gi/o agonists produced inconclusive results. One potential problem for the measurement of inhibition of the forskolin-stimulated responses by a Gi/o agonist is that forskolin stimulated adenylyl cyclase activity for the production of [^H]cAMP in both neurons and glial cells in mixed DRG cell cultures, while the Gi/o-coupled
GPCRs that mediate the inhibition of forskolin stimulation might not be expressed on
neurons and/or glial cells. Therefore, it reduced the sensitivity of the measurement in
mixed DRG cells and neuron-enriched cells. However, even in glial cell cultures that
had over >99% glial cell purity, the inhibition of the forskolin-stimulated responses
by a Gi/o agonist was still not prominent. Other methods are therefore needed to
characterize the expression of Gi/�receptor osn isolated DRG cells.
112 There are currently some other methods that are available and may be useful for the characterization of Gi/o-coupled GPCRs in cultures. Firstly, many Gi/o-GPCRs, such as opioid and CB receptors, are antigenic markers for the binding of specific antibodies (Ji et al.’ 1995; Ross et al, 2001). Detecting the presence of Gi/�-GPCR antibody immunoreactive cells will allow the characterization of Gi/�-GPC Ron DRG neurons and glial cells. However, with the drawbacks as discussed in Chapter 4 for the use of antibodies in cultures, unless these antibodies produced very specific staining patterns with appropriate negative and positive controls of cells that are immunoreactive for the antibody, the evidence for the presence of GI/o-GPCR in
DRG cell cultures is not persuasive with the use of immunocytochemical methods.
The use of RT-PCR methods, on the other hand, to detect the presence of Gi/o-GPCR
mRNA is relatively useful and the |x-OR receptor mRNA has been detected in mixed
DRG cells (Franklin et al, 2009). The mRNA expression of Gi/o-GPCR can be
characterized in glial cells because this cell preparation has over 99% purity of glial
cells. The presence of a PGR product, the mRNA of the receptor, would provide
evidence for the mRNA expression in glial cells. However, when the same procedure
is performed in neuron-enriched cells, the use of PGR method has limitations
because with around 70% purity of neurons, the cell type/s expressing the mRNA is
unknown.
113 And, last but not least, measurement on the change of membrane potential upon activation of a Gi/�GPC Rwould be useful to characterize the presence of Guo-
GPCRs on neurons and glial cells. This method employs the use of the fluorescent bisoxonol membrane potential dye DiBAC. The fluorescence of the dye will be enhanced when the dye enters the cell membrane as a result of membrane depolarization. Measurement can be made on the change of fluorescence upon activation by the agonist of Gi/o-coupled GPCR. This method has been used to characterize the presence of opioid receptors on DRG neurons (Cunha et al, 2010) and should be used in the future to reveal the presence of other pain-related Gj/�- receptors on DRG cells.
5.5 Summary
In this study, the presence of Gs-coupled GPCRs (p2-ARs and CGRP receptors, but not Pi-ARs)，were identified in neurons and glial cells. Consistent with the EP4 and
IP agonist-stimulated responses described previously (Ng et al.’ 2011), the CGRP- stimulated responses in mixed DRG cells was smaller than those in the glial cells.
This inhibitory pattern was not seen with P2-AR agonist stimulation. With all Gs-
GPCR agonist-stimulations, neuron-enriched cell preparations produced smaller
responses than mixed DRG cells and glial cells. The addition of NGF did not lead to
any increase in responses of neuron-enriched cells, but it facilitated neurite
114 outgrowth of DRG neurons. For the characterization of Gi/o-coupled GPCRs, agonist treatments failed to inhibit forskolin-stimulated responses in mixed DRG, neuron- enriched cells and glial cells, which prevented the characterization of Gi/�-coupled
GPCR expression. With strong evidence that the Gi/o-coupled GPCR being studied were expressed on DRG cells, other methods need to be identified in order to
characterize the presence of these receptors in DRG neurons and glial cells.
115 Table 5.1 Summary of the drugs used to characterize GPCRs in isolated cultures of DRG cells
GPCR Primary G-protein Agonist used Antagonist used coupling a2-AR Gi/o UK14304 (1 ^M) yohimbine (1 \M)
Pi-AR GS isoprenaline (1 yM) S-(-)-atenolol (1 |JM)
P2-AR Gs formoterol (1 ^M) ICI 118,551 (100 nM)
CGRP Gs CGRP (1 mM) CGRP8-37 (10 _)
CBl Gi/o R-(+)-methanandamide (100 nM) AM251 (10 pM)
CB2 GI/O L759656 (1 \M) AM630 (10 PM) li-OR Gi/o DAMGO (1 \M) naloxone (1 ^M)
ORLl Gi/o N/OFQ (1 ^M) 111397(300 nM)
5-HTIA GI/O 8-OH-DPAT (1 _ S-WAY100135 (500 nM)
(
116 • basal • cicaprost (40 nM) 4-1 • forskolin (1 ^M) I mmi
=• ### l圍i• A_ • - • _ •
Mixed Neuron Glia
(
117 • basal ### • CGRP (1 [Mi *** c • CGRP8.37 (10 nM) O • CGRP + CGRP8.37 i 3- • 3 O 園 •O _ AAA o ！2 圍 T
2. a g ** IS i• T I • 1' ^B AAA
- I I ## . I I QI 门 • ri _•_ 门 •_ Mixed Neuron Glia Figure 5.2 CGRP receptor-mediated adenylyl cyclase activity in isolated DRG
cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-
enriched cells, and glial cells. Cells were incubated with control solution (basal) or
with CGRP8.37 (10 \M) for 15 min before addition of CGRP (1 \M). Data are
represented as means 士 SEM, from three independent experiments. *** p < 0.01, ** p < 0.001 compared with basal activity in the same cell group using one-way
ANOVA. Np < 0.05, AAAp < 0.001 compared with the CGRP-stimulated responses
in the same cell group using one-way ANOVA. <0.01’ # �p < 0.001 compared
with the CGRP-stimulated responses in the mixed DRG cells using two-way
ANOVA with Bonferroni's posthoc test.
<
118 4i ### • basal T o • cicaprost (40 nM) � ^L •云一3. • CGRP (1 nM) T • 3 O 去•
2. II
|§ i 工 II
p 1- li 11 L • • ## # 圍•
QI 门 _ • r~i _ 门 _ •_ Mixed Neuron Glia Figure 5.3 CGRP-stimulated response patterns were similar to the cicaprost- stimulated response patterns. Adenylyl cyclase activity was determined in mixed
DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution (basal), IP receptor agonist (cicaprost, 40 nM) or CGRP receptor agonist
(CGRP, 1 loM). Data are represented as means 士 SEM，from three independent experiments. # p <0.05, ## p < 0.01, ### p < 0.001 compared with the same responses in the mixed DRG cells using two-way ANOVA with Bonferroni's posthoc test.
119 • CGRP I .S-i • +CGRP8.37 (10 ^iM)
！一 5 O [
0.0- I I I I I I I -9 -8 -7 -6 -5
/ log [CGRP] (M)
Figure 5.4 Log agonist concentration-response curve for CGRP in the presence of CGRP8-37 in mixed DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells. Cells were incubated with control solution (basal) and CGRPg-s?
(10 pM) for 15 mins before addition of increasing concentrations of the CGRP. Data are presented as means 士 SEM，from 3 independent experiments. * < 0.05，** <
0.01 and < 0.001 compared with agonist alone at the same concentration using two-way ANOVA. Error bars smaller than the symbol size were not shown.
120 • basal • isoprenaline (1 \M) ^ 1,5-1 • S-(-)-atenolol (1 ^M) o ... • iso + aten ^ *** T T offil.O- I • ## 5r CD = *** T n c I • ### S • g o • • *** 圓 • 差10,1 nil! nlii nlil Mixed Neuron Glia
Figure 5.5 Comparison of pi-AR-mediated adenylyl cyclase activity in isolated
DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron- enriched cells and glial cells. Cells were incubated with control solution (basal) or S-
(-)-atenolol (1 |JM) for 15 min before addition of PI/P2-AR agonist, isoprenaline (1
|JM). Data are represented as means 士 SEM，from four independent experiments. ** p < 0.01 and *** p < 0.001 compared with basal activity in the same cell group using one-way ANOVA. ##p< 0.01, ### p < 0.001 compared with the isoprenaline- stimulated responses in the mixed DRG cells using two-way ANOVA with
Bonferroni's posthoc test.
I
121 • basal 1 5i 口 formoterol (1 fjJVI) • ICI118551 (100 nM) 0 *** • form + ICI c T 1 I 10. _ ## Co = *** Q. > = A Q. § � AAA � _ S O n e 兰 T *** _ 八八A ^ oJlll niml n_.il Mixed Neuron Glia Figure 5.6 Comparison of p2-AR-mediated adenylyl cyclase activity in isolated
DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron- enriched cells, and glial cells. Cells were incubated with control solution (basal) or
ICIl 18551 (100 nM) for 15 min before addition of P2-AR specific agonist, formoterol (1 jiM). Data are represented as means 土 SEM，from four independent experiments. < 0.001 compared with basal activity in the same cell group using one-way ANOVA.� p~ < 0.001 compared with the formoterol-stimulated responses in the same cell group using one-way ANOVA. <0.01’ ### p < 0.001 compared with the isoprenaline-stimulated responses in mixed DRG cells using two-way
ANOVA with Bonferroni's posthoc test.
I
122 = • basal � 芸一 • cicaprost (40 nM) ^ 经 § • formoterol (1 ^M) "ag ?2 - 口 isoprenaline (1 [xM) I• II 1- iii I & _ • # III -J n_JII_ nRll n I .ill I Mixed Neuron Glia Figure 5.7 Comparison of cicaprost-stimulated and p-AR agonist-stimulated adenylyl cyclase activities in isolated DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution (basal), IP receptor agonist (cicaprost, 40 nM), non- selective P1/P2-AR agonist (isoprenaline, 1 jiM) or P2-AR agonist (formoterol, 1 |JM).
Data are represented as means 士 SEM，from four independent experiments. # p
<0.05，##p < 0.01 compared with the same responses in mixed DRG cells using two- way ANOVA with Bonferroni's posthoc test.
123 门 basal H 5 • isoprenaline (1 [M) § • iso + S-(-)-atenolol (1 iM) •云它 * • iso + ICI118551 (100 nM) 3 0 T T •§ 2 1.0- ^Ijli Q, g； B *** rm n c ^ ## A =o 三 ** 目 I O 0.5- _ S fii _ "S^ m AAA • AA _ 1*1 = 工 圍 A 圍 -0.01 nililll nljil n_J_ Mixed Neuron Glia Figure 5.8 Characterization of the p-AR that mediated the isoprenaline- stimulated responses. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution
(basal), S-(-)-atenolol (1 ^M) or ICI118551 (100 nM) for 15 min before the addition of non-specific P-AR agonist, isoprenaline (1 |JM). Data are represented as means 士
SEM, from four independent experiments. * p< 0.05，** p < 0.01 and *** p < 0.001 compared with basal activity in the same cell group using one-way ANOVA. ^ p <
0.01 and p < 0.001 compared with the isoprenaline-stimulated responses in the same cell group using one-way ANOVA. ## p < 0.01 compared with the isoprenaline-stimulated responses in the mixed DRG cells using two-way ANOVA with Bonferroni's posthoc test.
124 • basal 5"! • forskolin (1 iM) c • UK14304(1 ^iM) 芸一 4•丁 J • UK14304 + fors 另百 • yohimbine (1 [M) •o i X • UK14304 + yohimbine + fors p 3- B •
P： IIj Hi
Mixed Neuron Glia
Figure 5.9 Characterization of the az-AR-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution
(basal) or a2-AR specific antagonist, yohimbine (1 |iM) for 15 min before addition of a2-AR specific agonist, UK14304 (1 pM) alone or forskolin and UK14304. Data are represented as means 士 SEM, from three independent experiments.
I
125 • basal 5-1 • forskolin (1 |iM) c J • R-(+)-methanandamide (100 nM) ± T • meth + fors ^ g 圓 ra i • AM251 (10 [M) •g (5 3 _ 圏• • meth + AM251 + fors I•！ III PI III i 11 111 Mixed Neuron Glia
Figure 5.10 Characterization of the CBl receptor-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed
DRG cells, neuron-enriched cells and glial cells. Cells were incubated with control solution (basal), or CBl specific antagonist, AM251 (10 |JM) for 15 min before addition of CBl specific agonist, R-(+)-methanandainide (100 nM) alone or forskolin and R-(+)-methanandamide. Data are represented as means 士 SEM, from three independent experiments.
1
126 5- • basal • forskolin (1 ^M) § T • L759656 (1 \M) •云它 4- _ 工 T • L759656 + fors 3 O _ _ • • AM630 (10 譯） “o ffi 1 _II _ I• • L759656 + AM630 + fors II III _ I .
I QLDMM^MMI _nmrrn^mm_ III III Mixed Neuron Glia Figure 5.11 Characterization of the CB2 receptor-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed
DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution (basal) or CB2 specific antagonist, AM630 (10 [iM) for 15 min before addition of CB2 agonist, L79656 (1 ^M) alone or forskolin and AM630. Data are represented as means 土 SEM，from three independent experiments.
127 o basal o forskolin (1 J.lM) . 0 8-0H-DPAT (1 J.lM) C 8-0H-DPAT + fors o o ;:; .-. o S-WAY100135 (500 nM) o c: ::::s 0 o 8~OH-DPAT +(S)-WAY + fors -c .- o ~ ... Cl) c.> a.. C :E8 Mixed Neuron Glia Figure 5.12 Characterization of the 5-HT lA receptor-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells and glial cells. Cells were incubated with control solution (basal) or 5-HTIA receptor antagonist, S-WAYI00135 (500 nM) for 15 min before addition of 5-HT lA receptor agonist, 8-0H-DPAT (1 ~) alone or forskolin and 8-0H-DPAT. Data are represented as means ± SEM, from three independent experiments. 128 • basal g 口 forskolin (1 |iM) • DAMGO (1 |liM) g • DAMGO + fors ^ J • naloxone (1 jiM) 呂 § Jl • naloxone+ DAMGO + fors Ti^- ill I I II 圍圓• i基i h II • i _ i III 0 n M irn Ki ^ m 11H im 188 tggi • 1-1B rm 188 m B Mixed Neuron Glia Figure 5.13 Characterization of the ji-OR receptor-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution (basal) or |i-OR receptor antagonist, naloxone (1 (oM) for 15 min before addition of |a-OR receptor specific agonist, DAMGO (1 ^M) alone or forskolin and DAMGO. Data are represented as means 士 SEM, from three independent experiments. I 129 5-1 • basal - T • forskolin (1 ^M) o . J • N/OFQ (1 ijJVI) 云它 4•—丄 口 n/ofq + fors =.2 _ 肉 T • J11397 (300 nM) o 52 3- _ _ _ • J11397+ N/OFQ+fors __ 細jil QI 门_攀___n_nn•隣rffa•_ Mixed Neuron Glia Figure 5.14 Characterization of the ORLl receptor-mediated adenylyl cyclase activity in isolated DRG cells. Adenylyl cyclase activity was determined in mixed DRG cells, neuron-enriched cells, and glial cells. Cells were incubated with control solution (basal) or ORLl specific antagonist, J11397 (300 nM) for 15 min before addition of ORLl receptor agonist, N/OFQ (1 ^iM) alone or forskolin and N/OFQ. Data are represented as means 士 SEM，from three independent experiments. < 130 ['HJCAMP production > (% conversion) ['H]CAMP production � tT^^ (% control) ^ 3 0 S i S g \ •• \ �� J. •• \ =>• P V^ P 、•三 t \ ^^ � ’ [^H]cAMP production go (% conversion) 1 o “ to w ‘ \ V; I 1 1 V , [®H]cAMP production % % (% control) rn �� 々令] o S I S 8 \ • \ ™ , \ • \ ’ ~^ ^ \ �•- 2 〜% “ �� [^H]cAMP production Q 〜么 1 (% conversion) ,� ^. ‘ o � to w 一 、 々 .\ - 一 . \、•三 Fig. 5.15 Effect of NGF on Gg- and Gi/o-agonist-stimulated responses in DRG cells. DRG cells were pretreated with (closed bars) or without (open bars) NGF (50 ng/ml) for 48 h in (A) mixed DRG cells, and 24 h in (B) neuron-enriched cells and (C) glia cells. Adenylyl cyclase activity was determined in the three cell cultures on day 2 in response to incubation with control solution (basal), receptor-independent adenylyl cyclase activator (forskolin, 1 |AM), IP receptor agonist (cicaprost, 40 nM), EP4 agonists (ONO-AEL-329, 1 |JM; PGE2, 1 ^IM) and P2-AR agonist (formoterol, 1 joM). Data are represented as means 士 SEM, from three independent experiments. Mixed DRG cells were assayed on day 5 with Gi/o agonists in the (D) absence or (E) presence of NGF. Data are represented as means 士 SD, from one experiment 132 •m^i• n n •• Fig. 5.16 Effect of NGF on phenotypic properties of DRG cells. Immunofluorescence images showing representative fields of (A-B) mixed DRG cells and (C-D) neuron-enriched and (E-F) glial cells pretreated with (A,C,E) control solution and (B,D,F) NGF on day 2. Neurons were identified with anti-TUJ-1 antibodies (1:1000, green), glial cells were identified with anti-GFAP antibodies (1:500, red) and cell nuclei were stained with Hoechst 33342 stain (blue). Scale bar represents 25 |Lim. 133 Chapter 6 Conclusion and further studies In conclusion, this thesis has presented evidence that isolated DRG cell cultures may represent cells with neuropathic pain properties. Using immunocytochemical methods, we identified glial cell proliferation and an increase in GFAP fluorescence intensity in isolated DRG cell cultures. Proliferation of glial cells and increase in GFAP expression are characteristics of glial cell activation after nerve injury. These characteristics of glial cells are likely to contribute to the maintenance of neuropathic pain. Using pharmacological methods, the presence of P2-ARS and CGRP receptors were identified in both neurons and glial cells in addition to EP4 and IP receptors. Similar to EP4 and IP receptor agonist-stimulated responses, glial cell responses to CGRP were inhibited by co-culture with neurons. This study also revealed a novel neuron-glia interaction that neurons inhibited the adenylyl cyclase responses of glial cells to agonist stimulation. Further studies are required to determine the functions for the expression of pain- related receptors on glial cells and the functions and mechanisms of how neurons inhibited glial cell adenylyl cyclase activity. Our on-going investigations have suggested that both neuron-glial cell contact and neuron-derived soluble factor(s) are involved. As we could not identify the proportion of different types of glial cells in 134 isolated DRG cell cultures, we are currently unable to suggest the type(s) of glial cells that neurons are likely to inhibit. The use of isolated DRG cell cultures is useful by allowing studies on separated DRG cells to reveal possible neuron-glia interactions. However, currently, pure neuronal cell cultures cannot be obtained due to proliferation of the "contaminating" glial cells. In order to be sure that the agonist-stimulated responses in neuron- enriched cultures come from neurons but not the contaminating glial cells, embryonic DRG, which consists of neurons but nearly no glial cells, can be isolated and studied (Hall, 2006). In in vivo condition, DRG neurons and SGCs are in close association with each other, forming a unique functional complex. This physical relationship was however interrupted during the process of cell preparation to give dissociated cell cultures. In the future study of DRG cells in cultures, the physical relationship between neurons and glial cells will need to be carefully studied. Besides, one should not presume a similar cellular distribution of proteins such as receptors, transporters and enzymes in in vitro condition based on in vivo study data. ( In conclusion, this thesis suggested that DRG glial cells are clearly not passive bystanders but are actively responding to neuron injury as in isolated DRG cell 135 cultures. Besides, we showed that neuron-glia interactions played a role in modulating responses of neurons and glial cells. With considerable interest in studying neuron-glia interactions and the use of dissociated cultures of DRG and TG cells, the roles played by glial cells, which are capable of sensing nerve injury and expressing pain-related receptors, should be carefully considered. In the future, glial cells are likely to be the targets for the treatment of neuropathic pain by controlling their activation and the communication with neurons. 136 References Abadji V，Lin S, Taha G, Griffin G, Stevenson LA, Pertwee RG, et al. (1994). (R)- methanandamide: a chiral novel anandamide possessing higher potency and metabolic stability. JMedChem 37: 1889-1893. Ahluwalia J, Urban L, Capogna M, Bevan S, Nagy I (2000). Cannabinoid 1 receptors are expressed in nociceptive primary sensory neurons. Neuroscience 100: 685-688. Ahmed Z, Jacques SJ, Berry M, Logan A (2009). Epidermal growth factor receptor inhibitors promote CNS axon growth through off-target effects on glia. Neurobiol Dis36: 142-150. Aley KO, Levine JD (1999). Role of protein kinase A in the maintenance of inflammatory pain. JNeurosci 19: 2181-2186. Ariano MA, Briggs CA, McAfee DA (1982). Cellular localization of cyclic nucleotide changes in rat superior cervical ganglion. Cell Mol Neurobiol 2: 143-156- 156. Backstrom E，Chambers BJ, Kristensson K, Ljunggren HG (2000). Direct NK cell- mediated lysis of syngenic dorsal root ganglia neurons in vitro. J Immunol 165: 4895-4900. Barber R, Ray KP, Butcher RW (1980). Turnover of adenosine 3',5'-monophosphate in WI-38 cultured fibroblasts. Biochemistry-US 19: 2560-2567. Bhatheja K, Field J (2006). Schwann cells: Origins and role in axonal maintenance and regeneration. Int JBiochem Cell B 38: 1995-1999. Bhave G, Zhu W, Wang H, Brasier DJ, Oxford GS, Gereau RWt (2002). cAMP- dependent protein kinase regulates desensitization of the capsaicin receptor (VRl) by direct phosphorylation. Neuron 35: 721-731. Bley KR, Hunter JC，Eglen RM, Smith JA (1998). The role of IP prostanoid receptors in inflammatory pain. Trends Pharmacol Sci 19: 141-147. 137 Bond RA, Clarke DE (1988). Agonist and antagonist characterization of a putative adrenoceptor with distinct pharmacological properties from the a- and P-subtypes. Br J Pharmacol 95: 723-734. Bridges D, Rice AS, Egertova M, Elphick MR, Winter J, Michael GJ (2003). Localisation of cannabinoid receptor 1 in rat dorsal root ganglion using in situ hybridisation and immunohistochemistry. Neuroscience 119: 803-812. Bylund DB (1985). Heterogeneity of alpha-2 adrenergic receptors. Pharmacol Biochem Behav 22(5): 835-843. Bylund DB (1988). Subtypes of a2-adrenoceptors: pharmacological and molecular biological evidence converge. Trends Pharmacol Sci 9(10): 356-3(51. Bylund DB, Eikenberg DC, Hieble JP, Langer SZ, Lefkowitz RJ, Minneman KP, et al. (1994). International Union of Pharmacology nomenclature of adrenoceptors. Pharmacol Rev 46: 121-136. Campana WM (2007). Schwann cells: activated peripheral glia and their role in neuropathic pain. Brain Behav Immun 21: 522-527. Capuano A, De Corato A, Lisi L, Tringali G, Navarra P, Dello Russo C (2009). Proinflammatory-activated trigeminal satellite cells promote neuronal sensitization: relevance for migraine pathology. Mol Pain 5: 43. Ceruti S, Fumagalli M, Villa G，Verderio C，Abbracchio MP (2008). Purinoceptor- mediated calcium signaling in primary neuron-glia trigeminal cultures. Cell Calcium 43: 576-590. Ceruti S, Villa G, Fumagalli M, Colombo L, Magni G, Zanardelli M, et al. (2011). Calcitonin Gene-Related Peptide-Mediated Enhancement of Purinergic Neuron/Glia Communication by the Algogenic Factor Bradykinin in Mouse Trigeminal Ganglia from Wild-Type and R192Q Cav2.1 Knock-In Mice: Implications for Basic Mechanisms of Migraine Pain. JNeurosci 31: 3638-3649. Chen S, Rio C，Ji RR, Dikkes P, Coggeshall RE, WoolfCJ, et al. (2003). Disruption of ErbB receptor signaling in adult non-myelinating Schwann cells causes progressive sensory loss. Nat Neurosci 6: 1186-1193. 138 Chen Y, Sommer C (2006). Nociceptin and its receptor in rat dorsal root ganglion neurons in neuropathic and inflammatory pain models: implications on pain processing. J Peripher Nerv Syst 11: 232-240. Chiba T, Yamaguchi A, Yamatani T, Nakamura A, Morishita T, Inui T，et al. (1989). Calcitonin gene-related peptide receptor antagonist human CGRP-(8-37). Am J Physiol 256: E331-335. Cho HJ, Kim DS, Lee NH, Kim JK, Lee KM, Han KS, et al. (1997). Changes in the a2-adrenergic receptor subtypes gene expression in rat dorsal root ganglion in an experimental model of neuropathic pain. Neuroreport 8: 3119-3122. Cunha TM，Roman-Campos D, Lotufo CM, Duarte HL, Souza GR, Verri WA, Jr., et al. (2010). Morphine peripheral analgesia depends on activation of the PI3Kgamma/AKT/nN0S/N0/KATP signaling pathway. Proc Natl Acad Sci USA 107: 4442-4447. Disa J, Parameswaran N, Nambi P, Aiyar N (2000). Involvement of cAMP- dependent protein kinase and pertussis toxin-sensitive G-proteins in CGRP mediated JNK activation in human neuroblastoma cell line. Neuropeptides 34: 229-233. Doods H，Arndt K, Rudolf K, Just S (2007). CGRP antagonists: unravelling the role of CGRP in migraine. Trends Pharmacol Sci 28: 580-587. Doods H，Hallermayer G, Wu D，Entzeroth M, Rudolf K, Engel W, et al. (2000). Pharmacological profile of BIBN4096BS, the first selective small molecule CGRP antagonist. Br J Pharmacol 129: 420-423. Dooley CT, Spaeth CG, Berzetei-Gurske IP, Craymer K, Adapa ID, Brandt SR, et al. (1997). Binding and in vitro activities of peptides with high affinity for the nociceptin/orphanin FQ receptor, ORLl. J Pharmacol Exp Ther 283: 735-741. Drissi H, Lasmoles F, Le Mellay V, Marie PJ, Lieberherr M (1998). Activation of phospholipase C-pi via Goq/n during calcium mobilization by calcitonin gene-related peptide. J Biol Chem 273: 20168-20174. 139 Durham PL (2004). CGRP-receptor antagonists-a fresh approach to migraine therapy? N Engl J Med 350: 1073-1075. Eftekhari S, Edvinsson L (2010). Possible sites of action of the new calcitonin gene- related peptide receptor antagonists. Ther Adv Neurol Disord 3: 369-378. Elson K, Simmons A, Speck P (2004). Satellite cell proliferation in murine sensory ganglia in response to scarification of the skin. Glia 45: 105-109. Evans BA，Papaioannou M, Hamilton S, Summers RJ (1999). Alternative splicing generates two isoforms of the Ps-adrenoceptor which are differentially expressed in mouse tissues. Br J Pharmacol 127: 1525-1531. Ferreira SH (1980). Local analgesic effect of morphine on the hyperalgesia induced by cAMP, Ca2+, isoprenaline, and PGE2. Adv Prostaglandin Thromboxane Res 8: 1207-1215. Fichna J，Janecka A, Costentin J, Do Rego JC (2007). The endomorphin system and its evolving neurophysiological role. Pharmacol Rev 59: 88-123. Fields KL (1980). The study of Schwann cells using antigenic markers. Trends Neurosci 3: 236-238. Fields KL, Brockes JP, Mirsky R, Wendon LM (1978). Cell surface markers for distinguishing different types of rat dorsal root ganglion cells in culture. Cell 14: 43- 51. Fischbach T, Greffrath W, Nawrath H, Treede RD (2007). Effects of anandamide and noxious heat on intracellular calcium concentration in nociceptive DRG neurons of rats. JNeurophysiol 98: 929-938. Fitzgerald EM, Okuse K，Wood JN, Dolphin AC, Moss SJ (1999). cAMP-dependent phosphorylation of the tetrodotoxin-resistant voltage-dependent sodium channel SNS. J Physiol 5\6\ 433-446. Foord SM, Bonner TI, Neubig RR, Rosser EM, Pin JP, Davenport AP, et al. (2005). International Union of Pharmacology. XLVI. G protein-coupled receptor list. Pharmacol Rev 57: 279-288. 140 Franklin SL, Davies AM, Wyatt S (2009). Macrophage stimulating protein is a neurotrophic factor for a sub-population of adult nociceptive sensory neurons. Mol CellNeurosci A\\ 175-185. Gaoni Y, Mechoulam R (1964). Isolation, Structure, and Partial Synthesis of an Active Constituent of Hashish. J Am Chem Soc 86: 1646-1647. Gether U (2000). Uncovering molecular mechanisms involved in activation of G protein-coupled receptors. Endocr Rev 21: 90-113. Oilman AG (1987). G proteins: transducers of receptor-generated signals. Amu Rev Biochem 56: 615-649. Gladman SJ, Ward RE, Michael-Titus AT, Knight MM, Priestley JV (2010). The effect of mechanical strain or hypoxia on cell death in subpopulations of rat dorsal root ganglion neurons in vitro. Neuroscience 171: 577-587. Gold MS, Dastmalchi S, Levine JD (1997). a2-adrenergic receptor subtypes in rat dorsal root and superior cervical ganglion neurons. Pain 69: 179-190. Gold MS, Dastmalchi S, Levine JD (1996). Co-expression of nociceptor properties in dorsal root ganglion neurons from the adult rat in vitro. Neuroscience 71: 265-275. Graham RM, Perez DM, Hwa J, Piascik MT (1996). ai-adrenergic receptor subtypes. Molecular structure, function, and signaling. Circ Res 78: 737-749. Hall AK (2006). Rodent sensory neuron culture and analysis. Curr Protoc Neurosci Chapter 3: Unit 3 19. Hanani M (2005). Satellite glial cells in sensory ganglia: from form to function. Brain Res Rev 48: 457-476. Hanani M (2010). Satellite glial cells in sympathetic and parasympathetic ganglia: in search of function. Brain Res Rev 64: 304-327. Harper AA, Lawson SN (1985). Electrical properties of rat dorsal root ganglion neurones with different peripheral nerve conduction velocities. J Physiol 359: 47-63. 141 Hay DL, Conner AC, Howitt SG, Takhshid MA, Simms J, Mahmoud K, et al. (2004). The pharmacology of CGRP-responsive receptors in cultured and transfected cells. Peptides 25: 2019-2026. Hein L (2006). Adrenoceptors and signal transduction in neurons. Cell Tissue Res 326: 541-551. Henderson G, McKnight AT (1997). The orphan opioid receptor and its endogenous ligand - nociceptin/orphanin FQ. Trends Pharmacol Sci 18: 293-300. Howlett AC (1995). Pharmacology of cannabinoid receptors. Annu Rev Pharmacol Toxicol 35: 607-634. Howlett AC, Barth F, Bonner TI, Cabral G, Casellas P, Devane WA, et al. (2002). International Union of Pharmacology. XXVII. Classification of cannabinoid receptors. Pharmacol Rev 54: 161-202. Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ，et al. (1994). International Union of Pharmacology classification of receptors for 5- hydroxytryptamine (Serotonin). Pharmacol Rev 46: 157-203. Hoyer D, Hannon JP, Martin GR (2002). Molecular, pharmacological and functional diversity of 5-HT receptors. Pharmacol Biochem Behav 71: 533-554. Hsieh GC, Pai M, Chandran P，Hooker BA，Zhu CZ, Salyers AK, et al. (2011). Central and peripheral sites of action for CB receptor mediated analgesic activity in chronic inflammatory and neuropathic pain models in rats. Br J Pharmacol 162: 428- 440. Hucho T，Levine JD (2007). Signaling pathways in sensitization: toward a nociceptor cell biology. Neuron 55: 365-376. Hucho TB, Dina OA, Levine JD (2005). Epac mediates a cAMP-to-PKC signaling in inflammatory pain: an isolectin B4(+) neuron-specific mechanism. J Neurosci 25: 6119-6126. 142 Huffman JW (2000). The search for selective ligands for the CB2 receptor. Curr Pharm Des 6: 1323-1337. Husmann K, Sexton PM, Fischer JA, Born W (2000). Mouse receptor-activity- modifying proteins 1, -2 and -3: amino acid sequence, expression and function. Mol Cell Endocrinol 162: 35-43. Jasmin L，Vit JP, Bhargava A, Ohara PT (2010). Can satellite glial cells be therapeutic targets for pain control? Neuron Glia Biol 6: 63-71. lessen KR, Thorpe R, Mirsky R (1984). Molecular identity, distribution and heterogeneity of glial fibrillary acidic protein: an immunoblotting and immunohistochemical study of Schwann cells, satellite cells,, enteric glia and astrocytes. J Neurocytol 13: 187-200. Ji RR, Zhang Q, Law PY, Low HH, Elde R, Hokfelt T (1995). Expression of mu-, delta-, and kappa-opioid receptor-like immunoreactivities in rat dorsal root ganglia after carrageenan-induced inflammation. JNeurosci 15: 8156-8166. Julius D，Basbaum Al (2001). Molecular mechanisms of nociception. Nature 413: 203-210. Jung M, Calassi R, Rinaldi-Carmona M, Chardenot P, Le Fur G, Soubrie P, et al. (1997). Characterization of CBl receptors on rat neuronal cell cultures: binding and functional studies using the selective receptor antagonist SR 141716A. J Neurochem 68: 402-409. Khaira SK，Pouton CW, Haynes JM (2009). P2X2, P2X4 and P2Y1 receptors elevate 2+ intracellular Ca in mouse embryonic stem cell-derived GABAergic neurons. Br J Pharmacol 158: 1922-1931. Khanolkar AD, Abadji V, Lin S, Hill WA, Taha G, Abouzid K, et al. (1996). Head group analogs of arachidonylethanolamide, the endogenous cannabinoid ligand. J Med Chem 39: 4515-4519. Khasar SG, Lin YH, Martin A, Dadgar J, McMahon T, Wang D, et al. (1999a). A novel nociceptor signaling pathway revealed in protein kinase C epsilon mutant mice. Neuron 24(1): 253-260. 143 Khasar SG, McCarter G, Levine JD (1999b). Epinephrine Produces a p-Adrenergic Receptor-Mediated Mechanical Hyperalgesia and In Vitro Sensitization of Rat Nociceptors. JNeurophysiol 81: 1104-1112. Kurose H, Isogaya M, Kikkawa H, Nagao T (1998). Domains of Pi and p2 adrenergic receptors to bind subtype selective agonists. Life Sci 62: 1513-1517. LaMotte RH (2007). Acutely dissociated sensory neurons: normal or neuropathic? Focus on: "Dissociation of dorsal root ganglion neurons induces hyperexcitability that is maintained by increased responsiveness to cAMP and cGMP". J Neurophysiol 97: 1-2. • Lan R, Liu Q, Fan P, Lin S, Fernando SR, McCallion D, et al. (1999). Structure- activity relationships of pyrazole derivatives as cannabinoid receptor antagonists. J MedChem 42: 769-776. Lassen LH, Haderslev PA, Jacobsen VB, Iversen HK, Sperling B, Olesen J (2002). CGRP may play a causative role in migraine. Cephalalgia 22: 54-61. Ledda M, De Palo S, Pannese E (2004). Ratios between number of neuroglial cells and number and volume of nerve cells in the spinal ganglia of two species of reptiles and three species of mammals. Tissue Cell 36: 55-62. Lennerz JK, Ruhle V，Ceppa EP, Neuhuber WL, Bunnett NW, Grady EF，et al. (2008). Calcitonin receptor-like receptor (CLR), receptor activity-modifying protein 1 (RAMPl), and calcitonin gene-related peptide (CGRP) immunoreactivity in the rat trigeminovascular system: differences between peripheral and central CGRP receptor distribution. J Comp Neurol 507: 1277-1299. Leon LA, Hoffman BE, Gardner SD, Laping NJ, Evans C, Lashinger ES, et al. (2008). Effects of the P3-adrenergic receptor agonist disodium 5-[(2i?)-2-[[(2i?)-2-(3- chlorophenyl)-2-hydroxyethyl]amino]propyl]-l,3-beiizodioxole-2,2-dicarboxylate (CL-316243) on bladder micturition reflex in spontaneously hypertensive rats. J Pharmacol Exp Ther 326: 178-185. Levine JD, Fields HL, Basbaum Al (1993). Peptides and the primary afferent nociceptor. J Neurosci 13(6): 2273-2286. 144 Li J, Vause CV, Durham PL (2008). Calcitonin gene-related peptide stimulation of nitric oxide synthesis and release from trigeminal ganglion glial cells. Brain Res 1196: 22-32. Li JL, Ding YQ, Li YQ, Li JS, Nomura S, Kaneko T, et al. (1998). Immunocytochemical localization of mu-opioid receptor in primary afferent neurons containing substance P or calcitonin gene-related peptide. A light and electron microscope study in the rat. Brain Res 794: 347-352. Lindsay RM (1988). Nerve growth factors (NGF, BDNF) enhance axonal regeneration but are not required for survival of adult sensory neurons. J Neurosci 8: 2394-2405. Liu XY, Wu SX，Wang YY, Wang W，Zhou L, Li YQ (2005). Changes of 5-HT receptor subtype mRNAs in rat dorsal root ganglion by bee venom-induced inflammatory pain. Neurosci Lett 375: 42-46. Liu ZY, Zhuang DB, Lunderberg T, Yu LC (2002). Involvement of 5- hydroxytryptamine( 1 A) receptors in the descending anti-nociceptive pathway from periaqueductal gray to the spinal dorsal horn in intact rats, rats with nerve injury and rats with inflammation. Neuroscience 112: 399-407. Longhurst PA, Levendusky M (1999). Pharmacological characterization of P- adrenoceptors mediating relaxation of the rat urinary bladder in vitro. Br J Pharmacol 127: 1744-1750. Lorenz W, Lomasney JW, Collins S, Regan JW, Caron MG, Lefkowitz RJ (1990). Expression of three a2-adrenergic receptor subtypes in rat tissues: implications for receptor classification. Mol Pharmacol 38: 599-603. Ma W, Zhang Y, Bantel C, Eisenach JC (2005). Medium and large injured dorsal root ganglion cells increase TRPV-1, accompanied by increased azc-adrenoceptor co-expression and functional inhibition by clonidine. Pain 113: 386-394. Maze M, Tranquilli W (1991). Alpha-2 adrenoceptor agonists: defining the role in clinical anesthesia. Anesthesiology 74: 581-605. 145 Meng F, Xie GX, Thompson RC, Mansour A, Goldstein A, Watson SJ, et al. (1993). Cloning and pharmacological characterization of a rat kappa opioid receptor. Proc Natl Acad Sci USA 90: 9954-9958. Mika J, Li Y，Weihe E, Schafer MK (2003). Relationship of pronociceptin/orphanin FQ and the nociceptin receptor ORLl with substance P and calcitonin gene-related peptide expression in dorsal root ganglion of the rat. Neurosci Lett 348: 190-194. Milligan ED, Watkins LR (2009). Pathological and protective roles of glia in chronic pain. Nat Rev Neurosci 10: 23-36. Minneman KP, Esbenshade TA (1994). ai-adrenergic receptor subtypes. Annu Rev Pharmacol Toxicol 34: 117-133. Morara S, Wang LP, Filippov V，Dickerson IM, Grohovaz F, Provini L, et al. (2008). Calcitonin gene-related peptide (CGRP) triggers Ca^"^ responses in cultured astrocytes and in Bergmann glial cells from cerebellar slices. Eur J Neurosci 28: 2213-2220. Moreno MJ, Terron JA，Stanimirovic DB，Doods H, Hamel E (2002). Characterization of calcitonin gene-related peptide (CGRP) receptors and their receptor-activity-modifying proteins (RAMPs) in human brain microvascular and astroglial cells in culture. Neuropharmacology 42: 270-280. MoriokaN, Tanabe H, Inoue A, Dohi T，Nakata Y (2009). Noradrenaline reduces the ATP-stimulated phosphorylation of p38 MAP kinase via P-adrenergic receptors- cAMP-protein kinase A-dependent mechanism in cultured rat spinal microglia. Neurochemistry International 55: 226-234. Mosahebi A, Woodward B, Wiberg M, Martin R, Terenghi G (2001). Retroviral labeling of Schwann cells: in vitro characterization and in vivo transplantation to improve peripheral nerve regeneration. Glia 34: 8-17. Mulder AH, Burger DM, Wardeh G, Hogenboom F, Frankhuyzen AL ,(1991). Pharmacological profile of various K-agonists at K-�an d6-opioid receptors mediating presynaptic inhibition of neurotransmitter release in the rat brain. Br J Pharmacol 102: 518-522. 146 Munk SA, Harcourt D, Arasasingham P, Gluchowski C, Wong H, Burke J, et al. (1995). Analogs of UK 14,304: Structural features responsible for a! adrenoceptor activity. Bioorg Med Chem Lett 5: 1745-1750. Nagatomo T, Koike K (2000). Recent advances in structure, binding sites with ligands and pharmacological function of p-adrenoceptors obtained by molecular biology and molecular modeling. Life Sci 66: 2419-2426. Naito K, Nagao T, Otsuka M，Harigaya S, Nakajima H (1985). Studies on the affinity and selectivity of denopamine (TA-064), a new cardiotonic agent, for P-adrenergic receptors. Jpn J Pharmacol 38: 235-241. Newman-Tancredi A, Gavaudan S，Conte C，Chaput C, Touzard M，Verriele L, et al. (1998a). Agonist and antagonist actions of antipsychotic agents at 5-HTIA receptors: a [35S]GTPyS binding study. Eur J Pharmacol 355: 245-256. Newman-Tancredi A, Verriele L, Chaput C, Millan MJ (1998b). Labelling of recombinant human and native rat serotonin 5-HTIA receptors by a novel, selective radioligand, [^H]-S 15535: definition of its binding profile using agonists, antagonists and inverse agonists. Naunyn Schmiedebergs Arch Pharmacol 357: 205- 217. Ng KY, Wong YH, Wise H (2011). Glial cells isolated from dorsal root ganglia express prostaglandin E2 (EP4) and prostacyclin (IP) receptors. Eur J Pharmacol 661: 42-48. Ng KY, Wong YH, Wise H (2010). The role of glial cells in influencing neurite extension by dorsal root ganglion cells. Neuron Glia Biol 6: 19-29. Nicholson R，Dixon AK, Spanswick D, Lee K (2005). Noradrenergic receptor mRNA expression in adult rat superficial dorsal horn and dorsal root ganglion neurons. Neurosci Lett 380: 316-321. Nicholson R, Small J, Dixon AK, Spanswick D，Lee K (2003). Serotonin receptor mRNA expression in rat dorsal root ganglion neurons. Neurosci Lett 337: 119-122. Norton WT，Aquino DA, Hozumi I，Chiu FC, Brosnan CF (1992). Quantitative aspects of reactive gliosis: a review. Neurochem Res 17: 877-885. 147 Ohara PT, Vit JP, Bhargava A, Jasmin L (2008). Evidence for a role of connexin 43 in trigeminal pain using RNA interference in vivo. JNeurophysiol 100: 3064-3073. Ohara PT, Vit JP, Bhargava A, Romero M, Sundberg C, Charles AC, et al. (2009). Gliopathic pain: when satellite glial cells go bad. Neuroscientist 15: 450-463. Pannese E (1981). The satellite cells of the sensory ganglia. Adv Anat Embryol Cell Biol 65: 1-111. Pannese E (2010). The structure of the perineuronal sheath of satellite glial cells (SGCs) in sensory ganglia. Neuron Glia Biol 6: 3-10. Pertwee RG (2005). Inverse agonism and neutral antagonism at cannabinoid CBl receptors. Life Sci 16\ 1307-1324. Pertwee RG (1997). Pharmacology of cannabinoid CBl and CB2 receptors. Pharmacol Ther 74: 129-180. Pierce KL, Fremont RT, Lefkowitz RJ (2002). Seven-transmembrane receptors. Nat Rev Mol Cell Biol 3: 639-650. Pluteanu F, Ristoiu V, Flonta ML, Reid G (2002). ai-adrenoceptor-mediated depolarization and P-mediated hyperpolarization in cultured rat dorsal root ganglion neurones. Neurosci Lett 329: 277-280. Pol 0，Puig MM (2004). Expression of opioid receptors during peripheral inflammation. Curr Top Med Chem 4: 51-61. Probst WC, Snyder LA, Schuster DI, Brosius J, Sealfon SC (1992). Sequence alignment of the G-protein coupled receptor superfamily. DNA Cell Biol 11: 1-20. Puehler W, Zollner C, Brack A, Shaqura MA, Krause H，Schafer M, et al. (2004). Rapid upregulation of opioid receptor mRNA in dorsal root ganglia in response to peripheral inflammation depends on neuronal conduction. Neuroscience 129: 473- 479. 148 Pushpendran CK, Garci'a-Sainz JA (1984). RX781094 a potent and selective ai- adrenergic antagonist. Effects in adipocytes and hepatocytes. European Journal of Pharmacology 99: 337-339. Raye WS, Tochon-Danguy N, Pouton CW, Haynes JM (2007). Heterogeneous population of dopaminergic neurons derived from mouse embryonic stem cells: preliminary phenotyping based on receptor expression and function. Eur JNeurosci 25: 1961-1970. Raynor K, Kong H, Chen Y, Yasuda K, Yu L, Bell GI, et al. (1994). Pharmacological characterization of the cloned K-，5-，and |x-opioid receptors. Mol Pharmacol 45: 330-334. Ross RA, Brockie HC, Stevenson LA, Murphy VL, Templeton F, Makriyaimis A, et al. (1999). Agonist-inverse agonist characterization at CBl and CB2 cannabinoid receptors of L759633, L759656, and AM630. Br J Pharmacol 126: 665-672. Ross RA, Courts AA，McFarlane SM, Anavi-Goffer S, Irving AJ，Pertwee RG, et al. (2001). Actions of cannabinoid receptor ligands on rat cultured sensory neurones: implications for antinociception. Neuropharmacology 40: 221-232. Rowlands DK, Kao C, Wise H (2001). Regulation of prostacyclin and prostaglandin E2 receptor mediated responses in adult rat dorsal root ganglion cells, in vitro. Br J Pharmacol 133: 13-22. Sagar DR, Kelly S, Millns PJ, O'Shaughnessey CT, Kendall DA, Chapman V (2005). Inhibitory effects of CBl and CB2 receptor agonists on responses of DRG neurons and dorsal horn neurons in neuropathic rats. Eur JNeurosci 22: 371-379. Salomon Y (1991). Cellular responsiveness to hormones and neurotransmitters: conversion of [3H]adenine to [3H]cAMP in cell monolayers, cell suspensions, and tissue slices. Methods Enzymol 195: 22-28. Sato Y，Kurose H, Isogaya M，Nagao T (1996). Molecular characterization of pharmacological properties of T-0509 for beta-adrenoceptors. Eur J Pharmacol 315: 363-367. 149 Scholz J，Woolf CJ (2007). The neuropathic pain triad: neurons, immune cells and g\isi. Nat Neurosci 10: 1361-1368. Schon F, Kelly JS (1974). The characterisation of [3H]GABA uptake into the satellite glial cells of rat sensory ganglia. Brain Research 66: 289-300. Segond von Banchet G，Pastor A, Biskup C, Schlegel C, Benndorf K, Schaible HG (2002). Localization of functional calcitonin gene-related peptide binding sites in a subpopulation of cultured dorsal root ganglion neurons. Neuroscience 110: 131-145. Shi TS, Winzer-Serhan U, Leslie F，Hokfelt T (2000). Distribution and regulation of a2-adrenoceptors in rat dorsal root ganglia. Pain 84: 319-330. Simon B, Kather H, Kommerell B (1980). Activation of human colonic mucosal adenylate cyclase by prostaglandins. Adv Prostaglandin Thromboxane Res 8: 1617- 1620. Smith PB, Compton DR, Welch SP, Razdan RK, Mechoulam R, Martin BR (1994). The pharmacological activity of anandamide, a putative endogenous cannabinoid, in mice. J Pharmacol Exp Ther 270: 219-227. Sobue G, Yasuda T, Mitsuma T, Pleasure D (1986). Schwann cell galactocerebroside of unmyelinated fibers is inducible by derivatives of adenosine 3',5-monophosphate. Neurosci Lett 72: 253-257. St-Laurent J, Boulay ME, Prince P, Bissonnette E, Boulet LP (2006). Comparison of cell fixation methods of induced sputum specimens: an immunocytochemical analysis. J Immunol Methods 308: 36-42. Stein C, Zollner C (2009). Opioids and sensory nerves. Handb Exp Pharmacol. 495- 518. Suadicani SO, Cherkas PS, Zuckerman J，Smith DN, Spray DC, Hanani M (2010). Bidirectional calcium signaling between satellite glial cells and neurons in cultured mouse trigeminal ganglia. Neuron Glia Biol 6: 43-51. 150 Sugiura T, Kondo S, Sukagawa A, Nakane S, Shinoda A, Itoh K, et al. (1995). 2- Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun 215: 89-97. Summers RJ, McMartin LR (1993). Adrenoceptors and their second messenger systems. J Neurochem 60: 10-23. Supowit SC, Hallman DM, Zhao H, DiPette DJ (1998). a�-Adrenergi creceptor activation inhibits calcitonin gene-related peptide expression in cultured dorsal root ganglia neurons. Brain Research 782: 184-193. Takeda M, Takahashi M, Matsumoto S (2009). Contribution of the activation of satellite glia in sensory ganglia to pathological pain. Neurosci Biobehav Rev 33: 784- 792. Tang HB, Li YS, Nakata Y (2007). The release of substance P from cultured dorsal root ganglion neurons requires the non-neuronal cells around these neurons. J Pharmacol Sci 105: 264-271. Thalakoti S, Patil VV，Damodaram S, Vause CV, Langford LE, Freeman SE, et al. (2007). Neuron-glia signaling in trigeminal ganglion: implications for migraine pathology. Headache 47: 1008-1023; discussion 1024-1005. Todorovic S, Anderson EG (1992). Serotonin preferentially hyperpolarizes capsaicin-sensitive C type sensory neurons by activating 5-HTIA receptors. Brain Res 585:212-218. Tognetto M, Amadesi S, Harrison S, Creminon C, Trevisani M, Carreras M, et al. (2001). Anandamide excites central terminals of dorsal root ganglion neurons via vanilloid receptor-1 activation. J Neurosci 21: 1104-1109. Toma JS, McPhail LT，Ramer MS (2007). Differential RIP antigen (CNPase) expression in peripheral ensheathing glia. Brain Research 1137: 1-10. Ulrich CD, 2nd, Holtmann M, Miller LJ (1998). Secretin and vasoactive intestinal peptide receptors: members of a unique family of G protein-coupled receptors. Gastroenterology 114: 382-397. . . 151 Vause CV, Durham PL (2010). Calcitonin gene-related peptide differentially regulates gene and protein expression in trigeminal glia cells: findings from array analysis. Neurosci Lett Al?>\ 163-167. Vega JA, del Valle-Soto ME, Calzada B, Alvarez-Mendez JC (1991). Immunohistochemical localization of S-100 protein subunits (alpha and beta) in dorsal root ganglia of the rat. Cell Mol Biol 37: 173-181. Vit JP, Jasmin L, Bhargava A，Ohara PT (2006). Satellite glial cells in the trigeminal ganglion as a determinant of orofacial neuropathic pain. Neuron Glia Biol 2: 247-257. Vit JP, Ohara PT, Bhargava A, Kelley K, Jasmin L (2008). Silencing the Kir4.1 potassium channel subunit in satellite glial cells of the rat trigeminal ganglion results in pain-like behavior in the absence of nerve injury. J Neurosci 28: 4161-4171. Vroemen M, Weidner N (2003). Purification of Schwann cells by selection of p75 low affinity nerve growth factor receptor expressing cells from adult peripheral nerve. J Neurosci Methods 124: 135-143. Walker CS，Conner AC, Poyner DR，Hay DL (2010). Regulation of signal transduction by calcitonin gene-related peptide receptors. Trends Pharmacol Sci 31: 476-483. Wettschureck N, Offermanns S (2005). Mammalian G proteins and their cell type specific functions. Physiol Rev 85: 1159-1204. Wise H (2006). Lack of interaction between prostaglandin E2 receptor subtypes in regulating adenylyl cyclase activity in cultured rat dorsal root ganglion cells. Eur J Pharmacol 535: 69-77. Woolf CJ, Ma Q (2007). Nociceptors - noxious stimulus detectors. Neuron 55: 353- 364. Woolf CJ, Salter MW (2000). Neuronal plasticity: increasing the gain in pain. Science 288: 1765-1769. 152 Xie W, Strong JA, Zhang JM (2009). Early blockade of injured primary sensory afferents reduces glial cell activation in two rat neuropathic pain models. Neuroscience 160: 847-857. Xu XJ, Wiesenfeld-Hallin Z (1996). Calcitonin gene-related peptide (8-37) does not antagonize calcitonin gene-related peptide in rat spinal cord. Neurosci Lett 204: 185- 188. Yalcin I, Tessier LH, Petit-Demouliere N, Waltisperger E，Hein L, Freund-Mercier MJ, et al. (2010). Chronic treatment with agonists of P2-adrenergic receptors in neuropathic pain. Exp Neurol 221: 115-121. Yang SH, Chen YJ, Tung PY, Lai WL, Chen Y, Jeng CJ, et al. (2008). Anti-Thy-1 antibody-induced neurite outgrowth in cultured dorsal root ganglionic neurons is mediated by the c-Src-MEK signaling pathway. J Cell Biochem 103: 67-77. Ye Z, Wimalawansa SJ, Westlund KN (1999). Receptor for calcitonin gene-related peptide: localization in the dorsal and ventral spinal cord. Neuroscience 92: 1389- 1397. Zhang H, Mei X, Zhang P, Ma C, White FA, Donnelly DF, et al. (2009). Altered functional properties of satellite glial cells in compressed spinal ganglia. Glia 57: 1588-1599. Zhang J, Hoffert C, Vu HK, Groblewski T，Ahmad S, O'Donnell D (2003). Induction of CB2 receptor expression in the rat spinal cord of neuropathic but not inflammatory chronic pain models. Eur J Neurosci 17: 2750-2754. Zhang X, Chen Y，Wang C, Huang LY (2007a). Neuronal somatic ATP release triggers neuron-satellite glial cell communication in dorsal root ganglia. Froc Natl AcadSciUSA 104: 9864-9869. Zhang Z, Winbom CS, Marquez de Prado B, Russo AF (2007b). Sensitization of calcitonin gene-related peptide receptors by receptor activity-modifying protein-1 in the trigeminal ganglion. JNeurosci 27: 2693-2703. Zhou X, Wang L, Hasegawa H，Amin P, Han BX, Kaneko S, et al. (2010). Deletion of PIK3C3/Vps34 in sensory neurons causes rapid neurodegeneration by disrupting 153 the endosomal but not the autophagic pathway. Proc Natl Acad Sci USA 107: 9424- 9429. Zhou XF, Deng YS, Chie E, Xue Q, Zhong JH，McLachlan EM, et al. (1999). Satellite-cell-derived nerve growth factor and neurotrophin-3 are involved in noradrenergic sprouting in the dorsal root ganglia following peripheral nerve injury in the rat. Eur J Neurosci 11:1711-1722. 154 m^-m It • CUHmmiM...K Libraries “ 004806915 174 “