Content uploaded by Steven West
Author content
All content in this area was uploaded by Steven West on Mar 19, 2018
Content may be subject to copyright.
NEUROSCIENCE FOREFRONT REVIEW
CIRCUITRY AND PLASTICITY OF THE DORSAL HORN – TOWARD A
BETTER UNDERSTANDING OF NEUROPATHIC PAIN
S. J. WEST,
a
K. BANNISTER,
b
A. H. DICKENSON
b
AND
D. L. BENNETT
a
*
a
Neural Injury Group, Nuffield Department of Clinical
Neuroscience, University of Oxford, West Wing Level 6,
John Radcliffe Hospital, Oxford OX3 9DU, UK
b
Department of Neuroscience, Pharmacology and Physiology,
University College London, Gower Street, London WC1E 6BT, UK
Abstract—Maladaptive plasticity within the dorsal horn (DH)
of the spinal cord is a key substrate for development of neu-
ropathic pain following peripheral nerve injury. Advances in
genetic engineering, tracing techniques and opto-genetics
are leading to a much better understanding of the complex
circuitry of the spinal DH and the radical changes evoked
in such circuitry by nerve injury. These changes can be
viewed at multiple levels including: synaptic remodeling
including enhanced excitatory and reduced inhibitory drive,
morphological and electrophysiological changes which are
observed both to primary afferent inputs as well as DH neu-
rons, and ultimately circuit-level rewiring which leads to
altered connectivity and aberrant processing of sensory
inputs in the DH. The DH should not be seen in isolation
but is subject to important descending modulation from
the brainstem, which is further dysregulated by nerve injury.
Understanding which changes relate to specific disease-
states is essential, and recent work has aimed to stratify
patient populations in a mechanistic fashion. In this review
we will discuss how such pathophysiological mechanisms
may lead to the distressing sensory phenomena experi-
enced by patients suffering neuropathic pain, and the rela-
tionship of such mechanisms to current and potential
future treatment modalities. Ó2015 IBRO. Published by
Elsevier Ltd. All rights reserved.
Key words: neuropathic pain, plasticity, dorsal horn, desce-
nding control, primary afferent.
Contents
Introduction – the burden of neuropathic pain 254
Nerve injury models 255
The DH 255
DH input: Primary afferents 255
Initial injury reactions: Primary afferent signals 257
Changes in primary afferent input and projections 257
Spontaneous activity 257
Expression changes in primary afferents following
nerve injury 257
Sprouting of A-fiber terminals 259
Loss of C-fiber terminals 259
DH processing 259
Activity-dependent plasticity 260
Interneurons in the DH 261
Projection neurons in the DH 262
DH circuitry 262
Peripheral nerve injury: effects on the DH 263
Altered excitatory and inhibitory processing following
neuropathy 264
Altered structure and function of projection neurons 265
DH rewiring 265
Descending controls from the brain to the spinal cord 265
5HT 267
NA 268
Plasticity in descending controls following peripheral
nerve injury 268
Linking mechanisms to patients 269
Neuropathic mechanisms within patients 269
Conclusion 269
Acknowledgments 269
References 269
INTRODUCTION – THE BURDEN OF
NEUROPATHIC PAIN
Neuropathic pain is a type of chronic pain which occurs as
a consequence of a lesion or disease to the
somatosensory nervous system (Jensen et al., 2011). A
plethora of nerve damaging stimuli can result in neuro-
pathic pain, including trauma, metabolic syndromes,
http://dx.doi.org/10.1016/j.neuroscience.2015.05.020
0306-4522/Ó2015 IBRO. Published by Elsevier Ltd. All rights reserved.
*Corresponding author. Address: West Wing Level 6, Nuffield
Department of Clinical Neurosciences, John Radcliffe Hospital,
Headley Way, Oxford, Oxfordshire UK OX3 9DU, UK. Tel: +44-
1865-231512.
E-mail address: david.bennett@ndcn.ox.ac.uk (D. L. Bennett).
These authors have contributed equally to this work.
Abbreviations: 5HT, 5-hydroxytryptamine; AMPA, alpha-amino3-hydr
oxy-5-methyl-4-isoxazoleproprionic acid; BDNF, brain-derived
neurotrophic factor; CGRP, calcitonin gene related peptide; CVLM,
caudal ventrolateral medulla; DH, dorsal horn; DRG, dorsal root
ganglion; GABA, gamma-aminobutyric acid; GAD(65), glutamic acid
decarboxylase (65); IB4, isolectin B4; KCC2, potassium-chloride
cotransporter 2; LC, locus coeruleus; LTP, long-term potentiation;
mGlu, metabotropic glutamate receptor; NA, noradrenaline; NGF,
nerve growth factor; NK-1, neurokinin 1; NKCC1, sodium-potassium-
chloride cotransporter 1; NMDA, N-methyl-D-aspartate receptors; NS,
nociceptive specific; PAD, primary afferent depolarization; PAG,
periaqueductal gray; PKC
c
, protein kinase C gamma; RVM, rostral
ventromedial medulla; WDR, wide dynamic range neurons.
Neuroscience 300 (2015) 254–275
254
drug- or toxin-induced injury, hereditary disorders, malig-
nancy and infective or post-infective nerve damage. The
sensory manifestations expressed by patients are com-
plex, including lesion-induced reductions in somatosensa-
tion, combined with paradoxical sensory perceptions,
manifesting as spontaneous or ongoing sensations, with
pain as the dominating positive symptom. Furthermore,
evoked somatosensory stimulation often induces ampli-
fied painful responses following noxious or non-noxious
stimuli (Baron et al., 2010).
Neuropathic pain is common, affecting 6–8% of the
population, and in many cases treatment is inadequate,
leading to significant disability and suffering (Jensen
et al., 2007; Smith et al., 2007; Smith and Torrance,
2012). In a typical randomized control trial, less than
50% of neuropathic pain patients experience satisfactory
pain relief, with side-effects a common problem
(O’Connor and Dworkin, 2009; Dworkin et al., 2010);
and in the community setting it has been found that neu-
ropathic pain patients, on average, suffer with moderate
pain intensity, even though they are receiving the recom-
mended treatment for their condition (O’Connor, 2009).
Poor selection of treatment for individual patients
(Millan, 1999; O’Connor, 2009), and suboptimal pharma-
cological agents (Finnerup et al., 2010) have been sug-
gested to contribute to this lack of efficacy. This
ultimately stems from a lack of understanding of not only
the systems in which neuropathic pain manifests, but also
in how these systems change as a result of a lesion or dis-
ease to the somatosensory nervous system.
In this review, we attempt to address this shortfall
through consideration of the available evidence from pre-
clinical studies of neuropathic pain and correlate these
with clinical findings. We will focus on the dorsal horn
(DH), because this region is the first point of integration
of somatosensory information and it is a key region
where plasticity has been demonstrated (Latremoliere
and Woolf, 2009; Sandku
¨hler and Gruber-Schoffnegger,
2012). We will focus on the four major neuronal compo-
nents of the DH: primary afferent inputs, intrinsic inteneu-
rons and projection neurons of the DH, and descending
modulation from the brain to the DH (Todd, 2010).
An important theme to have emerged over the last two
decades has been the fact that neuropathy models, and in
particular traumatic nerve injury, is associated with a
vigorous glial response. Nerve injury results in a cellular
microglial response within somatotopically appropriate
regions of the DH of the spinal cord. This is not just an
epi-phenomenon, but a profound change in the
functional properties of these cells, and is important in
initiating and maintaining neuropathic pain following
traumatic nerve injury. A morphological and pro-
inflammatory phenotypic switch in astrocytes is also
apparent. The glial contribution to neuropathic pain has
recently been extensively reviewed (Beggs and Salter,
2013; Tsuda et al., 2013; Grace et al., 2014) and so we
will not cover this topic extensively, but will discuss some
issues relating to aberrant signaling between neurons and
glia relevant to excitability changes in the DH.
We begin with a brief overview of neuropathic pain
models, followed by a review of neuroplastic changes
within each of the neuronal compartments of the DH
after nerve injury.
NERVE INJURY MODELS
In order to understand the neurobiological mechanisms
that lead to neuropathic pain, nerve injury paradigms
have been developed. Broadly, these paradigms are
split into traumatic and non-traumatic models. Traumatic
injury paradigms include crush, transection, ligation or
any combination of these mechanical deformations to a
peripheral nerve (Jaggi et al., 2011). Whereas, non-
traumatic injury paradigms cover a more diverse set of
injury models, including the application of neurotoxic sub-
stances (such as chemotherapeutic agents), metabolic-
lesioning substances (such as streptozotocin), or cells
or proteins associated with various neuropathic pain
states (Authier et al., 2009; Jaggi et al., 2011).
Each model presents with different time of onset,
intensity and profile of behavioral hypersensitivity, which
reflects different combinations of underlying
mechanisms. Understanding when and where these
mechanisms present is an important challenge in
understanding neuropathic pain. Since the vast majority
of work has used traumatic nerve injury models, most of
our understanding is derived from these studies;
however, where possible we will contrast traumatic
injury with other models.
Most studies of neuropathic pain have used rodent
and there is currently a dearth of human post-mortem
tissue for anatomical studies, although increased tissue
banking will help to rectify this. An exciting development
in understanding pathophysiological changes within
human spinal cord is the recent improvements in spinal
imaging (Stroman et al., 2014), which may enable us to
study altered functional status in response to primary
afferent input or modulation of descending control
(Eippert et al., 2009; Rempe et al., 2014).
THE DH
The DH is divided into six parallel laminae based on
cytoarchitectonics, originally described by Rexed (1952),
and later extended to the rodent (Molander et al., 1984)
(Fig. 1A). It is composed principally of four major neuronal
components: primary afferent input, intrinsic neurons, pro-
jection neurons, and descending input from higher cen-
ters (Todd, 2010). Each of these systems display
profound changes following peripheral nerve lesions.
We begin by considering the primary afferent terminals
within the DH; an important component that transmits
sensory information to the DH, and the first component
disturbed by peripheral nerve injury.
DH input: Primary afferents
The most commonly used scheme to classify primary
afferent neurons has been by their myelination, diameter
and conduction velocity, which are inherently linked
(Table 1). Histochemical markers have been used to
delineate primary afferent neurons. All A-fiber neurons
appear to selectively express the heavy chain
S. J. West et al. / Neuroscience 300 (2015) 254–275 255
neurofilament (Lawson et al., 1984), whereas C-fibers can
be split broadly into three groups based on their expres-
sion of calcitonin gene-related peptide (CGRP; peptidergic
C-fibers), and isolectin B4 (IB4; non-peptidergic C-fibers)
(Snider and McMahon, 1998), as well as expression of tyr-
osine hydroxylase (which mark C-fiber low-threshold
mechanoreceptors) (Li et al., 2011); such divisions are
operationally helpful, however recent unbiased molecular
analysis indicates that the situation is clearly more com-
plex than these gross subdivisions (Usoskin et al., 2015).
Primary afferent neurons produce distinct termination
patterns within the DH that is based on the class of
neuron. Broadly, a graded pattern of termination is
seen, where C-fibers synapse in the superficial laminae
(I/II), Adfibers synapse in the superficial as well as
deeper laminae, and Abfibers are more restricted to
deeper laminae (III–V, see Fig. 1B) (Millan, 1999).
All primary afferent fibers use the neurotransmitter
glutamate. Glutamate has been shown to be released
after electrical and noxious stimulation, and its uptake
and release are decreased after dorsal rhizotomy (De
Biasi and Rustioni, 1988). Glutamate is found in many pri-
mary afferent terminals, both large and small, which end
in laminas I, III and IV and in small dorsal root ganglion
(DRG) cells (Battaglia and Rustioni, 1988; De Biasi and
Rustioni, 1988). Additionally, glutamate has also been
shown to be co-localized with peptides such as substance
P and/or CGRP (Battaglia and Rustioni, 1988; De Biasi
and Rustioni, 1988), and the release of such transmitters
after a noxious stimulus seems to show that they play a
role in nociception. Glutamate actions are exerted through
its ionotropic receptors kainate, the alpha-amino3-hydro
xy-5-methyl-4-isoxazoleproprionic acid (AMPA) receptor
and N-methyl-
D
-aspartate (NMDA) receptor and its meta-
botropic receptors (mGlu) (Watkins and Evans, 1981;
Woodruff et al., 1987; Dickenson, 1995).
Primary afferents form complex synaptic structures
within the DH. Synaptic glomeruli are complex
arrangements of synaptic contacts between primary
afferent fibers and DH neurons. These structures show
multiple connectivity patterns, including axo-dendritic,
axo-axonic and dendro-axonic synapses (Ribeiro-Da-
Silva and Coimbra, 1982; Ribeiro-Da-Silva et al., 1985).
Two type are recognized: type 1, which are formed from
IB4 + primary afferent fibers; and type 2, that are formed
by Adhair follicle axons (Ribeiro-Da-Silva and Coimbra,
1982; Ribeiro-Da-Silva et al., 1985, 1986). Peptide-
containing C-fibers may also form glomeruli within the
DH (Knyihar-Csillik et al., 1982), although many have
been observed to form simple synaptic contacts
(Ribeiro-Da-Silva et al., 1989).
These structures are important in primary afferent
depolarization (PAD), a form of primary afferent pre-
synaptic inhibition mediated through gamma-
aminobutyric acid (GABA) by GABAergic interneurons
(Rudomin and Schmidt, 1999). The mechanisms underly-
ing PAD and pre-synaptic inhibition are well understood.
Primary afferents possess a high intracellular chloride
concentration due to high sodium–potassium-chloride
cotransporter 1 (NKCC1) and low potassium-chloride
cotransporter 2 (KCC2) expression levels (Coull et al.,
2003; Price et al., 2009). GABAergic inteneurons, acting
via GABA
A
receptors (Eccles et al., 1963; Gallagher
et al., 1978), cause an efflux of chloride ions, and thus
depolarize the primary afferent fiber. Crucially, this is not
sufficient to release synaptic vesicles, but does result in
inactivation of sodium and calcium channels (Graham
and Redman, 1994), therefore preventing action poten-
tials from initiating synaptic transmission – a process
referred to as shunting.
Suprathreshold excitation of this mechanism, either
via increased efficacy of GABA on primary afferents, or
Fig. 1. Different primary afferent classes terminate in distinct regions of the dorsal horn. (A) The dorsal horn is divided into ten separate laminae
based on cytoarchitectonics. (B) Primary afferents terminate in the dorsal horn into specific laminae, with C fibers innervating superficial (I/II)
laminae, whereas A-fibers tend to innervate deeper (III/IV) laminae.
Table 1. Overview of peripheral nerve fiber properties
Fiber
type
Axon diameter
(
l
m)
Myelination Conduction
velocity (m/s)
C-fiber Thin (0.4–1.2) Unmyelinated Slow (0.5–2.0)
Adfiber Medium (2–6) Thinly
myelinated
Intermediate
(12–30)
Abfiber Fast (>10) Thickly
myelinated
Fast (30–100)
256 S. J. West et al. / Neuroscience 300 (2015) 254–275
alterations in the chloride gradient within primary
afferents, has been suggested to cause direct excitation
of primary afferent terminals, and may play a role in
neurogenic inflammation (Willis, 1999; Bardoni et al.,
2013).
Initial injury reactions: Primary afferent signals
Given that peripheral neuropathic pain develops following
injury to the axons of DRG cells, the early events in these
neurons must have a key role in the initiation of
neuropathic pain. Ultimately, axotomy results in
profound phenotypic changes in injured DRG cells, and
depending on the threshold used, altered expression in
up to a quarter of known genes (Perkins et al., 2014)
including: growth factor receptors, axon guidance mole-
cules, neurotransmitters and ion channels which will have
important consequences for DH function (Costigan et al.,
2002; Xiao et al., 2002).
In traumatic injury, initial transection of a nerve results
in a small burst of neuronal activity (Adrian, 1930; Wall
et al., 1974; Schneider, 1992), which quickly subsides,
and is followed by ‘positive’ and ‘negative’ molecular sig-
nals as a consequence of calcium flux and altered retro-
grade signaling within the axon.
The initial reaction to traumatic nerve injury involves
the propagation of calcium waves from the site of injury
toward the cell soma, and which are important for the
two dominant retrograde signaling complexes that form
following nerve injury: mitogen-activated protein kinases
(Zrouri et al., 2004), including JNK (Lindwall and Kanje,
2005); and nuclear localization signaling molecules such
as importins (Hanz et al., 2003). These signals are retro-
gradely transported via dynein motors to the soma, where
they induce profound changes in gene transcription
(Michaelevski et al., 2010; Perkins et al., 2014).
Concurrent with motor-dependent retrograde
signaling, the ongoing flux of neurotrophin support that
is derived from peripheral targets is altered in injured
axons. Treating uninjured animals with antibodies
against nerve growth factor (NGF) replicated some of
the axotomy-related changes to DRG protein expression
(Shadiack et al., 2001). Following injury, levels of NGF
appear to fall rapidly, with nerve transection exhibiting a
stronger reduction (Raivich et al., 1991). Consistent with
loss of NGF signaling, intrathecal injection of NGF after
nerve crush delays the neuronal responses of nerve injury
(Gold, 1997). Furthermore, supplementation of a syn-
thetic peptide mimicking NGF reduces pain behavior
associated with nerve injury(Colangelo et al., 2008;
Cirillo et al., 2009). These data in general support a role
for a loss of neurotrophic signaling in the onset of the neu-
ropathic pain phenotype, which is likely to be important for
central terminals of primary afferents following nerve
injury (Fig. 2A).
Changes in primary afferent input and projections
Spontaneous activity. Spontaneous activity in injured
sensory neurons has been observed in both animal
models and in patients following nerve injury, and is
believed to play an important role in the initiation and
maintenance of neuropathic pain. It has been shown
that 1–3 days following nerve injury a significant
proportion of A-fibers demonstrate spontaneous activity,
peaking at 1–2 weeks and then declining (Govrin-
Lippmann and Devor, 1978; Liu et al., 2000; Heinke
et al., 2004).
C-fibers also develop spontaneous activity at a lower
frequency, and importantly, spontaneous activity is not
restricted to injured afferents, but is also observed in
neighboring un-injured afferents, probably as a
consequence of the inflammatory nerve environment
(Hulse, 2010).
Spontaneous activity in primary afferents has also
been observed in rodent models of chemotherapy (Xiao
and Bennett, 2008) and painful diabetic neuropathy
(Khan et al., 2002; Sood et al., 2013) and crucially in neu-
ropathic pain states in humans (Ørstavik and Jørum,
2010). A recent clinical study highlights the importance
of ectopic activity in maintaining peripheral neuropathic
pain as peripheral nerve blocks using local anesthetic
were highly efficacious in reducing ongoing pain
(Haroutounian et al., 2014). In animal models, the behav-
ioral consequences of spontaneous activity have been
investigated. Giving brief and high-intensity C-fiber input
results in mechanical hypersensitivity in adult rats
(Hathway et al., 2009), suggesting continuous inputs
can sensitize central circuitry. This study further demon-
strated that C-fiber input can evoke a microglial response
within the DH.
Expression changes in primary afferents following
nerve injury. Traumatic nerve injury alters the pattern of
expression of neurotransmitters/neuromodulators within
sensory neurons. For instance C-fibers reduce their
expression of CGRP as well as show increased
expression of vasoactive intestinal polypeptide (Shehab
and Atkinson, 1986)(Fig. 2B). These effects seem to be
driven, at least in part, by altered neurotrophic support,
as inhibition of axonal transport with vinca alkaloids
results in a similar phenotype (Csillik et al., 1985;
Knyihar-Csillik et al., 1991).
A-fibers begin to express neuropeptide Y and brain-
derived neurotrophic factor (BDNF) following traumatic
nerve injury paradigms (Wakisaka et al., 1992; Obata
et al., 2006)(Fig. 2B). Interestingly some of these
changes are likely to be adaptive. For instance, pharma-
cological blockade (Intondi et al., 2008) and conditional
deletion (Solway et al., 2011) following nerve injury sug-
gests that neuropeptide Y actually suppresses pain
related hypersensitivity within the DH.
The
a
2
d-1 calcium channel subunit is upregulated on
primary afferent terminals after nerve injury (Li et al.,
2004), and appears to contribute to mechanical hypersen-
sitivity seen after nerve injury (Li et al., 2004, 2014).
Upregulation of the 2d-1 subunit has been linked with
enhanced neurotransmission of primary afferents as an
apparent increase in pre-synaptic terminals (Li et al.,
2014).
PAD has been hypothesized to be dysregulated after
nerve injury, via modified chloride ion gradients within
S. J. West et al. / Neuroscience 300 (2015) 254–275 257
258 S. J. West et al. / Neuroscience 300 (2015) 254–275
primary afferents, which could increase excitability in
primary afferent terminals, or even reach sufficient
intensity to excite these fibers (Cervero et al., 2003).
Alterations in the chloride ion potential in primary afferents
after nerve injury have been reported in spinal cord slices
(Pieraut et al., 2007), where the investigators observed a
shift in the GABA
A
receptor reversal potential toward
depolarized potentials in injured neurons. Recently, this
shift has been shown to occur in the DH in vivo (Wei
et al., 2013; Chen et al., 2014). Primary afferent neurons
in culture showed a shift in the GABA
A
receptor reversal
potential, as well as a loss in GABA-mediated conduc-
tance, which were due to increased NKCC1 activity and
reduced GABA receptor, respectively (Chen et al., 2014)
(Fig. 2C). NKCC1 has been shown to be specifically
upregulated following trigeminal nerve injury in primary
afferent neurons (Wei et al., 2013), and using fluorescent
voltage-sensitive dyes to record pre-synaptic activity in
trigeminal nuclear complexes, GABA
A
agonists failed to
reduce primary afferent terminal discharge in injured ani-
mals compared to sham-treated animals, and this was
reversed by the application of butmetanide, an NKCC1
inhibitor (Wei et al., 2013). Importantly, a conditional knock
out of the b-3 subunit of the GABA
A
receptor in the
voltage-gated sodium channel 1.8 (Na
V
1.8) expressing
primary afferents showed spontaneous mechanical and
heat hypersensitivity, which was not further altered by
nerve injury (Chen et al., 2014), which implies that the
hypersensitivity seen following nerve injury is at least par-
tially dependent on a loss of pre-synaptic inhibition
(Fig. 2C). These effects were recapitulated with the appli-
cation of BDNF, and inhibition of BDNF following nerve
injury prevented the change in GABA conductance,
GABA
A
receptor potential, and behavioral sensitivity, sug-
gesting BDNF plays an important role in regulating pre-
synaptic inhibition in the DH after nerve injury (Chen
et al., 2014). Taken together, these data indicate an impor-
tant role for pre-synaptic inhibition in the generation of
hypersensitivity following peripheral nerve injury.
Sprouting of A-fiber terminals. The sprouting of
A-fibers from deep to superficial laminae one week
following traumatic nerve injury has been proposed to
drive hypersensitivity after peripheral nerve injury (Woolf
et al., 1992; Proudlock et al., 1993). Here, Woolf et al.
demonstrated that exogenous cholera toxin bsubunit
transported by A-fibers, began to occupy the superficial
DH following injury. Single axon tracing was also per-
formed which suggested A-fibers began to sprout termi-
nals into superficial laminae. This sprouting was
suggested to contribute to allodynia following nerve injury,
as A-fiber sprouting into lamina I could potentially allow
innocuous signals to gain access to nociceptive projection
pathways. However, this study is now controversial and
the consensus is that, although this is an attractive
hypothesis, such sprouting of low-threshold A-fiber
mechanoreceptors does not occur.
Firstly, the tracing method used is not specific to
A-fibers following nerve injury (Shehab et al., 2003,
2004). Secondly, the findings of individual axon fills could
not be replicated. Following axotomy, low-threshold
A-fibers were not found to sprout into lamina 2, and the only
A-fibers found to project into this region were A-fiber noci-
ceptors which are known to project to lamina 2 (Woodbury
et al., 2008). Although evidence is lacking for wholesale
movement of A-fiber terminals between laminae this does
not negate more subtle changes for instance in pre-
synaptic release sites or synaptic size, features which have
not been definitively characterized as yet.
Loss of C-fiber terminals. Transganglionic
degenerative atrophy, which involves the degeneration
of primary afferent C-fibers within lamina II of the DH,
has been reported following nerve lesions (Knyihar-
Csillik and Csillik, 1981; Tajti et al., 1988), which is
reported to occur and recover following a crush injury
(Csillik and Knyihar-Csillik, 1981) as well as local admin-
istration of vinca alkaloids (Csillik and Knyihar-Csillik,
1982). Analysis of synapses from glomeruli derived from
C-fibers reveals a loss of synapses progressively over
the course of two weeks in transection (Castro-Lopes
et al., 1990) and a loss followed by recovery in constric-
tion injury (Bailey and Ribeiro-Da-Silva, 2006). These
data support the view that blockade of neurotrophic sig-
nals, by nerve lesion or blockade of axonal transport,
leads to a loss of C-fiber synapses within the DH.
Is it possible to reconcile this loss of input with the
gain-of-function associated with neuropathic pain
states? The non-peptidergic population of C-fibers
appear to be most vulnerable in terms of terminal
loss/withdrawal. This population is important in
mediating noxious mechanosensation (Cavanaugh
et al., 2009) as shown by targeted ablation. However,
their terminals express prostate acid phosphatase
(Zylka et al., 2008), an enzyme that generates adenosine,
which acts to suppress nociceptive signaling within the
DH; in which respect the loss of such terminals would
be maladaptive. The loss of C-fiber input may also help
to facilitate A-fiber access to nociceptive pathways,
through homeostatic adaptation of DH circuitry, which
has lost input (see Section DH rewiring).
DH PROCESSING
The DH comprises three basic neuronal subtypes: local
interneurons, propriospinal neurons and projection
3
Fig. 2. Primary afferent fibers show changes after nerve injury. (A) Following nerve transection, regular retrograde transport of trophic signals is
disrupted, leading to a rise in intracellular calcium, as well as the initiation of axonal transport of injury-related signals. (B) Primary afferent neurons
show changes in expression and activity. Injured C-fibers show increased VIP and reduced CGRP expression, and injured A-fibers display
increased NPY and BDNF expression. Injured A- and C-fibers as well as uninjured C-fibers show spontaneous activity. (C) Loss of primary afferent
depolarization results from a BDNF-mediated up-regulation of NKCC1, which results in increased chloride ion efflux, switching pre-synaptic
inhibition into excitation. NT, neurotrophic factor; IMP, importin; JNK, c-Jun N-terminal kinase; VIP, vasoactive intestinal peptide; CGRP, calcitonin
gene related peptide; NPY, neuropeptide Y; BDNF, brain-derived neurotrophic factor; NKCC1, sodium potassium chloride co-transporter 1.
S. J. West et al. / Neuroscience 300 (2015) 254–275 259
neurons, which form complex circuits within the DH
(Todd, 2010). Plasticity within such circuits following
repetitive nociceptive input (Ikeda et al., 2006;
Latremoliere and Woolf, 2009; Sandku
¨hler, 2009) can
act to facilitate or amplify incoming somatosensory inputs,
and ultimately modifying the processing and propagation
of somatosensory information to higher centers.
An early and still influential theory postulating the
importance of circuits involving DH interneurons was the
‘gate control’ theory of pain proposed by Melzack and
Wall (Melzack and Wall, 1965). Melzack and Wall argued
that spinal projection neurons transmitting nociceptive
information received input not only from nociceptors but
also from low-threshold Abafferents. This low-threshold
input is however gated by feed-forward activation of inhi-
bitory interneurons. Disinhibition would therefore lead to
these low-threshold inputs activating the transmission
neurons and the development of allodynia.
Understanding the function of these interneurons and
how they relate to pain perception will help us to recog-
nize how this system breaks down in various chronic pain
states.
Activity-dependent plasticity
A number of discoveries have led to the understanding
that activity-dependent plasticity in DH neurons plays a
key role in the generation of hypersensitivity. The first
process describes a progressive increase in DH output
with repetitive low-frequency stimulation of C-fibers, an
electrophysiological phenomenon referred to as wind up
(Mendell and Wall, 1965; Mendell, 1966). Later, it was
recognized that repeated C-fiber input following intense
and sustained noxious stimulation generated increased
excitability in flexor motorneurons, causing a reduced
threshold for reflex withdrawal, an increase in receptive
field size and novel responses to normally innocuous Ab
fiber inputs. This effect proved to be driven by plasticity
in the spinal cord, and was thus named central sensitiza-
tion (Woolf, 1983; Latremoliere and Woolf, 2009). This
work was extended by experiments which looked at
C-fiber-induced DH field potentials, which showed that
high-frequency C-fiber input was capable of significantly
increasing the amplitude of DH field potentials, reminis-
cent of long-term potentiation (LTP), and was thus called
spinal LTP (Liu and Sandku
¨hler, 1995; Drdla and
Sandku
¨hler, 2009)(Fig. 3A). These experiments were
later broadened by showing that neurons which project
to the periaqueductal gray (PAG) responded specifically
to low-frequency (2 Hz) C-fiber inputs with a prolonged
increase in excitatory postsynaptic potentials, that DH
field potentials are significantly increased with natural
low-frequency afferent barrages (Ikeda et al., 2006), and
that spinal LTP is heterosynaptic through induction of
C-fiber-induced LTP between intrinsic DH neuronal
connections (Fenselau et al., 2011).
These phenomena all demonstrate that activity-
dependent plasticity in the DH requires C-fiber input. It
has been previously demonstrated, using in vitro spinal
cord slices, that Ad- and C-fiber strength stimulation
produces prolonged excitatory post-synaptic potentials
in recorded DH (Yoshimura and Jessell, 1989) and ventral
horn (Thompson et al., 1990) neurons. C-fiber terminals
contain glutamate and neuropeptides such as substance
P and CGRP (Merighi et al., 1991), and DH neuron
C-fiber potentials contain both a glutamate (Thompson
et al., 1990) and neuropeptide (Nagy et al., 1993) compo-
nent. These prolonged potentials allow pronounced tem-
poral summation to occur, even at low frequencies,
leading to cumulative depolarization of dorsal and ventral
horn neurons (Sivilotti et al., 1993). The summation is not
linear, but progressively increases with continued input.
A large body of work now reveals that activity-dependent
plasticity comprises multiple distinct phases, each associ-
ated with specific mechanisms. Below we briefly review
these mechanisms, although more detailed accounts can
be found in the literature (Drdla and Sandku
¨hler, 2009;
Latremoliere and Woolf, 2009; Sandku
¨hler, 2009).
It is well established that the NMDA receptor plays an
integral role in activity-dependent plasticity in the DH,
since inhibition of NMDA currents can inhibit activity-
dependent plasticity (Dickenson and Sullivan, 1987;
Woolf and Thompson, 1991; Liu and Sandku
¨hler, 1995).
Its role in the progressive output of DH neurons with con-
tinued C-fiber input is dependent on its ligand- and
voltage-gated behavior (Dickenson, 1990). At resting
membrane potentials, the NMDA receptor is under
Mg
2+
blockade. This is lifted following depolarization
(Mayer et al., 1984), which then allows glutamate to acti-
vate the NMDA receptor, and results in an increased input
of Na
+
and Ca
2+
ions into the DH neuron. Much of the
acute response in DH neurons can be explained using
this model, as neuropeptide blockade and morphine
administration may inhibit the C-fiber-evoked slow synap-
tic potentials sufficiently to prevent NMDA receptor activa-
tion and thus stop amplification of the C-fiber input (Ma
and Woolf, 1995; Sivilotti et al., 1995).
Depolarization sufficient to induce NMDA receptor
activation occurs through glutamate acting on AMPA,
kainate receptors, and through neuropeptides such as
substance P (Afrah et al., 2002; Khasabov et al., 2002),
CGRP (Woolf and Wiesenfeld-Hallin, 1986; Sun et al.,
2003) and BDNF (Lever et al., 2001). mGlu Type I are
also important for the induction of DH activity-dependent
plasticity (Azkue et al., 2003; Derjean et al., 2003).
The rise in intracellular Ca
2+
is considered a key
trigger for subsequent plasticity of synapses within the
DH. This rise results from activation of NMDA- and
Ca
2+
-permeable AMPA-receptors, as well as voltage-
gated Ca
2+
channels and intracellular stores (Coderre
and Melzack, 1992; Morisset and Nagy, 1999; Polga
´r
et al., 2008). This influx of Ca
2+
ions results in the activa-
tion of a number of second messenger systems, including
protein kinase A and protein kinase C
Calcium/Calmodulin-dependent protein kinase II,
phosphatidylinositol-3-kinase and mitogen-activated pro-
tein kinase (Fig. 3B). The net effect of these pathways
is phosphorylation of postsynaptic receptors, recruitment
of new receptors, and the expression of novel genes.
These in turn result in altered synaptic and cellular
responses to inputs, and are believed to underlie, at least
in part, the hypersensitivity seen in different pain states
(Latremoliere and Woolf, 2009; Sandku
¨hler, 2009).
260 S. J. West et al. / Neuroscience 300 (2015) 254–275
Interneurons in the DH
The DH contains numerous interneurons, exceeding 95%
in each lamina (Todd et al., 2000; Spike et al., 2003;
Polgar et al., 2004), which is indicative of their important
role in DH processing. Understanding the function of
these neurons and how they relate to pain perception will
help us to recognize how this system breaks down in var-
ious chronic pain states. Here we review the physiological
and anatomical properties of this diverse collection of
cells.
Interneurons within the DH are often sub-divided into
morphological subtypes. Lamina I neurons have been
divided into four morphological subtypes: fusiform,
flattened, multipolar and pyramidal cells (Lima and
Coimbra, 1983, 1986). In lamina II a similar scheme
exists, consisting of central, islet, radial and vertical cells
Fig. 3. Activity-dependent plasticity in the dorsal horn. (A) An overview of the three historical accounts of activity-dependent plasticity: Wind up
describes an activity-dependent increase in WDR neuron output with continuous input, spinal LTP was identified as increased dorsal horn field
potentials following C-fiber input, central sensitization was initially described as changes in flexor motor neuron output following prolonged noxious
or C-fiber stimulation. (B) overview of the molecular mechanisms involved in activity-dependent plasticity within the dorsal horn. AMPA and
neuropeptide (GPCR receptors, 7-TM) receptors depolarize the post-synaptic neurons sufficiently to release the NMDA receptor of its magnesium
block, allowing calcium ions to enter the cell. Calcium activates a number of down-stream pathways (including protein kinase C and Calcium/
Calmodulin-dependent protein kinase II), which leads to altered synaptic and cellular responses to inputs. LTP, long-term potentiation; AMPA,
alpha-amino3-hydroxy-5-methyl-4-isoxazoleproprionic acid; NMDA, N-methyl-
D
-aspartate receptor; 7-TM, 7-trans membrance G-protein-coupled
receptor; DH, dorsal horn.
S. J. West et al. / Neuroscience 300 (2015) 254–275 261
(Grudt and Perl, 2002). However, a significant proportion
of cells continue to remain unclassified (Yasaka et al.,
2010). Fewer studies have looked at neurons in deeper
laminae, however one scheme used to divide lamina III–
V neurons into two subtypes based on axonal branching
has been proposed (Schneider, 1992).
Lamina II has been interrogated in detail by pain
researchers due to its important role in pain
transmission. Recent work has highlighted the
distribution of excitatory and inhibitory neurons of
distinct morphologies within lamina II (Heinke et al.,
2004; Punnakkal et al., 2014). Approximately 60% of
GABAergic neurons were islet cells, 5% vertical cells,
and the rest remained unclassified (Heinke et al., 2004).
An even distribution of glutamatergic neurons displayed
vertical, central, radial and unclassified cell morphologies,
but no excitatory neurons displayed islet morphology
(Punnakkal et al., 2014). These results suggest that neu-
ronal morphology is at least partially related to neuronal
phenotype and function.
Approximately 25%, 30% and 40% of neurons in
laminae I, II and III/IV are GABAergic (Todd and
Sullivan, 1990; Polgar et al., 2003). Recent work showed
the importance of dynorphin-expressing inhibitory
inteneurons in the DH for gating mechanical pain.
Ablation of the dynorphin lineage of GABAergic interneu-
rons resulted in spontaneous mechanical hypersensitivity
(Duan et al., 2014). Furthermore, this study demonstrated
the function of somatostain-positive glutamatergic neu-
rons in mechanical pain, by demonstrating that loss of this
neuronal subtype resulted in loss of noxious mechanical
sensations (Duan et al., 2014).
Another recent study demonstrated the importance of
glutamatergic DH neurons in mechanical and itch
sensation. Ablation of a subset of excitatory DH neurons
expressing the testicular orphan nuclear receptor 4
resulted in the complete absence of mechanical pain and
itch behavioral measures, as well as loss of a formalin test
response (Wang et al., 2013). Interestingly, a rewiring of
primary afferent C-fiber input to the medial inner lamina
2 region of the DH was seen, where IB4-positive afferents
were reduced and replaced with substance P afferent
arborizations, suggesting the loss of these excitatory DH
neurons led to altered input and circuitry.
Glutamate uncaging within DH slice preparations has
been used to understand properties of inhibitory and
excitatory DH neurons (Grudt and Perl, 2002; Kato
et al., 2007, 2009, 2013). One consistent finding was
the localized input of inhibitory interneurons versus the
diffuse input of excitatory interneurons. The extent of den-
drites in different planes of orientation was found to be
predictive of the degree of excitatory input in that plane
across all laminae. This was shown in the dorsoventral
axis to result in translaminar inputs (Kato et al., 2007,
2009, 2013), which bears relevance for the processing
of information from deep to superficial laminae, and vice
versa.
Projection neurons in the DH
Neurons projecting from the DH carry somatosensory
signals in ascending tracts to specific brain centers such
as the midbrain, cortical structures and the thalamus
(Todd, 2002; D’Mello and Dickenson, 2008). Many of
these neurons express the neurokinin 1 (NK-1) receptor;
the highest concentration of which is found in lamina I
(Nakaya et al., 1994; Todd, 2002).
Pathways through which second order neurons can
transmit pain-related signals from the spinal cord include
the spinothalamic, spinomedullary and spinobulbar
tracts (Dostrovsky and Craig, 2005). Most supraspinal
projections from the spinal cord originate either in lamina
I, and III–VI, with very few projections from lamina II in
lumbar DH (Todd, 2010). Projections from the spinal cord
innervate different supra-spinal structures including the
parabrachial area, nucleus of the solitary tract, the PAG,
caudal ventrolateral medulla (CVLM), and thalamic nuclei,
which are believed to code for specific dimensions of the
pain experience. In particular, the parabrachial area has
been linked to the affective and autonomic components
of pain, owing to its projections to forebrain structures
such as the amygdala and hypothalamus, and the thala-
mus has been associated with the sensory-
discriminative components of pain due to its projection
to primary and secondary somatosensory cortical regions.
The PAG and the CVLM engage other brainstem regions
(notably the rostral ventromedial medulla (RVM) and
nucleus raphe magnus), and can influence descending
controls that engage DH circuits (see
Section Descending controls) (Dostrovsky and Craig,
2005; Tracey and Mantyh, 2007; Todd, 2010).
Within the DH of the spinal cord there are two main
groups of cells involved in pain processing, as defined
by their receptive field characteristics: nociceptive
specific (NS) and wide dynamic range (WDR) neurons.
While NS neurons respond on the whole to high-intensity
input, WDR neurons respond to a much broader range
of stimuli and code stimulus intensity and wind up
(Dubner et al., 1989; Simone et al., 1991; Dougherty and
Willis, 1992). A recent study by Sikandar and colleagues
clearly demonstrated using lamina V WDR neuron record-
ings along with human quantitative sensory testing and
electroencephalography, that the response characteristics
of these DH neurons parallel psychophysical responses in
humans (Sikandar et al., 2013).
DH circuitry
A set of studies using paired recordings revealed a
number of connectivity motifs within the DH (Rexed,
1952; Lu and Perl, 2003, 2005). Approximately 10% of
randomly selected superficial DH interneurons showed
connectivity, indicating selective connectivity motifs within
this region. Lu et al. have proposed two canonical circuits:
an inhibitory connection between lamina II islet interneu-
rons and central interneurons (both which received
C-fiber inputs) (Lu and Perl, 2003)(Fig. 4A); and an
excitatory connectivity motif between central neurons
(receiving C-fiber input) to vertical interneurons (receiving
Adfiber input), and vertical interneurons to putative
lamina I projection neurons (which received C-fiber
inputs) (Lu and Perl, 2005)(Fig. 4B).
Yasaka et al. revealed a diverse set of inputs to all
morphologically defined lamina II interneurons (Yasaka
262 S. J. West et al. / Neuroscience 300 (2015) 254–275
et al., 2007). Consistent with the paired recording studies,
islet and central cells showed exclusively monosynaptic
C-fiber input only; however, vertical and radial interneu-
rons received monosynaptic inputs from both C and Ad
fibers. Furthermore, islet interneurons showed inhibitory
synaptic inputs following Adfiber stimulation, and central,
vertical and radial cells all showed inhibitory synaptic
input following both C and Adfiber stimulation (Yasaka
et al., 2007). These data suggest that many more connec-
tivity patterns exist within the DH.
Circuitry involved in Abprimary afferent processing
has been revealed in recent work. Miraucourt et al.
demonstrated that, following removal of glycine
inhibition in the DH, NS lamina I neurons could respond
synaptically to innocuous Abfiber input, and was
suggested to occur through the protein kinase C gamma
(PKC
c
)-expressing excitatory interneuron (Miraucourt
et al., 2007, 2009). These results suggest that PKC
c
interneurons are usually inhibited from activating lamina
I NS neurons through a gylcinergic inhibitory pathway.
This work was extended by Lu et al., who
demonstrated a polysynaptic pathway between Abinput
and lamina I projection neurons, via a bridge of
excitatory interneurons, including PKC
c
positive, central
and vertical neurons (Lu et al., 2013). The PKC
c
interneu-
ron, which received monosynaptic Abinput, was inhibited
by a glycinergic interneuron, which also received Abinput,
thus producing a feedforward inhibitory circuit (Lu et al.,
2013). Further experiments have highlighted the parallel
nature of this circuit, by showing that the larger somato-
statin subset of excitatory neurons present in lamina II
forms a network that transfers Abinput through lamina
II to lamina I, which included lamina II cells that were
PKC
c
positive (present on lamina II/III border), and other
cells with central and vertical morphologies (Duan et al.,
2014; Yasaka et al., 2014). Interestingly, monosynaptic
Abinput was present on both somatostain/PKC
c
-positive
interneurons on the lamina II/III border and on the
somatostatin-positive vertical neurons, indicating two
pathways for Abinput to reach lamina I. Furthermore, at
each successive step through the circuitry, inhibitory
interneurons, some of which received Abinput, formed
a feedforward synaptic gate with the somatostatin-
positive neurons (Duan et al., 2014)(Fig. 4C). This cir-
cuitry was shown to relate functionally to mechanical allo-
dynia, as ablation of somatostatin-positive neurons
resulted in loss of mechanical pain sensation in these
mice, whereas dynorphin cell ablation resulted in sponta-
neous mechanical hypersensitivity (Duan et al., 2014).
Finally, a recent study has shown that lamina I
projection neurons themselves are directly subject to
low-threshold, primary afferent-driven inhibitory input
(Luz et al., 2014)(Fig. 4C). These inhibitory inputs were
found to temporally precede high-threshold Adfiber
inputs, and acted to shunt excitatory post-synaptic poten-
tials generated from these fibers.
Peripheral nerve injury: effects on the DH
It is well acknowledged that central plasticity plays a role
in the hypersensitivity seen following peripheral nerve
injury (Dubner and Ruda, 1992; Woolf and Salter, 2000;
Fig. 4. Dorsal horn circuitry. A number of consistent connectivity motifs between interneurons identified within the superficial dorsal horn. (A)
Inhibitory islet cells were found to consistently contact central interneurons in lamina II, and both received C-fiber input. (B) Excitatory connectivity
motif connecting lamina II excitatory central cells, to excitatory vertical cells leading to lamina I projection neurons, all received C-fiber input, and
vertical cells received Adinput. (C) Two major pathways that gate Abinput to lamina I projection neurons. (1). Excitatory vertical cells receive Ab
input, which is gated by inhibitory interneurons. Abinput can also directly inhibit lamina I projection neurons. (2). A complex excitatory circuit from
protein kinase C/somatostatin cells to central to vertical cells, that can stimulate lamina I projection neurons is gated by dynorphin-expressing
inhibitory interneurons.
S. J. West et al. / Neuroscience 300 (2015) 254–275 263
Ji et al., 2003). The present challenge in pain research is
establishing which molecules, cells and circuits are
affected by neuropathic pain states to establish new ther-
apeutic strategies. Here we review evidence relating to
DH plasticity and the development of hypersensitivity fol-
lowing nerve injury.
Altered excitatory and inhibitory processing following
neuropathy. As previously reviewed, primary afferents
display spontaneous input after peripheral nerve injury,
and this is believed to contribute to activity-dependent
plasticity between primary afferent terminals and DH
neurons (Drdla and Sandku
¨hler, 2009; Latremoliere and
Woolf, 2009; Sandku
¨hler, 2009). Another important
observation is the reduction in GABA and glutamic acid
decarboxylase (GAD) following traumatic nerve injury
(Castro-Lopes et al., 1993; Moore et al., 2002), which
results in a reduction in inhibitory tone.
Previous work has investigated the loss of inhibitory
tone, with electrophysiological studies of lamina II
neurons showing reduced inhibitory post-synaptic
potentials in magnitude, incidence and duration following
partial peripheral nerve injuries (Moore et al., 2002;
Scholz et al., 2005)(Fig. 5A). Analysis of the inhibitory
post-synaptic potentials (Moore et al., 2002) suggested
a reduction in GABA release, and other studies have
found a reduction in GABA secretion from stimulated
spinal cord sections (Lever et al., 2003), and significant
reductions in GABA immunoreactivity within the DH
(Castro-Lopes et al., 1993; Ibuki et al., 1997). This evi-
dence suggests a loss of GABAergic neurons, which
has been demonstrated in some studies (Castro-Lopes
et al., 1993; Scholz et al., 2005), and some evidence sug-
gests that apoptosis drives the loss of inhibitory interneu-
rons (Moore et al., 2002; Scholz et al., 2005) following
peripheral nerve injury (Fig. 5B).
Recent work has highlighted the loss of putative
inhibitory terminals following peripheral nerve injury, that
mirrors the temporal and spatial alterations in presumed
IB4 + primary afferent synapses (Lorenzo et al., 2014).
GAD65 immunoreactive puncta, representing terminals
of inhibitory interneurons, were found to be significantly
reduced 3–4 weeks after nerve injury in the superficial
DH (Fig. 5B). Whether this reflects a dying back of pri-
mary afferent fibers in this region, and the subsequent
loss of pre-synaptic inhibitory control on these terminals
or represents reorganization of other circuits within the
DH remains unknown.
The exact nature of the loss of inhibitory tone remains
controversial, as other studies have suggested no loss of
GABAergic boutons or GABA
A
b3 subunits in denervated
DH following nerve injury (Polgar and Todd, 2008), and no
loss in the proportion of GABAergic neurons (Polgar et al.,
2003), nor the total number of neurons in lamina I-III
Fig. 5. Loss of inhibitory tone after peripheral nerve injury. (A) Electrophysiological studies have demonstrated a reduction in excitatory and
inhibitory post-synaptic potentials (EPSPs and IPSPs respectively). These have been related to reduced transmitter release, and with reductions in
GAD and GABA levels in inhibitory terminals. (B) Some studies have indicated a loss of inhibitory terminals as well as inhibitory interneurons in the
dorsal horn after nerve injury.
264 S. J. West et al. / Neuroscience 300 (2015) 254–275
(Polgar et al., 2004, 2005). Furthermore, in these studies
markers for apoptosis were not co-localized with neuronal
cells, but instead found in microglia (Polgar et al., 2005).
A different mechanism that may account for inhibitory
tone reduction without the frank loss of inhibitory
interneurons was recently described. Measurement of
miniature excitatory post-synaptic potentials were shown
to be reduced in neuropathic animals, and measurement
of pair-pulse ratios suggested a reduction in primary
afferent transmitter release may explain the decline in
inhibitory tone (Leitner et al., 2013). This study did not find
any changes in the density or morphology of dendritic
spines or the number of excitatory synapses on inhibitory
neurons, consistent with a pre-synaptic effect on primary
afferent neurotransmitter release (Fig. 5A).
A further promising mechanism that may account for
reduced inhibitory tone without a loss of GABAergic
neurons relates to the down-regulation of KCC2. This
ion pump, along with NKCC1, regulates the chloride
concentrations within neurons, and thus regulates
neuronal responses to GABA
A
activation. Following
nerve injury, a reduction in KCC2 leads to impaired
chloride homeostasis resulting in reduced inhibitory tone
(Coull et al., 2003). This mechanism has been suggested
to be driven by the microglial response seen after injury,
via BDNF signaling (Coull et al., 2005). Further work
has demonstrated that enhancing KCC2 activity after
nerve injury can restore normal response characteristics
of spinothalamic tract NS neurons (Lavertu et al., 2014),
and normalize behavioral hypersensitivity seen following
peripheral nerve injury (Gagnon et al., 2013).
Altered structure and function of projection
neurons. Anatomical evidence has revealed alterations
in dendritic spines on deep DH WDR projection
neurons, as determined by anatomical characteristics,
following traumatic nerve injury, and these effects
correlated with electrophysiological alterations of WDR
neurons showing hyper-excitability to a range of stimuli
(Tan and Zhao, 2011)Fig. 6A). Crucially, administration
of a drug that selectively inhibited Ras-related C3 botuli-
num toxin substrate 1 (Rac1), a key molecule involved
in dendritic spine plasticity, resulted in attenuation of
behavior sensitivity, dendritic spine remodeling, and
WDR neuronal hyper-excitability. Further work from this
group has shown a similar response to WDR neurons to
diabetic-induced neuropathy (Tan et al., 2012). Thus,
dendritic spine remodeling on WDR neurons may con-
tribute in general to the neuropathic pain phenotype.
Electrophysiological changes to WDR neurons that have
been observed include increased spontaneous activity,
enlargement of receptive field size, and reduced C-fiber
threshold with increased C-fiber responses (Liu et al.,
2011; Tan and Zhao, 2011).
Lamina I projection neurons appear to respond to
innocuous inputs after nerve injury, which may
contribute to mechanical allodynia in neuropathic pain
states. Using extracellular recording, lamina I projection
neurons of the spinao-paarabrachial pathway have been
shown to generate spontaneous activity, respond to
innocuous inputs, and produce an amplified response to
nociceptive inputs following peripheral nerve constriction
(Keller et al., 2007)(Fig. 6B). Further work has indicated
the ability for C-fiber input onto lamina I projection neu-
rons to induce LTP, even at physiologically active levels
in the intact cord (Ikeda et al., 2006), which can sensitize
other inputs (Fenselau et al., 2011), implying that sponta-
neous C-fiber input (Liu et al., 2000) may sensitize this
projection pathway. Consistent with this idea, ablation of
NK-1 receptor-expressing cells in the DH, many of which
are lamina 1 projection neurons (Todd et al., 2000),
reduces the mechanical hypersensitivity seen following
nerve injury (Nichols et al., 1999). Combined with the
rewiring of specific elements within the DH to gate innocu-
ous inputs to these projection neurons (see Section DH
rewiring below), the lamina I projection pathway appears
to play an important role in the production of neuropathic
symptoms.
DH rewiring. Many of the circuits previously reviewed
have been shown to be affected by nerve injury. Lu
et al. demonstrated a circuit between PKC
c
excitatory
interneurons in lamina II inner and glycinergic neurons
in lamina III, which produced a feed forward inhibitory
gate of Abinput, preventing Abpotentials reaching
lamina I projection neurons (Lu et al., 2013). Following
peripheral nerve injury, Lu et al. showed that the glyciner-
gic inhibitory interneuron connection to the PKC
c
excita-
tory interneuron became weaker, which thus allowed the
PKC
c
interneuron to excite lamina I projection neurons
with Abinput. Therefore, under neuropathic pain condi-
tions, the gate had been removed, allowing innocuous
input to gain access to nociceptive-specific projection
pathways (Lu et al., 2013)(Fig. 7).
Elaborating on this work, Duan et al. showed that
ablation of somatostatin-positive neuron results in
complete loss of mechanical hypersensitivity following
peripheral nerve injury, with no effect on heat
hyperalgesia (Duan et al., 2014). Ablation of the
dynorphin-positive inhibitory neurons did not change the
mechanical hypersensitivity after nerve injury (Fig. 7).
These results indicate that Abfibers can gain access
to lamina I projection neurons after nerve injury, and
that this may be mediated through two parallel
pathways. While it has been shown that the feedforward
inhibition has been reduced in these studies, allowing
Abinput to reach lamina I, these studies did not explore
why inhibition is reduced. This likely reflects many of the
changes to inhibitory tone previously reviewed (see
Section Altered excitatory and inhibitory processing),
however another possibility could relate to the observed
loss of C-fiber terminals (Bailey and Ribeiro-Da-Silva,
2006), which may interfere with these circuits.
DESCENDING CONTROLS FROM THE BRAIN
TO THE SPINAL CORD
Descending pathways from the brainstem play a pivotal
role in the modulation of pain processing (Fig. 8). Inputs
driving circuits in the amygdala, hypothalamus, frontal
lobe and anterior cingulate cortex activate areas of the
PAG and this in turn feeds into the RVM. The RVM also
S. J. West et al. / Neuroscience 300 (2015) 254–275 265
has reciprocal innervations from the dorsal lateral pontine
tegmentum, which sits rostrally to this and contains
clusters of noradrenergic nuclei of the A5, A7 and the
locus coeruleus (LC) cell groups. From the RVM
neurons project bilaterally via the dorsolateral and
ventrolateral funiculi to the DH, giving rise to descending
facilitatory and inhibitory pathways (Basbaum and
Fields, 1979). Such pathways have been identified as
key players in the complex synergistic pain processing
system, where the release of 5-hydroxytryptamine (5HT)
from the RVM and noradrenaline (NA) from the LC play
overriding pro-nociceptive and anti-nociceptive roles
respectively (Heinricher et al., 1992; Bannister et al.,
2011). In particular, the descending control system is
thought to be activated by a spinal-bulbo-spinal loop
whereby activated NK-1 expressing lamina I/III projection
Fig. 6. Deep and superficial neuron structure and function changes. (A) Lamina V neurons display altered dendritic spine dynamics, with increased
average length and spine head. Responses to brush, pressure and pinch are amplified, with longer after-discharges. (B) Lamina I neurons show
novel responses to brush inputs after injury, and exaggerated responses to noxious mechanical pinch stimuli. These neurons also show higher
levels of spontaneous activity.
Fig. 7. Abfibers gain access to lamina I projection neurons after nerve injury. Both identified pathways that can potentially lead Abinput to
projection neurons are affected. (1). Inhibitory vertical cells show reduced inhibitory drive to excitatory vertical cells, removing the inhibitory gate and
allowing Abstimulation to evoke activity in lamina I neurons. (2). The dynorphin inhibitory interneurons show reduced inhibitory tone, resulting in
increased access of Abinput to excitatory neurons in lamina II, resulting in increased lamina I projection neuron activation.
266 S. J. West et al. / Neuroscience 300 (2015) 254–275
neurons send signals to the parabrachial area and onto
the limbic system (Suzuki et al., 2002). A small population
of dopaminergic neurons in the PAG mediates anti-
nociception via participation in supraspinal nociceptive
responses after opiates; this dopaminergic system is
another network within the PAG involved in opiate-
induced anti-nociception (Flores et al., 2004). The
inhibitory PAG-RVM-DH descending pathway is well
characterized but the RVM is also involved in descending
facilitation of nociceptive processing (Porreca et al.,
2002).
The monoamine neurotransmitters NA and 5HT are
released from neurons in the brain and the peripheral
nervous system and are known to have a crucial
influence on mood and behavior. They are also
proposed to play complex, often overlapping,
modulatory roles in pain signaling.
5HT
Neurons of the nucleus raphe magnus, one of a group of
nuclei that comprise the RVM, are the principal source of
5HT release in the DH. DH-projecting 5HT-
immunoreactive neurons are also found within the
nucleus paragigantocellularis and ventral portions of the
nucleus gigantocellularis (Kwiat and Basbaum, 1992).
Axons of 5HT neurons in the higher raphe nuclei project
to, and terminate in, areas of the brain including the tha-
lamic nuclei, nucleus accumbens, hypothalamus, hip-
pocampus and amygdala. From the lower raphe nuclei
axons of descending 5HT neurons terminate in the cere-
bellum’s deep nuclei and cortex as well as innervating
the superficial lamina of the DH of the spinal cord. The
influence of 5HT is widespread due in part to the remark-
able diversity and dense population of 5HT receptors that
are differentially expressed on functionally distinct target
neurons throughout the nervous system (Hoyer et al.,
2002). In the DH of the spinal cord 5HT exerts complex
modulatory effects on nociceptive transmission (Millan,
2002). Complex because variant 5HT receptor subtypes
are coupled to contrasting intracellular mechanisms so
ultimately 5HT plays a bi-directional role in pain process-
ing by acting at specific receptor subtypes (Zeitz et al.,
2002; Suzuki et al., 2004). Anti-nociception caused by
opioid injection into the RVM is blocked by application of
a spinal 5HT7 receptor antagonist (Dogrul et al., 2009)
Fig. 8. Descending control from RVM and LC exert powerful control over dorsal horn processing and transmission. Descending inhibition is
predominantly noradrenergic as
a
2 adrenoceptors are activated, and involves opioidergic mechanisms. Descending inhibition acts via activation of
the 5HT3 receptor. INSET, Peripheral nerve injury causes a loss in descending inhibition and an increase in descending excitation of dorsal horn
circuits. In the spinal nerve ligation model, activity from injured myelinated L5 afferents induces spontaneous activity in intact L4 C-fibers. Enhanced
activity leads to homosynaptic and heterosynaptic potentiation (recruiting adjacent synapses). ACC, anterior cingulated cortex; Ce, central nucleus
of the amygdala; Hy, ventral medial nucleus of the hypothalamus; In, insula cortex; LC, locus coeruleus; PAG, periaqueductal grey; RVM, rostral
ventromedial medulla (Modified from Tracey and Mantyh, 2007).
S. J. West et al. / Neuroscience 300 (2015) 254–275 267
and agonist activation of post-synaptic 5HT1
A
receptors
inhibits excitability of spinothalamic neurons and excita-
tory interneuron’s (Benarroch, 2008). In the DH of the
spinal cord descending facilitatory influences are primarily
5HT3 receptor driven (Millan, 2002). That 5HT3 receptors
are excitatory is further confirmed by studies investigating
NK-1 expressing cells. The RVM contains high levels of
the NK-1 receptor and its endogenous ligand, substance
P. NK-1 expressing neurons in the superficial DH are
thought to drive descending facilitation via the brain stem
and enhance DH neuronal activity via 5HT3 receptors.
They form part of a major ascending pathway that sends
projections to the parabrachial nuclei and then onto the
hypothalamus and amygdala. When NK-1 expressing
cells are selectively ablated, 5HT3 receptor antagonism
no longer reduces the responses of deep DH neurons to
noxious natural stimulation in the complete freund’s adju-
vant inflammatory pain model (Suzuki et al., 2002).
NA
Neurons in the LC synthesize NA and from here axons of
neurons project to both sides of the brain ultimately
innervating the brain stem, cerebellum, hypothalamus,
amygdala, neocortex, thalamic nuclei and the spinal
cord (Roudet et al., 1994). In the brainstem the RVM
receives a dense noradrenergic projection arising from
the dorsolateral pontine cell groups in the LC A5 and A7
nuclei (Tanaka et al., 1996), and these provide the major
source of noradrenergic projections to the DH (Hentall
et al., 2003). In the spinal cord there are three main
classes of adrenergic receptor comprising 1, 2 and b
adrenoceptors (Millan, 2002). Noradrenergic projections
to the spinal cord are chiefly inhibitory as they exert
anti-nociceptive influences via spinal 2 adrenoceptors
(Byrum et al., 1984; Benarroch, 2008). Activity at these
receptors is enhanced during inflammation and applica-
tion of spinal 2 agonists reduces the response of deep
DH neurons in rats with acute and persistent pain
(Stanfa and Dickenson, 1994). In healthy humans
intrathecal, but not systemic, 2 agonists reduce thermal
and capsaicin-evoked pain (Eisenach, 1998).
Conversely spinal 1 adrenoceptors exert pro-nociceptive
influences on activation (Pertovaara, 2013) although this
story is more complex since activating post-synaptic 1
adrenoceptors may also contribute to anti-nociception
by causing the release of inhibitory neurotransmitters
(Benarroch, 2008).
Plasticity in descending controls following peripheral
nerve injury
Plasticity of descending modulation from the brainstem is
involved in central sensitization in abnormal pain states
and acts alongside the spinal events (Fig. 8).
Descending pain facilitation can promote a chronic pain
state that persists long after the initial lesion has healed.
Prolonged noxious stimulation also causes pronounced
changes in the activity of nociceptive modulatory
neurons in the RVM (Morgan and Fields, 1994), which
is implicated in the development of central sensitization
and secondary hyperalgesia following noxious stimulation
(Urban and Gebhart, 1997, 1999). Descending facilitatory
influences are also shown enhanced in neuropathy, pos-
sibly contributing to allodynia and hyperalgesia (Millan,
2002; Suzuki et al., 2004) and are purported to be
involved in the maintenance of nerve-injury-induced pain
(Burgess et al., 2002; Vera-Portocarrero et al., 2006).
DH neuronal responses to noxious stimuli are reduced
in normal injury-free animals after lidocaine injection into
the RVM and a greater effect is observed in nerve injured
animals. Here, descending modulation from the RVM
influences neuronal responses to non-noxious tactile
stimulation suggesting a possible mechanism for
mechanical allodynia (Bee and Dickenson, 2007). Nerve
injury is associated with enhanced descending 5HT exci-
tations, and specifically the 5HT3 receptor is implicated
(Porreca et al., 2002; Suzuki et al., 2004; Ossipov et al.,
2010). The contribution of descending serotonergic facili-
tation in neuropathic pain states is further confirmed by
experiments showing a significantly enhanced ability of
5HT3 antagonist ondansetron to suppress spinal
responses to mechanical punctate stimulation in spinal
nerve ligated animals (Suzuki et al., 2005). In osteoar-
thritic rats there is evidence of an adaptive change in
the excitatory serotonergic drive (Rahman et al., 2009).
Many drugs that have efficacy in neuropathic pain take
advantage of descending control systems. The state-
dependency of pregabalin and gabapentin is reliant on
upregulation of 5HT3 receptor mediated descending
facilitation (Suzuki et al., 2005). Peripheral nerve injury
can cause a loss of descending inhibition (Leong et al.,
2011). While 5HT actions are enhanced in some chronic
pain states, descending inhibitory noradrenergic pathways
may also undergo plastic changes and become down reg-
ulated. In nerve-injured animals, the efficacy of atipame-
zole, an 2 adrenoceptor antagonist, is shown reduced in
specific sensory modalities. In contrast, low-intensity
mechanical stimulation in sham-operated rats was shown
significantly enhanced (Rahman et al., 2008). These
results suggested a lack or suppression of descending
inhibition of DH neurons via spinal 2 adrenoceptors for
low-intensity mechanical stimuli. A loss of noradrenergic
influences was proposed and the authors suggested that
other descending modulatory pathways such as the sero-
tonergic system may explain lack of effect of atipamezole
on neuronal responses as reported by net serotonergic
facilitation in animals with peripheral nerve injury (Suzuki
et al., 2004). Enhanced efficacy of spinally administered
2 adrenoceptor agonists (Stanfa and Dickenson, 1994;
Mansikka and Pertovaara, 1995; Tsuruoka et al., 2003),
increased noradrenergic innervation to DH (Ma and
Eisenach, 2003) and upregulation of spinal 2 adrenocep-
tors and increased spinal cord NA content (Satoh and
Omote, 1996) all indicate profound alrerations to descend-
ing adrenergic inputs to the DH in neuropathic pain.
A recent paper addressed the clinically relevant
question of why only a minority of patients develop
chronic neuropathic pain following the same nerve
lesion (De Felice et al., 2011). The suggestion from the
authors that a descending noradrenergic mechanism
could actually protect from pain put brain to spinal cord
268 S. J. West et al. / Neuroscience 300 (2015) 254–275
pathways once again at the forefront of pain modulation,
and of great interest was the proposal that nerve injury-
induced pain could ultimately depend on descending
drives. The endogenous analgesic system, relating to
the engagement of inhibitory 2 adrenoceptors, is indicated
in shaping the spatial and temporal expression of the neu-
ropathic pain phenotype after nerve injury (Hughes et al.,
2013).
LINKING MECHANISMS TO PATIENTS
This review has described the physiology and anatomy of
the DH, and how these properties are modified following
injury to peripheral input. Deciphering how these
mechanisms relate to real patients, and how can we
determine which mechanisms are established within
different populations presents a challenge in bringing
these insights into the clinic. It is clear that patients will
have a combination of the changes that we have
presented in this review in that they will have damaged
and altered afferent fibers, as can be gauged by QST
as well as excitability changes in the spinal cord and
consequently in descending controls. For example
changes in spinal excitability leading to wind up and
central hypersensitivity is the most plausible explanation
for dynamic mechanical allodynia and increased levels
of pain despite the sensory loss that peripheral
neuropathy will involve. Finally there is good human
evidence for changes in descending controls as has
been shown by quantifying conditioned pain modulation
in neuropathic patients.
Neuropathic mechanisms within patients
In order to translate these finding to the clinic, a key
challenge to meet is establishing which of these
mechanisms are present within patients. Some progress
has been made in establishing subgroups of patients
using sensory profiles (Baron et al., 2012). Given the
huge difficulties in defining mechanisms in patients, the
concept that the sensory phenotype of the patient has
to reflect the particular underlying events represents a
feasible alternative approach to individualizing medicine.
Since the particular mechanisms that are driving the pain
phenotype can be modulated by drugs which target those
mechanisms, we can start to investigate whether pheno-
types are predictive of therapeutic success.
Stratifying patients into select groupings may help with
clinical trail design, help to prove efficacy for new drugs
that are only beneficial for a subset of patients, and help
to deliver personalized medicine to sufferers of chronic
pain. A recent example of the value of this approach is
a study on oxcarbazine in patients with neuropathic pain
(Demant et al., 2014). In a heterogenous population, the
drug was ineffective but in the subgroup of patients with
‘‘irritable nociceptors’’, where there is hypersensitivity
and remaining innervation, the drug was effective.
Descending modulation is increasingly being recognized
as clinically relevant in patient populations. Recently it
has been shown that conditioned pain modulation (a psy-
chophysical paradigm which tests the integrity of
descending pain modulatory pathways) was predictive
of the response to Duloxetine in painful diabetic neuropa-
thy (Yarnitsky et al., 2012).
In order to divide patients into sub-populations,
biomarkers of specific mechanisms are required. In
parallel with sensory profiling of patients, previous work
has focussed on systemic (cerebrospinal fluid and
blood), as well as local (skin and nerve) targets for both
biochemical, transcriptional and histological markers
(Uc¸ eyler and Sommer, 2012; Widerstro
¨m-Noga et al.,
2013). Recent suggestions include using microRNA pro-
files within the blood of patients, which may be used to
discern specific mechanisms (Andersen et al., 2014).
Specific disease states driving neuropathic pain may pre-
sent with specific mechanisms, thus the disease presen-
tation itself may be used to help offer the appropriate
treatment. And finally, the use of combination therapy
when treating neuropathic pain may be a useful strategy,
as the use of multiple pharmacological agents can target
multiple mechanisms, and often results in reduced dos-
ing, and thus a reduction in side effect profile without loss
of efficacy. Although this is intellectually appealing
enhanced efficacy of combined versus monotherapy for
neuropathic pain has not yet been firmly established
(Gilron et al., 2013; Finnerup et al., 2015).
CONCLUSION
The review has highlighted the key drivers of peripheral
neuropathic pain. This begins with a lesion to the
somatosensory nervous system that subsequently
drives plasticity within the DH, which contributes to the
altered sensations experienced by neuropathic patients.
Much progress has been made in revealing
pathophysiological mechanisms associated with
neuropathic pain. Advances in transgenic technology
and the use of optogenetics to activate particular
neuronal populations are likely to drive further advances
in understanding the changes at circuit level which drive
and then maintain neuropathic pain. The future
challenge will be to investigate the clinical relevance of
such changes and wherever possible to determine if the
primary pathophysiological mechanisms can be
recognized on an individualized patient basis.
Acknowledgments—The authors would like to acknowledge
Ryan Patel (Neuroscience Physiology & Pharmacology Dept.,
UCL) for his contribution to the production of Fig. 8. This work
has received support from the Wellcome Trust-funded London
Pain Consortium (ref. no. 083259). Corresponding author David
Bennett is a senior Wellcome Trust Clinical Scientist Fellow
(ref. no. 095698z/11/z).
REFERENCES
Adrian ED (1930) The effects of injury on mammalian nerve fibres. P
Roy Soc Lond Bio Charact 106:596–618.
Afrah AW, Fiska
˚A, Gjerstad J, Gustafsson H, Tjølsen A, Olgart L,
Stiller C-O, Hole K, Brodin E (2002) Spinal substance P release
in vivo during the induction of long-term potentiation in dorsal horn
neurons. Pain 96:49–55.
S. J. West et al. / Neuroscience 300 (2015) 254–275 269
Andersen HH, Duroux M, Gazerani P (2014) MicroRNAs as
modulators and biomarkers of inflammatory and neuropathic
pain conditions. Neurobiol Dis 71C:159–168.
Authier N, Balayssac D, Marchand F, Ling B, Zangarelli A, Descoeur
J, Coudore F, Bourinet E, Eschalier A (2009) Animal models of
chemotherapy-evoked painful peripheral neuropathies.
Neurotherapeutics 6:620–629.
Azkue JJ, Liu X-G, Zimmermann M, Sandku
¨hler J (2003) Induction of
long-term potentiation of C fibre-evoked spinal field potentials
requires recruitment of group I, but not group II/III metabotropic
glutamate receptors. Pain 106:373–379.
Bailey AL, Ribeiro-Da-Silva A (2006) Transient loss of terminals from
non-peptidergic nociceptive fibers in the substantia gelatinosa of
spinal cord following chronic constriction injury of the sciatic
nerve. NSC 138:675–690.
Bannister K, Sikandar S, Bauer CS, Dolphin AC, Porreca F,
Dickenson AH (2011) Pregabalin suppresses spinal neuronal
hyperexcitability and visceral hypersensitivity in the absence of
peripheral pathophysiology. Anesthesiology 115:144–152.
Bardoni R, Takazawa T, Tong C-K, Choudhury P, Scherrer G,
MacDermott AB (2013) Pre- and postsynaptic inhibitory control in
the spinal cord dorsal horn. Ann N Y Acad Sci 1279:90–96.
Baron R, Binder A, Wasner G (2010) Neuropathic pain: diagnosis,
pathophysiological mechanisms, and treatment. Lancet Neurol
9:807–819.
Baron R, Fo
¨rster M, Binder A (2012) Subgrouping of patients with
neuropathic pain according to pain-related sensory abnormalities:
a first step to a stratified treatment approach. Lancet Neurol
11:999–1005.
Basbaum AI, Fields HL (1979) The origin of descending pathways in
the dorsolateral funiculus of the spinal cord of the cat and rat:
Further studies on the anatomy of pain modulation – Basbaum –
2004 – The Journal of Comparative Neurology – Wiley Online
Library. J Comp Neurol 187:513–531.
Battaglia G, Rustioni A (1988) Coexistence of glutamate and
substance P in dorsal root ganglion neurons of the rat and
monkey. J Comp Neurol 277:302–312.
Bee LA, Dickenson AH (2007) Rostral ventromedial medulla control
of spinal sensory processing in normal and pathophysiological
states. NSC 147:786–793.
Beggs S, Salter MW (2013) The known knowns of microglia-neuronal
signalling in neuropathic pain. Neurosci Lett 557A:37–42.
Benarroch EE (2008) Descending monoaminergic pain modulation:
bidirectional control and clinical relevance. Neurology
71:217–221.
Burgess SE, Gardell LR, Ossipov MH, Malan TP, Vanderah TW, Lai
J, Porreca F (2002) Time-dependent descending facilitation from
the rostral ventromedial medulla maintains, but does not initiate,
neuropathic pain. J Neurosci 22:5129–5136.
Byrum CE, Stornetta R, Guyenet PG (1984) Electrophysiological
properties of spinally-projecting A5 noradrenergic neurons. Brain
Res 303:15–29.
Castro-Lopes JM, Coimbra A, Grant G, Arvidsson J (1990)
Ultrastructural changes of the central scalloped (C1) primary
afferent endings of synaptic glomeruli in the substantia gelatinosa
Rolandi of the rat after peripheral neurotomy. J Neurocytol
19:329–337.
Castro-Lopes JM, Tavares I, Coimbra A (1993) GABA decreases in
the spinal cord dorsal horn after peripheral neurectomy. Brain Res
620:287–291.
Cavanaugh DJ, Lee H, Lo L, Shields SD, Zylka MJ, Basbaum AI,
Anderson DJ (2009) Distinct subsets of unmyelinated
primary sensory fibers mediate behavioral responses to noxious
thermal and mechanical stimuli. Proc Natl Acad Sci
106:9075–9080.
Cervero F, Laird J, Garcı
´a-Nicas E (2003) Secondary hyperalgesia
and presynaptic inhibition: an update. Eur J Pain 7:345–351.
Chen JT-C, Guo D, Campanelli D, Frattini F, Mayer F, Zhou L, Kuner
R, Heppenstall PA, Knipper M, Hu J (2014) Presynaptic
GABAergic inhibition regulated by BDNF contributes to
neuropathic pain induction. Nat Commun 5:5331.
Cirillo G, Cavaliere C, Bianco MR, De Simone A, Colangelo AM,
Sellitti S, Alberghina L, Papa M (2009) Intrathecal NGF
administration reduces reactive astrocytosis and changes
neurotrophin receptors expression pattern in a rat model of
neuropathic pain. Cell Mol Neurobiol 30:51–62.
Coderre TJ, Melzack R (1992) The role of NMDA receptor-operated
calcium channels in persistent nociception after formalin-induced
tissue injury. J Neurosci 12:3671–3675.
Colangelo AM, Bianco MR, Vitagliano L, Cavaliere C, Cirillo G, De
Gioia L, Diana D, Colombo D, Redaelli C, Zaccaro L, Morelli G,
Papa M, Sarmientos P, Alberghina L, Martegani E (2008) A new
nerve growth factor-mimetic peptide active on neuropathic pain in
rats. J Neurosci 28:2698–2709.
Costigan M, Befort K, Karchewski L, Griffin RS, D’Urso D, Allchorne
A, Sitarski J, Mannion JW, Pratt RE, Woolf CJ (2002) Replicate
high-density rat genome oligonucleotide microarrays reveal
hundreds of regulated genes in the dorsal root ganglion after
peripheral nerve injury. BMC Neurosci 3:16.
Coull JAM, Beggs S, Boudreau D, Boivin D, Tsuda M, Inoue K,
Gravel C, Salter MW, De Koninck Y (2005) BDNF from microglia
causes the shift in neuronal anion gradient underlying neuropathic
pain. Nature 438:1017–1021.
Coull JAM, Boudreau D, Bachand K, Prescott SA, Nault F, Sı
´kA,De
Koninck P, De Koninck Y (2003) Trans-synaptic shift in anion
gradient in spinal lamina I neurons as a mechanism of
neuropathic pain. Nature 424:938–942.
Csillik B, Knyihar-Csillik E (1981) Regenerative synaptoneogenesis in
the mammalian spinal cord: dynamics of synaptochemical
restoration in the Rolando Substance after transganglionic
degenerative atrophy. J Neural Transm 52:303–317.
Csillik B, Knyihar-Csillik E (1982) Reversibility of microtubule
inhibitor-induced transganglionic degenerative atrophy of central
terminals of primary nociceptive neurons. NSC 7:1149–1154.
Csillik B, Schwab ME, Thoenen H (1985) Transganglionic regulation
of central terminals of dorsal root ganglion cells by nerve growth
factor (NGF). Brain Res 331:11–15.
D’Mello R, Dickenson AH (2008) Spinal cord mechanisms of pain. Br
J Anaesth 101:8–16.
De Biasi S, Rustioni A (1988) Glutamate and substance P coexist in
primary afferent terminals in the superficial laminae of spinal cord.
Proc Natl Acad Sci U S A 85:7820–7824.
De Felice M, Sanoja R, Wang R, Vera-Portocarrero L, Oyarzo J, King
T, Ossipov MH, Vanderah TW, Lai J, Dussor GO, Fields HL, Price
TJ, Porreca F (2011) Engagement of descending inhibition from
the rostral ventromedial medulla protects against chronic
neuropathic pain. Pain 152(12):2701–2709.
Demant DT, Lund K, Vollert J, Maier C, Segerdahl M, Finnerup NB,
Jensen TS, Sindrup SH (2014) The effect of oxcarbazepine in
peripheral neuropathic pain depends on pain phenotype: a
randomised, double-blind, placebo-controlled phenotype-
stratified study. Pain 155(11):2263–2273.
Derjean D, Bertrand S, Le Masson G, Landry M, Morisset V, Nagy F
(2003) Dynamic balance of metabotropic inputs causes dorsal
horn neurons to switch functional states. Nat Neurosci 6:274–281.
Dickenson AH (1990) A cure for wind up: NMDA receptor antagonists
as potential analgesics. Trends Pharmacol Sci 11:307–309.
Dickenson AH (1995) Spinal cord pharmacology of pain. Br J Anaesth
75:193–200. Available at: http://www.ncbi.nlm.nih.gov/pubmed/
7577253.
Dickenson AH, Sullivan AF (1987) Evidence for a role of the NMDA
receptor in the frequency dependent potentiation of deep rat
dorsal horn nociceptive neurones following C fibre stimulation.
Neuropharmacology 26:1235–1238.
Dogrul A, Ossipov MH, Porreca F (2009) Differential mediation of
descending pain facilitation and inhibition by spinal 5HT-3 and
5HT-7 receptors. Brain Res 1280:52–59.
Dostrovsky JO, Craig AD (2005) Ascending projection systems. In:
The textbook of pain, 5 ed. (Mcmahon SB, Klotzenburg M, eds),
pp 187–203. Churchill Livingstone.
Dougherty PM, Willis WD (1992) Enhanced responses of
spinothalamic tract neurons to excitatory amino acids
270 S. J. West et al. / Neuroscience 300 (2015) 254–275
accompany capsaicin-induced sensitization in the monkey. J
Neurosci 12:883–894.
Drdla R, Sandku
¨hler J (2009) Long-term potentiation in superficial
spinal dorsal horn: a pain amplifier. Synaptic Plast Pain:201–218.
Duan B, Cheng L, Bourane S, Britz O, Padilla C, Garcia-Campmany
L, Krashes M, Knowlton W, Velasquez T, Ren X, Ross SE, Lowell
BB, Wang Y, Goulding M, Ma Q (2014) Identification of spinal
circuits transmitting and gating mechanical pain. Cell
159:1417–1432.
Dubner R, Kenshalo DR, Maixner W, Bushnell MC, Oliveras JL
(1989) The correlation of monkey medullary dorsal horn neuronal
activity and the perceived intensity of noxious heat stimuli. J
Neurophysiol 62:450–457.
Dubner R, Ruda MA (1992) Activity-dependent neuronal plasticity
following tissue injury and inflammation. Trends Neurosci
15:96–103.
Dworkin RH, O’Connor AB, Audette J, Baron R, Gourlay GK,
Haanpa
¨a
¨ML, Kent JL, Krane EJ, LeBel AA, Levy RM (2010)
Recommendations for the pharmacological management of
neuropathic pain: an overview and literature update. Mayo Clin
Proc 85:S3–S14.
Eccles JC, Schmidt R, Willis WD (1963) Pharmacological studies on
presynaptic inhibition. J Physiol 168:500–530.
Eippert F, Finsterbusch J, Bingel U, Bu
¨chel C (2009) Direct evidence
for spinal cord involvement in placebo analgesia. Science
326:404.
Eisenach JC (1998) Additives for epidural analgesia for labor: why
bother? Reg Anesth Pain Med 23:531–532.
Fenselau H, Heinke B, Sandku
¨hler J (2011) Heterosynaptic long-term
potentiation at GABAergic synapses of spinal lamina I neurons. J
Neurosci 31:17383–17391.
Finnerup NB et al (2015) Pharmacotherapy for neuropathic pain in
adults: a systematic review and meta-analysis. Lancet Neurol
14(2):162–173.
Finnerup NB, Sindrup SH, Jensen TS (2010) The evidence for
pharmacological treatment of neuropathic pain. Pain
150:573–581.
Flores JA, Banoua El F, Gala
´n-Rodrı
´guez B, Fernandez-Espejo E
(2004) Opiate anti-nociception is attenuated following lesion of
large dopamine neurons of the periaqueductal grey: critical role
for D1 (not D2) dopamine receptors. Pain 110:205–214.
Gagnon M, Bergeron MJ, Lavertu G, Castonguay A, Tripathy S,
Bonin RP, Pe
´rez-Sa
´nchez J, Boudreau D, Wang B, Dumas L,
Valade I, Bachand K, Jacob-Wagner M, Tardif C, Kianicka I,
Isenring P, Attardo G, Coull JAM, De Koninck Y (2013) Chloride
extrusion enhancers as novel therapeutics for neurological
diseases. Nat Med 19:1524–1528.
Gallagher JP, Higashi H, Nishi S (1978) Characterization and ionic
basis of GABA-induced depolarizations recorded in vitro from cat
primary afferent neurones. J Physiol 275:263–282.
Gilron I, Jensen TS, Dickenson AH (2013) Combination
pharmacotherapy for management of chronic pain: from bench
to bedside. Lancet Neurol 12:1084–1095.
Gold BG (1997) Axonal regeneration of sensory nerves is delayed by
continuous intrathecal infusion of nerve growth factor. NSC
76:1153–1158.
Govrin-Lippmann R, Devor M (1978) Ongoing activity in severed
nerves: source and variation with time. Brain Res 159:406–410.
Grace PM, Hutchinson MR, Maier SF, Watkins LR (2014)
Pathological pain and the neuroimmune interface. Nat Rev
Immunol 14:217–231.
Graham B, Redman S (1994) A simulation of action potentials in
synaptic boutons during presynaptic inhibition. J Neurophysiol
71:538–549.
Grudt TJ, Perl ER (2002) Correlations between neuronal morphology
and electrophysiological features in the rodent superficial dorsal
horn. J Physiol 540:189–207.
Hanz S, Perlson E, Willis D, Zheng J-Q, Massarwa R, Huerta JJ,
Koltzenburg M, Kohler M, van-Minnen J, Twiss JL, Fainzilber M
(2003) Axoplasmic importins enable retrograde injury signaling in
lesioned nerve. Neuron 40:1095–1104.
Haroutounian S, Nikolajsen L, Bendtsen TF, Finnerup NB, Kristensen
AD, Hasselstrøm JB, Jensen TS (2014) Primary afferent input
critical for maintaining spontaneous pain in peripheral neuropathy.
Pain 155:1272–1279.
Hathway GJ, Vega-Avelaira D, Moss A, Ingram R, Fitzgerald M
(2009) Brief, low frequency stimulation of rat peripheral C-fibres
evokes prolonged microglial-induced central sensitization in
adults but not in neonates. Pain 144:110–118.
Heinke B, Ruscheweyh R, Forsthuber L, Wunderbaldinger G,
Sandku
¨hler J (2004) Physiological, neurochemical and
morphological properties of a subgroup of GABAergic spinal
lamina II neurones identified by expression of green fluorescent
protein in mice. J Physiol 560:249–266.
Heinricher MM, Morgan MM, Fields HL (1992) Direct and indirect
actions of morphine on medullary neurons that modulate
nociception. NSC 48:533–543.
Hentall ID, Mesigil R, Pinzon A, Noga BR (2003) Temporal and
spatial profiles of pontine-evoked monoamine release in the rat’s
spinal cord. J Neurophysiol 89:2943–2951.
Hoyer D, Hannon JP, Martin GR (2002) Molecular, pharmacological
and functional diversity of 5-HT receptors. Pharmacol Biochem
Behav 71:533–554.
Hughes SW, Hickey L, Hulse RP, Lumb BM, Pickering AE (2013)
Endogenous analgesic action of the pontospinal noradrenergic
system spatially restricts and temporally delays the progression of
neuropathic pain following tibial nerve injury. Pain
154:1680–1690.
Hulse R, Wynick D, Donaldson LF (2010) Intact cutaneous C fibre
afferent properties in mechanical and cold neuropathic allodynia.
Eur J Pain 14:565.e1–565.e10.
Ibuki T, Hama AT, Wang XT, Pappas GD, Sagen J (1997) Loss of
GABA-immunoreactivity in the spinal dorsal horn of rats with
peripheral nerve injury and promotion of recovery by adrenal
medullary grafts. NSC 76:845–858.
Ikeda H, Stark J, Fischer H, Wagner M, Drdla R, Ja
¨ger T, Sandku
¨hler
J (2006) Synaptic amplifier of inflammatory pain in the spinal
dorsal horn. Science 312:1659–1662.
Intondi AB, Dahlgren MN, Eilers MA, Taylor BK (2008) Intrathecal
neuropeptide Y reduces behavioral and molecular markers of
inflammatory or neuropathic pain. Pain 137:352–365.
Jaggi AS, Jain V, Singh N (2011) Animal models of neuropathic pain.
Fundam Clin Pharmacol 25:1–28.
Jensen MP, Chodroff MJ, Dworkin RH (2007) The impact of
neuropathic pain on health-related quality of life review and
implications. Neurology 68:1178–1182.
Jensen TS, Baron R, Haanpa
¨a
¨M, Kalso E, Loeser JD, Rice ASC,
Treede R-D (2011) A new definition of neuropathic pain. Pain
152:2204–2205.
Ji R-R, Kohno T, Moore KA, Woolf CJ (2003) Central sensitization
and LTP: do pain and memory share similar mechanisms? Trends
Neurosci 26:696–705.
Kato G, Kawasaki Y, Ji RR, Strassman AM (2007) Differential wiring
of local excitatory and inhibitory synaptic inputs to islet cells in rat
spinal lamina II demonstrated by laser scanning photostimulation.
J Physiol 580:815–833.
Kato G, Kawasaki Y, Koga K, Uta D, Kosugi M, Yasaka T, Yoshimura
M, Ji R-R, Strassman AM (2009) Organization of intralaminar and
translaminar neuronal connectivity in the superficial spinal dorsal
horn. J Neurosci 29:5088–5099.
Kato G, Kosugi M, Mizuno M, Strassman AM (2013) Three-
dimensional organization of local excitatory and inhibitory inputs
to neurons in laminae III–IV of the spinal dorsal horn. J Physiol
591:5645–5660.
Keller AF, Beggs S, Salter MW, De Koninck Y (2007) Transformation
of the output of spinal lamina I neurons after nerve injury and
microglia stimulation underlying neuropathic pain. Mol Pain 3:27.
Khan GM, Chen SR, Pan HL (2002) Role of primary afferent nerves in
allodynia caused by diabetic neuropathy in rats. NSC
114:291–299.
Khasabov SG, Rogers SD, Ghilardi JR, Peters CM, Mantyh PW,
Simone DA (2002) Spinal neurons that possess the substance P
S. J. West et al. / Neuroscience 300 (2015) 254–275 271
receptor are required for the development of central sensitization.
J Neurosci 22:9086–9098.
Knyihar-Csillik E, Csillik B (1981) Selective ‘‘labeling’’ by
transsynaptic degeneration of substantia gelatinosal cells: an
attempt to decipher intrinsic wiring in the Rolando substance of
primates. Neurosci Lett 23:131–136.
Knyihar-Csillik E, Csillik B, Rakic P (1982) Periterminal synaptology
of dorsal root glomerular terminals in the substantia gelatinosa of
the spinal cord in the rhesus monkey. J Comp Neurol
210:376–399.
Knyihar-Csillik E, Kreutzberg GW, Raivich G, Csillik B (1991)
Vasoactive intestinal polypeptide in dorsal root terminals of the
rat spinal cord is regulated by the axoplasmic transport in the
peripheral nerve. Neurosci Lett 131:83–87.
Kwiat GC, Basbaum AI (1992) The origin of brainstem noradrenergic
and serotonergic projections to the spinal cord dorsal horn in the
rat. Somatosens Mot Res 9:157–173.
Latremoliere A, Woolf CJ (2009) Central sensitization: a generator of
pain hypersensitivity by central neural plasticity. J Pain
10:895–926.
Lavertu G, Co
ˆte
´SL, De Koninck Y (2014) Enhancing K-Cl co-
transport restores normal spinothalamic sensory coding in a
neuropathic pain model. Brain 137:724–738.
Lawson SN, Harper AA, Harper EI, Garson JA, Anderton BH (1984) A
monoclonal antibody against neurofilament protein specifically
labels a subpopulation of rat sensory neurones. J Comp Neurol
228:263–272.
Leitner J, Westerholz S, Heinke B, Forsthuber L, Wunderbaldinger G,
Ja
¨ger T, Gruber-Schoffnegger D, Braun K, Sandku
¨hler J (2013)
Impaired excitatory drive to spinal GABAergic neurons of
neuropathic mice. PLoS One 8:e73370.
Leong ML, Gu M, Speltz-Paiz R, Stahura EI, Mottey N, Steer CJ,
Wessendorf M (2011) Neuronal loss in the rostral ventromedial
medulla in a rat model of neuropathic pain. J Neurosci
31:17028–17039.
Lever I, Cunningham J, Grist J, Yip PK, Malcangio M (2003) Release
of BDNF and GABA in the dorsal horn of neuropathic rats. Eur J
Neurosci 18:1169–1174.
Lever IJ, Bradbury EJ, Cunningham JR, Adelson DW, Jones MG,
Mcmahon SB, Marvizo
´n JC, Malcangio M (2001) Brain-derived
neurotrophic factor is released in the dorsal horn by distinctive
patterns of afferent fiber stimulation. J Neurosci 21:4469–4477.
Li C-Y, Song Y-H, Higuera ES, Luo ZD (2004) Spinal dorsal horn
calcium channel alpha2delta-1 subunit upregulation contributes to
peripheral nerve injury-induced tactile allodynia. J Neurosci
24:8494–8499.
Li K-W, Yu YP, Zhou C, Kim D-S, Lin B, Sharp K, Steward O, Luo ZD
(2014) Calcium channel
a
2d1 proteins mediate trigeminal
neuropathic pain states associated with aberrant excitatory
synaptogenesis. J Biol Chem 289(10):7025–7037.
Li L, Rutlin M, Abraira VE, Cassidy C, Kus L, Gong S, Jankowski MP,
Luo W, Heintz N, Koerber HR, Woodbury CJ, Ginty DD (2011)
The functional organization of cutaneous low-threshold
mechanosensory neurons. Cell 147:1615–1627.
Lima D, Coimbra A (1983) The neuronal population of the marginal
zone (lamina I) of the rat spinal cord. A study based on
reconstructions of serially sectioned cells. Anat Embryol
167:273–288.
Lima D, Coimbra A (1986) A Golgi study of the neuronal population of
the marginal zone (lamina I) of the rat spinal cord. J Comp Neurol
244:53–71.
Lindwall C, Kanje M (2005) Retrograde axonal transport of JNK
signaling molecules influence injury induced nuclear changes in p-
c-Jun and ATF3 in adult rat sensory neurons. Mol Cell Neurosci
29:269–282.
Liu CN, Wall PD, Ben-Dor E, Michaelis M, Amir R, Devor M (2000)
Tactile allodynia in the absence of C-fiber activation: altered firing
properties of DRG neurons following spinal nerve injury. Pain
85:503–521.
Liu F-Y, Qu X-X, Cai J, Wang F-T, Xing G-G, Wan Y (2011)
Electrophysiological properties of spinal wide dynamic range
neurons in neuropathic pain rats following spinal nerve ligation.
Neurosci Bull 27:1–8.
Liu XG, Sandku
¨hler J (1995) Long-term potentiation of C-fiber-
evoked potentials in the rat spinal dorsal horn is prevented by
spinal N-methyl-
D
-aspartic acid receptor blockage. Neurosci Lett
191:43–46.
Lorenzo L-E, Magnussen C, Bailey AL, Louis MS, De Koninck Y,
Ribeiro-da-Silva A (2014) Spatial and temporal pattern of changes
in the number of GAD65-immunoreactive inhibitory terminals in
the rat superficial dorsal horn following peripheral nerve injury.
Mol Pain 10:57.
Lu Y, Dong H, Gao Y, Gong Y, Ren Y, Gu N, Zhou S, Xia N, Sun Y-Y,
Ji R-R, Xiong L (2013) A feed-forward spinal cord glycinergic
neural circuit gates mechanical allodynia. J Clin Invest
123:4050–4062.
Lu Y, Perl ER (2003) A specific inhibitory pathway between
substantia gelatinosa neurons receiving direct C-fiber input. J
Neurosci 23:8752–8758.
Lu Y, Perl ER (2005) Modular organization of excitatory circuits
between neurons of the spinal superficial dorsal horn (laminae I
and II). J Neurosci 25:3900–3907.
Luz LL, Szucs P, Safronov BV (2014) Peripherally driven low-
threshold inhibitory inputs to lamina I local-circuit and projection
neurones: a new circuit for gating pain responses. J Physiol
592:1519–1534.
Ma Q-P, Woolf CJ (1995) Involvement of neurokinin receptors in the
induction but not the maintenance of mechanical allodynia in rat
flexor motoneurones. J Physiol 486:769–777.
Ma W, Eisenach JC (2003) Cyclooxygenase 2 in infiltrating
inflammatory cells in injured nerve is universally up-regulated
following various types of peripheral nerve injury. NSC
121:691–704.
Mansikka H, Pertovaara A (1995) The role of
a
2-adrenoceptors of
the medullary lateral reticular nucleus in spinal antinociception in
rats. Brain Res Bull 37:633–638.
Mayer ML, Westbrook GL, Guthrie PB (1984) Voltage-dependent
block by Mg
2+
of NMDA responses in spinal cord neurones.
Nature 309(5965):261–263.
Melzack R, Wall PD (1965) Pain mechanisms: a new theory. Science
150:971–979.
Mendell LM (1966) Physiological properties of unmyelinated fiber
projection to the spinal cord. Exp Neurol 16:316–332.
Mendell LM, Wall PD (1965) Responses of single dorsal cord
cells to peripheral cutaneous unmyelinated fibres. Nature
206:97–99.
Merighi A, Polak JM, Theodosis DT (1991) Ultrastructural
visualization of glutamate and aspartate immunoreactivities in
the rat dorsal horn, with special reference to the co-localization of
glutamate, substance P and calcitonin-gene related peptide. NSC
40:67–80.
Michaelevski I, Segal-Ruder Y, Rozenbaum M, Medzihradszky KF,
Shalem O, Coppola G, Horn-Saban S, Ben-Yaakov K, Dagan SY,
Rishal I, Geschwind DH, Pilpel Y, Burlingame AL, Fainzilber M
(2010) Signaling to transcription networks in the neuronal
retrograde injury response. Sci Signal 3:ra53.
Millan MJ (1999) The induction of pain: an integrative review. Prog
Neurobiol 57:1–164.
Millan MJ (2002) Descending control of pain. Prog Neurobiol
66:355–474.
Miraucourt LS, Dallel R, Voisin DL (2007) Glycine inhibitory
dysfunction turns touch into pain through PKCgamma
interneurons. PLoS One 2:e1116.
Miraucourt LS, Moisset X, Dallel R, Voisin DL (2009) Glycine
inhibitory dysfunction induces a selectively dynamic, morphine-
resistant, and neurokinin 1 receptor- independent mechanical
allodynia. J Neurosci 29:2519–2527.
Molander C, Xu Q, Grant G (1984) The cytoarchitectonic organization
of the spinal cord in the rat. I. The lower thoracic and lumbosacral
cord. J Comp Neurol 230:133–141.
Moore KA, Kohno T, Karchewski LA, Scholz J, Baba H, Woolf CJ
(2002) Partial peripheral nerve injury promotes a selective loss of
272 S. J. West et al. / Neuroscience 300 (2015) 254–275
GABAergic inhibition in the superficial dorsal horn of the spinal
cord. J Neurosci 22:6724–6731.
Morgan MM, Fields HL (1994) Pronounced changes in the activity of
nociceptive modulatory neurons in the rostral ventromedial
medulla in response to prolonged thermal noxious stimuli. J
Neurophysiol 72:1161–1170.
Morisset V, Nagy F (1999) Ionic basis for plateau potentials in deep
dorsal horn neurons of the rat spinal cord. J Neurosci
19:7309–7316.
Nagy I, Maggi CA, Dray A, Woolf CJ, Urban L (1993) The role of
neurokinin and N-methyl-
D
-aspartate receptors in synaptic
transmission from capsaicin-sensitive primary afferents in the rat
spinal cord in vitro. NSC 52:1029–1037.
Nakaya Y, Kaneko T, Shigemoto R, Nakanishi S, Mizuno N (1994)
Immunohistochemical localization of substance P receptor in the
central nervous system of the adult rat. J Comp Neurol
347:249–274.
Nichols ML, Allen BJ, Rogers SD, Ghilardi JR, Honore P, Luger NM,
Finke MP, Li J, Lappi DA, Simone DA, Mantyh PW (1999)
Transmission of chronic nociception by spinal neurons expressing
the substance P receptor. Science 286:1558–1561.
O’Connor AB (2009) Neuropathic pain: quality-of-life impact, costs
and cost effectiveness of therapy. Pharmacoeconomics
27:95–112.
O’Connor AB, Dworkin RH (2009) Treatment of neuropathic pain: an
overview of recent guidelines. Am J Med 122:S22–S32.
Obata K, Yamanaka H, Kobayashi K, Dai Y, Mizushima T, Katsura H,
Fukuoka T, Tokunaga A, Noguchi K (2006) The effect of site and
type of nerve injury on the expression of brain-derived
neurotrophic factor in the dorsal root ganglion and on
neuropathic pain behavior. NSC 137:961–970.
Ossipov MH, Dussor GO, Porreca F (2010) Central modulation of
pain. J Clin Invest 120:3779–3787.
Perkins JR, Antunes-Martins A, Calvo M, Grist J, Rust W, Schmid R,
Hildebrandt T, Kohl M, Orengo C, McMahon SB, Bennett DLH
(2014) A comparison of RNA-seq and exon arrays for whole
genome transcription profiling of the L5 spinal nerve transection
model of neuropathic pain in the rat. Mol Pain 10:7.
Pertovaara A (2013) The noradrenergic pain regulation system: a
potential target for pain therapy. Eur J Pharmacol 716(1):2–7.
Pieraut S, Laurent-Matha V, Sar C, Hubert T, Me
´chaly I, Hilaire C,
Mersel M, Delpire E, Valmier J, Scamps F (2007) NKCC1
phosphorylation stimulates neurite growth of injured adult
sensory neurons. J Neurosci 27:6751–6759.
Polgar E, Gray S, Riddell JS, Todd AJ (2004) Lack of evidence for
significant neuronal loss in laminae I-III of the spinal dorsal horn of
the rat in the chronic constriction injury model. Pain 111:144–150.
Polgar E, Hughes D, Riddell J, Maxwell D, Puskar Z, Todd A (2003)
Selective loss of spinal GABAergic or glycinergic neurons is not
necessary for development of thermal hyperalgesia in the chronic
constriction injury model of neuropathic pain. Pain 104:229–239.
Polgar E, Hughes DI, Arham AZ, Todd AJ (2005) Loss of neurons
from laminas I-III of the spinal dorsal horn is not required for
development of tactile allodynia in the spared nerve injury model
of neuropathic pain. J Neurosci 25:6658.
Polgar E, Todd AJ (2008) Tactile allodynia can occur in the spared
nerve injury model in the rat without selective loss of GABA or
GABA(A) receptors from synapses in laminae I-II of the ipsilateral
spinal dorsal horn. NSC 156:193–202.
Polga
´r E, Watanabe M, Hartmann B, Grant SG, Todd AJ (2008)
Expression of AMPA receptor subunits at synapses in laminae I-
III of the rodent spinal dorsal horn. Mol Pain 4:5.
Porreca F, Ossipov MH, Gebhart GF (2002) Chronic pain and
medullary descending facilitation. Trends Neurosci 25:319–325.
Price TJ, Cervero F, Gold MS, Hammond DL, Prescott SA (2009)
Chloride regulation in the pain pathway. Brain Res Rev
60:149–170.
Proudlock F, Spike RC, Todd AJ (1993) Immunocytochemical study
of somatostatin, neurotensin, GABA, and glycine in rat spinal
dorsal horn. J Comp Neurol 327:289–297.
Punnakkal P, Schoultz von C, Haenraets K, Wildner H, Zeilhofer HU
(2014) Morphological, biophysical and synaptic properties of
glutamatergic neurons of the mouse spinal dorsal horn. J
Physiol 592:759–776.
Rahman W, Bauer CS, Bannister K, Vonsy J-L, Dolphin AC,
Dickenson AH (2009) Descending serotonergic facilitation and
the antinociceptive effects of pregabalin in a rat model of
osteoarthritic pain. Mol Pain 5:45.
Rahman W, D’Mello R, Dickenson AH (2008) Peripheral nerve injury-
induced changes in spinal alpha(2)-adrenoceptor-mediated
modulation of mechanically evoked dorsal horn neuronal
responses. J Pain 9:350–359.
Raivich G, Hellweg R, Kreutzberg GW (1991) NGF receptor-
mediated reduction in axonal NGF uptake and retrograde
transport following sciatic nerve injury and during regeneration.
Neuron 7:151–164.
Rempe T, Wolff S, Riedel C, Baron R, Stroman PW, Jansen O,
Gierthmu
¨hlen J (2014) Spinal fMRI reveals decreased
descending inhibition during secondary mechanical
hyperalgesia. PLoS One 9:e112325.
Rexed B (1952) The cytoarchitectonic organization of the spinal cord
in the cat. J Comp Neurol 96:415–495.
Ribeiro-Da-Silva A, Castro-Lopes JM, Coimbra A (1986) Distribution
of glomeruli with fluoride-resistant acid phosphatase (FRAP)-
containing terminals in the substantia gelatinosa of the rat. Brain
Res 377:323–329.
Ribeiro-Da-Silva A, Coimbra A (1982) Two types of synaptic
glomeruli and their distribution in laminae I-III of the rat spinal
cord. J Comp Neurol 209:176–186.
Ribeiro-Da-Silva A, Pignatelli D, Coimbra A (1985) Synaptic
architecture of glomeruli in superficial dorsal horn of rat spinal
cord, as shown in serial reconstructions. J Neurocytol
14:203–220.
Ribeiro-Da-Silva A, Tagari P, Cuello AC (1989) Morphological
characterization of substance P-like immunoreactive glomeruli in
the superficial dorsal horn of the rat spinal cord and trigeminal
subnucleus caudalis: a quantitative study. J Comp Neurol 281.
497–415.
Roudet C, Mouchet P, Feuerstein C, Savasta M (1994) Normal
distribution of alpha 2-adrenoceptors in the rat spinal cord and its
modification after noradrenergic denervation: a quantitative
autoradiographic study. J Neurosci Res 39:319–329.
Rudomin P, Schmidt RF (1999) Presynaptic inhibition in the
vertebrate spinal cord revisited. Exp Brain Res 129:1–37.
Sandku
¨hler J (2009) Models and mechanisms of hyperalgesia and
allodynia. Physiol Rev 89:707–758.
Sandku
¨hler J, Gruber-Schoffnegger D (2012) Hyperalgesia by
synaptic long-term potentiation (LTP): an update. Curr Opin
Pharmacol 12:18–27.
Satoh O, Omote K (1996) Roles of monoaminergic, glycinergic and
GABAergic inhibitory systems in the spinal cord in rats with
peripheral mononeuropathy. Brain Res 728:27–36.
Schneider SP (1992) Functional properties and axon terminations of
interneurons in laminae III–V of the mammalian spinal dorsal horn
in vitro. J Neurophysiol 68(5):1746–1759.
Scholz J, Broom DC, Youn D-H, Mills CD, Kohno T, Suter MR, Moore
KA, Decosterd I, Coggeshall RE, Woolf CJ (2005) Blocking
caspase activity prevents transsynaptic neuronal apoptosis and
the loss of inhibition in lamina II of the dorsal horn after peripheral
nerve injury. J Neurosci 25:7317–7323.
Shadiack AM, SUN Y, Zigmond RE (2001) Nerve growth factor
antiserum induces axotomy-like changes in neuropeptide
expression in intact sympathetic and sensory neurons. J
Neurosci 21:363–371.
Shehab S, Atkinson ME (1986) Vasoactive intestinal polypeptide
(VIP) increases in the spinal cord after peripheral axotomy of the
sciatic nerve originate from primary afferent neurons. Brain Res
372:37–44.
Shehab SAS, Spike RC, Todd AJ (2003) Evidence against cholera
toxin B subunit as a reliable tracer for sprouting of primary
S. J. West et al. / Neuroscience 300 (2015) 254–275 273
afferents following peripheral nerve injury. Brain Res
964:218–227.
Shehab SAS, Spike RC, Todd AJ (2004) Do central terminals of intact
myelinated primary afferents sprout into the superficial dorsal
horn of rat spinal cord after injury to a neighboring peripheral
nerve? J Comp Neurol 474:427–437.
Sikandar S, Ronga I, Iannetti GD, Dickenson AH (2013) Neural
coding of nociceptive stimuli-from rat spinal neurones to human
perception. Pain 154:1263–1273.
Simone DA, Sorkin LS, Oh U, Chung JM, Owens C, LaMotte RH,
Willis WD (1991) Neurogenic hyperalgesia: central neural
correlates in responses of spinothalamic tract neurons. J
Neurophysiol 66:228–246.
Sivilotti LG, Gerber G, Rawat B, Woolf CJ (1995) Morphine
selectively depresses the slowest, NMDA-independent
component of C-fibre-evoked synaptic activity in the rat spinal
cord in vitro. Eur J Neurosci 7:12–18.
Sivilotti LG, Thompson SW, Woolf CJ (1993) Rate of rise of the
cumulative depolarization evoked by repetitive stimulation of
small-caliber afferents is a predictor of action potential windup in
rat spinal neurons in vitro. J Neurophysiol 69:1621–1631.
Smith BH, Torrance N (2012) Epidemiology of neuropathic pain and
its impact on quality of life. Curr Pain Headache Rep 16:191–198.
Smith BH, Torrance N, Bennett MI, Lee AJ (2007) Health and quality
of life associated with chronic pain of predominantly neuropathic
origin in the community. Clin J Pain 23:143–149.
Snider WD, McMahon SB (1998) Tackling pain at the source: new
ideas about nociceptors. Neuron 20:629–632.
Solway B, Bose SC, Corder G, Donahue RR, Taylor BK (2011) Tonic
inhibition of chronic pain by neuropeptide Y. Proc Natl Acad Sci
108:7224–7229.
Sood A, Cunningham C, Lin S (2013) The BB wistar rat as a diabetic
model for fracture healing. ISRN Endocrinol 2013:349604.
Spike RC, Puskar Z, Andrew D, Todd AJ (2003) A quantitative and
morphological study of projection neurons in lamina I of the rat
lumbar spinal cord. Eur J Neurosci 18:2433–2448.
Stanfa LC, Dickenson AH (1994) Electrophysiological studies on the
spinal roles of endogenous opioids in carrageenan inflammation.
Pain 56:185–191.
Stroman PW et al (2014) The current state-of-the-art of spinal cord
imaging: methods. NeuroImage 84:1070–1081.
Sun R-Q, Lawand NB, Willis WD (2003) The role of calcitonin gene-
related peptide (CGRP) in the generation and maintenance of
mechanical allodynia and hyperalgesia in rats after intradermal
injection of capsaicin. Pain 104:201–208.
Suzuki R, Morcuende S, Webber M, Hunt SP, Dickenson AH (2002)
Superficial NK1-expressing neurons control spinal excitability
through activation of descending pathways. Nat Neurosci
5:1319–1326.
Suzuki R, Rahman W, Hunt SP, Dickenson AH (2004) Descending
facilitatory control of mechanically evoked responses is enhanced
in deep dorsal horn neurones following peripheral nerve injury.
Brain Res 1019:68–76.
Suzuki R, Rahman W, Rygh LJ, Webber M, Hunt SP, Dickenson AH
(2005) Spinal-supraspinal serotonergic circuits regulating
neuropathic pain and its treatment with gabapentin. Pain
117:292–303.
Tajti J, Fischer J, Knyihar-Csillik E, Csillik B (1988) Transganglionic
regulation and fine structural localization of lectin-reactive
carbohydrate epitopes in primary sensory neurons of the rat.
Histochemistry 88:213–218.
Tan AM, Chang Y-W, Zhao P, Hains BC, Waxman SG (2011) Rac1-
regulated dendritic spine remodeling contributes to neuropathic
pain after peripheral nerve injury. Exp Neurol 232(2):222–223.
Tan AM, Samad OA, Fischer TZ, Zhao P, Persson AK, Waxman SG
(2012) Maladaptive dendritic spine remodeling contributes to
diabetic neuropathic pain. J Neurosci 32:6795–6807.
Tanaka M, Matsumoto Y, Murakami T, Hisa Y, Ibata Y (1996) The
origins of catecholaminergic innervation in the rostral
ventromedial medulla oblongata of the rat. Neurosci Lett
207:53–56.
Thompson SW, King AE, Woolf CJ (1990) Activity-dependent
changes in rat ventral horn neurons in vitro; summation of
prolonged afferent evoked postsynaptic depolarizations produce a
d-2-amino-5-phosphonovaleric acid sensitive windup. Eur J
Neurosci 2:638–649.
Todd AJ (2002) Anatomy of primary afferents and projection
neurones in the rat spinal dorsal horn with particular emphasis
on substance P and the neurokinin 1 receptor. Exp Physiol
87:245–249.
Todd AJ (2010) Neuronal circuitry for pain processing in the dorsal
horn. Nat Rev Neurosci 11:823–836.
Todd AJ, McGill MM, Shehab SA (2000) Neurokinin 1 receptor
expression by neurons in laminae I, III and IV of the rat spinal
dorsal horn that project to the brainstem. Eur J Neurosci
12:689–700.
Todd AJ, Sullivan AC (1990) Light microscope study of the
coexistence of GABA-like and glycine-like immunoreactivities in
the spinal cord of the rat. J Comp Neurol 296:496–505.
Tracey I, Mantyh PW (2007) The cerebral signature for pain
perception and its modulation. Neuron 55:377–391.
Tsuda M, Beggs S, Salter MW, Inoue K (2013) Microglia and
intractable chronic pain. Glia 61:55–61.
Tsuruoka M, Matsutani K, Inoue T (2003) Coeruleospinal inhibition of
nociceptive processing in the dorsal horn during unilateral
hindpaw inflammation in the rat. Pain 104:353–361.
Uc¸ eyler N, Sommer C (2012) Cytokine-related and histological
biomarkers for neuropathic pain assessment. Pain Manag
2:391–398.
Urban MO, Gebhart GF (1997) Characterization of biphasic
modulation of spinal nociceptive transmission by neurotensin in
the rat rostral ventromedial medulla. J Neurophysiol
78:1550–1562.
Urban MO, Gebhart GF (1999) Supraspinal contributions to
hyperalgesia. Proc Natl Acad Sci U S A 96:7687–7692.
Usoskin D, Furlan A, Islam S, Abdo H, Lo
¨nnerberg P, Lou D, Hjerling-
Leffler J, Haeggstro
¨m J, Kharchenko O, Kharchenko PV,
Linnarsson S, Ernfors P (2015) Unbiased classification of
sensory neuron types by large-scale single-cell RNA
sequencing. Nat Neurosci 18:145–153.
Vera-Portocarrero LP, Xie JY, Yie JX, Kowal J, Ossipov MH, King T,
Porreca F (2006) Descending facilitation from the rostral
ventromedial medulla maintains visceral pain in rats with
experimental pancreatitis. Gastroenterology 130:2155–2164.
Wakisaka S, Kajander KC, Bennett GJ (1992) Effects of peripheral
nerve injuries and tissue inflammation on the levels of
neuropeptide Y-like immunoreactivity in rat primary afferent
neurons. Brain Res 598:349–352.
Wall PD, Waxman S, Basbaum AI (1974) Ongoing activity in
peripheral nerve: injury discharge. Exp Neurol 45:576–589.
Wang X, Zhang J, Eberhart D, Urban R, Meda K, Solorzano C,
Yamanaka H, Rice D, Basbaum AI (2013) Excitatory superficial
dorsal horn interneurons are functionally heterogeneous and
required for the full behavioral expression of pain and itch. Neuron
78:312–324.
Watkins JC, Evans RH (1981) Excitatory amino acid transmitters.
Annu Rev Pharmacol Toxicol 21:165–204.
Wei B, Kumada T, Furukawa T, Inoue K, Watanabe M, Sato K,
Fukuda A (2013) Pre- and post-synaptic switches of GABA
actions associated with Cl-homeostatic changes are induced in
the spinal nucleus of the trigeminal nerve in a rat model of
trigeminal neuropathic pain. Neuroscience 228:334–348.
Widerstro
¨m-Noga E, Pattany PM, Cruz-Almeida Y, Felix ER, Perez S,
Cardenas DD, Martinez-Arizala A (2013) Metabolite
concentrations in the anterior cingulate cortex predict high
neuropathic pain impact after spinal cord injury. Pain
154:204–212.
Willis WD (1999) Dorsal root potentials and dorsal root reflexes: a
double-edged sword. Exp Brain Res 124:395–421.
Woodbury CJ, Kullmann FA, McIlwrath SL, Koerber HR (2008)
Identity of myelinated cutaneous sensory neurons projecting to
274 S. J. West et al. / Neuroscience 300 (2015) 254–275
nocireceptive laminae following nerve injury in adult mice. J Comp
Neurol 508:500–509.
Woodruff GN, Foster AC, Gill R, Kemp JA, Wong EH, Iversen LL
(1987) The interaction between MK-801 and receptors
for N-methyl-
D
-aspartate: functional consequences.
Neuropharmacology 26:903–909.
Woolf C, Wiesenfeld-Hallin Z (1986) Substance P and calcitonin
gene-related peptide synergistically modulate the gain of the
nociceptive flexor withdrawal reflex in the rat. Neurosci Lett
66:226–230.
Woolf CJ (1983) Evidence for a central component of post-injury pain
hypersensitivity. Nature 306:686–688.
Woolf CJ, Salter MW (2000) Neuronal plasticity: increasing the gain in
pain. Science 288:1765.
Woolf CJ, Shortland P, Coggeshall RE (1992) Peripheral nerve injury
triggers central sprouting of myelinated afferents. Nature
355:75–78.
Woolf CJ, Thompson SW (1991) The induction and maintenance of
central sensitization is dependent on N-methyl-
D
-aspartic acid
receptor activation; implications for the treatment of post-injury
pain hypersensitivity states. Pain 44:293–299.
Xiao H-S, Huang Q-H, Zhang F-X, Bao L, Lu Y-J, Guo C, Yang L,
Huang W-J, Fu G, Xu S-H, Cheng X-P, Yan Q, Zhu Z-D, Zhang X,
Chen Z, Han Z-G, Zhang X (2002) Identification of gene
expression profile of dorsal root ganglion in the rat peripheral
axotomy model of neuropathic pain. Proc Natl Acad Sci U S A
99:8360–8365.
Xiao WH, Bennett GJ (2008) Chemotherapy-evoked neuropathic
pain: abnormal spontaneous discharge in A-fiber and C-fiber
primary afferent neurons and its suppression by acetyl-L-
carnitine. Pain 135:262–270.
Yarnitsky D, Granot M, Nahman-Averbuch H, Khamaisi M,
Granovsky Y (2012) Conditioned pain modulation predicts
duloxetine efficacy in painful diabetic neuropathy. Pain
153:1193–1198.
Yasaka T, Kato G, Furue H, Rashid MH, Sonohata M, Tamae A,
Murata Y, Masuko S, Yoshimura M (2007) Cell-type-specific
excitatory and inhibitory circuits involving primary afferents in the
substantia gelatinosa of the rat spinal dorsal horn in vitro. J
Physiol 581:603–618.
Yasaka T, Tiong SY, Polga
´r E, Watanabe M, Kumamoto E, Riddell
JS, Todd AJ (2014) A putative relay circuit providing low-threshold
mechanoreceptive input to lamina I projection neurons via vertical
cells in lamina II of the rat dorsal horn. Mol Pain 10:3.
Yasaka T, Tiong SYX, Hughes DI, Riddell JS, Todd AJ (2010)
Populations of inhibitory and excitatory interneurons in lamina II of
the adult rat spinal dorsal horn revealed by a combined
electrophysiological and anatomical approach. Pain
151:475–488.
Yoshimura M, Jessell TM (1989) Primary afferent-evoked synaptic
responses and slow potential generation in rat substantia
gelatinosa neurons in vitro. J Neurophysiol 62:96–108.
Zeitz KP, Guy N, Malmberg AB, Dirajlal S, Martin WJ, Sun L,
Bonhaus DW, Stucky CL, Julius D, Basbaum AI (2002) The 5-
HT3 subtype of serotonin receptor contributes to nociceptive
processing via a novel subset of myelinated and unmyelinated
nociceptors. J Neurosci 22:1010–1019.
Zrouri H, Le Goascogne C, Li WW, Pierre M, Courtin F (2004) The
role of MAP kinases in rapid gene induction after lesioning of the
rat sciatic nerve. Eur J Neurosci 20:1811–1818.
Zylka MJ, Sowa NA, Taylor-Blake B, Twomey MA, Herrala A, Voikar
V, Vihko P (2008) Prostatic acid phosphatase is an
ectonucleotidase and suppresses pain by generating adenosine.
Neuron 60:111–122.
Ørstavik K, Jørum E (2010) Microneurographic findings of relevance
to pain in patients with erythromelalgia and patients with diabetic
neuropathy. Neurosci Lett 470:180–184.
(Accepted 8 May 2015)
(Available online 16 May 2015)
S. J. West et al. / Neuroscience 300 (2015) 254–275 275
