A gastrin-releasing peptide receptor mediates the
itch sensation in the spinal cord
Yan-Gang Sun1& Zhou-Feng Chen1,2,3
Itching, or pruritus, is defined as an unpleasant cutaneous sen-
sation that serves as a physiological self-protective mechanism to
prevent the body from being hurt by harmful external agents.
Chronic itch represents a significant clinical problem resulting
from renal diseases and liver diseases, as well as several serious
skin diseases such as atopic dermatitis1–3. The identity of the itch-
specific mediator in the central nervous system, however, remains
elusive. Here we describe that the gastrin-releasing peptide recep-
tor (GRPR) plays an important part in mediating itch sensation in
the dorsal spinal cord. We found that gastrin-releasing peptide is
specifically expressed in a small subset of peptidergic dorsal root
ganglion neurons, whereas expression of its receptor GRPR is
showed comparable thermal, mechanical, inflammatory and
induction of scratching behaviour was significantly reduced in
GRPR mutant mice in response to pruritogenic stimuli, whereas
normal responses were evoked by painful stimuli. Moreover,
direct spinal cerebrospinal fluid injection of a GRPR antagonist
significantly inhibited scratching behaviour in three independent
itch models. These data demonstrate that GRPR is required for
mediating the itch sensation rather than pain, at the spinal level.
Our results thus indicate that GRPR may represent the first mole-
cule that is dedicated to mediating the itch sensation in the dorsal
target for antipruritic drug development.
GRPR, a G-protein coupled receptor, is a mammalian homologue
besin in rodents elicited grooming behaviour6,7, which is relevant to
the hypothesis that bombesin is a mediator of pruritus8. There is,
however, little evidence in support of this hypothesis. Indeed, bom-
besin-evoked grooming behaviour has been generally regarded as a
manifestation of stress response and thermoregulation9,10.
a subset of small and medium-sized dorsal root ganglion (DRG)
neurons (Fig.1a), andthat it iscolocalized withperipherin, amarker
for unmyelinated fibres (Fig. 1b). GRP is also colocalized with pep-
tidergic markers CGRP and substance P, but it neither co-stained
with IB4, a nonpeptidergic marker, nor with myelinated marker
NF200 (Fig. 1c, d, e, and data not shown). Furthermore, an approx-
imate 80% of GRP1neurons express TRPV1, a vanilloid receptor
important for detecting thermal and chemical stimuli, in the DRGs
to the lamina I and II outer layer (IIo) (Fig. 1f). GRP1fibres in the
dorsal horn largely diminished after dorsal rhizotomy (Supplemen-
tary Fig. 2), indicating that GRP arises from primary afferents. We
found no GRP mRNA expression in the superficial dorsal horn in
adult mice (data not shown). In contrast, we found GRPR1neurons
in laminaI of the dorsal horn (Fig. 1g, h) but none in the deep dorsal
1Department of Anesthesiology,2Department of Molecular Biology and Pharmacology,3Department of Psychiatry, Washington University School of Medicine Pain Center, St Louis,
Missouri 63110, USA.
Figure 1 | Expression pattern of GRP and GRPR in adult DRG and dorsal
horn of the spinal cord. a, GRP is detected in small to medium-sized DRG
neurons by immunocytochemistry. b–e, Double-staining of GRP with
peripherin (b), NF200 (c), IB4 (d) and CGRP (e) in DRGs. GRP (red) is
localized in peripherin1(green) DRG neurons (b). Double-staining of GRP
(red) and NF200 (green) in adult mouse DRGs indicates that GRP and
NF200 expression do not overlap (c). GRP is present in adult DRG neurons
labelled with a CGRP antibody (green in e), but not with IB4 (green in
d). Arrows indicate double-labelled neurons (b, e). f, GRP1fibres (red) are
GRPR is present in the superficial dorsal horn of the spinal cord. Higher
magnification of the dorsal horn is shown in h. Arrows indicate the GRPR1
neurons (h). Scale bars, 50mm (a, b, c, d, e, h); 100mm (f and g).
Vol 448|9 August 2007|doi:10.1038/nature06029
horn or the ventral horn of the spinal cord (Fig. 1g, and data not
shown). Several studies have suggested an itch-dedicated pathway in
the peripheral and central nervous system that involves a group of
itch-specific unmyelinated C-fibres and spinothalamic lamina Ineu-
rons in the spinal cord11,12. Thus, our results indicate that the GRP/
or itch sensation at the spinal level.
To assess whether GRPR may contribute to pain sensation, we
examined thermal, mechanical, inflammatory and neuropathic pain
responses of GRPR mutant mice. GRPR mutant mice and wild-type
micedidnotdifferintheHargreave’s pawwithdrawal testofthermal
nociception (Fig. 2a). And in the hotplate and tail-flick tests, GRPR
control across a range of temperatures (48uC, 52uC and 56uC for
hotplate, and 48uC, 50uC and 52uC for tail-flick tests; Fig. 2b, c).
Mechanical pain sensitivity was measured using graded von Frey
filaments. The mechanical threshold in GRPR mutant mice was sim-
ilar to that in wild-type mice (Fig. 2d). These results indicate that
GRPR signalling is not required for the transmission of acute pain
messages. To measure the response of GRPR mutant mice to inflam-
matory pain, flinching and licking behaviours of GRPR mutant
mice were compared with their control littermates after injection
of 5% formalin (10ml) into the right hindpaw of the animals. No
statistically significant differences in the first phase (0–10min) or in
the second phase (10–60min) between GRPR mutant and wild-type
mice were found (Fig. 2e). In addition, mechanical hypersensitivity
produced by hindpaw injection of complete Freund’s adjuvant was
indistinguishable between GRPR mutant mice and wild-type mice
(Supplementary Fig. 3). GRPR mutant and wild-type mice also
showed comparable mechanical hyperalgesia in a neuropathic pain
gene expression profile did not reveal significant changes in the dor-
sal spinal cord of GRPR mutant mice (Supplementary Fig. 5). Taken
together, these data suggest that GRPR is not essential for transmis-
sion of noxious information.
To ascertain whether GRPR is required for itch sensation, we
studied the effects of several pruritogenic agents in GRPR mutant
mice and wild-type mice. Intradermal injection of compound 48/80
(100mg per50ml), which degranulates mast cells to release histam-
viour in GRPR mutant mice, the number of scratches was signifi-
cantly smaller compared with that of wild-type mice (P,0.05,
as a pruritogenic agent. PAR2, a G-protein coupled receptor, is
expressed inprimary sensory neurons and isactivated byproteinases
Latency of tail
Phase IPhase II
Duration of licking
and flinching (s)
0 10 20 30 40 50 60 70
Figure 2 | Pain behaviours and locomotor activity are normal in GRPR
mutant mice. a–c, Responses to noxious thermal stimulation were
measured by paw withdrawal latency (Hargreaves test; a), hotplate (b) and
water immersion tail-flick latency (c). There were no significant differences
in thermal pain responses between wild-type (n58; white bars) and GRPR
mutant mice (n57; black bars); Student’s t-test; P.0.05. d, Sensitivity to
paw withdrawal threshold on exposure to von Frey filaments was
comparable to wild-type mice (n58; white bars); Student’s t-test; P.0.05.
e, Spontaneous pain responses in the first (0–10min) and second phase
(10–60min) of the formalin test are comparable between wild-type (n58;
white bars) and GRPR mutant mice (n59; black bars); Student’s t-test;
P.0.05. f, Locomotor activity measured by ambulations distance is
comparablebetween wild-type (n58; open circles) and GRPR mutantmice
(n57; filled circles); repeated measures ANOVA (ANOVA); P.0.05. All
data are presented as means6s.e.m. Error bars represent s.e.m.
0510 15 20 25 30 35
0510 15 20 25 30 35
0510 15 20 25 30 35
Number of scratches
Figure 3 | Scratching behaviour is reduced in GRPR mutant mice. a, The
scratching behaviour induced by intradermal injection of compound 48/80
(100mg per 50ml) is significantly decreased in GRPR mutant mice (n517;
filled circles) compared with wild-type mice (n514; open circles); repeated
measures ANOVA (ANOVA); *P,0.05. b, The scratching behaviour
50ml) in wild-type (n511; open circles) and GRPR mutant mice (n511;
filled circles) shows GRPR mutant mice have a severely blunted response;
repeated measures ANOVA (ANOVA); *P,0.05. c, The scratching
behaviour induced by intradermal injection of chloroquine (200mg per
50ml) in wild-type (n511; open circles) and GRPR mutant mice (n511;
filled circles) shows GRPR mutant mice have a significantly reduced
response; repeated measures ANOVA (ANOVA); **P,0.01. All data are
presented as means6s.e.m. Error bars represent s.e.m.
NATURE|Vol 448|9 August 2007
such as trypsin and tryptase14,15. PAR2 mediates itch in human skin,
and PAR2 agonist (SLIGRL-NH2) is a histamine-independent itch
inducer16,17. In wild-type mice, intradermal injection of SLIGRL-
NH2 (100mg per 50ml) induced robust scratching behaviour within
30min. In contrast, the number of scratches decreased significantly
quine model to assess scratching behaviour in GRPR mutant
mice18,19. The scratching behaviour induced by intradermal injection
of chloroquine (200mg per 50ml) was significantly reduced in GRPR
mutant mice compared with wild-type mice (P,0.01, Fig. 3c). The
reduction in the number of scratches was not due to a defect in
locomotor activity because there was no significant difference in
spontaneous locomotor activity between wild-type and GRPR
differ significantly between GRPR mutant and wild-type mice after
intradermal injection of vehicle (data not shown). These results sug-
gest that GRPR is a receptor that is important to the transmission of
itch information in the spinal cord circuits.
We postulated that if GRPR mediates itch sensation, then activa-
tion of GRPR at the spinal level may initiate the transmission of itch
signals,which wouldsubsequently resultinscratching behaviour. To
test this hypothesis, we intrathecally administered the GRPR agonist
GRP18-27(ref. 20) and found that GRP18-27indeed induced scratch-
ing behaviour in a dose-dependent manner (Fig. 4a, Supplementary
Fig. 7). At these doses, intrathecal GRP18-27did not affect the pain
sensitivity testedbytail-flick assay(datanotshown). WhenGRP18-27
was injected into GRPR mutant mice intrathecally, the level of
scratching behaviour was significantly lower compared to wild-type
antagonist21(10min before injection of GRP18-27) significantly
inhibited GRP18-27-induced scratching behaviour (Fig. 4b), whereas
this antagonist itself did not induce any scratching behaviour (data
not shown). Finally, we intrathecally injected the GRPR antagonist
10min before intradermal injection of each of the pruritogenic
ing behaviour (Fig. 4c, d, e). In contrast, the algesiogenic agent
(formalin)-induced scratching behaviour was not significantly affec-
ted by the GRPR antagonist (Supplementary Fig. 6b). Thus, our
results suggest that GRP acts through the spinal GRPR to mediate
the transmission of information that drives scratching behaviour in
response to pruritogenic stimuli only.
Taken together, our data provide genetic, pharmacological and
behavioural evidence supporting the hypothesis that GRPR is a spe-
cific itch-signalling molecule in the spinal cord. Although itch and
pain are intimately related and may share similar peripheral and
central mechanisms2,22,23, we believe that the reduced scratching
behaviour of GRPR mutant mice reflects a defect in itching, rather
than a defect in pain sensation, for several reasons. First, our results
showed that pain sensitivities were not compromised in GRPR
mutant mice. This is also supported by the observation that GRPR
mutant mice showed normal response to foot shock24. Second, the
scratching behaviour responding to the pruritogenic agents was sig-
nificantly reduced in GRPR mutant mice, but remained normal in
activation of the GRPR receptor at the spinal level induced dose-
dependent scratching behaviour without affecting pain behaviour.
One interesting observation is that GRPR mutant mice showed dif-
ferential sensitivity in different itch models: the reduced scratching
behaviours of GRPR mutant mice are much more notable in PAR2
and chloroquine models than in the compound 48/80 model. Unlike
compound 48/80, PAR2 agonist SLIGRL-NH2 and chloroquine are
believed to act through a histamine-independent mechanism17,18.
through distinct types of central mediators in the spinal cord in
GRP18-27 + saline
GRP18-27 + antagonist
Number of scratches
Number of scratches in 30 min
Number of scratches in 30 min
SalineAntagonist Saline AntagonistSaline Antagonist
Figure 4 | EffectsofGRPRagonistandantagonistonscratchingbehaviour.
a, The scratching behaviour is monitored for 30min after intrathecal
dependentscratching behaviourinwild-typemice (n57; opencircles).The
number of scratches was markedly reduced in GRPR mutant mice (n57;
filled circles) when compared to wild-type littermate controls; repeated
measures ANOVA (ANOVA); **P,0.01. b, The effect of GRP18-27
(1.0nmol per 2.5ml) was significantly blocked by intrathecal injection of
GRPR antagonist (D-Phe-6-Bn(6-13)OMe, 1.0nmol per 2.5ml, n58; filled
circles) compared with vehicle (2.5ml, n58; open circles); repeated
measures ANOVA (ANOVA); *P,0.05. c–e, Compared with the control
group (white bars), intrathecal injection of the GRPR antagonist (1.0nmol,
black bars) significantly inhibited scratching behaviour induced by
intradermal injection of compound 48/80, PAR2 agonist SLIGRL-NH2, or
chloroquine (n510, 9 and 10 for compound 48/80, SLIGRL-NH2, and the
chloroquine model, respectively). Student’s t-test comparing genotypes
within treatment; *P,0.05, **P,0.01. All data are presented as
means6s.e.m. Error bars represent s.e.m.
NATURE|Vol 448|9 August 2007
response to distinct types of pruritogenic stimuli. Alternatively,
GRPR signalling may be differentially recruited in response to differ-
ent types of pruritogenic stimuli. It is also possible that other bom-
besin-like receptors in the spinal cord may produce a compensatory
effect in GRPR mutant mice25.Although our results suggest that
GRPR is specifically involved in itch not pain sensation, the possibil-
ity that GRPR-expressing neurons are involved in pain sensation can
not be excluded, given the fact that the selective histamine-sensitive
neurons also respond to capsaicin11,12. Nevertheless, our results
strongly suggest the existence of an itch-specific GRPR-mediated
molecular signalling pathway in the dorsal horn of the spinal cord.
Our study should have important clinical implications for the treat-
ment of pruritus. For example, chloroquine has been used to treat
malaria fever in Africa, but it often induced intractable pruritus in
these patients26. GRP/GRPR signalling pathway may provide a cent-
ral target for therapeutic treatment of chronic and intractable prur-
itus, without compromising the ability to sense acute noxious
stimulus, which itself would be detrimental to the patient.
the behaviour experiments27. Mice were shaved at the back of the neck where
intradermalinjectionswerethengiven. Hindlimbscratchingbehaviour directed
intervals17. Pain behaviours were assessed as described previously28. Locomotor
activity of mice was evaluated by an open-field test over a 1-h period in trans-
parent (48325321cm) polystyrene enclosures as previously described29.
Immunocytochemical staining and in situ hybridization. Immunocyto-
chemical staining and in situ hybridization were performed as previously
described28. The following primary antibodies were used: rabbit anti-GRP
(Immunostar), guinea pig anti-CGRP (Peninsula), mouse anti-peripherin
afferents were identified with IB4–biotin (Vector Laboratories) staining and
were visualized with avidin coupled to FITC (Vector Laboratories). For dou-
ble-staining, secondary antibodies (Jackson ImmunoResearch), including don-
to FITC for CGRP, and donkey anti-mouse IgG coupled to FITC for NF200 or
peripherin were used.
Data analysis. Statistical comparisons were performed with two-way ANOVA
(ANOVA) and Student’s t-test. All data are expressed as the mean6standard
error of the mean (s.e.m.) and error bars represent s.e.m. P,0.05 was consid-
ered statistically significant.
Full Methods and any associated references are available in the online version of
the paper at www.nature.com/nature.
Received 4 August 2006; accepted 18 June 2007.
Published online 25 July 2007.
1. Paus, R., Schmelz, M., Biro, T. & Steinhoff, M. Frontiers in pruritus research:
scratching the brain for more effective itch therapy. J. Clin. Invest. 116, 1174–1186
Ikoma, A., Steinhoff, M., Stander, S., Yosipovitch, G. & Schmelz, M. The
neurobiology of itch. Nature Rev. Neurosci. 7, 535–547 (2006).
Greaves, M. W. Itch in systemic disease: therapeutic options. Dermatol. Ther. 18,
Roesler, R., Henriques, J. A. & Schwartsmann, G. Gastrin-releasing peptide
receptor as a molecular target for psychiatric and neurological disorders. CNS
Neurol. Disord. Drug Targets 5, 197–204 (2006).
Battey, J. & Wada, E. Two distinct receptor subtypes for mammalian bombesin-
like peptides. Trends Neurosci. 14, 524–528 (1991).
Cowan, A., Khunawat, P., Zhu, X. Z. & Gmerek, D. E. Effects of bombesin on
behavior. Life Sci. 37, 135–145 (1985).
7. Gmerek, D. E. & Cowan, A. Studies on bombesin-induced grooming in rats.
Peptides 4, 907–913 (1983).
Gmerek, D. E. & Cowan, A. Bombesin—a central mediator of pruritus? Br. J.
Dermatol. 109, 239 (1983).
J. Neurosci. 4, 2956–2962 (1984).
10. Moody, T. W. & Merali, Z. Bombesin-like peptides and associated receptors
within the brain: distribution and behavioral implications. Peptides 25, 511–520
11. Schmelz, M., Schmidt, R., Bickel, A., Handwerker, H. O. & Torebjork, H. E. Specific
C-receptors for itch in human skin. J. Neurosci. 17, 8003–8008 (1997).
12. Andrew, D. & Craig, A. D. Spinothalamic lamina I neurons selectively sensitive to
histamine: a central neural pathway for itch. Nature Neurosci. 4, 72–77 (2001).
13. Kuraishi, Y., Nagasawa, T., Hayashi, K. & Satoh, M. Scratching behavior induced
by pruritogenic but not algesiogenic agents in mice. Eur. J. Pharmacol. 275,
14. Nystedt, S., Emilsson, K., Wahlestedt, C. & Sundelin, J. Molecular cloning of a
potential proteinase activated receptor. Proc. Natl Acad. Sci. USA 91, 9208–9212
15. Steinhoff, M. et al. Agonists of proteinase-activated receptor 2 induce
inflammation by a neurogenic mechanism. Nature Med. 6, 151–158 (2000).
16. Steinhoff, M. et al. Proteinase-activated receptor-2 mediates itch: a novel
pathway for pruritus in human skin. J. Neurosci. 23, 6176–6180 (2003).
17. Shimada,S.G.,Shimada,K.A.&Collins,J.G.Scratching behaviorinmiceinduced
by the proteinase-activated receptor-2 agonist, SLIGRL-NH2. Eur. J. Pharmacol.
530, 281–283 (2006).
18. Inan, S. & Cowan, A. Kappa opioid agonists suppress chloroquine-induced
scratching in mice. Eur. J. Pharmacol. 502, 233–237 (2004).
19. Green, A. D., Young, K. K., Lehto, S. G., Smith, S. B. & Mogil, J. S. Influence of
genotype, dose and sex on pruritogen-induced scratching behavior in the mouse.
Pain 124, 50–58 (2006).
20. Ladenheim, E. E., Taylor, J. E., Coy, D. H., Moore, K. A. & Moran, T. H. Hindbrain
GRP receptor blockade antagonizes feeding suppression by peripherally
administered GRP. Am. J. Physiol. 271, R180–R184 (1996).
21. Wang, L. H. et al. des-Met carboxyl-terminally modified analogues of bombesin
function as potent bombesin receptor antagonists, partial agonists, or agonists.
J. Biol. Chem. 265, 15695–15703 (1990).
22. Stander,S.&Schmelz, M.Chronicitchandpain—similaritiesanddifferences.Eur.
J. Pain 10, 473–478 (2006).
23. McMahon, S. B. & Koltzenburg, M. Itching for an explanation. Trends Neurosci. 15,
24. Shumyatsky, G. P. et al. Identification of a signaling network in lateral nucleus of
111, 905–918 (2002).
25. Kroog, G. S., Jensen, R. T. & Battey, J. F. Mammalian bombesin receptors. Med.
Res. Rev. 15, 389–417 (1995).
26. Sowunmi, A., Walker, O. & Salako, L. A. Pruritus and antimalarial drugs in
Africans. Lancet 2, 213 (1989).
27. Hampton, L. L. et al. Loss of bombesin-induced feeding suppression in gastrin-
28. Chen, Z. F. et al. The paired homeodomain protein DRG11 is required for the
projection of cutaneous sensory afferent fibers to the dorsal spinal cord. Neuron
31, 59–73 (2001).
29. Wozniak, D. F.et al.Apoptotic neurodegeneration induced by ethanol inneonatal
mice is associated with profound learning/memory deficits in juveniles followed
by progressive functional recovery in adults. Neurobiol. Dis. 17, 403–414 (2004).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank J. Yin, C.-S. Qiu and K.-H. Zhang for technical
assistance and J. Battey for providing GRPR mutant mice. We are grateful to
A. Basbaum, E. Carstens, L. Hampton and J. Battey for critical comments on the
manuscript. We also thank D. H. Coy for providing the GRPR antagonist. The work
was supported by an NIH RO1 to Z.F.C.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Correspondence and requests for materials should be addressed to Z.F.C.
NATURE|Vol 448|9 August 2007
Animals. GRPR mutant mice and littermate wild-type mice were used in the
behaviour experiments27. Mice were backcrossed to C57BL/6J strain for more
antagonist experiment. Male mice between 8 and 12 weeks of age were acclima-
tized to the experimental room and were used for experiments. All behavioural
All the experiments were performed in accordance with the guidelines of the
National Institutes of Health and the International Association for the Study of
Pain and were approved by the Animal Studies Committee at the Washington
University School of Medicine.
Drugs. GRP18-27(0.03, 0.1, 1.0nmol, Bachem), and the GRPR antagonist,
D-Phe-6-Bn(6-13)OMe, (1.0nmol, kindly provided by D. H. Coy) were dis-
solved in sterile saline and administered intrathecally with a volume of 5ml,
unless specified. Compound 48/80 (100mg, Sigma-Aldrich), chloroquine
(200mg, Sigma-Aldrich), and PAR2 agonist, H-Ser-Leu-Ile-Gly-Arg-Leu-NH2,
(SLIGRL-NH2, 100mg, Bachem) were dissolved in sterile saline and adminis-
tered by intradermal injection at a volume of 50ml.
Itch behaviour tests. Prior to experiments, mice were given 30min to acclimate
to a small plastic chamber (15326312 cm). For intradermal injections of
drugs, mice were briefly removed from the chamber and intradermally injected
with drugs. Hindlimb scratching behaviour directed towards the shaved area at
the back of the neck was observed for 30min at 5-min intervals. One scratch is
between those two movements17.
Pain behaviours. Pain behaviours were assessed as described previously28.
Thermal sensitivity was determined using hotplate (48, 52, or 56uC), paw-flick
For the hotplate test, the latency for the mouse to lick its hindpaw or jump was
recorded. For the Hargreaves test, thermal sensitivity was measured using a
Hargreaves-type apparatus (IITC Inc.). The latency for the mouse to withdraw
from the heat source was recorded. For the water immersion tail-flick test, tails
were dipped beneath the water in a temperature-controlled water bath (IITC
Inc.). The latency to withdrawal was measured with a 10-s cutoff. Mechanical
sensitivity was assessed using a set of calibrated von Frey filaments (Touch-Test
kit, Stoelting). Each filament was applied 5 consecutive times and the smallest
filamentthat evoked reflexive flinchesofthe pawon3 of the5 trialswastaken as
alin test by intraplantar injection of formalin (Sigma, 10ml of 5% formalin in
saline) into the plantar surface of the right hindpaw. The total time spent in
licking and flinching of the injected paw was monitored for 60min.
into the lumbar region of unanaesthetized mice were performed as described
previously30. For the GRPR agonist experiment, different doses of GRP18-27or
GRPR antagonist was injected 10min before either intrathecal injection of
GRP18-27or intradermal injection of the pruritogenic agents, and the number
of scratching responses was counted for 30min at 5-min intervals after the
injection of the pruritogenic agents or GRP18-27.
30. Hylden, J. L. & Wilcox, G. L. Intrathecal morphine in mice: a new technique. Eur. J.
Pharmacol. 67, 313–316 (1980).