THE JOURNAL OF COMPARATIVE NEUROLOGY 367:342-351 (lY96)

Localisation of NK1 Receptor

Immunoreactivity to Neurons and
Interstitial Cells of the Guinea-pig
Gastrointestinal Tract
A.L. PORTBURY, J.B. FURNESS, H.M. YOUNG, B.R. SOUTHWELL, AND S.R. VlGNA
Department of Anatomy and Cell Biology, University of Melbourne, Australia, 3052
(A.L.P., J.B.F., H.M.Y., B.R.S.) and Department of Cell Biology,
Duke University Medical Centre, Durham, North Carolina 27710 (S.R.V.)

ABSTRACT
Tachykinins, including substance P, neurokinin A, and neuropeptides K and y, are expressed
widely in the peripheral nervous system where they affect smooth muscle contraction, exocrine
gland secretion, vascular permeability, and neurotransmission. Substance P, the preferred
ligand for the NK1 receptor, is found in high concentrations in the enteric nervous system. In
the present study, the localisation and distribution of the NK1 receptor was studied throughout
the gastrointestinal tract of the guinea-pig by using a polyclonal antiserum raised against the
C-terminal 15 amino acids of the NKI receptor. Co-localisation with other neuronal markers
was examined in the ileum. Nerve cell bodies reactive for the NK1 receptor were found in the
myenteric plexus of all regions and the submucous plexus of the small and large intestines. In
the small intestine, the interstitial cells of Cajal were also immunoreactive. Immunoreactivity
was largely confined to cell surfaces. Almost all immunoreactive myenteric nerve cells had
Dogie1 type I morphology, and most of these were immunoreactive for nitric oxide synthase, a
transmitter of inhibitory neurons to the muscle and of descending interneurons. Neuropeptide
Y-containing secretomotor neurons in the submucous and myenteric plexuses also exhibited
NK1 receptor immunoreactivity. NK1 receptors were present on a minority of tachykinin
immunoreactive neurons of submucous ganglia. The results suggest that receptors on the
longitudinal muscle might not be conventional NK1 receptors, that excitation of the circular
muscle of the ileum is indirect, perhaps via the interstitial cells of Cajal, and that enteric
inhibitory neurons may be excited via NK1 receptors. z I!)'~w\Viky-Liss;, Inc.

Indexing terms: enteric nervous system, intestinal motility, substance P, tarhykinin receptor,
immunohistochemistry

Tachykinins are a group of structurally related peptides
which share the common C-terminal sequence, Phe-X-Gly-
Leu-Met-NH?, and includes substance P, neurokinin A
(NKA), and neurokinin B (NKB). Three receptors for
tachykinins have been identified on the basis of agonist and
antagonist potencies, namely neurokinin 1 (NKl), NK2,
and NK3 (Guard and Watson, 1991; Nakanishi, 1991;
Maggi et al., 1993; Mussap et al., 1993; Regoli et al., 1994).
More recently, the structures of each of the receptors have
been determined and each has been shown to be a G
protein-coupled receptor of the 7 transmembrane spanning
class (Sasai and Nakanishi, 1989; Yokota et al., 1989;
Hershey and Krause, 1990; Shigemoto et al., 1990). Knowl-
edge of the amino acid sequences of receptors permits the
generation of specific antibodies which can be used in
immunohistochemical studies to determine the localisation
of the receptors (Moussaoui et al., 1992; V i s a et al., 1994).
For example, antibodies to the NKI receptor stain the
plasma membrane of rat kidney epithelial cells expressing
the native NK1 receptor as well as neurons of the dorsal
horn of the rat spinal cord and ganglion cells in the
myenteric plexus of the rat ileum (Moussaoui et al., 1992;
Vigna et al., 1994; Sternini et al., 1995).
Although preparations of intestine, particularly the il-
eum of the guinea-pig, have been used in numerous pharma-
cological assays of tachykinin analogue activities, it has
been difficult to determine the physiological sites of action
of tachykinins, which receptors are on different cell types,
Accepted Novernher 5,1995.
Addrcss reprint requests to A.L. Portbury, Department of Anatomy and
Cell Riology,University of Melbourne, Parkvillr. 3052, Victoria, Australia.

o 1996 WILEY-LISS, INC.

NK1 RECEPTORS IN THE INTESTINE
or even whether there may be other receptor types, such as
the septide-preferring receptor, in the gut (Mag@ et al.,
1993; Burcher and Stamatakos, 1994).
Immunoreactivity for substance P, the preferred ligand
for the NK1 receptor, is found throughout the guinea-pig
ileum in a variety of myenteric nerve cells, including
putative sensory neurons, ascending interneurons, circular
muscle excitatory motor neurons, and longitudinal muscle
excitatory motor neurons (Costa et al., 1992). The tachyki-
nins appear to act as excitatory transmitters to both the
circular and longitudinal muscle layers and also to affect
motility indirectly via activation of enteric neurons (Bartho
et al., 1982; Bartho and Holzer, 1985; Costa et al., 1985;
Holzer, 1989; Maggi et al., 1993). Thus, the effects of
tachykinins on motility of the guinea-pig ileum are partly
neurogenic and partly direct (Maggi et al., 1990). In the
circular muscle of the guinea-pig ileum, NK1 and NK2
receptors are thought to mediate the direct response of the
muscle cells while NK3 receptors activate motor neurons
which release acetylcholine and possibly endogenous tachy-
kinins to contract the muscle (Bartho et al., 1982;Maggi et
al.; 1990; Maggi et al.; 1994). The longitudinal and circular
muscle both appear to contain a novel, septide-preferring
binding site (Chassaing et al., 1992; Petitet et al., 1992;
Patacchini et al., 1993).Substance P and related tachykinins
also enhance fluid secretion across the intestinal mucosa
(Brown and Miller, 1991).In the guinea-pig small intestine,
the secretory effect has a direct component, and an indirect,
nerve-mediated component, implying that tachykinin recep-
tors are on both mucosal epithelial cells and neurons of
secretomotor pathways iKeast et al.,1985; Perdue et al.,
1987).
Although these studies indicate that tachykinin receptors
are present in the gastrointestinal tract, they do not provide
direct information concerning the cellular localisation of
the receptors. In the present study, a polyclonal antiserum,
raised against the C-terminal 15 amino acids of the NK1
receptor from rat brain, has been used to study the
localisation of the NK1 receptor throughout the guinea-pig
gastrointestinal tract. Double staining has been used to
identify the nerve cell types that have NKl receptor
immunoreactivity in the small intestine.
MATERIALS AND METHODS
Tissue was obtained from guinea-pigs of both sexes, in
the weight range 200- 400 g , which were stunned by a blow
to the head and killed by severing the carotid arteries. All
procedures have been approved by the University of Mel-
bourne Animal Experimentation Ethics Committee.
Tissue preparation
Wholemounts. Fresh tissue was removed and placed in
phosphate-buffered saline (PBS; 0.9% NaCl in 0.01 M
sodium phosphate buffer, pH 7.01, containing the muscle
relaxant nicardipine M, Sigma Chemical Co. St.
Louis, MOj. Segments of intestine were cut along the
mesenteric border, opened out and the contents flushed
away with PBS. The tissue was stretched tautly, pinned on
balsa board, mucosal surface facing down, and immersed in
Zamboni’s fixative (2% formaldehyde and 0.2% picric acid
in 0.1 M sodium phosphate buffer, pH 7.0) for 24 hours at
4°C. Following fixation, the tissue was cleared of fixative
with 3 x 10 minute washes in dimethylsulphoxide (DMSO),
followed by 3 x 10 minute washes in YBS. Wholemount
343
preparations of longitudinal muscle and myenteric plexus
as well as submucous plexus were prepared.
Frozen sections. Animals were anaesthetised with SO-
dium pentobarbitone (15 mg/kg subcutaneously),and intra-
muscular fentanyl (0.16 mgikg) plus fluanisone (5.0 mg/
kg) and perfused transcardially with Zamboni’s fixative.
Segments of tissue from the stomach, pancreas, duodenum,
ileum, caecum, and proximal and distal colon were removed
and placed in Zamboni’s fixative overnight at 4°C.Tissue
was cleared in DMSO as described above and placed in
PBS-sucrose-azide (PBS containing 0.1%’ sodium azide and
30% sucrose as a cryoprotectant) and stored at 4°C over-
night. The following day, small segments of tissue were
transferred to a mixture of PBS-sucrose-azide and OCT
compound (Tissue Tek, Elkhart, IN) in a ratio of 1:l for a
further 24 hours before being embedded in 100% OCT.
Sections of 10 pm thickness were cut and collected on slides
coated with aminopropyltriethoxy-silane (AES; Sigma
Chemical Co., St. Louis, MO) and left to dry for 1 hour at
room temperature. In the pancreas, neurons are found in
interlobular ganglia. To ensure that sections of pancreas
that were processed for immunohistochemistry contained
neurons, serial sections were taken through the tissue and
every second section stained with toluidine Blue. Only
sections of pancreas containing neurons were processed for
immunohistochemistry .
Seriaithin sections. Tissue that was to be examined as
semithin sections was processed as above for wholemounts
and incubated in anti-NK1 receptor (NKlr) antiserum and
the staining revealed using the DAB reaction. Following
this, wholemount preparations were dehydrated and embed-
ded flat in Araldite resin and 1 bm sections were cut
longtudinally, in the plane of the myenteric plexus, to
reveal immunoreactive cells.
NK1 receptor immunohistochemistry
Tissue sections or wholemounts were incubated in 10%
normal sheep serum in PBS for 1hour at room temperature
prior to exposure to primary antisera. The primary antise-
rum to NKlr was raised in a rabbit against the C-terminal
15 amino acids of the rat NK1 receptor (antiserum code
11886.5, see Vigna et al., 1994, for further characterisation
of this antiserum) and was diluted at 1:1,000 for whole-
mounts and 1:2,000for sections in buffered PBS-azide, pH
7.0. Tissue was incubated for 48 hours at room tempera-
ture, and then excess serum was washed off with 3 x 10
minute changes in PBS. Tissue was then incubated in goat
anti-rabbit IgG coupled to biotin (Vector Laboratories,
Burlingame, CA) at a dilution of 1 2 0 0 for 2 hours. Follow-
ing incubation in the secondary immunoglobulin, the tissue
was washed in 3 x 10 minute changes of PBS and incubated
in avidin-biotin horseradish-peroxidase complex (Vector
Laboratories, Burlingame, CA). The horseradish peroxi-
dase was revealed by the diaminobenzidine (DAB) reaction.
Double staining
Following fixation and clearing, wholemounts of myen-
teric and submucous plexuses were incubated in 10%
blocking serum, of the same species in which the secondary
antibody had been raised, for 1 hour at room temperature.
Excess serum was removed from the tissue which was
subsequently incubated in mixtures of primary antisera,
diluted appropriately (see below), for 48 hours at room
temperature, after which excess antisera were washed from
the wholemounts with 3 x 10 minute changes in PBS.
344
Tissue was then incubated in a series of secondary antibod-
ies to obtain differential labelling of each primary antibody
on the tissue. To ensure that no cross reactivity occurred,
one or other primary antiserum was omitted in control
tests. No cross reactivity was observed.
Double staining carried out in the myenteric plexus
involved antisera against the following combinations of
antigens: NKlr plus nitric oxide synthase (NOS), NKlr
plus neuropeptide Y (NPY), and NKlr plus a mixture of
NOS and NPY. Double staining carried out in the submu-
cous plexus involved the following combinations: NKlr plus
NPY, and NKlr plus vasoactive intestinal peptide (VIP)
and NKlr plus tachykinin (TK).
NKlr plus NOS irnmunoreactivity. The primary anti-
NOS serum was a sheep polyclonal raised against the full
length sequence of neuronal NOS (code 205, a gift from Dr.
P.C. Emson) and was incubated at 12,000 with the anti-
NKlr antiserum (1:1,000)in PBS. Tissue was subsequently
incubated in donkey anti-rabbit IgG coupled to biotin (RPN
1004; Amersham, Melbourne, Australia), diluted at 1:200
in PBS for 2 hours a t room temperature. Tissue was then
washed with 3 x 10 minute changes of PBS and subse-
quently incubated in a mixture of streptavidin coupled to
Texas Red (RPN 1233; Amersham, Melbourne, Australia)
diluted at 150, plus donkey anti-sheep antibody coupled to
FITC (code 713-095-147; Jackson ImmunoResearch, West
Grove, PA) diluted at 1:150 for 90 minutes at room
temperature.
NKlr plus NPY irnrnunoreactiuity. The primary anti-
NPY serum was a sheep polyclonal (code E2210, see
Furness et al., 1985) and was diluted a t 1:800 and mixed
with the anti-NKlr antiserum (1:lOOO) in PBS. Subsequent
labelling of the tissue to reveal both antisera was as
described above for the NKlr plus NOS combination.
NKlrplus NOS and NPY. This combination of antisera
was used to determine if the NOS and NPY subgroups of
NKlr immunoreactive neurons accounted for all the NKlr
neurons in the myenteric plexus. The sera used were as
described above, and the second labelling systems were also
as outlined above.
NKlr plus VIP. The primary antibody against VIP was
a mouse monoclonal (code 55, see McConalogue et al., 1995)
and was used at 1:2,000 with the anti-NKlr antiserum
(1:1,000)in PBS. The tissue was then incubated in donkey
anti-rabbit IgG coupled to biotin (RPN 1004; Amersham,
Melbourne, Australia), diluted at 1:200 in PBS, for 2 hours
at room temperature. Tissue was then incubated in horse
anti-mouse coupled to AMCA (code 8104; Vector Laborato-
ries, Burlingame, CA) at a dilution of 1 5 0 , plus streptavidin
coupled to Texas Red (RPN 1233; Amersham, Melbourne,
Australia) at a dilution of 1 5 0 in PBS for 1 hour at room
temperature.
NKlr plus TK. The primary antibody used was a rat
monoclonal raised against substance P (code NC1/34HL;
see Cue110 et a]., 1979).This antibody was characterised by
radioimmunoassay for binding to synthetic substance P, but
cross-reactivity with other tachykinins was not determined.
Therefore, in this study this antibody is referred to as
tachykinin (TK) antibody. The TK antibody was diluted a t
1:200 with the anti-NKlr antiserum (1:1,000) in PBS. The
tissue was then incubated in donkey anti-rabbit I& coupled
to biotin (code BlOOO; Vector Laboratories, Burlingame,
CA) diluted at 1 2 0 0 in PBS for 2 hours at room tempera-
ture. Tissue was then incubated in donkey anti-rat IgG
coupled to FITC (code 27674; Jackson ImmunoResearch,
West Grove, PA) diluted at 1 : l O O in PBS plus streptavidin
A.L. PORTBURY ET AL.
coupled to Texas Red (RPN 1233; Amersham, Melbourne,
Australia) at a dilution of 1:50 in PBS for 1 hour at room
temperature.
RESULTS
Distribution of NKl receptor
immunoreactivity throughout the guinea-pig
gastrointestinal tract
NKlr immunoreactivity was examined in wholemounts
and sections from the stomach, duodenum, ileum, caecum,
proximal and distal colon, and in sections of the pancreas.
NKlr immunoreactive nerve cell bodies were found in the
myenteric plexus of all gut resons examined (Figs. IA-E,
2A-El. The highest frequency of occurrence of NKlr
immunoreactive nerve cells was in the duodenum and
ileum and the lowest was in the proximal colon. Nerve cell
bodies were stained in the submucous plexus (Fig. 2B) of all
regions except the gastric corpus, a regon where submu-
cous ganglia are rare or absent. The intensity of staining of
submucous neurons was usually greater than that observed
in the myenteric plexus. No stained nerve terminals were
observed in either the muscle layers or the ganglia (Figs.
2A,I3,G, 3A) and no staining was observed in association
with blood vessels in the submucous plexus. In sections, a
dense line of immunoreactivity was seen in the small
intestine at the level of the deep muscular plexus (Fig. 2G).
In wholemounts, it was possible to discern the shapes of
individual cells in this layer; they conform in appearance to
the interstitial cells of Cqjal (Fig. 2H). In the duodenum,
NKlr immunoreactivity was also observed in cells resem-
bling the interstitial cells of Cajal at the level of the
myenteric plexus (Fig. 1D). In the pancreas, no immunore-
activity was localised to nerve cell bodies. The longitudinal
and circular muscle layers were not stained in any of the
regions examined (Fig. 2G).
Nerve cell morphology and pattern of NKlr
immunoreactivily in the ileum
Most NKlr immunoreactive nerve cell bodies in the
myenteric plexus had Dogie1type I morphology with numer-
ous lamellar dendrites protruding from the cell body (Figs.
BA,C,D,E, 3A, 4A,B). Immunoreactive cells appeared to
have only one axon. Although the initial segments of the
axons were immunoreactive for NK1 receptors, this ex-
tended only as far as approximately 4 cell body diameters
(Fig. 3B) and, as mentioned, there was no staining of
terminals in either the ganglia or the muscle. A small
proportion of nerve cells had the morphology of secretomo-
tor neurons (see below) with filamentous dendrites that
often arose predominantly from one end of the neuron
(Figs. 3B, 4A). At the other end, each of these cell bodies
had a very prominent and intensely immunoreactive axon
(Fig. 3B). Semi thin (1.0 pm) sections through stained
neurons and confocal microscopy revealed that the staining
was mostly confined to the cell surface (Figs. 2D,F, 3A,A”).
The surface staining was concentrated in small patches
around the cell perimeter (Fig. 2F). The staining intensity
of cell bodies was equivalent to that of the dendrites and the
initial parts of axons.
Co-localisation studies: ileum
Co-localisation of NKlr immunoreactivity with immuno-
reactivity for NOS, NPY, VIP, and TK was examined.
Colocalisation with NOS immunoreactivity was examined
NKl RECEPTORS IN THE INTESTINE:
Fig. 1. S K l r immunoreactivity in nerve cells in various reginna of'
the guinea-pig gastrointestinal tract. A Wholemount preparation of a
myenteric ganglion in the corpus of the stomach showing NKlr
immunoreactive nerve cell bodies. Cells show Dogiel type I morphology
and immunorcactivity is largely confined to the cell surface membrane.
B,C: Wholemount preparations of myenteric ganglia in the antrum of
the stomach. Staining appears particularly punctate in this region with
large aggregations of immunoreactivity dotting the cell surface mem-
because most myenteric NKlr immunoreactive cells were
similar in shape to NOS cells (Furness et al., 1994). NPY
immunoreactivity was examined because a small propor-
tion of the myenteric and many submucous NKlr immuno-
reactive cells exhibited the morphological characteristics of
the secretornotor neurons which are NPY immunoreactive.
NPY immunoreactive secretornotor neurons have filamen-
tous dendrites that often arise predominantly from one end
of the neuron and at the other end these cell bodies have a
prominent and anally projecting axon (Furness et al.,
1985). NPY immunoreactivity, TK immunoreactivity, and
VIP immunoreactivity were used in the submucous plexus
to identify the NKlr immunoreactive cells found there.
These three classes of neurons make up nearly 90% of
neurons found in the submucous plexus (Bornstein and
Furness, 1988).
Myenteric plexus. Based on counts of 899 nerve cells
*
from four animals, 63 t 7% (mean standard deviation) of
NKlr immunoreactive cells in the myenteric plexus of the
small intestine were also immunoreactive for NOS (Figs.
3A,A',A", 4h,A',B,B'). These cells were all Dogiel type I
cells. About one third of NKlr immunoreactive cells were
not double labelled (Fig. 4A,A') and these appeared to be
closer to the longitudinal muscle than nerve cells that were
double labelled with NOS. Moreover, 24 t 7% (666 nerve
cells in four animals) of NOS immunoreactive neurons were
not double labelled for NKlr.
Every NPY immunoreactive secretornotor neuron that
was examined was also NKlr immunoreactive (39 nerve
345
brane. D: Wholemount preparation of a myenteric ganglion in the
duodenum illustrating immunoreactive nerve cells (arrow heads) and
interstitial cells of Cajal (small arrows) at the level of the myenteric
ganglia. The nerve cell bodies exhibit Dugiel type I morphology with
flattened lamellar dendrites and a prominant axon. E: Wholemount
preparation of the myenteric ganglia of the distal colon showing three
NK1 receptor immunoreactive neurons. Scale bars, 20 bm for A,R,C,E;
50 )*m for D.
cells) (Fig. 3B,B'), accounting for approximately 7% of all
NKlr cells in the myenteric plexus of the small intestine.
Along with the secretomotor neurons, there was a very
small number of Dogiel type I cells with faint NPY immuno-
reactivity that were also stained for NKlr. To determine if
all NKlr immunoreactive neurons contained either NOS or
NPY, tissue was incubated in antisera against all three
substances (NOS, NKlr, and NPY). It was found that only
72 2 8% (787 nerve cells) of cells wit,hNKlr immunoreactiv-
ity were also labelled for NOS or NPY.
Submucous plexus. Of NKlr irnmunoreactive neurons
in the submucous plexus 72 '-c 6% (478 nerve cells in four
animals) were also immunoreactive for NPY (Fig. 4C,C'i.
All of the NPY immunoreactive neurons (292 nerve cells)
also contained NKlr immunoreactivlty. There was no
overlap between NKlr and VIP staining of neurons in the
submucous plexus (400 nerve cells with immunoreactivity
for NKlr); 3.5 ? 3% of NKlr immunoreactive submucous
neurons were also immunoreactive for TK (545 nerve cells),
while 23 2 6% of TK cells were immunoreactive for NKlr
(131 nerve cellsj (Fig. 4D,D').
DISCUSSION
Receptors on muscle
This study revealed NK1 receptors on neurons, but not
on smooth muscle, throughout the guinea-pig gastrointesti-
nal tract. Immunoreactivity also occurred on the intersti-
 Fig. 2. NKlr immunoreactivity in nerve cells and interstitial cells in
the ileum. A Wholemount preparation of a myenteric ganglion showing
NKlr immunoreactive nerve cell bodies. The majority have Dogie1 I
morphology with flattened lamellar dendrites (arrows).B: Wholemount
preparation of submucous ganglia. NKlr immunoreactive nerve cell
bodies are evenly distributed within each ganglion, but no nerve fibres
show NKlr immunoreactivity. C-F: Examples of the appearance of
nerve cells immunoreactive for NKlr. C: High magnification fluores-
cence micrograph of a cell body with prominent lamellar dendrites
(arrows). D: Confocal microscope image (projection of 28 optical
sections, 0.3 pm z steps) of an immunoreactive neuron showing the
surface distribution of immunoreactivity (arrow heads) and numerous
lamellar dendrites. E: Light micrograph of an immunoreactive neuron
after processing using avidin-biotin HRP and the DAB reaction. This
neuron also has prominent lamellar dendrites (arrows).F: A 1pm resin
section through the cell bodies of two NKlr immunoreactive neurons
demonstrating the punctate surface staining (arrow heads). G: Confo-
cal microscope image o f a frozen section through the wall of the small
intestine. An immunoreactive ganglion cell (asterisk) is present in the
myenteric plexus (mp) and there is a dense layer of immunoreactivity at
the level ofthe deep muscular plexus (icc), where the interstitial cells of
Cajal are found. Note the absence of nerve fibres and terminals in the
circular muscle layer icm). H: Wholemount preparation of circular
muscle. Interstitial cells on the surface of the muscle apposed to the
submucous plexus are immunoreactive for NKlr and form a highly
branching network but muscle cells in the same field of view are
unreactive. Scale bars, 50 pm for A,B; 20 pm for C,G,H; 10 pm for
D,E,F.
KKI RECEPTORS IN THE INTESTINE
Fig. 3. Confocal microscope optical sections of neurons in t h e
myenteric plexus showing the colocalisation of N K l r immunoreactivity
with XOS immunoreactivity ( A - A 1 and NPY immunoreactivity (B-
E' 1. A: Single image of 0.5 IJ-moptical thickness of a N K l r immunoreac-
tive nerve cell body illustrating the surface staining observed when
using this antiserum. The nerve cell has Dogie1 I morphology. A
lamellar dendrite emerges from the right hand side of the cell. There
are no terminals with N K l r immunoreactivity on or around the nerve
cell body. A': The same cell also shows immunoreactivity for NOS.
There are numerous NOS immunorcactive terminals around the cell.
A : A compositc image showing N K l r and NOS iminunoreactivities.
347
The different distribution of the two substances is clearly evident with
N K l r immunoreactivity confined to the cell perimeter and KOS
immunoreactivity labelling thc cytoplasm of the cell body, but not t h e
nucleus. B,B': Projection of 15 optical sections (0.5 IJ-mz steps) showing
a N K l r immunoreactive nerve cell body (B, asterisk) which is also
immunoreactive for NPY [B', asterisk). N K l r immunoreactivity is
concentrated around the cell body as well as in discrete patches on its
surface. Note that nnly the initial part of the axon is immunoreactive
for N K l r . The same neuron shows NPY immunoreactivity and t h e axon
is traceable to the edge of t h e section. The bright varicose fibre in this
field of view is a n axon from a neighbouring cell. Scale bars, 20 IJ-m.
348
Fig. 4. Paired micrographs showing the colocalisation of NKlr
immunoreactivity with NOS, NPY and TK in wholemount preparations
of mycnteric and submucous ganglia. A,A'; B,B': In the myenteric
plexus, many nerve cells had both NKlr and NOS immunoreactivity.
However, the overlap was not complete. A,A': Two NKlr immunoreac-
tive cells have NOS immunoreactivity but a third (arrow)is not reactive
tial cells of Cajal at the level of the deep muscular plexus in
the small intestine. In the ileum, the lack of detectable
receptors on the muscle contrasts with pharmacological
evidence that the muscle of the guinea-pig small intestine is
contracted by tachykinins acting at NK1 receptors on the
muscle itself. Although there is only one known gene for the
NK1 receptor, results from pharmacological studies indi-
cate that there may be a number of receptor subtypes
(Maggi et al., 1993). Differential splicing could lead to tho
generation of NK1 receptors with different pharmacological
A.L. PORTBURY ET AL.
for NOS. B!R': A nerve cell reactive for both antigens and an example of
a NOS immunoreactim neuron iarrow) that does not show NKlr
immunorcactivity. C,C': In the siihmucous plexus, all of the NPY
immunoreactive neurons also showed NKlr immunoreactivity. D,D': A
TK immunoreactive neuron in the submucous plexus showing NKlr
immunoreactivity.Scale bars, 20 pm for A,A',B,B': 10 p m for C,C',D,D'.
and antigenic properties. A splice variant of the human
NKlr with a truncated C-terminal has been identified
(Fong et al., 1992). This truncation, after amino acid 311,
would render the receptor unreactive t o the antiserum used
in the present work, which was raised against amino acids
393-407. Kage et al. (1993) have also identified a shortened
N K l receptor in rat, which they deduce to have a C-
terminal truncation of 60 to 70 amino acids. It is also
feasible that post-translational modification, e.g., sulfation,
could alter the antigenicity of the receptor.
N K l RECEPTORS IN THE INTESTINE
In the lon,gitudinal muscle of the guinea-pig small intes-
tine, excitatory receptors, detected pharmacologically or in
ligand binding studies, were initially deduced to be N K l
receptors, but more recent studies suggest that these differ
from NK1 receptors elsewhere. Petitet et al. (19923 found
that two presumed NKl receptor agonists, septide and
[Progl-SPwere equally potent in the longitudinal muscle,
but septide was about 200 times less potent than [Pro"l-SP
in displacing [Pro9]-SPbinding. Moreover, the NK1 recep-
tor antagonist, GR 71,251, was over ten times as effective in
antagonising septide-induced contractions, compared to its
antagonism of [Prog]-SP-inducedcontractions. Thus, it is
possible that the dominant receptor type is a septide-
preferring receptor, distinct from the established NKl
receptor, and that GR 71,251 is an effective antagonist of
the septide receptor. Other pharmacological and ligand
binding data also point to the presence of a novel receptor
on the longitudinal muscle that differs from NK1 receptors
elsewhere (Chassaing et al., 1992; Ireland et al., 1992;
Mussap and Burcher, 1994). The lack of immunohistochemi-
cal localisation of NK1 receptors in the longitudinal muscle
in this study supports the contention that these muscle
receptors are a subtype of NK1 receptors.
In the circular muscle of the guinea-pig small intestine,
direct contractile effects appear to be mediated through
both NK1 (LPro9]-SPactivated) and NK2 receptors (acti-
vated by p-alaR-NKA-4-10and antagonised by MEN 10207;
Maggi et al., 1990). The present work suggests that contrac-
tions evoked by NK1 receptors may be a consequence of the
activation of interstitial cells of Cajal at the inner surface of
the circular muscle, which, in contrast to the circular mus-
cle, were strongly iminunoreactive for N K l receptors. The
autoradiographic localisation of tachykinin binding sites in
dog small intestine are consistent with the present observa-
tions; Mantyh et al. (1988) found that the NK1 ligand,
Bolton Hunter Substance P, bound to sites forming a thin
band at the inner surface of the circular muscle, but not to
the bulk of the muscle. The resolution of the autoradiogra-
phy was not sufficient to identify the cell type to which the
label bound, but the location of the label coincides with that
of the interstitial cells. N K l r immunoreactivity is also
found on interstitial cells at the inner surface of the circular
muscle in the rat ileum (Vigna et al., 1994; Sternini et al.,
1995). The interstitial cells are linked to the muscle via
tight junctions and appear to drive contractile activity in
the muscle (Hara et al., 1986; Rarajas-Lopez et al., 1989;
Langton et a]., 1989; Ward and Sanders, 1990; Ward et al.,
1994). They may in fact be a specialised type of smooth
muscle cell (Zhou and Komuro, 1992; Torihashi et al.,
1993). It is notable that the circular muscle is less sensitive
to tachykinins after the mucosal layer is removed; this loss
of sensitivity could result from injury or inadvertent removal
of interstitial cells at the inner surface of the muscle (Bauer
and Kuriyama, 1982; Costa et al., 1985; Maggi et al., 1990).
The muscle in other regions of the guinea-piggastrointes-
tinal tract also has excitatory receptors for tachykinins.
However, in only a few instances have the receptor types
been pharmacologically classified. In the duodenum and
proximal colon, such experiments point to the existence of
both NK1 and NK2 receptors o n the muscle (Mag@et al.,
1994; Zagorodnyuk et al., 1995). In the proximal colon, the
relative apparent affinities of septide and LSarg]-SPsulphone
suggest that a subtype of the NK1 receptor is present, being
perhaps a septide-preferring receptor (Maggi et al., 1994).
Thus, the lack of staining of muscle throughout the gastro-
349
intestinal tract may be because the muscle has a NKlr
subtype that is not recognised by the antiserum used in the
present study.
Receptors on enteric neurons
Although NK3 receptors are the type most commonly
detected on myenteric neurons (Laufer et al., 1985; Guard
and Watson, 19871, functional N K l receptors also appear to
be present. Guard et al. (19911 reported transmitter release
from myenteric neurons in response to NK1 receptor
activation and Burcher and Stamatakos (1994) found that
septide stimulates inhibitory neurons, but it was not deter-
mined whether this was due to the agonist action of septide
at NK1 receptors or a t septide preferring receptors of the
type found in the muscle. The present study revealed NK1
receptors on NOS immunoreactive nerve cells in the guinea-
pig ileum. NOS is known to be contained in enteric inhihi-
tory neurons ofthe guinea-pig intestine (Costa et al., 1992),
so the localisation of NK1 receptors is consistent with the
pharmacological observations of Burcher and Stamatakos
(1994). Our results show that 76%, of NOS neurons were
also immunoreactive for the NK1 receptor. Since 19%of all
myenteric nerve cell bodies in the guinea-pig ileum contain
NOS (Furness et al., 19941, we estimate that 14% of all
myenteric neurons show immunoreactivity for both NOS
and the NKI receptor. However, the results of this study
also demonstrated that NOS immunoreactive neurons com-
prised only 63%)of all NKI receptor immunoreactive neu-
rons. Thus, neurons expressing the NK1 receptor comprise
approximately 23% of all myenteric nerve cell bodies in the
guinea-pig ileum.
In submucous ganglia of the ileum, N K l receptors were
found on all NPY immunoreactive neurons, which are
known to be cholinergic secretomotor neurons, as well as on
a minority of substance P neurons, which are probably
sensory (Bornstein and Furness, 19883, but they were not
found on the VIP immunoreactive, non-cholinergx secreto-
motor neurons. NPY immunoreactive neurons comprise
approximately 30%of all submucous neurons in the guinea-
pig ileum (Bornstein and Furness, 1988). Since only 72% of
the NK1 receptor immunoreactive neurons also showed
immunoreactivity for NPY, it can be estimated that 42% of
submucous nerve cell bodies in the guinea-pig ileum ex-
press the NK1 receptor. NK1 receptors have also been
located on submucous nerve cells by Burcher and Bornstein
(1988) and SP has been shown to stimulate secretomotor
neurons of the guinea-pig small intestine (Keast et al.,
1985). Tachykinin containing nerve terminals surround
submucous nerve cells, so it is possible that substance P
contributes to excitatory transmission to the neurons via
NK1 receptors.
In conclusion, the immunohistochemical localisation of
NK1 receptors assists the interpretation of pharmacologi-
cal and ligand binding studies. The localisation supports
recent pharmacological and binding studies that suggest
that tachykinin receptors on longitudinal muscle are a sub-
type of NK1 receptors; it suggests that the circular muscle
may be indirectly excited via interstitial cells of Cajal; and it
confirms that NK1 receptors are present on different
subpopulations of myenteric and submucous nerve cells.
ACKNOWLEDGMENTS
This work was supported by a grant from the NH&MRC
ofAustralia. We thank Dr. Elizabeth Burcher for her expert
350
advice on the manuscript. We thank Dr. P.C. Emson for
kindly providing the NOS antiserum.
LITERATURE CITED
Barajas-Lbpcz, C., I. Rerezin, E.E. Daniel. and J.D. Huizinga 11989)
Pacemaker activity recorded in interstitial cells of Cajal of the gastroin-
testinal tract. Am. J. Physiol. 257:C830-835.
Barthb, P.. P Holzer, J. Donnerer, and F.Lembeck 119821 Evidence for the
involvement of substance P in the atropine-resistant peristalsis of the
guinea-pig ileum. Neurosci. Lett. .72:69-74.
Bartho, L., and P . Holzer (1985) Search for aphysiological role of substance
P i n gastrointestinal motility. Neuroscience 16:l-32.
Bauer, V.. and H. Kuriyania 119821 The nature of non-cholinergic. non-
adrenergic transmission in longitirdinal and circular muscles of the
guinea-pig ileum. J . Physiol. [London, 332:375-391.
Bornstein, J.C., and tJ.B. Furness (1988) Correlated electrophysiological and
histochemical studies of submucous neurons and their contribution t o
understanding enteric neural circuits. J. Auton. Kcrv. Syst. 25.1-1 3.
Brown, D.R., and R.J. Miller (1991) Neurohormonal control of' Huid and
electrolyte transport in intesl.ina1 rnucosa. In M. Firld and R.A. FrizLell
(eds]:Handbook of Physiology: Thc Gastrointestinal System IV. Wash-
ington: American Physiologrul Society; pp. 517-589.
Burcher, E., and J.C. Bornstein (1988) Localizatiun of substance P binding
sites in submucous plexus of guinea-pig ileum using whole-mount
auto-radiography. Synapse 2232-239.
Burcher. E., and C. Stamatakos (1994) Septide but not substance k'
stimulates inhihitory neurons in guinea-pig ileum. Eur. J. Pharmacol.
258:RS-RlO.
Chassaing, G., S. Lavielle, A. Brunissen, A. Carruette, C. Garret, F. Petitet,
J.C. Beaujouan, Y. Torrens, and J . Glowinski (1992) [Pro91SP and
[pGlu". P r o W P (6-11) interact with two different receptors in the
guinea-pig ileum as demonstrated with new SP antagonists. Neuropep-
tides 23:73-79.
Costa, M., S.J.H. Brookes, S. Waterman, and R. Mayo 11992) Enteric
neuronal circuitry and transmitters controlling intestinal motor func-
tion. In G.E. Holle cedi: Advances in the 1nnervat.ion of the Gastrointes-
tinal Tract. Amsterdam: Elsevier Science Publishers, pp. 115-121.
Costa, M., J.B. Furness. C.O. Pullin, and d. Bornstein 119%) Substance P
enteric neurons mediate non-cholinergic transmission to the circular
muscle of the guinea-pig intestine. Naunyn-Schniiedeherg's Arch. Phar-
macol. 328:446-453.
Cuello. A.C., G. Galfre. and C. Milstein (1979) Detection of' substance P in
the central nervous system by a monoclonal antibody. Proc. Natl. Acad.
Sci. U S A . 76:3532-3536.
Fnng, T.M., S.A. Anderson, H. Yu, R.R.C. Huang, and C.D. Strader (1992)
Differential activation of intracellular effector by two isufornrs nf human
neurokinin-1 receptor. Mol. Pharniacol. 42.24-30.
Furness, J.B., M. Costa, I.L. Cibbins, I.J. Llewellyii-Smith, and J.R. Oliver
(1985) Neurocheniically similar inyenteric and submucous neurons
directly traced to the mucosa of the small intestine. Cell Tissue Res.
241:155-163.
Furness, J.B., Z.S.Li, H.M. Young. and U. Forsterniann (1994) Nitric oxide
syuthase in the pnteric nervous system of the guinea-pig: A quantitative
description. Cell Tissue Res. 277:139-149.
Guard, S., A.T. McKnight, K.J. Watling, and S.P. Watson t1991) Evidence
for two types of tachykinin receptors on cholincrgic neurnns iJf the
guinea-pigmyenteric plexus. Ann. N.Y. Acad. Sci. 632:4U0-403.
Guard, S.. and S.P.Watson (1987) Evidence for neurokinin-3 receptor-
mediated tachykinin release in the guinea-pig ileum. Eur. J . Pharmacol.
I44:409412.
Guard, S., and S.P. Watson (1991) 'I'achykinin receptor types. Classification
and membrane signalling mechanisms. Neurochem. Intern. Itl:149-165.
Hara, Y., M. Kubota, and J.H. Szurszewski (1986) Electrophysiology of
smooth muscle of the small intestine of some mammals. J. Physiol.
372:601-520.
Hershey, AD.. and J.E. Krause (1990) Molecular characterization of a
functional cDNA encoding the rat substance P receptor. ScienceZ47.958-
962.
llolzer. P. 11989) Ascending enteric reflex: Multiple neurotransmitter
systcms and interactions. Am. J. Physiol. 256:540-545
Ireland, S.J., I.J.M. Beresford, and J . Stahles (19921 Cornparison of the
apparent allinity of GR 82, 334 at NKI receptors in various tissue
prrparations. Ar. J. Pharmac. 706.93P.
A.L. PORTBURY ET AL.
Kage, K., S.E. Leeman, and N.D. Sovd 11993) Biochemical characterization
of two dilferent forms of the substance P receptor in rat suhmaxillary
gland. J. Neurochem. 60:347-351.
Keast, J.R., J.B. Furness, and M. Costa 11985) Different substance P
receptors are found on mucosal epithelial cells and submucous neurons
of the guinea-pig small intestine. Naunyn-Sclimiedeherg's Arch. Pharma-
col. .329:382-387.
Langton, P., S.M. Ward. A. Carl, M.A. Norell, and K.M. Sanders (1989)
Spontaneous electrical activity of interhtial cells of Cajal isolated from
canine proximal colon. Proc. Natl. Acad. Sci. U.S.A. #6:7280-7284.
Laufer, R., L. Wormser, Z.Y. Friedman, C. Gilon, M. Chorcv, and Z. Selinger
(1 985) Keurolrinin R i s t i preferred agonist for a neuronal substance P
receptor and its action is antagonized by enkephalin. Proc. Natl Acad.
Sci. U.S.A. 82.74447448,
Maggi, C.A., R. Patacchini, A. Giachetti. and A. Meli 11990) Tachykinin
receptors in the circular muscle of the ,qiine:i-pig ileum. Rr. J. Pharma-
col. 101:Y96-1000.
Maggi, C.A., R. Patacchini, S. Mcini, and S. Giuliani 11993) Nitric Oxide is
the mediator of tachykinin NK:, receptor-induced relaxation in the
circular muscle ofthe guinea-pig ileum. Eur. J. Pharmacol. 240:45-50
Maggi, C.A.. R. Patacchini, S. Meini. L. Quartara, A. Sisto, E. Potier, S.
Giulinni, and 4 . Giachetti (1994) Cornparison of tachykinin NK, and
NK2 receptors in the circular mu or the guinea-pig ileuni and
proximal colorr. Br. .J. Pharmacol. 112:150-160.
Maggi,C.A., R. Patacchini, P. Rovero, and A. Giachetti 11993) Tachykinin
receptors and tachykinin receptor ant.agon ists. ,J. Auton. Pharmacol.
13:83-93.
Mantyh. P.W.. C.R. Mantyh, T. Gates, S.R.Vigna, and J.E. Maggin 11988)
Rrceptor hinding sites for substance P and substance K in the canine
gastrointestinal tract and their pnssihle role in inflammatory bowel
disease. Neuroscience 3817-837.
McConalogue, K., J.B. Furness, M.A. Vremec, J . J . Holst, K. Tornge, and
P.D. Marley (19951 Histochemical, pharmacologd, biochemical and
chromiitiigraphic evidence that pituitar); adenylyl cyclase activating
peptide is involved in inhihitoq neurotransmission in the taenia of the
guinea-pig caecum. J. Autonom. New Sys. 50:311- 322.
Moussaoui, S.M.. E. IIermans. B. Mathieu, B. Bonici, F. Clerc, F. Guinet, C.
Garret, and P.M. Laduron (1992: Polyclonal antibodies against the rat
NK, receptor. Characterization and localization in the spinal cord.
Neurti. Rep. 3:1073-1076.
Mussap, C.J., and E. Burcher (1994) Binding sites for L12i11-Bolton-Hunter
scyliorhinin 11in guinea-pig ileum: A radioligand binding, functional and
autoradiographic study. Naiinyn-Schiniedeherg's Arch. Pharmacol. 350:
301-309.
Mussap, C J . , U.P Geraghty, and E.Burcher (1993) Tdchykinin receptors: 4
radioligand binding perspective. J . Neurochem. 60.1987-2009.
Nakanishi, S. (1991) Mammalian tachykinin receptors. Annu. Rev. Neuro-
sci. 74:123-136.
Patacchini. R., S. Meini, S. Giuhdni, and C.A. Maggi (19931 Evidence! for the
presence of a septide-sensitive tachykinin receptor in the circular muscle
of the guinea-pig ileum Neuropeptides 24.231.
Perdue, M.H., R. Galbraith, and J.S. Davison (1987) Evidence for substance
P a s a functional ncurotransmitter in guinea pig small intestinal mucosa.
Regul. Pept. 18:63-74.
Petitet, F., M.Saffroy, Y. Torrens, S. Lavielle, G. Chassaing, L). Loeuillet, J.
Glowinski. arid J.C. Beaujouan 11992) Possible existence of a uew
tachykinin receptor subtype in the guinea-pig ileum. Pcptides 13383-
388.
Regoli, D., A. Boudon, and J.-L. Fauchere (19941 Receptors and antagonists
for substance P and related peptides. Pharmacol. Rev. 46:551-599.
Sasai, Y., and S. Nakanishi (1989, Molecular characterization of rat sub-
stance K receptor and its mRNAs. Biochem. Biophys. Res. Commun.
I 65695-702.
Shigemoto, R., Y. Yokota, K. 'I'suchida, and S. Nakanishi (1990) Cloningand
expression of' a rat neuromedin K receptor cDNA. J. B i d Chern.
265623428,
Sterriini, C., D. Su, P.D. Ganip. and N.W. Bunnet (1995) Cellular sites of
expression of the neurokinin-1 rrceplor in thc rat gastrointestinal tract.
J Comp. Nrurol. 2.58:531 -540.
Torihashi, S., S Kobayashi, W.T. Gerthoffer, and K.M. Sanders (1993)
Interstitial cells in deep muscular plexus of canine small intestine may be
specialized smooth muscle cells. Am. J. Physiol. 265:G638-G645.
Vigna, S.R., J.J. Bowden. U.M. McDonald, J. Fisher, A. Okaniiitn, D.C.
McVey, D.G. Payan, and N.W. Bunnett 11994) Characterization of
NK1 RECEPTORS IN THE INTESTINE
antibodies to the rat substance P (NK-11 receptor and to a chimeric
substance P receptor expressed in mammalian cells. J. Ncurosci. 14:834-
845.
Ward, S.M., A.J. Burns. S. 'Torihashi, and K.M. Sanders (1994) Mutation of
the proto-oncogene c-kit blocks development of interstitial cells and
electrical rhylhmicity in murine intestine. J. Physiol. (Lnndon) 480:91-
97.
Ward, S.M., and K.M. Sanders (1990) Paccmaker activity in septa1 struc-
tures of canine colonic circular rnusclc. Am. J. F'hysiol. 259:G264-273.
Yokota, Y.. Y. Sasai, K. 'Tanaka, T. Fujiwara, K. Tsuchida, R. Shigemoto, A.
351
Kakizuka, H. Ohkubo. and S. Nakanishi (1989)Molecular characteriza-
tion of a functional cDNA for rat substancc I' receptor. J. Biol. Chem.
264:17649-17652.
Zagorodnyuk. V., P. Santicioli, C.A. Maggi, and A. Giachetti (1995) Evidence
that tachykinin NK1 and NK2 receptors mediate non-adrenergic, non-
choliiiergic excitation and contraction in the circular muscle of the
guinea-pigduodcnum. Ur. J. Pharmacol. 115237-246.
Z h o u , D . 3 , and T. Komuro (19921 Interstitial cells associated with the deep
muscular plexus of the guinea-pig small int,estine, with special rcfcrcnce
to the interstitial rclls of'Cajal. Cell Tissue Res. 268905-216.