Department of Neuroscience (D.R.L., V.S.S.) and Division of
Neuroscience Research in Psychiatry (M.J.C., E.E.E.), University of
Minnesota, Minneapolis, Minnesota
 |
Introduction |
The
neurokinin3 (NK3) receptor
is a member of the neurokinin family of G protein-coupled receptors
that also includes NK1 and
NK2 receptors. The major endogenous mammalian
ligands for these receptors are substance P, neurokinin A, and
neurokinin B (NKB). The endogenous ligand that has the highest affinity
for the NK3 receptor is NKB. Although substance P
and neurokinin A bind to the NK3 receptor (Guard
et al., 1989
; Sadowski et al., 1993), they exhibit higher affinity for
the NK1 and NK2 receptors, respectively.
NK3 receptors are distributed widely throughout
the body. Within the nervous system, NK3
receptors are expressed by neurons at all levels of the brain and
spinal cord (Ding et al., 1996), as well as neurons in the enteric
nervous system (Guard et al., 1989; Mann et al., 1997). Several other
tissues, including vascular smooth muscle (Mastrangelo et al., 1986),
kidney (Buell et al., 1992), and iris sphincter muscle (Medhurst et
al., 1997), also express the NK3 receptor.
NK3 receptors couple most efficiently to the
Gq/11 family of G proteins. Activation of this G
protein family initiates an intracellular signaling pathway that
includes the formation of inositol-1,4,5-trisphosphate
(IP3) and diacylglycerol via the activation of
phospholipase C, and the release of calcium from intracellular
stores. Although the formation of inositol phosphates (Guard et al.,
1988; Krause et al., 1997) and the mobilization of intracellular
calcium (Pinnock et al., 1994) have been observed after the activation
of NK3 receptors, intracellular effector systems
further downstream have not been described to date.
In several tissues, the activation of NK3
receptors results in physiological effects that are blocked with the
inhibition of nitric-oxide (NO) synthase (NOS), suggesting the
involvement of NO. Via this mechanism, NO has been shown to contribute
to NK3 receptor agonist-induced relaxation of rat
vascular smooth muscle (Mizuta et al., 1995) and guinea pig colon (Jin
et al., 1993; Maggi et al., 1993a), as well as the induction of
thermal hyperalgesia in the rat spinal cord (Linden and Seybold, 1999). Coupling between NK3 receptors and NOS within the
same cell, however, is not likely within the gastrointestinal tract
because NOS is not contained in enteric neurons that express
NK3 receptors (Mann et al., 1997). Furthermore,
tetrodotoxin and vasoactive intestinal peptide receptor antagonists
block the production of NO and longitudinal muscle relaxation induced
by NK3 receptor activation (Jin et al., 1993).
Therefore, NK3 receptor-mediated production of NO
may occur transneuronally in the gastrointestinal tract.
In the superficial dorsal horn of the rat spinal cord, however, the
neuronal isoform of NOS (nNOS) is cocontained in neurons that express
NK3 receptors (Seybold et al., 1997). These
morphological data increase the likelihood that the production of NO
after NK3 receptor activation in the spinal cord
may occur within single cells. The occurrence of three endogenous
ligands is also important to the physiology of spinal
NK3 receptors. Although substance P and
neurokinin A have 10- to 100-fold lower potency at
NK3 receptors than NKB (Buell et al., 1992;
Nakajima et al., 1992), the release of substance P and neurokinin A
from primary afferent neurons is greatly increased during persistent
inflammation (Hope et al., 1990; Schaible et al., 1990). The increased
concentration of lower-affinity agonists may result in the activation
of spinal NK3 receptors under pathophysiological
conditions. Thus, the efficiency of NK3 receptor
coupling to effector systems may determine the extent to which other
endogenous ligands exert physiological effects via
NK3 receptors.
We used a mammalian cell line that stably expressed both the
NK3 receptor and nNOS to investigate the
hypothesis that NK3 receptors can couple
intracellularly to the activation of NOS and to explore the efficiency
of coupling of NK3 receptors to the generation of
intracellular messengers. NO production was detected directly, by
measuring the formation of L-citrulline, another product of
NOS, as well as indirectly, by measuring cGMP production in a cell line
that naturally expresses soluble guanylyl cyclase. The data presented
here support the hypothesis that NK3 receptors
couple to the production of NO through the activation of nNOS within
the same cell.
| |
Experimental Procedures |
Cell Culture.
Chinese hamster ovary (CHO) cells stably
expressing human NK3 receptors (CHO
hNK3; Krause et al., 1997) were kindly supplied by Dr. James Krause. These cells were stably transfected using DEAE/dextran methods (Promega, Madison, WI) with the gene encoding the
rat isoform of nNOS (received as a gift from Drs. David Bredt and
Solomon Snyder, Johns Hopkins University, Baltimore, MD) that was
subcloned into the pREP4 vector (InVitrogen, Carlsbad, CA). Transfected
cells (CHO hNK3/nNOS) were grown in minimum
essential medium-
(Life Technologies, Gaithersburg, MD) supplemented
with 10% heat-inactivated fetal bovine serum (Life Technologies), 50 µg/ml geneticin (Life Technologies), and 50 µg/ml hygromycin B (Calbiochem, La Jolla, CA).
Rat lung fibroblast (RFL-6) cells were grown in Dulbecco's modified
Eagle's medium (Life Technologies) supplemented with 10% bovine calf
serum (Hyclone, Logan, UT), 100 U/ml penicillin (Life Technologies),
and 100 µg/ml streptomycin (Life Technologies). RFL-6 and CHO
hNK3/nNOS cell lines were grown in
75-cm2 polystyrene flasks (Sarsteadt, Newton, NC)
at 37°C in an atmosphere of 5% CO2 and were
used for experiments when they reached confluency. Generally,
experiments were conducted 4 days after cells were subcultured. All
experimental protocols were conducted with the cells incubated in a
HEPES-based buffer (110 mM NaCl, 5.4 mM KCl, 1.8 mM
CaCl2, 1.0 mM MgSO4, 20 mM
HEPES, 25 mM glucose, and 60 mM sucrose, pH 7.4, 335-340 mOsM).
Immunocytochemistry on CHO Cells.
Stable transfection of CHO
hNK3 cells with the gene encoding nNOS was
confirmed by immunocytochemical detection of the NOS protein. CHO
hNK3/nNOS cells were cultured on glass
coverslips in minimum essential medium-. Two days after
plating, the cells on the glass coverslip were rinsed with PBS (0.05 M,
pH 7.35), fixed with 4% paraformaldehyde in PBS for 30 min at room
temperature, and again rinsed with PBS. Cultures were then incubated
for 45 min at room temperature with PBS containing 0.1% Triton X-100, 10% normal goat serum, and 0.1% sodium azide. Excess solution was
removed, and the cultures were incubated for 45 min at room temperature
with a rabbit anti-nNOS antibody (1:100, R-20; Santa Cruz
Biotechnology, Santa Cruz, CA) in PBS containing Triton-X, normal goat
serum, and sodium azide. Some glass coverslips were incubated under the
same conditions with antibody that had been preincubated with the
immunizing peptide (0.02 mg/ml) for 4 h. After three 5-min washes
with PBS, cultures were incubated with Alexa 488-labeled goat
anti-rabbit antiserum (1:300; Molecular Probes, Eugene, OR) for 45 min
at room temperature. After three 5-min washes with PBS to remove
unbound antibodies, glass coverslips were inverted on slides over a
drop of glycerol:PBS (5:1, v/v) containing 0.1%
p-phenylenediamine to preserve the fluorophores. Immunoreactivity was detected with a microscope equipped for the visualization of fluorescence.
Measurement of Total Inositol Phosphates.
Total inositol
phosphates in CHO hNK3/nNOS cells were measured
as an estimate of the formation of IP3 (Berridge
et al., 1983). Experimental protocols used in our laboratory have been
described previously (Parsons et al., 1995). Briefly, the pool of
phosphoinositide in CHO hNK3/nNOS cells was
labeled by adding 2.5 µCi/ml
myo-[3H]inositol (Amersham,
Arlington Heights, IL) to the culture medium 16 h before the
experiments were initiated. At the time of the experiment, CHO
hNK3/nNOS cells were harvested, washed twice, resuspended in HEPES buffer with 10 mM LiCl, and added to 24-well plates at a density of 5 × 105 cells/well.
Test compounds were added directly to each well, and cells were
incubated with the agonist for 30 min at 37°C. After the incubation
period, cells were lysed with perchloric acid, and inositol phosphates
in the samples were separated from free inositol by anion exchange
chromatography (Berridge et al., 1983) using columns of AG1-X8 resin
(100-200 mesh, formate form; Bio-Rad, Hercules, CA).
[14C]Inositol 1-phosphate was added to each
sample to assess recovery of inositol phosphates during anion exchange
chromatography. The amount of radioactivity in each sample was
determined with a Beckman LS 6500 scintillation counter (Fullerton,
CA). Data were corrected for recovery of inositol phosphates and
expressed as fold-increase over the radioactivity due to
3H-labeled inositol phosphates obtained in
corresponding control samples within each experiment. The average
recovery of [14C]inositol 1-phosphate was
76 ± 4% (n = 3 experiments), and the average
amount of radioactivity due to 3H-labeled
inositol phosphates in control samples was 2063 ± 289 dpm
(n = 3 experiments).
Measurement of Concentration of Intracellular Calcium
([Ca2+]i).
[Ca2+]i was measured in
suspensions of CHO hNK3/nNOS cells with the
fluorescent calcium indicator dye Fura-2 (Grynkiewicz et al., 1985).
Protocols used in this study have been described previously (Wotta et
al., 1998). Briefly, CHO hNK3/nNOS cells were
harvested and incubated with 5 µM Fura-2/acetoxymethyl ester
(Calbiochem), 0.4% Pluronic F127 (Calbiochem), 1 mM probenecid, and
0.02 mg/ml BSA at 37°C for 30 min. The cells were then rinsed in
calcium-free HEPES buffer containing 1 mM probenecid and centrifuged,
and the cell pellet was resuspended at a concentration of 1 × 106 cells/ml in calcium-free HEPES buffer
containing 1 mM probenecid. CaCl2 was added to a
final concentration of 1.8 mM, and the cells were allowed to
equilibrate for at least 5 min before recordings began. Measurements of
fluorescence were made in a Perkin-Elmer Cetus model LS50B fluorescence
spectrophotometer (Beaconsfield, UK) using excitation wavelengths of
340 and 380 nm and an emission wavelength of 510 nm. Cells were kept in
suspension at 37°C with continuous stirring. Test compounds, diluted
in HEPES buffer, were added directly to the cell suspension to the
final concentrations indicated in the text. For each recording, maximal
and minimal fluorescence values were determined using 60 µg/ml
digitonin and 10 mM EGTA, respectively. Autofluorescence for each
wavelength was measured in a suspension of cells that was not loaded
with Fura-2. Data were collected using FLWinLab software provided by the equipment manufacturer and were analyzed off-line.
[Ca2+]i was also measured
in single CHO hNK3/nNOS cells using a
microfluorometer (Photoscan; Photon Technology International, South Brunswick, NJ) to monitor the emission of the
Ca2+-sensitive fluorescent chelator Indo-1
(Grynkiewicz et al., 1985). Cells were harvested 4 days before
subculture and plated at a density of 1 × 104 cells/3 ml onto glass coverslips. The cells
were allowed to adhere to the glass for at least 2 h. They were
then washed with HEPES buffer and incubated with 3 µM
Indo-1/acetoxymethyl ester in HEPES buffer containing 2% BSA for 45 to
60 min at 37°C in an atmosphere of 5% CO2.
Protocols for calcium imaging of single cells in our laboratory have
been described previously (Stucky et al., 1996). Measurements of
fluorescence of Indo-1 were made using the excitation wavelength of 360 nm and emission wavelengths of 405 and 485 nm. Test compounds were
diluted to final concentrations indicated in HEPES buffer, and the
buffer was superfused over the glass coverslip at a rate of
approximately 1.2 ml/min at 37°C. For each recording, maximal and
minimal fluorescence values were determined using 3 µM ionomycin and
10 mM EGTA, respectively. After recording from each cell, the
background fluorescence at each wavelength was measured in a field of
the same dimensions that contained no cells or debris. Because
superfusion of the glass coverslip with test compounds stimulated every
cell, subsequent recordings would have been biased. Therefore, only one
cell was studied per glass coverslip.
For both experimental protocols,
[Ca2+]i values were
calculated from the equation:
[Ca2+]i = KD
(R 
Rmin)/(Rmax R) (Grynkiewicz et al., 1985), where the dissociation
constant (KD) of the calcium-dye
complex used for Fura-2 was 225 nM and that for Indo-1 was 250 nM
(Grynkiewicz et al., 1985). The value for was determined as the
ratio of the fluorescence in the absence and presence of a saturating
concentration of Ca2+ at 380 and 485 nm for use
with Fura-2 and Indo-1, respectfully. R was the excitation
fluorescence ratio of 340 nm/380 nm corrected for autofluorescence for
Fura-2 and the emission fluorescence ratio of 405 nm/485 nm corrected
for background fluorescence for Indo-1.
Rmin and
Rmax are the R values
determined in the absence and presence, respectively, of a saturating
concentration of calcium.
Measurement of L-[3H]Citrulline
Formation.
The activity of nNOS was measured by monitoring the
conversion of L-[3H]arginine to
L-[3H]citrulline. Procedures for
this assay have been described previously (Wotta et al., 1998).
Briefly, CHO hNK3/nNOS cells were harvested, washed twice, resuspended in HEPES buffer, and added to 12- × 75-mm tubes (5 × 105 cells/tube).
L-[2,3,4,5-3H]Arginine
monohydrochloride (0.3 µCi/tube; Amersham) was added to each tube,
followed immediately by various concentrations of agonists. After a 1-h
incubation, the formation of
L-[3H]citrulline was inhibited by
the addition of a solution that yielded final concentrations of 357 mM
unlabeled L-arginine and 71 mM EDTA. The samples were
centrifuged, and the supernatant was aspirated to remove excess
L-[3H]arginine. The cell pellets
were lysed with 0.3 M perchloric acid and neutralized with the addition
of 0.15 M potassium carbonate. L-[14C]Citrulline was added to each
sample so we could monitor the recovery of
L-[3H]citrulline during anion
exchange chromatography.
L-[3H]Citrulline in the suspension
was separated from L-arginine by anion exchange
chromatography using columns of AG50W-X8 resin (100-200 mesh, hydrogen
form; Bio-Rad). The amount of radioactivity in each sample was
determined with a Beckman LS 6500 scintillation counter. Data are
expressed as a percentage of the radioactivity due to
L-[3H]citrulline obtained in
response to a standard concentration of
MePhe7-NKB within each experiment. The average
recovery of L-[14C]citrulline was
86 ± 3% (n = 10 experiments).
Measurement of cGMP Formation.
Because CHO cells lack
guanylyl cyclase, the assay required the use of a donor-detector cell
protocol. This paradigm involves the generation of NO in one cell
(donor) with activation of guanylyl cyclase in another cell (detector)
by diffusible NO. Procedures for this assay have been described
previously (Wotta et al., 1998). Briefly, CHO
hNK3/nNOS cells were harvested, washed twice,
resuspended in HEPES buffer, and added (5 × 105 cells/well) to harvested, washed, and
resuspended NO detector RFL-6 cells (2 × 105 cells/well, unless otherwise indicated) on
24-well plates. The plates were incubated at 37°C in a shaking water
bath (45 rpm) for the duration of the experiment. Test compounds were
added directly to each well as described in Results. cGMP
was measured by radioimmunoassay. The samples and cGMP standards were
diluted with H2O and acetylated by the addition
of triethylamine:acetic anhydride (2:1; 2 µl to 100-µl sample).
125I-cGMP (final concentration, 15 pM; NEN Life
Science Products, Boston, MA) and antiserum generated to cGMP
(1:240,000; a generous gift of Dr. Thomas Gettys, Duke University,
Durham, NC) were added to acetylated samples and standards in an NaAC
buffer, pH 6.2. The samples were equilibrated overnight at 4°C, and
the cGMP-antibody complex was precipitated with 2 ml of ice-cold
ethanol. Radioactivity of the pellet was determined with a gamma
counter (1272; Clinigamma, Wallac, Finland). Quantitation of cGMP (in
picomoles per well) was calculated from a standard curve generated
using nonlinear regression to the formula C = Cmin + [(Cmax Cmin)/(1 + eIC50 [cGMP])/slope)] (Prism v. 2.0; GraphPad Software, San
Diego, CA), where C is counts per minute measured for the sample
pellet, Cmin is counts per minute measured for
nonspecific binding, and Cmax is counts per
minute measured for total binding. A least-squares nonlinear regression
analysis of the data resulted in an R2
value of 0.982 ± 0.005 (n = 13 experiments).
Cross-reactivity of the cGMP antiserum with various purine nucleosides
was determined in competitive binding assays. When possible, the
concentration of ligand that competed for 50% of the
125I-cGMP binding sites
(IC50) was determined with nonlinear regression to the formula presented above. cAMP competed for the
125I-cGMP binding sites with an affinity that was
more than three orders of magnitude less than that of cGMP
(IC50 of cGMP = 5.5 ± 0.4 pM;
IC50 of cAMP = 9.5 ± 0.2 nM;
n = 3 independent experiments). The antiserum showed no
cross-reactivity with guanosine, GMP, GDP, GTP, adenosine, AMP, ADP, or
ATP at concentrations of 10 µM.
Materials.
MePhe7-NKB, an
NK3 receptor agonist, substance P, and neurokinin
A (Research Biomedicals International, Natick, MA), were dissolved and
stored in 1 mM acetic acid. SR 142801, a nonpeptide
NK3 receptor antagonist (Sanofi Recherche,
Montpellier, France), was dissolved and stored in 10% (w/v)
2-hydroxypropyl--cyclodextrin (Research Biochemicals International).
Calcium chelators BAPTA-AM (Calbiochem) and EGTA (Sigma Chemical Co.,
St. Louis, MO) were dissolved and stored in DMSO (Sigma Chemical Co.)
and HEPES buffer, respectively. Thapsigargin, an endoplasmic reticular
Ca2+-ATPase inhibitor (Calbiochem), was dissolved
and stored in DMSO. An NOS inhibitor,
NG-nitro-L-arginine
(Calbiochem), was dissolved in HEPES buffer. 2-(4-Carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO; Calbiochem), an NO scavenger, was dissolved and stored in DMSO.
Triethylamine, acetic anhydride, and BSA used in the cGMP
radioimmunoassay were obtained from Sigma Chemical Co. Bovine 
-globulin used in the cGMP radioimmunoassay was obtained from ICN
(Costa Mesa, CA). All purine nucleosides (Sigma Chemical Co.) used as
standards or for competitive binding in the radioimmunoassay were
dissolved and stored in H2O. All compounds were
diluted with HEPES buffer to the appropriate working concentration for
each assay. Chemicals used to dissolve test compounds were less than 0.1% final concentration in contact with cells. HEPES buffer was used
for all control and basal experiments.
Data Analysis.
Within individual experiments, treatments
were repeated in triplicate, and results were averaged to obtain one
value. Reported data are the mean ± S.E. for n
independent experiments. Unless otherwise stated, statistical analyses
were done with ANOVA followed by Student-Newman-Keuls multiple
comparisons test or with Student's unpaired t test where
appropriate. Significance was determined at a level of
P < .05.
Nonlinear regression to one-site sigmoidal concentration-response
curves and determination of EC50 values were
determined using an equation in Prism (GraphPad Software). In some
cases, the concentration-response data were analyzed according to a
two-site model by fitting the following equation to the data by
nonlinear regression analysis (Prism):
where Max denotes the overall maximal cellular response,
fractionH is the proportion of the response
mediated by the high-potency component, and EC50H
and EC50L denote the high- and low-potency EC50 values, respectively. To determine whether
regression to a two-site model was an appropriate fit of the data, the
sum-of-squares for one-site and two-site regressions were compared with
an F test and accepted if P < .05 (Prism).
For all concentration-response data, analyses were performed on each
experiment, and reported EC50 values are the
mean ± S.E. for n independent experiments. Intrinsic efficacy was
calculated for each agonist according to the method of Ehlert (1985)
using the equation:
| |
Results |
Transfection and Confirmation.
CHO cells that stably express
the human NK3 receptor were used to explore the
coupling of these receptors to the generation of NO. CHO
hNK3 cells were additionally transfected with the
gene encoding nNOS. The success of the transfection was confirmed by the generation of NO in the transfected cells (see later) as well as by
detection of nNOS immunoreactivity. Immunoreactive NOS was present in
all cells (Fig. 1A). Immunoreactivity was
specific for NOS because preabsorption of the antiserum with the
immunizing peptide (0.02 mg/ml) completely eliminated staining (Fig.
1B). CHO cells that were not transfected with nNOS did not stain for the enzyme (data not shown), providing further evidence that the antiserum was specific for NOS.
View larger version (49K):
[in this window]
[in a new window]
|
Fig. 1.
A, digital image of type I NOS immunoreactivity in
CHO hNK3/nNOS cells. Note that the immunofluorescence is
associated with every cell. B, preabsorption of the antiserum with the
immunizing peptide eliminated staining. Bar, 30 µm.
|
|
To determine whether transfection with nNOS impaired signal
transduction, agonist-stimulated inositol phospholipid hydrolysis was
examined. To examine the activation of phosphoinositol hydrolysis by
NK3 receptors, we assayed the effect of
MePhe7-NKB, an NK3
receptor-specific agonist, on the accumulation of total inositol
phosphates within cells. Levels of total inositol phosphates increased
2.0 ± 0.3-fold over basal levels after stimulation with 300 nM
MePhe7-NKB. The response to
MePhe7-NKB was concentration-dependent, and the
concentration needed for half-maximal response
(EC50) was 8.3 ± 4.1 nM. The
EC50 value reported for
MePhe7-NKB is approximately 20-fold less than
previously reported for this cell line (Krause et al. 1997). Although
transfection of the CHO hNK3 cells with the nNOS gene may have changed
the efficiency of coupling to intracellular effectors, the cells are
still able to couple to intracellular effector systems.
NK3 Receptor-Mediated Formation of
L-[3H]Citrulline.
Activation of nNOS
results in the conversion of L-arginine into NO and
L-citrulline. Thus, the conversion of
L-[3H] arginine to
L-[3H]citrulline was used to
investigate whether NK3 receptors couple to the
activation of nNOS. Incubation with MePhe7-NKB
(300 nM) for 60 min significantly increased the level of L-[3H]citrulline in CHO
hNK3/nNOS cells (12,512 ± 938 dpm) compared with basal levels (4404 ± 417 dpm; n = 10 experiments). MePhe7-NKB stimulation of
L-[3H]citrulline
production was concentration-dependent (Fig.
2). A two-site model for the
concentration-response relationship fit the experimental data better
than a one-site model (F test). Thus, nonlinear regression
of the data yielded a high-potency response with an
EC50 value of 0.13 ± 0.04 nM and a
lower-potency response with an EC50 value of
55 ± 36 nM (n = 3 experiments). Substance P and
neurokinin A also stimulated the formation of
L-[3H]citrulline, each
with only a single potency (Fig. 2 and Table 1).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 2.
The addition of NK3 receptor agonists to
CHO hNK3/nNOS cells stimulated the production of
L-[3H]citrulline in a concentration-dependent
manner. Concentration-response curves for MePhe7-NKB ( ),
neurokinin A ( ), and substance P ( ) in increasing levels of
L-[3H]citrulline. Varying concentrations of
agonists were incubated with cell suspensions for 60 min. A two-site
model (see Experimental Procedures) for the
concentration-response relationship for MePhe7-NKB fit the
experimental data better than a one-site model (F test).
The experimentally determined EC50 value for the
high-potency component of the cellular response was 0.13 ± 0.04 nM and for the lower-potency component was 55 ± 36 nM. Neurokinin
A and substance P stimulated the formation of
L-[3H]citrulline with EC50 values
of 288 ± 97 and 4946 ± 1045 nM, respectively. Data are the
mean ± S.E. percent of the radioactivity due to
L-[3H]citrulline obtained in response to 300 nM MePhe7-NKB within each experiment, for three or four
independent experiments.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1
Summary of the Ki and EC50 values from
competitive binding studies and functional assays of the production of
L-citrulline and cGMP by various NK3 receptor
agonists
The data reported here summarize experiments presented in Figs. 2 and 4
and incorporate data from a previous report to emphasize the
amplification of NOS signaling and the difference in potencies between
agonists. EC50 values are given as mean ± S.E. from three
or four independent experiments. Intrinsic efficacy is calculated as
described in the text.
|
|
NK3 Receptor-Mediated Formation of cGMP.
The low
sensitivity of the citrulline assay, resulting from the existence of a
high concentration of endogenous L-arginine, necessitates
measurement of the accumulation of L-citrulline during long
time intervals after the addition of agonist. Because the apparent
two-site action of MePhe7-NKB on
L-[3H]citrulline formation may be
an effect of a lengthy incubation with the agonist, the activation of
nNOS was also investigated using an indirect approach that more closely
reflected changes in production of NO at early time points. Rat lung
fibroblast (RFL-6) cells contain high levels of soluble guanylyl
cyclase. Because NO readily diffuses across membranes and because
soluble guanylyl cyclase is activated by NO, levels of cGMP can be used to estimate activation of NOS. When incubated separately, neither CHO
hNK3/nNOS cells nor RFL-6 cells showed increased
production of cGMP in the presence of MePhe7-NKB
(0.1-300 nM). However, when CHO hNK3/nNOS cells
were coincubated with RFL-6 cells, MePhe7-NKB
stimulated the production of cGMP in a time-dependent manner (Fig.
3). The time course showed an initial
increase in cGMP that reached a peak of 5.17 ± 1.45 pmol/well at
1 min. This level of cGMP is significantly increased over the basal
level of 0.79 ± 0.27 pmol cGMP/well. The peak level of cGMP then
declined to a plateau of 1.71 ± 0.32 pmol/well that lasted at
least 15 min. Using the response at 1 min,
MePhe7-NKB stimulated an increase in cGMP levels
in a concentration-dependent manner (Fig.
4) with an EC50
value of 77.9 ± 25.4 pM. Substance P and neurokinin A also
stimulated the production of cGMP (Fig. 4 and Table 1).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 3.
Time course of MePhe7-NKB (100 pM)-induced increase in cGMP. MePhe7-NKB was added to a
mixture of CHO hNK3/nNOS and RFL-6 cells, and the cells
were lysed after different time intervals. cGMP in the lysate was
determined by radioimmunoassay. Data are the mean ± S.E. for
three independent experiments.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 4.
The addition of MePhe7-NKB to CHO
hNK3/nNOS cells stimulated the production of cGMP in RFL-6
cells in a concentration-dependent manner. Concentration-response
curves for MePhe7-NKB (), neurokinin A (), and
substance P () in increasing levels of cGMP. Varying concentrations
of agonists were incubated with cell suspensions for 1 min. The
experimentally determined EC50 values were 0.078 ± 0.025 nM for MePhe7-NKB, 47 ± 28 nM for neurokinin A,
and 200 ± 47 nM for substance P. Data are the mean ± S.E.
percentage of the cGMP formed in response to 3 nM
MePhe7-NKB within each experiment, for three independent
experiments.
|
|
Pharmacological tools that disrupt either the generation or action of
NO were used to confirm a causal relationship between the activation of
nNOS by agonists at NK3 receptors and cGMP
formation. The increase in cGMP in response to
MePhe7-NKB (100 pM) was blocked by a 5-min
pretreatment with
NG-nitro-L-arginine
(100 µM), a competitive NOS inhibitor (Fig. 5). The cGMP response was also inhibited
by a 5-min pretreatment with cPTIO (500 µM), an NO scavenger (Fig.
5). These data indicate that cGMP formation in response to stimulation
of NK3 receptors is the result of activation of
NOS.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 5.
NK3 receptor-mediated production of cGMP
is through the production of the diffusible messenger NO. Mixtures of
CHO hNK3/nNOS and RFL-6 cells were preincubated for 5 min
at 37°C with or without
NG-nitro-L-arginine (100 µM)
or cPTIO (500 µM). Cells were then incubated with or without
MePhe7-NKB (100 pM) for 1 min. Data are the mean ± S.E. for three independent experiments and were analyzed using two-way
ANOVA and Student-Newman-Keuls multiple comparisons test to determine
significant differences among groups. *, different from
basal levels of similarly treated cells.  , different from
MePhe7-NKB response in untreated cells.
|
|
To confirm that activation of nNOS after the addition of
MePhe7-NKB was mediated by
NK3 receptors, we pretreated cells with the NK3 receptor-specific, nonpeptide antagonist SR
142801. SR 142801 treatment for 5 min shifted the
MePhe7-NKB concentration-response curve to the
right in a concentration-dependent manner. In the presence of 10 nM SR
142801, MePhe7-NKB had an
EC50 value of 0.82 ± 0.35 nM. In the
presence of 30 nM SR 142801, MePhe7-NKB had an
EC50 value of 3.7 ± 1.6 nM. When treated
with 100 nM SR 142801, the EC50 value was
10.4 ± 4.7 nM. A Schild plot of these data is presented in Fig.
6. A straight line with a slope of 1.2 fit the data (r = 0.997), confirming a competitive mode of antagonism by SR 142801 and an experimentally determined
pA2 value of 9.0. These values are similar to
those previously reported for this compound in antagonizing
NK3 receptors in the guinea pig longitudinal
muscle preparation (Emonds-Alt et al., 1995).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6.
Schild plot for SR 142801 antagonism of
MePhe7-NKB-induced cGMP production. Mixtures of CHO
hNK3/nNOS and RFL-6 cells were incubated with SR 142801 (10, 30, or 100 nM) at 37°C for 5 min before the addition of varying
concentrations of MePhe7-NKB for 1 min.
Concentration-response functions of MePhe7-NKB in the
presence of 10, 30, or 100 nM SR 142801 were determined by nonlinear
regression from three independent experiments. The concentration ratio
(CR) was determined as the mean EC50 values in the presence
of SR 142801 divided by the EC50 value determined in Fig.
2. A plot of log(CR 1) against the logarithm of the antagonist
concentration regressed to a straight line (r = 0.997) with a slope of 1.2 and a pA2 value of 9.0.
|
|
Although the production of
L-[3H]citrulline in response to
MePhe7-NKB was best fit by a two-site model, the
production of cGMP in response to MePhe7-NKB
corresponded to a one-site concentration-response relationship. One
possible explanation of this discrepancy is that the concentration of
soluble guanylyl cyclase, contained in the RFL-6 cells, is limiting.
This would result in a masking of the appearance of a second,
low-potency component of the cGMP response. To test this hypothesis, we
determined whether a 10-fold increase in the density of RFL-6 cells
altered the concentration-response curve (Table
2). The 10-fold increase in the
concentration of RFL-6 cells increased the basal level of cGMP and the
cGMP level in response to MePhe7-NKB (100 pM) by
10-fold. There was, however, no difference in the
EC50 value determined in the presence of 2 × 105 versus 2 × 106
RFL-6 cells, and a one-site model for the concentration-response curve
fit the data better than a two-site model.
View this table:
[in this window]
[in a new window]
|
TABLE 2
Measures dependent on RFL-6 cell concentration
CHO hNK3/nNOS cells were incubated with various concentrations
of MePhe7-NKB (0.001-10 nM) for 1 min in wells that contained
2 × 105 RFL-6 cells and in wells that contained 2 × 106 RFL-6 cells. The concentration-response functions and
EC50 values were determined by nonlinear regression. Data are
the mean ± S.E. for three independent experiments and were
analyzed using an unpaired t test to determine significant
differences.
|
|
Calcium Dependence of NK3 Receptor-Mediated Production
of NO.
Because NK3 receptors were linked to
the activation of nNOS and because this process is thought to be
dependent on an increased concentration of intracellular calcium
(Knowles et al., 1989), we investigated the dependence of
NK3 receptor-mediated activation of NOS on
calcium. BAPTA-AM was used to chelate intracellular calcium.
Pretreatment of the cells with BAPTA-AM (100 µM) for 15 min blocked
the production of cGMP 1 min after stimulation with
MePhe7-NKB (0.1 nM; Fig.
7). These data indicate that increased
production of cGMP after stimulation with
MePhe7-NKB was indeed a calcium-dependent
process.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 7.
NK3 receptor-mediated activation of NOS
is calcium-dependent. In two conditions, mixtures of CHO
hNK3/nNOS and RFL-6 cells were incubated at 37°C with or
without either BAPTA-AM (100 µM for 15 min) or EGTA (3 mM for 5 min).
In a third condition, a suspension of CHO hNK3/nNOS cells
was incubated at 37°C for 15 min with thapsigargin (500 nM), washed,
and subsequently added to RFL-6 cells. The cell mixtures were then
incubated with or without MePhe7-NKB (100 pM) for 1 min.
Data are the mean ± S.E. for three independent experiments and
were analyzed using two-way ANOVA and Student-Newman-Keuls multiple
comparisons test to determine significant differences among groups.
*, different from basal levels of similarly treated cells.
, different from MePhe7-NKB response in untreated
cells.
|
|
To investigate the source of calcium required for the generation of NO,
EGTA was used to chelate extracellular calcium, and thapsigargin was
used to deplete intracellular stores of calcium. A 5-min treatment with
3 mM EGTA, a concentration sufficient to reduce levels of extracellular
calcium well below the basal concentration of free intracellular
calcium (Bers, 1982), did not alter the increase in cGMP 1 min after
the addition of MePhe7-NKB (100 pM; Fig. 5).
However, when CHO hNK3/nNOS cells were treated
with thapsigargin (500 nM) for 15 min, washed, and subsequently incubated with RFL-6 cells, there was no increase in cGMP in response to MePhe7-NKB (100 pM; Fig. 7). The duration of
treatment and concentration of thapsigargin is sufficient to inhibit
endoplasmic reticular Ca2+-ATPase and effectively
deplete calcium from intracellular stores without affecting resting
levels of free intracellular calcium (Thomas and Hanley, 1994). These
data indicate that the production of cGMP 1 min after
MePhe7-NKB treatment is dependent on release of
calcium from intracellular stores and not on influx of extracellular
calcium. Effects at later time points in the response (i.e., plateau)
were not tested.
NK3 Receptor-Mediated Calcium Signaling.
Because
MePhe7-NKB activation of nNOS was
calcium-dependent, we investigated the calcium signaling of CHO
hNK3/nNOS cells in response to
MePhe7-NKB. A suspension of cells was stimulated
with MePhe7-NKB (30 nM), and the change in
[Ca2+]i was measured over
time. There was an initial rapid and transient increase in
[Ca2+]i, followed by a
lower plateau that did not return to baseline levels during a 15-min
incubation (Fig. 8A). Basal
[Ca2+]i in CHO hNK3/nNOS
cells was 119 ± 29 nM.
[Ca2+]i increased to
1.28 ± 0.15 µM at the peak of the response and plateaued at
240 ± 45 nM (measured 2.5 min after the addition of agonist;
n = 4 experiments). On the basis of the peak increase in intracellular calcium, MePhe7-NKB increased
[Ca2+]i in a
concentration-dependent manner (Fig. 9)
with an EC50 value of 16.1 ± 4.7 nM
(n = 4 experiments).
View larger version (7K):
[in this window]
[in a new window]
|
Fig. 8.
Representative profiles demonstrating the
contribution of extracellular calcium to the change in
[Ca2+]i in response to an NK3
receptor agonist. Cells were loaded with Fura-2, and then
MePhe7-NKB was added directly to the suspension. In the
control condition (A), MePhe7-NKB (30 nM) induced a
biphasic increase in [Ca2+]i with a transient
initial peak followed by a prolonged increase in
[Ca2+]i that lasted at least 15 min. The
addition of EGTA (3 mM; B) reduced the initial peak and abolished the
sustained increase in [Ca2+]i.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 9.
Maximal increase in [Ca2+]i
in a suspension of CHO hNK3/nNOS cells in response to
MePhe7-NKB was used to define the response to agonist, and
the concentration-response function was determined by nonlinear
regression. The experimentally determined EC50 value was
16.1 ± 4.7 nM.
|
|
Although extracellular calcium was not required for the activation of
NOS and the production of cGMP, we explored whether extracellular
calcium still contributed to the changes in
[Ca2+]i observed on the
activation of NK3 receptors. In the presence of
EGTA (3 mM), the initial increase in free intracellular calcium was
significantly reduced, and the plateau response was abolished (Fig.
8B). Cells incubated for 3 min with EGTA had basal levels of
intracellular calcium of 73.9 ± 11.2 nM. This increased to 890 ± 166 nM after the agonist addition. The intracellular
calcium level returned to a basal level of 65.1 ± 11.8 nM,
measured 2.5 min after the agonist addition (n = 4 experiments). These data indicate that extracellular calcium was
responsible for the sustained increase in
[Ca2+]i.
An apparent discrepancy exists between the data for cGMP production and
calcium signaling. A concentration of MePhe7-NKB
(0.1 nM) that produced a maximal increase in cGMP (Fig. 4) did not
cause a significant increase in
[Ca2+]i in suspensions of
CHO hNK3/nNOS cells (Fig. 9). Stimulation of
these cells with low concentrations of agonist may increase [Ca2+]i that is below the
detection limit of measurement of
[Ca2+]i in a suspension
of CHO hNK3/nNOS cells. To test this hypothesis, we compared changes in
[Ca2+]i in response to
0.1 nM MePhe7-NKB in single cells to the response
to 30 nM, which maximally increased
[Ca2+]i in cell
suspension. A cell was defined as responsive to agonist if
MePhe7-NKB produced an increase in
[Ca2+]i that was 2-fold
greater than the basal
[Ca2+]i. Representative
profiles of responses to 0.1 and 30 nM MePhe7-NKB
are presented in Fig. 10. Most cells
that responded to 0.1 nM MePhe7-NKB had multiple
rapid and transient increases in
[Ca2+]i (11 of 13 cells;
Fig. 10A). Two cells responded with a single peak of longer duration
that slowly decayed to basal
[Ca2+]i. Conversely,
cells generally responded to 30 nM MePhe7-NKB
with a single peak (14 of 17, Fig. 10B). Only three cells responded to
30 nM agonist with multiple peaks. Response profiles were similar to
those reported previously (Pinnock et al., 1996). The proportion of
cells responding to 0.1 nM MePhe7-NKB (13 of 18;
72%) was significantly lower than the proportion of cells responding
to 30 nM MePhe7-NKB (17 of 17; 100%; Fisher's
exact test). Every cell tested first with 100 pM
MePhe7-NKB (n = 18) responded to
subsequent stimulation with 30 nM MePhe7-NKB,
which indicates that every cell contains functional
NK3 receptors.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 10.
Representative profiles of change in
[Ca2+]i in single CHO hNK3/nNOS
cells in response to an NK3 receptor agonist. Cells
attached to a glass coverslip were loaded with Indo-1 and superfused
with agonist for 2 min, as indicated by the time bar. A, cells
responded to 100 pM MePhe7-NKB most frequently with
multiple transient increases in [Ca2+]i. B,
response to 30 nM MePhe7-NKB was most frequently
characterized by a single peak of longer duration that slowly decayed
to basal [Ca2+]i.
|
|
| |
Discussion |
Although diverse physiological effects of
NK3 receptor activation are mediated by NO, the
data presented here represent direct demonstration that
NK3 receptor activation couples to the production of NO via an intracellular pathway. In the present study, NO production after NK3 receptor activation in CHO cells is
concurrent with increased levels of inositol phosphates and
intracellular calcium. It is reasonable to assume that the production
of NO in this system occurs through the release of calcium from
intracellular stores through IP3-gated calcium
channels. This is likely to occur after NK3
receptor activation of Gq, stimulation of
phospholipase C, and the formation of IP3. In
support of the hypothesis that NK3 receptor-induced IP3 formation and subsequent
release of calcium from intracellular stores form the signaling pathway
responsible for NOS activation, peak NO production in this study was
dependent on the release of calcium from intracellular stores.
Furthermore, the increase in
[Ca2+]i, and cGMP
occurred over a similar time course.
Although the release of calcium from intracellular stores was shown to
contribute to the activation of NOS by NK3
receptors within the same cell, our experiments do not exclude the
involvement of other second messenger systems in the generation of NO.
cAMP has been shown to mediate the production of NO in intestinal
smooth muscle and epithelial cells (Scott-Burden et al., 1994; Tamaoki et al., 1995) as well as to enhance receptor-mediated production of NO
in nervous tissue (Toms and Roberts, 1994). NK3
receptors in another transfected cell line have been linked to a modest increase in cAMP (Nakajima et al., 1992). Thus, cAMP may have contributed to the production of NO in the present study. Another intracellular signaling molecule that may play a role in the formation of NO after the activation of NK3 receptors is
protein kinase C (PKC). PKC is activated by diacylglycerol, which also
is a product of phospholipase C. Active PKC generates an influx of
extracellular calcium in several systems (DiRiemer et al., 1985) and is
responsible for -adrenergic-mediated activation of NOS in primary
cultures of mouse cortical cells (Agullo and Garcia, 1992). Because the increase in NO that occurred at 1 min after the activation of NK3 receptors was dependent on the release of
calcium from intracellular stores and not the influx of extracellular
calcium, it seems unlikely that PKC-activated calcium influx is
involved in the acute increase in cGMP. However, PKC may contribute to
the sustained production of NO because the elevation in
[Ca2+]i that occurred
through 15 min was dependent on the influx of extracellular calcium.
Finally, although NO more commonly modulates arachidonic acid-mediated
cell signaling (for a review, see Salvemini, 1997), several reports
indicate a reciprocal modulation of NO-cGMP signaling by arachidonic
acid (Reiser, 1990). Because CHO hNK3 cells
couple to arachidonic acid release (Krause et al., 1997), arachidonic
acid may play a role in the activation, or modulation, of NOS in the
present study. The CHO hNK3/nNOS cells prepared in the present study
provide a model for examination of the role of the relative
contribution of intracellular signaling pathways in coupling
NK3 receptors to the generation of NO.
Consistent with previous reports (Drapeau et al., 1987; Krause et al.,
1997), MePhe7-NKB was more potent than neurokinin
A, which was more potent than substance P, in NK3
receptor-mediated production of NO. Potency for a biological effect,
such as the production of NO, is dependent on two properties of
agonists: affinity and intrinsic efficacy. It is generally assumed that
neurokinin A and substance P are full agonists at
NK3 receptors because they are able to generate maximal biological responses (Maggi et al., 1993b). A more appropriate estimate of intrinsic efficacy, however, is the comparison of concentration-response and concentration-occupancy curves. The estimation of intrinsic efficacy according to the method developed by
Ehlert (1985) allows for the comparison of intrinsic efficacies for
"apparent" full agonists. Using the
Ki values obtained in radioligand
binding assays by Krause et al. (1997) and the
EC50 values obtained in the present study for the
production of citrulline and cGMP by substance P, neurokinin A, and
MePhe7-NKB, the intrinsic efficacy of each
agonist may be calculated (Table 1). Using this method of analysis, it
is apparent that MePhe7-NKB has a higher
intrinsic efficacy than neurokinin A and substance P at
NK3 receptors. Furthermore, it is likely that the
differences in potency among the agonists in the production of NO are a
combination of the differences in affinity and intrinsic efficacy.
The two-site nature of MePhe7-NKB-induced
production of L-citrulline is not apparent in
concentration-response curves for either neurokinin A or substance P. This indicates that a property intrinsic to
MePhe7-NKB may allow it to interact with the
NK3 receptor in two different ways. There are two
explanations for this phenomenon. First, it is possible that
MePhe7-NKB may have affinity for two binding
sites on the NK3 receptor. If this is the case,
the affinity of MePhe7-NKB for the two binding
sites must be similar because radioligand binding of
[125I]MePhe7-NKB to
NK3 receptors does not reveal binding to two
sites (Sadowski et al., 1993; D. R. Linden and V. A. Seybold,
unpublished observations). Second, it is possible that
MePhe7-NKB is able to activate more than one
intracellular signaling pathway, with different potencies, that
converge on the activation of nNOS. Defining the mechanisms responsible
for the two-site nature of MePhe7-NKB requires
further exploration.
Whereas a two-site model is more appropriate for the data obtained for
MePhe7-NKB-induced production of
L-[3H]citrulline, a one-site model
is adequate for the data obtained for
MePhe7-NKB-induced production of cGMP. This
phenomenon was not biased by the concentration of guanylyl cyclase in
the assay because increasing the level of the effector enzyme (i.e.,
increasing the RFL-6 cells/well) did not alter the
MePhe7-NKB concentration-response relationship.
The activation of guanylyl cyclase may be mediated solely by the
high-potency component of NK3 receptor-induced
NOS activation because the EC50 values for the
production of cGMP and the high-potency component of
L-citrulline formation were the same. Thus, the NO produced
by 100 pM MePhe7-NKB maximally activated guanylyl
cyclase, precluding the observation of another low-potency component of
the response at higher agonist concentrations.
Both neurokinin A and substance P stimulated the production of NO
through the activation of NK3 receptors. This
finding may have implications on the interpretation of the physiology
of spinal cord neurokinin receptors. During the persistent activation
of nociceptors, the release of substance P and neurokinin A from primary afferent neurons is greatly increased (Hope et al., 1990; Schaible et al., 1990). Under this condition, these peptides may also
cause the production of NO via NK3 receptors.
Substance P- and neurokinin A-containing terminals occur in the region
of dendrites and cell bodies that express NK3
receptors (Hökfelt et al., 1975; Ding et al., 1996). Therefore,
it is feasible that physiological concentrations of substance P or
neurokinin A may activate NK3 receptors to
produce NO, which contributes to hyperalgesia at the level of the
spinal cord (Meller and Gebhart, 1994). Although the study of this cell
signaling pathway in spinal neurons would better address the physiology
of the NK3 receptor, the heterogeneous population
of cells in the spinal cord would not eliminate the possibility of
transneuronal activation of nNOS by NK3
receptors. Therefore, the use of the CHO hNK3/nNOS cells provided a
model for the study NK3 receptor-mediated
activation of nNOS through intracellular pathways.
In conclusion, this study presents evidence that
NK3 receptors join a long list of other G
protein-coupled receptors linked to the activation of NOS (for a
review, see Christopoulos and El-Fakahany, 1999). Results of this study
show that NK3 receptors couple to the generation
of NO within the same cell. The marked amplification of this signaling
pathway provides the potential for physiological concentrations of
endogenous substance P or neurokinin A to stimulate NO formation at
NK3 receptors in vivo despite its low affinity at
this particular receptor subtype. Furthermore, there may be differences
between neurokinin receptor agonists, such as intrinsic efficacy, that
would explain differences between various cellular responses downstream
of receptor binding.
We acknowledge Drs. Ann M. Parsons, David K. Waid, and Arthur
Christopoulos for valuable input into the manuscript.
NK, neurokinin;
NO, nitric oxide;
nNOS, neuronal NO synthase;
CHO, Chinese hamster ovary;
RFL, rat fetal lung;
IP3, inositol-1,4,5-trisphosphate;
BAPTA-AM, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic
acid;
cPTIO, 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide;
PKC, protein kinase C.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
![<UP>Response = Max × </UP><FENCE><FR><NU><UP>fraction<SUB>H</SUB></UP></NU><DE><UP>1 + 10</UP><SUP>(<UP>log EC</UP><SUB><UP>50H − log </UP>[<UP>agonist</UP>]</SUB>)</SUP></DE></FR><UP> + </UP><FR><NU><UP>1 − fraction<SUB>H</SUB></UP>)</NU><DE><UP>1 + 10</UP><SUP>(<UP>log EC</UP><SUB><UP>50L − log </UP>[<UP>agonist</UP>]</SUB>)</SUP></DE></FR>i</FENCE>](/content/vol293/issue2/fulltext/559/img001.gif)
