Activation of the neurokinin-1 receptor by substance P triggers the release of substance P from cultured adult rat dorsal root ganglion neurons

To investigate whether SP induces its own release from cultured DRG neurons, we examined the effects of SP on the release of SP in a dose- and time-dependent manner. Based on the amount of the SP release induced by various chemicals in our previous study [5, 6, 18], we selected 200 pg/dish of SP as an appropriate concentration for our experimental conditions for investigating the possibility of self-induced SP release. A time-course of SP release induced by SP (200 pg/dish) from cultured DRG neurons is shown in Fig. 1A. As a peak of SP release was observed after the 60 min incubation, we decided to use the 60 min incubation with SP (200 pg/dish) as an experimental condition for examining various drugs on the self-induced SP release. As shown in Fig. 1B, SP evoked a dose-dependent release of SP during a 60 min incubation of cultured DRG neurons.

It is well known that all three neurokinin receptors (neurokinin-1, -2 and -3 receptors) are expressed in DRG neurons. We therefore investigated whether neurokinin-1 and/or other neurokinin receptor(s) are involved in the SP release induced by itself. The increase in the SP release evoked by itself was partially significantly attenuated by 1 μM CP-96,345 (a selective antagonist of neurokinin-1 receptor) [19] and by 100 nM GR94800 (a selective antagonist of neurokinin-2 receptor) [20], not by 1 μM SB222200 (a selective antagonist of neurokinin-3 receptor) [20] as shown in Fig. 1C, whereas these antagonists did not have any effect when used alone (Data not shown). Based on the results shown in Fig. 1C, both the neurokinin-1 and -2 receptors seem to be involved in the process of SP release, however, the detailed pharmacological action(s) of the neurokinin-2 receptor in the substance P release will be examined in future experiments.

We next investigated the time-dependent changes in the expression of SP and its neurokinin-1 receptor in cultured DRG cells incubated in serum-free DMEM (containing peptidase inhibitors) with or without SP (200 pg/dish) in the presence/absence of 1 μM CP-96,345. The specific type of cells indicated by arrows in Figs 2A and 2B can easily be identified to be DRG neurons by the morphology (characteristic phase bright cell bodies bearing either bi- or multi-polar neurites observed by using the Olympus IX71 inverted microscope with a 20X Ph1 objective) and by the Schwann cell marker S100 (Rabbit anti-cow S100; 1:1,000 dilution; Dako, Glostrup, Denmark)- and the astrocyte marker GFAP (Mouse anti-glial fibrillary acidic protein; 1:800 dilution; Chemicon, Temecula, CA)-negative expression (Data not shown). As shown in Figs 2Ai–iv, the neurokinin-1 receptors were distributed on the membrane and in the cytoplasm of smaller DRG neurons in a naive state. The ratio of the number of neurons expressing the neurokinin-1 receptor on their membranes to the total number of neurokinin-1 receptor-positive neurons in a randomly selected field in each image from three separate experiments was simultaneously calculated. The percentage of neurons expressing neurokinin-1 receptor on their membrane in a naive state was 91 ± 5%, 82 ± 4% and 74 ± 4%, 71 ± 3% in Figs 2Ai–iv, respectively. Interestingly, a significant reduction in the percentage of neurons expressing neurokinin-1 receptor on their membranes was observed after the time-dependent stimulation of 200 pg/dish SP (85 ± 4%, 71 ± 7% and 56 ± 5%, 32 ± 2% in Figs 2Av–viii, respectively) in comparison to their respective controls (p > 0.05, p > 0.05 and p < 0.05, p < 0.001, respectively). Furthermore, the reduction induced by SP was attenuated by the pretreatment (10 min) with 1 μM CP-96,345 (86 ± 4%, 85 ± 7% and 84 ± 2%, 84 ± 5% in Figs 2Aix–xii, respectively). On the other hand, we observed the SP content to be mainly distributed in the cytoplasm of the smaller DRG neurons (Fig. 2B). The long-term (60 to 360 min) exposure of the DRG neurons to SP (200 pg/dish) thus resulted in a reduction of the SP content in their cytoplasm (Figs 2Bv–viii), whereas the pretreatment with 1 μM CP-96,345 blocked the reduction of the SP content (Figs 2Bix–xii). The SP-evoked expression change of the neurokinin-1 receptor from the membrane to the intracellular compartment in somas of DRG neurons should therefore be considered as the receptor internalization (a pharmacologically specific index of neurokinin-1 receptor-SP interaction).

Based on the neurokinin-1 receptor localization results shown in Fig. 2A, we attempted to further quantify the levels of neurokinin-1 receptor in the cytosolic and membrane fractions of cultured DRG neurons. As shown in Figs 3A and 3B, the time-dependent (10–360 min) exposure of cultured DRG neurons to SP (200 pg/dish) resulted in a significant decrease of the neurokinin-1 receptor expression in the membrane fractions (104 ± 15%, 74 ± 4% and 42 ± 5%, 14 ± 1% in Figs 3A and 3B, respectively, compared with the control at 10 min). The SP-induced decrease of that was completely blocked by the 10 min pretreatment with 1 μM CP-96,345 (102 ± 9%, 100 ± 7% and 74 ± 11%, 61 ± 14% in Figs 3A and 3B, respectively, compared with the control at 10 min). Therefore the activation of neurokinin-1 receptor by SP is accompanied by the internalization of its receptors, however, we could not identify the proteins detected by anti-substance P receptor antibody in the cytosol as a functional neurokinin-1 receptor. Interestingly, we observed a gradual decrease of the neurokinin-1 receptor expression in the membrane fraction from the untreated DRG neurons (the control). These data suggest the endogenous SP released from the DRG neurons could enhance the turnover of neurokinin-1 receptor from the membrane to the cytosol as the released SP is not taken into the DRG cells. However, after the induction of internalization by the stimulation of SP, the level of the cytosolic proteins detected by anti-substance P receptor antibody does not change, thus suggesting a part of internalized neurokinin-1 receptor protein could be degraded to maintain the amount of neurokinin-1 receptor to some extent within the cytosol.

The data shown in Figs 1C and 2 indicate that the activation of neurokinin-1 receptor is involved in the SP release induced by SP. We therefore selected another potent neurokinin-1 receptor agonist GR73632 (showing no cross-reactivity with anti-SP serum in radioimmunoassay) to further investigate the molecular mechanisms of the SP release via the activation of the neurokinin-1 receptor in cultured DRG neurons. As shown in Fig. 4A, it was observed that a 60 min incubation with GR73632 (1–100 nM) stimulated a significant increase in the SP release in a dose-dependent manner from the cultured DRG neurons. The increases in the SP release induced by GR73632 at various concentrations were almost completely blocked by the 10 min pretreatment with CP-96,345. Based on the results shown in Fig. 4B, we thought that the 60 min incubation with 10 nM GR73632 was an appropriate condition in our experiments. However, we observed that a time-dependent treatment (10 min to 24 h) of GR73632 (10 nM) did not cause any detectable change in the total amount of SP content from cultured DRG neurons and the culture medium (Fig. 4C). In addition, significant changes of the preprotachykinin mRNA expression were not caused by the time-dependent (10–360 min) exposure of cultured DRG neurons to 10 nM GR73632 (Data not shown). Therefore, the effect of GR73632 under our experimental conditions should not be considered to cause a rapid increase in the synthesis of the SP content in cultured DRG neurons.

Whether the complex process of SP release induced by GR73632 requires the activation of MAP kinases, COXs, PLC or PKCs, PKA in cultured DRG neurons, still remains to be elucidated. As shown in Fig. 5A, both U0126 and SB203580 exhibited a significant inhibitory effect on the SP release evoked by GR73632, whereas SP600125 stimulated an increase in the SP release. When these three inhibitors [6] of MAP kinases were used alone, only a c-Jun NH₂-terminal kinase (JNK) inhibitor SP600125 had a weak tendency to increase the SP release (Fig. 5A). We also observed that the GR73632-induced SP release (Figs 5B and 5D) was significantly attenuated by NS-398 (a highly selective inhibitor of COX-2) [5], indomethacin (a non-selective inhibitor of COX1/2) [6] or by PKCε translocation inhibitor peptide [21], Gö6976 (a selective inhibitor for PKCα-, βI-isozymes) [22] or bisindolylmaleimide I (a broad inhibitor for PKCα-, βI-, βII-, γ-, δ-, ε-isozymes) [22]. However, neither U73122 (a selective inhibitor of PLC) [23] nor H89 (a selective inhibitor of PKA) [4] influenced the SP release induced by GR73632 (Figs 5C and 5D).

To clarify the possible signal transduction pathway(s) involved in the SP release via the activation of neurokinin-1 receptor, the activation status of COX-2 was assessed using specific antibodies for COX-2 after the stimulation of SP or GR73632 in the absence or presence of various inhibitors. The time-dependent exposure of DRG neurons to SP (200 pg/dish) resulted in the significant increase of de novo protein synthesis of COX-2 (Fig. 6A). The 60 min incubation with 10 nM GR73632 also up-regulated the expression of COX-2 protein (Fig. 6B), whereas the increase in the expression of COX-2 protein evoked by GR73632 was significantly attenuated by the pretreatment with CP-96,345, U0126, NS-398 or three inhibitors for PKC isozymes (Gö6976, bisindolylmaleimide I and PKCε translocation inhibitor peptide), respectively. However, neither SB203580 nor SP600125 influenced the up-regulation of COX-2 expression induced by GR73632.

According to a previously described method [5, 6, 23], DRGs of young adult Wistar rats (6–9 weeks of age) were dissociated into single isolated neurons and non-neuronal cells by the treatment of collagenase (Sigma, St. Louis, MO, USA) and trypsin (Invitrogen, Burlington, ON, Canada). The cells (3 DRGs/dish) were maintained at 37°C in a water-saturated atmosphere with 5% CO₂ for 5 days before the initiation of the experiments. All procedures for animal experiments were performed according to the Guide for Animal Experimentation, Hiroshima University, and the Committee of Research Facilities for Laboratory Animal Sciences, Graduate School of Biomedical Sciences, Hiroshima University, Japan.

Except for some cultured cells treated by peptidase inhibitors containing 1 μM phosphoramidon (Sigma), 4 μg/ml bacitracin (Sigma) and 1 μM captopril (Sigma) alone (as a control), other cultured cells were exposed to SP (Peptide Institute, Osaka, Japan) or to GR73632 (Sigma), either alone or together with various inhibitors such as Gö6976 (Calbiochem, Darmstadt, Germany), PKCε translocation inhibitor peptide (Calbiochem) and bisindolylmaleimide I (Calbiochem), indomethacin (Sigma) and SB222200 (Sigma), GR94800 (Sigma) and U73122 (Sigma), SP600125 (Sigma) and H89 (Seikagaku, Tokyo, Japan) in 1 ml serum-free DMEM (Nissui, Tokyo, Japan) containing peptidase inhibitors for a designated period of time at 37°C in a water-saturated atmosphere with 5% CO₂. Thereafter, the SP content collected from the culture medium and the cultured DRG neurons was measured by a highly sensitive radioimmunoassay [5, 18], respectively. For examining the amount of SP-induced SP release in the present experiments, we developed a new computational method. Briefly, SP at a specified concentration (100, 200 and 800, 1,000 pg/dish) was used to stimulate two groups of cultured DRG neurons in both the absence and presence of various antagonists for three neurokinin receptors (neurokinin-1, -2 and -3 receptors). The SP content was immediately collected from the culture medium (1 ml) after the SP stimulation for the first group, and the amount of SP content was examined from the culture medium (1 ml) after the SP stimulation lasted for 10, 60, 180, 360 minutes, respectively, for the second group. The numerical difference in the SP content between the two groups is considered to be the amount of SP release induced by this specified concentration of SP during a specific time period from cultured DRG neurons.

Immunocytochemical staining for the neurokinin-1 receptor and SP in cultured DRG neurons on coverglasses was performed with a standard immunoperoxidase technique [Histofine Simple Stain Rat MAX-PO (MULTI) kit; Nichirei, Tokyo, Japan] according to the manufacturer's instructions. Briefly, 4% paraformaldehyde-fixed cultured DRG cells on coverglasses were incubated with anti-neurokinin-1 receptor (1:2,000 dilution; Sigma) or anti-SP serum (1:1,000 dilution; a gift of Dr. J.S. Hong, National Institute of Environmental Health Sciences, NIH, USA) [33]. After the treatment with Histofine simple stain rat MAX-PO (MULTI), color development (brown) was performed using a DAB substrate kit (Nichirei), and the coverglasses were counterstained with hematoxylin (blue). According to the manufacturer's instructions for the datasheet of anti-substance P receptor antibody (S8305, Sigma), it is guaranteed that the antibody specifically recognizes the neurokinin-1 receptor peptide (rat, amino acids 393–407) in immunoblotting. Immunocytochemical controls demonstrating antibody specificity for the neurokinin-1 receptor and SP included immunostaining cultured cells on coverglasses, but the primary antibody was omitted. The omission of the primary antibody resulted in no staining in the cells.

After a 10 min pretreatment with the presence or absence of 1 μM CP-96,345 (Pfizer, Groton, CT, USA), the cultured DRG cells were incubated in serum-free DMEM (containing peptidase inhibitors) with or without SP (200 pg/dish) for 10, 60, 180, 360 minutes, respectively. The isolation of cytosolic and membrane fractions from these DRG cells was performed with a standard cell compartment kit fractionation procedure (Cell compartment kit; Qiagen, Tokyo, Japan) according to the manufacturer's instructions. Protein concentrations were determined, and then the neurokinin-1 receptor levels in the same amounts of cytosolic and membrane proteins were analyzed separately by a Western blot analysis.

At the end of the SP release experiments, the cell samples were processed for Western blot analysis as previously described [6]. Primary antibodies were raised against COX-2 (1:1,000 dilution; COX-2 polyclonal antibody; Cayman Chemical, Ann Arbor, MI), β-actin (1:5,000 dilution; the mouse monoclonal antibody for β-actin; Sigma) or anti-substance P receptor (1:2,000 dilution; Sigma). The horseradish peroxidase-conjugated anti-rabbit and anti-mouse secondary antibodies (1:2,000 dilution; Cell Signaling Technology, Beverly, MA) were used for chemiluminescence detection according to the manufacturer's instructions, respectively.