Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The relationships between steroid hormones and brain functions have mostly been considered within the framework of endocrine mechanisms as genomic responses elicited by secretory products from steroidogenic endocrine glands, transport through bloodstream, and exerting actions on the brain. Since the biosynthesis of steroid hormones in the brain (Compagnone and Mellon, 2000) and their rapid nongenomic actions (Harrison and Simmonds, 1984; Maggi and Perez, 1984) were reported, specific targets for so-called "neurosteroids" in plasma membranes have been postulated. From current studies the neurosteroids were found to have allosteric actions on ligand-gated channels such as -aminobutyric acid_A (Majewska, 1999), NMDA (Gibbs et al., 1999), and nicotinic receptors (Buisson and Bertrand, 1999). In addition to these findings, however, a series of reports has shown that -receptor, a drug receptor, might be another target of the neurosteroids (Maurice et al., 1999). In these studies, neurosteroids and -compounds showed allosteric modulations of NMDA responses (Monnet et al., 1995).

The variety of pharmacological studies, including ligand binding and drug discrimination studies, suggested the existence of at least two subtypes of -receptors, termed 1 and 2 (Quirion et al., 1992). The pharmacological actions through the 1 subtype receptor are characterized to have marked (+)-ligand stereoselectivity and PTX sensitivity (Itzhak and Stein, 1991; Monnet et al., 1992). In addition, some neurosteroids have affinity to this site. The 2 subtype, on the other hand, has different stereoselectivity from the 1 subtype and shows no PTX sensitivity (Hellewell et al., 1994; Bowen et al., 1995). Most recent attempt has resulted in the purification from guinea pig liver of [³H]pentazocine binding proteins, 1 binding protein (SBP) with the pharmacological profile of the 1 subtype and the cloning of the corresponding cDNA (Hanner et al., 1996). However, because this protein with single transmembrane structure appears to be different from typical G protein-coupled receptors, we have speculated that another 1-receptor different from SBP exists in the brain and it is coupled with G_i from reconstitution experiments using brain membranes and recombinant G proteins (Tokuyama et al., 1999). Recently, we also found that the intraplantar (i.pl.) injection of 1 agonists caused nociceptive flexor responses through G_i-mediated mechanisms. Here, we report that neurosteroids share pharmacological actions with the proposed metabotropic -receptor in in vivo peripheral nociception tests, and that there also might exist another type of neurosteroid receptor.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male ddY mice or mutant mice weighing 20 to 22 g were used. They were kept in a room maintained at 21 ± 2°C with free access to a standard laboratory diet and tap water in a group of 10 animals. Procedures were approved by Nagasaki University Animal Care Committee and complied with the recommendations of the International Association for the Study of Pain (Zimmermann, 1983).

Drugs. Neurosteroids, such as DHEAS, pregnenolone sulfate (PREGS), progesterone (PROG), histamine, and diphenhydramine (DPH) were purchased from Sigma (St. Louis, MO). (+)-Pentazocine and N,N-dipropyl-2-(4-methoxy-3-(2-phenylethoxy)phenyl)-ethylamine monohydrochloride (NE-100) were synthesized in Santen Pharmaceutical Co. (Osaka, Japan). (+)-Pentazocine, (+)-3-(hydroxyphenyl)-N-(1-propyl)piperidine, BD1063, pertussis toxin (PTX), U-73122, U-73343, and thapsigargin were purchased from Funakoshi (Tokyo, Japan). CP-99994 and CP-100263 were generously provided by Pfizer Pharmaceuticals (Tokyo, Japan). All drugs were dissolved in physiological saline. G_i1 antisense oligodeoxynucleotide (AS-ODN: 5'-AGA CCA CTG CTT TGT A-3') or G_i1 missense oligodeoxynucleotide (MS-ODN: 5'-AGC ACA CGT CTT GTT-3') were synthesized, freshly dissolved in physiological saline, and used for intrathecal (i.t.) injection in a volume of 2 µl on the 1st, 3rd, and 5th day. SBP AS-ODNs (5'-CCA CGG CAT TCT AGC GGG CA-3') were synthesized, freshly dissolved in physiological saline, and used for i.t. injection in a volume of 2 µl on the 1st, 2nd, 3rd, 4th, and 5th day. On the 6th day flexor responses were tested. Other details were as previously reported (Ueda and Inoue, 2000).

Evaluation of Nociceptive Flexor Response. All experiments were performed in compliance with the relevant laws and institutional guidelines. Experiments were performed, as described previously (Inoue et al., 1998; Ueda, 1999; Inoue and Ueda, 2000; Ueda and Inoue, 2000). Briefly, mice were lightly anesthetized with ether and held in a cloth sling with their four limbs hanging free through holes. The sling was suspended on a metal bar. All limbs were tied with strings, and three were fixed to the floor, whereas the other one was connected to an isotonic transducer and recorder. Two polyethylene cannulae (0.61 mm in outer diameter) filled with drug solution connected to a microsyringe were inserted into the planta of hind paw. Because we used light and soft polyethylene cannulae, they did not fall off the paw during the experiments. Because the intensity of flexor responses differs from mouse to mouse, we used the biggest response among spontaneous and nonspecific flexor responses occurring immediately after cannulation as the maximal reflex. One cannula was filled with -agonists, neurosteroids, or saline, and the other with inhibitors or antagonists. All experiments were started after complete recovery (20-30 min) from the light ether anesthesia and i.pl. injection of saline did not show any significant flexor responses. Neurosteroids or -compounds were given i.pl. 10 and 5 min before and 5, 10, 20, and 30 min after inhibitor (or antagonist) or vehicle injection. In most experiments, the results were expressed as percentage of analgesia, using the following equation: [1  neurosteroid-induced response (in amplitude) after test drug administration/the average of twice control responses) × 100 (%)]. All animals were used for only one experiment by an observer who did not know what kind of pretreatments had been given.

Immunoblot Analysis. Dorsal root ganglions (DRGs) were removed from mice immediately after the use for nociception tests and homogenized in sample buffer (2% SDS, 10% glycerol, 50 mM Tris-HCl, pH 6.8). SDS-PAGE by using 12% polyacrylamide gel and immunoblot analysis was performed as described (Yoshida and Ueda, 1999). The proteins separated by SDS-PAGE were transferred on Immobilon (Nihon Millipore Ltd., Yonezawa, Japan). Antisera for G_i1 were purchased from PerkinElmer Life Science Products (Boston, MA) (AS/7). Anitsera for SBP were raised in New Zealand White rabbits against a synthetic docosapeptide (CSEVFYPGETVVHGPGEATAVE) (Yamamoto et al., 1999), which detected only 29-kDa protein in the microsomal, mitochondrial, and synaptosomal fractions from rat brains. To visualize immunoreactive bands, an enhanced chemiluminescent substrate for detection of horseradish peroxidase, SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, IL), was used. The intensities of immunoreactive bands were analyzed by NIH imaging after scanning exposed films.

Statistical Analysis. Statistical analyses between two groups were conducted using a two-tailed, unpaired Student's t test. Analyses among multigroup data were conducted using analysis of variance, followed by Student-Newman-Keuls test. The criterion of significance was set at p < 0.05. All results are expressed as the mean ± S.E.M.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Peripheral Flexor Responses by -Agonists and Their in Vivo Signaling. The local application of 100 pmol of SA4503, a specific 1 agonist (Matsuno et al., 1996), into the planta of the hind limb produced a very short-acting flexor response, and stable responses were observed when this compound was repeatedly challenged at 5-min intervals (Fig. 1A). As shown in Fig. 1B, the consecutive challenges of increasing doses of SA4503 showed dose-dependent responses between 0.1 and 100 pmol (i.pl.). Similar dose-dependent flexor responses were observed with (+)-pentazocine and (+)-3-(hydroxyphenyl)-N-(1-propyl)piperidine, other representative 1 agonists, whereas not with either NE-100 or BD1063, both representative 1 antagonists (Matsumoto et al., 1995). SA4503 (100 pmol)-induced responses were antagonized in a dose-dependent manner by NE-100 in doses of 100 pmol and 1 nmol (Fig. 1C). The complete antagonism of (+)-pentazocine (100 pmol) responses by NE-100 or BD1063 was also observed when the responses at 30 min after the antagonist challenge were compared (Fig. 1D).

View larger version (32K): [in this window] [in a new window]   Fig. 1.   -Receptor-mediated nociceptive flexor responses. A, representative trace of SA4503-induced flexor responses. SA4503 (100 pmol) was given intraplantarly every 5 min, as indicated by the arrow. B, dose-response curves of several -ligands. Results (percentage of maximal reflex) represent the mean ± S.E.M. from at least six separate experiments. C, blockade of SA4503-induced responses by NE-100. Control was evaluated by the average of twice SA4503-induced responses in the beginning of each experiment. NE-100 (NE) or vehicle (Veh) was pretreated intraplantarly 5 min after the second challenge of SA4503, and the results showed the SA4503-induced response (percentage of control) at each time after the pretreatment. Data represent the mean ± S.E.M. from at least six separate experiments. D, blockade of (+)-pentazocine-induced responses by NE-100 or BD1063. Experiments were carried out, as shown in C. Results represent the (+)-pentazocine-induced responses 30 min after Veh, NE, or BD1063 (BD) injection, and data are expressed as the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with vehicle treatment.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   In vivo signaling of -agonist-induced responses. A, PTX sensitivity. PTX was pretreated at a dose of 10 ng (i.pl.). Data (percentage of control) represent the mean ± S.E.M. from at least six separate experiments. B, blockade of SA4503-induced responses at various doses by the treatment with AS-ODN for Gi1. Results (percentage of maximal reflex)represent the mean ± S.E.M. from six separate experiments. Inset, representative data of Western blot analysis. C, blockade by PLC inhibitor. U-73122 (a PLC-inhibitor) or U-73343 (its inactive isomer) was pretreated at a dose of 10 pmol (i.pl.). D, lack of effects of thapsigargin. Experiments were carried out, as shown in A and D. Results represent SA4503-induced responses 30 min after vehicle (Veh) or thapsigargin (Thap) injection, and data are expressed as the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with vehicle-treatment. E and F, involvement of NK1 receptor activation in SA4503- (E) or (+)-pentazocine-induced responses (F). CP-99994 (an NK1 receptor antagonist) or CP-100263 (its inactive isomer) was pretreated at a dose of 10 pmol (i.pl.). Data (percentage of control) represent the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with vehicle treatment. G, lack of involvement of SBP in (+)-pentazocine responses. Results (percentage of maximal response) represent the mean ± S.E.M. from six separate experiments. Inset, representative data of Western blot analysis. Other details are described in the legend of Fig. 1.

Peripheral Flexor Responses by Neurosteroids and Their in Vivo Signaling. DHEAS in a dose of 1 fmol produced stable flexor responses upon repeated challenges at 5-min intervals for 30 min (data not shown). The DHEAS (1 fmol)-induced response was observed within 10 s after the cessation of infusion (Fig. 3A). Similar responses were also observed with PREGS. DHEAS- or PREGS-induced responses were dose-dependent between 0.1 and 1000 or 0.1 and 100 amol, respectively (Fig. 3B). When PROG (1 fmol) was given after twice control DHEAS (1 fmol) challenges, the flexor responses were rapidly attenuated (Fig. 3A), and complete reduction was observed 20 or 30 min after the PROG infusion (Fig. 3C). The injection of 1 fmol of PROG did not produce any nociceptive responses (data not shown). The dose-response curve with DHEAS was shifted to the right in the presence of 10 amol of PROG (Fig. 3D). However, DHEAS (1 fmol)-induced responses were not affected by NE-100 in a dose of 1 pmol (data not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Neurosteroid-induced nociceptive flexor responses. A, representative trace of DHEAS-induced flexor responses and their blockade by progesterone. DHEAS (1 fmol) or PROG (1 fmol) was given intraplantarly, as indicated by the arrow. B, dose-response curves of DHEAS- or PREGS-induced responses. Results (percentage of maximal reflex) represent the mean ± S.E.M. from at least six separate experiments. C, antagonism of DHEAS-induced responses by PROG. Data (percentage of control) represent the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with vehicle (Veh) treatment. D, competitive blockade of DHEAS-induced responses by PROG. Results represent the DHEAS-induced dose-dependent responses 30 min after Veh or PROG, and data are expressed as the mean ± S.E.M. from at least six separate experiments. E, PTX insensitivity and PLC sensitivity. Results represent DHEAS-induced responses 30 min after vehicle (Veh), PTX (10 ng), U-73122 (10 pmol), or U-73343 (10 pmol) injection, and data are expressed as the mean ± S.E.M. from at least six separate experiments. *p < 0.05, compared with Veh treatment. F, blockade by thapsigargin. Thapsigargin was pretreated at a dose of 100 pmol (i.pl.). Other details are given in the legend of Fig. 1.

Blockade of Low Dose of DHEAS-Induced Response by DPH. On the analogy of the flexor responses by Compound 48/80, a histamine releaser from mast cells (Ueda and Inoue, 2000), which were blocked by thapsigargin, DPH, a representative histamine H1 receptor antagonist, was tested to see any influence on DHEAS-induced flexor responses. As shown in Fig. 4A, both histamine- (100 pmol) and DHEAS (1 fmol)-induced flexor responses were inhibited by 200 pmol of DPH.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   DHEAS-responses in the presence of DPH. A, blockade of histamine or low dose of DHEAS (1 fmol)-induced responses by DPH. Results represent histamine- (100 pmol) or DHEAS (1 fmol)-induced responses 20 min after vehicle (Veh) or DPH (200 pmol) injection, and data are expressed as the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with Veh treatment. B, dose-dependent responses by higher doses of DHEAS in the presence of DPH. Results represent DHEAS-induced dose-dependent responses 20 min after Veh or DPH (200 pmol) injection, and data are expressed as the mean ± S.E.M. from at least six separate experiments. C, blockade by NE-100 on DHEAS-induced responses in the presence of DPH. Results represent DHEAS-induced responses in the presence of DPH 20 min after vehicle (Veh) or NE-100 (NE; 1 nmol) injection. D, blockade of (+)-pentazocine-induced responses by PROG. Results represent (+)-pentazocine-induced responses 20 min after Veh or PROG (1 nmol) injection. E and F, inhibition of xestospongin C (E) and EGTA (F) on DHEAS-induced responses in the presence of DPH. Xestospongin C or EGTA was pretreated at a dose of 10 pmol (i.pl.) or 2 nmol, respectively. All data are expressed as the mean ± S.E.M. from at least six separate experiments. *p < 0.05, **p < 0.01, compared with Veh treatment. Other details are given in the legend of Fig. 1.

Flexor Responses by Higher Doses of DHEAS Insensitive to DPH. To examine the neuronal contribution to DHEAS-induced flexor responses, experiments using a wide range of DHEAS doses were carried out in the presence of 200 pmol of DPH. DHEAS caused a dose-dependent flexor response in doses between 1 pmol to 1 nmol, whereas it did not in doses between 1 amol and 100 fmol (Fig. 4B). The repeated challenges of DHEAS (1 nmol) in the presence of 200 pmol of DPH caused stable flexor responses for 30 min (data not shown). Such responses by DHEAS (1 nmol) in the presence of DPH were significantly inhibited by 1 nmol of NE-100 (Fig. 4C). On the other hand, (+)-pentazocine-stimulated flexor responses were also blocked by 1 nmol of PROG (Fig. 4D). Table 1 shows a close similarity in characteristics between DHEAS-induced flexor responses in the presence of DPH and SA4503-induced ones. The DHEAS (1 nmol)-induced responses in the presence of DPH were blocked by PTX (10 ng), U-73122 (10 pmol), and CP-99994 (10 pmol), but not by their inactive isomers U-73343 (10 pmol) or CP-100263 (10 pmol). Further experiments revealed that they were blocked by xestospongin C (Fig. 4E) and EGTA (Fig. 4F).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Dehydroepiandrosterone and its metabolite DHEAS elicited rapid nongenomic actions. Most of underlying mechanisms have been postulated to be through allosteric modulation of ionotropic receptors (Buisson and Bertrand, 1999; Gibbs et al., 1999; Majewska, 1999). Similarly, there are also accumulating reports that neurosteroids share pharmacological actions with -compounds (Monnet et al., 1994, 1995). In such studies, the mode of action of neurosteroids and -compounds is also conceived to be allosteric modulation of NMDA responses. Furthermore, the -agonist effect on NMDA response was abolished by PTX treatment (Monnet et al., 1994). These findings suggest that neurosteroids and -compounds share the same receptors, which may be coupled with PTX-sensitive G protein(s). This view was supported by the present experiments. We have used this test for assessing many biologically active substances (Inoue et al., 1998; Ueda, 1999; Inoue and Ueda, 2000; Ueda and Inoue, 2000), and we could study the in vivo signaling of receptor-mediated flexor responses, on the analogy of nociceptin mechanisms (Inoue et al., 1998) (Fig. 5). Here, we could also detect the flexor responses by -agonists. Because the response was abolished by treatment with PTX, it is suggested that G_i-coupled 1-receptor is involved in this action. Further experiments suggested that this metabotropic 1-receptor also shares common signaling mechanisms through PLC and SP release with nociceptin receptor (Inoue et al., 1998), being consistent with previous reports (Morin-Surun et al., 1999). Therefore, it is suggested that the site of action is likely attributed to nociceptor endings of SP-containing polymodal C fiber. This view was further supported by two other findings with AS-ODN for G_i1 and thapsigargin. As shown in Fig. 2B, the AS-ODN pretreatments (i.t.) abolished the SA4503-induced responses. Since the intrathecally administrated ODN is unlikely to be diffused to peripheral sites, it is evident that the loss of SA4503 responses is attributed to the reduction of G_i1 expression in nociceptor endings as well as DRGs (Fig. 2B). Similar discussion has also been done elsewhere (Ueda, 1999). Most recently, we have reported that thapsigargin blocked Compound 48/80-induced histamine release from mast cells, but did not affect the nerve ending mechanism through inositol trisphosphate (InsP₃) receptor (Ueda and Inoue, 2000). This difference reflects that thapsigargin selectively inhibits Ca²⁺ mobilization through InsP₃ receptor channel in peripheral cells, but not in nerve endings of sensory neurons. We have proposed the existence and function of plasma membrane-type InsP₃ receptor rather than endoplasmic reticulum-type one in biochemical experiments measuring ⁴⁵Ca²⁺ release from resealed vesicles (Ueda et al., 1996). In addition, we have observed the existence of InsP₃ receptor on the plasma membrane of a peripheral nerve ending by immunoelectron microscopic experiments (Shimohira et al., 2000). Moreover, the immunoelectron microscopic existence of plasma membrane-type IP₃ receptor has been also demonstrated in monkey and rat photoreceptor outer segments (Wang et al., 1999). Thus, the lack of effects of thapsigargin also strengthens the view that the site of action of -agonists is on the presynaptic nociceptor endings (Fig. 5).

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Proposed model of peripheral pain-producing mechanism. DHEAS has two sites of actions, on the mast cell (NS2 site) and on the nociceptor ending (NS1/ site) containing SP. -Compounds such as SA4503 and (+)-pentazocine share the latter site, which is coupled with Gi and PLC. A small number of InsP3s (IP3) generated by PLC, which is weakly stimulated by Gi, gates InsP3 receptor (IP3R) channel to cause Ca2+ influx. This leads to a release of a number of SP molecules that in turn stimulate NK1 receptor, which couples to Gq/11 and PLC. Much more InsP3s are generated by PLC, which is strongly stimulated by Gq/11, causing intense Ca2+ influx. Following mechanisms (indicated by dotted arrows) through voltage-dependent Ca2+ and Na+ channels are expected to generate action potentials, although sufficient data have not yet been obtained. Other details are described in the text.

Similar approaches were then carried out with neurosteroids. Both DHEAS and PREGS induced flexor responses. However, the doses required for significant actions were extremely low compared with agonists. Compared with the dose of compounds required for inducing 50% of maximal response, it was 30 pmol for -agonists, and 1 and 10 amol for PREGS and DHEAS, respectively (Fig. 3B). DHEAS-induced action was blocked by an equimolar amount of PROG. However, marked differences were observed in the sensitivity to various inhibitors. U-73122 blocked this action but PTX did not. Thapsigargin also showed a potent blockade, thus suggesting the involvement of non-neuronal peripheral mechanisms. Indeed, DHEAS-induced flexor responses were abolished by DPH. Thus, it is evident that DHEAS-induced action was mediated through a histamine release from mast cells (Fig. 5). Collectively, it is suggested that nociceptive responses by low doses of neurosteroids are mediated through different mechanisms from the -agonist-induced ones. We called this putative neurosteroid site in mast cells NS2 receptor, which is stimulated by low doses (1 amol-10 fmol) of neurosteroids, and distinguished it from NS1/-receptor, which exists on nociceptor endings and is stimulated by higher doses (1 pmol-1 nmol) of neurosteroids or -agonists. Because of the rapid flexor responses by neurosteroids, the NS2 receptor is characterized to be a nongenomic receptor. In addition, the addition of U-73122 or thapsigargin significantly attenuated the responses, but PTX had no effect on it, G_q, or other unknown PLC-activating mechanisms and Ca²⁺ signaling would be involved (Fig. 5).

The possibility that other chemical mediators are involved in the nociception by lower doses of neurosteroids cannot be excluded. However, because we did not intend to damage the tissue to release inflammatory mediators, it is more likely that histamine released from mast cells is the best candidate of various mediators, including prostaglandins, potassium, and ATP from many damaged cells in the vicinity of nociceptor endings and 5-hydroxytryptamine from platelets in blood vessels. The complete blockade of such responses by DPH may strengthen this view. On the other hand, we demonstrated that the nociception by 1000 times higher doses of neurosteroids in the presence of H1 receptor antagonist may be attributed to an SP release from nociceptor endings, and they share the common receptor mechanisms to -agonists. As mentioned above, we called this site NS1/ receptor. As seen in Figs. 4C and 3D, the responses were blocked by -antagonist NE-100, whereas -agonist actions were antagonized by PROG. The pharmacological characterization by use of signal transduction-modifying drugs, such as xestospongin C, EGTA (Fig. 4, E and F), PTX, and PLC inhibitor (Table 1) also supported this view. SP as well as glutamate is a representative neurotransmitter related to pain transmission in polymodal nociceptor endings, and we have a number of data supporting this view in experiments involving intrathecal injection of NK1 antagonist. In such studies the nociception induced by bradykinin, a representative pain-producing substance, histamine, or SP was all abolished by the intrathecal injection of CP-99994, an NK1 antagonist (Inoue et al., 1999; Ueda and Inoue, 2000; Ueda et al., 2000). As shown in Table 1, DHEAS plus DPH-induced nociception was completely blocked by NK1 antagonist, just like in the case with -agonists. All these findings strongly support the hypothesis that NS1/-receptor is on the nociceptor endings and mediates the nociceptive responses by -agonists and high doses (1 pmol-1 nmol) of neurosteroids.

It should be noted that PROG blocked both neurosteroid actions through NS1/ and NS2 receptors. In the previous studies, PROG antagonized 1-receptor-mediated responses in NMDA-evoked [³H]norepinephrine release (Monnet et al., 1995), and PREG-stimulated tubulin polymerization through a microtubule-associated protein-2 (Murakami et al., 2000). Thus, the present study might add another target, NS2 site for the antagonist activity by PROG.

Our concern is to know whether endogenous neurosteroids may also stimulate the nociceptor ending through metabotropic 1-receptor or NS1/-receptor, although it is plausible that NS2 receptor-mediated mechanisms would be driven in vivo because of high sensitivity to neurosteroids. We have to consider two sources for neurosteroids: a peripheral tissue-derived one and an extravasated one from the blood vessels. Recently, it was reported that peripheral-type benzodiazepine receptor, a key molecule essential for neurosteroid biosynthesis (Krueger and Papadopoulos, 1990), is expressed in Schwann cells in myelinating sciatic nerves, and its expression increases with the nerve injury (Lacor et al., 1999). On the other hand, the plasma concentration of DHEAS or PROG was reported to be 1 or 10 nM, respectively (Schwarz et al., 1989; Robel et al., 1999). However, it remains to be determined how much of extravasated neurosteroids correspond to local concentrations for nociceptor endings.

The next concern is to clarify whether SBP can recognize neurosteroids. The treatment with AS-ODN for SBP did not attenuate the -agonist-induced flexor responses. On the other hand, we have obtained data that recombinant SBP did not show the -agonist-stimulation of guanosine-5'-O-(3-thio)triphosphate binding to G_i1, G_i2, G_s, G₁₁, or G_oA in baculovirus/Sf21 cell expression system (Ueda et al., 2001). Taking the present finding into account that -agonist-induced flexor responses were abolished by AS-ODN for G_i1, it is evident that SBP is not involved in the NS1/-receptor-mediated mechanisms.

In conclusion, we revealed two types of neurosteroid receptors coupling to PLC/Ca²⁺ mobilization pathway in the peripheral nociception test. One is the receptor that recognizes both neurosteroids and -compounds, which has been previously postulated (Su et al., 1988; Monnet et al., 1992), whereas the other would be a novel type. The characterized properties here will be valuable for future study of cloning of these neurosteroid receptors.