Introduction

Top Abstract Introduction Methods Results Discussion References

Substance P has long been known to be an important neurotransmitter implicated in the transmission of nociception (Henry, 1980). It is a member of the tachykinin family of neuropeptides that include neurokinin A and neurokinin B. Substance P is co-localized with neurokinin A in small diameter, high threshold nociceptive C-fibers and is released in response to a variety of noxious stimuli (Otsuka and Yoshioka, 1993). Three subclasses of G-protein coupled tachykinin receptors have been described in the CNS and periphery: NK₁, NK₂ and NK₃ with substance P, neurokinin A and neurokinin B as their respective preferred endogenous ligands (Mussap et al., 1993). The NK₁ receptor is widely distributed in the central nervous system (Otsuka and Yoshioka, 1993). In the spinal cord, NK₁ binding sites are widely distributed throughout the dorsal (particularly laminae I and II where substance P containing primary afferent neurones terminate) and ventral horn (Yashpal et al., 1990).

The recent development of nonpeptide, highly selective NK₁ receptor antagonists that penetrate into the CNS has helped to further elucidate the role of substance P in nociception. Most of these studies have focused on animal models of inflammatory pain. Thus, it has been reported that selective but chemically unrelated NK₁ receptor antagonists such as, CP 99,994, SR 140333, L-733,060 and LY303870 can dose-dependently block the second phase of the formalin response (Iyengar et al., 1997; Rupniak et al., 1996; Seguin et al., 1995). Other studies have shown that NK₁ receptor antagonists can also prevent the development of inflammation-induced thermal and mechanical hyperalgesia (Traub, 1996; Sluka et al., 1997). However, the role of substance P in other types of pain remains to be elucidated.

Conventional analgesics, such as opiates and NSAIDS have limited therapeutic value in the management of diabetes-induced neuropathic pain (Galer, 1995; James and Page, 1994). This has led to the use of adjuvant analgesics for the management of this condition. At present tricyclic antidepressants are currently the first choice in the treatment of painful diabetic neuropathy (Galer, 1995; James and Page, 1994). However, no agent is fully effective in all patients and undesirable side effects are common (Galer, 1995; James and Page, 1994). Streptozocin is a selective pancreatic -cell toxin that has been used to induce experimental diabetes in laboratory animals (Tomlinson et al., 1992). The resultant loss of endogenous insulin induced by streptozocin mimics the characteristics of type I or insulin-dependent diabetes. Streptozocin-induced diabetes has recently been described as a model of chronic pain. It has been reported that streptozocin administration leads to mechanical, thermal and chemical hyperalgesia as well as mechanical hypersensitivity detected using von Frey hairs (Courteix et al., 1993a; Calcutt et al., 1996). The mechanical hyperalgesia induced by streptozocin is sensitive to tricyclic antidepressants, clonidine and lidocaine treatment (Courteix et al., 1994). However, it is less sensitive to aspirin and morphine (Courteix et al., 1994). Thus, these profiles mimic their respective clinical activities observed in the treatment of diabetic neuropathy. The most common symptoms of diabetic neuropathy appear to be spontaneous burning pain and mechanical hypersensitivity (where normally innocuous tactile stimuli induce pain) in the feet or lower limbs. Our study investigates the role of peripheral and central NK₁ receptors in the maintenance of diabetesinduced mechanical hypersensitivity in the rat.

	Methods

Top Abstract Introduction Methods Results Discussion References

Mongolian gerbils (40-70 g) of either sex (Bantin and Kingman, Hull, UK) were housed in groups of eight. Male Sprague Dawley rats (200-250 or 250-300 g), obtained from Bantin and Kingman were housed in groups of six. Male Dunkin-Hartley guinea pigs (350-400 g) were obtained from Interfauna (Huntington, UK). All animals were kept under a 12-hr light/dark cycle (lights on at 07 hr 00 min) with food and water ad libitum. All experiments were carried out by an observer blind to drug treatments.

Receptor Binding Assays

Tachykinin receptor binding assays. Tachykinin NK₁ receptor binding assays were carried out as described previously (Boyle et al., 1994). Human lymphoma IM9 cells were grown in RPMI 1640 culture medium supplemented with 10% fetal calf serum and 2 mM glutamine and maintained under an atmosphere of 5% CO₂. Cells were harvested for experiments by centrifugation at 1000 × g for 3 min. Pelleted cells were washed once by resuspension into assay buffer (50 mM Tris HCl pH 7.4, 3 mM MnCl₂, 0.02% bovine serum albumin, 40 µg/ml bacitracin, 2 µg/ml chymostatin, 2 µM phosphoramidon, 4 µg/ml leupeptin) and repeating the centrifugation step before resuspending at a concentration of 2.5 × 10⁶ cells/ml assay buffer. Cells (200 µl) were incubated with [¹²⁵I]Bolton-Hunter substance P (0.05-0.1 nM) in the presence and absence of varying concentrations of test compounds for 50 min at 21°C. Nonspecific binding (10% of the total binding observed under these conditions) was defined by 1 µM [Sar⁹,Met(0₂)¹¹]substance P. Reactions were terminated by rapid filtration under vacuum onto GFC filters presoaked in 0.2% polyethylene imine for 1 to 2 hr, using a Brandel cell harvester. Filters were washed with 6 × 1 ml ice-cold Tris HCl (50 mM, pH 7.4) and bound radioactivity determined using a gamma counter. Results were analyzed using iterative curve fitting procedures in RS1 (BBN Software Products, Cambridge, MA) or Graphpad Inplot. Binding affinity data were expressed as the geometric mean and range of IC₅₀ values determined from three to six separate experiments unless otherwise indicated.

Selectivity assays. To determine selectivity, 23 binding assays for non tachykinin receptors or binding sites were carried out using standard methodology as described in the literature. The species and radioligands used are shown in table 1.

Electrophysiology

Coronal slices of brainstem 350-µm thick containing the locus ceruleus were cut from the brains of guinea pigs. One such slice was placed in a Perspex recording chamber where it was pinned down on to a Sylgard base. The slice was placed in a perfusion chamber and perfused with ACSF flowing at 4 ml/min. The ACSF contained (in mM) NaCl 126; KCl 5; NaH₂PO₄ 1.2; MgCl₂ 1.3; CaCl₂ 2.4; NaHCO₃ 26 and glucose 10. It was gassed with a carbogen mixture (95% oxygen and 5% carbon dioxide). All experiments were performed at 37°C. Extracellular recordings were made from spontaneously firing individual neurones in the locus ceruleus using low resistance (5 -10 megohm when filled with ACSF) glass microelectrodes. Details of the subsequent data analysis have been reported elsewhere (Boden et al., 1991). The agonist, in this case [Sar⁹ Met(O₂)¹¹] substance P, was applied for 1 min. Slices were pretreated with antagonists for 30 min before addition of agonist in the continued presence of the antagonist.

Induction of [Sar⁹,Met(O₂)¹¹] Substance P-Induced Foot Tapping

Gerbils were briefly anesthetized with an isoflurane O₂/NO₂ mixture. An incision was made into the scalp to expose the skull. [Sar⁹,Met(O₂)¹¹] substance P was administered i.c.v. in a volume of 5 µl by vertical insertion of a cuffed 27-gauge needle to a depth of 4.5 mm below bregma. Animals were placed individually into observation boxes and duration of hind paw tapping was recorded for 5 min immediately after recovery of the animals righting reflex. For antagonism studies PD 156982 (1-100 mg/kg) was administered s.c. at 0.25, 0.5, 1 and 2 hr before, or i.c.v. (10-100 µg/animal) 0.25 h before, the injection of [Sar⁹,Met(O₂)¹¹] substance P.

Effect of CI-1021 on Hypersensitivity Induced by Central Administration of Substance P or [Ala⁸]NK A(4-10)

Rats (200-250 g) were administered with substance P or [Ala⁸]NK A(4-10) i.t., in a volume of 10 µl using a 100 µl Hamilton syringe, by exposing the spine under brief isoflurane O₂/NO₂ anesthesia. Injections of agonists were made into the i.t. space between lumbar region 5-6 with a 10 mm long 27-gauge needle, with penetrations being judged successful by a tail flick response. The wound was sealed with an autoclip and rats appeared fully awake within 2 to 3 min after injection. Baseline PWT to von Frey hairs were determined before drug treatment. CI-1021 (formally PD 154075) or its vehicle was administered s.c. 30 min before i.t. substance P (8 µg/animal) or [Ala⁸]NK A(4-10) (0.2 µg/animal). PWT were determined again at various times up to 1 hr post i.t. injection.

Effect of PD 156982 on Hypersensitivity Induced by Administration of Substance P into the Hind Paw of Rats

Rats (200-250 g) were administered with substance P in a volume of 50 µl into the plantar surface of the left hind paw using a 100 µl Hamilton syringe. Baseline PWT, using von Frey hairs were determined before drug treatment. PD 156892 or its vehicle was administered s.c. 30 min before intraplantar substance P (8 µg/animal). PWT were determined again at 10 and 20 min post i.t. injection.

Development of Diabetes in the Rat

Diabetes was induced in 40 rats (250-300 g) by a single i.p. injection of streptozocin (75 mg/kg) as described previously (Courteix et al., 1993a). Control animals (n = 8) received a similar administration of isotonic saline.

Effect of CI-1021 and PD 156982 on the Maintenance of Streptozocin-Induced Mechanical Hypersensitivity

CI-1021 was examined between 11 to 12 days post streptozocin. PD 156982 was examined in the same set of animals between 14 to 15 days post streptozocin. On each test day baseline PWT to von Frey hairs were determined before drug treatment. CI-1021 (3-100 mg/kg), PD 156982 (10-100 mg/kg) or their vehicle was administered s.c. and PWT reexamined every hour for up to 5 hr.

Evaluation of Mechanical Hypersensitivity

Mechanical hypersensitivity was measured using Semmes-Weinstein von Frey hairs (Stoelting, Wood Dale, IL). Animals were placed into wire mesh bottom cages allowing access to the underside of their paws. Animals were habituated to this environment before the start of the experiment. Mechanical hypersensitivity was tested by touching the plantar surface of the animals hind paw with von Frey hairs in ascending order of force (0.7, 1.2, 1.5, 2, 3.6, 5.5, 8.5, 11.8, 15.1 and 29 g) for up to 6 sec. Once a withdrawal response was established, the paw was retested, starting with the next descending von Frey hair until no response occurred. The highest force of 29 g lifted the paw as well as eliciting a response, thus represented the cut-off point. The lowest amount of force required to elicit a response was recorded as the PWT in grams.

Drugs Used

CI-1021 and PD 156982 were synthesized at Parke-Davis Neuroscience Research Centre (Cambridge, UK). Both compounds were dissolved in polyethyleneglycol-200 for systemic administration. For i.c.v. administration PD 156982 was dissolved in hydroxypropyl--cyclodextrin (20% in saline). [Sar⁹,Met(O₂)¹¹] substance P (Sigma, Poole, UK) was dissolved in saline with 0.01% acetic acid. Streptozocin (Aidrich, Gillingham, UK), substance P (Sigma, Poole, UK) and [Ala⁸]NK A(4-10) (RBI, Natick, MA) were dissolved in 0.9% w/v NaCl. Systemic drug administrations were made in a volume of 2 ml/kg in gerbils and 1 ml/kg in rats. The i.c.v., i.t. and intraplantar administrations were made in respective volumes of 5, 10 and 50 µl.

Statistics

Data obtained from the foot tapping model were subjected to unpaired Student t test comparing to the relevant vehicle-treated group. The mechanical hypersensitivity results obtained using the von Frey hairs were subjected to an individual Mann Whitney U test and compared to vehicle-treated animals.

	Results

Top Abstract Introduction Methods Results Discussion References

Radioligand binding studies. PD 156982 exhibited high affinity for tachykinin NK₁ receptors present in human lymphoma IM9 cells with an IC₅₀ value (mean with minimum and maximum range shown in parentheses) of 1.4 nM (0.4-4.3). PD 156982 also possessed high affinity for NK₁ receptors in the guinea pig, dog, ferret and hamster, with IC₅₀ values of 9.6 (8.1-14), 9.6 (4.2-19), 21 (10-45) and 10 nM (5.2-36), respectively. In contrast, affinities of PD 156982 for rodent NK₁ receptors in the rat and mouse were at least two orders of magnitude lower than human receptors [IC₅₀ values 820 nM (630-1050) and 530 nM (340-820) respectively].

Electrophysiology. The locus ceruleus preparation contains functional NK₁ receptors which are excited in a concentration-dependent manner by the selective NK₁ agonist [Sar⁹ Met(O₂)¹¹]substance P. PD 156982 (300 nM) blocked the response to [Sar⁹ Met(O₂)¹¹]substance P in an agonist surmountable manner (fig. 1a), indicative of competitive antagonism, yielding an equilibrium constant (K_e) of 13.9 nM (fig. 1b). K_e values were within the range 13.9 to 36 nM for the series of experiments, in accord with the binding affinity for PD 156982 at the guinea pig and human NK₁ receptor.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of PD 156982 on the [Sar9Met(O2)11] Substance P-induced increase in firing rate of a locus ceruleus neurone. (a) PD 156982 antagonizes the concentration-dependent increase in the spontaneous firing rate of a locus ceruleus neurone produced by [Sar9 Met(O2)11] substance P. The trace shows a rate meter histogram of firing rate (Hz) plotted against time of the experiment (min). [Sar9 Met(O2)11] substance P produced a concentration dependent increase in firing with a peak effect at 10 nM. Application of PD 156982 (300 nM) led to a reduction in the basal firing rate and also a block of the [Sar9 Met(O2)11] substance P response. (b) Concentration-response curves drawn from the data in (a), showing that in this experiment the block produced by PD 156982 (300 nM) was surmountable with a Ke of 13.9 nM.

Effect of PD 156982 on hypersensitivity induced by administration of substance P into the hind paw of rats. Substance P (8 µg) injected into the plantar surface of the rat hind paw induced mechanical hypersensitivity which was present for at least 20 min post injection (table 2). The s.c. administration of PD 156982 dose-dependently (10-100 mg/kg) blocked the development of this hypersensitivity with a MED of 30 mg/kg (table 2).

Effect of PD 156982 on [Sar⁹,Met(O₂)¹¹] substance P-induced foot tapping in the gerbil. The dose of 30 nmol/animal of [Sar⁹,Met(O₂)¹¹] substance P was chosen for antagonism studies as it represents a submaximal dose (Singh et al., 1997). The systemic administration of PD 156982 (1-100 mg/kg, s.c.), failed to block the [Sar⁹,Met(O₂)¹¹] substance P-induced foot tapping response at any pretreatment time (fig. 2a). However, when PD 156982 was administered i.c.v. 15 min before [Sar⁹,Met(O₂)¹¹] substance P, it dose-dependently (10-100 µg/animal) antagonized the foot tapping response, with a MED of  10 µg/animal (fig. 2b).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of PD 156982 on [Sar9, Met(O2)11] substance P-induced foot tapping in the gerbil. PD 156982 was administered either (a) s.c. at various pretreatment times before, or (b) centrally at 0.25 hr before 30 nmol [Sar9, Met(O2)11]substance P (i.c.v). The duration of foot tapping was recorded for 5 min immediately after recovery from anesthesia. Vehicle treatments were administered for each separate time point in (a) and remained consistent throughout. For clarity only the vehicle response for the first time point is shown. Results are shown as the mean (vertical bars show S.E.M.) of 5 to 10 animals per group. *P < .05, **P < .01, Significantly different from the vehicle (Veh.) treated control group (unpaired Student t test).

Effect of CI-1021 on hypersensitivity induced by central administration of substance P or [Ala⁸]NK A(4-10). The i.t. administration of substance P or [Ala⁸]NK A(4-10) induced mechanical hypersensitivity which was present post injection for up to 40 and 60 min respectively (fig. 3). Pretreatment with CI-1021 dose-dependently (10-100 mg/kg, s.c.) blocked the development of substance P-induced hypersensitivity with a MED of 30 mg/kg (fig. 3a). The top dose of 100 mg/kg completely blocked the development of mechanical hypersensitivity (fig. 3a). In contrast, similar administration of 100 mg/kg CI-1021 failed to show any block of the [Ala⁸]NK A(4-10)-induced hypersensitivity (fig. 3b).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effect of CI-1021 on the development of mechanical hypersensitivity induced by central administration of either (a) substance P or (b) [Ala8]NK A(4-10). Baseline (BL) PWT to von Frey hairs were determined before drug treatment. CI-1021 or its vehicle was administered s.c. 30 min before i.t. substance P (8 µg/animal) or [Ala8]NK A(4-10) (0.2 µg/animal). Results are expressed as median force (g) required to induce a withdrawal in 8-10 animals per group (vertical bars represent first and third quartiles). *P < .05, **P < .01, ***P < .005 significantly different from ipsilateral paw of agonist + vehicle- (Veh.) treated animals (Mann Whitney U test).

Effect of CI-1021 and PD 156982 on diabetes-induced mechanical hypersensitivity. The single injection of streptozocin (75 mg/kg, i.p.) induced mechanical hypersensitivity which developed fully within 11 days post treatment. The administration of CI-1021 dose-dependently (3-100 mg/kg, s.c.) blocked the maintenance of this hypersensitivity with a MED of 10 mg/kg (fig. 4a). The dose of 100 mg/kg completely antagonized the mechanical hypersensitivity, with PWT similar to those of nondiabetic control animals (fig. 4a). The antihypersensitive action of 100 mg/kg CI-1021 was apparent for approximately 4 hr. In contrast, similar administration of PD 156982 (10-100 mg/kg, s.c.) failed to block the hypersensitivity exhibited by streptozocin animals (fig. 4b).

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of (a) CI-1021 and (b) PD 156982 on the maintenance of diabetes-induced mechanical hypersensitivity. Baseline (BL) PWT to von Frey hairs were determined before drug administration. CI-1021 or PD 156982 was administered s.c. and PWT reexamined for up to 5 hr. Results are expressed as median force (g) required to induce a paw withdrawal in eight animals per group (vertical bars represent first and third quartiles). *P < .05 **P < .01, ***P < .005 significantly different comparing vehicle- (Veh.) treated streptozocin group at each time point (Mann Whitney U test).

	Discussion

Top Abstract Introduction Methods Results Discussion References

PD 156982: a peripherally selective NK₁ receptor antagonist. The radioligand binding results presented here show that PD 156982 possesses nanomolar affinity for human and guinea pig NK₁ receptors. However, as with many other NK₁ receptor antagonists, it has approximately two orders of magnitude lower affinity for the rodent NK₁ receptor. This confirms previous reports that guinea pig and human tachykinin NK₁ receptors are similar to each other but different to rodent NK₁ receptors (Maggi, 1994). PD 156982 displays a high degree of selectivity for the human type NK₁ receptor over NK₂ and NK₃ receptor types. The electrophysiological data indicate that PD 156982 is a competitive antagonist at central NK₁ receptors in the guinea pig. It was found to be inactive in most binding assays, suggesting that it is a selective NK₁ receptor antagonist. Previously, it has been shown that central activation of NK₁ receptors induces a characteristic foot tapping response in the gerbil (Graham et al., 1993; Bristow and Young, 1994; Rupniak and Williams, 1994). The antagonism of this response by systemic administration of NK₁ receptor antagonists has been suggested to reflect their ability to penetrate into the brain (Graham et al., 1993; Bristow and Young, 1994; Rupniak and Williams, 1994). Thus, the failure of PD 156982 to block the foot tapping response after systemic administration indicates that it does not penetrate into the CNS.

Effect of peripherally selective and CNS penetrating NK₁ receptor antagonists on diabetes-induced mechanical hypersensitivity. CI-1021 is a selective NK₁ receptor antagonist, but unlike PD 156982, it can penetrate into the CNS (Singh et al., 1997). Our results show that systemic administration of CI-1021 can block the maintenance of diabetes-induced mechanical hypersensitivity. This is consistent with a recent study showing that the rat selective but chemically unrelated NK₁ receptor antagonist RP-67580 can reduce diabetes-induced mechanical hyperalgesia (Courteix et al., 1993b). Our results further show that PD 156982 can block mechanical hypersensitivity mediated by peripheral NK₁ receptors in the rat. This indicates that in vivo, PD 156982 is an effective antagonist at peripheral NK₁ receptors. Thus, its failure to block diabetes-induced hypersensitivity highlights the importance of the central NK₁ receptor in the maintenance of this state of hypersensitivity.