- Email: [email protected]
Somatostatin receptors on peripheral primary afferent terminals: inhibition of sensitized nociceptors
Pain 90 (2001) 233±244
www.elsevier.nl/locate/pain
Somatostatin receptors on peripheral primary afferent terminals: inhibition of sensitized nociceptors Susan M. Carlton a,*, Junhui Du a, Elyad Davidson b, Shengtai Zhou a, Richard E. Coggeshall a a
Department of Anatomy and Neurosciences, Marine Biomedical Institute, University of Texas Medical Branch, Galveston, TX 77555-1069, USA b Department of Anesthesia and Critical Care Medicine, Hadassah Hebrew University, School of Medicine, Jerusalem, Israel Received 6 March 2000; received in revised form 30 June 2000; accepted 9 August 2000
Abstract Somatostatin (SST) is in primary afferent neurons and reduces vascular and nociceptive components of in¯ammation. SST receptor (SSTR) agonists provide analgesia following intrathecal or epidural administration in humans, but neurotoxicity in the central nervous system (CNS) has been reported in experimental animals. With the rationale that targeting peripheral SSTRs would provide effective analgesia while avoiding CNS side effects, the goals of the present study are to investigate the presence of SSTRs on peripheral primary afferent ®bers and determine the behavioral and physiological effects of the SST agonist octreotide (OCT) on formalin-induced nociception and bradykinin-induced primary afferent excitation and sensitization in the rat. The results demonstrate that: (1) SSTR2as are present on 11% of peripheral primary afferent sensory ®bers in rat glabrous skin; (2) intraplantar injection of OCT reduces formalin-induced nociceptive behaviors; (3) OCT reduces, in a dose-dependent fashion, responses to thermal stimulation in C-mechanoheat sensitive ®bers; and (4) OCT reduces the responses of C-mechanoheat ®bers to bradykinin-induced excitation and sensitization to heat. Each of these actions can be reversed following co-injection of OCT with the SSTR antagonist cyclo-somatostatin (c-SOM). Thus, activation of peripheral SSTRs reduces both in¯ammatory pain and the activity of sensitized nociceptors, avoids deleterious CNS side effects and may be clinically useful in the treatment of pain of peripheral origin. q 2001 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Primary afferent; Thermal hyperalgesia; Nociception; Peptide
1. Introduction Somatostatin (SST) is a widespread, biologically active polypeptide (Reichlin, 1983) that is found in a subset of primary afferent neurons (HoÈkfelt et al., 1975). SST and its agonists such as Octreotide (OCT, Sandostatin) (Penn et al., 1992) act at speci®c receptors which are at present divided into ®ve subtypes, SSTR1-5 with the SSTR2 receptor subdivided into a and b subtypes (Patel et al., 1995). Based on structural similarity and reactivity for analogues, the ®ve subtypes are grouped into two general categories, with SSTR2, SSTR3 and SSTR5 in one group and SSTR1 and SSTR4 in the other (Viollet et al., 1995). Interest in SST and its receptors has been kindled by human studies demonstrating that intrathecal, intracerebroventricular, epidural or systemic administration of SST or OCT, provide analgesia in many clinical situations, including post-operative pain (Chrubasik et al., 1984, 1985; Taura et al., 1994), bone * Corresponding author. Tel.: 11-409-772-2124; fax: 11-409-762-9382. E-mail address: [email protected] (S.M. Carlton).
pain (Burgess et al., 1996), cancer pain (Chrubasik et al., 1984; Meynadier et al., 1985; Mollenholt et al., 1994; Penn et al., 1992), pain of osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994), intractable non-malignant pain (Paice et al., 1996), visceral pain (Plourde et al., 1993), and migraine headache (Kapicioglu et al., 1997). This analgesia is often obtained when opioids are no longer effective (Chrubasik et al., 1984; Meynadier et al., 1985). In experimental paradigms, the predominant effect of SSTR activation in the spinal cord is inhibition (Chapman and Dickenson, 1992; Helmchen et al., 1995; Miletic and Randic, 1982; Murase et al., 1982; Randic and Miletic, 1978; Sandkuhler et al., 1990), however, excitation has also been reported (Randic and Miletic, 1978; Wiesenfeld-Hallin, 1985). In some studies, central administration of SST is associated with spinal cord damage including nucleolysis of dorsal and ventral horn cells, pyknotic neurons, focal demyelination and in¯ammation (Gaumann et al., 1989; Gaumann and Yaksh, 1988), and the amount of damage is dose-dependent (Mollenholt et al., 1988). Thus, the usefulness of central routes of admin-
0304-3959/01/$20.00 q 2001 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(00)00407-3
234
S.M. Carlton et al. / Pain 90 (2001) 233±244
istration of SST in treating deleterious pain states is problematic. Another approach to the development of SST as a therapeutic agent is to administer it or its agonists peripherally since this avoids the side effects that follow epidural or intrathecal administration. That this approach might be fruitful is shown in recent studies demonstrating that SST injected locally in the knee joint provides useful clinical analgesia in osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994). Furthermore, it has also been shown that close arterial injection of an SST agonist in rat will inhibit mechanoreceptor activity in response to noxious joint rotation in the normal and in¯amed knee joint (Heppelmann and Pawlak, 1997). To investigate whether peripheral SSTR activation is critical in controlling nociceptor excitability, we determine (1) the association of SSTR2a with peripheral terminals of primary afferent ®bers, (2) the effects of a locally administered SSTR agonist and antagonist on the spontaneous behavioral responses that arise in the formalin test, and (3) the effects of an SSTR agonist and antagonist on responses to thermal stimulation of normal and sensitized nociceptors in rat glabrous skin. The ®ndings support the hypothesis that activation of peripheral SSTRs on cutaneous primary afferents reduces in¯ammatory nociceptive behaviors, and nociceptor excitation and sensitization. 2. Materials and methods All experiments were approved by the University Animal Care and Use Committee and followed the guidelines for the ethical care and use of laboratory animals (Zimmermann, 1983). 2.1. Anatomical studies Electron microscopic analyses of SSTR2a immunostaining: Four rats (male Sprague±Dawley, 250±300 g) were deeply anesthetized with pentobarbital (70 mg/kg, i.p.) and perfused through the left ventricle with heparinized saline followed by a mixture of 2.5% glutaraldehyde, 1% paraformaldehyde and 0.1% picric acid in 0.1 M phosphate buffer (PB), pH 7.4 at 48C. Glabrous skin from the plantar surface of the hind toes was removed from each animal. The skin was cut into blocks approximately 1mm thick with a razor blade. Prior to immunostaining, the skin was placed in 1% sodium borohydride in PB for 1 h to preserve immunoreactivity and then rinsed in graded alcohols to increase antibody penetration. All tissues were immunostained with an ABC kit (Vector Laboratories, Inc., Burlingame, CA) using a previously described protocol (Coggeshall and Carlton, 1998). The antibody was raised in guinea pig, directed against the 2a subunit of the SSTR (Gramsch Labs, Dachau-Indersdorf, Germany) and used at a dilution of 1:40 000. To control for non-speci®c immunostaining, tissue was treated the same way as described above except
that the primary antibody was omitted. Controls for antibody speci®city included preabsorbing the antibody with an excess of antigen (100 mg/ml, obtained from Gramsch Labs) and immunostaining with this solution. Extensive characterization of the SSTR2a antibody, including dot-blot and Western blot analyses, has been previously reported (Schulz et al., 1998). After immunostaining, the tissues were placed in 1% osmium tetroxide in PB, dehydrated, and embedded in a mixture of Epon and Araldite. Semithin serial sections (0.5 mm) were cut at right angles to the dermal-epidermal junction until tissue appeared. Ultrathin sections were then collected and mounted on slot grids and examined with a JEOL CX100 electron microscope. Labeled and unlabeled axon pro®les were counted in one ultrathin section from each block of skin (n 3 blocks per animal) to estimate the percentages of immunostained axon pro®les. 2.2. Behavioral studies 2.2.1. Habituation A total of 62 male Sprague±Dawley rats (150±200 g) were used in this study. Animals were housed in groups of three in plastic cages with soft bedding under a reversed light/dark cycle of 12:12 h. Following arrival at the animal care facility, the animals were acclimated for at least 3 days before behavioral testing was initiated. Rats were habituated to the behavioral testing procedures by placing them on a wire screen platform in Plexiglas cages (8 £ 8 £ 18 cm) for 1 h. Each rat was habituated two times before being placed in an experimental group. 2.2.2. Drug injections Octreotide acetate (OCT, Sandostatin) was diluted in phosphate-buffered saline (PBS, Gibco, pH 7.4) to three different concentrations (0.2, 2.0, 20 mM). Thirty microliters of 2.5% formalin (FM) followed by 50 ml of one of the above OCT concentrations was drawn into a 100-ml Hamilton syringe. For injection, a 28-gauge needle was attached with PE20 tubing to the Hamilton syringe and 80ml of this solution was injected subcutaneously in the hind paw (n 10 per group). The needle punctured the plantar skin and was guided forward in the subcutaneous space to a site just proximal to the pads. Each animal was used only once, and the investigator was unaware of the drugs injected with the FM. To determine if OCT had a systemic effect, 50 ml of 20 mM OCT (the standard dose we used in this study) was injected subcutaneously into one hindpaw (n 9), followed immediately by an injection of 30 ml of 2.5% FM into the contralateral hindpaw, and spontaneous pain behaviors for the FMinjected paw were recorded. To determine drug speci®city, cyclo-somatostatin (c-SOM, Bachem, Bioscience, Inc.), an SSTR antagonist, was dissolved in PBS and 1.3 mM c-SOM (30 ml) was injected into the hindpaw 20 min prior to injection of 50 ml of 20 mM OCT 1 30 ml of 2.5% FM (n 9).
S.M. Carlton et al. / Pain 90 (2001) 233±244
2.2.3. Behavioral testing Spontaneous behaviors resulting from the FM injections were assessed by counting the number of ¯inches and seconds an animal spent lifting and/or licking (L/L) the injected paw in 5-min intervals. A ¯inch was de®ned as a spontaneous, rapid jerk of the foot whether the foot was on the screen or held in the air. The time course of the nociceptive behaviors was plotted as the mean number of ¯inches and amount of time spent L/L in 5-min intervals for 50 min. There were two phases in FM-induced spontaneous pain behaviors. The ®rst phase occurred from 0 to 10 min after injection and the second from 10 to 30 min after injection. The following formulae were used for averaging the data in each phase: (1) ®rst phase: [total number of ¯inches or L/L (s)]/2 (two 5-min intervals in 10 min); (2) second phase: [total number of ¯inches or L/L (s)]/4. 2.3. Electrophysiological studies 2.3.1. In vitro skin-nerve recordings Single unit recordings from C-mechanoheat (CMH) sensitive ®bers in glabrous skin were obtained using a modi®ed in vitro skin-nerve preparation (Fig. 1) (Du et al., 2000; Kress et al., 1992; Reeh, 1986). Twenty-six male Sprague± Dawley rats (200±300 g) were killed with an overdose of CO2 and the glabrous skin of the hindpaw was dissected from each animal with the attached medial and lateral plantar nerves. The preparation was then placed corium side up in an organ bath and superfused (15 ml/min, 348C) with an oxygen-saturated, modi®ed synthetic interstitial ¯uid solution (SIF, in mM: NaCl, 123; KCl, 3.5; MgSO4, 0.7; CaCl2, 2.0; Na gluconate, 9.5; NaH2PO4, 1.7; glucose, 5.5; sucrose, 7.5; HEPES, 10; pH 7.40 ^ 0.05). The plantar nerves were placed in a separate chamber containing a top layer of mineral oil and a bottom layer of SIF. The nerves were desheathed and teased apart on a mirror stage under a dissecting microscope. Small nerve bundles were repeatedly split with sharpened forceps until single unit activity was obtained.
235
2.3.2. Thermal stimulation A feedback-controlled lamp placed beneath the organ bath supplied radiant heat to each receptive ®eld. The beam was focused through the translucent bottom of the bath onto the epidermal side of the skin. A thermocouple was placed in the corium above the light beam to measure intracutaneous temperature. A standard heat ramp was applied to each unit, starting from an adapting temperature of 348C and rising to 478C in 10 s (478C on the corium side was equivalent to 518C on the epidermal side). The temperature at which the second spike was elicited by the heat stimulus was de®ned as heat threshold. 2.3.3. Mechanical stimulation The mechanical threshold for each unit was determined using calibrated von Frey ®laments (Stoelting, Inc.). The ®laments were applied to the receptive ®eld on the corium side of the skin in a uniform fashion, starting with ®laments delivering relatively small amounts of force and ascending (0.1±166.7 mN) until an action potential could be consistently evoked. 2.3.4. Chemical stimulation To investigate responses of units to various drugs, a small plastic ring (5 mm diameter) was placed over the receptive ®eld of each unit and the SIF in the ring was replaced with SIF containing the SSTR agonist OCT, or bradykinin (BK), or BK 1 OCT, or BK 1 OCT 1 the antagonist c-SOM, or OCT 1 naloxone. All drugs dissolved in SIF were buffered to pH 7.40 ^ 0.05. 2.3.5. Effects of an SSTR agonist on thermal responses To investigate whether activation of SSTRs could inhibit responses of CMH ®bers to heat, SIF in the ring was replaced with an ascending series of OCT concentrations ranging from 0.002 to 200 mM made in SIF (pH 7.4). Each concentration was applied for 1 min and the intertrial interval was 5±5.5 min. Threshold temperatures at which nociceptor activity was elicited and discharge rates were measured before and after application of OCT. 2.3.6. Effects of an SSTR agonist and antagonist on BKinduced excitation and sensitization To determine if OCT effected peripheral sensitization, three different paradigms were used with BK as the sensitizing agent. Following characterization of each unit, which included testing for a heat response (Heat 1), all CMH units were exposed to 10 mM BK (BK1) for 1 min, then following a 5-min washout period, the unit was exposed a second time to 10 mM BK alone (BK2) or BK2 1 20 mM OCT, or BK2 1 20 mM OCT 1 1.3mM c-SOM for 1 min, followed within 30 s by retesting of the heat response (Heat 2).
Fig. 1. Schematic drawing of our in vitro skin-nerve preparation in which the glabrous skin of the hindpaw is dissected with the medial and lateral plantar nerves intact. The various forms of stimulation that are applied to the receptive ®eld of each unit are indicated.
2.3.7. Effects of an SSTR agonist and an opiate antagonist on thermal responses To determine if OCT effects were mediated all or in part
236
S.M. Carlton et al. / Pain 90 (2001) 233±244
by opiate receptors, 3 CMH units were tested for a heat response, then exposed to 20 mM OCT 1 2 mM naloxone for 1 min, followed within 30 s by retesting of the heat response. 2.3.8. Neurophysiological recordings Neural activity was recorded using a DAM80 Differential Ampli®er (World Precision Instruments) attached to a monopolar gold wire electrode, with the reference electrode positioned nearby. Units were ®rst identi®ed by manual probing of the glabrous skin with a blunt glass rod. Only units responding to mechanical probing with a clearly de®ned receptive ®eld were studied in detail. Action potentials were analyzed on a Dell computer with a custom-made template-matching program that allowed discrimination of single-unit activity based on the amplitude and wave form of each action potential (Forster and Handwerker, 1990). The conduction velocity of each unit was determined by monopolar electrical stimulation (0.1 ms duration, train frequency 1 Hz) at the most mechanosensitive site in the receptive ®eld of each unit using a Te¯on-coated steel electrode (5 MV impedance, 250 mm shaft diameter). Conduction velocity was determined from the latency of the action potential and the distance from the stimulation electrode to the recording site (measured in millimeters). Units with a conduction velocity of less than 1.6 m/s were classi®ed as C ®bers. 2.4. Statistical analysis Behavioral and electrophysiological data were expressed as mean ^ standard error of the mean (SEM) and evaluated using the Sigmastat program (Jandel Corporation). In the behavioral studies, differences between groups were evalu-
Table 1 Percentages of SSTR2a-labeled axons in the dermal-epidermal junction Animal no.
Labeled axons
Total no. of axons
Percent labeled
Rat 1 Rat 2 Rat 3 Mean ^ SD
21 19 23
198 167 203
10.6 11.4 11.3 11.1 ^ 0.4
ated using a t-test if a normality test was passed or the Kruskal±Wallis test if not. In the electrophysiological studies, differences in discharge rates or threshold temperatures were evaluated with either a Friedman's ANOVA, Kruskal±Wallis test followed by Dunn's post hoc analysis, Wilcoxon signed ranks test or the Mann±Whitney U-test where appropriate. P , 0:05 was considered signi®cant. 3. Results 3.1. Anatomical studies Electron microscopic analyses of SSTR2a immunostaining in peripheral primary afferent terminals demonstrated immunostaining of unmyelinated sensory axons at the dermal-epidermal junction. Reaction product for the SSTR2a was localized on discrete patches on the axonal membrane (Fig. 2A) or diffusely distributed within the axoplasm. Unlabeled axons were also observed (Fig. 2A,B). Counts of labeled and unlabeled axons demonstrated that 11.1 ^ 0.4% of the unmyelinated axons were positively stained for the SSTR2a (Table 1). On occasion, reaction product was observed in some Schwann cell processes (Fig. 2B). Incubating tissue in solutions in which the
Fig. 2. Electron micrographs of Remak bundles at the dermal-epidermal junction which contains (A) one axon immunostained for the SSTR2a (arrow) and several unlabeled axons (arrowheads), and (B) a labeled Schwann cell process (arrow) and unlabeled axons (arrowheads). Bar: 0.25 mm.
S.M. Carlton et al. / Pain 90 (2001) 233±244
237
Fig. 3. Dose±response relationship. Note that animals receiving an intraplantar injection of 20 mM OCT 1 FM show a signi®cant decrease in phase 2 but not phase 1 lifting/licking behavior compared with animals receiving PBS 1 FM (A). Flinching behavior (B) is not affected. (*P , 0:05, significant difference from PBS 1 FM group, Kruskal±Wallis test).
primary antibody had been omitted or in which the antibody had been pre-absorbed with the appropriate peptide resulted in no immunoreaction product in axons or Schwann cells. 3.2. Behavioral studies Intraplantar injection of 2.5% FM resulted in animals attending to the injected paw with L/L behavior and ¯inching (Fig. 3A,B). A biphasic response was observed in the present study with the animals demonstrating these nociceptive behaviors in both phase 1 and phase 2. Compared with FM alone, animals injected with 20 mM OCT 1 2.5% FM had signi®cantly reduced L/L behavior during phase 2, but phase 1 L/L behavior was unaffected (Fig. 3A). There was no change in ¯inching behavior at any dose of OCT (Fig. 3B). The time course study performed with 20 mM OCT 1 2.5% FM demonstrated that L/L behaviors were signi®cantly reduced at 15, 20 and 25 min during phase 2 (Fig. 4). Intraplantar injection of the SSTR antagonist cyclo-somatostatin (1.3 mM, c-SOM), followed by 20 mM OCT 1 2.5%FM, reversed the OCT effects (Fig. 5). That OCT acted locally rather than systemically was demonstrated in animals injected with 20 mM OCT in one hindpaw and 2.5% FM into the contralateral hindpaw (Fig. 5). These animals had nociceptive behaviors similar to animals injected with PBS 1 2.5% FM.
Fig. 4. (A) A time course study demonstrating that co-injection of 20 mM OCT with 2.5% FM signi®cantly attenuates lifting/licking behaviors at 15, 20 and 25 min post-injection during phase 2 (*P , 0:05, signi®cant difference from PBS 1 FM, Kruskal±Wallis). (B) Flinching behavior did not change following intraplantar injection of 20 mM OCT 1 FM.
Fig. 5. Intraplantar injection of 1.3 mM c-SOM blocked the OCT-induced attenuation of FM behaviors such that these animals displayed behaviors that were similar in magnitude to animals receiving PBS 1 FM (PBS 1 FM, and 20 mM OCT 1 FM groups are reprinted from Fig. 3 for comparison). The anti-nociceptive effect of OCT was not due to a systemic effect since animals injected with 20 mM OCT in one hindpaw and 2.5% FM into the contralateral hindpaw (`Contral') had nociceptive behaviors no different from those seen in animals injected with formalin alone. (*P , 0:05, signi®cant difference from the PBS 1 FM group, Kruskal± Wallis).
3.3. Electrophysiological studies Units responding to both mechanical and heat stimuli with conduction velocities below 1.6 m/s were studied (C mechanoheat-sensitive or CMH ®bers). The mean conduction velocity of these ®bers (n 57) was 0.63 ^ 0.03 m/s and ranged from 0.26 to 1.5 m/s. The median force for activation with von Frey ®laments was 27.9 mN, ranging from 1.6 to 166 mN. Some of the drug application paradigms required repeated heating of these CMH ®bers, and it was determined that there was no change in the mean heatevoked discharge rate if intertrial intervals were 5 min or longer (Friedman's ANOVA, Fig. 6). Prior to each drug application, background activity was measured for 1 min for each unit and this activity was subtracted from the activ-
Fig. 6. Exposing units to heat with 5-min intertrial intervals results in responses that are not signi®cantly different from each other indicating that changes in heat responses following drug application are not due to repeated heating of the units (Friedman's ANOVA, P 0:6).
238
S.M. Carlton et al. / Pain 90 (2001) 233±244
ity obtained during the 1 min drug exposure. Prior to heat stimulation, background activity was measured for 10 s for each unit and this activity was subtracted from the activity obtained during the 10-s heat stimulus. 3.3.1. Effects of OCT on thermal responses One-minute applications of ascending concentrations of OCT (0.002±200 mM) to the receptive ®elds of CMH units demonstrated a dose-dependent inhibition of responses to thermal stimuli (Figs. 7 and 8). In particular, 2.0, 20, 60 and 200 mM OCT resulted in signi®cant decreases of 62, 71, 95 and 96%, respectively, in responses to thermal stimuli when compared with control responses (Kruskal± Wallis test followed by Dunn's post hoc analysis, P , 0:05, Fig. 7). Based on these ®ndings, 20.0 mM OCT was chosen as the concentration to test the effect of OCT on BK-induced excitation and sensitization of CMH units. The thermal responses of three additional CMH ®bers were analyzed before and after the exposure of their receptive ®elds to 2 mM naloxone 1 20.0 mM OCT. The OCTinduced inhibition of the heat response in these units was not reversed by naloxone (data not shown). 3.3.2. Effect of BK Application of 10 mM BK to the receptive ®eld of CMH units (n 37) for 1 min resulted in an increase in the mean discharge rate to 0.24 ^ 0.06 impulses/s above background. Of the 37 ®bers tested, 15 (41%) were categorized as `BK responders', meeting the criteria of having an increase in discharge rate that was equal to or greater than 2 standard deviations above mean background for the total population.
Fig. 7. Application of OCT to the receptive ®eld of nociceptors produces a dose-dependent reduction in their responses to thermal stimuli. An ascending series of OCT concentrations, ranging from 0.002 to 200 mM, was applied for 1 min to the receptive ®elds of CMH units and then heat was applied to the receptive ®eld. Note that in a dose-dependent fashion, OCT decreased the mean discharge rates in response to heat. Compared with control values, 2, 20, 60 and 200 mM OCT resulted in signi®cant reductions in the heat response (Kruskal±Wallis test followed by a Dunn's post hoc analysis. N, numbers of individually recorded ®bers at each dose).
Fig. 8. The top trace shows the response of a C nociceptor to a 10-s heat pulse following exposure of the receptive ®eld to synthetic interstitial ¯uid (SIF). The middle trace shows the decrease in response to heat of the same unit following exposure of its receptive ®eld to SIF 1 20 mM OCT. The bottom trace shows our standard 10-s heat pulse.
3.3.3. Effects of OCT and c-SOM on BK-induced excitation Each of the 37 units tested with BK (BK1) was randomly placed into one of three groups and then exposed to either (1) 10 mM BK2, (2) 10 mM BK2 1 20 mM OCT, or (3) 10 mM BK2 1 20 mM OCT 1 1.3 mM c-SOM. Statistical analysis demonstrated that there was no difference in mean responses to BK1 among these three groups (Kruskal±Wallis test, P 0:96). For group 1 (n 15), the second application of BK (BK2) was applied 5 min after the ®rst and the mean discharge rate generated by the second application was signi®cantly reduced to 0.07 ^ 0.03 impulses/s (Wilcoxon signed rank test, P , 0:05), representing a 70% decrease from the BK1 response (Figs. 9 and 10B,C) which demonstrates the well-known phenomenon of tachyphylaxis (Mizumura and Kumazawa, 1996). If BK2 was applied in
Fig. 9. Application of 10 mM BK (BK1) to the receptive ®eld of CMH units results in excitation. A second application of BK (BK2) produces less excitation of the units than BK1 due to tachyphylaxis or BK-induced desensitization. (*P , 0:05, signi®cant difference from BK1, Wilcoxon signed rank test).
S.M. Carlton et al. / Pain 90 (2001) 233±244
239
Fig. 10. Changes in group 1 (see text) BK-induced excitation. (A) The paradigm used to demonstrate BK-induced tachyphylaxis and sensitization to heat for group 1 is shown. (B) The time course of CMH responses (n 15) to BK1 is shown. (C) The time course of CMH responses to BK2 is shown. Note that the background prior to BK2 is higher than that prior to BK1 due to exposure to BK1 and furthermore, that the response to BK2 is much less than that to BK1, demonstrating BK-induced tachyphylaxis.
the presence of OCT (group 2, n 12), the BK-induced excitation was reduced to essentially zero (Fig. 11). If BK2 was applied in the presence of both OCT and c-SOM (group 3, n 10), the inhibitory effect of OCT was reversed in that the units responded with a discharge rate of 0.05 ^ 0.06 impulses/s, which was not signi®cantly different from BK2 (Fig. 11). These results demonstrate that local OCT inhibits the BK-induced excitation of CMHs, and this inhibition can be reversed by the SST antagonist c-SOM. 3.3.4. Effects of OCT and c-SOM on units sensitized to heat by BK There was great variability in the magnitude of individual heat responses. Therefore, in order to determine if a heat response was altered following exposure to various drugs, a power analysis using a power of 0.80, an alpha of 0.05, and a standard deviation of 0.64 (obtained from the 37 units used in this part of the study) was performed. Based on this criterion, if a unit displayed an increase in discharge rate that was .30% of its original heat response, the unit was considered a `responder'. Following application of 10 mM BK2 (group 1), 8/15 units (53%) had an increase in Heat 2 responses that were equal to or greater than 30% of Heat 1. The mean discharge
rate increased 46%, from 0.98 ^ 0.01 to 1.43 ^ 0.28 impulses/s (Wilcoxon signed rank test, P , 0:01), and the
Fig. 11. Changes in BK-induced excitation: units were divided into three groups and discharge rates measured following exposure to either BK (BK2, group 1), BK2 1 OCT (group 2) or BK2 1 OCT 1 c-SOM (group 3). In the presence of OCT, BK-induced excitation was essentially eliminated. In contrast, when the BK antagonist c-SOM was added to the BK2 1 OCT, it reversed the effect of OCT such that the mean discharge rate of this group was not signi®cantly different (NS) from that of BK2. (*P , 0:05, signi®cant difference from BK2, Mann±Whitney U-test).
240
S.M. Carlton et al. / Pain 90 (2001) 233±244
mean threshold decreased from 41.7 ^ 0.4 to 40.2 ^ 0.68C comparing Heat 1 with Heat 2, and these changes were signi®cant (Fig. 12, Wilcoxon signed rank test, P , 0:05). Thus as expected, BK sensitized CMH units in this paradigm. When BK2 was applied in the presence of OCT in group 2 (n 12), 11/12 units (92%) demonstrated decreased responses to heat, with the mean discharge rate decreasing from 1.16 ^ 0.25 (Heat 1) to 0.83 ^ 0.17 impulses/s (Heat 2) and the mean temperature threshold increasing from 41.2 ^ 0.7 (Heat 1) to 42.5 ^ 0.78C (Heat 2). Comparing group 1 with group 2 demonstrated that addition of OCT to BK signi®cantly reduced the Heat 2 response (Fig. 13A, Mann±Whitney U-test, P , 0:05) and signi®cantly increased heat threshold (Fig. 13B, Mann±Whitney U-test, P , 0:05). When BK2 was applied in the presence of both OCT and c-SOM in group 3 (n 10), the inhibitory effect of OCT was reversed in 9/10 units (90%) with the mean discharge rate increasing from 0.87 ^ 0.15 (Heat 1) to 1.9 ^ 0.34 impulses/s (Heat 2) and the mean threshold temperature decreasing from 41.7 ^ 0.7 (Heat 1) to
40 ^ 0.48C (Heat 2). Comparing group 1 with group 3 demonstrated that addition of OCT 1 c-SOM to BK resulted in no signi®cant difference a mean discharge rate (Fig. 13A) or mean threshold temperature (Fig. 13B). These results demonstrate that OCT inhibits the thermal responses of nociceptors sensitized by BK, and this inhibition can be reversed by the SSTR antagonist c-SOM. 4. Discussion The present study demonstrates that approximately 11% of peripheral unmyelinated sensory axons at the dermalepidermal junction of the glabrous skin of the rat hindpaw express the SSTR2a. Due to the fact that there are no sympathetic efferents at the dermal-epidermal junction in the rat glabrous skin (Coggeshall et al., 1997), the labeled unmyelinated axons must be sensory axons. This provides a reasonable anatomical basis for the behavioral and physiological changes that follow intraplantar application of the
Fig. 12. Changes in sensitized heat responses induced by BK (note that the paradigm used is illustrated in Fig. 10A) Exposure to BK resulted in a signi®cant increase in the mean discharge rate (A) and a signi®cant decrease in the mean threshold temperature for heat activation (B) for group 1 units (Wilcoxon signed rank test, *P , 0:05, **P , 0:01). The time course of unit activation for Heat 1 (C) and Heat 2 (D) are shown. Background discharge rates prior to Heat 2 are higher than that prior to Heat 1 because of exposure to BK.
S.M. Carlton et al. / Pain 90 (2001) 233±244
Fig. 13. Changes in sensitized heat responses induced by BK: Application of BK2 1 OCT to receptive ®elds of nociceptors resulted in a 42% decrease in discharge rate in response to heat (A) and a signi®cant increase in threshold to activation (B) when compared with BK2. If c-SOM was applied with BK2 1 OCT, the OCT effect on the heat response was reversed and discharge rates (A) and threshold temperatures (B) were not signi®cantly different (NS) from BK2. (*P , 0:05, signi®cant difference from BK2, Mann±Whitney U).
SSTR agonist OCT. It is, of course, not certain that the OCT is working exclusively through SSTR2a, since OCT will activate SSTR2, SSTR3 and SSTR5 receptor subtypes (Patel et al., 1995; Viollet et al., 1995), and mRNA for SSTR3 has been localized in DRG cells (Senaris et al., 1995). The fact that there are several SSTR subtypes that may be activated by OCT may explain the discrepancy between the relatively low percentage of SSTR2a-positive ®bers observed in the anatomical study and the relatively high percentage of ®bers that respond to OCT in the skinnerve preparation. More precise delineations of the distribution and function of the different SSTR subtypes will await development of more antibodies and speci®c agonists and antagonists. It is of interest, however, that the DRG neurons that express the SSTR2a are restricted to a subpopulation of cells with a medium diameter (Schulz et al., 1998), that also label positively for the Griffonia simplicifolia isolectin B4 (IB4) and have very little overlap with DRG cells labeled for CGRP, TrkA, or SST itself (Michael et al., 1999). Thus, it appears that SSTR2a -containing sensory neurons fall into the non-peptidergic class. Whether other SSTRs subtypes are located on different functional groups of DRG cells remains to be determined. The behavioral ®ndings in the present study are that intraplantar injection of the SSTR agonist OCT reduces FMinduced phase 2 L/L behaviors and that c-SOM, an SSTR
241
antagonist, reverses the OCT effect. The lack of OCT effect on phase 1 behaviors suggests that SSTR activation may primarily reduce in¯ammation which is believed to be one of the major factors underlying phase 2 formalin pain behaviors (Dubuisson and Dennis, 1977). It is not clear why ¯inching and L/L behavior were differentially affected. However, the lack of effect of OCT on ¯inching behavior suggests that the neurocircuitry underlying this presumed segmentally organized re¯exive behavior may be distinct from that underlying L/L behavior, which is more likely to involve supraspinal mechanisms. This hypothesis is supported by a previous study demonstrating preservation of ¯inching behavior in the ®rst phase of the formalin test in spinally transected animals (Wheeler-Aceto and Cowan, 1991). Control studies demonstrate that OCT produces this effect through local and not systemic actions since injection of OCT into one hindpaw and FM into the contralateral hindpaw results in nociceptive behaviors similar to those seen in animals injected with FM alone. Demonstration of an analgesic effect following SSTR activation is consistent with earlier ®ndings demonstrating that intraplantar injection of SST reduces mechanical hyperalgesia in carrageenan-induced in¯ammation (Corsi et al., 1997) and systemic injection of an SST analogue produces antinociceptive effects in the tail¯ick and hot plate tests (Eschalier et al., 1991). Interestingly, these changes are attributed to opioid actions (Corsi et al., 1997) or to central rather than peripheral actions (Eschalier et al., 1991), but our preliminary naloxone studies and the localization of SSTRs on sensory axons suggests that this is a direct action on peripheral SSTRs. More precise evaluations of changes in nociceptor activity following SSTR activation can be obtained from single afferent ®ber recordings. Using an in vitro skin-nerve preparation, we determined the effects of OCT on normal and sensitized nociceptors. Exposure of CMH units to the SSTR agonist OCT reduces, in a dose-dependent manner, the mean discharge rate in response to noxious heat. In a separate experiment, units are sensitized by exposure to BK, a known in¯ammatory mediator and sensitizing agent (Beck et al., 1974; Chahl and Iggo, 1977; Mizumura and Kumazawa, 1996). BK application results in excitation of 41% of the ®bers tested, con®rming previous studies (Koltzenburg et al., 1992; Lang et al., 1990), OCT signi®cantly attenuates the thermal responses of BK-sensitized nociceptors, and cSOM reverses the OCT effects. It is known that BK stimulation of DRG cells is associated with membrane depolarization and Ca 21 in¯ux (Burgess et al., 1989). In contrast, SST stimulation of DRG cells is associated with a decrease in Ca 21 channel conductance (Taddese et al., 1995). Furthermore, pre-treatment of cultured DRG cells with SST will inhibit BK-stimulated arachidonate release and these ®ndings support a coupling of SSTRs to Ca 21 channel inactivation through pertussis toxin sensitive G-proteins (Gammon et al., 1990). Calcium channel inactivation is known to interfere with release of endogenous substances such as
242
S.M. Carlton et al. / Pain 90 (2001) 233±244
excitatory peptides. Results from Heppelmann and Pawlak (1997) and this study suggest that SST-induced Ca 21 channel inactivation may also interfere with sensory neuron responses to mechanical and thermal stimulation, respectively. Our hypothesis is that OCT is directly activating SSTRs on primary afferents. Another possibility, however, is that SST is activating opioid receptors. Such an indirect action is suggested by observations that SST or its analogs bind to opioid receptors (Gulya et al., 1986; Maurer et al., 1982; Pelton et al., 1986; Pugsley and Lippmann, 1978; Terenius, 1976), and naloxone, a widely used opioid antagonist, prevents the analgesia that follows intraventricular injection of SST (Rezek et al., 1978). On the other hand, naloxone has no effect on responses of dorsal horn cells to SST application (Randic and Miletic, 1978; Sandkuhler et al., 1990), and clinical studies show that naloxone has no effect on the analgesia provided by epidural application of SST (Chrubasik et al., 1984, 1985). Furthermore, SST ligands are often effective in patients who are no longer responding to opioids) (Mollenholt et al., 1994), (Meynadier et al., 1985; Paice et al., 1996) and the receptors for SST and opioids seem to be on different populations of DRG cells (Taddese et al., 1995). In the present study, 2 mM naloxone, a dose that has been previously shown to reverse the inhibitory effects of morphine and selective m and k agonists on the spontaneous ®ring of peripheral ®bers in UV-irradiated skin (Andreev et al., 1994), did not reverse the inhibitory effects of OCT. Accordingly, most of the evidence indicates that SST and opioids are acting through different systems. A ®nal point is that SST is an endogenous anti-in¯ammatory peptide (SzolcsaÂnyi et al., 1998a). Administration of systemic SST or its ligands, aside from the profound effects on gastrointestinal and endocrine systems (Lamberts et al., 1996; Reichlin, 1983), depresses in¯ammatory responses to various stimuli (Karalis et al., 1994; Matucci-Cerinic et al., 1995; SzolcsaÂnyi et al., 1998a,b). Of particular interest is the ®nding that in¯ammation in one part of the body releases enough SST systemically to depress in¯ammatory responses in another body region (SzolcsaÂnyi et al., 1998a,b). Thus, it is likely that during neurogenic in¯ammation, the release of SST, an inhibitory anti-in¯ammatory peptide, occurs concomitantly with the release of CGRP and SP, which are excitatory, pro-in¯ammatory peptides (SzolcsaÂnyi et al., 1998a,b), and that the balance of these peptides determines the degree of both the pain and edema of in¯ammation (Heppelmann and Pawlak, 1997). This offers the hope that it may be possible to promote enhanced release of endogenous inhibitory compounds to aid in reducing the pain, swelling and other long-term deleterious changes that accompany in¯ammation. 5. Conclusion In summary, we demonstrate the presence of SSTRs on
peripheral primary afferents. Activation of these receptors with OCT diminishes the spontaneous nociceptive behaviors that follows intraplantar FM injection, and reduces the responses of nociceptors excited and sensitized by BK. These ®ndings are consistent with earlier observations that local administration of SST or SSTR agonists in the knee joint signi®cantly improves the pain of osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994) and diminishes the single unit ®ring that results from noxious movements of the normal and in¯amed joint (Heppelmann and Pawlak, 1997). Peripheral administration would largely avoid the reported CNS neurotoxicity of SST without losing the analgesic properties associated with SSTR activation. Thus, SST or more long-lasting SSTR analogues may be of considerable use in controlling nociceptive input from the periphery, both in the normal state and after primary afferent sensitization. Acknowledgements The authors thank Kristin Sawyer and Vicki Wilson for their excellent secretarial assistance, and Zhixia Ding and Greg Hargett for excellent technical assistance. This work was supported by NIH NS11255 to S.M.C. and R.E.C., NS27910 and NS40700 to S.M.C and NS10161 to R.E.C. References Andreev NY, Urban L, Dray A. Opioids suppress spontaneous activity of polymodal nociceptors in rat paw skin induced by ultraviolet irradiation. Neuroscience 1994;58:793±798. Beck PW, Handwerker HO, Hermann O. Bradykinin and serotonin effects on various types of cutaneous nerve ®bres. P¯uÈgers Arch 1974;347:209±222. Burgess GM, Mullaney J, McNeil M, Dunn P, Rang HP. Second messengers involved in the action of bradykinin on cultured sensory neurons. J Neurosci 1989;9:3314±3325. Burgess JR, Shepherd JJ, Murton FJ, Parameswaran V, Greenaway TM. Effective control of bone pain by octreotide in a patient with metastatic gastrinoma. Med J Aust 1996;164:725±727. Chahl LA, Iggo A. The effects of bradykinin and prostaglandin E1 on rat cutaneous nerve activity. Br J Pharmacol 1977;59:343±347. Chapman V, Dickenson AH. The effects of sandostatin and somatostatin on nociceptive transmission in the dorsal horn of the rat spinal cord. Neuropeptides 1992;23:147±152. Chrubasik J, Meynadier J, Blond S, Scherpereel P, Ackerman E, Weinstock M, Bonath K, Cramer H, Wunsch E. Somatostatin. A potent analgesic. Lancet 1984;1:1208±1209. Chrubasik J, Meynadier J, Scherpereel P, Wunsch E. The effect of epidural somatostatin on postoperative pain. Anesth Analg 1985;64:1085±1088. Coggeshall RE, Carlton SM. Ultrastructural analysis of NMDA, AMPA and kainate receptors on unmyelinated and myelinated axons in the periphery. J Comp Neurol 1998;391:78±86. Coggeshall RE, Zhou S, Carlton SM. Opioid receptors on peripheral sensory axons. Brain Res 1997;764:126±132. Corsi MM, Ticozzi C, Netti C, Fulgenzi A, Tiengo M, Gaja G, Guidobono F, Ferrero ME. The effect of somatostatin on experimental in¯ammation in rats. Anesth Analg 1997;85:1112±1115. Du J, Koltzenburg M, Carlton SM. Glutamate-induced excitation and sensitization of nociceptors in rat glabrous skin. Pain 2000;89:187±198.
S.M. Carlton et al. / Pain 90 (2001) 233±244 Dubuisson D, Dennis SG. The formalin test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats. Pain 1977;4:161±174. Eschalier A, Aumaitre O, Ardid D, Fialip J, Duchene-Marullaz P. Longlasting antinociceptive effect of RC-160, a somatostatin analog, in mice and rats. Eur J Pharm 1991;199:119±121. Forster C, Handwerker HO. Automatic classi®cation and analysis of microneurographic spike data using a PC/AT. J Neurosci Methods 1990;31:109±118. Gammon CM, Lyons SA, Morell P. Modulation by neuropeptides of bradykinin-stimulated second messenger release in dorsal root ganglion neurons. Brain Res 1990;518:159±165. Gaumann DM, Yaksh TL. Intrathecal somatostatin in rats: antinociception only in the presence of toxic effects. Anesthesiology 1988;68:733±742. Gaumann DM, Yaksh TL, Post C, Wilcox GL, Rodriguez M. Intrathecal somatostatin in cat and mouse studies on pain, motor behavior, and histopathology. Anesth Analg 1989;68:623±632. Gulya K, Pelton JT, Hruby VJ, Yamamura HI. Cyclic somatostatin octapeptide analogues with high af®nity and selectivity toward mu opioid receptors. Life Sci 1986;38:2221±2229. Helmchen C, Fu QG, Sandkuhler J. Inhibition of spinal nociceptive neurons by microinjections of somatostatin into the nucleus raphe magnus and the midbrain periaqueductal gray of the anesthetized cat. Neurosci Lett 1995;187:133±136. Heppelmann B, Pawlak M. Inhibitory effect of somatostatin on the mechanosensitivity of articular afferents in normal and in¯amed knee joints of the rat. Pain 1997;73:377±382. HoÈkfelt T, Elde R, Hohansson O, Luft R, Arimura A. Immunohistochemical evidence for the presence of somatostatin, a powerful inhibitory peptide, in some primary sensory neurons. Neurosci Lett 1975;1:231± 235. Kapicioglu S, Gokce E, Kapicioglu Z, Ovali E. Treatment of migraine attacks with a long-acting somatostatin analogue (octreotide, SMS 201-995). Cephalalgia 1997;17:27±30. Karalis K, Mastorakos G, Chrousos GP, Tolis G. Somatostatin analogues suppress the in¯ammatory reaction in vivo. J Clin Invest 1994;93:2000± 2006. Koltzenburg M, Kress M, Reeh PW. The nociceptor sensitization by bradykinin does not depend on sympathetic neurons. Neuroscience 1992;46:465±473. Kress M, Koltzenburg M, Reeh PW, Handwerker HO. Responsiveness and functional attributes of electrically localized terminals of cutaneous C®bers in vivo and in vitro. J Neurophysiol 1992;68:581±595. Lamberts SWJ, van der Lely A-J, De Herder WW, Ho¯and LJ. Octreotide. N Engl J Med 1996;334:246±254. Lang E, Novak A, Reeh PW, Handwerker HO. Chemosensitivity of ®ne afferents from rat skin in vitro. J Neurophysiol 1990;63:887±901. Matucci-Cerinic M, Marabini S. Somatostatin treatment for pain in rheumatoid arthritis: a double blind versus placebo study in knee involvement. Med Sci Res 1998;16:233±234. Matucci-Cerinic M, Borrelli F, Generini S, Cantelmo A, Marcucci I, Martelli F, Romagnoli P, Bacci S, Conz A, Marinelli P, Marabini S. Somatostatin-induced modulation of in¯ammation in experimental arthritis. Arthritis Rheum 1995;38:1687±1693. Maurer R, Gaehwiler BH, Buescher HH, Hill RC, Roemer D. Opiate antagonistic properties of an octapeptide somatostatin analog. Proc Natl Acad Sci USA 1982;79:4815±4817. Meynadier J, Chrubasik J, Dubar M, Wunsch E. Intrathecal somatostatin in terminally ill patients. A report of two cases. Pain 1985;23:9±12. Michael GJ, Tabrizi I, Saggu R. Localization and Regulation of the Somatostatin SST2a receptor in subpopulations of primary sensory neurons. Soc Neurosci Abstr 1999;25:409. Miletic V, Randic M. Neonatal rat spinal cord slice preparation: postsynaptic effects of neuropeptides on dorsal horn neurons. Dev Brain Res 1982;2:432±438. Mizumura K, Kumazawa T. Modi®cation of nociceptor responses by in¯ammatory mediators and second messengers implicated in their
243
action ± a study in canine testicular polymodal receptors. Prog Brain Res 1996;113:115±141. Mollenholt P, Post C, Rawal N, Freedman J, HoÈkfelt T, Paulsson I. Antinociceptive and `neurotoxic' actions of somatostatin in rat spinal cord after intrathecal administration. Pain 1988;32:95±105. Mollenholt P, Rawal N, Gordh T, Olsson Y. Intrathecal and epidural somatostatin for patients with cancer. Analgesic effects and postmortem neuropathologic investigations of spinal cord and nerve roots. Anesthesiology 1994;81:534±542. Murase K, Nedeljkov V, Randic M. The actions of neuropeptides on dorsal horn neurons in the rat spinal cord slice preparation: an intracellular study. Brain Res 1982;234:170±176. Paice JA, Penn RD, Kroin JS. Intrathecal octreotide for relief of intractable nonmalignant pain: 5-year experience with two cases. Neurosurgery 1996;38:203±207. Patel YC, Greenwood MT, Panetta R, Demchyshyn L, Nizik H, Srikant CB. The somatostatin receptor family. Life Sci 1995;57:1249±1265. Pelton JT, Kazmierski W, Guyla K, Yamamura HI, Hruby VJ. Design and synthesis of conformationally constrained somatostatin analogues with high potency and speci®city for m opioid receptors. J Med Chem 1986;29:2370±2375. Penn RD, Paice JA, Kroin JS. Ocetreotide: a potent new non-opiate analgesic for intrathecal infusion. Pain 1992;49:13±19. Plourde V, Lembo T, Shui Z, Parker J, Mertz H, Tache Y, Sytnik B, Mayer E. Effects of the somatostatin analogue octreotide on rectal afferent nerves in humans. Am J Physiol 1993;265:G742±G751. Pugsley TA, Lippmann W. Effect of somatostatin analogues and 17-adihydroequilin on rat brain opiate receptors. Res Commun Chem Pathol Pharm 1978;21:153±157. Randic M, Miletic V. Depressant actions of methionine-enkephalin and somatostatin in cat dorsal horn neurones activated by noxious stimuli. Brain Res 1978;152:196±202. Reeh PW. Sensory receptors in mammalian skin in an in vitro preparation. Neurosci Lett 1986;66:141±146. Reichlin S. Medical progress: somatostatin. N Engl J Med 1983;309:1495± 1501. Rezek M, Havlicek V, Leybin L, LaBella FS, Friesen H. Opiate-like naloxone-reversible action of somatostatin given intracerebrally. Can J Phys Pharm 1978;56:227±231. Sandkuhler J, Fu Q-G, Helmchen C. Spinal somatostatin superfusion in vivo affects activity of cat nociceptive dorsal horn neurons: comparison with spinal morphine. Neuroscience 1990;34:565±576. Schulz S, Schreff M, Schmidt H, Handel M, Przewlocki R, Hollt V. Immunocytochemical localization of somatostatin receptor SST2a in the rat spinal cord and dorsal root ganglia. Eur J Neurosci 1998; 10:3700±3708. Senaris RM, Schindler M, Humphrey PPA, Emson PC. Expression of somatostatin receptor 3 mRNA in the motorneurones of the rat spinal cord, and the sensory neurones of the spinal ganglia. Mol Brain Res 1995;29:185±190. Silveri F, Morosini P, Brecciaroli D, Cervini C. Intra-articular injection of somatostatin in knee osteoarthritis: clinical results and IGF-1 serum levels. Int J Clin Pharm Res 1994;14:79±85. SzolcsaÂnyi J, Helyes Z, Oroszi G, Nemeth J, PinteÂr E. Release of somatostatin and its role in the medication of the anti-in¯ammatory effect induced by antidromic stimulation of sensory ®bres of rat sciatic nerve. Br J Pharmacol 1998a;123:936±942. SzolcsaÂnyi J, PinteÂr E, Helyes Z, Oroszi G, Nemeth J. Systemic anti-in¯ammatory effect induced by counter-irritation through a local release of somatostatin from nociceptors. Br J Pharmacol 1998b;125:916±922. Taddese A, Nah S-Y, McCleskey EW. Selective opioid inhibition of small nociceptive neurons. Science 1995;270:1366±1369. Taura P, Planella V, Balust J, Beltran J, Anglada T, Carrero E, Burgues S. Epidural somatostatin as an analgesic in upper abdominal surgery: a double-blind study. Pain 1994;59:135±140. Terenius L. Somatostatin and ACTH are peptide with partial antagonistlike selectivity for opiate receptors. Eur J Pharm 1976;38:211±213.
244
S.M. Carlton et al. / Pain 90 (2001) 233±244
Viollet C, Prevost G, Maubert E, Faivre-Bauman A, Gardette R, Kordon C, Loudes C, Slama A, Epelbaum J. Molecular pharmacology of somatostatin receptors. Clin Pharmacol 1995;9:107±113. Wheeler-Aceto H, Cowan A. Standardization of the rat formalin test for the evaluation of analgesics. Psychopharmacology 1991;104:35±44.
Wiesenfeld-Hallin Z. Intrathecal somatostatin modulates spinal sensory and re¯ex mechanisms: behavioral and electrophysiological studies in the rat. Neurosci Lett 1985;62:69±74. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109±110.
www.elsevier.nl/locate/pain
Somatostatin receptors on peripheral primary afferent terminals: inhibition of sensitized nociceptors Susan M. Carlton a,*, Junhui Du a, Elyad Davidson b, Shengtai Zhou a, Richard E. Coggeshall a a
Department of Anatomy and Neurosciences, Marine Biomedical Institute, University of Texas Medical Branch, Galveston, TX 77555-1069, USA b Department of Anesthesia and Critical Care Medicine, Hadassah Hebrew University, School of Medicine, Jerusalem, Israel Received 6 March 2000; received in revised form 30 June 2000; accepted 9 August 2000
Abstract Somatostatin (SST) is in primary afferent neurons and reduces vascular and nociceptive components of in¯ammation. SST receptor (SSTR) agonists provide analgesia following intrathecal or epidural administration in humans, but neurotoxicity in the central nervous system (CNS) has been reported in experimental animals. With the rationale that targeting peripheral SSTRs would provide effective analgesia while avoiding CNS side effects, the goals of the present study are to investigate the presence of SSTRs on peripheral primary afferent ®bers and determine the behavioral and physiological effects of the SST agonist octreotide (OCT) on formalin-induced nociception and bradykinin-induced primary afferent excitation and sensitization in the rat. The results demonstrate that: (1) SSTR2as are present on 11% of peripheral primary afferent sensory ®bers in rat glabrous skin; (2) intraplantar injection of OCT reduces formalin-induced nociceptive behaviors; (3) OCT reduces, in a dose-dependent fashion, responses to thermal stimulation in C-mechanoheat sensitive ®bers; and (4) OCT reduces the responses of C-mechanoheat ®bers to bradykinin-induced excitation and sensitization to heat. Each of these actions can be reversed following co-injection of OCT with the SSTR antagonist cyclo-somatostatin (c-SOM). Thus, activation of peripheral SSTRs reduces both in¯ammatory pain and the activity of sensitized nociceptors, avoids deleterious CNS side effects and may be clinically useful in the treatment of pain of peripheral origin. q 2001 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Primary afferent; Thermal hyperalgesia; Nociception; Peptide
1. Introduction Somatostatin (SST) is a widespread, biologically active polypeptide (Reichlin, 1983) that is found in a subset of primary afferent neurons (HoÈkfelt et al., 1975). SST and its agonists such as Octreotide (OCT, Sandostatin) (Penn et al., 1992) act at speci®c receptors which are at present divided into ®ve subtypes, SSTR1-5 with the SSTR2 receptor subdivided into a and b subtypes (Patel et al., 1995). Based on structural similarity and reactivity for analogues, the ®ve subtypes are grouped into two general categories, with SSTR2, SSTR3 and SSTR5 in one group and SSTR1 and SSTR4 in the other (Viollet et al., 1995). Interest in SST and its receptors has been kindled by human studies demonstrating that intrathecal, intracerebroventricular, epidural or systemic administration of SST or OCT, provide analgesia in many clinical situations, including post-operative pain (Chrubasik et al., 1984, 1985; Taura et al., 1994), bone * Corresponding author. Tel.: 11-409-772-2124; fax: 11-409-762-9382. E-mail address: [email protected] (S.M. Carlton).
pain (Burgess et al., 1996), cancer pain (Chrubasik et al., 1984; Meynadier et al., 1985; Mollenholt et al., 1994; Penn et al., 1992), pain of osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994), intractable non-malignant pain (Paice et al., 1996), visceral pain (Plourde et al., 1993), and migraine headache (Kapicioglu et al., 1997). This analgesia is often obtained when opioids are no longer effective (Chrubasik et al., 1984; Meynadier et al., 1985). In experimental paradigms, the predominant effect of SSTR activation in the spinal cord is inhibition (Chapman and Dickenson, 1992; Helmchen et al., 1995; Miletic and Randic, 1982; Murase et al., 1982; Randic and Miletic, 1978; Sandkuhler et al., 1990), however, excitation has also been reported (Randic and Miletic, 1978; Wiesenfeld-Hallin, 1985). In some studies, central administration of SST is associated with spinal cord damage including nucleolysis of dorsal and ventral horn cells, pyknotic neurons, focal demyelination and in¯ammation (Gaumann et al., 1989; Gaumann and Yaksh, 1988), and the amount of damage is dose-dependent (Mollenholt et al., 1988). Thus, the usefulness of central routes of admin-
0304-3959/01/$20.00 q 2001 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(00)00407-3
234
S.M. Carlton et al. / Pain 90 (2001) 233±244
istration of SST in treating deleterious pain states is problematic. Another approach to the development of SST as a therapeutic agent is to administer it or its agonists peripherally since this avoids the side effects that follow epidural or intrathecal administration. That this approach might be fruitful is shown in recent studies demonstrating that SST injected locally in the knee joint provides useful clinical analgesia in osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994). Furthermore, it has also been shown that close arterial injection of an SST agonist in rat will inhibit mechanoreceptor activity in response to noxious joint rotation in the normal and in¯amed knee joint (Heppelmann and Pawlak, 1997). To investigate whether peripheral SSTR activation is critical in controlling nociceptor excitability, we determine (1) the association of SSTR2a with peripheral terminals of primary afferent ®bers, (2) the effects of a locally administered SSTR agonist and antagonist on the spontaneous behavioral responses that arise in the formalin test, and (3) the effects of an SSTR agonist and antagonist on responses to thermal stimulation of normal and sensitized nociceptors in rat glabrous skin. The ®ndings support the hypothesis that activation of peripheral SSTRs on cutaneous primary afferents reduces in¯ammatory nociceptive behaviors, and nociceptor excitation and sensitization. 2. Materials and methods All experiments were approved by the University Animal Care and Use Committee and followed the guidelines for the ethical care and use of laboratory animals (Zimmermann, 1983). 2.1. Anatomical studies Electron microscopic analyses of SSTR2a immunostaining: Four rats (male Sprague±Dawley, 250±300 g) were deeply anesthetized with pentobarbital (70 mg/kg, i.p.) and perfused through the left ventricle with heparinized saline followed by a mixture of 2.5% glutaraldehyde, 1% paraformaldehyde and 0.1% picric acid in 0.1 M phosphate buffer (PB), pH 7.4 at 48C. Glabrous skin from the plantar surface of the hind toes was removed from each animal. The skin was cut into blocks approximately 1mm thick with a razor blade. Prior to immunostaining, the skin was placed in 1% sodium borohydride in PB for 1 h to preserve immunoreactivity and then rinsed in graded alcohols to increase antibody penetration. All tissues were immunostained with an ABC kit (Vector Laboratories, Inc., Burlingame, CA) using a previously described protocol (Coggeshall and Carlton, 1998). The antibody was raised in guinea pig, directed against the 2a subunit of the SSTR (Gramsch Labs, Dachau-Indersdorf, Germany) and used at a dilution of 1:40 000. To control for non-speci®c immunostaining, tissue was treated the same way as described above except
that the primary antibody was omitted. Controls for antibody speci®city included preabsorbing the antibody with an excess of antigen (100 mg/ml, obtained from Gramsch Labs) and immunostaining with this solution. Extensive characterization of the SSTR2a antibody, including dot-blot and Western blot analyses, has been previously reported (Schulz et al., 1998). After immunostaining, the tissues were placed in 1% osmium tetroxide in PB, dehydrated, and embedded in a mixture of Epon and Araldite. Semithin serial sections (0.5 mm) were cut at right angles to the dermal-epidermal junction until tissue appeared. Ultrathin sections were then collected and mounted on slot grids and examined with a JEOL CX100 electron microscope. Labeled and unlabeled axon pro®les were counted in one ultrathin section from each block of skin (n 3 blocks per animal) to estimate the percentages of immunostained axon pro®les. 2.2. Behavioral studies 2.2.1. Habituation A total of 62 male Sprague±Dawley rats (150±200 g) were used in this study. Animals were housed in groups of three in plastic cages with soft bedding under a reversed light/dark cycle of 12:12 h. Following arrival at the animal care facility, the animals were acclimated for at least 3 days before behavioral testing was initiated. Rats were habituated to the behavioral testing procedures by placing them on a wire screen platform in Plexiglas cages (8 £ 8 £ 18 cm) for 1 h. Each rat was habituated two times before being placed in an experimental group. 2.2.2. Drug injections Octreotide acetate (OCT, Sandostatin) was diluted in phosphate-buffered saline (PBS, Gibco, pH 7.4) to three different concentrations (0.2, 2.0, 20 mM). Thirty microliters of 2.5% formalin (FM) followed by 50 ml of one of the above OCT concentrations was drawn into a 100-ml Hamilton syringe. For injection, a 28-gauge needle was attached with PE20 tubing to the Hamilton syringe and 80ml of this solution was injected subcutaneously in the hind paw (n 10 per group). The needle punctured the plantar skin and was guided forward in the subcutaneous space to a site just proximal to the pads. Each animal was used only once, and the investigator was unaware of the drugs injected with the FM. To determine if OCT had a systemic effect, 50 ml of 20 mM OCT (the standard dose we used in this study) was injected subcutaneously into one hindpaw (n 9), followed immediately by an injection of 30 ml of 2.5% FM into the contralateral hindpaw, and spontaneous pain behaviors for the FMinjected paw were recorded. To determine drug speci®city, cyclo-somatostatin (c-SOM, Bachem, Bioscience, Inc.), an SSTR antagonist, was dissolved in PBS and 1.3 mM c-SOM (30 ml) was injected into the hindpaw 20 min prior to injection of 50 ml of 20 mM OCT 1 30 ml of 2.5% FM (n 9).
S.M. Carlton et al. / Pain 90 (2001) 233±244
2.2.3. Behavioral testing Spontaneous behaviors resulting from the FM injections were assessed by counting the number of ¯inches and seconds an animal spent lifting and/or licking (L/L) the injected paw in 5-min intervals. A ¯inch was de®ned as a spontaneous, rapid jerk of the foot whether the foot was on the screen or held in the air. The time course of the nociceptive behaviors was plotted as the mean number of ¯inches and amount of time spent L/L in 5-min intervals for 50 min. There were two phases in FM-induced spontaneous pain behaviors. The ®rst phase occurred from 0 to 10 min after injection and the second from 10 to 30 min after injection. The following formulae were used for averaging the data in each phase: (1) ®rst phase: [total number of ¯inches or L/L (s)]/2 (two 5-min intervals in 10 min); (2) second phase: [total number of ¯inches or L/L (s)]/4. 2.3. Electrophysiological studies 2.3.1. In vitro skin-nerve recordings Single unit recordings from C-mechanoheat (CMH) sensitive ®bers in glabrous skin were obtained using a modi®ed in vitro skin-nerve preparation (Fig. 1) (Du et al., 2000; Kress et al., 1992; Reeh, 1986). Twenty-six male Sprague± Dawley rats (200±300 g) were killed with an overdose of CO2 and the glabrous skin of the hindpaw was dissected from each animal with the attached medial and lateral plantar nerves. The preparation was then placed corium side up in an organ bath and superfused (15 ml/min, 348C) with an oxygen-saturated, modi®ed synthetic interstitial ¯uid solution (SIF, in mM: NaCl, 123; KCl, 3.5; MgSO4, 0.7; CaCl2, 2.0; Na gluconate, 9.5; NaH2PO4, 1.7; glucose, 5.5; sucrose, 7.5; HEPES, 10; pH 7.40 ^ 0.05). The plantar nerves were placed in a separate chamber containing a top layer of mineral oil and a bottom layer of SIF. The nerves were desheathed and teased apart on a mirror stage under a dissecting microscope. Small nerve bundles were repeatedly split with sharpened forceps until single unit activity was obtained.
235
2.3.2. Thermal stimulation A feedback-controlled lamp placed beneath the organ bath supplied radiant heat to each receptive ®eld. The beam was focused through the translucent bottom of the bath onto the epidermal side of the skin. A thermocouple was placed in the corium above the light beam to measure intracutaneous temperature. A standard heat ramp was applied to each unit, starting from an adapting temperature of 348C and rising to 478C in 10 s (478C on the corium side was equivalent to 518C on the epidermal side). The temperature at which the second spike was elicited by the heat stimulus was de®ned as heat threshold. 2.3.3. Mechanical stimulation The mechanical threshold for each unit was determined using calibrated von Frey ®laments (Stoelting, Inc.). The ®laments were applied to the receptive ®eld on the corium side of the skin in a uniform fashion, starting with ®laments delivering relatively small amounts of force and ascending (0.1±166.7 mN) until an action potential could be consistently evoked. 2.3.4. Chemical stimulation To investigate responses of units to various drugs, a small plastic ring (5 mm diameter) was placed over the receptive ®eld of each unit and the SIF in the ring was replaced with SIF containing the SSTR agonist OCT, or bradykinin (BK), or BK 1 OCT, or BK 1 OCT 1 the antagonist c-SOM, or OCT 1 naloxone. All drugs dissolved in SIF were buffered to pH 7.40 ^ 0.05. 2.3.5. Effects of an SSTR agonist on thermal responses To investigate whether activation of SSTRs could inhibit responses of CMH ®bers to heat, SIF in the ring was replaced with an ascending series of OCT concentrations ranging from 0.002 to 200 mM made in SIF (pH 7.4). Each concentration was applied for 1 min and the intertrial interval was 5±5.5 min. Threshold temperatures at which nociceptor activity was elicited and discharge rates were measured before and after application of OCT. 2.3.6. Effects of an SSTR agonist and antagonist on BKinduced excitation and sensitization To determine if OCT effected peripheral sensitization, three different paradigms were used with BK as the sensitizing agent. Following characterization of each unit, which included testing for a heat response (Heat 1), all CMH units were exposed to 10 mM BK (BK1) for 1 min, then following a 5-min washout period, the unit was exposed a second time to 10 mM BK alone (BK2) or BK2 1 20 mM OCT, or BK2 1 20 mM OCT 1 1.3mM c-SOM for 1 min, followed within 30 s by retesting of the heat response (Heat 2).
Fig. 1. Schematic drawing of our in vitro skin-nerve preparation in which the glabrous skin of the hindpaw is dissected with the medial and lateral plantar nerves intact. The various forms of stimulation that are applied to the receptive ®eld of each unit are indicated.
2.3.7. Effects of an SSTR agonist and an opiate antagonist on thermal responses To determine if OCT effects were mediated all or in part
236
S.M. Carlton et al. / Pain 90 (2001) 233±244
by opiate receptors, 3 CMH units were tested for a heat response, then exposed to 20 mM OCT 1 2 mM naloxone for 1 min, followed within 30 s by retesting of the heat response. 2.3.8. Neurophysiological recordings Neural activity was recorded using a DAM80 Differential Ampli®er (World Precision Instruments) attached to a monopolar gold wire electrode, with the reference electrode positioned nearby. Units were ®rst identi®ed by manual probing of the glabrous skin with a blunt glass rod. Only units responding to mechanical probing with a clearly de®ned receptive ®eld were studied in detail. Action potentials were analyzed on a Dell computer with a custom-made template-matching program that allowed discrimination of single-unit activity based on the amplitude and wave form of each action potential (Forster and Handwerker, 1990). The conduction velocity of each unit was determined by monopolar electrical stimulation (0.1 ms duration, train frequency 1 Hz) at the most mechanosensitive site in the receptive ®eld of each unit using a Te¯on-coated steel electrode (5 MV impedance, 250 mm shaft diameter). Conduction velocity was determined from the latency of the action potential and the distance from the stimulation electrode to the recording site (measured in millimeters). Units with a conduction velocity of less than 1.6 m/s were classi®ed as C ®bers. 2.4. Statistical analysis Behavioral and electrophysiological data were expressed as mean ^ standard error of the mean (SEM) and evaluated using the Sigmastat program (Jandel Corporation). In the behavioral studies, differences between groups were evalu-
Table 1 Percentages of SSTR2a-labeled axons in the dermal-epidermal junction Animal no.
Labeled axons
Total no. of axons
Percent labeled
Rat 1 Rat 2 Rat 3 Mean ^ SD
21 19 23
198 167 203
10.6 11.4 11.3 11.1 ^ 0.4
ated using a t-test if a normality test was passed or the Kruskal±Wallis test if not. In the electrophysiological studies, differences in discharge rates or threshold temperatures were evaluated with either a Friedman's ANOVA, Kruskal±Wallis test followed by Dunn's post hoc analysis, Wilcoxon signed ranks test or the Mann±Whitney U-test where appropriate. P , 0:05 was considered signi®cant. 3. Results 3.1. Anatomical studies Electron microscopic analyses of SSTR2a immunostaining in peripheral primary afferent terminals demonstrated immunostaining of unmyelinated sensory axons at the dermal-epidermal junction. Reaction product for the SSTR2a was localized on discrete patches on the axonal membrane (Fig. 2A) or diffusely distributed within the axoplasm. Unlabeled axons were also observed (Fig. 2A,B). Counts of labeled and unlabeled axons demonstrated that 11.1 ^ 0.4% of the unmyelinated axons were positively stained for the SSTR2a (Table 1). On occasion, reaction product was observed in some Schwann cell processes (Fig. 2B). Incubating tissue in solutions in which the
Fig. 2. Electron micrographs of Remak bundles at the dermal-epidermal junction which contains (A) one axon immunostained for the SSTR2a (arrow) and several unlabeled axons (arrowheads), and (B) a labeled Schwann cell process (arrow) and unlabeled axons (arrowheads). Bar: 0.25 mm.
S.M. Carlton et al. / Pain 90 (2001) 233±244
237
Fig. 3. Dose±response relationship. Note that animals receiving an intraplantar injection of 20 mM OCT 1 FM show a signi®cant decrease in phase 2 but not phase 1 lifting/licking behavior compared with animals receiving PBS 1 FM (A). Flinching behavior (B) is not affected. (*P , 0:05, significant difference from PBS 1 FM group, Kruskal±Wallis test).
primary antibody had been omitted or in which the antibody had been pre-absorbed with the appropriate peptide resulted in no immunoreaction product in axons or Schwann cells. 3.2. Behavioral studies Intraplantar injection of 2.5% FM resulted in animals attending to the injected paw with L/L behavior and ¯inching (Fig. 3A,B). A biphasic response was observed in the present study with the animals demonstrating these nociceptive behaviors in both phase 1 and phase 2. Compared with FM alone, animals injected with 20 mM OCT 1 2.5% FM had signi®cantly reduced L/L behavior during phase 2, but phase 1 L/L behavior was unaffected (Fig. 3A). There was no change in ¯inching behavior at any dose of OCT (Fig. 3B). The time course study performed with 20 mM OCT 1 2.5% FM demonstrated that L/L behaviors were signi®cantly reduced at 15, 20 and 25 min during phase 2 (Fig. 4). Intraplantar injection of the SSTR antagonist cyclo-somatostatin (1.3 mM, c-SOM), followed by 20 mM OCT 1 2.5%FM, reversed the OCT effects (Fig. 5). That OCT acted locally rather than systemically was demonstrated in animals injected with 20 mM OCT in one hindpaw and 2.5% FM into the contralateral hindpaw (Fig. 5). These animals had nociceptive behaviors similar to animals injected with PBS 1 2.5% FM.
Fig. 4. (A) A time course study demonstrating that co-injection of 20 mM OCT with 2.5% FM signi®cantly attenuates lifting/licking behaviors at 15, 20 and 25 min post-injection during phase 2 (*P , 0:05, signi®cant difference from PBS 1 FM, Kruskal±Wallis). (B) Flinching behavior did not change following intraplantar injection of 20 mM OCT 1 FM.
Fig. 5. Intraplantar injection of 1.3 mM c-SOM blocked the OCT-induced attenuation of FM behaviors such that these animals displayed behaviors that were similar in magnitude to animals receiving PBS 1 FM (PBS 1 FM, and 20 mM OCT 1 FM groups are reprinted from Fig. 3 for comparison). The anti-nociceptive effect of OCT was not due to a systemic effect since animals injected with 20 mM OCT in one hindpaw and 2.5% FM into the contralateral hindpaw (`Contral') had nociceptive behaviors no different from those seen in animals injected with formalin alone. (*P , 0:05, signi®cant difference from the PBS 1 FM group, Kruskal± Wallis).
3.3. Electrophysiological studies Units responding to both mechanical and heat stimuli with conduction velocities below 1.6 m/s were studied (C mechanoheat-sensitive or CMH ®bers). The mean conduction velocity of these ®bers (n 57) was 0.63 ^ 0.03 m/s and ranged from 0.26 to 1.5 m/s. The median force for activation with von Frey ®laments was 27.9 mN, ranging from 1.6 to 166 mN. Some of the drug application paradigms required repeated heating of these CMH ®bers, and it was determined that there was no change in the mean heatevoked discharge rate if intertrial intervals were 5 min or longer (Friedman's ANOVA, Fig. 6). Prior to each drug application, background activity was measured for 1 min for each unit and this activity was subtracted from the activ-
Fig. 6. Exposing units to heat with 5-min intertrial intervals results in responses that are not signi®cantly different from each other indicating that changes in heat responses following drug application are not due to repeated heating of the units (Friedman's ANOVA, P 0:6).
238
S.M. Carlton et al. / Pain 90 (2001) 233±244
ity obtained during the 1 min drug exposure. Prior to heat stimulation, background activity was measured for 10 s for each unit and this activity was subtracted from the activity obtained during the 10-s heat stimulus. 3.3.1. Effects of OCT on thermal responses One-minute applications of ascending concentrations of OCT (0.002±200 mM) to the receptive ®elds of CMH units demonstrated a dose-dependent inhibition of responses to thermal stimuli (Figs. 7 and 8). In particular, 2.0, 20, 60 and 200 mM OCT resulted in signi®cant decreases of 62, 71, 95 and 96%, respectively, in responses to thermal stimuli when compared with control responses (Kruskal± Wallis test followed by Dunn's post hoc analysis, P , 0:05, Fig. 7). Based on these ®ndings, 20.0 mM OCT was chosen as the concentration to test the effect of OCT on BK-induced excitation and sensitization of CMH units. The thermal responses of three additional CMH ®bers were analyzed before and after the exposure of their receptive ®elds to 2 mM naloxone 1 20.0 mM OCT. The OCTinduced inhibition of the heat response in these units was not reversed by naloxone (data not shown). 3.3.2. Effect of BK Application of 10 mM BK to the receptive ®eld of CMH units (n 37) for 1 min resulted in an increase in the mean discharge rate to 0.24 ^ 0.06 impulses/s above background. Of the 37 ®bers tested, 15 (41%) were categorized as `BK responders', meeting the criteria of having an increase in discharge rate that was equal to or greater than 2 standard deviations above mean background for the total population.
Fig. 7. Application of OCT to the receptive ®eld of nociceptors produces a dose-dependent reduction in their responses to thermal stimuli. An ascending series of OCT concentrations, ranging from 0.002 to 200 mM, was applied for 1 min to the receptive ®elds of CMH units and then heat was applied to the receptive ®eld. Note that in a dose-dependent fashion, OCT decreased the mean discharge rates in response to heat. Compared with control values, 2, 20, 60 and 200 mM OCT resulted in signi®cant reductions in the heat response (Kruskal±Wallis test followed by a Dunn's post hoc analysis. N, numbers of individually recorded ®bers at each dose).
Fig. 8. The top trace shows the response of a C nociceptor to a 10-s heat pulse following exposure of the receptive ®eld to synthetic interstitial ¯uid (SIF). The middle trace shows the decrease in response to heat of the same unit following exposure of its receptive ®eld to SIF 1 20 mM OCT. The bottom trace shows our standard 10-s heat pulse.
3.3.3. Effects of OCT and c-SOM on BK-induced excitation Each of the 37 units tested with BK (BK1) was randomly placed into one of three groups and then exposed to either (1) 10 mM BK2, (2) 10 mM BK2 1 20 mM OCT, or (3) 10 mM BK2 1 20 mM OCT 1 1.3 mM c-SOM. Statistical analysis demonstrated that there was no difference in mean responses to BK1 among these three groups (Kruskal±Wallis test, P 0:96). For group 1 (n 15), the second application of BK (BK2) was applied 5 min after the ®rst and the mean discharge rate generated by the second application was signi®cantly reduced to 0.07 ^ 0.03 impulses/s (Wilcoxon signed rank test, P , 0:05), representing a 70% decrease from the BK1 response (Figs. 9 and 10B,C) which demonstrates the well-known phenomenon of tachyphylaxis (Mizumura and Kumazawa, 1996). If BK2 was applied in
Fig. 9. Application of 10 mM BK (BK1) to the receptive ®eld of CMH units results in excitation. A second application of BK (BK2) produces less excitation of the units than BK1 due to tachyphylaxis or BK-induced desensitization. (*P , 0:05, signi®cant difference from BK1, Wilcoxon signed rank test).
S.M. Carlton et al. / Pain 90 (2001) 233±244
239
Fig. 10. Changes in group 1 (see text) BK-induced excitation. (A) The paradigm used to demonstrate BK-induced tachyphylaxis and sensitization to heat for group 1 is shown. (B) The time course of CMH responses (n 15) to BK1 is shown. (C) The time course of CMH responses to BK2 is shown. Note that the background prior to BK2 is higher than that prior to BK1 due to exposure to BK1 and furthermore, that the response to BK2 is much less than that to BK1, demonstrating BK-induced tachyphylaxis.
the presence of OCT (group 2, n 12), the BK-induced excitation was reduced to essentially zero (Fig. 11). If BK2 was applied in the presence of both OCT and c-SOM (group 3, n 10), the inhibitory effect of OCT was reversed in that the units responded with a discharge rate of 0.05 ^ 0.06 impulses/s, which was not signi®cantly different from BK2 (Fig. 11). These results demonstrate that local OCT inhibits the BK-induced excitation of CMHs, and this inhibition can be reversed by the SST antagonist c-SOM. 3.3.4. Effects of OCT and c-SOM on units sensitized to heat by BK There was great variability in the magnitude of individual heat responses. Therefore, in order to determine if a heat response was altered following exposure to various drugs, a power analysis using a power of 0.80, an alpha of 0.05, and a standard deviation of 0.64 (obtained from the 37 units used in this part of the study) was performed. Based on this criterion, if a unit displayed an increase in discharge rate that was .30% of its original heat response, the unit was considered a `responder'. Following application of 10 mM BK2 (group 1), 8/15 units (53%) had an increase in Heat 2 responses that were equal to or greater than 30% of Heat 1. The mean discharge
rate increased 46%, from 0.98 ^ 0.01 to 1.43 ^ 0.28 impulses/s (Wilcoxon signed rank test, P , 0:01), and the
Fig. 11. Changes in BK-induced excitation: units were divided into three groups and discharge rates measured following exposure to either BK (BK2, group 1), BK2 1 OCT (group 2) or BK2 1 OCT 1 c-SOM (group 3). In the presence of OCT, BK-induced excitation was essentially eliminated. In contrast, when the BK antagonist c-SOM was added to the BK2 1 OCT, it reversed the effect of OCT such that the mean discharge rate of this group was not signi®cantly different (NS) from that of BK2. (*P , 0:05, signi®cant difference from BK2, Mann±Whitney U-test).
240
S.M. Carlton et al. / Pain 90 (2001) 233±244
mean threshold decreased from 41.7 ^ 0.4 to 40.2 ^ 0.68C comparing Heat 1 with Heat 2, and these changes were signi®cant (Fig. 12, Wilcoxon signed rank test, P , 0:05). Thus as expected, BK sensitized CMH units in this paradigm. When BK2 was applied in the presence of OCT in group 2 (n 12), 11/12 units (92%) demonstrated decreased responses to heat, with the mean discharge rate decreasing from 1.16 ^ 0.25 (Heat 1) to 0.83 ^ 0.17 impulses/s (Heat 2) and the mean temperature threshold increasing from 41.2 ^ 0.7 (Heat 1) to 42.5 ^ 0.78C (Heat 2). Comparing group 1 with group 2 demonstrated that addition of OCT to BK signi®cantly reduced the Heat 2 response (Fig. 13A, Mann±Whitney U-test, P , 0:05) and signi®cantly increased heat threshold (Fig. 13B, Mann±Whitney U-test, P , 0:05). When BK2 was applied in the presence of both OCT and c-SOM in group 3 (n 10), the inhibitory effect of OCT was reversed in 9/10 units (90%) with the mean discharge rate increasing from 0.87 ^ 0.15 (Heat 1) to 1.9 ^ 0.34 impulses/s (Heat 2) and the mean threshold temperature decreasing from 41.7 ^ 0.7 (Heat 1) to
40 ^ 0.48C (Heat 2). Comparing group 1 with group 3 demonstrated that addition of OCT 1 c-SOM to BK resulted in no signi®cant difference a mean discharge rate (Fig. 13A) or mean threshold temperature (Fig. 13B). These results demonstrate that OCT inhibits the thermal responses of nociceptors sensitized by BK, and this inhibition can be reversed by the SSTR antagonist c-SOM. 4. Discussion The present study demonstrates that approximately 11% of peripheral unmyelinated sensory axons at the dermalepidermal junction of the glabrous skin of the rat hindpaw express the SSTR2a. Due to the fact that there are no sympathetic efferents at the dermal-epidermal junction in the rat glabrous skin (Coggeshall et al., 1997), the labeled unmyelinated axons must be sensory axons. This provides a reasonable anatomical basis for the behavioral and physiological changes that follow intraplantar application of the
Fig. 12. Changes in sensitized heat responses induced by BK (note that the paradigm used is illustrated in Fig. 10A) Exposure to BK resulted in a signi®cant increase in the mean discharge rate (A) and a signi®cant decrease in the mean threshold temperature for heat activation (B) for group 1 units (Wilcoxon signed rank test, *P , 0:05, **P , 0:01). The time course of unit activation for Heat 1 (C) and Heat 2 (D) are shown. Background discharge rates prior to Heat 2 are higher than that prior to Heat 1 because of exposure to BK.
S.M. Carlton et al. / Pain 90 (2001) 233±244
Fig. 13. Changes in sensitized heat responses induced by BK: Application of BK2 1 OCT to receptive ®elds of nociceptors resulted in a 42% decrease in discharge rate in response to heat (A) and a signi®cant increase in threshold to activation (B) when compared with BK2. If c-SOM was applied with BK2 1 OCT, the OCT effect on the heat response was reversed and discharge rates (A) and threshold temperatures (B) were not signi®cantly different (NS) from BK2. (*P , 0:05, signi®cant difference from BK2, Mann±Whitney U).
SSTR agonist OCT. It is, of course, not certain that the OCT is working exclusively through SSTR2a, since OCT will activate SSTR2, SSTR3 and SSTR5 receptor subtypes (Patel et al., 1995; Viollet et al., 1995), and mRNA for SSTR3 has been localized in DRG cells (Senaris et al., 1995). The fact that there are several SSTR subtypes that may be activated by OCT may explain the discrepancy between the relatively low percentage of SSTR2a-positive ®bers observed in the anatomical study and the relatively high percentage of ®bers that respond to OCT in the skinnerve preparation. More precise delineations of the distribution and function of the different SSTR subtypes will await development of more antibodies and speci®c agonists and antagonists. It is of interest, however, that the DRG neurons that express the SSTR2a are restricted to a subpopulation of cells with a medium diameter (Schulz et al., 1998), that also label positively for the Griffonia simplicifolia isolectin B4 (IB4) and have very little overlap with DRG cells labeled for CGRP, TrkA, or SST itself (Michael et al., 1999). Thus, it appears that SSTR2a -containing sensory neurons fall into the non-peptidergic class. Whether other SSTRs subtypes are located on different functional groups of DRG cells remains to be determined. The behavioral ®ndings in the present study are that intraplantar injection of the SSTR agonist OCT reduces FMinduced phase 2 L/L behaviors and that c-SOM, an SSTR
241
antagonist, reverses the OCT effect. The lack of OCT effect on phase 1 behaviors suggests that SSTR activation may primarily reduce in¯ammation which is believed to be one of the major factors underlying phase 2 formalin pain behaviors (Dubuisson and Dennis, 1977). It is not clear why ¯inching and L/L behavior were differentially affected. However, the lack of effect of OCT on ¯inching behavior suggests that the neurocircuitry underlying this presumed segmentally organized re¯exive behavior may be distinct from that underlying L/L behavior, which is more likely to involve supraspinal mechanisms. This hypothesis is supported by a previous study demonstrating preservation of ¯inching behavior in the ®rst phase of the formalin test in spinally transected animals (Wheeler-Aceto and Cowan, 1991). Control studies demonstrate that OCT produces this effect through local and not systemic actions since injection of OCT into one hindpaw and FM into the contralateral hindpaw results in nociceptive behaviors similar to those seen in animals injected with FM alone. Demonstration of an analgesic effect following SSTR activation is consistent with earlier ®ndings demonstrating that intraplantar injection of SST reduces mechanical hyperalgesia in carrageenan-induced in¯ammation (Corsi et al., 1997) and systemic injection of an SST analogue produces antinociceptive effects in the tail¯ick and hot plate tests (Eschalier et al., 1991). Interestingly, these changes are attributed to opioid actions (Corsi et al., 1997) or to central rather than peripheral actions (Eschalier et al., 1991), but our preliminary naloxone studies and the localization of SSTRs on sensory axons suggests that this is a direct action on peripheral SSTRs. More precise evaluations of changes in nociceptor activity following SSTR activation can be obtained from single afferent ®ber recordings. Using an in vitro skin-nerve preparation, we determined the effects of OCT on normal and sensitized nociceptors. Exposure of CMH units to the SSTR agonist OCT reduces, in a dose-dependent manner, the mean discharge rate in response to noxious heat. In a separate experiment, units are sensitized by exposure to BK, a known in¯ammatory mediator and sensitizing agent (Beck et al., 1974; Chahl and Iggo, 1977; Mizumura and Kumazawa, 1996). BK application results in excitation of 41% of the ®bers tested, con®rming previous studies (Koltzenburg et al., 1992; Lang et al., 1990), OCT signi®cantly attenuates the thermal responses of BK-sensitized nociceptors, and cSOM reverses the OCT effects. It is known that BK stimulation of DRG cells is associated with membrane depolarization and Ca 21 in¯ux (Burgess et al., 1989). In contrast, SST stimulation of DRG cells is associated with a decrease in Ca 21 channel conductance (Taddese et al., 1995). Furthermore, pre-treatment of cultured DRG cells with SST will inhibit BK-stimulated arachidonate release and these ®ndings support a coupling of SSTRs to Ca 21 channel inactivation through pertussis toxin sensitive G-proteins (Gammon et al., 1990). Calcium channel inactivation is known to interfere with release of endogenous substances such as
242
S.M. Carlton et al. / Pain 90 (2001) 233±244
excitatory peptides. Results from Heppelmann and Pawlak (1997) and this study suggest that SST-induced Ca 21 channel inactivation may also interfere with sensory neuron responses to mechanical and thermal stimulation, respectively. Our hypothesis is that OCT is directly activating SSTRs on primary afferents. Another possibility, however, is that SST is activating opioid receptors. Such an indirect action is suggested by observations that SST or its analogs bind to opioid receptors (Gulya et al., 1986; Maurer et al., 1982; Pelton et al., 1986; Pugsley and Lippmann, 1978; Terenius, 1976), and naloxone, a widely used opioid antagonist, prevents the analgesia that follows intraventricular injection of SST (Rezek et al., 1978). On the other hand, naloxone has no effect on responses of dorsal horn cells to SST application (Randic and Miletic, 1978; Sandkuhler et al., 1990), and clinical studies show that naloxone has no effect on the analgesia provided by epidural application of SST (Chrubasik et al., 1984, 1985). Furthermore, SST ligands are often effective in patients who are no longer responding to opioids) (Mollenholt et al., 1994), (Meynadier et al., 1985; Paice et al., 1996) and the receptors for SST and opioids seem to be on different populations of DRG cells (Taddese et al., 1995). In the present study, 2 mM naloxone, a dose that has been previously shown to reverse the inhibitory effects of morphine and selective m and k agonists on the spontaneous ®ring of peripheral ®bers in UV-irradiated skin (Andreev et al., 1994), did not reverse the inhibitory effects of OCT. Accordingly, most of the evidence indicates that SST and opioids are acting through different systems. A ®nal point is that SST is an endogenous anti-in¯ammatory peptide (SzolcsaÂnyi et al., 1998a). Administration of systemic SST or its ligands, aside from the profound effects on gastrointestinal and endocrine systems (Lamberts et al., 1996; Reichlin, 1983), depresses in¯ammatory responses to various stimuli (Karalis et al., 1994; Matucci-Cerinic et al., 1995; SzolcsaÂnyi et al., 1998a,b). Of particular interest is the ®nding that in¯ammation in one part of the body releases enough SST systemically to depress in¯ammatory responses in another body region (SzolcsaÂnyi et al., 1998a,b). Thus, it is likely that during neurogenic in¯ammation, the release of SST, an inhibitory anti-in¯ammatory peptide, occurs concomitantly with the release of CGRP and SP, which are excitatory, pro-in¯ammatory peptides (SzolcsaÂnyi et al., 1998a,b), and that the balance of these peptides determines the degree of both the pain and edema of in¯ammation (Heppelmann and Pawlak, 1997). This offers the hope that it may be possible to promote enhanced release of endogenous inhibitory compounds to aid in reducing the pain, swelling and other long-term deleterious changes that accompany in¯ammation. 5. Conclusion In summary, we demonstrate the presence of SSTRs on
peripheral primary afferents. Activation of these receptors with OCT diminishes the spontaneous nociceptive behaviors that follows intraplantar FM injection, and reduces the responses of nociceptors excited and sensitized by BK. These ®ndings are consistent with earlier observations that local administration of SST or SSTR agonists in the knee joint signi®cantly improves the pain of osteo- and rheumatoid arthritis (Matucci-Cerinic and Marabini, 1998; Silveri et al., 1994) and diminishes the single unit ®ring that results from noxious movements of the normal and in¯amed joint (Heppelmann and Pawlak, 1997). Peripheral administration would largely avoid the reported CNS neurotoxicity of SST without losing the analgesic properties associated with SSTR activation. Thus, SST or more long-lasting SSTR analogues may be of considerable use in controlling nociceptive input from the periphery, both in the normal state and after primary afferent sensitization. Acknowledgements The authors thank Kristin Sawyer and Vicki Wilson for their excellent secretarial assistance, and Zhixia Ding and Greg Hargett for excellent technical assistance. This work was supported by NIH NS11255 to S.M.C. and R.E.C., NS27910 and NS40700 to S.M.C and NS10161 to R.E.C. References Andreev NY, Urban L, Dray A. Opioids suppress spontaneous activity of polymodal nociceptors in rat paw skin induced by ultraviolet irradiation. Neuroscience 1994;58:793±798. Beck PW, Handwerker HO, Hermann O. Bradykinin and serotonin effects on various types of cutaneous nerve ®bres. P¯uÈgers Arch 1974;347:209±222. Burgess GM, Mullaney J, McNeil M, Dunn P, Rang HP. Second messengers involved in the action of bradykinin on cultured sensory neurons. J Neurosci 1989;9:3314±3325. Burgess JR, Shepherd JJ, Murton FJ, Parameswaran V, Greenaway TM. Effective control of bone pain by octreotide in a patient with metastatic gastrinoma. Med J Aust 1996;164:725±727. Chahl LA, Iggo A. The effects of bradykinin and prostaglandin E1 on rat cutaneous nerve activity. Br J Pharmacol 1977;59:343±347. Chapman V, Dickenson AH. The effects of sandostatin and somatostatin on nociceptive transmission in the dorsal horn of the rat spinal cord. Neuropeptides 1992;23:147±152. Chrubasik J, Meynadier J, Blond S, Scherpereel P, Ackerman E, Weinstock M, Bonath K, Cramer H, Wunsch E. Somatostatin. A potent analgesic. Lancet 1984;1:1208±1209. Chrubasik J, Meynadier J, Scherpereel P, Wunsch E. The effect of epidural somatostatin on postoperative pain. Anesth Analg 1985;64:1085±1088. Coggeshall RE, Carlton SM. Ultrastructural analysis of NMDA, AMPA and kainate receptors on unmyelinated and myelinated axons in the periphery. J Comp Neurol 1998;391:78±86. Coggeshall RE, Zhou S, Carlton SM. Opioid receptors on peripheral sensory axons. Brain Res 1997;764:126±132. Corsi MM, Ticozzi C, Netti C, Fulgenzi A, Tiengo M, Gaja G, Guidobono F, Ferrero ME. The effect of somatostatin on experimental in¯ammation in rats. Anesth Analg 1997;85:1112±1115. Du J, Koltzenburg M, Carlton SM. Glutamate-induced excitation and sensitization of nociceptors in rat glabrous skin. Pain 2000;89:187±198.
S.M. Carlton et al. / Pain 90 (2001) 233±244 Dubuisson D, Dennis SG. The formalin test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats. Pain 1977;4:161±174. Eschalier A, Aumaitre O, Ardid D, Fialip J, Duchene-Marullaz P. Longlasting antinociceptive effect of RC-160, a somatostatin analog, in mice and rats. Eur J Pharm 1991;199:119±121. Forster C, Handwerker HO. Automatic classi®cation and analysis of microneurographic spike data using a PC/AT. J Neurosci Methods 1990;31:109±118. Gammon CM, Lyons SA, Morell P. Modulation by neuropeptides of bradykinin-stimulated second messenger release in dorsal root ganglion neurons. Brain Res 1990;518:159±165. Gaumann DM, Yaksh TL. Intrathecal somatostatin in rats: antinociception only in the presence of toxic effects. Anesthesiology 1988;68:733±742. Gaumann DM, Yaksh TL, Post C, Wilcox GL, Rodriguez M. Intrathecal somatostatin in cat and mouse studies on pain, motor behavior, and histopathology. Anesth Analg 1989;68:623±632. Gulya K, Pelton JT, Hruby VJ, Yamamura HI. Cyclic somatostatin octapeptide analogues with high af®nity and selectivity toward mu opioid receptors. Life Sci 1986;38:2221±2229. Helmchen C, Fu QG, Sandkuhler J. Inhibition of spinal nociceptive neurons by microinjections of somatostatin into the nucleus raphe magnus and the midbrain periaqueductal gray of the anesthetized cat. Neurosci Lett 1995;187:133±136. Heppelmann B, Pawlak M. Inhibitory effect of somatostatin on the mechanosensitivity of articular afferents in normal and in¯amed knee joints of the rat. Pain 1997;73:377±382. HoÈkfelt T, Elde R, Hohansson O, Luft R, Arimura A. Immunohistochemical evidence for the presence of somatostatin, a powerful inhibitory peptide, in some primary sensory neurons. Neurosci Lett 1975;1:231± 235. Kapicioglu S, Gokce E, Kapicioglu Z, Ovali E. Treatment of migraine attacks with a long-acting somatostatin analogue (octreotide, SMS 201-995). Cephalalgia 1997;17:27±30. Karalis K, Mastorakos G, Chrousos GP, Tolis G. Somatostatin analogues suppress the in¯ammatory reaction in vivo. J Clin Invest 1994;93:2000± 2006. Koltzenburg M, Kress M, Reeh PW. The nociceptor sensitization by bradykinin does not depend on sympathetic neurons. Neuroscience 1992;46:465±473. Kress M, Koltzenburg M, Reeh PW, Handwerker HO. Responsiveness and functional attributes of electrically localized terminals of cutaneous C®bers in vivo and in vitro. J Neurophysiol 1992;68:581±595. Lamberts SWJ, van der Lely A-J, De Herder WW, Ho¯and LJ. Octreotide. N Engl J Med 1996;334:246±254. Lang E, Novak A, Reeh PW, Handwerker HO. Chemosensitivity of ®ne afferents from rat skin in vitro. J Neurophysiol 1990;63:887±901. Matucci-Cerinic M, Marabini S. Somatostatin treatment for pain in rheumatoid arthritis: a double blind versus placebo study in knee involvement. Med Sci Res 1998;16:233±234. Matucci-Cerinic M, Borrelli F, Generini S, Cantelmo A, Marcucci I, Martelli F, Romagnoli P, Bacci S, Conz A, Marinelli P, Marabini S. Somatostatin-induced modulation of in¯ammation in experimental arthritis. Arthritis Rheum 1995;38:1687±1693. Maurer R, Gaehwiler BH, Buescher HH, Hill RC, Roemer D. Opiate antagonistic properties of an octapeptide somatostatin analog. Proc Natl Acad Sci USA 1982;79:4815±4817. Meynadier J, Chrubasik J, Dubar M, Wunsch E. Intrathecal somatostatin in terminally ill patients. A report of two cases. Pain 1985;23:9±12. Michael GJ, Tabrizi I, Saggu R. Localization and Regulation of the Somatostatin SST2a receptor in subpopulations of primary sensory neurons. Soc Neurosci Abstr 1999;25:409. Miletic V, Randic M. Neonatal rat spinal cord slice preparation: postsynaptic effects of neuropeptides on dorsal horn neurons. Dev Brain Res 1982;2:432±438. Mizumura K, Kumazawa T. Modi®cation of nociceptor responses by in¯ammatory mediators and second messengers implicated in their
243
action ± a study in canine testicular polymodal receptors. Prog Brain Res 1996;113:115±141. Mollenholt P, Post C, Rawal N, Freedman J, HoÈkfelt T, Paulsson I. Antinociceptive and `neurotoxic' actions of somatostatin in rat spinal cord after intrathecal administration. Pain 1988;32:95±105. Mollenholt P, Rawal N, Gordh T, Olsson Y. Intrathecal and epidural somatostatin for patients with cancer. Analgesic effects and postmortem neuropathologic investigations of spinal cord and nerve roots. Anesthesiology 1994;81:534±542. Murase K, Nedeljkov V, Randic M. The actions of neuropeptides on dorsal horn neurons in the rat spinal cord slice preparation: an intracellular study. Brain Res 1982;234:170±176. Paice JA, Penn RD, Kroin JS. Intrathecal octreotide for relief of intractable nonmalignant pain: 5-year experience with two cases. Neurosurgery 1996;38:203±207. Patel YC, Greenwood MT, Panetta R, Demchyshyn L, Nizik H, Srikant CB. The somatostatin receptor family. Life Sci 1995;57:1249±1265. Pelton JT, Kazmierski W, Guyla K, Yamamura HI, Hruby VJ. Design and synthesis of conformationally constrained somatostatin analogues with high potency and speci®city for m opioid receptors. J Med Chem 1986;29:2370±2375. Penn RD, Paice JA, Kroin JS. Ocetreotide: a potent new non-opiate analgesic for intrathecal infusion. Pain 1992;49:13±19. Plourde V, Lembo T, Shui Z, Parker J, Mertz H, Tache Y, Sytnik B, Mayer E. Effects of the somatostatin analogue octreotide on rectal afferent nerves in humans. Am J Physiol 1993;265:G742±G751. Pugsley TA, Lippmann W. Effect of somatostatin analogues and 17-adihydroequilin on rat brain opiate receptors. Res Commun Chem Pathol Pharm 1978;21:153±157. Randic M, Miletic V. Depressant actions of methionine-enkephalin and somatostatin in cat dorsal horn neurones activated by noxious stimuli. Brain Res 1978;152:196±202. Reeh PW. Sensory receptors in mammalian skin in an in vitro preparation. Neurosci Lett 1986;66:141±146. Reichlin S. Medical progress: somatostatin. N Engl J Med 1983;309:1495± 1501. Rezek M, Havlicek V, Leybin L, LaBella FS, Friesen H. Opiate-like naloxone-reversible action of somatostatin given intracerebrally. Can J Phys Pharm 1978;56:227±231. Sandkuhler J, Fu Q-G, Helmchen C. Spinal somatostatin superfusion in vivo affects activity of cat nociceptive dorsal horn neurons: comparison with spinal morphine. Neuroscience 1990;34:565±576. Schulz S, Schreff M, Schmidt H, Handel M, Przewlocki R, Hollt V. Immunocytochemical localization of somatostatin receptor SST2a in the rat spinal cord and dorsal root ganglia. Eur J Neurosci 1998; 10:3700±3708. Senaris RM, Schindler M, Humphrey PPA, Emson PC. Expression of somatostatin receptor 3 mRNA in the motorneurones of the rat spinal cord, and the sensory neurones of the spinal ganglia. Mol Brain Res 1995;29:185±190. Silveri F, Morosini P, Brecciaroli D, Cervini C. Intra-articular injection of somatostatin in knee osteoarthritis: clinical results and IGF-1 serum levels. Int J Clin Pharm Res 1994;14:79±85. SzolcsaÂnyi J, Helyes Z, Oroszi G, Nemeth J, PinteÂr E. Release of somatostatin and its role in the medication of the anti-in¯ammatory effect induced by antidromic stimulation of sensory ®bres of rat sciatic nerve. Br J Pharmacol 1998a;123:936±942. SzolcsaÂnyi J, PinteÂr E, Helyes Z, Oroszi G, Nemeth J. Systemic anti-in¯ammatory effect induced by counter-irritation through a local release of somatostatin from nociceptors. Br J Pharmacol 1998b;125:916±922. Taddese A, Nah S-Y, McCleskey EW. Selective opioid inhibition of small nociceptive neurons. Science 1995;270:1366±1369. Taura P, Planella V, Balust J, Beltran J, Anglada T, Carrero E, Burgues S. Epidural somatostatin as an analgesic in upper abdominal surgery: a double-blind study. Pain 1994;59:135±140. Terenius L. Somatostatin and ACTH are peptide with partial antagonistlike selectivity for opiate receptors. Eur J Pharm 1976;38:211±213.
244
S.M. Carlton et al. / Pain 90 (2001) 233±244
Viollet C, Prevost G, Maubert E, Faivre-Bauman A, Gardette R, Kordon C, Loudes C, Slama A, Epelbaum J. Molecular pharmacology of somatostatin receptors. Clin Pharmacol 1995;9:107±113. Wheeler-Aceto H, Cowan A. Standardization of the rat formalin test for the evaluation of analgesics. Psychopharmacology 1991;104:35±44.
Wiesenfeld-Hallin Z. Intrathecal somatostatin modulates spinal sensory and re¯ex mechanisms: behavioral and electrophysiological studies in the rat. Neurosci Lett 1985;62:69±74. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109±110.