Neuropeptides (1992) 23,147-l 52 0 Longman Group UK Ltd 1992
The Effects of Sandostatin and Somatostatin on Nociceptive Transmission in the Dorsal Horn of the Rat Spinal Cord V. CHAPMAN and A. H. DICKENSON Department of Pharmacology, request to VC)
University College London, Gower St, London, UK. (Reprint
Abstract -The role of somatostatin and a stable analogue, sandostatin (Octreotide), on the responses of spinal cord neurones in vivo was investigated in the rat. Electrical C-fibre stimulation was used as a model of acute nociception and the response to subcutaneous formalin was used as a model of longer term events. lntrathecal pre-treatment with sandostatin and somatostatin did not alter the C-fibre response, wind up or A6 responses of the cells. However, intrathecal pre-treatment with sandostatin and somatostatin inhibited both the first and second phases of the formalin response dose dependently. Thus, sandostatin (20 pg) and somatostatin (150 pg) inhibited the first phase (66 f 12% inhibition and 52 f 13% respectively) and second phase (91 f 2% inhibition and 39 + 16% inhibition respectively). The second phase of the formalin response was more sensitive to somatostatin and sandostatin than the first. Sandostatin was approximately 400 times more potent than somatostatin on the second phase of the response. Subcutaneous sandostatin (100 mg/kg) significantly inhibited both the first and second phase of the formalin response whereas the local peripheral administration of sandostatin (20 pg) only inhibited the second phase of the formalin response. Introduction
The distribution of somatostatin within the dorsal horn of the spinal cord has led to speculation that somatostatin has a role in the transmission of nociceptive inputs. The peptide is present in small diameter cells in the dorsal root ganglion (1) and afferent terminals in the substantia gelatinosa of the spinal cord (2). The primary sensory neurons have been
Date received 2 June 1992 Date accepted 30 June 1992
confirmed as the origin of spinal cord somatostatin in addition to ascending and descending somatostatin pathways (3). The dorsal horn somatostatin content has been shown to be 15 times higher than the ventral horn (3) and the dorsal horn content of somatostatin is highest in lamina II (4). Somatostatin has been shown to be released from the dorsal horn of the spinal cord following noxious thermal stimulation (5). In vitro studies have shown that iontophoretic application of somatostatin results in the hyperpolarisation of dorsal horn neurons and a reduction in spontaneous firing (6). 147
148 Behavioural studies have previously shown that the intrathecal application of somatostatin results in an initial excitatory response followed by antinociception which was only observed in the presence of toxic effects (7). This excitatory response (biting and scratching) to intrathecally applied somatostatin has been observed in other behavioural studies (8,9) but in the absence of antinociception. It has previously been shown that a large array of intrathecally applied drugs including morphine result in a biting and scratching response (see references in 10). The basis of this behaviour has been difficult to ascertain and is not necessarily indicative of pain. A safety margin for the intrathecal administration of somatostatin in rats was defined by Mollenholt el al (11). The threshold intrathecal dose of somatostatin for antinociception was 10 ug as compared to 30 pg which gave rise to chronic motor impairment associated with necrotic changes and loss of an immunohistochemical marker for motoneurones. A subsequent study showed that intrathecally applied somatostatin produced a dose dependent antinociception in the tail flick test which could be separated from a transient motor dysfunction (< 30 min) and high dose induced neuronal necrosis (12). The dose response curve for somatostatin on cortical neurones is biphasic, lower concentrations are excitatory whereas higher concentrations are inhibitory. At higher doses long lasting and often irreversible neuronal depolarization indicative of neuronal membrane damage were produced (13). Although there have been many claims that somatostatin has toxic effects, human studies have shown that both intrathecal and epidural application of somatostatin is effective in the management of acute, post-operative and chronic cancer pain (14,15). This study investigates the effect of intrathecal somatostatin on the response of nociceptive dorsal horn neurones in the spinal cord of the anaesthetized rat. We have used a model of acute and longer term pain to establish whether the conflicting roles of somatostatin previously reported are due to the different models of pain studied. Since somatostatin is an unstable peptide and is rapidly degraded, we have used somatostatin and a stable analogue sandostatin ( 16) to investigate the role of somatostatin in the spinal transmission of nociceptive information. Sandostatin has previously been shown in clinical studies to be a potent analgesic (16).
NEUROPEPTIDES
Methods The techniques used have been previously described (17, 18). Sprague-Dawley rats (200-250 g) were anaesthetized with 2-3% halothane in 66% N2 0 -33% 02. A laminectomy was performed exposing segments L l-L3 and the anaesthesia was then maintained with 1.5% halothane. Glass-coated tungsten electrodes were used to record extracellularly the responses of nociceptive convergent multireceptive dorsal horn neurones (responding to pinch, heat and touch) to A/3 and C fibre afferent inputs from the skin of the hindpaw following transcutaneous electrical stimulation (2 ms wide pulses at 0.5 Hz). Responses were elicited by a train of 16 stimuli (3 times threshold for Al.3and C fibres) at 5 min intervals. The initial response (input) was calculated as the number of action potentials produced by the first stimulation at C-fibre strength multiplied by the total number of stimuli (16). The frequency dependent potentiation of dorsal horn nociceptive neurones to repetitive C-fibre stimulation (wind up) (17) was studied. Wind up was taken as the difference between the total number of action potentials at C-fibre latenties (90 msec and later) produced by the train of 16 stimuli and the input. In addition to this model of acute nociceptive transmission we also used the formalin response as the model of longer term pain (18). Formalin (50 fl of 5% formaldehyde in saline) was injected subcutaneously at 20 min post drug treatment into the centre of the receptive field on the toe of the hindpaw. The response to formalin consists of a prolonged firing of the dorsal horn neurones. The firing of single dorsal horn neurones to formalin was counted, the first 10 min of the formalin response was taken as the first peak and the remaining 50 min of firing was taken as the second peak. After 3 stable control responses, the effect of intrathecal somatostatin (15 E and 150 c(%/50u.l) and sandostatin (Octreotide) (2 ug and 20 pg/50 pl) were studied on the deep convergent dorsal horn response to electrica stimulation over a time course of 30 min. Intrathecal somatostatin (15 clg, n = 11 and 150 pg, n = 8) and sandostatin (0.02 pg, n = 10, 0.2pg,n=7,2Clg,n=6and20pg,n=4)weregiven as 30 min pre-treatments prior to the peripheral injection of formalin. Subcutaneous administration of sandostatin (1 mgkg, 10 mgkg and 100 mg/kg) and the local
SANLIOSTATIN
AND
SOMATOSTATIN
EFFECT
ON UT
SPINAL
The mean maximal effects of intrathecal somatostatin and sandostatin on the C fibre, AP fibre and wind up of the population of cells in the study. The results (mean and SE.) are expressed as percentages of the control response. Table
Cfibre Response
Treatment
Somatostatin (15 I%) Somatostatin (150 I%) Sandostatin (2 %) Sandostatin (20 M)
APfibre Response
llSrt4
98 f 6
108 f9
87fll
96If 15 109 f 29
88k21 115f4
Wind up Response
no. of cells
156f21
9
80+9
6
lOSf40
6
65+23
4
peripheral administration of sandostatin (0.2 ccg, 2 clg, 20 pg) were also given as 30 min pre-treatments prior to the peripheral injection of formalin.
149
CORD
Results
Intrathecal pre-treatment with somatostatin and sandostatin did not result in a significant change in the wind up of the dorsal horn neurones, C fibre or AP fibre response to electrical stimulation (Table). The control response of a deep convergent dorsal horn neurones to the peripheral injection of formalin which is representative of the population of control responses is shown in Figure 1. The intrathecal pretreatment with the lowest doses of sandostatin (0.02 c(g) and somatostatin (15 B) did not significantly inhibit the response to formalin. By contrast, pre-treatment with higher doses of intrathecal sandostatin (20 clg) and somatostatin (150 pg) did significantly reduce the first peak of the formalin response, 66% inhibition, p = 0.06 and 52% inhibition, p < 0.02 respectively.
10011
4
5C l-
time
’
Fig. 1 The control response of a deep convergent dorsal horn neurone (representative of the population of the cells studied) to the subcutaneous injection of formalin (50 pl of 5% formaldehyde). The spikes per second are plotted on the vertical axis and the time (seconds) on the horizontal axis. The response was recorded over a period of 60 min, the fist 10 min of response was taken as the first phase and the remainder as the second phase.
150
NEUROPEPTIDES
100-
0.06) whereas the lower dose of somatostatin (15 pg) did not have a significant effect on the second phase of the response. Subcutaneous pre-treatment with the highest dose of sandostatin (100 mgkg) significantly inhibited both the first and second phase of the formalin response (43% inhibition, p = 0.04 and 40% inhibition, p = 0.03 respectively). The peripheral administration of sandostatin (20 pg) significantly inhibited only the second phase of the formalin response (43%, p = 0.016). p =
So-
60-
0” E Y
40-
z 6Q m-
Discussion
o-
1 ,
.Ol
.I
1
10
loo
“7
1000
DOSE&is) Fig. 2 The dose response relationship of somatostatin and sandostatin on the second phase of the formalin response. The inhibitory action of sandostatin on the second phase of the formalin response was approximately 400 times greater than somatostatin. o--o-o Sandostatin (n ranged from 4-10) O-O-C Somatostatin (n ranged from S-10).
The second phase of the formalin response was more sensitive than the first phase to the inhibitory effects of somatostatin and sandostatin. Sandostatin was found to be approximately 400 times more potent than somatostatin as shown by the dose response relationship of somatostatin and sandostatin on the second phase of the formalin response (Fig. 2). Sandostatin (2 and 20 pg) significantly inhibited the second phase of the formalin response (59% inhibition, p < 0.01 and 91% inhibition, p < 0,001 respectively) in a dose dependent manner. The formalin response of a single dorsal horn neurone pre-treated with 20 pg of intrathecal sandostatin is shown in Figure 3, where both the first and second phase of the response are greatly reduced as compared to the control response (Fig. 1). Intrathecal somatostatin (150 cl9>significantly inhibited the second phase of the formalin response (39% inhibition,
Sandostatin and somatostatin were seen to have an inhibitory action when applied onto the spinal cord in the model of longer term nociception (formalin response) but not in the model of acute pain. The highest dose of somatostatin studied was not within the safety margin as defined by Mollenholt et al (11). At this dose we did not observe any change in the response of the neurones to electrical stimulation as compared to the controls, indicating that the presence of the somatostatin on the spinal cord was not causing neurotoxic effects on the primary afferent fibres nor these sensory neurones, at least over the period of study. In addition we were able to record from other neurons once the intrathecally applied somatostatin was washed off. Other studies have suggested that the neurotoxicity associated with somatostatin is selective for the motor neurones (12). The electrophysiological model used in this study does not depend on a motor response to pain and therefore even if either somatostatin or sandostatin did cause a neurotoxic effect on motoneurones it will not influence our results. Sandostatin was considerably more potent than somatostatin, which is likely to be a consequence of sandostatin not being enzymatically degraded and therefore having a long duration of action (19). Sandostatin did not appear to have any toxic effects on the spinal cord, nor any action on acute inputs since the C fibre, wind up and Al3 fibre response to electrical stimulation was unaltered (as compared to controls) in the presence of all intrathecal doses. The peripheral administration of sandostatin only inhibited the second phase of the formalin response.
SANDOSTATIN
AND
SOMATOSTATIN
EFFECT
ON RAT
SPINAL
CORD
151
Fig. 3 The formalin response of a representative deep convergent dorsal horn neurone pre-treated with 20 pg of intrathecal sandostatin. The spikes per second are plotted on the vertical axis and the time (seconds) on the horizontal axis. Both the first and second phases of the response are significantly inhibited in the presence of intrathecal sandostatin (20 pg) as compared to the control response.
It is plausible that peripheral somatostatin receptors are only effective during pain states associated with inflammation. The peripheral receptors may therefore inhibit the actions of proalgesic mediators such as those active in the second phase of the formalin response (20) but not influence the direct activation of the nociceptors by the electrical stimulation. The subcutaneous administration of a high dose of sandostatin resulted in a significant inhibition of both phases of the formalin response This suggests that systemic sandostatin was able to cross the blood brain barrier and have a central site of action since an effect on both peaks of the formalin response was only seen with spinal application. The subcutaneous dose of sandostatin used was substantially greater than the highest intrathecal dose studied but the intrathecal dose inhibited the formalin response to a greater degree; it therefore appears unlikely that the
inhibitory action of spinal sandostatin can be attributed to leakage into the systemic circulation and a peripheral site of action. The spinal inhibitory action of somatostatin and sandostatin is in agreement with previous in vitro studies (6). The ability of intrathecal somatostatin and sandostatin to inhibit the formalin response but not the response to electrical stimulation may reflect the different nociceptive intensities ofthese two models. The nociceptive response to formalin is of a lower intensity but longer duration of action and inhibited by somatostatin and sandostatin, whereas electrical activation of the nociceptors results in a higher intensity of nociceptive transmission but over a shorter time course which is not inhibited by somatostatin and sandostatin. Another possibility is that different central mediators are involved in the two different states, and we have recently shown that central
152 indomethacin has an identical differential effect on the electrical and formalin evoked responses (21). This study shows a marked antinociceptive role of somatostatin and sandostatin at the spinal level and also a weaker inhibitory action at the peripheral nociceptor terminal but only in nociceptive states associated with peripheral inflammation. These results support previous clinical studies which have shown that spinal somatostatin (14, 15) and sandostatin (16,22) are potent analgesics which are effective in the management of cancer pain. However, although our results show that a genuine antinociception can be observed without overt toxic effects on sensory neurones the neurotoxic effects of spinal somatostatin on motoneurones in the rat (12) has to be borne in mind in the clinical use of these agents.
NEUROPEPTIDES
7.
8.
9.
10. 11.
12.
13.
Acknowledgements This work was supported by the MRC and a SERC-CASE award with the Sandoz Institute for Medical Research to V. Chapman.
14.
15.
References 1. Tuscherer, M. M. and Seybold, V. S. (1985). Immunohistochemical studies of substance P, cholecystokininoctapeptide and somatostatin in dorsal root ganglia of the rat. Neuroscience 14: 593-605. 2. Hokfelt, T., Elde, R., Johansson, O., Luft, R., Nilsson, G. and Arimura, A. (1976). Immunohistochemical evidence for separate populations of somatostatin-containing and substance P containing primary afferent neurons in the rat. Neuroscience 1: 131-136. 3. Stine, S. M., Yang, H. and Costa, E. (1982). Evidence for ascending and descending intraspinal as well as primary sensory somatostatin projections in the rat spinal cord. Journal of Neurochemistry 38: 1144-1150. 4. Polack, J. M. and Bloom, S. R. (1986). Somatostatin localisation in tissues. Stand. J. Gastroenterol2 1: 1 l-2 1. 5. Kuraishi, Y., Hirota, N., Sato, Y., Satoh, M. and Takagi, H. (1985). Evidence that substance P and somatostatin transmit separate information related to pain in the spinal dorsal horn. Brain Research 325: 294-298. 6. Murase, K., Nedeljov, V. and Randic, M. (1982). The actions of neuroneutides on dorsal horn neurons in the rat soinal cord
16.
17.
18.
19.
20.
21.
22.
slice preparation: an intrathecal study. Brain Research 234: 170-176. Gaumann,D.M. andYaksh,T.L. (1988).Intrathecalsomatostatin in rats: Antinociceptive only in the presence of toxic effects. Anesthesiology 68: 733-742. Seybold, V. S., Hylden, J. L. K. and Wilcox, G. L. (1982). Intrathecal substance P and somatostatin in rats: Behaviours indicative of sensation. Peptides 3: 49-54. Wiesenfeld-Hallin, 2. (1985). Intrathecal somatostatin modulates spinal sensory and reflex mechanisms: behavioural and electrophysiological studies in the rat. Neuroscience Letters 62: 69-74. Iversen, L. (1989). Central actions of substance P and related tachykinins. Journal of Psychopharmacology 3(l): 1-6. Mollenholt, P., Post, C., Rawal, N., Freedman, J., Hokflet, T. and Paulsson, I. (1988). Antinociceptive and neurotoxic actions of somatostatin in rat spinal cord after intrathecal administration. Pain 32: 95-105. Mollenholt, P., Post, C. I., Pat&son, I. and Rawal, N. (1990). Intrathecal and epidural somatostatin in rats: can antinociception, motor effects and neurotoxicity be separated. Pain 43: 363-370. Delfs, J. R. and Dichter, M. A. (1983). Effects of somatostation on mammalian cortical neurons in culture: physiological actions and unusual dose response characteristics. Neuroscience 4: 1176-l 188. Chrubasik, J., Meynadier, J., Scherpereel, P. and Wtlnsch, E. (1985). The effect of epidural somatostatin on postoperative pain. Anesth. Analg. 64: 1085-1088. Meynadier, J., Chrubasik, J., Dubar, M. and Wiinsch, E. (1985). Intrathecal somatostatin in terminally ill patients. A report of two cases. Pain 23: 9-12. Penn, R. D., Paice, J. A. and Kroin, J. S. (1992). Octreotide: a potent new non-opiate analgesic for intrathecal infusion. Pain 49: 13-19. Dickenson, A. H. and Sullivan, A. F. (1990). Differential effects of excitatory amino acid antagonists on dorsal horn nociceptive neurones in the rat. Brain Research 506: 3 l-39. Haley, J. E., Sullivan, A. F. and Dickenson, A. H. (1990). Evidence for spinal N-methyl-D-aspartate receptor involvement in prolonged chemical nociception in the rat. Brain Research 5 18: 2 18-226. Pless, J., Bauer, W., Briner, U. et al. (1986). Chemistry and pharmacology of SMS 201-995, an acting octapeptide analogue of somatostatin. Stand. J. Gastroenterol 119: 54-64. Dickenson, A. H. (1991). Recent advances in the physiology and pharmacology of pain: plasticity and its implications for clinical analgesia. J. Psychopharmacology 5: 342-35 1. Chapman, V. and Dickenson, A. H. (1992). The spinal and peripheral roles of bradykinin and prostaglandins in nociceptive processing in the rat. European J. Pharmacology (In press). Penn, R. D., Paice, J. A. and Kroin, J. S. (1990). Intrathecal octreotide for cancer pain. The Lancet 335: 738.
The Effects of Sandostatin and Somatostatin on Nociceptive Transmission in the Dorsal Horn of the Rat Spinal Cord V. CHAPMAN and A. H. DICKENSON Department of Pharmacology, request to VC)
University College London, Gower St, London, UK. (Reprint
Abstract -The role of somatostatin and a stable analogue, sandostatin (Octreotide), on the responses of spinal cord neurones in vivo was investigated in the rat. Electrical C-fibre stimulation was used as a model of acute nociception and the response to subcutaneous formalin was used as a model of longer term events. lntrathecal pre-treatment with sandostatin and somatostatin did not alter the C-fibre response, wind up or A6 responses of the cells. However, intrathecal pre-treatment with sandostatin and somatostatin inhibited both the first and second phases of the formalin response dose dependently. Thus, sandostatin (20 pg) and somatostatin (150 pg) inhibited the first phase (66 f 12% inhibition and 52 f 13% respectively) and second phase (91 f 2% inhibition and 39 + 16% inhibition respectively). The second phase of the formalin response was more sensitive to somatostatin and sandostatin than the first. Sandostatin was approximately 400 times more potent than somatostatin on the second phase of the response. Subcutaneous sandostatin (100 mg/kg) significantly inhibited both the first and second phase of the formalin response whereas the local peripheral administration of sandostatin (20 pg) only inhibited the second phase of the formalin response. Introduction
The distribution of somatostatin within the dorsal horn of the spinal cord has led to speculation that somatostatin has a role in the transmission of nociceptive inputs. The peptide is present in small diameter cells in the dorsal root ganglion (1) and afferent terminals in the substantia gelatinosa of the spinal cord (2). The primary sensory neurons have been
Date received 2 June 1992 Date accepted 30 June 1992
confirmed as the origin of spinal cord somatostatin in addition to ascending and descending somatostatin pathways (3). The dorsal horn somatostatin content has been shown to be 15 times higher than the ventral horn (3) and the dorsal horn content of somatostatin is highest in lamina II (4). Somatostatin has been shown to be released from the dorsal horn of the spinal cord following noxious thermal stimulation (5). In vitro studies have shown that iontophoretic application of somatostatin results in the hyperpolarisation of dorsal horn neurons and a reduction in spontaneous firing (6). 147
148 Behavioural studies have previously shown that the intrathecal application of somatostatin results in an initial excitatory response followed by antinociception which was only observed in the presence of toxic effects (7). This excitatory response (biting and scratching) to intrathecally applied somatostatin has been observed in other behavioural studies (8,9) but in the absence of antinociception. It has previously been shown that a large array of intrathecally applied drugs including morphine result in a biting and scratching response (see references in 10). The basis of this behaviour has been difficult to ascertain and is not necessarily indicative of pain. A safety margin for the intrathecal administration of somatostatin in rats was defined by Mollenholt el al (11). The threshold intrathecal dose of somatostatin for antinociception was 10 ug as compared to 30 pg which gave rise to chronic motor impairment associated with necrotic changes and loss of an immunohistochemical marker for motoneurones. A subsequent study showed that intrathecally applied somatostatin produced a dose dependent antinociception in the tail flick test which could be separated from a transient motor dysfunction (< 30 min) and high dose induced neuronal necrosis (12). The dose response curve for somatostatin on cortical neurones is biphasic, lower concentrations are excitatory whereas higher concentrations are inhibitory. At higher doses long lasting and often irreversible neuronal depolarization indicative of neuronal membrane damage were produced (13). Although there have been many claims that somatostatin has toxic effects, human studies have shown that both intrathecal and epidural application of somatostatin is effective in the management of acute, post-operative and chronic cancer pain (14,15). This study investigates the effect of intrathecal somatostatin on the response of nociceptive dorsal horn neurones in the spinal cord of the anaesthetized rat. We have used a model of acute and longer term pain to establish whether the conflicting roles of somatostatin previously reported are due to the different models of pain studied. Since somatostatin is an unstable peptide and is rapidly degraded, we have used somatostatin and a stable analogue sandostatin ( 16) to investigate the role of somatostatin in the spinal transmission of nociceptive information. Sandostatin has previously been shown in clinical studies to be a potent analgesic (16).
NEUROPEPTIDES
Methods The techniques used have been previously described (17, 18). Sprague-Dawley rats (200-250 g) were anaesthetized with 2-3% halothane in 66% N2 0 -33% 02. A laminectomy was performed exposing segments L l-L3 and the anaesthesia was then maintained with 1.5% halothane. Glass-coated tungsten electrodes were used to record extracellularly the responses of nociceptive convergent multireceptive dorsal horn neurones (responding to pinch, heat and touch) to A/3 and C fibre afferent inputs from the skin of the hindpaw following transcutaneous electrical stimulation (2 ms wide pulses at 0.5 Hz). Responses were elicited by a train of 16 stimuli (3 times threshold for Al.3and C fibres) at 5 min intervals. The initial response (input) was calculated as the number of action potentials produced by the first stimulation at C-fibre strength multiplied by the total number of stimuli (16). The frequency dependent potentiation of dorsal horn nociceptive neurones to repetitive C-fibre stimulation (wind up) (17) was studied. Wind up was taken as the difference between the total number of action potentials at C-fibre latenties (90 msec and later) produced by the train of 16 stimuli and the input. In addition to this model of acute nociceptive transmission we also used the formalin response as the model of longer term pain (18). Formalin (50 fl of 5% formaldehyde in saline) was injected subcutaneously at 20 min post drug treatment into the centre of the receptive field on the toe of the hindpaw. The response to formalin consists of a prolonged firing of the dorsal horn neurones. The firing of single dorsal horn neurones to formalin was counted, the first 10 min of the formalin response was taken as the first peak and the remaining 50 min of firing was taken as the second peak. After 3 stable control responses, the effect of intrathecal somatostatin (15 E and 150 c(%/50u.l) and sandostatin (Octreotide) (2 ug and 20 pg/50 pl) were studied on the deep convergent dorsal horn response to electrica stimulation over a time course of 30 min. Intrathecal somatostatin (15 clg, n = 11 and 150 pg, n = 8) and sandostatin (0.02 pg, n = 10, 0.2pg,n=7,2Clg,n=6and20pg,n=4)weregiven as 30 min pre-treatments prior to the peripheral injection of formalin. Subcutaneous administration of sandostatin (1 mgkg, 10 mgkg and 100 mg/kg) and the local
SANLIOSTATIN
AND
SOMATOSTATIN
EFFECT
ON UT
SPINAL
The mean maximal effects of intrathecal somatostatin and sandostatin on the C fibre, AP fibre and wind up of the population of cells in the study. The results (mean and SE.) are expressed as percentages of the control response. Table
Cfibre Response
Treatment
Somatostatin (15 I%) Somatostatin (150 I%) Sandostatin (2 %) Sandostatin (20 M)
APfibre Response
llSrt4
98 f 6
108 f9
87fll
96If 15 109 f 29
88k21 115f4
Wind up Response
no. of cells
156f21
9
80+9
6
lOSf40
6
65+23
4
peripheral administration of sandostatin (0.2 ccg, 2 clg, 20 pg) were also given as 30 min pre-treatments prior to the peripheral injection of formalin.
149
CORD
Results
Intrathecal pre-treatment with somatostatin and sandostatin did not result in a significant change in the wind up of the dorsal horn neurones, C fibre or AP fibre response to electrical stimulation (Table). The control response of a deep convergent dorsal horn neurones to the peripheral injection of formalin which is representative of the population of control responses is shown in Figure 1. The intrathecal pretreatment with the lowest doses of sandostatin (0.02 c(g) and somatostatin (15 B) did not significantly inhibit the response to formalin. By contrast, pre-treatment with higher doses of intrathecal sandostatin (20 clg) and somatostatin (150 pg) did significantly reduce the first peak of the formalin response, 66% inhibition, p = 0.06 and 52% inhibition, p < 0.02 respectively.
10011
4
5C l-
time
’
Fig. 1 The control response of a deep convergent dorsal horn neurone (representative of the population of the cells studied) to the subcutaneous injection of formalin (50 pl of 5% formaldehyde). The spikes per second are plotted on the vertical axis and the time (seconds) on the horizontal axis. The response was recorded over a period of 60 min, the fist 10 min of response was taken as the first phase and the remainder as the second phase.
150
NEUROPEPTIDES
100-
0.06) whereas the lower dose of somatostatin (15 pg) did not have a significant effect on the second phase of the response. Subcutaneous pre-treatment with the highest dose of sandostatin (100 mgkg) significantly inhibited both the first and second phase of the formalin response (43% inhibition, p = 0.04 and 40% inhibition, p = 0.03 respectively). The peripheral administration of sandostatin (20 pg) significantly inhibited only the second phase of the formalin response (43%, p = 0.016). p =
So-
60-
0” E Y
40-
z 6Q m-
Discussion
o-
1 ,
.Ol
.I
1
10
loo
“7
1000
DOSE&is) Fig. 2 The dose response relationship of somatostatin and sandostatin on the second phase of the formalin response. The inhibitory action of sandostatin on the second phase of the formalin response was approximately 400 times greater than somatostatin. o--o-o Sandostatin (n ranged from 4-10) O-O-C Somatostatin (n ranged from S-10).
The second phase of the formalin response was more sensitive than the first phase to the inhibitory effects of somatostatin and sandostatin. Sandostatin was found to be approximately 400 times more potent than somatostatin as shown by the dose response relationship of somatostatin and sandostatin on the second phase of the formalin response (Fig. 2). Sandostatin (2 and 20 pg) significantly inhibited the second phase of the formalin response (59% inhibition, p < 0.01 and 91% inhibition, p < 0,001 respectively) in a dose dependent manner. The formalin response of a single dorsal horn neurone pre-treated with 20 pg of intrathecal sandostatin is shown in Figure 3, where both the first and second phase of the response are greatly reduced as compared to the control response (Fig. 1). Intrathecal somatostatin (150 cl9>significantly inhibited the second phase of the formalin response (39% inhibition,
Sandostatin and somatostatin were seen to have an inhibitory action when applied onto the spinal cord in the model of longer term nociception (formalin response) but not in the model of acute pain. The highest dose of somatostatin studied was not within the safety margin as defined by Mollenholt et al (11). At this dose we did not observe any change in the response of the neurones to electrical stimulation as compared to the controls, indicating that the presence of the somatostatin on the spinal cord was not causing neurotoxic effects on the primary afferent fibres nor these sensory neurones, at least over the period of study. In addition we were able to record from other neurons once the intrathecally applied somatostatin was washed off. Other studies have suggested that the neurotoxicity associated with somatostatin is selective for the motor neurones (12). The electrophysiological model used in this study does not depend on a motor response to pain and therefore even if either somatostatin or sandostatin did cause a neurotoxic effect on motoneurones it will not influence our results. Sandostatin was considerably more potent than somatostatin, which is likely to be a consequence of sandostatin not being enzymatically degraded and therefore having a long duration of action (19). Sandostatin did not appear to have any toxic effects on the spinal cord, nor any action on acute inputs since the C fibre, wind up and Al3 fibre response to electrical stimulation was unaltered (as compared to controls) in the presence of all intrathecal doses. The peripheral administration of sandostatin only inhibited the second phase of the formalin response.
SANDOSTATIN
AND
SOMATOSTATIN
EFFECT
ON RAT
SPINAL
CORD
151
Fig. 3 The formalin response of a representative deep convergent dorsal horn neurone pre-treated with 20 pg of intrathecal sandostatin. The spikes per second are plotted on the vertical axis and the time (seconds) on the horizontal axis. Both the first and second phases of the response are significantly inhibited in the presence of intrathecal sandostatin (20 pg) as compared to the control response.
It is plausible that peripheral somatostatin receptors are only effective during pain states associated with inflammation. The peripheral receptors may therefore inhibit the actions of proalgesic mediators such as those active in the second phase of the formalin response (20) but not influence the direct activation of the nociceptors by the electrical stimulation. The subcutaneous administration of a high dose of sandostatin resulted in a significant inhibition of both phases of the formalin response This suggests that systemic sandostatin was able to cross the blood brain barrier and have a central site of action since an effect on both peaks of the formalin response was only seen with spinal application. The subcutaneous dose of sandostatin used was substantially greater than the highest intrathecal dose studied but the intrathecal dose inhibited the formalin response to a greater degree; it therefore appears unlikely that the
inhibitory action of spinal sandostatin can be attributed to leakage into the systemic circulation and a peripheral site of action. The spinal inhibitory action of somatostatin and sandostatin is in agreement with previous in vitro studies (6). The ability of intrathecal somatostatin and sandostatin to inhibit the formalin response but not the response to electrical stimulation may reflect the different nociceptive intensities ofthese two models. The nociceptive response to formalin is of a lower intensity but longer duration of action and inhibited by somatostatin and sandostatin, whereas electrical activation of the nociceptors results in a higher intensity of nociceptive transmission but over a shorter time course which is not inhibited by somatostatin and sandostatin. Another possibility is that different central mediators are involved in the two different states, and we have recently shown that central
152 indomethacin has an identical differential effect on the electrical and formalin evoked responses (21). This study shows a marked antinociceptive role of somatostatin and sandostatin at the spinal level and also a weaker inhibitory action at the peripheral nociceptor terminal but only in nociceptive states associated with peripheral inflammation. These results support previous clinical studies which have shown that spinal somatostatin (14, 15) and sandostatin (16,22) are potent analgesics which are effective in the management of cancer pain. However, although our results show that a genuine antinociception can be observed without overt toxic effects on sensory neurones the neurotoxic effects of spinal somatostatin on motoneurones in the rat (12) has to be borne in mind in the clinical use of these agents.
NEUROPEPTIDES
7.
8.
9.
10. 11.
12.
13.
Acknowledgements This work was supported by the MRC and a SERC-CASE award with the Sandoz Institute for Medical Research to V. Chapman.
14.
15.
References 1. Tuscherer, M. M. and Seybold, V. S. (1985). Immunohistochemical studies of substance P, cholecystokininoctapeptide and somatostatin in dorsal root ganglia of the rat. Neuroscience 14: 593-605. 2. Hokfelt, T., Elde, R., Johansson, O., Luft, R., Nilsson, G. and Arimura, A. (1976). Immunohistochemical evidence for separate populations of somatostatin-containing and substance P containing primary afferent neurons in the rat. Neuroscience 1: 131-136. 3. Stine, S. M., Yang, H. and Costa, E. (1982). Evidence for ascending and descending intraspinal as well as primary sensory somatostatin projections in the rat spinal cord. Journal of Neurochemistry 38: 1144-1150. 4. Polack, J. M. and Bloom, S. R. (1986). Somatostatin localisation in tissues. Stand. J. Gastroenterol2 1: 1 l-2 1. 5. Kuraishi, Y., Hirota, N., Sato, Y., Satoh, M. and Takagi, H. (1985). Evidence that substance P and somatostatin transmit separate information related to pain in the spinal dorsal horn. Brain Research 325: 294-298. 6. Murase, K., Nedeljov, V. and Randic, M. (1982). The actions of neuroneutides on dorsal horn neurons in the rat soinal cord
16.
17.
18.
19.
20.
21.
22.
slice preparation: an intrathecal study. Brain Research 234: 170-176. Gaumann,D.M. andYaksh,T.L. (1988).Intrathecalsomatostatin in rats: Antinociceptive only in the presence of toxic effects. Anesthesiology 68: 733-742. Seybold, V. S., Hylden, J. L. K. and Wilcox, G. L. (1982). Intrathecal substance P and somatostatin in rats: Behaviours indicative of sensation. Peptides 3: 49-54. Wiesenfeld-Hallin, 2. (1985). Intrathecal somatostatin modulates spinal sensory and reflex mechanisms: behavioural and electrophysiological studies in the rat. Neuroscience Letters 62: 69-74. Iversen, L. (1989). Central actions of substance P and related tachykinins. Journal of Psychopharmacology 3(l): 1-6. Mollenholt, P., Post, C., Rawal, N., Freedman, J., Hokflet, T. and Paulsson, I. (1988). Antinociceptive and neurotoxic actions of somatostatin in rat spinal cord after intrathecal administration. Pain 32: 95-105. Mollenholt, P., Post, C. I., Pat&son, I. and Rawal, N. (1990). Intrathecal and epidural somatostatin in rats: can antinociception, motor effects and neurotoxicity be separated. Pain 43: 363-370. Delfs, J. R. and Dichter, M. A. (1983). Effects of somatostation on mammalian cortical neurons in culture: physiological actions and unusual dose response characteristics. Neuroscience 4: 1176-l 188. Chrubasik, J., Meynadier, J., Scherpereel, P. and Wtlnsch, E. (1985). The effect of epidural somatostatin on postoperative pain. Anesth. Analg. 64: 1085-1088. Meynadier, J., Chrubasik, J., Dubar, M. and Wiinsch, E. (1985). Intrathecal somatostatin in terminally ill patients. A report of two cases. Pain 23: 9-12. Penn, R. D., Paice, J. A. and Kroin, J. S. (1992). Octreotide: a potent new non-opiate analgesic for intrathecal infusion. Pain 49: 13-19. Dickenson, A. H. and Sullivan, A. F. (1990). Differential effects of excitatory amino acid antagonists on dorsal horn nociceptive neurones in the rat. Brain Research 506: 3 l-39. Haley, J. E., Sullivan, A. F. and Dickenson, A. H. (1990). Evidence for spinal N-methyl-D-aspartate receptor involvement in prolonged chemical nociception in the rat. Brain Research 5 18: 2 18-226. Pless, J., Bauer, W., Briner, U. et al. (1986). Chemistry and pharmacology of SMS 201-995, an acting octapeptide analogue of somatostatin. Stand. J. Gastroenterol 119: 54-64. Dickenson, A. H. (1991). Recent advances in the physiology and pharmacology of pain: plasticity and its implications for clinical analgesia. J. Psychopharmacology 5: 342-35 1. Chapman, V. and Dickenson, A. H. (1992). The spinal and peripheral roles of bradykinin and prostaglandins in nociceptive processing in the rat. European J. Pharmacology (In press). Penn, R. D., Paice, J. A. and Kroin, J. S. (1990). Intrathecal octreotide for cancer pain. The Lancet 335: 738.
