Pain, 51 (1992) 343-347 0 1992 Elsevier Science
343 Publishers
B.V. All rights reserved
0304-3959/92/$05.00
PAIN 02175
Intrathecal somatostatin in the guinea pig: effects on spinal cord blood flow, histopathology and motor function Peter Mollenholt,
Claes Post, Ivar Paulsson
and Narinder
Rawal
Department of Anesthesiology and Intenske Care, drebro Medical Center Hospital, &ebro (Sweden) and Astra Alab AB, Research and Development Laboratories and Astra Toxicological Laboratories, Siidertiilje (Sweden) (Received
11 February
1992, revision
received
7 July 1992, accepted
20 July 1992)
In the present investigation, the vasoconstrictive, motor and neurodegenerative effects of intratheSummary cal somatostatin (SST) were assessed in guinea pigs implanted with lumbar intrathecal catheters. Five consecutive dose increments of SST (5, 10, 15, 30 and 60 pg) to a total of 120 pg during the period of 16 f 3 min, resulted in a moderate (< 20%), gradual decrease of the spinal blood flow monitored with the laser-doppler method. A subsequent injection of clonidine (50 pg) or norepinephrine (10 pg) resulted in a more pronounced decrease of spinal blood flow (35% and 79%, respectively). Three consecutive, daily intrathecal injections of 30 or 60 pg SST did not cause any loss of weight support or paralysis of the hind limbs. There were no histopathological changes in the white or gray matter of the thoracic and lumbar sections of the spinal cords. It is concluded that SST, in the doses studied, is not neurodegenerative in guinea pigs. These findings are in contrast to those previously seen in rats. The implication of this study may be the necessity to use several alternate animal species in order to evaluate the antinociceptive and neurodegenerative properties of the peptides administered by the intrathecal route and the choice of dose to be compared across species. Key words: Peptides; Somatostatin;
Spinal cord; Vasoconstriction;
Introduction
The presence of somatostatin (SST) in primary sensory neurons (C afferents) projecting to the neuroaxis supports its role as a putative neurotransmitter or neuromodulator for pain (Hokfelt et al. 1976). Experimental studies in rats have shown that intrathecal (i.t.) administration of SST is associated with antinociception and motor blockade (Spampinato et al. 1988; Mollenholt et al. 1990) and also with neurodegeneration (Mollenholt et al. 1988). It has been suggested (Freedman et al. 1988; Long 1988) that the ‘neurotoxic’ effects of i.t. SST in rat spinal cord may partly be due to ischemia resulting from vasoconstriction of spinal vessels. However, the susceptibility to SST-induced ischemia appears to be species dependent (Gordin et
Neurotoxicity;
Intrathecal;
(Guinea pig>
al. 1978). This has also been demonstrated for substance P receptor antagonists (Post and Folkers 1985; Post et al. 1987). The present study addresses the issue of whether circulatory events and neuronal lesions following i.t. SST are as pronounced in guinea pigs as in rats.
Materials
and methods
Preparation of drug solution Somatostatin (Ferring AB, Sweden), clonidine (Boehringer Ingelheim, Germany) and norepinephrine bitartrate (Sigma, USA) were dissolved in 0.9% saline. All solutions were prepared immediately prior to injection.
Experiment 1 Correspondence to: Peter and Intensive Care, brebro Grebro, Sweden.
Mollenholt, Dept. of Anesthesiology Medical Center Hospital, S-701 85
Animals and animal preparation. Male guinea pigs (weight range: 250-300 gl of the Dunkin-Hartley strain (Sahlins, Sweden) were used. Experimental work was reviewed by an Institutional Animal Care and Use Committee. The animals were anesthetized with 5%
344 enflurane (Efrane’“) in O2 +N,O (1 I/min + 1 I/min of each) for approximately 1.5 min. When fully anesthetized, the enflurane concentration was reduced to t.S?. with the same concentration ratio of 0, and N@. The laser-doppler experiments were carried out in animals with their spinal cords exposed. For this. they were placed in a David Kopf stereotaxic frame (Tajunga. CA; model 900). which was comhined with a model 980 spinal unit. The spinal column was dissected free. and the cord was exposed by bilateral laminectomy over the lumbar L,-L, segments. Care was taken not to damage the dura mater. The spinal column was fixed in the spinal unit in order to minimize respiratory movements of the cord. A PE-IO catheter, previously stretched to double its length to make it more flexible and non-traumatizing, was then inserted through a slit in the dura mater under a dissection microscope. ’ l~u.~~r-~iopp~ermethod. Blood Bow in the spinal cord was monitored with the laser-doppier technique using a Perimed model Pf 2 laser-doppler instrument (Perimed. Sweden) (Tentand 1982). The laser-doppler probe (model Pf 103) was positioned by a micromanipulator 0.5-l mm above the darsolateral surface of the spinal cord, close to the tip of the catheter. The operated field and probe were covered with a 2.5? agar gel (in 0.9Oh saline) to reduce water evaporation. Recordings were performed with a time-constant of 3 set and the gain was set to t or 3. Readings were recorded on a strip-chart recorder, and the results calculated as blood flow relative to the pre-injection value. Each animal served as its own control. When a steady signal was achieved, which usually occurred within 15-X min, a control volume of 25 @I of 0.9% saline was injected intrathecatly. This resulted in a minor and transient decrease in spinal blood flow. After recovery of blood flow to baseline, 5 consecutive i.t. injections of SST were administered. The doses were 5, IO. 15. 30 and 60 pg (2 wg/plI). This was followed by injection of saline (5-22.5 ~1) to flush the catheter. After the final SST dose, 2 animals were injected Lt. with clonidine (50 gg) and 5 animals with norepinephrine (10 pg). These injections were given as positive controls with known vasoconstrictive substances. Both drugs were dissolved in IO ~1 of saline followed by 15 ~1 of saline to flush the catheter. The total volume of drug and saline flush was kept constant at 25 ~1 except for the 60 Pg SST dose where it was 45 ~1. Injections were made only after a steady blood flow signal resulting from the previous dose was obtained.
100
3 OEI
-
80 -
-3
8-6 02
iia 35
c
2
0 \o
60 -
40 -
t NaCl
I
1
I
1
5
10
15
30
1
60
SSTir-ls)
Fig. I. Log dose-response curve of laser-doppler measurements of spinal cord blood flow. The results are given as means + S.E.M., from 7 animals, in per cent of the pre-drug values. The treatment and saline control values were compared by I-way ANOVA 1* * P < 0.01).
at a nominal thickness of 5 km and stained with Nissl’s stain. Microscopic criteria for intactness of the spinal cord were, inter alia, absence of neuronal necrosis, with pyknosis and hyperchromasia of the nucleus; early degenerative changes including chromatolysis; signs of gliosis and signs of vascular damage. Neither the experimenter nor the pathologist were aware of the treatment until the entire experiment had been completed.
Statistical methods Statistical analysis was performed by analysis (ANOVA). Values are presented as means + S.E.M.
of
variance
Results Experiment 2 An&u& and animal prep~rut~on. Male Dunkin-Hartley guinea pigs (weight range: 2.50-300 gf were anesthetized with 60 mg/kg pentobarbital injected intraperitoneally (i.p.) fMebumalB’, ACO. Sweden). The chronic PE-IO catheter (previously stretched to double its length) was then inserted it. with the tip placed in the lumbar rrgion (Yaksh and Rudy 1976). All the animals showed normal motor behavior after the recovery period. Drug treatment. Twelve guinea pigs were randomly assigned to 1 of 2 treatment groups (6 animals/group). One group receiving 30 p.g and the other 60 pg SST/anima~ intrathecaliy f2 pg/pul) followed in both groups by S-10 ~1 saline/animal to flush the catheter. These injections were administered daily for 3 consecutive days. One control animal received 25 ~1 it. saline daily instead of SST. Mofor effecfs. Motor effect was defined as loss of weight support or flaccid paralysis of the hind limbs. The time of onset and duration of motor blockade were noted every 5 min for a period of 60 min. Histological eduation. All 16 animals were killed on the 4th day try an i.p. overdose of pentobarhitat. The spinal cords were excised after bilateral laminectomy where care was taken to avoid producing traumatic lesions and immediately immersed into a neutral, buffered 3.7% formaldehyde solution. Specimens from the lumbar and lower thoracic region were embedded in Paraplast (Lancer, USA) and cut
Expe~rn~~t 1 Laser-doppler
experiments. The Lt. application of saline caused a small, brief drop in spinal blood flow. The subsequent i.t. SST dose escalation (5, 10, 15, 30 and 60 pg) resulted in a gradual decrease in spinal blood flow which reached significance (P < 0.01) at the total dose of 120 @g/animal. At this dose, the spinal blood flow decreased to 81 rt 11% of its pre-injection value (Fig. 1). A subsequent i.t. application of 50 pg clonidine in 2 animals reduced their spinal blood flow to 65 + 11% (Fig. 2) and a subsequent it. application of 10 icl.gnorepinephrine in another 5 animals, resulted in a reduction of the spinal blood flow to 21 3r:15% (Fig. 3). The steady, lowest blood flow signal corresponding to the maximal decrease in flow level, was obtained within 2-3 min following each dose. Accordingly, the average period of measurement was 16 I!I3 min/animal.
34.5
20 1
OJt
NaCl
1
I
I
5
10
15
t
,
I
60 Clonidine (50 w9)
30
SST k4
Fig. 2. Laser-doppler measurements of spinal cord blood flow. The effect of 120 pg SST and 50 pg clonidine. The results are given as means+S.E.M. Data from 2 animals in per cent of the pre-drug values. The clonidine treatment values were compared with the SST treatment values and saline control values by l-way ANOVA (** P < 0.01).
Experiment 2 Motor effects. No pain related behavior (e.g., caudally directed bites and scratches, vocalization) or motor block was noted in control animals or in animals receiving 30 and 60 pg SST. Histopathological findings. No signs of necrosis or degenerative changes were seen in the neurons at any level of the spinal cords investigated. Discussion The present experiments show the vasoconstrictive effects of i.t. SST on the spinal blood vessels of the
NaCl
I
I
I
I
5
10
15
30
SST(w)
I
guinea pig. The decrease in blood flow was moderate (20%) but significant. In the dose range investigated, SST had no motor blocking or neurodegenerative effects in this species. In our previous studies in rats, we established the threshold i.t. SST dose for antinociception and ‘neurotoxicity’ associated with permanent motor impairment, neuronal necrosis and loss of an immunohistochemical marker for motoneurons in the spinal cord (Mollenholt et al. 1988). Furthermore, it was possible to separate the antinociceptive and motor blocking actions of SST following both i.t. and epidural administration (Mollenholt et al. 1990). However, the margin of safety (antinociceptive vs. ‘toxic’ effects) was quite narrow. In the present investigation, no behavior indicative of sensation or motor block was noted following 3 consecutive, daily i.t. injections of SST. Although the doses were of the magnitude that produce irreversible paraparesis and extensive neuronal necrosis in the rat (Long 19881, no neurodegeneration was noted in any of the guinea pigs, In the present study, histopathological evaluation was made 24 h after the 3rd i.t. injection of SST. It may be argued that this 24-h period was too short for the ‘neurotoxic’ lesions to develop fully. However, our earlier studies in rats showed histopathological changes within 72 h after a single drug exposure (Mollenholt et al. 1988). In order to ascertain that the spinal cord blood flow in the guinea pig responds adequately to locally applied vasoconstrictive drugs, the SST injection was followed by clonidine (Fig. 2) or norepinephrine (Fig. 3) given as positive controls. A rapid reduction in the local blood flow followed these injections indicative that the spinal microcirculation responded as described previously (Gordh et al. 1986) ‘. Although the SST-dependent vasoconstriction has been described previously by many investigators (Neff et al. 1985; Tornebrandt et al. 1987; Tydtn et al. 1979; Vora et al. 19861, the mechanism of this action is speculative. The relatively rapid onset of vasoconstriction ( < 1 min) (Freedman et al. 1988) may suggest that SST acts directly on the vascular smooth muscle (Tyden et al. 1979). Considering the differential distribution of peptidergic fibres (SST, substance P, Met-enkephalin, oxytocin) to intermediolateral (IML) cell column of the spinal cord (Holets and Elde 19891, it is possible that the dose-dependent enhancement of IML cell column function may evoke significant hemodynamic effects due to a sympathetic stimulation via SST afferent fibers
t
60 Norepinephrine (10 cls)
Fig. 3. Laser-doppler measurements of spinal cord blood flow. The effects of 120 pg SST and 10 pg norepinepbrine. Data from 5 animals. The results are given as means f S.E.M. The norepinephine treatment values were compared with SST treatment values and saline control values by l-way ANOVA (* * P < 0.01).
’ In preliminary experiments, we have also measured the spinal cord blood flow effects in rat with the laser-doppler method after i.t. injection of epinephrine (Post, unpublished). We found a dose-dependent flow inhibition in the range of 0.025-0.5 ng.
346
that preferentially surround sympathoadrenai neurons (Gaumann et al. 1989). Furthermore, SST coexists within neurons that also produce adrenergic transmitter substances (norepinephrine) and other transmitters (GABA: y-aminobutyric acid) (Elde et al. 1985). Direct spinal norepinephrine release or enhancement of norepinephrine release by GABA inhibition may thus contribute to the spinal ischemia. SST may also decrease the release of substance P (Gammon et al. 1990) and thus disrupt the vascular tone maintained by this and several other tachykinins (D’OrlCans-Juste et al. 1985). In our previous study, the it. SST ‘neurotoxic’ doses (30 pg) resulted in pronounced vasoconstriction in rat spinal cord (Freedman et al. 1988). In the guinea pig, however, the Lt. dose of 120 pg decreased the spinal blood flow by less than 20%. The reason for this difference is unclear. Considering the body weight and CSF volume in rat and guinea pig, the concentration of i.t. administered SST at the dorsal horn area should have been roughly equal in both species. However, as pointed out by Vaught and Scott (1987), caution must be taken in extrapolating the toxic actions of neurokinins in the rat to other animal species. This was also recently demonstrated for substance P antagonists (Post and Folkers 1985; Post et al. 1987). In man SST doses up to 3000 pg did not result in any noticeable neurological lesions (Desborough et al. 1989). However, Gaumann et al. 0989) have demonstrated that an i.t. SST dose of 50 pg is neurotoxic in the mouse and 8000 pg is neurotoxic in the cat. We agree with Yaksh and Gaumann (1990) that there is significant homology across species in the biochemists of the mammalian spinal cord and that the random administration of experimental drugs to humans (by any route) is unjustifiable. However, the commonly used local anesthetics, if given intrathecally in concentrations approaching their solubility limit, cause irreversible neurologic injury and histological changes in rabbits (Ready et al. 1985). Furthermore, the ‘neurotoxic’ injury in the rat sciatic nerve was described following the extrafascicular application of clinically used concentrations of local anesthetics (Myers et al. 1986). Continuous infusion of epidural morphine can cause demyelination of posterior spinal columns in cancer patients (Coombs et al. 1985) but, interestingly, not in guinea pigs (Edwards et al. 1986). Neither of these findings diminished the further clinicat use of local anesthetics or intrathecal/epiduraI administration of morphine. We conclude that, although gradual, i.t. application of 120 pg SST resulted in a moderate (< 20%) decrease in spinal cord blood flow in the guinea pigs, there were no motor or histopathological changes. The implication of this study may be in the choice of a relevant dose to be compared across animal species and in the necessity to use several, alternative species
in order to evaluate the antinociceptive and neurodegenerative properties of the it. administered peptides.
Acknowledgements
Ing-Marie Dimgren and Sonja Malmqvist are gratefully acknowledged for typing the manuscript.
References Coombs, D.W.. Fratkin, J.D., Meir, F.A.. Nierenberg. D.W. and Saunders, R.L., Neuropathologic lesions and CSF morphine concentrations during chronic continuous intraspinal morphine infusion: a clinical and post-mortem study, Pain, 22 (19851 337-351. Desborough, J.P.. Edlin. S.A., Burrin, J.M., Bloom, S.R., Morgan, M. and Hall. G.M., Hormonal and metabolic responses to cholecystectomy: comparison of extradural somatostatin and diamorphine, Br. J. Anaesth.. 63 (1989) 50X-515. D’Orleans-Juste, P.. Dion, S.. Mizrahi, J. and Regoli. D., Effects of peptides and non-peptides on isolated arterial smooth muscles: role of endothelium, Eur. J. Pharmacoi., 114 (1985) 9-21. Edwards, W.T., DeGirolami, U., Burney, R.G., Cappadona, D. and Brickley, R., Histopathologic changes in the epidural space of the guinea pig during long-term morphine infusion, Reg. Anesth.. 11 (1986) 14-19. Elde. R., Johansson. 0. and Hokfelt. T., Immunocytochemical studies of somatostatin neurons in brain, Adv. Exp. Med. Biol., 188 f1985)167-181. Freedman, J., Post, C.. Kghrstriim, J.. Ghlen, A., Mollenholt, P., Owman. C., Alari. L. and Hokfelt. T., Vasoconstrictor effects in spinal cord of the substance P antagonist (p-Arg’, o-Trp’,‘, Leu”)-substance P (Spantide) and somatostatin and interaction with thyrotropin releasing hormone, Neuroscience, 27 (19881 267-278. Gammon, CM.. Lyons. S.A. and Morell, P., Modulation by neuropeptides of bradykinin-stimulated second messenger release in dorsal root ganglion neurons, Brain Res., 518 (1990) 159-165. Gaumann, D.M., Yaksh, T.L., Post, C., Wilcox. G.L. and Rodrigues, M.. lntrathecal somatostatin in cat and mouse. Studies on pain, motor behavior, and histopathnlogy, Anesth. Analg.. 68 (1989) 623-632. Cordh. T., Feuk. U. and Norlin, K., Effects of epidural clonidine on spinal cord blood flow and regional and central hemodynamics in pigs. Anesth. Analg.. 65 (1986) 1312-1318. Gordin. A.. Eriksson, L.. Blom. A.K., Taskinen, M.R. and Fyhrquist. F.. Lack of behavioural effects following intraventricular infusion of somatostatin in the conscious goat, Pharmacol. Biochem. Behav.. 9 119781255-257. Hiikfelt. T.. Elde. R., Johansson, 0.. Luft, R., Nilsson, G. and Arimura, A.. Immunohistochemical evidence for separate populations of somatostatin-containing primary neurons in the rat. Neuroscience, l(1976) 131-136. Holets, V. and Elde. R.. The differential distribution and relationship of serotoninergic and peptidergic fibers to sympathoadrenal neurons in the intermediolateral cell column of the rat: a combined retrograde axonal transport and immunofluorescence study, Neuroscience, 7 (1982) 1155-l 174. Long, J.B.. Spinal subarachnoid injection of somatostatin causes neurological deficits and neuronal injury in rats, Eur. J. Pharmaccl.. 149 (I988) 287-296. Molienholt. P., Post, C.. Rawal, N., Freedman J., H&felt. T. and Paulsson, I., Antinociceptive and ‘neurotoxic’ actions of somato-
347 statin in rat spinal cord after intrathecal administration, Pain, 32 (1988) 95-105. Mollenholt, P., Post, C., Paulsson, I. and Rawal, N., Intrathecal and epidural somatostatin in rats: can antinociception, motor effects and neurotoxicity be separated?, Pain, 43 (1990) 363-370. Myers, R.R., Kalichman, M.W., Reisner, L.S. and Powell, H.C., Neurotoxicity of local anesthetics: altered perineurial permeability, edema, and nerve fiber injury, Anesthesiology, 64 (1986) 29-35. Neff, M., Metry, J.M., Frick, P.. Anliker, M. and Knoblauch, M.. The microvasculature of the small-intestinal mucosa of the rat: quantification of hemodynamic effects of topically applied cimetidine, ranitidine, somatostatin, and vasopressin, Stand. J. Gastroenterol., 20 (Suppl. 112) (1985) 6-11. Post, C. and Folkers, K., Behavioural and antinociceptive effects of intrathecally injected substance P analogues in mice, Eur. J. Pharmacol., 113 (1985) 335-343. Post, C., Freedman, J., Paulsson, I. and Hokfelt, T., Antinociceptive effects in mice after intrathecal injection of a substance P receptor antagonist, Spantide: lack of neurotoxic action, Reg. Peptides, 18 (1987) 243-252. Ready, L.B., Plumer, M.H.. Haschke, R.H., Austin, E. and Sumi, M., Neurotoxicity of intrathecal local anesthetics in rabbits, Anesthesiology, 63 (1985) 364-370.
Spampinato, S., Romualdi, P., Candeletti. S.. Cavicchini, E. and Ferri, S., Distinguishable effects of intrathecal dynorphins, somatostatin, neurotensin and s-calcitonin on nociception and motor function in the rat, Pain, 35 (1988) 95-104. Tenland, T., On Laser-Doppler Flowmetry. Methods and Microvascular Applications, Doctoral dissertation, Linkoping University Medical Dissertations, Linkiiping, 1982. No. 136. Tiirnebrandt, K., Nobin, A. and Owman, C., Contractile and dilatory action of neuropeptides on isolated human mesenteric blood vessels, Peptides. 8 (1987) 251-256. Tydin, G., Samnegird, H., Thulin. L., Muhrbeck, 0. and Efendic. S., Circulatory effects of somatostatin in anesthetized man, Acta Chir. Stand., 145 (1979) 443-446. Vaught, J.L. and Scott, R., Species differences in the behavioral toxicity produced by intrathecal substance P antagonists: Relationship to analgesia, Life Sci., 40 (1987) 175-181. Vora, J.P., Owens, D.R., Ryder, R., Atiea, J., Luzio, S. and Hayes, T.M., Effects of somatostatin on renal function, Br. Med. J., 292 (1986) 1701-1702. Yaksh, T.L. and Gaumann, D.M., Spinal somatostatin studies in animals (in response), Anesth. Analg., 70 (1990) 227-229. Yaksh, T.L. and Rudy, T.A., Chronic catheterization of spinal subarachnoidal space, Physiol. Behav.. 17 (1976) 1031-1036.
343 Publishers
B.V. All rights reserved
0304-3959/92/$05.00
PAIN 02175
Intrathecal somatostatin in the guinea pig: effects on spinal cord blood flow, histopathology and motor function Peter Mollenholt,
Claes Post, Ivar Paulsson
and Narinder
Rawal
Department of Anesthesiology and Intenske Care, drebro Medical Center Hospital, &ebro (Sweden) and Astra Alab AB, Research and Development Laboratories and Astra Toxicological Laboratories, Siidertiilje (Sweden) (Received
11 February
1992, revision
received
7 July 1992, accepted
20 July 1992)
In the present investigation, the vasoconstrictive, motor and neurodegenerative effects of intratheSummary cal somatostatin (SST) were assessed in guinea pigs implanted with lumbar intrathecal catheters. Five consecutive dose increments of SST (5, 10, 15, 30 and 60 pg) to a total of 120 pg during the period of 16 f 3 min, resulted in a moderate (< 20%), gradual decrease of the spinal blood flow monitored with the laser-doppler method. A subsequent injection of clonidine (50 pg) or norepinephrine (10 pg) resulted in a more pronounced decrease of spinal blood flow (35% and 79%, respectively). Three consecutive, daily intrathecal injections of 30 or 60 pg SST did not cause any loss of weight support or paralysis of the hind limbs. There were no histopathological changes in the white or gray matter of the thoracic and lumbar sections of the spinal cords. It is concluded that SST, in the doses studied, is not neurodegenerative in guinea pigs. These findings are in contrast to those previously seen in rats. The implication of this study may be the necessity to use several alternate animal species in order to evaluate the antinociceptive and neurodegenerative properties of the peptides administered by the intrathecal route and the choice of dose to be compared across species. Key words: Peptides; Somatostatin;
Spinal cord; Vasoconstriction;
Introduction
The presence of somatostatin (SST) in primary sensory neurons (C afferents) projecting to the neuroaxis supports its role as a putative neurotransmitter or neuromodulator for pain (Hokfelt et al. 1976). Experimental studies in rats have shown that intrathecal (i.t.) administration of SST is associated with antinociception and motor blockade (Spampinato et al. 1988; Mollenholt et al. 1990) and also with neurodegeneration (Mollenholt et al. 1988). It has been suggested (Freedman et al. 1988; Long 1988) that the ‘neurotoxic’ effects of i.t. SST in rat spinal cord may partly be due to ischemia resulting from vasoconstriction of spinal vessels. However, the susceptibility to SST-induced ischemia appears to be species dependent (Gordin et
Neurotoxicity;
Intrathecal;
(Guinea pig>
al. 1978). This has also been demonstrated for substance P receptor antagonists (Post and Folkers 1985; Post et al. 1987). The present study addresses the issue of whether circulatory events and neuronal lesions following i.t. SST are as pronounced in guinea pigs as in rats.
Materials
and methods
Preparation of drug solution Somatostatin (Ferring AB, Sweden), clonidine (Boehringer Ingelheim, Germany) and norepinephrine bitartrate (Sigma, USA) were dissolved in 0.9% saline. All solutions were prepared immediately prior to injection.
Experiment 1 Correspondence to: Peter and Intensive Care, brebro Grebro, Sweden.
Mollenholt, Dept. of Anesthesiology Medical Center Hospital, S-701 85
Animals and animal preparation. Male guinea pigs (weight range: 250-300 gl of the Dunkin-Hartley strain (Sahlins, Sweden) were used. Experimental work was reviewed by an Institutional Animal Care and Use Committee. The animals were anesthetized with 5%
344 enflurane (Efrane’“) in O2 +N,O (1 I/min + 1 I/min of each) for approximately 1.5 min. When fully anesthetized, the enflurane concentration was reduced to t.S?. with the same concentration ratio of 0, and N@. The laser-doppler experiments were carried out in animals with their spinal cords exposed. For this. they were placed in a David Kopf stereotaxic frame (Tajunga. CA; model 900). which was comhined with a model 980 spinal unit. The spinal column was dissected free. and the cord was exposed by bilateral laminectomy over the lumbar L,-L, segments. Care was taken not to damage the dura mater. The spinal column was fixed in the spinal unit in order to minimize respiratory movements of the cord. A PE-IO catheter, previously stretched to double its length to make it more flexible and non-traumatizing, was then inserted through a slit in the dura mater under a dissection microscope. ’ l~u.~~r-~iopp~ermethod. Blood Bow in the spinal cord was monitored with the laser-doppier technique using a Perimed model Pf 2 laser-doppler instrument (Perimed. Sweden) (Tentand 1982). The laser-doppler probe (model Pf 103) was positioned by a micromanipulator 0.5-l mm above the darsolateral surface of the spinal cord, close to the tip of the catheter. The operated field and probe were covered with a 2.5? agar gel (in 0.9Oh saline) to reduce water evaporation. Recordings were performed with a time-constant of 3 set and the gain was set to t or 3. Readings were recorded on a strip-chart recorder, and the results calculated as blood flow relative to the pre-injection value. Each animal served as its own control. When a steady signal was achieved, which usually occurred within 15-X min, a control volume of 25 @I of 0.9% saline was injected intrathecatly. This resulted in a minor and transient decrease in spinal blood flow. After recovery of blood flow to baseline, 5 consecutive i.t. injections of SST were administered. The doses were 5, IO. 15. 30 and 60 pg (2 wg/plI). This was followed by injection of saline (5-22.5 ~1) to flush the catheter. After the final SST dose, 2 animals were injected Lt. with clonidine (50 gg) and 5 animals with norepinephrine (10 pg). These injections were given as positive controls with known vasoconstrictive substances. Both drugs were dissolved in IO ~1 of saline followed by 15 ~1 of saline to flush the catheter. The total volume of drug and saline flush was kept constant at 25 ~1 except for the 60 Pg SST dose where it was 45 ~1. Injections were made only after a steady blood flow signal resulting from the previous dose was obtained.
100
3 OEI
-
80 -
-3
8-6 02
iia 35
c
2
0 \o
60 -
40 -
t NaCl
I
1
I
1
5
10
15
30
1
60
SSTir-ls)
Fig. I. Log dose-response curve of laser-doppler measurements of spinal cord blood flow. The results are given as means + S.E.M., from 7 animals, in per cent of the pre-drug values. The treatment and saline control values were compared by I-way ANOVA 1* * P < 0.01).
at a nominal thickness of 5 km and stained with Nissl’s stain. Microscopic criteria for intactness of the spinal cord were, inter alia, absence of neuronal necrosis, with pyknosis and hyperchromasia of the nucleus; early degenerative changes including chromatolysis; signs of gliosis and signs of vascular damage. Neither the experimenter nor the pathologist were aware of the treatment until the entire experiment had been completed.
Statistical methods Statistical analysis was performed by analysis (ANOVA). Values are presented as means + S.E.M.
of
variance
Results Experiment 2 An&u& and animal prep~rut~on. Male Dunkin-Hartley guinea pigs (weight range: 2.50-300 gf were anesthetized with 60 mg/kg pentobarbital injected intraperitoneally (i.p.) fMebumalB’, ACO. Sweden). The chronic PE-IO catheter (previously stretched to double its length) was then inserted it. with the tip placed in the lumbar rrgion (Yaksh and Rudy 1976). All the animals showed normal motor behavior after the recovery period. Drug treatment. Twelve guinea pigs were randomly assigned to 1 of 2 treatment groups (6 animals/group). One group receiving 30 p.g and the other 60 pg SST/anima~ intrathecaliy f2 pg/pul) followed in both groups by S-10 ~1 saline/animal to flush the catheter. These injections were administered daily for 3 consecutive days. One control animal received 25 ~1 it. saline daily instead of SST. Mofor effecfs. Motor effect was defined as loss of weight support or flaccid paralysis of the hind limbs. The time of onset and duration of motor blockade were noted every 5 min for a period of 60 min. Histological eduation. All 16 animals were killed on the 4th day try an i.p. overdose of pentobarhitat. The spinal cords were excised after bilateral laminectomy where care was taken to avoid producing traumatic lesions and immediately immersed into a neutral, buffered 3.7% formaldehyde solution. Specimens from the lumbar and lower thoracic region were embedded in Paraplast (Lancer, USA) and cut
Expe~rn~~t 1 Laser-doppler
experiments. The Lt. application of saline caused a small, brief drop in spinal blood flow. The subsequent i.t. SST dose escalation (5, 10, 15, 30 and 60 pg) resulted in a gradual decrease in spinal blood flow which reached significance (P < 0.01) at the total dose of 120 @g/animal. At this dose, the spinal blood flow decreased to 81 rt 11% of its pre-injection value (Fig. 1). A subsequent i.t. application of 50 pg clonidine in 2 animals reduced their spinal blood flow to 65 + 11% (Fig. 2) and a subsequent it. application of 10 icl.gnorepinephrine in another 5 animals, resulted in a reduction of the spinal blood flow to 21 3r:15% (Fig. 3). The steady, lowest blood flow signal corresponding to the maximal decrease in flow level, was obtained within 2-3 min following each dose. Accordingly, the average period of measurement was 16 I!I3 min/animal.
34.5
20 1
OJt
NaCl
1
I
I
5
10
15
t
,
I
60 Clonidine (50 w9)
30
SST k4
Fig. 2. Laser-doppler measurements of spinal cord blood flow. The effect of 120 pg SST and 50 pg clonidine. The results are given as means+S.E.M. Data from 2 animals in per cent of the pre-drug values. The clonidine treatment values were compared with the SST treatment values and saline control values by l-way ANOVA (** P < 0.01).
Experiment 2 Motor effects. No pain related behavior (e.g., caudally directed bites and scratches, vocalization) or motor block was noted in control animals or in animals receiving 30 and 60 pg SST. Histopathological findings. No signs of necrosis or degenerative changes were seen in the neurons at any level of the spinal cords investigated. Discussion The present experiments show the vasoconstrictive effects of i.t. SST on the spinal blood vessels of the
NaCl
I
I
I
I
5
10
15
30
SST(w)
I
guinea pig. The decrease in blood flow was moderate (20%) but significant. In the dose range investigated, SST had no motor blocking or neurodegenerative effects in this species. In our previous studies in rats, we established the threshold i.t. SST dose for antinociception and ‘neurotoxicity’ associated with permanent motor impairment, neuronal necrosis and loss of an immunohistochemical marker for motoneurons in the spinal cord (Mollenholt et al. 1988). Furthermore, it was possible to separate the antinociceptive and motor blocking actions of SST following both i.t. and epidural administration (Mollenholt et al. 1990). However, the margin of safety (antinociceptive vs. ‘toxic’ effects) was quite narrow. In the present investigation, no behavior indicative of sensation or motor block was noted following 3 consecutive, daily i.t. injections of SST. Although the doses were of the magnitude that produce irreversible paraparesis and extensive neuronal necrosis in the rat (Long 19881, no neurodegeneration was noted in any of the guinea pigs, In the present study, histopathological evaluation was made 24 h after the 3rd i.t. injection of SST. It may be argued that this 24-h period was too short for the ‘neurotoxic’ lesions to develop fully. However, our earlier studies in rats showed histopathological changes within 72 h after a single drug exposure (Mollenholt et al. 1988). In order to ascertain that the spinal cord blood flow in the guinea pig responds adequately to locally applied vasoconstrictive drugs, the SST injection was followed by clonidine (Fig. 2) or norepinephrine (Fig. 3) given as positive controls. A rapid reduction in the local blood flow followed these injections indicative that the spinal microcirculation responded as described previously (Gordh et al. 1986) ‘. Although the SST-dependent vasoconstriction has been described previously by many investigators (Neff et al. 1985; Tornebrandt et al. 1987; Tydtn et al. 1979; Vora et al. 19861, the mechanism of this action is speculative. The relatively rapid onset of vasoconstriction ( < 1 min) (Freedman et al. 1988) may suggest that SST acts directly on the vascular smooth muscle (Tyden et al. 1979). Considering the differential distribution of peptidergic fibres (SST, substance P, Met-enkephalin, oxytocin) to intermediolateral (IML) cell column of the spinal cord (Holets and Elde 19891, it is possible that the dose-dependent enhancement of IML cell column function may evoke significant hemodynamic effects due to a sympathetic stimulation via SST afferent fibers
t
60 Norepinephrine (10 cls)
Fig. 3. Laser-doppler measurements of spinal cord blood flow. The effects of 120 pg SST and 10 pg norepinepbrine. Data from 5 animals. The results are given as means f S.E.M. The norepinephine treatment values were compared with SST treatment values and saline control values by l-way ANOVA (* * P < 0.01).
’ In preliminary experiments, we have also measured the spinal cord blood flow effects in rat with the laser-doppler method after i.t. injection of epinephrine (Post, unpublished). We found a dose-dependent flow inhibition in the range of 0.025-0.5 ng.
346
that preferentially surround sympathoadrenai neurons (Gaumann et al. 1989). Furthermore, SST coexists within neurons that also produce adrenergic transmitter substances (norepinephrine) and other transmitters (GABA: y-aminobutyric acid) (Elde et al. 1985). Direct spinal norepinephrine release or enhancement of norepinephrine release by GABA inhibition may thus contribute to the spinal ischemia. SST may also decrease the release of substance P (Gammon et al. 1990) and thus disrupt the vascular tone maintained by this and several other tachykinins (D’OrlCans-Juste et al. 1985). In our previous study, the it. SST ‘neurotoxic’ doses (30 pg) resulted in pronounced vasoconstriction in rat spinal cord (Freedman et al. 1988). In the guinea pig, however, the Lt. dose of 120 pg decreased the spinal blood flow by less than 20%. The reason for this difference is unclear. Considering the body weight and CSF volume in rat and guinea pig, the concentration of i.t. administered SST at the dorsal horn area should have been roughly equal in both species. However, as pointed out by Vaught and Scott (1987), caution must be taken in extrapolating the toxic actions of neurokinins in the rat to other animal species. This was also recently demonstrated for substance P antagonists (Post and Folkers 1985; Post et al. 1987). In man SST doses up to 3000 pg did not result in any noticeable neurological lesions (Desborough et al. 1989). However, Gaumann et al. 0989) have demonstrated that an i.t. SST dose of 50 pg is neurotoxic in the mouse and 8000 pg is neurotoxic in the cat. We agree with Yaksh and Gaumann (1990) that there is significant homology across species in the biochemists of the mammalian spinal cord and that the random administration of experimental drugs to humans (by any route) is unjustifiable. However, the commonly used local anesthetics, if given intrathecally in concentrations approaching their solubility limit, cause irreversible neurologic injury and histological changes in rabbits (Ready et al. 1985). Furthermore, the ‘neurotoxic’ injury in the rat sciatic nerve was described following the extrafascicular application of clinically used concentrations of local anesthetics (Myers et al. 1986). Continuous infusion of epidural morphine can cause demyelination of posterior spinal columns in cancer patients (Coombs et al. 1985) but, interestingly, not in guinea pigs (Edwards et al. 1986). Neither of these findings diminished the further clinicat use of local anesthetics or intrathecal/epiduraI administration of morphine. We conclude that, although gradual, i.t. application of 120 pg SST resulted in a moderate (< 20%) decrease in spinal cord blood flow in the guinea pigs, there were no motor or histopathological changes. The implication of this study may be in the choice of a relevant dose to be compared across animal species and in the necessity to use several, alternative species
in order to evaluate the antinociceptive and neurodegenerative properties of the it. administered peptides.
Acknowledgements
Ing-Marie Dimgren and Sonja Malmqvist are gratefully acknowledged for typing the manuscript.
References Coombs, D.W.. Fratkin, J.D., Meir, F.A.. Nierenberg. D.W. and Saunders, R.L., Neuropathologic lesions and CSF morphine concentrations during chronic continuous intraspinal morphine infusion: a clinical and post-mortem study, Pain, 22 (19851 337-351. Desborough, J.P.. Edlin. S.A., Burrin, J.M., Bloom, S.R., Morgan, M. and Hall. G.M., Hormonal and metabolic responses to cholecystectomy: comparison of extradural somatostatin and diamorphine, Br. J. Anaesth.. 63 (1989) 50X-515. D’Orleans-Juste, P.. Dion, S.. Mizrahi, J. and Regoli. D., Effects of peptides and non-peptides on isolated arterial smooth muscles: role of endothelium, Eur. J. Pharmacoi., 114 (1985) 9-21. Edwards, W.T., DeGirolami, U., Burney, R.G., Cappadona, D. and Brickley, R., Histopathologic changes in the epidural space of the guinea pig during long-term morphine infusion, Reg. Anesth.. 11 (1986) 14-19. Elde. R., Johansson. 0. and Hokfelt. T., Immunocytochemical studies of somatostatin neurons in brain, Adv. Exp. Med. Biol., 188 f1985)167-181. Freedman, J., Post, C.. Kghrstriim, J.. Ghlen, A., Mollenholt, P., Owman. C., Alari. L. and Hokfelt. T., Vasoconstrictor effects in spinal cord of the substance P antagonist (p-Arg’, o-Trp’,‘, Leu”)-substance P (Spantide) and somatostatin and interaction with thyrotropin releasing hormone, Neuroscience, 27 (19881 267-278. Gammon, CM.. Lyons. S.A. and Morell, P., Modulation by neuropeptides of bradykinin-stimulated second messenger release in dorsal root ganglion neurons, Brain Res., 518 (1990) 159-165. Gaumann, D.M., Yaksh, T.L., Post, C., Wilcox. G.L. and Rodrigues, M.. lntrathecal somatostatin in cat and mouse. Studies on pain, motor behavior, and histopathnlogy, Anesth. Analg.. 68 (1989) 623-632. Cordh. T., Feuk. U. and Norlin, K., Effects of epidural clonidine on spinal cord blood flow and regional and central hemodynamics in pigs. Anesth. Analg.. 65 (1986) 1312-1318. Gordin. A.. Eriksson, L.. Blom. A.K., Taskinen, M.R. and Fyhrquist. F.. Lack of behavioural effects following intraventricular infusion of somatostatin in the conscious goat, Pharmacol. Biochem. Behav.. 9 119781255-257. Hiikfelt. T.. Elde. R., Johansson, 0.. Luft, R., Nilsson, G. and Arimura, A.. Immunohistochemical evidence for separate populations of somatostatin-containing primary neurons in the rat. Neuroscience, l(1976) 131-136. Holets, V. and Elde. R.. The differential distribution and relationship of serotoninergic and peptidergic fibers to sympathoadrenal neurons in the intermediolateral cell column of the rat: a combined retrograde axonal transport and immunofluorescence study, Neuroscience, 7 (1982) 1155-l 174. Long, J.B.. Spinal subarachnoid injection of somatostatin causes neurological deficits and neuronal injury in rats, Eur. J. Pharmaccl.. 149 (I988) 287-296. Molienholt. P., Post, C.. Rawal, N., Freedman J., H&felt. T. and Paulsson, I., Antinociceptive and ‘neurotoxic’ actions of somato-
347 statin in rat spinal cord after intrathecal administration, Pain, 32 (1988) 95-105. Mollenholt, P., Post, C., Paulsson, I. and Rawal, N., Intrathecal and epidural somatostatin in rats: can antinociception, motor effects and neurotoxicity be separated?, Pain, 43 (1990) 363-370. Myers, R.R., Kalichman, M.W., Reisner, L.S. and Powell, H.C., Neurotoxicity of local anesthetics: altered perineurial permeability, edema, and nerve fiber injury, Anesthesiology, 64 (1986) 29-35. Neff, M., Metry, J.M., Frick, P.. Anliker, M. and Knoblauch, M.. The microvasculature of the small-intestinal mucosa of the rat: quantification of hemodynamic effects of topically applied cimetidine, ranitidine, somatostatin, and vasopressin, Stand. J. Gastroenterol., 20 (Suppl. 112) (1985) 6-11. Post, C. and Folkers, K., Behavioural and antinociceptive effects of intrathecally injected substance P analogues in mice, Eur. J. Pharmacol., 113 (1985) 335-343. Post, C., Freedman, J., Paulsson, I. and Hokfelt, T., Antinociceptive effects in mice after intrathecal injection of a substance P receptor antagonist, Spantide: lack of neurotoxic action, Reg. Peptides, 18 (1987) 243-252. Ready, L.B., Plumer, M.H.. Haschke, R.H., Austin, E. and Sumi, M., Neurotoxicity of intrathecal local anesthetics in rabbits, Anesthesiology, 63 (1985) 364-370.
Spampinato, S., Romualdi, P., Candeletti. S.. Cavicchini, E. and Ferri, S., Distinguishable effects of intrathecal dynorphins, somatostatin, neurotensin and s-calcitonin on nociception and motor function in the rat, Pain, 35 (1988) 95-104. Tenland, T., On Laser-Doppler Flowmetry. Methods and Microvascular Applications, Doctoral dissertation, Linkoping University Medical Dissertations, Linkiiping, 1982. No. 136. Tiirnebrandt, K., Nobin, A. and Owman, C., Contractile and dilatory action of neuropeptides on isolated human mesenteric blood vessels, Peptides. 8 (1987) 251-256. Tydin, G., Samnegird, H., Thulin. L., Muhrbeck, 0. and Efendic. S., Circulatory effects of somatostatin in anesthetized man, Acta Chir. Stand., 145 (1979) 443-446. Vaught, J.L. and Scott, R., Species differences in the behavioral toxicity produced by intrathecal substance P antagonists: Relationship to analgesia, Life Sci., 40 (1987) 175-181. Vora, J.P., Owens, D.R., Ryder, R., Atiea, J., Luzio, S. and Hayes, T.M., Effects of somatostatin on renal function, Br. Med. J., 292 (1986) 1701-1702. Yaksh, T.L. and Gaumann, D.M., Spinal somatostatin studies in animals (in response), Anesth. Analg., 70 (1990) 227-229. Yaksh, T.L. and Rudy, T.A., Chronic catheterization of spinal subarachnoidal space, Physiol. Behav.. 17 (1976) 1031-1036.
