- Email: [email protected]
Endothelin-1-induced nociception
Life Sciences, Vol. 49, pp. PL-61 -PL-65 Printed in the U.S.A.
Pergamon Press
PHARMACOLOGY LETTERS Accelerated Communication
ENDOTHELIN-I-INDUCED NOCICEPTION
Robert B. Raffa, James J. Schupsky, Rebecca P. Martinez and Henry I. Jacoby Drug Discovery Research, The R.W. Johnson Pharmaceutical Research Institute, Spring House, PA 19477-0776, U.S.A. (Submitted July 9, 1991; accepted July I0, 1991; received in final form July 12, 1991)
Abstract. Intracerebroventricular (i.c.v.) or intrathecal (i.t.) administration of morphine to mice antagonized the abdominal constriction induced by an i.p. injection of endothelin-1 (ET-1; 0.1 mg/kg). The EDS0 values (95% confidence intervals) were 39.3 (16.5-80.2) ng and 1.5 (0.8-4.9) ng, respectively. The antagonism of ET-l-induced abdominal constriction by morphine was blocked by naloxone (1.0 mg/kg, s.c.) or by 24 h pretreatment with ~-funaltrexamine ([~-FNA; 8.84 I~g, i.c.v.). These results demonstrate for the first time that the stimulus resulting from an i.p. injection of ET-1 is transmitted via ascending (pain) pathways that are subject to attenuation by opioid (~t) receptor activation. Hence, ET-l-induced abdominal constriction is a new pain model which, given the other pharmacology of ET-1, might represent a unique model with potential specific utility for anginal or other visceral pain. Inlroduclion
We recently reported that the intraperitoneal injection of either endothelin-1 (ET-1), ET-2, ET-3 or big-ET- 1[ 1-38] in mice produces an abdominal constriction response (1). The effect of big-ET1[ 1-38] was blocked by pretreatment with the enzyme inhibitor phosphoramidon, indicating that the big-ET-l[1-38] must first be enzymatically cleaved to ET-1 in order to elicit the response. ET1-induced abdominal constriction is not blocked by atropine (10 mg/kg, s.c.) or the K+-channel opener cromakalim (0.043 mg/kg, s.c.), suggesting that the stimulus is not secondary to the release of acetylcholine or due to a direct vasoconstrictor activity. The implication of ET- 1 in the pathogenesis of coronary vasospasm (2), a possible causative factor in variant angina, raised the possibility that endothelin-induced abdominal constriction was a response to a nociceptive stimulus and, therefore, that this test represents a novel pain model. In the present study, we addressed the issue of an ET-l-induced nociceptive stimulus by examining whether the afferent signal in ET-l-induced abdominal constriction is transmitted via central pain pathways, i.e., can be blocked by spinal (intrathecal) or supraspinal (intracerebroventricular) administration of morphine. We report that ET-1-induced abdominal constriction in mice was potently blocked by either spinal (i.t.) or supraspinal (i.c.v.) administration of morphine (EDS0 = 1.5 ng and 39.3 ng, respectively). Further, the effect of morphine was naloxone-reversible and was significantly attenuated in animals pretreated with the ~t-opioid receptor alkylating agent I~-funaltrexamine ( [~FNA).
0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc
PL-62
ET-l-lnduced Nociception
Vol. 49, No. 11, 1991
Methods
Male, virus-free albino CD-1 ® mice, 18-25 g (Charles River Laboratories; Kingston Facility, NY) were group-housed 8-10 per plastic cage and maintained in a climate controlled room on a 12 h light/dark cycle (lights on at 0715 h). Food and water were available ad libitum up to the time of the test. Each animal was used once. All testing was performed in accordance with the recommendations and policies of the International Association For The Study of Pain (IASP) and National Institutes of Health (NIH) and Johnson & Johnson Guidelines for the care and use of laboratory animals. The procedure was similar to that described by Collier et al. (3) with minor modifications. Each mouse received an i.p. injection (10 ml/kg) of 0.1 mg/kg ET-1 (Peptide Institute, Inc.; Osaka, Japan) and was then placed into a large glass jar (14 cm diameter; 3 mice per jar) and observed for 10 rain for the occurrence of an abdominal constriction response (defined as a wave of constriction and elongation passing caudally along the abdominal wall, accompanied by a twisting of the trunk and followed by extension of the limbs). The data are reported according to the formula: Inhibition of response (% analgesia) = 100 x (number of non-responders)/(number in group). The mice received i.c.v, or i.t. injections (5 lal) of graded doses of morphine sulfate or vehicle (double-distilled water) 15 min prior to the i.p. injection of ET-1. The EDS0 value (dose of morphine sulfate that reduced ET- 1-induced abdominal constriction to 50%) and the corresponding 95% confidence intervals were calculated using probit analysis, including a ~2 test for linearity. Some animals were injected with naloxone hydochloride (1.0 mg/kg, s.c.) 5 min prior to morphine. Other animals were treated 24 h prior with I~-FNA (8.84 txg, i.c.v.). Results
The results of the first set of experiments are shown in Fig. 1. Morphine produced dose-related antinociception by both the i.c.v, route and, more potently, by the i.t. route (Fig. 1; Table 1). Injection of vehicle alone had no effect (data not shown).
1007 90 to c
o D_ to
80"
ED50 = 0 . 0 0 2
70"
/
60 o
50"
c
40
o _Q cC
/
/
o
o
20"
/
10
ED50 = 0.04
30
0 0.0001
0.001
Morphine
0.01
O. 1
(/~g/5/J,L)
Fig. 1 Dose-related inhibition of ET-1-induced abdominal constriction by morphine administered by either the i.c.v (filled circles) or the i.t. (open circles) route.
Vol. 49, No. ii, 1991
ET-l-lnduced Nociception
PL-63
TABLE I
Inhibition of ET-l-induced abdominal constriction by spinal (i.t.) or supraspinal (i.c.v.) administration of morphine
ED50 (95% confidence interval)
Rou~
i.c.v.
39.3 (16.5-80.2) ng
i.t.
1.5 (0.8-4.9) ng
As shown in Fig. 2, pretreatment with naloxone (1.0 mg/kg, s.c.) significantly (P<0.05) reduced the antinociception produced by morphine (0.3 Ixg, i.c.v.) from 100% to 20%. Pretreatment with [~-FNA (8.84 ~tg, 24 h prior) significantly (P<0.05) reduced the antinociception produced by morphine (0.3 ~tg, i.c.v.) from 100% to 33.3%. Pretreatment with either naloxone or [~-FNA alone had no effect on ET-l-induced abdominal constriction (data not shown).
100 90 80 70
g C
60 50 40 30 20 10 0 MS
+Nx
+B-FNA
Fig. 2 The effect of 5 rain prior treatment with naloxone (1.0 mg/kg, s.c.; +Nx) or by 24 h pretreatment with [~-FNA (8.84 I~g, i.c.v.; +[~-FNA) on morphine (0.3 ~tg, i.c.v.; MS) antagonism of ET-1induced abdominal constriction. The asterisks signify a significant (P<0.05, Fisher's Exact test) attenuation of the morphine effect.
PL-64
ET-l-lnduced Nociception
Vol. 49, No. ii, 1991
Discussion
That the i.p. administration of ET-1, ET-2, ET-3 or big-ET-1[ 1-38] to mice elicits an abdominal constriction response was previously reported by our laboratory (1). This response does not appear to be mediated through cholinergic pathways, since it is not antagonized by atropine at doses that block acetylcholine(5.5 mg/kg)-induced abdominal constriction (1), nor to be caused by a direct vasoconstrictor activity, since it is not blocked by the K+-channel opener cromakalim (0.043 mg/kg, s.c.). In addition, the abdominal constriction elicited by ET-1 is apparently mediated through a different mechanism than overtly similar responses to acetylcholine (ACh) or phenyl-p-quinone (PQ), since compounds have been identified which block ACh- and PQinduced, but not ET-1-induced abdominal constriction and vice versa (manuscript in preparation). The abdominal constriction induced by ET-1 is of particular interest, since ET-1 has recently been implicated in the pathogenesis of coronary vasospasm (2), a possible cause of variant angina. Hence, the pain associated with angina (or other vascular or visceral pain) might be related to this action of ET-1 either secondary to ET-l-induced vasospasm and resultant hypoxia or to a direct nociceptive action of ET-1 or endogenous factor released by ET-1. The possibility that ET-1 produces a nociceptive stimulus was tested directly in the present study. The major finding of this study is that the i.p. administration of ET-1 to mice generates an afferent stimulus that is transmitted rostrally via ascending pathways that are blocked by morphine administered either spinally or supraspinally. The possibility that the effect of morphine could be due to some peripheral effect on smooth muscle, skeletal muscle or other peripheral action was avoided by administering the morphine directly into the brain or spinal cord as well as by 24 h pretreatment with 13-FNA. Hence, the afferent signal is transmitted via pathways that involve central (spinal and supraspinal) components associated with the transmission of pain. That the effect of morphine was mediated through opioid receptors was demonstrated by the reversal of this effect by the opioid antagonist naloxone. Further, the attenuation of morphineinduced antinociception by pretreatment with the selective opioid rt-receptor antagonist fI-FNA (4) confirmed this finding and, further, implicates a role for the ~ subtype of opioid receptor in the mediation of this action of morphine. The mechanism by which ET-1 elicits either the abdominal constriction or the nociceptive stimulus is not presently known. In rat tissue, displaceable [125I]endothelin binding is widely distributed in various tissues, including arteries, heart, kidney, lung, adrenal gland, intestine and brain (5). Putative endothelin receptors have also been localized autoradiographically in frozen sections of human brain using [125I]endothelin (6). The non-vascular pattern of the distribution of binding sites found in these studies suggests that endothelin(s) may function, in addition to their possible involvement in the regulation of vascular tone, as modulators of neuronal function. It appears unlikely that systemically injected ET crosses the blood-brain barrier (5). A recent report by Yoshizawa and coworkers (7) demonstrated that ET-l-like immunoreactivity is localized to dot- and fibre-like structures and neurons in the dorsal horn of porcine spinal cord and that the ventral root depolarization produced by ET-1 in newborn rat spinal cord is depressed by the substance P antagonist spantide ([D-Arg1,D-Trp7,9,Leu11]-substance P), suggesting that endothelins may release substance P. Thus, speculatively, the stimulus associated with ET-1induced abdominal constriction may involve the release of substance P. Taken together, the findings of the present study suggest that the i.p. injection of ET-1 (or other endothelins or endothelin precursors; see ref. 1) produce(s) a noxious stimulus that is transmitted centrally via pain pathways. This stimulus can be blocked by morphine at opioid (naloxonesensitive) receptors at either the spinal or supraspinal level. Additionally, the morphine-induced antinociception apparently involves the I1 subtype of opioid receptors, a conclusion that agrees with many other studies using different stimuli. These findings further suggest that ET-l-induced abdominal constriction might be a novel model with particular potential application to the study of anginal or, because of the wide distribution of endothelin receptors, other visceral pain.
Vol. 49, No. ii, 1991
ET-l-lnduced Nociception
PL-65
References
1. R.B. RAFFA and H.I. JACOBY, Life Sci./Pharmacol. Lett. 48 PL85-PL90 (1991). 2. H. KURIHARA, M. YOSHIZUMI, T. SUGIYAMA, K. YAMAOKI, R. NAGAI, F. TAKAKU, H. SATOH, J. INUI, M. YANAGISAWA, T. MASAKI and Y. YAZAKI, J. Cardiovasc. Pharmacol. 13(Suppl. 5) S132-S 137 (1989). 3. H.O.J. COLLIER, L.C. DINNEEN, C.A. JOHNSON and C. SCHNEIDER, Br. J. Pharmacol. Chemother. 32 295-310 (1968). 4. S.J. WARD, PORTOGHESE, P.S. and TAKEMORI, A.E., J. Pharmacol. Exp. Ther. 220 494-498 (1982). 5. C. KOSEKI, I. MASASHI, Y. HIRATA, M. YANAGISAWA and T. MASAKI, Am. J. Physiol. 256 R858-R866 (1989). 6. C.R. JONES, C.R. HILEY, J.T. PELTON and M. MOHR, Neurosci. Lett. 97 276-279 (1989). 7. T. YOSHIZAWA, S. KIMURA, I. KANAZAWA, Y. UCHIYAMA, M. YANAGISAWA and T. MASAKI, Neurosci. Lett. 102 179-184 (1989).
Pergamon Press
PHARMACOLOGY LETTERS Accelerated Communication
ENDOTHELIN-I-INDUCED NOCICEPTION
Robert B. Raffa, James J. Schupsky, Rebecca P. Martinez and Henry I. Jacoby Drug Discovery Research, The R.W. Johnson Pharmaceutical Research Institute, Spring House, PA 19477-0776, U.S.A. (Submitted July 9, 1991; accepted July I0, 1991; received in final form July 12, 1991)
Abstract. Intracerebroventricular (i.c.v.) or intrathecal (i.t.) administration of morphine to mice antagonized the abdominal constriction induced by an i.p. injection of endothelin-1 (ET-1; 0.1 mg/kg). The EDS0 values (95% confidence intervals) were 39.3 (16.5-80.2) ng and 1.5 (0.8-4.9) ng, respectively. The antagonism of ET-l-induced abdominal constriction by morphine was blocked by naloxone (1.0 mg/kg, s.c.) or by 24 h pretreatment with ~-funaltrexamine ([~-FNA; 8.84 I~g, i.c.v.). These results demonstrate for the first time that the stimulus resulting from an i.p. injection of ET-1 is transmitted via ascending (pain) pathways that are subject to attenuation by opioid (~t) receptor activation. Hence, ET-l-induced abdominal constriction is a new pain model which, given the other pharmacology of ET-1, might represent a unique model with potential specific utility for anginal or other visceral pain. Inlroduclion
We recently reported that the intraperitoneal injection of either endothelin-1 (ET-1), ET-2, ET-3 or big-ET- 1[ 1-38] in mice produces an abdominal constriction response (1). The effect of big-ET1[ 1-38] was blocked by pretreatment with the enzyme inhibitor phosphoramidon, indicating that the big-ET-l[1-38] must first be enzymatically cleaved to ET-1 in order to elicit the response. ET1-induced abdominal constriction is not blocked by atropine (10 mg/kg, s.c.) or the K+-channel opener cromakalim (0.043 mg/kg, s.c.), suggesting that the stimulus is not secondary to the release of acetylcholine or due to a direct vasoconstrictor activity. The implication of ET- 1 in the pathogenesis of coronary vasospasm (2), a possible causative factor in variant angina, raised the possibility that endothelin-induced abdominal constriction was a response to a nociceptive stimulus and, therefore, that this test represents a novel pain model. In the present study, we addressed the issue of an ET-l-induced nociceptive stimulus by examining whether the afferent signal in ET-l-induced abdominal constriction is transmitted via central pain pathways, i.e., can be blocked by spinal (intrathecal) or supraspinal (intracerebroventricular) administration of morphine. We report that ET-1-induced abdominal constriction in mice was potently blocked by either spinal (i.t.) or supraspinal (i.c.v.) administration of morphine (EDS0 = 1.5 ng and 39.3 ng, respectively). Further, the effect of morphine was naloxone-reversible and was significantly attenuated in animals pretreated with the ~t-opioid receptor alkylating agent I~-funaltrexamine ( [~FNA).
0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc
PL-62
ET-l-lnduced Nociception
Vol. 49, No. 11, 1991
Methods
Male, virus-free albino CD-1 ® mice, 18-25 g (Charles River Laboratories; Kingston Facility, NY) were group-housed 8-10 per plastic cage and maintained in a climate controlled room on a 12 h light/dark cycle (lights on at 0715 h). Food and water were available ad libitum up to the time of the test. Each animal was used once. All testing was performed in accordance with the recommendations and policies of the International Association For The Study of Pain (IASP) and National Institutes of Health (NIH) and Johnson & Johnson Guidelines for the care and use of laboratory animals. The procedure was similar to that described by Collier et al. (3) with minor modifications. Each mouse received an i.p. injection (10 ml/kg) of 0.1 mg/kg ET-1 (Peptide Institute, Inc.; Osaka, Japan) and was then placed into a large glass jar (14 cm diameter; 3 mice per jar) and observed for 10 rain for the occurrence of an abdominal constriction response (defined as a wave of constriction and elongation passing caudally along the abdominal wall, accompanied by a twisting of the trunk and followed by extension of the limbs). The data are reported according to the formula: Inhibition of response (% analgesia) = 100 x (number of non-responders)/(number in group). The mice received i.c.v, or i.t. injections (5 lal) of graded doses of morphine sulfate or vehicle (double-distilled water) 15 min prior to the i.p. injection of ET-1. The EDS0 value (dose of morphine sulfate that reduced ET- 1-induced abdominal constriction to 50%) and the corresponding 95% confidence intervals were calculated using probit analysis, including a ~2 test for linearity. Some animals were injected with naloxone hydochloride (1.0 mg/kg, s.c.) 5 min prior to morphine. Other animals were treated 24 h prior with I~-FNA (8.84 txg, i.c.v.). Results
The results of the first set of experiments are shown in Fig. 1. Morphine produced dose-related antinociception by both the i.c.v, route and, more potently, by the i.t. route (Fig. 1; Table 1). Injection of vehicle alone had no effect (data not shown).
1007 90 to c
o D_ to
80"
ED50 = 0 . 0 0 2
70"
/
60 o
50"
c
40
o _Q cC
/
/
o
o
20"
/
10
ED50 = 0.04
30
0 0.0001
0.001
Morphine
0.01
O. 1
(/~g/5/J,L)
Fig. 1 Dose-related inhibition of ET-1-induced abdominal constriction by morphine administered by either the i.c.v (filled circles) or the i.t. (open circles) route.
Vol. 49, No. ii, 1991
ET-l-lnduced Nociception
PL-63
TABLE I
Inhibition of ET-l-induced abdominal constriction by spinal (i.t.) or supraspinal (i.c.v.) administration of morphine
ED50 (95% confidence interval)
Rou~
i.c.v.
39.3 (16.5-80.2) ng
i.t.
1.5 (0.8-4.9) ng
As shown in Fig. 2, pretreatment with naloxone (1.0 mg/kg, s.c.) significantly (P<0.05) reduced the antinociception produced by morphine (0.3 Ixg, i.c.v.) from 100% to 20%. Pretreatment with [~-FNA (8.84 ~tg, 24 h prior) significantly (P<0.05) reduced the antinociception produced by morphine (0.3 ~tg, i.c.v.) from 100% to 33.3%. Pretreatment with either naloxone or [~-FNA alone had no effect on ET-l-induced abdominal constriction (data not shown).
100 90 80 70
g C
60 50 40 30 20 10 0 MS
+Nx
+B-FNA
Fig. 2 The effect of 5 rain prior treatment with naloxone (1.0 mg/kg, s.c.; +Nx) or by 24 h pretreatment with [~-FNA (8.84 I~g, i.c.v.; +[~-FNA) on morphine (0.3 ~tg, i.c.v.; MS) antagonism of ET-1induced abdominal constriction. The asterisks signify a significant (P<0.05, Fisher's Exact test) attenuation of the morphine effect.
PL-64
ET-l-lnduced Nociception
Vol. 49, No. ii, 1991
Discussion
That the i.p. administration of ET-1, ET-2, ET-3 or big-ET-1[ 1-38] to mice elicits an abdominal constriction response was previously reported by our laboratory (1). This response does not appear to be mediated through cholinergic pathways, since it is not antagonized by atropine at doses that block acetylcholine(5.5 mg/kg)-induced abdominal constriction (1), nor to be caused by a direct vasoconstrictor activity, since it is not blocked by the K+-channel opener cromakalim (0.043 mg/kg, s.c.). In addition, the abdominal constriction elicited by ET-1 is apparently mediated through a different mechanism than overtly similar responses to acetylcholine (ACh) or phenyl-p-quinone (PQ), since compounds have been identified which block ACh- and PQinduced, but not ET-1-induced abdominal constriction and vice versa (manuscript in preparation). The abdominal constriction induced by ET-1 is of particular interest, since ET-1 has recently been implicated in the pathogenesis of coronary vasospasm (2), a possible cause of variant angina. Hence, the pain associated with angina (or other vascular or visceral pain) might be related to this action of ET-1 either secondary to ET-l-induced vasospasm and resultant hypoxia or to a direct nociceptive action of ET-1 or endogenous factor released by ET-1. The possibility that ET-1 produces a nociceptive stimulus was tested directly in the present study. The major finding of this study is that the i.p. administration of ET-1 to mice generates an afferent stimulus that is transmitted rostrally via ascending pathways that are blocked by morphine administered either spinally or supraspinally. The possibility that the effect of morphine could be due to some peripheral effect on smooth muscle, skeletal muscle or other peripheral action was avoided by administering the morphine directly into the brain or spinal cord as well as by 24 h pretreatment with 13-FNA. Hence, the afferent signal is transmitted via pathways that involve central (spinal and supraspinal) components associated with the transmission of pain. That the effect of morphine was mediated through opioid receptors was demonstrated by the reversal of this effect by the opioid antagonist naloxone. Further, the attenuation of morphineinduced antinociception by pretreatment with the selective opioid rt-receptor antagonist fI-FNA (4) confirmed this finding and, further, implicates a role for the ~ subtype of opioid receptor in the mediation of this action of morphine. The mechanism by which ET-1 elicits either the abdominal constriction or the nociceptive stimulus is not presently known. In rat tissue, displaceable [125I]endothelin binding is widely distributed in various tissues, including arteries, heart, kidney, lung, adrenal gland, intestine and brain (5). Putative endothelin receptors have also been localized autoradiographically in frozen sections of human brain using [125I]endothelin (6). The non-vascular pattern of the distribution of binding sites found in these studies suggests that endothelin(s) may function, in addition to their possible involvement in the regulation of vascular tone, as modulators of neuronal function. It appears unlikely that systemically injected ET crosses the blood-brain barrier (5). A recent report by Yoshizawa and coworkers (7) demonstrated that ET-l-like immunoreactivity is localized to dot- and fibre-like structures and neurons in the dorsal horn of porcine spinal cord and that the ventral root depolarization produced by ET-1 in newborn rat spinal cord is depressed by the substance P antagonist spantide ([D-Arg1,D-Trp7,9,Leu11]-substance P), suggesting that endothelins may release substance P. Thus, speculatively, the stimulus associated with ET-1induced abdominal constriction may involve the release of substance P. Taken together, the findings of the present study suggest that the i.p. injection of ET-1 (or other endothelins or endothelin precursors; see ref. 1) produce(s) a noxious stimulus that is transmitted centrally via pain pathways. This stimulus can be blocked by morphine at opioid (naloxonesensitive) receptors at either the spinal or supraspinal level. Additionally, the morphine-induced antinociception apparently involves the I1 subtype of opioid receptors, a conclusion that agrees with many other studies using different stimuli. These findings further suggest that ET-l-induced abdominal constriction might be a novel model with particular potential application to the study of anginal or, because of the wide distribution of endothelin receptors, other visceral pain.
Vol. 49, No. ii, 1991
ET-l-lnduced Nociception
PL-65
References
1. R.B. RAFFA and H.I. JACOBY, Life Sci./Pharmacol. Lett. 48 PL85-PL90 (1991). 2. H. KURIHARA, M. YOSHIZUMI, T. SUGIYAMA, K. YAMAOKI, R. NAGAI, F. TAKAKU, H. SATOH, J. INUI, M. YANAGISAWA, T. MASAKI and Y. YAZAKI, J. Cardiovasc. Pharmacol. 13(Suppl. 5) S132-S 137 (1989). 3. H.O.J. COLLIER, L.C. DINNEEN, C.A. JOHNSON and C. SCHNEIDER, Br. J. Pharmacol. Chemother. 32 295-310 (1968). 4. S.J. WARD, PORTOGHESE, P.S. and TAKEMORI, A.E., J. Pharmacol. Exp. Ther. 220 494-498 (1982). 5. C. KOSEKI, I. MASASHI, Y. HIRATA, M. YANAGISAWA and T. MASAKI, Am. J. Physiol. 256 R858-R866 (1989). 6. C.R. JONES, C.R. HILEY, J.T. PELTON and M. MOHR, Neurosci. Lett. 97 276-279 (1989). 7. T. YOSHIZAWA, S. KIMURA, I. KANAZAWA, Y. UCHIYAMA, M. YANAGISAWA and T. MASAKI, Neurosci. Lett. 102 179-184 (1989).