0022-3565/05/3123-1257–1265$20.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 312, No. 3 Copyright © 2005 by The American Society for Pharmacology and Experimental Therapeutics 76224/1193072 JPET 312:1257–1265, 2005 Printed in U.S.A. Differential Mechanisms of Antianalgesia Induced by Endomorphin-1 and Endomorphin-2 in the Ventral Periaqueductal Gray of the Rat Maia Terashvili, Hsiang-en Wu, Randy J. Leitermann, Han-Sen Sun, Andrew D. Clithero, and Leon F. Tseng Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin Received August 13, 2004; accepted November 11, 2004 Downloaded from ABSTRACT The effects of pretreatment with endomorphin-1 (EM-1) and en- ␮-opioid receptor antagonist 3-methoxynaltrexone (6.4 pmol) se- domorphin-2 (EM-2) given into the ventral periaqueductal gray lectively blocked EM-2- but not EM-1-induced antianalgesia. Pre- (vPAG) to induce antianalgesia against the tail-flick (TF) inhibition treatment with dynorphin A(1–17) antiserum reversed only EM-2- produced by morphine given into the vPAG were studied in rats. but not EM-1-induced antianalgesia. Pretreatment with antiserum jpet.aspetjournals.org Pretreatment with EM-1 (3.5–28 nmol) given into vPAG for 45 min against ␤-endorphin, [Met5]enkephalin, [Leu5]enkephalin, sub- dose-dependently attenuated the TF inhibition produced by mor- stance P or cholecystokinin, or with ␦-opioid receptor antagonist phine (9 nmol) given into vPAG. Similarly, pretreatment with EM-2 naltrindole (2.2 nmol) or ␬-opioid receptor antagonist norbinaltor- (1.7–7.0 nmol) for 45 min also attenuated the TF inhibition induced phimine (6.6 nmol) did not affect EM-2-induced antianalgesia. It is by morphine; however, a high dose of EM-2 (14 nmol) did not concluded that EM-2 selectively releases dynorphin A(1–17) by attenuate the morphine-produced TF inhibition. The attenuation of stimulation of a novel subtype of ␮-opioid receptor, tentatively ␮ morphine-produced TF inhibition induced by EM-2 or EM-1 pre- designated as 3 in the vPAG to induce antianalgesia, whereas treatment was blocked by pretreatment with ␮-opioid antagonist the antianalgesia induced by EM-1 is mediated by the stimulation at ASPET Journals on January 20, 2020 Ϫ ϩ ␮ ␮ ( )-naloxone (55 pmol) but not nonopioid ( )-naloxone (55 pmol). of another subtype of 1-or 2-opioid receptor. However, pretreatment with a morphine-6␤-glucuronide-sensitive ␮ ␮ Endogenous opioid tetrapeptides, endomorphin-1 (EM-1) The -opioid receptors have been classified into 1- and ␮ and endomorphin-2 (EM-2), have been found to be highly 2-opioid receptors (Pasternak et al., 1980; Pasternak and selective for ␮-opioid receptors (Zadina et al., 1997). Both Wood, 1986). Recent studies indicate that the antinociceptive EM-1 and EM-2 potently compete with ␮-opioid receptor effect produced by EM-1 and EM-2 is mediated by the binding with no appreciable affinity with ␦- and ␬-opioid stimulation of different subtypes of ␮-opioid receptors. This receptors and selectively activate ␮-opioid receptor mediated view is evidenced by the findings that pretreatment with G-proteins (Goldberg et al., 1998; Narita et al., 1998; Monory ␮ -receptor antagonist naloxonazine or a novel morphine-6␤- 35 ␥ 1 et al., 2000). The increases of the [ S]GTP S binding and glucuronide-sensitive ␮-opioid receptor antagonist 3-me- antinociceptive effects induced by both EM-1 and EM-2 are thoxynaltrexone (Brown et al., 1997) selectively attenuates ␮ selectively blocked by the pretreatment with selective -opi- EM-2- but not EM-1-produced antinociception, suggesting oid receptor antagonist ␤-funaltrexamine or D-Phe-Cys-Tyr- that the EM-2-produced antinociception is mediated by the D-Trp-Orn-Thr-Pen-Thr-NH , indicating that the effects are 2 stimulation of ␮ and a novel ␮-opioid subtype receptor mediated by the stimulation of ␮-opioid receptors. 1 (Sakurada et al., 1999, 2000, 2001). Pretreatment with 2 D-Pro -endomorphin-1 blocks the antinociception produced This work was supported by Grant DA3811 and DA12588 from the National by endomorphin-1 but not endomorphin-2, and pretreatment Institutes of Health, National Institute on Drug Abuse (to L.F.T.). A prelimi- 2 nary report of these results was presented at the 33rd Annual Meeting of the with D-Pro -endomorphin-2 blocks the antinociception pro- Society for Neuroscience, New Orleans, LA, November 8–12, 2003. duced by endomorphin-2 but to a much less extent endomor- Article, publication date, and citation information can be found at http://jpet.aspetjournals.org. phin-1 (Hung et al., 2002). Pretreatment with antisense de- doi:10.1124/jpet.104.076224. oxynucleotide against Gi2 protein blocks the antinociception ABBREVIATIONS: EM-1, endomorphin-1; EM-2, endomorphin-2; [35S]GTP␥S, guanosine 5Ј-O-(3-[35S]thio)triphosphate; nor-BNI, norbinaltorphi- mine; Dyn, dynorphin A(1–17); vPAG, ventral periaqueductal gray; TF, tail-flick; NTI, naltrindole; CCK, cholecystokinin; NRS, normal rabbit serum; A/S Dyn, antiserum against dynorphin A(1–17); ANOVA, analysis of variance. 1257 1258 Terashvili et al. induced by EM-2 but not EM-1, whereas pretreatment with Assessment of Analgesia. Analgesic responses were measured antisense deoxynucleotide against Gi1 and Gi3 proteins with the tail-flick (TF) test (D’Amour and Smith, 1941). To measure blocks both antinociception induced by EM-1 and EM-2, in- the latency of the TF response, rats were gently held by hand and ␮ their tail positioned on the apparatus (model TF6; EMDIE Instru- dicating that -opioid-coupled G-proteins such as Gi2 is reg- ulated by EM-2 but not EM-1, whereas G and G are ment Co., Maidens, VA). The TF response was elicited by applying i1 i3 radiant heat to the dorsal surface of the tail. The intensity of the heat regulated by both EM-1 and EM-2 (Sanchez-Blazquez et al., ␬ stimulus was set to provide a predrug TF response time of 3 to 4 s. 1999). Blockade of -opioid receptors by pretreatment with The cutoff time was set at8stominimize tissue damage. norbinaltorphimine (nor-BNI) attenuates only EM-2- but not Experimental Protocol. Groups of rats were pretreated with EM-1-produced antinociception. These findings strongly sup- EM-1, EM-2, or vehicle before microinjection of morphine (9 nmol), port the view that EM-1 and EM-2 stimulate different sub- and the TF response was measured at 5, 10, 20, 30, 40, 60, 90, and types of ␮-opioid receptors to produce their pharmacological 120 min thereafter. The following three experiments were per- functions. The stimulation of this EM-2-sensitive ␮-opioid formed. 1) Determine the time course and dose-response effects of receptor subsequently induces the release of dynorphin A(1– the development of EM-2-induced antianalgesia against morphine- 17) (Dyn) acting on ␬-opioid receptors for producing antino- produced antinociception. Groups of rats were pretreated with EM-2 ciception (Tseng et al., 2000; Ohsawa et al., 2001). (7.0 nmol) given into the vPAG at different times (30, 45, 90, and 180 We have previously demonstrated that intrathecal pre- min, respectively) before microinjection of morphine (9 nmol) given into vPAG, and the antinociception produced by morphine was de- treatment of mice with a small dose of EM-2 attenuates the termined by the TF test. A similar protocol was also used to deter- analgesia produced by intrathecal administration of mor- mine the dose response of EM-1-induced antianalgesia. 2) Determine Downloaded from phine; the phenomenon has been defined as antianalgesia the type of receptors involved in the EM-2- and EM-1-induced anti- (Wu et al., 2003). The antianalgesia induced by EM-2 against analgesia against morphine-produced analgesia. Rats were pre- morphine-produced analgesia has been postulated to be spe- treated with opioid receptor antagonists, (Ϫ)-naloxone (50 min), (ϩ)- cifically mediated by the stimulation of a novel subtype of naloxone (50 min), 3-methoxynaltrexone (25 min) (Sakurada et al., ␮ ␮ 2000), nor-BNI (24 h) (Wu et al., 2003), or naltrindole (NTI) (10 min) -opioid receptor, tentatively designated as 3. This view is (Dervisogullari et al., 1996) into the vPAG before vPAG injection of supported by the findings that EM-2 pretreatment induces jpet.aspetjournals.org antianalgesia, which is blocked by the pretreatment with EM-2 followed by vPAG injection of morphine 45 min thereafter. The 3-methoxynaltrexone. Pretreatment with EM-2 subse- TF response was then measured 20 min after morphine injection. Other groups of rats were pretreated with (Ϫ)-naloxone (50 min), quently causes the release of Dyn to induce antianalgesia. 3-methoxynaltrexone (25 min), or vehicle given into the vPAG before The ventral periaqueductal gray (vPAG) area of the mes- vPAG injection of EM-1 followed by vPAG injection of morphine encephalon is a primary site sensitive to opioid agonists for 45 min thereafter. 3) Determine what endogenous opioid peptides producing analgesia (Smith et al., 1988; Yaksh et al., 1988). were released by EM-2 for the induction of antianalgesia. Rats Microinjection of morphine into the vPAG consistently pro- were pretreated into the vPAG with antiserum against Dyn, at ASPET Journals on January 20, 2020 duces analgesia, which is blocked by the pretreatment with [Met5]enkephalin, [Leu5]enkephalin, ␤-endorphin, cholecystoki- naloxone. The analgesia produced by opioid agonists from the nin (CCK), normal rabbit serum (NRS), or substance P given into vPAG is mediated by the activation of the spinopetal de- the vPAG 1 h prior to EM-2 vPAG injection, followed by vPAG scending pain control pathways, which are initiated from the injection of morphine 45 min thereafter. The tail-flick response vPAG via the rostral ventromedial medulla projecting to the was then measured 20 min after morphine injection. A similar protocol using antiserum against dynorphin A(1–17) (A/S Dyn) spinal and trigeminal dorsal horns (Basbaum and Fields, was also used to determine whether Dyn is involved in EM-1- 1984).