Substance P releases and augments the morphine-evoked release of adenosine from spinal cord

The effects of substance P on the morphine-evoked release of adenosine were examined. Substance P alone produced a multiphasic effect on release of adenosine, with release occurring at low nanomolar concentrations and at a micromolar concentration, but not at intermediate concentrations. An inactive dose of substance P augmented the morphine-evoked release of adenosine at a nanomolar (cid:14) . (cid:14) . concentration of morphine. Release of adenosine by substance P alone 1 nM or substance P r morphine 100 nM r 10 nM was Ca 2 q -dependent and originated from capsaicin-sensitive nerve terminals. q 1997 Elsevier Science B.V.

Substance P is present in small diameter unmyelinated primary afferent nerve terminals within the dorsal spinal cord and is involved in the transmissionrmodulation of w x nociceptive information 15 . Substance P depolarizes projection neurons and interneurons within the dorsal horn, and such postsynaptic actions have received emphasis with w x respect to pain transmission mechanisms 15 . There is also some evidence that substance P can modulate primary w x afferent function 10,17 . Substance P is released from w x primary afferent neurons by noxious stimulation 15 and release is increased under conditions of inflammation w x 6,19 . Opioids have been known for some time to inhibit the release of substance P from sensory nerve terminals w x contributing to antinociception 9 , but more recent studies report dual effects of opioids on substance P release with stimulatory and inhibitory effects being due to actions on w x different opioid receptor populations 21 . At supraspinal w x sites, substance P releases endogenous opioids 8,14 , and this contributes to some behavioural effects of substance P. Multiple forms of interactions appear to occur between opioids and substance P in relation to pain mechanisms. Within the spinal cord, release of adenosine mediates a component of morphine-induced antinociception. In behavioural studies, spinal opioid-induced antinociception is w x antagonized by pretreatment with methylxanthines 2,4 , while in neurochemical studies, opioids stimulate the release of adenosine in both in vivo and in vitro spinal cord w x preparations 22 . The morphine-evoked release of adenosine from dorsal spinal cord synaptosomes occurs at nanomolar concentrations in the presence of elevated K q concentrations; this release occurs via activation of mw x opioid receptors 2 . The present study determined whether substance P can induce adenosine release directly, and whether it augments morphine-evoked release of adenosine from dorsal spinal cord synaptosomes in a manner similar to K q . Ž Male Sprague-Dawley rats 250-325 g; Charles River, . Quebec, Canada were used. Adenosine release from dorsal spinal cord synaptosomes was examined in a synaptow x somal suspension as described previously in detail 2 . For intrathecal pretreatment with capsaicin, an acute cannula was inserted into the spinal subarachnoid space under w x halothane anaesthesia as described previously 22 . Cap-Ž . saicin 60 mg in 20 ml 60% dimethylsulfoxidersaline or vehicle was injected over a 7-10-min interval prior to cannula withdrawal. Animals were allowed to recover at least 7 days before being used in neurochemical experiments. Any animal displaying motor deficits as a result of this procedure was excluded. For Ca 2q -free experiments, 0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.
Ž . PII S 0 0 0 6 -8 9 9 3 9 7 0 0 4 7 3 -3 ( ) synaptosomes were prepared in a Krebs-Henseleit medium from which Ca 2q was omitted. Ca 2q was added back to synaptosomes during the incubation stage. All experiments included a time s 0 determination of adenosine generated by the experimental procedure, and this was subsequently subtracted from all other values. Adenosine release values are expressed as pmol adenosine released per mg protein.
Statistical comparisons were made using analysis of variance and Student Newman Keuls test for post hoc comparisons.
Substance P released adenosine in a multiphasic manner, enhancing release at 0.1-1 nM, and again at 1 mM but Ž . not at intermediate concentrations Fig. 1A . The extent of the adenosine released by substance P at both concentrations is comparable to that produced by maximum depolarq Ž w x. ization with K cf. 3 . Two threshold concentrations of Ž . substance P 0.01 nM and 100 nM were combined with morphine. Substance P at 100 nM enhanced release of Ž . adenosine by 10 nM morphine Fig. 1B , as does 6 mM q Ž w x. K cf. 3 . No augmentation of release was observed Ž . with 0.01 nM substance P data not shown . The release of Ž . adenosine evoked by substance P 1 nM and substance Ž . P r morphine 100 nMr10 nM appears to originate from capsaicin-sensitive nerve terminals, as release from cap-Ž . saicin-pretreated rats was significantly reduced Fig. 2A . Such release was Ca 2q -dependent, as no release occurred 2q Ž . when Ca was omitted from the medium Fig. 2B . These characteristics of release are identical to those observed for q Ž . morphine in the presence of 6 mM K Fig. 2A,B . The present study demonstrates that substance P can release adenosine from dorsal spinal cord synaptosomes in a Ca 2q -dependent manner. Substance P depolarizes a range of neuronal types by decreasing K q conductances, leads to enhanced Ca 2q entry via voltage-gated Ca 2q channels, and 2q w x induces Ca release from intracellular stores 15 . Substance P releases a number of neurotransmitters from spinal cord preparations; in some cases release is Ca 2q -dew x 2q w x pendent 11 , but in others, it is Ca -independent 10,18 , perhaps reflecting an involvement of different neurokinin receptors in these responses. An interesting feature of the adenosine release induced by substance P is its multiphasic nature. The neurokinin receptor subtype mediating release of adenosine by substance P at nanomolar concentrations is likely a neurokinin-1 receptor based on the potency of w x the effect 15 ; other subtypes may mediate the inhibitory phase and subsequent stimulatory phase at higher concentrations. Micromolar concentrations of substance P previously have been shown to release glutamate, acetylcholine and gamma-aminobutyric acid from spinal cord preparaw x tions 10,11,18 .
The capsaicin-sensitivity of the substance P-induced release of adenosine suggests that release occurs from small diameter primary afferent nerve terminals, as the capsaicin pretreatment schedule used here results in degenw x eration of C fibre profiles in the substantia gelatinosa 16 . A number of observations suggest that substance P can exert actions on afferent nerve terminals within the spinal cord. Thus, substance P releases glutamate from primary w x afferents 10 , alters primary afferent nerve terminal exw x citability 17 , and depolarizes sensory neuron cell bodies w x 20 . Ligand binding studies have failed to demonstrate any ( ) loss of substance P receptors in the dorsal horn following w x capsaicin pretreatment or rhizotomy 13,25 , but postsynaptic upregulation may have obscured a change in a small population of receptors. More recently, in situ hybridization analysis of mRNA for substance P receptors and immunohistochemistry of the substance P receptor itself in the spinal cord showed no evidence of substance P w x receptors on primary afferent nerve terminals 1 , and it was suggested that effects of substance P on C fibres are mediated indirectly by actions on interneurons. In the present study, release occurs from a synaptosomal suspension where anatomical juxtapositions are largely not retained. This observation initially suggests that a direct effect on synaptosomes occurs, perhaps by a direct depolarization. However, an indirect effect via release of endogenous opioids also is possible. Thus, spinal administration of substance P can produce a delayed analgesia which Ž is blocked both by naloxone suggesting release of endoge- . w x Ž nous opioids 5,23 , and by caffeine suggesting an adeno- . w x sine link also occurs 24 . The present demonstration that the effect of a nanomolar concentration of morphine is enhanced by substance P indicates that an amplification mechanism could occur in the synaptosomal suspension due to simple diffusion of a mediator without necessarily requiring an anatomical juxtaposition. Opioid-induced rew x lease of adenosine is capsaicin-sensitive 22 , and this would then account for the capsaicin-sensitivity of the adenosine released by substance P and the substance Prmorphine combination.
The interaction between substance P and morphine in releasing adenosine is of interest from a functional point of view. Substance P is released by acute noxious sensory w x stimulation 15 , and this could interact subsequently with morphine to augment antinociception. The spinal administration of low doses of substance P has been shown to potentiate antinociception by morphine using the thermal threshold tail flick test, and this exhibits a bell-shaped w x dose-response curve as does adenosine release 12 . Augmentation of the action of morphine could occur either by substance P releasing adenosine directly with adenosine w x subsequently enhancing the action of morphine 3 , or substance P enhancing the ability of morphine to release adenosine and accentuating the component of opioid action w x due to adenosine release 2,4 . Interestingly, under conditions of inflammation where release of substance P is w x enhanced 6,19 , morphine exhibits an enhanced spinal w x antinociception 7 . A substance P-adenosine-opioid axis could contribute to changes which occur under conditions of inflammation as well.