Noxious mechanical stimulation of the hind paws of the anaesthetized rat fails to elicit release of immunoreactive β-endorphin in the periaqueductal grey matter

Neuroscience Letters, 149 (1993) 205-208 © 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/93l$ 06.00

205

NSL 09242

Noxious mechanical stimulation of the hind paws of the anaesthetized rat fails to elicit release of immunoreactive fl-endorphin in the periaqueductal grey matter A.W. Duggan a, EJ. Hope a, C.W. Lang a and B. Bjelke b ~Departmentof Preclinical Veterinary Sciences, Universityof Edinburgh, Edinburgh (UK) and bDepartmentof Histology and Neurobiology, Karolinska Institutet, Stockholm (Sweden) (Received 1 September 1992; Revised version received 23 October 1992; Accepted 23 October 1992)

Key words." fl-Endorphin release; Noxious stimulation; Periaqueductal grey; Antibody microprobe As a test of the hypothesis that an animal responds to a severe peripheral painful stimulus by a central release offl-endorphin, antibody microprobes were inserted stereotactically into the midbrain of urethane anesthetized rats. These microprobes bore antibodies to fl-endorphin immobilized to their outer surfaces. While microprobes were in the brain for periods of 10 to 30 min either no stimulus was delivered or alligator clamps were applied to both hind paws. Microprobes were then incubated with 125I-fl-endorphin. Quantitative image analysis of microprobe autoradiographs showed no differences between the no-stimulus and noxious-stimulus groups. Thus these experiments found no evidence for fl-endorphin release following a severe peripheral painful stimulus.

Acute noxious foot shock to rats produces a naloxone reversible analgesia [14, 15] and elevated circulating levels of immunoreactive (ir)fl-endorphin [10, 17]. Elevated levels of circulating fl-endorphin however, do not appear to produce analgesia [17] and a central release of opioids including fl-endorphin has been proposed as responsible for opioid analgesia produced by a variety of procedures. The sites where such release could occur are many including areas within the brain but also in the superficial dorsal horn of the spinal cord from activity in descending fibres [19]. The opioid fl-endorphin has a relatively restricted distribution within the brain. The cell bodies synthesizing fl-endorphin are found predominantly in the hypothalamic arcuate nuclei [12] and the axons arborize mainly in the hypothalamus, midbrain and medulla. Prominent among these are fibres coursing through and terminating in the periaqueductal gray matter (PAG). These PAGterminating fibres have been implicated in mechanisms of analgesia. Thus electrical stimulation in the ventral PAG of the rat produces naloxone reversible analgesia

Correspondence: A.W. Duggan, Department of Preclinical Veterinary Sciences, University of Edinburgh, Summerhall, Edinburgh EH9 1QH, UK.

[1, 14]. The site of release of an opioid peptide produced by such stimulation is not known with certainty but Herz and Millan [10] have suggested that the resultant analgesia derives from a release offl-endorphin within the PAG from direct stimulation of nerve terminals adjacent to the stimulating electrodes. One recent study [7] measured fl-endorphin levels in the PAG of rats following subcutaneous injection of formalin into the forepaws and found elevated amounts at 60 and 120 minutes post injection. Elevated levels of any compound in an area of the brain when measured by homogenization and extraction probably mainly represents that present within neurones. Thus elevated levels could equally result from a phase of inhibition of release just prior to extraction or from increased synthesis following a previous period of release. Difficulties have arisen from measuring fl-endorphin in cerebrospinal fluid (CSF). Early reports found that electrical stimulation in the region of the periaqueductal grey matter of humans, adequate to produce pain relief, was associated with elevated levels of fl-endorphin in CSF. A subsequent study however found that the contrast media used as aids to ventricular cannulation interfered with the radioimmunoassay for fl-endorphin and that increased levels could not be linked to pain relief [2]. Another procedure used for pain relief, transcutaneous vibratory

206 stimulation, has been recently shown not to produce changes in CSF levels offl-endorphin [8]. These difficulties associated with indirect measures of release indicate the need for a more direct approach. Although many of the experiments cited have used electrical stimulation, and this is relevant to clinical practice, the present experiments have examined a question fundamental to an understanding of the physiology of fl-endorphin: does a severe peripheral noxious stimulus evoke release of fl-endorphin in the region of the periaqueductal grey? This has been studied in anaesthetised rats by using antibody microprobes inserted through the cerebral cortex into the PAG and delivering a severe noxious mechanical stimulus to the hind paws. Although these experiments have had to be conducted with the constraint of anaesthesia, ethical considerations prevent the conduct of release experiments in conscious animals by any method when repeated noxious stimuli are used. Experiments were performed on 6 rats anaesthetised with intraperitoneal urethane (1.5 g/kg initially and supplemented when necessary). The trachea was cannulated and blood pressure measured with a cannula in a femoral artery. Animals were mounted in a stereotaxic head frame after 2% lidocaine had been sprayed into the ears, and bilateral bone flaps were removed over the sites of proposed microprobe entry. The dura mater was removed at these sites. The exposed areas of cerebral cortex were intermittently irrigated with warm Ringer's solution. When microprobes had been introduced, small pieces of plastic film were placed over the adjacent cerebral cortex to minimize drying. A controlled electric blanket was used to maintain body temperature in the range 36-38°C. The area chosen for penetration of the PAG was A.P. +2.2 mm using the stereotaxic atlas of Paxinos and Watson. Microprobes were prepared by immobilizing antibodies to fl-endorphin to the outer surfaces of glass micropipettes [4]. Briefly the micropipettes were coated successively with: a polymer derived from gamma amino propyltriethoxysilane, glutaraldehyde, protein A and finally antibodies. In vitro tests showed that incubation of microprobes in fl-endorphin, 10-8 M, for 30 minutes at 37°C resulted in near complete suppression of the binding of ~25I-fl-endorphin (Amersham). With such sensitivity microprobes are estimated to detect l 0 -17 mol of ligand bound over 100 p m of length [3] and this is one order of magnitude better than that of microprobes used to detect release of substance P [5]. Microprobes were inserted into the brain two at a time using stepping motor microdrives. In 5 experiments the microprobes were inserted 7 mm from the brain surface at approximately the same anterior-posterior level, but

A o

PAG

,!\ 625 •

{3 o')

~.%~.....,~_~

.~

1250

(.9 1875 •

2500 7

Depth within the brain (mm)

Fig. 1. A: the mean image scan of 38 microprobes inserted 7 m m into the brain along the track indicated in B, and no peripheral stimulus was applied, is plotted together with the mean scan of 31 microprobes present in the brain while a severe noxious mechanical stimulus was applied to both hind paws. The image analysis was performed at 30 p m intervals and for both groups a line has joined the mean density obtained at each point. For the no stimulus group the +S.E.M. has been plotted at each analysis point and the -S.E.M. plotted for the noxious stimulus group. Ordinate Grey scale: integrals obtained by transverse integration of optical density of microprobe images in 30 ,urn intervals. Abscissa Depth within the brain of analysis points. B: The t-statistics derived from the differences of mean density (grey scale) at each analysis point for the no stimulus and the noxious stimulus groups have been plotted along the line of introduction of microprobes into the midbrain. As the t-values are all less than 2, the observed differences are not significant at the P < 0.05 level.

from opposite sides, at an angle of 15° to the vertical. Such a track is illustrated in Fig. 1. In one experiment microprobes were introduced 10 mm into the brain. Microprobes remained in the brain for periods of 10 to 30 minutes, and either no peripheral stimulus was applied, or alligator clamps were applied to all of the digits of both hind paws and to the central pads with a sequence

207

of 3 rain on, 2 min off. This stimulus was experienced as very painful by the experimenters when applied to the fingers. Approximately twenty microprobes were inserted into the brain in each experiment and animals were killed by an intravenous injection of pentobarbitone sodium. Following removal from the brain, microprobes were first washed in cold phosphate buffered saline (PBS) and then inserted into a solution of ~25I-fl-endorphin in PBS containing bovine serum albumin 2%. After 24 h incubation at 60°C, microprobes were again washed and the tips were then carefully broken off and glued to a sheet of paper and placed in an X-ray film cassette with a sheet of monoemulsion film (Kodak NMC). Exposures varied from 3 to 10 days. Autoradiographs of microprobes were scanned with an image analysis system using an Imaging Technology PC Visionplus frame grabber board operating within an AT based computer [6, 9]. This performed microdensitometric estimates of microprobe images and compiled integrals derived by transverse integrations across each microprobe in 30 ~tm intervals. These integrals were stored on hard disc and a sorting program subsequently selected microprobes meeting designated criteria and plotted the mean integrals (+ S.E.M.) for each group, in 30 pm intervals, and determined the significance of differences between defined groups of microprobes at each interval (see Fig. 1). Fig. 1A illustrates the mean image analysis of 38 microprobes inserted 7 mm into the midbrain in the absence of any active peripheral stimulus and that of 31 microprobes present in the same area during noxious pinching of the skin. The two group analyses are virtually identical at all sites in the brain examined including the PAG. This is further shown in Fig. 1B which plots the statistics derived from the means of these two analyses in 30 pm intervals and these are placed along a microprobe track. This shows that at each site sampled, noxious stimulation of the hind limbs failed to produce elevated levels of ir-fl-endorphin and hence failed to release this neuropeptide. In one experiment microprobes were inserted 10 mm into the midbrain and hence the areas contacted at points along such microprobes differed from those inserted 7 mm, preventing addition of these results. In this experiment the mean image analysis of 12 microprobes inserted in the absence of stimulation, showed no significant differences from that of 14 microprobes inserted while alligator clips were applied bilaterally to the hind paws. The mechanical stimulus used in the present experiment was more severe than that which produces a release of ir substance P in the spinal cord of the anaesthetised cat [5] and rat (Hope and Lang, unpublished) and yet

failed to produce a detectable release of ir-fl-endorphin in the PAG. Anaesthesia may have reduced the responses of fl-endorphin releasing neurones but many naloxone sensitive events have been described in anaesthetised animals [11, 13, 18]. The results do not favour the response of an animal to a severe painful peripheral stimulus [16] being mediated in part by a release offl-endorphin within the PAG. 1 Akil, H., Young, E., Walker, J.M. and Watson, S.J., The many possible roles of opioids and related peptides in stress-induced analgesia, Ann. NY Acad. Sci., 467 (1986) 140-153. 2 Dionne, R.A., Mueller, G.E, Young, R.F., Greenberg, R.E, Hargreaves, K.M., Gracely, R. and Dubner, R., Contrast medium causes the apparent increase in fl-endorphin levels in human cerebrospinal fluid following brain stimulation, Pain, 20 (1984) 313 321. 3 Duggan, A.W., Antibody microprobes. In J. Stamford (Ed.), Monitoring Neuronal Activity: A Practical Approach, Oxford University Press, Oxford, 1991. 4 Duggan, A.W., Hendry, I.A., Green, J.L., Morton, C.R. and Hutchison, W.D., The preparation and use of antibody microprobes, J. Neurosci. Methods, 23 (1988) 241-247. 5 Duggan, A.W., Hendry, I.A., Green, J.L., Morton, C.R. and Hutcbison, W.D., Cutaneous stimuli releasing immunoreactive substance P in the dorsal horn of the cat, Brain Res., 451 (1988) 261273. 6 Duggan, A.W., Hope, EJ., Jarrott, B., Schaible, H.-G. and Fleetwood-Walker, S.W., Release, spread and persistence of immunoreactive neurokinin A in the dorsal horn of the cat following noxious cutaneous stimulation. Studies with antibody microprobes, Neuroscience, 35 (1990) 195-202. 7 Facchinetti, F., Tassinari, G., Porto, C.A., Galetti, A. and Genazzani, A.R., Central changes offl-endorphin-like immunoreactivity during rat tonic pain differ from those of purified fl-endorphin, Pain, 49 (1992) 113-116. 8 Guieu, R., Tardy-Gervet, M.-F. and Giraud, E, Met-enkephalin and fl-endorphin are not involved in the analgesic action of transcutaneous vibratory stimulation, Pain, 48 (1992) 83 88. 9 Hendry, I.A., Morton, C.R. and Duggan, A.W., Analysis of antibody microprobe autoradiographs by computerized image processing, J. Neurosci. Methods, 23 0988) 249 256. l0 Herz, A. and Millan, M.J., Endogenous opioid peptides in the descending control of nociceptive responses of spinal dorsal horn neurons. In H.L. Fields and J.M. Besson (Eds.), Progress in Brain Research, Vol. 77, 1988, pp. 263-274. 11 Kayser, V., Benoist, J.M., Neil, A., Gautron, M. and Gilbaud, G., Behavioural and electrophysiological studies on the paradoxical antinociceptive effects of an extremely low dose of naloxone in an animal model of acute and localized inflammation, Exp. Brain Res., 73 0988) 402--410. 12 Khachaturian, H., Lewis, M.E., Kang, T. and Watson, S.J., fl-Endorphin, ~-MSH, ACTH and related peptides. In A. Bj6rklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 4, Elsevier, Amsterdam, 1985, pp. 216~272. 13 Le Bars, D., Chitour, D., Kraus, E., Dickenson, A.H. and Besson, J.-M., Effect of naloxone upon diffuse noxious inhibitory controls (DNIC) in the rat, Brain Res., 204 (1981) 387~02. 14 Lewis, J.W., Cannon, J.T. and Liebeskind, J.C., Opioid and nonopioid mechanisms of stress analgesia, Science, 208 0980) 623~24. 15 Lewis, J.W., Terman, G.W., Watkins, L.R., Mayer, D.J. and Lie-

208 beskind, J.C., Opioid and non-opioid mechanisms of foot shockinduced analgesia: role of the dorsolateral funiculus, Brain Res., 267 (1983) 139 144. 16 Mayer, D.J., Endogenous analgesia systems: neural and behavioral mechanisms. In J.J. Bonica, J.C. Liebeskind and D.G. Albe-Fessard (Eds.), Advances in Pain Research and Therapy, Vol. 3, Raven, New York, 1979, pp. 385~,10. 17 MiUan, M.J., Przewlocki, R., Jerlicz, M.H., Gramsch, C., Hollt, H. and Herz, A., Stress-induced release of brain and pituitary fl-endorphin: major role of endorphins in generation of hyperthermia not analgesia, Brain Res., 208 (1981) 325 328.

18 Morton, C.R., Zhao, Z.Q. and Duggan, A.W., A function of opioid peptides in the spinal cord of the cat: intracellular studies of motoneurones during naloxone administration, Neuropeptides, 3 (1982) 83-90. 19 Zorman, G., Belcher, G., Adams, J.E. and Fields, H.L., Lumbar intrathecal naloxone blocks analgesia produced by microstimulation of the ventro medial medulla in the rat, Brain Res., 236 (1982) 77 84.