Increased [Met]enkephalin and decreased substance p in spinal cord following thermal injury to one limb

Q306-4522/90 $3.00+0.00 PergamoaPressplc

Neuroscfence Vol.36,No. 3.pp.731-736,1990

F&ted in ‘&eat Britain

@ 1990fBR0

INCREASED ~~T~ENKE~~AL~N AND INCREASED SUBSTANCE P IN SPINAL CORD FOLLOWING THERMAL INJURY TO ONE LIMB M, L. DE CEBALL~S,* P. JENNERQ and C. D. MAKDEN§ *Department of Neuropharmacology, Cajal Institute, C.S.I.C., Velazquez X44, Madrid 28006, Spain tMRC Movement Disorders Research Group, University Department of Neuroiogy and Parkinson’s Disease Society Research Centre, Institute of Psychiatry and King’s CoBege HospitaI Medical School, Denmark HiII, London SE5 6LX, U.K. &lniversity Department of Clinicat Neurology, Institute of Neurology, National HospitaI, Queen Square, London WC1 3BG, U.K. A~-Tk~s~ injuryto orxe hind limb of rats was induoed by ~rne~~o~ into water at 62°C. Both a mifd (15 s) or severe (30 s) lesion caused ingammstion of the limb when observed 24 h later; but at this time the animals used the: injured limb when they walked. Animals with a severe lesion of the injured limb subsequently withdrew it from use when walking. Limb withdrawal did not ooxu following a mild lesion. At 24 h following the lesion, lumbar spinal cord levels of ~M~t$enkephal~~, as measured by radioimmunoas~y* were &sated (70%) ~~~~~~~~in b&h ~rn~~~~, ipsi- and ~~t~lateral ta the lesion. At seven days following either mild or severe hind limb lesion ~Met]enkephal~n levels were elevated only in the ipsilateral lumbar hemisegment. At that time no changes in thoracic [Met]enkephalin levels were observed. Substance P levels were decreased (2~25%) bilaterally in the lumbar cord 24th following a severe limb lesion, but no change was observed at seven days in any cord segment following a mild or severe I&on. Changes in spinal cord [Metfenkepbalin confent occur in response to thermal injury to one hind limb. However, the changes do not appear to be related to the withdrawal of the damaged limb from use following a severe lesion. Peptide changes in the spinal cord may reflect pain or injury to the damaged limb following a thermal Lesion. In contrast, limb withdrawal may be a physioiogical rest mechanism r&ted to altered basal ganglia peptide fun&~on.

It has been suggested &at changes in basai ganglia enkephalin content may be responsible for animats

withdrawing an injured limb from the normal pattern of locomotion, so providing a physiological rest rn~~~~rn to &ow heafi~g.~ Rats receiving a s~ere thermal lesion to a hind paw initially walked on the injured limb, but in the next few days withdrew it from use. Twenty-four hours following the thermal lesion, at a time when the Zesioned Emb was stib used for walking, there was a reduction in [Met]- and [Leulenkephalin content of the periaqueductal gray and a bilateral reduction in [leu]enkeplralin content of the gfobxz pa%dus, bxit no change in e~ke~ha~~~ fevels in caudate-putamen. One week after the lesion, when animals showed complete withdrawal of the injured limb when walking there was a reduction of both [Met]- and ~Leu]en~~~ha~in content in periaqueductaf gray and globus pahldus; in addition, there was now a reduction in the [MetJenkephahn content of caudate-putamen The decrease was more

$To whom correspondence should be addressed at: King’s College London, Biomedical Sciences Division, Pharmacology Group, Manresa Road, London SW3 6LX, U.K. ~~b~~~~~~~~~~ BDTA. ethvfenediamj~~tra-acetate: HPLC. higk performance ‘liqAd &hromatograFby; IR, immunoreactivity; RIG, radioimmunoassay; SP, substance P.

marked in basal ganglia areas ~nt~~a~~~~ to the h&n5 These resdts suggested, but did not prove, that the basal ganglia enkephalin content may be an important factor in limb withdrawal after injury. Neuropeptide aitemtions in the spinal cord might also contribute to the abnormal locomotor behaviour following limb injury. An increase in wetlenkephahn content of spinai cord has beea reported fo~~ow~n~ induction of locafized pain and in~ammatio~ of a limb by administration of Freund’s adjuvant.3+s7 Some of these animals also withdrew the affected limb from use. Po,fToting dea~e~nta~on or h~~s~tjon of the spinal cord, the affected limb may not be used for a brief period .2,10*26 Golderberger” reported that after deafferentation most animals soon walked normally9 but some of the animals displayed a threeIegged walk, which gradually appeared eight to IO days pastoperatively. Deafferentation ahers the peptide content of spinal cord,6,22 [Met]enkephahn and substance P @PI levels are decreased. Traumatic spina injury causing motor ~rnp~i~ent also induces a locahzed decrease of SP content in the spinal cord related to the severity of the inj~ry.~ On the other hand, ~MetJenke~h~ju ievels are either unaltered or decreased after t~~at~~ spinal injury, irrespective of the severity of the lesion.9

731

The present work was designed to study the time course of the changes in peptide levels in the spinal cord following thermal injury to one hind limb. in relation to focomotnr use of the limb. Peptide concentrations were measured after a mild thermat injury, which did not modify ~~comot~~n, and after a severe thermai lesion, which induced withdrawal of the hmb from use. In addition, peptide content was assessed shortly after injury, when all animals used the inflamed paw.

Male Wistar rats (Cajal Institute, ISO-200~ at the start of the experiments) were anaesthetized with chioral hydrate (4~mg/kg~ i.p.). The right hind paw was lesioned by immersion into hot water at 62-C. Groups of animals were maintained for 24 h or one week after a mild therma injury (immersion for 1.5s) or severe injury (30 s immersion), before being killed. ~rnl~~jateIy after the lesion and at daily laterals tbereafter for six days, the animals were treated with ampicillin (1.5m&at, i.p.) and an ointment (triamcinalone acetonide I mg, neomycin sulphate 2.5 mg, gramicidine 0.25 mg and nystatin 100,000 W/g) was also applied to the lesioned limb. For biochemical studies only animafs showing withdrawat of the injured limb from use seven days fobowing 30s ~rnrnersi~o were employed. The condition of the animals was carefully monitored by the investigators at the Cajal Institute. Any animal showing weight loss or general debility was immediately killed. Animal numbers were restricted to the minimum required by the experimental protocol,

Animais were killed by decapitation, the spinal cord was removed and quickly dissected on a coid phte. In one experiment the whole tharacic and lumbar segments (2 cm) were dissected. In another experiment the lumbar enlargement was further dissected into two hemisegments. The segment was halved inta right and left sides through the medial f&sure. Dissected spinal cord areas were imme~ately frozen on dry ice. The sampies were boiled for L5mix in fOOvoi of a mixture of 1N acetic acid and 0.02 N R&l which contained 0.1% of 2-mercaptoethanoi. Samples were sonicated and centrifuged (12,OOOg, IO mm), the supernatant was decanted (S,), and the pehet resus~nd~ again in 100 val of the acid mixture and centrifuged again, retaining the supernatant (S,). Al&tots of the combined supernatants (S, + S,) were freeze-dried and stored at -7O’C. ~~et]~nke~h~~n was measured by rad~o~mmun~ssay {RfAf after high ~rfo~a~~ liquid chroma~o~~h~ @iPtC) separation of the extracts. In brief, samples were injected onto a Spherisorb 5 GDS 2 column (Phase Separation) and eluted with 0.5 M pyridine 0.5 M formic acid buffer, plti 4. containing 15% of n-propanol at a flow rate of f mlimin. Fractions of Zml ~o~s~~d~ng to @I] ~~et]enke~ha~~~ pak were collected and freeze-dried. The recovery of f”H] ~et]enke~~~ljn from the chromatographic separation was greater than 90%. In the [Metjenkephalin immunoassay the samples were reconstituted in the appropriate volume of RIA buffer (50 mM sad&m phosphate buffer, pH 7.4, containing 0.2% gelatin and 1OmM EDTA). An ahquot of the samples or standards were incubated with lOOn of antiserum ~Amersham ~~~ernat~ona1~and IoOlti of I”zsI][Met]enkeph&in ~~~rn~ iodinated by the &&amine T method1 in a final volume of 6OOpl. Coded samples were assayed’in duplicate. The tubes were incubated at 4°C for 18-24 h. Separation of bound and free peptidc was achieved by addition of 250 ~1 of dextran-coated charcoal (1.6% activated charcoal and 0.16% dextran in RIA bufferf. After

centrifugation (5OQOg, IO min) the supernatants woe dwanted and counted far I min in a ~atn~~~~counter. The a~ti~rum against ~~et]enkephaljn displayed no crossreactivity with dynorphin l-13. [Metfenkephaiin-Arg-GlyLeu, iMet~enke~halin”Ar~-Phe. ~ho~e~vstokin~n-~. neurotensin‘ or- SF.’ Howe&, a 2% cross-reactivity with ~~u]enke~~a~n was observed. No interference from the latter peptide should be expected since all the samples were subjected to HPLC purification prior to the immunoassays. The RIA sensitivity was 2.5 fmohtuhe (15041 displacement of bound tracer).

SP-~rnrn~~o~ct~v~ty @P-N) was measured by RIA of the tissue extracts. In some experiments SP-IR was cbaracterized by WPLC separation. The HPLC buffer used was identical to that described above but, containing 25% ~-~ro~anol, and eluted at a flow rate of 0.6 ml/mm Oneminute fractions were collected, and after freeze-drying they were assayed for SP-XR. The column was equilibrated with synthetic SP and pH]SP, the recovery for [33]SP was greater than 90%. O&y one eomp~nent of SP-XR was obtained with spinal cord samples and the peak had the same Position as synthetic SP. Samples were reconstituted in RIA buffer (50 mM sodium phosphate buffer, pH 7.2, containing 0.3% bovine serum albumin and 1OmM EDTA), and 5OpI incubated with iFOpl of tmI]SP ~6~~rn~ iodinated by the chfummine T method) and lffopl of rabbit an&SR serum, at a final dilution of t:5O,ooO, in a total volume af 850 ~1. After l&-24 h incubation, bound and free peptide was separated with 5flO~l dextran-coated charcoal. The antisera used in the SP RIA had no detectable gross-reactivity with the following peptides [Met]enkephaIin-Ark-Abe, [Metlenkepbalin-Arg-Gly-Leu, eledoisin, physaelemin, neurokinin A and negligible cross-reactivity (0.02%) with kassinin was detected. The sensitivity of the assay was 13 fmol/tube. RESULTS

Animals in which t&z right hind limb was immersed in hot water (UT) for i 5 s showed reddening of the skin and mild inflammation of the paw for two days after the lesion, but they used the paw normally when they walked. Animals subjected to a 30-s exposure in hot water had severe immediate inflammation but, this disappeared over the following days. After one week SD% of animals had wjthd~awn the lesioned limb from use; they walked and ran with the injured limb flexed, and held to the trunk, so that it did not touch the floor. Some of the animals (40%) also displayed self-mutilation of the affected paw that was not observed in animals subjected to the mild thermal lesion. Spinal cord [Metfenkephalin

and substance P lewfs

In control animals [~et]enkepha~~~ levels were higher in the lumbar enlargement of the spinal cord compared to the thoracic region (Fig. 1). These results arc in accordance with the segmental distribution of the ~ptide,7~1*,~~In animals which had received a thermal Iesion (30 s) one week before, and had withdrawn the limb from use [Metlenkephalin levels were increased in the lumbar spiaal cord but not in the thoracic region (Fig. 1).

733

Peptide levels following thermal injury

40C

,~

In control rats there was no difference in SP levels between the right and the left sides (Fig. 3). Acute thermal injury (30 s immersion 24 h previously) reduced SP levels in the ipsi- and contralateral hemisegments. In animals lesioned one week before (15 or 30 s immersion in hot water), there was no significant difference in SP levels in the ipsi- or contralateral lumbar spinal cord hemisegments.

/

--

DISCUSSION

200

Thoracic

Lumbar

Fig. I. Effect of severe thermal injury to a right hind paw on spinal cord [Met]enkephalin levels. Lesioned animals received a thermal injury (30 s immersion into hot water) one week prior to killing. Thoracic and lumbar refers to the spinal cord segment studied. Open columns, control animals; filled columns, lesioned animals. Results are mean ±S.E.M. of five to seven different determinations. *P < 0.05 lesioned compared to controls (Student's t-test). In subsequent experiments, peptide changes in spinal hemisegments were studied only in the lumbar region. In the control rats, the mean concentrations of [Met]enkephalin on the right and left sides were identical. Shortly after the thermal injury (30 s immersion 2 4 h previously) [Met]enkephalin content increased ipsi- and contralateral to the limb lesion in the lumbar spinal cord (Fig. 2). However, in animals lesioned one week before, either for 15 s and using the limb, or for 30 s and showing limb withdrawal, [Met]enkephalin levels were elevated to the same extent only in the hemisegment of the lumbar cord ipsilateral to the thermal lesion (Fig. 2).

Thermal injury to one hind limb induces inflammation of the paw, but initially the limb is used normally when walking. Depending upon the severity of the lesion, rats may subsequently withdraw the limb from use when walking. In this study we have examined the changes in spinal cord [Met]enkephalin and SP levels induced by a thermal lesion to one hind paw at two time points: (1) 24 h after the injury when the animal still walked on four legs and (2) seven days after the lesion when the animals had withdrawn the limb from use when walking. In the initial experiment a severe lesion to the hind paw causing limb withdrawal produced an increase in [Met]enkephalin levels in the whole lumbar cord seven days later but not in the thoracic region. [Met]Enkephalin immunoreactivity in spinal cord is thought to be contained within interneurons distributed throughout the spinal cord, predominantly in the substantia gelatinosa of the dorsal horn, 16'26 and terminals of a long bulbospinal pathwaY./7 Since [Met]enkephalin concentration was not altered in the thoracic region, thermal lesioning may primarily affect the [Met]enkephalin containing spinal interneurons. The second experiment addressed the issue of whether altered locomotor function was related to the changes in [Met]enkephalin-IR in lumbar spinal cord.

[ ] Control [ ] 80sec,-24h CONTRALATERAL /

IPSILATERAL

(,)

(,) o

iii,; ~I (")

(') C')

800-

;<:.:

i?.':? %:::!

[ ] 30sec ,1W

[:<":.S

(.)

[ ] 15 sec,lW

fD'. ......... I~:'!{

(,)

i:-".{ ,...-.,

aoo

,.. .-?

.....

!:.".:i

...] i ;-.',.';'<

i

Fig. 2. Effect of thermal lesion on lumbar spinal cord [Met]enkephalin levels. Lesioned animals received a thermal injury to the right hind limb. Open columns, control animals; stippled columns, animals lesioned for 30 s, 24 h post-lesion; cross-hatched columns, animals lesioned for 15 s one week before; hatched columns, animals lesioned for 30s one week before. Results are mean ±S.E.M. of five to eight determinations. *P < 0.05 lesioned compared to controls (Student's t-test).

734

M. L. DE CEBALLOS et al.

I

IPSILATERAL

I

CONTRALATERAL

0

control

8i

3088c,-24h

B

l!MC .lW

q 3u6ec ,lW

Fig. 3. SP-IR in lumbar spinal cord following a thermal lesion to the right hind limb. Open columns, control animals; stippled columns, animals lesioned for 30 s, 24 h post-lesion; cross-hatched columns, animals lesioned for 15 s one week before; hatched columns, animais tesioned for 30 s one week before. Results are mean + S.E.M. of five to eight different determinations. *P < 0.05lesioned compared to controls (Student’s t-test).

For that purpose animals subjected to thermal injuries of differing severity were compared. In this case [Metlenkephalin levels were measured in the two hemisegments of the lumbar cord. Twenty-four hours following a severe thermal injury, when all the animals used the inflammed paw, [Metlenkephalin-IR in both hemisegments of the lumbar cord was increased compared to controls. These changes may be due to the acute pain or injury induced by the lesion. [Met]Enkephalin content has also been found to be increased in the dorsal horn and ventral half of the spinal cord both ipsi- and contralateral to the inflammed paw of experimentally induced arthritic rats.4,7 At seven days following thermal lesioning, the changes in [Metlenkephalin levels were more localized. In animais either with a mild lesion walking on four legs or with a severe lesion walking on three legs, [Metlenkephalin-IR was increased in the segment of lumbar spinal cord ipsilateral to the injured limb. Spinal cord [Metlenkephalin immunostaining is increased in the rat mutant known as “mutilated foot”,*4 an autosomal recessive neurological disorder.” Affected animals showed decreased perception of noxious stimuli, self-mutilation of the limbs and ataxia. Furthermore, the density of [Metfenkephalin immunostaining of neurons in human spinal cord, following limb amputation, was enhanced ipsilateral to the surgery.” However, in the present experiments with thermal injury to a limb there was no difference between the effects of a mild or severe lesion, and the results were independent of use of the limb on walking. In contrast to [Metlenkephalin, a reduction in SP-IR was observed only 24 h after a severe thermal lesion in both hemisegments of the lumbar cord; there were no changes at later times, regardless of whether

the animals had a normal or abnormal gait. SP has been reported to be decreased in spinal cord following dorsal rhizotomy,i5,*’ periheral deafferentation>27 in the “mutilated foot” rat24,30and in familia1 dysautonomia.28 In all these instances the reduction of SP-IR occurred with loss of nociception. Human amputees also had decreased SP immunostaining in the dorsal horn of the spinal cord. ” This evidence suggests that the changes in SP observed in the present experiments may be due to alteration of sensory input into spinal cord. SP (and [Metlenkephalin) have been suggested to be involved in the transmission of pain perception. The excitatory neuropeptide SP, contained in primary afferents, is released by chemical or mechanical noxious stimulation either in ttitrd2 or in z&w.23.33 Further, Helme er uI.‘~ reported SP release from skin nerves by nociceptive thermal stimulus. Thus, the decrease in SP-IR in the spinal cord may have been due to the release of the peptide as a consequence of the thermal stimulation. The bilateral changes in SP levels are more difficult to interpret. However, changes in SP in the spinal cord after chemical nociceptive stimulus were bilateraLz3 and bilateral el~trophysiologica1 changes were observed in the spinal cord neurons after exposure to a unilateral thermal injury.32 Numerous electrophysiological and biochemical studies have shown that spinal cord enkephalinergic systems also participate in the control of nociceptive transmission (see Ref. 1 for review), through a presynaptic inhibition of SP release from sensory afferent fibres.” According to the same argument used in the case of SP, increased spinal cord [Metl~kephalin concentration would mean accumulation of the peptide, which is not released, as a consequence of loss of SP input.

735

Peptide levels FoLIowing thermal injury

than a. secondary response to the changes in use of the

CONCLUSION

affected limb during locomotion rapid and long-lasting change in lumbar r~et~e~~e~ha~~n content occurs as a con~quen~ of A

a thermal fesion to one hind paw. These changes in peptides in the spinal cord appear to k a direct consequence of alterations in sensory neurons rather

in such animals.

A~~~~~~edge~~~~-W~ grateMy ~~~~~i~~~ Prof. 3. Deg Rio and Dr De Fe&x for the& helpful discussions. XI& study was supported by the Medical Research Council and by the funds of the Bethlem Royal and Maudsley Hospitals, and King’s College Hospital.

REFERENCES 1. Basbaum A. I. and Fields H. L. (1984) Endogenous pain control systems: brainstem spinal pathways and endorphin

circuitry. A. Rev. Neurozi. 7, 309-338. 2. Berman D. and Rodin B. E. (1982) T&e influence of housing ~nd~~o~s on au~on~rn~ Fo~~o~n~ dotsai r%zotumy in

rats. Pain 13, JQ7-31 I. 3. Ccsselin F., Muntastruc J. L., Gros C., Bourgoin S. and Hamon M. (1980) Met-Enkephalin levels and opiafe receptors in the spinal cord of chronic suffering rats. Bruin Re$, 191, 289-293, 4. Cesselia F., Baurgoin S., Arfaud F. and flamon M. (1984) Basic reguiatoq mechan%ms of “in z&o” release of rne~-enkepba~~~From the dorszd zone of rat spinal cord. J: ~e~~~~. 43, 763-773. 5. De Ceballos M. L., Baker M., Rose S., Jenner P. and Marsden C. D, 11986) Do enkephaIins in basal ganglia mediate a physiological motor rest mechanism? Muu. Dis. 1, 223-233. 6. Di Giulio A. M.. Manteaazza P.. Dona M. and Gorio A. (19851 Perilrheral nerve lesions cause simultaneous alferations of substance P and enk-~h~~n IeveIs in the spinal cord.‘&& R& 342,4f?5-408. 7j. Faccini E., X.Gxmaki H., Govoni S., M&ale C., Spano P, F., CoveI% and Trabucchi M. (19%) Avant %res mediate the increase of met-enkephalin elicited in rat spinal cord by localised pain. P&z 1%.25-3f. 8. Faden A. I., Jacobs T. P. and Helke C, J, (1985) Changes in substance P and somatostatin in the spinal cord after traumatic spinal injury in the rat. Neurqvep~ides 6, 21Sz225. 9. Faden A. f.. Molineaax C, J.. Rosenberaer 1. G.. Jacobs T. P. and Cox B. M. t1985) Endoaenous onioid ~mrnuno~~t~v~f~ in raf spinal cord folbfiowi;g fra~a~i~ injury. Alu?. N~~~~~ X7,38&39& ” z iu. Golderberger M, and Murray M. (1974) Restitution of function aad collateral sprouting in the cat spinal cord: the deafferented animal. J. camp. Neurol. ls8, 37-54. 11. Colderberger M. (1977) Locomotor recovery after unilateral hind Ximb deaffereatation. Bra& Res. 123, 59-74. 12. Helke C. J., Jacobowitz D. M. and Thoa N. B. (I981) Gapsaicin and potassium evoked substance P release frum the nucleus fracfus solifarius and spinal trigeminat nucfeus in Y&Q. Efi Sci. 29%177%If%. If. Helke C. J., Neil J., Massari V. J. and Loewy A. D. (1982) Substance P neurons project from t&e ventral medutlar to the intermediolateral cell column and ventral horrx in the rat. &&I Res. 243, 147-152. 14. Hclme R. D., Koschorke 0, M. and Zimmermann M. (1986) Immunoreactive substance P release from skin nerves in tke rat by naxious thennaf stimulation. Neurusri. -Leu. 63, 295-299. I5 NiiikfeIt T., Kdlerfh f. O., Nilsson G. and Pemow B. (1975) Subsfance P: Iocalization in f&e OenfraI nervous syskm and in some primary sensory neurones. SGience i9U, 889-8%. 16. Hilkfelt T., Ljundahl A., Tennius L., Elde R. and Nilson G. (1977) Tmmunohistochemical analysis of pepfide pathways possibly related to pain and analgesia: cnkephalin and substance F. prproc.naln, Acad. Sci. U.S.A. 74, 3081-3085. 13, H&felt T., Tr?rsnius t., Kuypers H. G. J. M. and Dann 0. (1979) Evidence for onkephalin immunoreacfive neurons in fhe medulla ~bIo~ta prujecfions to the spinal cord. ~e~~o~~~.teu. 14 5560. f8. Howe 1. R., Yaksh T. L. and Go V. L. (1987) The effect of u&feral dorsal root ~n~lion~omi~ or ventrat rhizotomies on qadrenocepfor binding to, and the substance P, enkephalin, and neurotensin content of, the cat lumbar spinal cord. Neuroscience 21, 385-394. 19. Hunt S. P., Rossor M. N., Emson P. C. and Clement-Jones V. (1982) Substance P and enkephalin in spinal cord after limb amputation. Lrmeer 8279, 1023. 20. Jacobs 1. M., Scaravifli F., Duchen L. W. and Merfin J. (1981) A new neuroiogicai rat mutant ‘fnufifafed foot”. J. Atrar. 132, 525-543. 21. Jesse11T. M. and Iversen L. L. (1977) Opiate analgesics inhibit substance P release from rat trieeminal nucleus. Nature 268, 549-55 1. 22. JSS& T., TSUWOA., Kanaa~~ I. and Ofsuh M. (I%??) Substance P: deptetion in fhe dorsal horn of rat spinal card after sections af the peripheral processes of primary sensory nem-cpnwB&n Rrs. l$& X7-259, 23. Kantner R. M., Kirby M. L. and Goldstein B. D. (1985) Increase in substance Pin the dorsal horn during a chemogenic nociceptive stimulus. Brain Res. 338, 196,-199. 24. Kar S., Gibson S. J., Scaravilli F., Jacobs J. M., Aber V. R. and Pcrlak J. M. (1989) Reduced numbers ofcalcifonin sac-re-lated pe@ide {CGRP) and fa~h~~nin-immuno~acf~ve sensory neurones associafed with greafer enkepha& immunor~~t~vity in the dorsal horn of a mutant rat with hereditary sensory neuruFathy. C& T&S.Rex. 255,45&-46& 25. Ljungdahl A., HBkfelt T. and Nitsson G. (1978) Distribution of substanceP immunoreactivitv * in the central nervous system of the rat. 1: Cell bodies and nerve terminals. Neuroscience 3, 861-943, 26. Lombard M.-C.. Nashold B. S. Jr, Albe-Fessard D., Salman N. and Sakr C. (1979) DeafTererntation hypemensitivity in the rat affer dorsai rhiiotomy: a possible animal model of chrortic pain. Prrin 6, 163-174. 27. McGregor G. P., Gibson S. J,, &bate 1. M., Blank M. A., Cbristo~es N. D., Waft P. D., Polak J. M, and B&m S. R. (1984) Effit of peripheral nerve section and nerve crush. on spinal cord neuropeptides in the rat; increased VIP and PHI in the dorsal horn. Neuroscience 13, 207-216. 28. F”r?arsonJ., Brnndeis L. and Cue110 A. C. (1982) De@tion of substance P-containing axons in substantia gelatinosa of patients with d~~njs~d pain sensitivity, Nature 295. 61-63. 29. Sar N., Stumpf W. Es, M&r R. I., Cbz& K. 3. and -Cuatreczsas P+ jl97g:t ~rnrn~o~s~~~e~~ Iocalisatiou uf enkephafin in rat brain and spinaf cord. J. camp. Ne~of. 182, 17-38. 30. Scaravilli F. (1983) Reduced substance P in hereditary sensory neurapathy in the mf rat. &ain Res. 263, 147-150.

736

M. L.

DECEBALLOS

et al.

31. Takahashi T. and Osuka M. (1975) Regional distribution of substance P in the spinal cord and nerve roots of the cat and the effect of dorsal root section. Brain Res.87,l--l I. 32. Woolf C. J. (1983) Evidence for a central component of post-injury pain hy~rsensiti~ty. Nature 306, 686488. 33. Yaksh T. L., Jesse11T. M., Gamse R., Mudge A. W. and Leeman S. E. (1980) Intrathecal morphine inhibits substance P release from mammalian spinal cord in uiuo. Nufure 286, 155-157. (Accepted

12 February

1990)