UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Supraspinal involvement in acupuncture analgesia Yee, K.C. 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_fall_yee_k_c.pdf [ 4.3MB ]
JSON: 831-1.0086657.json
JSON-LD: 831-1.0086657-ld.json
RDF/XML (Pretty): 831-1.0086657-rdf.xml
RDF/JSON: 831-1.0086657-rdf.json
Turtle: 831-1.0086657-turtle.txt
N-Triples: 831-1.0086657-rdf-ntriples.txt
Original Record: 831-1.0086657-source.json
Full Text

Full Text

SUPRASPINAL INVOLVEMENT IN ACUPUNCTURE ANALGESIAbyK.C.YEEM.D., Zuzhen Medical School, 1954M.S., The University of Washington, 1983Ph.D., Columbia University, 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDIVISION OF NEUROLOGICAL SCIENCES,DEPARTMENT OF PSYCHIATRYWe accept this thesis as conformingto the required standardTHE OF BRITISH COLUMBIAMay 1992(. K.C.Yee, 1992Signature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of /t-2OThe University of British ColumbiaVancouver, CanadaDate 1’ “Y)’°\DE-6 (2/88)Signature(s) removed to protect privacyABSTRACTElectroacupuncture (EA) on Zusanli (st. 36) and Shangjuxu (st. 38)(10 Hz, 1.0 ms) was found to produce a long-lasting inhibition of widedynamic range neurones in the dorsal horn of the spinal cord and toprolong the latency of the tail-flick reflex in the lightlyanaesthetized rat. This inhibition was effectively produced at astimulation intensity which excited only A13 fibres. The effects of EAwere eliminated by cold-blocking the spinal cord rostral to therecording site suggesting a supraspinal involvement in the EA-inducedinhibition of spinal cord nociceptive transmission. EA also facilitatedthe discharge of non-clock-like dorsal raphe neurones (NCL). Bilaterallesions of the ventrolateral tract (VLT), but not the dorsolateralfuniculi (DLF), blocked this effect suggesting that the ascending arm ofthe loop is via the VLT. The descending arm is located in the DLF sincebilateral lesions of the DLF blocked the effects of EA in the spinalcord.Evidence in the literature suggests that the dorsal raphe nucleus(DRN) may be involved in the above supraspinal loop as well as in anascending inhibitory pathway to the nucleus parafascicularis (NPF).Examination of the DRN revealed three types of neurones: clock-like(CL), NCL and non-clock-like non-responding neurones (NCLN). The NCLneurones were excited by noxious and non-noxious natural peripheralstimuli as well as EA. The other neurones were non-responsive to thesestimuli. NCL neurones of the DRN were also antidromically activated by11AbstractNPF stimulation indicating that the projection from the DRN to the NPFis direct. Stimulation of the DRN produced an inhibition of NPFneurones with sudden onset and offset and a duration correlated with thelength of stimulation. EA also produced long-lasting inhibition ofthese cells. The inhibitory pathway from the DRN to the NPF, which isactivated by EA and presumably mediated by NCL neurones, would appear tobe serotonergic. The evidence for this is that the inhibition evoked byDRN stimulation or EA is enhanced by alaproclate, a 5-FIT uptake blocker,and blocked by 5,7-DHT, a 5-HT neurotoxic agent, or cyproheptadine, aserotonin antagonist.J.G.Sinclair, Ph.D.Supervisor111TABLE OF CONTENTSABSTRACT.iiTABLE OF CONTENTS ivLIST OF ABBREVIATIONS viiLIST OF FIGURESLIST OF TABLES xivACKNOWLEDGEMENTS xvINTRODUCTION 1I. Rationale 5II. Specific aims 6BACKGROUND 8I. History of acupuncture 8II. Basic concepts of acupuncture meridians and points 10III. Clinic application of acupuncture anaesthesia 12IV. Acupuncture study in Western countries 13V. Peripheral receptors and afferent fibres 15Vi. Spinal cold interactions 15VII. Ascending-tract cells 17(a) Spinothalamic tract (STT) 17(b) Spinoreticular tract (SRT) 19(c) Spinomesencephalic tract (SMT) 20(d) Spinocervical tract (SCT) 21(e) Dorsal column pathway (DCP) 22(f) Postsynaptic dorsal column pathway (PSDC) 23(g) Spinohypothalaniic pathway (SHP) 23VIII. Descending modulation of nociceptivetransmission 24IX. Ascending modulation of nociceptive transmission 28ivTABLE OF CONTENTX. Drugs used to modify 5-HT transmission .30(a) 5,7-DI-IT 30(b) cyproheptadine 31(c) Alaproclate 32METHODS AND MATERIALS 35I. Electrode preparation 35(a)Carbon-fibre electrodes 35(b)Multibarrelled microiontophoretic electrodes 35(c)EA electrodes 36(d)TF-EMG electrodes 37(e)Peripheral nerve recording electrodes 37(f)Spinal cord lesion electrodes 37II. Equipment 37III. Software and statistical analysis 38IV. Experimental protocols 39Experiment 1. The effects of EA on the spinal cordWDR dorsal horn neurones and the tail-flickreflex latency (TF-EMG) 39Experiment 2. The effects of stimulating acupuncturevs non-acupuncture points on the inhibitionof the spinal cord WDR dorsal horn neurones 41Experiment 3. The types of afferent fibresactivated by EA 42Experiment 4. Tests for a supraspinal involvementin EA-evoked inhibition of spinal cordnociceptive transmission 43Experiment 5. Ascending and descending tractsin EA-evoked inhibition of spinal cordnociceptive transmission 43Experiment 6. Characteristics of the DRN neuronesin response to natural noxious and non-noxiousperipheral stimuli and EA 44Experiment 7. Characteristics of the NPF neuronesin response to natural noxious and non-noxiousperipheral stimuli and EA 45Experiment 8. Test of a direct projection fromthe DRN to the NPF 47Experiment 9. Effects of the DRN stimulationon NPF neuronal activity 48Experiment 10. Tests for a serotonergic projectioninvolvement in the ascending pathway from the DRNto the NPF 491) Serotonin neurotoxin study 492) lontophoretic studies 51V. Histology 55VTABLE OF CONTENTRESULTS .57FA on spinal cord WDR dorsal horn neuronesand the tail-flick latency (experiment 1) 57The effects of stimulating acupuncture vsnon-acupuncture points on the inhibitionof the spinal cord WDR dorsal horn neurones(experiment 2) 60Types of afferent fibres activatedby EA (experiment 3) 62A supraspinal involvement in EA-evoked inhibitionof the spinal cord nociceptive transmission(experiment 4) 65Ascending and descending tracts in EA-evokedinhibition of spinal cord nociceptivetransmission (experiment 5) 69The characteristics of the DRN neurones in responseto natural peripheral stimuli and EA (experiment 6) 71Test of a direct projection from the DRNto the NPF (experiment 7) 76The characteristics of the NPF neurones in responseto natural peripheral stimuli and EA (experiment 8) 81The effects of DRN stimulation on NPF neuronesand the TF-EMG (experiment 9) 88Serotonergic involvement in the inhibitoryprojection from the DRN to the NPF (experiment 10) 911) Studies on 5,7-DHT pretreated animals 912) lontophoretic studies 97DISCUSSION 118CONCLUSION AND SUMMARY 137BIBLIOGRAPHY . . .143APPENDIX 167viLIST OF ABBREVIATIONSAA acupuncture analgesiaB betaC centigradeCL clock-like neurones in the DRNcm centimeterDCP dorsal column pathwayEA el ectroacupuncture5,7-DHT 5,7-dihydroxytryptamineDLF dorsolateral funiculusDRN dorsal raphe nucleus6 deltaFig figureg gram(s)GABA gamma-aminobutyric acidHRP horseradish peroxidaseHz hertzi .p. intraperitonealkg kilogramKHz kilohertzM molarrn metermA milliamperesrng milligramviiLIST OF ABBREVIATIONSmm minutemicrometermM millimolarmm millimeterms millisecondnA nanoampereNaC1 sodium chlorideNCL non-clock-like neurones in the DRNNCLN non-clock-like-non-responding neuronesin the DRNNC11 centromedian nucleusNOC nucleus gigantocellularisNLA nucleus lateral anterialisNRM nucleus raphe magnusNPF nucleus parafascicularisPAG periaqueductal grayPCA p-chloroamphetamirieP-CPA p-chlorophenylalaninePSDC postsynaptic dorsal column pathwayR-S raphe-spinals secondSCT spinocervical tractSE standard error of the meanSHP spmnohypothalamic pathwayviiiLIST OF ABBREVIATIONSSMT spinornesencephalic tractSPA stimulation-produced antinociceptionSRI spinoreticular tractSIT spiriothalamic tractTF-EMG tail-fl ick-electromyogramVLT ventrolateral funiculusWDR wide dynamic rangeixLIST OF FIGURESFig. 1. [A inhibition of a spinal cord WDR neurone and TF-EMGFig. 2. EA reproducibly inhibits WDR dorsal horn neurones and prolongsthe tail-flick reflex.Fig. 3. The inhibition of WDR neurones is specific to acupuncturepoints.Fig. 4. Compound action potentials recorded from the tibial nerve uponEA stimulation.Fig. 5. The effects of EA and cold-block on the responses of a spinalcord WDR neurone to noxious radiant heat applied to the tail.Fig. 6. Cold-blocking the spinal cord eliminates the effects of EA.Fig. 7. Cold-blocking the spinal cord eliminates the effects of EA evenafter baseline adjustment.Fig. 8. Bilateral lesions of the VLT block the effects of [A on dorsalraphe neuronal nociceptor-evoked activity and the tail-flickreflex.Fig. 9. Bilateral lesions of the DLF block the effect of EA on thetail-flick reflex but do not alter the enhancement by [A onnociceptor-evoked activity in dorsal raphe neurones.Fig. 10. The extent of lesions noted in individual animals are shown forthe DLF and VLT.Fig. 11. Oscilloscope traces illustrating the effect of noxious heat(50°C) applied to the tail on a clock-like and non-clock-likeneurone.xList of FiguresFig. 12. Three dimensional representation of DRN CL, NCL and NCLNneurones.Fig. 13. EA enhances the nociceptor-evoked activity of a DRN neurone.Fig. 14. The effect of LA on the noxious radiant heat-evoked activity innon-clock-like DRN neurones.Fig. 15. The effect of LA on the noxious radiant heat-evoked andspontaneous activity in clock-like DRN neurones.Fig. 16. An example of a NCL DRN neurone exhibiting high frequencyfollowing and collision.Fig. 17. Oscilloscope trace illustrating the spontaneous and nociceptorevoked activity of an NPF neurone.Fig. 18. LA inhibits the nociceptor-evoked activity of a NPF neurone.Fig. 19. The effect of EA on the noxious radiant-heat evoked activity inNPF neurones.Fig. 20. The effect of EA on the activity of a NPF and DRN neuronerecorded simultaneously.Fig. 21. The effect of the DRN stimulus on the noxious radiant heat-evoked activity in NPF neurones and the TF-EMG latency.Fig. 22. Peristimulus histograms of 6 glutamate-driven NPF neuronestested by single pulse stimulation (1.0 s Hz, 0.2 ms, 0.6 mA)in the DRN.Fig. 23. An illustration of serotonergic immunofluorescence reaction inthe DRN region of a control rat and a 5,7-DHT treated animal.xiList of FiguresFig. 24. EA fails to inhibit the evoked activity of NPF neurones inresponse to noxious radiant heat in 5,7-Dill treated animals.Fig. 25 Stimulation of the DRN fails to inhibit the noxious radiantheat-evoked activity in NPF neurones or increase the tail-flick latency in 5,7-DHT treated animals.Fig. 26. An example of the effect of iontophoretic application ofalaproclate at 30 and 45 nA on a NPF neurone.Fig. 27. The effects of iontophores-is of alaproclate on NPF neuronesactivated by glutamate.Fig. 28. lontophoretic application of alaproclate enhances theinhibition produced by 5-HT but not GABA on NPF neurones.Fig. 29. An example of the effect of iontophoretic application of 5-HT,alaproclate and GABA on a NPF neurone.Fig. 30. The effect of iontophoretically applied alaproclate on noxiousradiant heat-evoked activation of NPF neurones and inhibitionby DRN stimulation.Fig. 31. An example of the effect of iontophoretically appliedalaproclate and DRN stimulation on noxious radiant heatevoked activation of a NPF neurone.xiiList of FiguresFig. 32.. The effects of alaproclate on DRN- and EA evoked-inhibition ofa glutamate-driven NPF neurone.Fig. 33. The duration of the NPF inhibition produced by DRN stimulationof 1, 2, 4 and 30 s.Fig. 34. The magnitude of the NPF inhibition produced by DRNstimulation.Fig. 35. Alaproclate iontophoretically applied at 30 nA enhances themagnitude but not the duration of EA-evoked inhibition of NPFneurones.Fig. 36. Cyproheptadine blocks the DRN-evoked inhibition of a glutamate-driven NPF neurone.Fig. 37. Cyproheptadine blocks the DRN-evoked inhibition of glutamatedriven NPF neurones.Fig. 38. Cyproheptadine blocks EA-evoked inhibition of a glutamatedriven NPF neurone.Fig. 39. Schematic arrangement of neurones activated by EA and a noxiousstimulus.xiiiLIST OF TABLESTable 1. Latencies and types of compound action potentials recordedfrom two sets of electrodes at various current intensities.Table 2. Characteristics of DRN neurones.Table 3. Changes in weight, behavioural activity and indirectimmunofluorescence reaction in 5,7-DHT treated rats.Table 4. lontophoretic application of alaproclate and Na on NPFneurones activated by glutamate.Table 5. The effect of DRN stimulation on NPF neurones driven byglutamate.Table 6. The effect of alaproclate on EA-induced inhibition ofglutamate-driven NPF neurones.Table 7. Cyproheptadine blocks the DRN-evoked inhibition of glutamatedriven NPF neurones.xivACKNOWLEDGEMENTSI am deeply grateful to Dr. J. G. Sinclair for his guidance,support and patience throughout this research.I also wish to thank the other members of my research committee,Drs. J. A. Pearson and [I. McLennan, for their help and advice. I alsowish to extend my thanks to Dr. Steven Vincent, Mr. Roland Burton andJeff Hockings for their valuable assistance.xvINTRODUCTIONAlthough acupuncture has been practiced for centuries and theneuronal mechanisms of acupuncture have been investigated over the pastforty years, the ascending pathways of electroacupuncture (EA) are notwell established. The focus of this study is to re-examine the validityof the proposed supraspinal loop on the spinal effects of EA in the rat,to characterize neurones in the dorsal raphe nucleus (DRN) and nucleusparafascicularis (NPF) in response to natural peripheral stimuli and EA,to determine whether there is a direct projection from the DRN to theNPF and to determine whether this ascending pathway is serotonergic.EA has been shown to inhibit wide dynamic range (WDR) neurones ofthe spinal cord in cats (Pomeranz et al., 1977; Pomeranz and Cheng,1979; Wu et al., 1986). There were, however, several discrepanciesbetween these studies. Pomeranz and Cheng (1979) elicited a relativelylow magnitude, delayed onset but prolonged inhibition, whereas Wu et al.(1986), using a higher stimulus intensity, observed a greater inhibitionwith immediate onset but which was abolished almost immediately with theoffset of the [A stimulus. The finding that the [A-induced inhibitiondisappeared immediately after discontinuing EA is not consistent withthe prolonged time course of acupuncture analgesia in humans (Yee,1973). Further Pomeranz et al. (1977) implicated supraspinal structuresin the long-lasting inhibition of the WDR neurones produced by [A. Theyfound that the inhibition was abolished by spinal cord transection,midcollicular decerebration or hypophysectomy and suggested that the1Introductioneffect was mediated by the release of a pituitary hormone. Wu et al.(1986) found the results were inconsistent on two neurones tested for asupraspinal involvement in EA. The inhibition on one cell was blockedafter cold-blocking the spinal cord while the inhibition on another cellwas not altered after transection of the dorsal half of the spinal cord.The inconsistencies noted above as well as the lack of informationavailable on the rat encouraged me to re-examine the supraspinalinvolvement of EA in attenuating spinal cord nociceptive transmissionand the time course of this inhibition by EA.In other studies on the cat, Du and Chao (1976) and Shen (1975)clearly showed that EA inhibited the viscerosoinatic reflex through asupraspinal loop. In addition, Shen et al. (1975), based on spinal cordlesion studies, concluded that the ascending arm of the loop was locatedin the ventrolateral tract (VLT) and the descending arm in thedorsolateral funiculus (DLF).Anatomical and physiological studies have shown that theperiaqueductal gray (PAG) neurones project directly to the nucleus raphemagnus (NRM) in both the cat (Hoistege, 1991> and rat (Pomeroy andBehbehani, 1979). Further, Liu et al. (1986) compared the effects of EAand PAG stimulation on NRM neuronal activity in the rat. They foundmost neurones to be responsive to both types of stimuli with an increasein activity and an inhibition of nociceptor-evoked responses. Bilateraldestruction of the PAG markedly reduced the effect of EA indicating thatEA activates NRM neurones partly through the PAG. Du and Chao (1976)2Introductionreported that a lesion in the NRM resulted in a significant decrease ofthe inhibitory effect of the viscerosomatic reflex during acupuncture.Their evidence indicates that the NRM is involved in the descendinginhibitory effects in acupuncture analgesia. A lesion of NRM alsoreduced the inhibitory effects of viscerosomatic reflexes produced bythe stimulation of the PAG (Du and Zhao, 1986).Cellular origins of projecting fibres in the DLF are concentratedin the NRM and in the adjacent nucleus reticularis paragigantocellularisin the rat (Basbaum and Fields, 1979; Pomeroy and Behbehani, 1974,1979). Xiang et al. (1986) found that the great majority of raphespinal (R-S) neurones did respond to noxious or EA stimuli. Thereceptive fields of the R-S neurones were very wide, covering almost allof the body. The effect of EA on these R-S neurones was mainly toincrease their firing rates and to inhibit their nociceptive-evokedresponses. After transection of DLF, the R-S neurones could still beactivated by EA indicating that the ascending projection of EA is not inthe DLF of the spinal cord (Xiang et al., 1986).Aghajanian et al. (1978) reported that the spontaneously firingneurones in the URN of the rat typically have a regular or irregularrhythm with a slow firing rate. However a stimulus to the sciatic nerve(constant current pulses of 50 #A, 0.5 ms, 1 Hz) was shown to produce aninhibitory response in the regular firing neurones. The mean durationof total suppression was 316 ms There was no suppression found inirregular firing neurones. These results would appear to be in conflict3Introductionwith the report of Shima et al. (1987) although different species wereused. It should be mentioned t.hat most investigators did not findregular firing neurones to respond to a noxious natural stimulus such aspinching (Aghajariian, et al., 1977a,b, 1978; Haigler, 1976; Wang andAghajanian, 1977a,b).Shima and his colleagues (1987) classified DRN neurones into twogroups with different patterns of firing in the anaesthetized cat:regularly firing (clock-like) and irregularly firing (non-clock-like)neurones. Clock-like neurones (CL) did not change their firing rate tonociceptive and non-nociceptive stimulation. On the other hand, manynon-clock-like neurones (NCL) responded to nociceptive and nonnociceptive stimulation. Also, NCL neurones were activated by theadministration of morphine which was dose-related and reversed by thenarcotic antagonist, naloxone. In the report of Shima et al. (1987)there was no information on whether NCL neurones of the DRN areserotonerg ic.In addition to the involvement of DRN neurones in descendingeffects, electrophysiological, autoradiographic and degeneration studiesshow an ascending projection from the ORN to the NPF of the thalamus(Bobillier et al, 1975; Conrad et al., 1974; Pierce et al., 1976) andthe cortex (Olpe, 1981). Electrophysiological evidence in the rat andthe cat shows the NPF receiving nociceptive input that arises fromspinothalamic pathways (Albe-Fessard et al., 1962; Chang, 1986).4IntroductionThus the DRN may modulate nociceptive transmission via ascendingas well as descending projections. Qiao et al. (1983), Zhang etal.(1986) and Qiao and Dafny (1988) reported that electrical stimulationof the DRN had a marked inhibitory effect upon nociceptive neurones inthe NPF. This inhibitory effect persisted, although reduced, aftertransection of the dorsal half of the spinal cord at T3-4, suggestingthat the inhibitory effect is not totally due to a reduction of theascending input from the spinal cord. The nociceptive discharges of theNPF can also be inhibited by activation of acupuncture point “Zusanli”in the rat (Chang, 1986) or “[legu” in the rabbit (Zhang et al., 1986).Therefore it is of interest to determine whether EA activates DRNneurones which project to the NPF. There is a suggestion byinvestigators at the Beijing Medical College (1986) that about half ofthe CL ORN neurones are facilitated by EA.Zhang et al. (1986) reported that the nociceptive discharges ofthe NPF were inhibited after intraventricular injection of 100 jig ofserotonin (5—hydroxytryptamine, 5-HT). Andersen and Dafny (1983a)reported that NPF neurones in animals treated with 5,7-dihydroxytryptamine (5,7-Dill), a serotonin neurotoxic agent, were notinhibited by DRN stimulation. They therefore proposed that the pathwayfrom the DRN to the NPF is serotonergic.I. Rationale:5IntroductionThe inconsistencies noted in the time course of inhibition of WDRdorsal horn neurones of the spinal cord produced by EA and the lack ofinformation available on the rat encouraged us to re-examine thesupraspinal involvement of EA in attenuating spinal cord nociceptivetransmission.There are conflicts in the reports studying CL and NCL neurones ofthe DRN as to which type respond to noxious or non-noxious stimuli.There is no information showing what kind of neurones in the DRN projectto the NPF, whether a direct projection from the DRN to the NPF occurs,whether EA activates this pathway and, if so, whether serotonin isinvolved in this ascending inhibitory pathway. Therefore, it isnecessary to characterize the DRN neurones in response to naturalnoxious and non-noxious as well as LA stimulation, to examine theprojection from the DRN to the NPF and to test for a serotonergicinvolvement.II. Specific Aims of This Work Are to Determine:1. The effect of EA on WDR dorsal horn neurones of the spinal cord andthe tail-flick reflex latency.2. The types of afferent fibres activated by LA.3. The effects of stimulating acupuncture vs non-acupuncture points onthe inhibition of the spinal cord WDR dorsal horn neurones.4. If there is a supraspinal loop involved in LA-evoked inhibition ofthe spinal cord nociception transmission.6Introduction5. Whether the ascending and descending LA neuronal activity travelsvia the VU or the IJLF if there is a supraspinal loop.6. The characteristics of neurones in the DRN responding to naturalnoxious and non-noxious peripheral stimuli and [A.7. The effects of URN stimulation on NPF neuronal activity.8. The effects of EA on NPF neuronal activity.9. Whether there is a direct projection from the DRN to the NPF.10. Whether the ascending pathway from the DRN to the NPF isserotonergic. To this end, responses of NPF neurones to DRNstimulation or EA will be tested using the following agents: 5-HT,a 5-HT uptake inhibitor (alaproclate); a 5-HT neurotoxin (5,7-dihydroxytryptamine) and a 5-HT antagonist (cyproheptadine).7BACKGROUNDI. History of AcupunctureThe information regarding acupuncture history in this backgroundwas gleaned from following books: “Acupuncture Manual, a WesternApproach” (Chu, et al., 1979) “Acupuncture, Textbook and Atlas” (Stuxand Pomeranz, 1987) and “Academy of Traditional Chinese Medicine”(Shanghai Traditional Chinese Medical College, 1975).Acupuncture is one of the oldest medical treatments in the world.Early in primitive society stone needles were used and can be traced asfar back as the New Stone Age, about 4,000-10,000 years ago. Flintneedles dating from 7000 to 5000 B.C. have been found which may suggestthat Neolithic humans used these needles to relieve pain and sickness.The emperors, Fu Hsi (accession 2852 B.C.) and Shen Nung(accession 2737 B.C.), were responsible for the development of Chinesecivilization, agriculture and medicine. Huang Ti, “the Yellow Emperor”(accession 2697 B.C., translated by H.C. Lu, 1973), conducted many wide-ranging dialogues on the subject with his minister, Chi Po, which werelater gathered into the celebrated classic known as the Nei Ching [TheYellow Emperor’s Treatise on Internal Medicine) which includes the SuWen and the Ling Shu. The Su Wen describes the entire field ofmedicine; the Ling Shu is a supplement and includes the first-knowndiscussion of acupuncture therapy.8BackgroundBetween 700-221 B.C. the theory of Jingluo (meridians) wasestablished. Hua To (A.D. 110-207), the first-known chinese surgeon,used acupuncture for headache and herb anesthesia for minor surgery.The “Zheri-jiu Jiayijing”, the first classic book on acupuncture,was written between 256-260 A.D. In the Tan dynasty (A.D. 617-907), the“Chai Kin Yao Fan” and “Chat Kin Vi Fan” were written by Sun Szu Mo inwhich all essential acupuncture methods, points and contraindicationswere discussed. During this period, an Acupuncture Institute wasestablished at the Imperial Medical College in Peking. This is theearliest medical school in China.One of the most important achievements in acupuncture practicetook place in the eleventh century during the Sung dynasty (A.D. 960-1206) when Wang Wei wrote the Tun Jen Ching in which he standardizedacupuncture points and suggested using a bronze model of the human formto demonstrate the location of the meridian and acupuncture points.This hollow bronze figure soon became a highly effective teaching device(Ornstein, 1976).In Europe, physicians first learned of acupuncture from the Germanbotanist, physician and traveler, Eugene Kampfer (1651-1716). InLondon, the surgeon, Croley (1802), reported some success usingacupuncture for headache, backache and rheumatism. John Elliotson(1791-1868) used the acupuncture technique at Saint Thomas Hospital inLondon. Sir William Osler (1849-1919) recommended acupuncture therapygBackgroundfor lumbago in his classic work, The Principles and Practice of Medicine(1892).II. Basic Concepts of Acupuncture Meridians and PointsHealth depends upon the flow of chi, meaning “life-force” alonginvisible pathways called meridians. Chi can be understood collectivelyas the vital energetic or functional components necessary for organiclife. Disease or pain is the result of a blockage of chi at one or moremeridians. By inserting needles into various acupoints, theacupuncturist dissipates the excess or replenishes the undersupply ofchi to the vital organs. Thus acupuncture restores health or dynamicequilibrium by regulating or rebalancing chi.The routes of energy circulation are known as the meridians. Themeridian is the simple name for channel (vertical paths of circulation)and collateral (horizontal paths of circulation). There are 12 mainmeridians and branch meridians. Internally, the meridians connect tothe zang—fu (organs) and superficial circulation throughout the body.The 12 meridians directly connected with the organs are as follows:lung, large intestine, stomach, spleen, heart, small intestine, bladder,kidney, pericardium, liver, gallbladder and triple warmer. The triplewarmer, although named as an organ, is not a organ by Westerndefinition. The energy circulation is said to start in the lungmeridian and successively flow through each of the meridians and their10Backgroundassociated organs to complete the cycle by going through the livermeridian.Acupuncture points are key points along the energy pathways of thebody known as meridians. The size of the meridians do not remainconsistent throughout whole body. In some places the meridians widenand in other places they become narrow. The places on the meridian thatare wider and where reactions appear on the skin surface are known asacupuncture points. Acupuncture points are regarded as “gates” oropenings by which the meridians communicate with the externalenvironment. Treating the meridians by stimulating the affectedacupuncture points serves to alleviate the dysfunction. Therefore theacupuncture points, in addition to being points at which one’s conditioncan be diagnosed, are points at which treatment can be applied. In thecase of abdominal pain, for example, the acupuncture points used todiagnose and to treat the problem will vary according to the location ofthe pain and which meridians are in the vicinity as well as whether itis a stomach condition or a liver condition.There are more than 360 acupuncture points on the body. These aremost commonly located in depressed spaces in the joints, creases in skinat joints, clefts between muscle and bone, in places where nerve trunkscome close to the surface or in places where cutaneous nerves reach theskin and the muscles. In a sense these are all places in the body thatare structurally more sensitive to a physical stimulus than other areas.11BackgroundIII. Clinic Application of Acupuncture Analgesia (AA)Acupuncture has a comparatively good analgesic effect. This haslong been known and used to treat a variety of inflanimations and pains.More than 2,000,000 operations have been conducted in China under AA.These include more than 100 different types of minor and majoroperations, such as abdominal tubal ligation, cesarean section, subtotalgastrectomy, splenectomy and open heart surgery. AA has the followingcharacteristics: patients remain in a conscious state throughout theoperation; all mental, sensory and motor functions are normal except fora dull pain sensation during the operation. Little or no anaesthetic isneeded, precluding the possibility of postoperative drug side-effectsapart from anaesthetic accidents. Postoperative pain is mild, andgenerally there are no such reactions as nausea and vomiting.Problems remaining to be solved in AA are incomplete analgesia andcontrol of visceral reaction. EA has not yet been improved to such alevel that it can ensure complete painlessness. Therefore the judicioususe of certain adjuvants to increase its analgesic action isjustifiable. The adjuvants most frequently employed inelectroacupuncture (EA) are central nervous system sedatives andanalgesics. Local anaesthetics may also be used. It is consideredimportant to estimate the effectiveness of AA before the operation, inorder to select suitable cases. Such patients respond to EA with anincreased pain threshold. Generally, the results of analgesia are goodBackgroundin patients in whom the pain threshold is high or can be elevated byneedi ing.IV. Acupuncture Studies in Western CountriesAcupuncture has been practiced for centuries in China, but onlyrecently it has begun to be accepted in the Western countries. Severalhospitals have been using EA for surgical procedures in the UnitedStates (Gaary,1975; Ledergerber,1976). Gaary (1975) reported that 56surgical procedures such as hip pinning, multiple dental extractions andremoval of 2 tibial staples were performed using EA. All patients werealert, cooperative and responsive throughout the procedures.Ledergerber (1976) used EA (0.2 ms, 0.93 mA, 0.25 - 30 Hz) for theinduction of labor and delivery. Of the 15 cases, 6 were completelysuccessful requiring no medication in labor and no local or regionalblock for forceps or episiotomy, 3 were partially successful and 6 weretotal failures. Richter et al. (1975) reported on 125 patientsundergoing open heart surgery using [A in Germany. They indicated thatone of the important advantages was a considerable reduction of the useof analgesics in the postoperative stage.Acupuncture has also been reported to be effective therapy for thetreatment of substance abuse and AIDS (Smith, 1990) and chronic severedepression (Jacob, 1990).13BackgroundAcupuncture has been accepted slowly in North America perhapsbecause of insufficient scientific explanations to describe itsmechanisms of action. Some studies on acupuncture mechanisms have beenconducted at a number of laboratories in the U.S.A. (Brockhaus andElger, 1990; Chapman et al.,1977, 1980; Davis, 1973; El-Etr and Pesch,1973; Gaw et aL, 1975; Greguss, 1973; Hynynen et at., 1981; Kitahata,1975, 1977; I..ee et al., 1975; Linzer and Atta, 1973; Looney 1973a, b,1975; Matsumato et al.,1973; Mayer and Liebeskind, 1974; Mayer et aL,1976; Numoto and Donaghy, 1973; Oleson and Kroening, 1983; Oleson etal., 1978; Philips and Rusy, 1973; Wagman 1973), Canada (Cheng andPomranz, 1979, 1980; McLennan et al., 1977; Pomeranz, 1973, 1977;Pomeranz, et al., 1977; Pomeranz and Cheng, 1979; Pomeranz and Paley,1979), Japan (Takeshige, 1981, 1985; Takeshige et al., 1976, 1980,1981), Finland (Duggan 1978) and Sweden (Anderson and Holmgren, 1975).Three national symposia on acupuncture in China in the last 25 yearshave illustrated that acupuncture analgesia involves the central nervoussystem. The proceedings of two of these have been published (Advancesin Acupuncture and Acupuncture Anaesthesia, 1979; Research onAcupuncture, Moxibustion and Acupuncture Anesthesia, 1986). Thestructures and background most pertinent to the present study aredescribed below.14BackgroundV. Peripheral Receptors and Afferent FibresCutaneous neurones have large (Aa13) and small (A6) myelinatedfibres as well as unmyelinated C-fibres. The conduction velocity ofthese fibres are 30-100 m/s, 4-30 rn/s and less than 2.5 m/s,respectively (Boivie and Perl, 1975; Gasser, 1950). Muscle nerves aredescribed as groups I, II, III and IV (Lloyd and Chang, 1948). GroupsIII and IV fibres are similar to A6 and C-fibres.Several types of sensory receptors are located at the peripheralends of the above nerves. These are extensively described by Willis andCoggeshall (1991). Mechanoreceptors which are activated by individualhairs in mammals are coded by two general classes of receptors: rapidlyadapting and slowly adapting (Brown and Iggo, 1967; Burgess et al.,1968). A number of these receptors respond to non-noxious stimuli.Nociceptors signalling the presence of damaging stimuli may besubdivided into mechanical, thermal and polymodal.VI. Spinal Cord InteractionsPrimary afferent fibres enter the spinal cord for the most part bythe dorsal roots. Upon entering the spinal cord the small fibresbifurcate sending axonal trajectories rostrally and caudally through thetract of Lissauer. In a transverse section, Lissauer’s tractcorresponds to a columnar area of fibres located between the dorsal rootentry zone and the superficial boundary of the dorsal horn gray matter.15BackgroundThe large fibres project more medially and also bifurcate sendingrostral and caudal projections in the dorsal columns. In each casecollaterals from the main axon branch off, enter the gray matter of thespinal cord and make synaptic connections (Brown, 1981).The scheme used by Rexed (1952) is a useful one and widely used toindicate the location of neurones in the spinal cord. He organized thespinal cord gray matter into 10 laminae: the first 6 are located in thedorsal horn, 7-9 in the ventral horn and lamina 10 surrounds the centralcanal. Although certain types of cells tend to be more prominent in onelamina than another, there is much overlap and the position of a cell ina lamina cannot be used to indicate its function.Collaterals from small fibres terminate mainly in the superficiallaminae of the dorsal horn, lamina I (the marginal layer) and 11 (thesubstantia gelatinosa). The A6 fibres primarily terminate in lamina Iwhile C-fibres end mainly in laminae II (Light and Perl, 1977, 1979).In contrast, the large-diameter fibres terminate mainly in lamina IIIand deeper laminae of the dorsal horn (Brown, 1981).Sensory processing in the spinal Cord results from interactionsamong primary afferent fibres, interneurones, ascending-tract cells anddescending-tract cells which modulate spinal cord neurones and afferentterminals. The majority of neurones in the spinal cord areinterneurones; it is estimated that 1% are ascending-tract cells and 2%are motoneurones (K. Chung et al., 1984).16BackgroundSpinal cord neurones responding to a noxious stimulus aregenerally of two types: WOR and nociceptive specific (NS). WDR cells,also referred to as lamina V-type, multireceptive or convergentneurones, are numerous and respond to both noxious and non-noxiousnatural stimuli. They can be found in most laminae of the spinal cordbut tend to be concentrated in lamina V. NS cells as the name implies,respond only to a noxious peripheral stimulus and are found mainly inlamina I (Christensen and Pen., 1970).VII. Ascending-Tract CellsPerception and behavioural response to sensory informationrequires that the sensory signals be transmitted in ascending pathwaysto supraspinal levels. It seems evident that the various tracts areinvolved in particular functions, but neurones in several tracts havebeen shown experimentally to respond to a noxious stimulus. There arespecies differences and due to the bulk of the literature most of thefollowing review will be restricted to the rat.(a) Spinothalamic Tract (STT)The STT cells are primarily located in the dorsal horn of thespinal cord although some cells are in the intermediate zone and ventralhorn (Willis and Coggeshall, 1991). Using retrograde tracing methods ofinjecting horseradish peroxidase (HRP) into various regions of the17Backgroundthalamus, one finds the largest concentration of cells in the cervicalsegments of the spinal cord. In lower spinal cord segments cells arefound mainly in the lumbo-sacral enlargement. Almost all the labeledcells at these levels are found contralateral to the injection site(Giesler et al., 1979). Cells projecting to the lateral thalamus arefound in the marginal layer, nucleus proprius and medial intermediategroups.STT cells projecting to the medial thalamus are in the medial baseof the dorsal horn and the intermediate gray. Axons to the lateralthalamus ascend more laterally than those to the medial thalamus(Giesler et al., 1981). SIT axons pass dorsolateral to the inferiorolivary nucleus in the medulla and via the medial lemniscus to thethai ainus.Target cells in the thalamus for SIT axons include the VLPnucleus, the centrolateral thalamic nucleus of the intralaminar complex[as well as adjacent parts of the medial dorsal and parafascicularnuclei] and the posterior complex (Lund and Webster, 1967; Peschanski etal., 1983).STT cells, as determined by conduction velocities, have a widerange of axon diameters including large and small myelinated axons andunmyelinated axons (Trevino et ai., 1973).STT cells in rats usually respond to both noxious and non-noxiousnatural stimuli (Giesler et al., 1976) and would be classified as WDR18Backgroundneurones. However relatively little work has been done on these cellsin the rat but they have been studied extensively in the primate (Willisand Coggeshall, 1991). Of particular importance to the present studyare the inputs to the medial thalamus. Some STT cells in laniinae IV-VIsend collaterals to the medial thalamus and some in the deeper layers ofthe spinal cord gray matter project only to the medial thalamus. Sincemany of these neurones are of the high-threshold type and thus unlikelyto provide sensory discrimination information, it is speculated thatthey may instead trigger a motivational-affective response (Willis andCoggeshall, 1991).(b) Spinoreticular Tract (SRT)Neurones in this tract, as the name suggests, originate in thespinal cord and terminate in the reticular formation. The axons projectin the ventrolateral white matter. Like the STT, the majority (85%,Chaouch, 1983; 74%, Kevetter and Willis, 1983) of SRT cells originate inthe spinal cord contralateral to their termination in the brain. Mostcells are located in laminae V, VII and VIII.One part of the SRT prajects to the lateral reticular nucleus, aprecerebellar nucleus. The medial part of the SRT innervates neuronesin the caudal brainstem, which may be involved in descending sensorymotor control systems, or project to higher levels including themidbrain and diencephalon. Using 2 different tracers, Kevetter and19BackgroundWillis (1982, 1983) found cells projecting to the reticular formationand the thalamus with about 10% of the cells projecting to both sites.The projections to the thalamus are primarily in the medial thalamus(Peschanski and Besson, 1984).The majority axons of almost all SRT cells are myelinated withconduction velocities ranging from 2-96 rn/s (Fields et al., 1977b; Maunzet al., 1978). Again, most SRI neurones are of the WDR variety.(c) Spinomesencephalic Tract (SMT)Using HRP injections to the midbrain tegmentum in the rat,Menetrey et al, (1980) found SMT cells, mostly contralateral, in laminaI, the lateral spinal nucleus, lamiria V and laminae at all levels of thespinal cord. R. P. Liu, (1983) found that the PAG projections werelargely in the ventrolateral PAG (including the DRN). Interestingly,many of the lamina I neurones projected to the PAG (Swett et al., 1985).Pechura and Liu (1986) have shown that some SMT neurones haveprojections both to the PAG and medullary reticular formation.Yezierski et al. (1991) found that 74% of SMT cells had a contralateralprojection and 26 % an ipsilateral projection.SMT axons ascend in the white matter of the ventral part of thespinal cord along with SIT and SRT axons. However some SMT lamina Ineurones project in the DLF. The PAG is innervated by SMT cells at allrostro-caudal levels (Zemlan et al., 1978; Swett et al., 1985). In the20Backgroundrat, SMT cells include the low-threshold, WDR and high-threshold variety(Menetrey et aL, 1980).In general, the main difference between these tract cells is theirpoint of termination. SIT, SRT and SMT cells have a somewhat similardistribution, the ascending projection for all is predominantlycontralateral in the VLT and they seem to be subject to similarperipheral inputs.(d) Spinocervical Tract (SCT)SCT cells project in the DLF of the spinal cord to synapse withneurones in the lateral cervical nucleus (LCN) which is located in thecervical region of the spinal cord, just ventrolateral to the dorsalhorn. The axons of LCN neurones decussate and ascend into the brainstemjoining the medial lemniscus on their way to the thalamus.The SCT does not seem to be well developed in the rat and probablyfor that reason has not received much attention in this species. It hasbeen studied much more extensively in the cat.Injections of HRP into the LCN results in labeled neuronesconcentrated in the ipsilateral nucleus proprius, particularly lamina IV(Giesler et al., 1978; Craig, 1976; Brown et aL, 1980). A few neuroneswere found in various laminae contralaterally.Axons of SCT cells terminate in the LCN but some fibres are notedto be collaterals of large fibres that continue to ascern in the DLF21Background(Enevoldson and Gordon, 1989). The targets of these continuing fibresare not known but some appear to synapse in the dorsal column nuclei(Enevoidson and Gordon, 1989).Brown and Franz (1969) characterized SCT neurones in the cat.They found that 30% of the cells were excited just by hair movement, 48%were excited by hair movement and pressure while 21% were excited bypressure and pinch. Consistent with this finding, Brown et al. (1975)reported that 29% of Sd cells were activated by A fibres while 71% wereexcited by both A and C-fibres. Cervero et al. (1977) found almost allSCT cells were responsive to non-noxious stimuli but most were alsoexcited (6l%) or excited and inhibited by noxious stimuli.The SCT is thought to be an important tactile pathway due to thevigorous activity of cells to small movements of hair. However, thefact that many also respond to a noxious stimulus and indirectly projectto the thalamus suggests that they may also be involved in nociception.(e) Dorsal Column Pathway (DCP)This pathway consists of branches of primary afferent fibres whichascend in the dorsal column. The dorsal column may be divided into thefasciculus cuneatus (containing fibres from the midthoracic to uppercervical levels) and the fasciculus gracilis (where fibres originatebelow the midthoracic level). These fibres terminate in nuclei with thesame name in the caudal medulla. Together they are called the dorsal22Backgroundcolumn nuclei. Only a portion of the primary afferents in the DCP reachthe DC nuclei, the majority terminating in the gray matter of the spinalcord. The DC also contain axons from propriospinal neurones andpostsynaptic DC neurones. Fibres from the DC nuclei project to thecontralateral thalamus via the medial lemniscus.The majority of cutaneous fibres in the DC contain rapidlyadapting hair follicle receptors at their peripheral terminals (Brown,1968).(f) Postsynaptic Dorsal Column Pathway (PSDC)Neurones forming this pathway in the DC are spinal cord neurones.HRP injections into the DC nuclei in the rat show that the cell bodiesare located in the nucleus proprius below the substantia gelatinosa(Giesler et aL, 1984).In the rat, Giesler and Cliffer (1985) found that the majority ofthese cells responded exclusively to non-noxious stimuli. The majorityresponded to non-noxious stimuli and strong mechanical stimuli. Almostnone responded to noxious heat and thus, this pathway is not consideredto play a major role for nociception in the rat.(g) Spinohypothalamic Pathway (SHP)Recent studies in Giesler’s laboratory (Burstein, et aL, 1987,1990) have revealed a bilateral projection to the medial and lateral23Backgroundhypothalamus in the rat. The cells of origin are in the deep dorsalhorn and lateral spinal nucleus and in laniinae I and VII. Some STTcells in the rat have collaterals to the hypothalamus. The terminationof these ascending SHP fibres has not yet been reported. SHP neuronesrespond to noxious heat. It has been speculated that these cells, areinvolved in autonomic and endocrine integration including viscerosomaticresponses to noxious stimuli. However it is also of interest that thehypothalamus supplies a major afferent input to the PAG (Beitz, 1982).Electrical stimulation of certain regions of the hypothalamus alsoproduces analgesia so it is possible that this loop is involved in thedescending modulating system for nociceptive transmission.VIII. Descending Modulation of Nociceptive TransmissionThe PAG has received much attention in analgesia studies sinceReynolds (1969) showed that electrical stimulation can produce a markedanalgesia without apparently modifying other sensory modalities. Thiseffect has since been referred to as “stimulation-produced analgesia”(SPA). Shortly thereafter, Guilbaud et a]. (1973) showed thatstimulation of the PAG produced a selective inhibition of nociceptivecells in the dorsal horn of the spinal cord. Further, discrete lesionsof the DLF blocked the SPA (Basbaum et al., 1976). The ventrolateralregion of the PAG, including the URN, was generally found to be the mosteffective site to produce SPA (Guilbaud et al., 1973). Mayer et a].,24Background(1974) found that there was no correlation between SPA and self-stimulation behaviour.Midbrain stimulation in the vicinity of DRN in chronicallyimplanted, awake cats and rats evoked profound analgesia to peripherallyapplied noxious stimuli (Aimone, et aL, 1987; Hung, C. et al., 1982;Liebeskind et al., 1973, 1983; Oliver et al., 1975, 1979). Du et al.(1978) reported that stimulation of the DRN also produced inhibitoryeffects on viscero-somatic reflexes in the cat. The peripheral field ofanalgesia sometimes includes the entire body.DRN neurones in the PAG, however, do not project to the spinalcord (Kneisley, et al., 1976; Kuypers, and Maisky, 1975). Rather, DRNneurones have been shown to project to and excite neurones in theventromedial medulla that contain the NRM and the adjacent reticularformation, the nucleus paragigantocellularis (Pomeroy and Behbehani,1979; Vanegas et al., 1984; Gallager and Pert, 1978). These neurones,in turn, project to various regions of the spinal cord via the DLF(Basbaum, et al., 1978). Many of these descending neurones contain 5-HT. The termination is most dense in the superficial regions of thedorsal horn which is, of course, the site of termination of the smallnociceptive primary afferent fibres (Basbaum et al., 1978).Consistent with their role in the decending inhibitory pathway fornociceptive transmission, stimulation of the NRM producesantinociception (Oleson et al., 1978; Proudfit and Anderson, 1975).25BackgroundAlso, lesions of the NRM and adjacent reticular formation antagonizesthe antinociception produced by glutamate injections into the PAG(Behbehani and Fields, 1979).The neurotransmitters released from PAG neurones to excite NRMcells are not known. However, one candidate is neurotensin which ispresent in a number of the projecting neurones (Beitz, 1982). Inaddition, this substance produces a dose-dependent antinociception wheninjected into the ventromedial medulla (Fang et al., 1987). Theexcitatory amino acids, glutamate and aspartate, are other candidates(Aimone and Gebhart, 1986).Bennett and Mayer (1979) and Mayer and Liebeskind (1974) reportedthat the analgesia produce by the simulation of the PAG was equal to orgreater than that produced by 10 mg/kg morphine.The DRN, which is located in the ventromedial region of the PAG,possesses the largest clusters of 5-HT neurones in the brain (Dahlstromand Fuxe, 1964; Descarries, et al., 1982). Numerouselectrophysiological studies have been performed on neurones containedin this nucleus. Aghajanian and his colleagues, working on chioralhydrate anaesthetized rats, have concentrated on the slow, regularfiring neurones and have provided the following evidence that these areserotonergic: they are located in the vicinity of clusters of 5-HTneurones demonstrated histochemically (Aghajanian and Haigler, 1974),they are antidromically activated by stimulating the 5-HT ascending26Backgroundpathway in the ventromedial tegmentum (VMT, Wang and Aghajanian, [977b)and these cells cannot be located after treatment with the tryptophanhydroxylase inhibitor, p-chlorophenylalanine (PCPA, Sheard et al., 1972)or the 5-HT neurotoxic agent, 5,7-dihydroxytryptamine (5,7-DHT,Aghajanian et al., 1978). These neurones are reliably inhibited bydrugs that would be expected to enhance 5-HT neurotransrnission (Sheardet al., 1978; Bradshaw et al., 1983; Brarnwell and Gonye. 1976). Theconduction velocity of their axons can be calculated to be in the rangeof 0.3-1.5 rn/s with the majority below 1 rn/sec in the rat (Wang andAghajanian, 1977 b). Sanders et al. (1980), however, found thatantidromic activation of DRN neurones by VMT stimulation was not a validcriterion for identifying 5-HI neurones.The great majority of these cells were reported to be resistent tonatural influences such as light flashes or noxious stimuli (Aghajanianand Wang, 1978; Haigler, 1976; Mosko and Jacobs, 1974). HoweverAghajanian et al. (1978), in a poststimulus histogram analysis, foundthese neurones to exhibit a transient inhibition following a lowfrequency and low intensity stimulus to the sciatic nerve.Aghajanian et al. (1978) also reported on two other types ofneurones located in the DRN. They referred to Type 2 neurones asquiescent or having a very slow discharge rate and Type 3 as having arelatively higher discharge rate, and irregular in rhythm. Neither ofthese types of cells were considered to be serotonergic in that they27Backgroundcould not be activated by VMT stimulation nor were they eliminated by5,7-DHT pretreatment. Type 2 cells could be activated by peripheralstimulation and Type 3 cells were either unaffected or showed an overallenhanced response which was not time-locked to the stimulus.In both the anaesthetized (Nakahama et al., 1981) andunanaesthetized (Shima et al., 1986) cat, regularly firing CL neuroneswere located in the DRN. Interestingly, almost no CL neurones wereactivated by noxious natural stimuli whereas about half of the NCLneurones were exitated by noxious stimuli. In addition, Shima et al.(1987) found that CL neurones were not activated by intravenous morphinewhile the majority of the nociceptive NCL neurones were responsive.Thus only the NCL neurones activated by noxious stimuli were responsiveto morphine. These investigators did not carry out studies to determinewhich type of DRN neurone was serotonergic.IX. Ascending Modulation of Nociceptive TransmissionEvidence indicates that the DRN is involved in ascending as wellas descending projections to modulate nociceptor-driven transmission.Oleson and Liebeskind (1976) recorded evoked potential and multiple-unitresponses in the medial thalamus as well as behavioural responses inawake, partially restrained rats. A noxious stimulus evoked nocifensivebehaviour, an evoked potential and increased multiple unit activity inthe medial thalamus. In the great majority.of rats, stimulation in the28Backgroundmidbrain PAG markedly reduced the behavioural responses as well as theevoked neurophysiological responses.More recently, several papers have been published from Dafny’slaboratory on the inhibition of NPF neurones by DRN stimulation(Andersen and Dafny, 1982, 1983a,b, ). They have presented someevidence that this is a 5-I-IT mediated inhibition. As mentioned earlier,some NPF neurones are activated by a noxious stimulus.Numerous investigators have shown that the NPF in the medialthalamus receives nociceptive information (Albe-Fessard and Kruger,1962; Andersen and Dafny, 1983b; Benabid et al., 1983; Dong et al.,1978). Several reports indicate that neurones in this nucleus havelarge receptive fields (Albe-Fessard and Kruger, 1962; Dong et al.,1978; Nyguist and Greenhoot, 1974 and Peschanski et al., 1981). This,plus reports that these cells are not intensity coded and respond tohigh intensity stimulation has lead to the suggestion that they are notimportant in sensory discrimination but rather they may be involved inidentifying novel sensory stimuli, especially harmful ones (Peschanskiet al., 1981). This is controversial, however, since Dong et a]. (1978)did find the neurones in this region of the cat to be intensity coded.Reports on rats are consistent in that at least 75% of the cells respondto a noxious stimulus (Benabid et al., 1983; Peschanski, et al., 1981).Consistent with this work is the study by Conrad et a]. (1974).They performed an autoradiographic and degeneration study to map the29Backgroundprojections from the DRN. The major ascending projections were found tosweep ventrally from the nucleus, course rostrally through the ventraltegmentum and into the medial forebrain bundle. Projections to the NPFwere noted in this study.X. Drugs Used to Modify 5-HT Transmission(a) 5,7-ONTBjorklund et al. (1974) have reviewed the properties of 5,7-DHT asa 5-HT neurotoxin. It appears that 5,7-DKT uses the neuronal uptakesystem and thus concentrates within the neurones and produces itsneurotoxicity. Although 5,7-DHT is most effective on 5-HT neurones italso produces a lower toxicity to noradrenaline neurones. Howeverexperimentally these neurones can be protected by pretreatment withdesimipramine, a relatively specific noradrenaline uptake blocker.Surprisingly dopamine neurones do not seem to be affected by 5,7-DHT.This neurotoxic agent would produce its maximal effect within one dayafter a 200 ug intraventricular injection and maintain the 5-HTdepletion (70% - 90%) at almost the maximal level for 30 days(Bjorklund, et al., 1974).(b) CyproheptadineStone et al. (1961) showed cyproheptadine was effective inblocking 5-HT actions in organ systems. They found it to be 165 times30Backgroundmore potent against 5-HT than noradrenaline. It failed to block theactions of acetyicholine but did have an antihistaminic effect.Segal (1975) examined the inhibition produced by DRN stimulationon rat hippocampal pyramidal cells. He presented evidence that thepathway was serotonergic including the finding that in 4 of 5 casescyproheptadine blocked the effect. It is however not clear what doselevel was used. Segal (1976) also found cyproheptadine to block theinhibition in these cells when lontophoretically released.Wang and Aghajanian (1977a) presented evidence for a directinhibitory DRN-amygdala serotonergic pathway in the rat. However,cyproheptadine (2-12 mg/kg, i.v.) or iontophoretically applied did notblock the inhibition.Olpe (1981) found DRN stimulation to produce an inhibition ofrostral and posterior cingulate cortical neurones in chloral hydrateanaesthetized rats. Cyproheptadine (10 mg/kg, i.p.) was required toconsistently block the effects. Doses of 1.0 and 3.0 mg/kg affectedonly 1 of 4 cells tested. lontophoretically applied cyproheptadineblocked the 5HT-induced depression on the same cells while beingineffective on GABA-induced inhibition.Interestingly, McLennan et al. (1977) found cyproheptadine (1.0mg/kg, i.v,) completely blocked the effects of EA in the rabbit.The type of 5-HT receptors blocked by cyproheptadine is not clear.There has been a lot of confusion concerning the nomenclature of 5-HT-31Backgroundreceptors (Glennon, 1986) but the classification proposed by Bradley etal. (1986) seems now to be largely accepted (Bonate, 1991). Under thissystem there are four main 5-HT receptors, 5-HT1, 5-HI2, 5-HI3 and 5-HT4with the 5-HT1 receptors further subdivided into 5-HT1a, 5HT1b, 5-HT1cand 5-HT1d. Cyproheptadine is generally regarded as a 5-HT2 antagonist(Bonate, 1991). However, cyproheptadine and mianserin share a similarbinding profile and also have a high affinity for the 5-HTic site(Asarch et al, 1985; Glennon, 1986). On the other hand, Peroutka (1986)reports a cyproheptadine affinity for 5-HT1a > 5-HT1b > 5HTic. Thus,cyproheptadine would appear to be a 5-HI2 antagonist, the effect on 5-HT1 receptor subtypes is not presently clear. Cyproheptadine cannot beregarded as a specific 5-HT antagonist since it can block otherreceptors as well, particularly histamine H1 receptors.(c) AlaproclateThe major method of inactivation of 5-I-IT neurotransmission isthrough a neuronal uptake mechanism and thus drugs which block thisuptake would be expected to enhance 5-HI synaptic activity. It isuseful to use a drug which has a high specificity for the uptake systembecause other neurotransmitters are also inactivated through uptakesystems.Alaproclate, a monocyci ic compound [2-(4-chlorophenyl)-1, 1-dimethylethyl-2-aminopropanoatej was first reported by Lindberg et al.32Background(1978) to be a specific 5-HI uptake blocker. They found alaproclate tohave an IC50 of 1.4 x i0 M for the inhibition of 5-HI accumulation insynaptosomes. It was found to be 100 times more potent in blocking theuptake of 5-HT than noradrenaline.Ogren et al. (1984), in an in vivo study, reported that a dose of40-60 mg/kg completely blocked the depletion induced by 4-methyl-a-ethy-m-tyramine (H 75/12) in rats. Clomipramine, another 5-HT uptakeblocker, even at high doses failed to completely block the depletion.The same dose of alaproclate as mentioned above had no effect on thedepletion of noradrenaline or dopamine induced by 3-hydroxy-4-methyl-a-ethyl-phenethyalamine HC1 (H 77/77). Alaproclate showed a regionalselectivity in blocking 5-HT uptake. Alaproclate was found to be mostpotent in the hippocampus (ED50 = 4 rug/kg) and hypothalamus (ED50 = 8mg/kg) followed by the striatum (ED50 = 12 mg/kg) and cerebral cortex(ED50 18 mg/kg). It exhibited a low potency in the spinal cord (ED50> 30 mg/kg). Alaproclate failed to have an effect at concentrations.: of< 10 #M on the following receptors examined: 5-HT, histamine H1; a, a2-adrenergic; dopamine D2 and muscarinic. Ogren et al. (1985) reportedthat the apparent potentiation of the muscarinic response was via aserotonergic mechanism.In another study, Ogren and Holm (1980) found alaproclate to beeffective in the hot-plate test but not the tail-flick test. Later Eideand Hole (1988) confirmed that alaproclate was ineffective in the tail33Backgroundflick test when administered in a single dose but was effective whenadministered chronically. Ogren and Berge (1985) found that pchioroamphetamine (PCA), a 5—HT releasing compound, produced analgesiain the hot plate test in rats which was blocked by alaproclate (20mg/kg) while desipramine was ineffective Interestingly,cyproheptadine (1.0 mg/kg) was also ineffective in blocking the responseof PCA but the dose may have been too low.34METHODS AND MATER IA[SI. Electrode Preparation(a) Carbon-fibre electrodesAll single unit recordings and DRN stimulaton in this study weremade through carbon-fibre electrodes. They were prepared according tothe method of Armstrong-James and Millar (1979). A carbon fibre (8 jim)of suitable length (>10 cm) was inserted into a glass capillary filledwith acetone (KIMAX-51 capillaries of 10 cm length and 0.8 mm (o.d.);Kimble Products) and pulled on a vertical microelectrode puller(Narashigi). The electrode formed had several cm of carbon fibreprotruding from the tip which was then cut with scissors to a few mmfrom the glass tip. Under a light microscope, the fibre was positionedinto a silver loop containing a drop of 1 M chromic acid. The fibre wasetched to a point by passing a current of 0.12 - 0.30 mA (AC) so thatnot more than 15 jim of carbon fibre protruded from the microelectrode.The electrodes had an impedance which ranged from 200 kf2 to 2 M2.(b) Multibarrel led inicrolontophoretic electrodesFive-barrel microelectrodes were constructed from glasscapillaries with an outside diameter of 1.0 mm (Glass Company ofAmerica, Omega Dot Brand). These capillaries contained a single glassfibre strand to faciljtate their filling by capillary action. Four35Methods and Materialsindividual capillaries were bent to an obtuse angle at about 0.5 cm fromone end. The length of these capillaries from the bend to the oppositeend was about 4.0 cm. Four of these capillaries were glued together(cold cure denture material) with a 10 cm (0.8 mm, o.d..) centre glasscapillary which contained a carbon-fibre. This centre capillaryprotruded about 3.0 cm beyond the bent end of the outer barrels. Thelower end of this assembly was also held together by glue. After 24 hrsof curing at room temperature, the capillary assembly was pulled into amultibarrelled microelectrode using the Narashigi vertical puller.Here, the unit was held at each end by the central capillary tube,heated via a coil midway between the glued ends and gently twisting 180°while being allowed to fall 0.5 - 1 cm by gravity as the glass melted.The heat source was turned off and the glass allowed to cool. It wasthen pulled in the normal fashion as single electrode. Under a lightmicroscope the carbon-fibre of the central barrel was trimmed and etchedto a point as described in (a). The surrounding barrels were filledwith appropriate drug solutions.(c) [A electrodesStainless steel uninsulated needles (China National ChemicalsImport and Export Corporation) 1.0 - 2.0 cm long, 34 gauge (0).36Methods and Materials(d) Tail-Fl ick-Electromyogram (TF-Et4G) electrodesStainless steel needles 2-4 cm long, 32 G and insulated except forthe tip.(e) Peripheral nerve recording electrodesSilver chiorided hook electrodes, 24 G.(f) Spinal cord lesion electrodesConcentric bipolar platinum electrodes (David Kopf Instruments,Model NE-100) were used to lesion the DLF or the VLT of the spinal cord.II. EquipmentThe recording microelectrode was secured to a hydraulic microdriveholder (David Kopf Instruments) which, in turn, was mounted to a fineadjustable electrode carrier. The signal picked up by the recordingmicroelectrode was fed into a high impedance preamplifer ( WPI,Microprobe system, M-707A), bandpass filtered at 3 KHz, amplified andsubsequently displayed on a oscilloscope (Tektronix, M-Dll). The outputof the oscilloscope was usually fed to an four-channel tape recorder(Teac, A-3440), a window discriminator (Digitimer D130) set for period2.5 s and range 0.25 s, a D.C driver amplifier of a polygraph (Grass M79D) and a IBM computer via an A/D interface.37Methods and MaterialsIII.. Software and Statistical AnalysisExtensive software was written in Turbo Pascal 5.5, which allowedthe microcomputer system to be used for data collection, display,analysis and real time control of equipment (see Appendix 1).Various versions of peristimulus and latency histogram programswere written to allow analysis of the neuronal responses to differenttreatments. In most cases the histograms consisted of 200 - 300 binswith a bin width of 0.5 ms - 10 s. Data collected by these programswere then stored onto floppy diskettes. Programs provided the storeddata to display and calculated the required statistical measures.Other programs controlled the application of drugs byiontophoresis while simultaneously recording and displaying the neuronalfiring rate visually, both as a time versus rate display andnumerically. Again, the collected data were stored on floppy diskettesfor recall by other programs for further display and analysis. Theseprograms allowed a number of drug application trials to be averaged.Graphs were plotted on a Sun (3/60) computer system. All softwareprograms are listed in Appendix 1.The data were analyzed statistically using the one way ANOVA testand the computed F ratio was used to determine significant differenceamong group means (Devore, 1982). The significant difference betweenpairs of means was then determined using Fisher’s Least Significant38Methods and MaterialsDifference multiple range test. A significant difference betweencompared values was accepted at p<0.05.IV. Experimental ProtocolsExperiment 1. The Effect of [A on the Spinal Cord WDR Dorsal HornNeurones and the Tail-Flick Reflex Latency (TF-EMG).Male Wistar rats (200-300g) were anaesthetized with urethane (1.0g/kg, i.p.) and supplemented during surgery with the addition ofhalothane. A laminectomy was performed in the 110 - L6 region of thespinal cord. The animals were then rigidly placed in a stereotaxicheadholder and spinal frame (Narishige M-11A). A cold-block device,described by Sinclair et al. (1980), was positioned on the spinal cordimmediately rostral to the thoraco-lumbar junction. Two stainless steelelectrodes were placed bilaterally in the abductor caudal dorsalismuscles at the base of the tail to record the tail-flick reflex latency(TF-EMG; Peets and Pomeranz, 1987). EA electrodes were inserted about0.4 cm apart bilaterally into the m. tibialis anterior 3.0 - 5.0 mmdeep. The sites correspond approximately to the classical humanacupuncture points Zusanli’ (St. 36) and HShangjuxufl (st.38). Rectaltemperature was monitored by an electronic thermometer and automaticallymaintained within physiological limits by a feedback-controlled directcurrent heating pad. The animals exhibited no signs of discomfort39Methods and Materialsfollowing removal of the halothane but they did respond with reflexmovements to a noxious stimulus.A tail-flick reflex was produced by focusing a halogen projectorlamp on a thermocouple placed on the tail 3.0 cm from the tip. Theintensity of the lamp was feedback-controlled to maintain a temperatureof 50°C. The TF-EMG latency, measured at 2 mm intervals, was the timebetween the onset of the lamp and tail-flick. A latency histogramcomputer program (first channel) with bin width 0.5 s was used in theseexperiments (S4, see Appendix 1). The cut-off time was 8.0 s and theaverage latency of the TF-EMG was between 4 and 5 s in control animals.In the same animals, a carbon-fibre microelectrode was loweredinto the dorsal horn at a depth of 0.3 mm to 0.7 mm below the dorsalsurface of the lumbar spinal cord to record single unit activity fromWDR neurones. As the electrode was advanced, mechanical stimuli (fingerpressure) were applied to the body or the tail. Neurones exhibiting WDRcharacteristics to mechanical stimuli were selected for further study ifthey also responded to noxious radiant heat of the tail. Peristiinulushistograms (second channel) with bin width 0.5 s were computed from 4responses to noxious radiant heat applied to the tail at 2 mm intervals(S4, see Appendix 1). The spontaneous activity was averaged in the 20 sprior to the radiant heat and the mean evoked activity was averaged inthe 10 s following the onset of the heat pulse.40Methods and MaterialsOnce reproducible TF-EMG latency and dorsal horn neurone controlreadings had been obtained, EA, consisting of pulses at 10 Hz, 1.0 msand an intensity which produced a slight toe twitch, was applied for 10mm. These parameters were also used whenever EA was applied insubsequent experiments. The TF-EMG latency and dorsal horn neuronalactivity were monitored for 20 mm following the cessation of EA. Inthe initial group of animals the above protocol was presented a secondtime to ensure the reproducibility of EA effects.A graph was constructed for each animal by averaging thespontaneous and evoked activity of 4 consecutive responses and plottingthe control value at time 0 and subsequent averaged values in the middleof the collection period. Similarly, the TF-EMG latericies were averagedand plotted at the same time points. The graphs illustrated depict theaveraged responses from six animals. This method was used inconstructing graphs of evoked activity in other experiments as well.Experiment 2. The Effects of Stimulating Acupuncture vs Non-AcupuncturePoints on the Inhibition of the Spinal Cord WDR Dorsal Horn Neurones.To determine whether EA-induced inhibition of WDR dorsal hornneurones was restricted to acupuncture points, EA electrodes werepositioned bilaterally in acupuncture points described above and about0.4 cm apart in non-acupuncture points in the gastrocnemius muscles. Ineach animal of this group the effects of EA applied to acupuncture41Methods and Materialspoints were determined on the nociceptor-evoked WDR dorsal horn neuronalactivity as described in Experiment 1. Following recovery, theexperiment was repeated except that an identical stimulus wasbilaterally applied to the non-acupuncture points.Experiment 3. The Types of Afferent Fibres Activated by EA.Following Experiment 2, the same animals were used with the EAelectrodes left in place in an attempt to determine what type of fibreswere activated by [A stimulation. Compound action potentials wererecorded through two electrodes placed 20 mm apart on the tibial nerve.A tight ligature was tied around the tibial nerve as high as possible toensure that the potentials recorded were travelling in the afferentdirection. The intensity of the EA stimulus (1.0 ms at 1.0 Hz) wasgradually adjusted upward. The stimulation current required to producethreshold responses for AB and A& fibres was noted and the potentialsaveraged. The potentials produced at the EA stimulus intensity usedpreviously in the animal were then averaged. Peristimulus histogram L.was used in this experiment (see Appendix 1). The conduction velocitiesof AB and A& fibres were determined from the latency differences whenthe potentials were recorded from the two sets of electrodes on thetibial nerve.42Methods and MateriaTsExperIment 4. Tests for a Supraspinal Involvement in EA-EvokedInhibition of Spinal Cord Nociceptive Transmission.In another group, once the effects of EA were initially determinedon TF-EMG and dorsal horn neuronal activity, the spinal cord was cold-blocked. When reproducible dorsal horn neurone and TF-EMG effects wereattained with the spinal cord blocked, the effects of EA were againdetermined to check for a supraspinal involvement. Additional controlexperiments were necessary to compensate for the baseline changesproduced by the cold-block. This involved conducting experiments asdescribed above except the noxious radiant heat intensity was readjustedafter the application of the cold block so that the evoked activity wascomparable to the control levels. Then the effects of EA were againtested.Experiment 5. Ascending and Descending Tracts in EA-Evoked Inhibition ofSpinal Cord Nociceptive Transmission.To determine whether the ascending and descending pathways of theEA supraspinal involvement were located in the DLF or VLT the followingsets of experiments were performed. The animals were prepared forrecording TF-EMG latency and from neurones in the DRN responding to anoxious peripheral stimulus applied at 2 mm intervals. Neurones in theDRN were chosen for this study since the nucleus is locatedsupraspinally and has been implicated in EA (G. Zhang, et al., 1986). A43Methods and Materialssearch for neurones took place in the DRN as described in the nextsection. Neurones were selected for the study if they showed anincrease in neuronal firing rate during noxious radiant heat applied tothe tail. Neurones used in this study turned out to be of the NCLvariety (see next section).In one group of animals the effects of EA were determined on theTF-EMG latency and DRN neuronal activity. An electrolytic lesion wasthen made in the left VLT of the spinal cord by passing a DC current(0.9 mA) for 30 seconds though a concentric bipolar electrode. Once theeffects of this lesion were determined on the [A-induced changes on theTF-EMG latency and DRN activity, a similar lesion was made in the VLT ofthe contralateral side. Again the effects of EA on the TF-EMG latencyand the DRN neurone activity were evaluated.In another group of animals, the same procedure was used exceptthat an electrolytic lesion was applied unilaterally and thenbilaterally to the DLF of the spinal cord.Experiment 6. Characteristics of DRN Neurones in Response to NaturalNoxious and Non-Noxious Peripheral Stimuli and [A.Rats used in these experiments were anaesthetized and rigidly fixed in astereotaxic headholder as described above. After making an incision inthe scalp and retracting the skin, a hole (3 mm diameter) was drilled in44Methods and Materialsthe midline 7.8 mm caudal to Bregma. Exposed tissue was covered withagar (4% in saline) to prevent drying and to dampen brain tissue motion.A carbon-fibre microelectrode was then directed into the DRN (Bregma -7.3 to -8.3 mm, -0.5 to + 0.5 mm of the midline, 6.0 to 7.0 mm ventralto cortical surface) according to the atlas of Paxinos and Watson (1982)to record single unit extracellular activity. As the electrode wasadvanced a silent period occurred which represented penetration of theaqueduct. Neurones recorded 0.1 - 0.3 mm below the silent area werecharacterized as DRN cells. DRN neurones were found to exhibit a slowregular CL or irregular NCL neuronal discharge pattern as reported byShima et al. (1986) in the cat. In each case they were characterizedaccording to their responses to natural non-noxious (touch or lightpressure on the extremities and the body) and noxious (pinching of thetail) stimuli. NCL neurones which did not response to peripheralstimuli were designated as NCLN neurones. Thus a survey of DRN neuroneswithin the DRN was made by inserting the microelectrode in a grid of 200ani intervals within the 1 mm2 above the DRN. EA was tested on severalneurones belonging to each cell category in the DRN.Experiment 7. Test of a Direct Projection from the DRN to the NPF.The animals in this group were prepared as described inExperiments6•Two microelectrodes were directed into the DRN and the45Methods and MaterialsNPF to record single unit extracellular activity of DRN and NPF neurones(as described in Experiment 8). Once an NPF and a CL, NCL or NCLN DRNneurone had been isolated, the NPF was electrically stimulated (0.2 ms,0.3 mA) through the recording electrode. If the DRN neurone projectsdirectly to the NPF it should be antidromically activated. Thestimulator was triggered by a spontaneous action potential in the DRNneurone. The stimulus output was delivered to the electrode in the NPFafter a preselected delay. The delay was varied to demonstratecollision in an antidromically activated neurone. Computer program L.was used in this experiment.Units were classified as being antidromically activated if theydisplayed constant latency at threshold, followed a three pulsestimulation of the NPF at a frequency greater than 100 Hz anddemonstrated collision between the NPF stimulus-evoked action potentialand a spontaneous action potential. Collision was considered to haveoccurred if the potential failed when an orthodromic action potentialpreceded it by less than twice the propagation time between thestimulating and recording electrodes plus the absolute refractory periodof the unit at the site of stimulation.46Methods and MaterialsExperiment 8. Characteristics of NPF Neurones Responding to NaturalNoxious and Non-Noxious Peripheral Stimuli and EA.Rats were prepared for these experiments in a similar manner tothat described for Experiment 6 except that a hole was drilled in themidline 4.3 mm caudal to Bregma. A microelectrode was then directed tothe NPF (Bregma -3.8 - 4.8 mm, lateral 0.5 - 1.5 mm, 5.5 - 6.5 mmventral to cortical surface) to record single unit extracellularactivity. Electrophysiologically, when the NPF was entered at the depthof 5.5 mm-6.0 mm there was generally a high incidence of spontaneousactivity. A single cell was isolated and its discharge pattern wasnoted. If the cell responded to a tail pinch it was monitored for 10mm to verify stability and then it was subjected to further study.Control data were obtained on the responses of the cell to noxiousradiant heat applied to the tail at 2 mm interval. LA was then appliedfor 10 mm. The spontaneous and noxious radiant heat-evoked activitywere monitored for 20 mm following the cessation of EA. This protocolwas presented a second time to ensure the reproducibility of EA effects.Another group of 4 rats was prepared as described in Experiment 7.A multibarrelled microiontophoretic electrode containing glutamate and0.95 % NaCl was directed into the NPF and a single microelectrode wasalso directed into the DRN. Single unit extracellular glutamate-drivenactivity was recorded in a NPF neurone. Spontaneous activity of a DRNneurone was recorded. EA was then administered while simultaneously47Methods and Materialsrecording the activity of these cells. Computer program S41 was used inthis experiment (Appendix 1).Experiment 9. Effects of DRN Stimulation on NPF Neuronal Activity andTF-EtIG.This group of animals was prepared for recording the TF-EMG andsingle units in the NPF and in the DRN as previously described inExperiments 1, 6 and 7. The DRN electrode was then switched from arecording to a stimulating electrode. Control responses were obtainedfrom a NPF neurone in response to noxious radiant heat applied to thetail at 2 mm intervals. After at least 8 responses, the DRN waselectrically stimulated for 30 s (square wave pulses, 10 Hz, 0.2 ms, 0.3mA) through the carbon-fibre microelectrode. This stimulus occurred 10s following the end of the last control noxious radiant heat pulse. Theactivity of the NPF neurone and the TF-EMG were monitored until recoverywas seen. The above protocol was presented a second time to ensure thereproducibility of the DRN stimulus effects. Computer program S4 wasused in this experiment.Another group of animals was tested for the effect of single pulseDRN stimulation on NPF neurones. Here a five-barrel microelectrodecontaining two barrels of NaCl and two barrels of glutamate was used torecord the activity of an NPF neurone whose background activity wassubstantially increased by the lontophoretic release of glutamate (45 -48Methods and Materials50 nA). The DRN was stimulated with a single pulse delivered to the DRN(0.2 ms, 0.6 mA) at 1.0 Hz. Peristimulus histograms were constructed(bin width 1 ms) comprised of 400 sweeps (S46, see Appendix 1).Additional experiments involving DRN stimulation are described inExperiment 10.Experiment 10. Tests for a Serotonergic Involvement in the AscendingPathway from the DRN to the NPF.The responses of NPF neurones to DRN stimulation or EA were testedusing the following agents: a 5-HT neurotoxin, 5,7-dihydroxytryptamineceatinine sulfate, (5,7-DHT; Sigma); 5-hydroxytryptamine creatininesufate (5-HT; Sigma); a 5-I-IT uptake inhibitor, alaproclate I-IC] (Astra)and a 5-HT antagonist, cyproheptadine HC1 (Merck Sharp and Dohme).1) Serotonin neurotoxin studyThis group of rats was treated with 5,7-DHT to destroyserotonergic neurones in the brain (Bjorklund et aL, 1974). Sincecatecholaminergic neurones can also be affected by 5,7-DHT, animals werepretreated with 25 mg/kg (i.p.) of desimipramine hydrochloride (Sigma)before administration of 5,7-DHT to block uptake into catecholaminergicneurones (Bjorklund et al., 1974). The rats were anaesthetized withhalothane and placed in a stereotaxic instrument. Two small holes weredrilled in the skull (0.9 mm posterior to Bregma and 2.0 mm lateral to49Methods and Materialsthe midline) and a 31-gauge cannula was lowered 3.5 miii from the surfaceof the cortex into the left lateral ventricle. Then an infusion pumpwas used to deliver 100 g of 5,7-DHT in 200 #1 saline containing 0.1%ascorbic acid (to prevent oxidation) over a 10 mm period. The cannulawas slowly removed and the hole was plugged with dental cement. Thesame procedure was repeated at the right lateral ventricle. The animalswere monitored for the behavioral signs of hyperaggresiveness,hyperactivity as well as weight loss which occur within 24 hours and isalways indicative of serotonergic neuronal destruction in the brain(Bjorklund et al., 1974). Dr. Steven Vincent (Department of Psychiatry,University of British Columbia), using the immunofluorescence method(Appendix 2), kindly examined the PAG region from two of the treatedanimals after 2 weeks of treatment and compared them to two of thecontrol animals. The animals were allowed 2 weeks for the neurotoxicaction to occur before the start of the electrophysiologicalexperiments.In one set of experiments, 2 - 3 weeks following 5,7-DI-ITtreatment, 6 animals were used for recording the activity of DRN and NPFneurones as described in Experiments 6 and 7, respectively. They werecharacterized in response to natural noxious and non-noxious stimuli andto EA.A separate group of six 5,7-DI-IT treated animals was prepared as inExperiments 8 and 9 to examine the effect of DRN stimulation on the50Methods and Materialsactivity of the NPF neurones and to test DRN neurones for antidromicactivation from the NPF.2) lontophoretic StudiesThe initial study was designed to examine the effect ofalaproclate on NPF neurones. The rats were surgically prepared forrecording of NPF neurones as previously described. A five-barrelmicroelectrode consisting of a central recording carbon-fibre electrodeand outer barrels filled with glutamate (0.1 M, pH 4.5, Sigma),alaproclate (0.1 M, pH 4.5) and two barrels filled with 0.95 NaC1 wasdirected into the NPF. This electrode Was connected to a Dagan 6400 SixChannel Micro-Tontophoresis Current Generator with automatic currentbalancing capabilities. One NaCl barrel was used for current balancing.A retaining current of 10 nA was applied to all drug barrels. A singleNPF neurone was isolated which responded to a noxious pinch of the tail.The experimental protocol was to excite the neurone under study byejecting pulses of glutamate for 10 s with 20 s intervals. Alaproclatewas then tested on the background activity by ejecting the drug atcurrents of 15, 30 and 45 nA for periods of 150 s. A current effect wastested by ejecting Na from the other NaCl barrel using a current of 45nA.In other experiments, the five-barrel electrode contained 5-HT(0.05 M, pH 5,) in addition to one barrel each of glutamate, alaproclate51Methods and MateriaTsand NaCL as described above. In preTiminary tests for each neurone acurrent was estabTished which when applied to the 5-HI barrel reducedthe glutamate-evoked activity to about 50% of control. The sameprotocol was then used in testing a number of cells. The procedureincluded pulsing glutamate for 10 s with 20 s intervals, applying 5-HIfor 260 s in which alaproclate was released concurrently for the final150 s. Following recovery, alaproclate at the same current was testedalone on the glutamate-evoked responses.In an attempt to obtain information on the specificity ofalaproclate, similar experiments were performed examining theinteraction between iontophoretically released GABA and alaproclate.Thus the five-barrel electrodes were.the same as described above exceptthat GABA (0.5 M, pH 4.5, Sigma) was substituted for 5-HT GABAreleased with a current which decreased the glutamate-evoked responsesto about 50% of control was applied for 150 s. After recovery,alaproclate was applied for 290 s with GABA again applied concurrentlyfor the final 150 s.In another set of experiments the effect of alaproclate wasexamined on the nociceptor-evoked activity of NPF neurones. Inaddition, the combined effects of alaproclate and DRN stimulation wastested on the nociceptor-evoked activity of these neurones. The animalswere prepared for NPF neuronal recording but, in addition, a stimulatingelectrode was positioned in the DRN as described in Experiment 8. The52Methods and Materialsfive-barrel electrode directed into the NPF contained two barrels ofglutamate and one each of alaproclate and NaCl. A cell was locatedwhich responded to noxious radiant heat (50°C) applied to the tail.After having established that alaproclate ejected at 30 nA did not alterglutamate-evoked responses in the cell, the drug at this current wastested on the nociceptor-evoked activity which was elicited at 90 sintervals. Here alaproclate was ejected for 240 s beginning immediatelyafter an evoked response. After recovery, the procedure was repeated toensure reproducibility. Similarly, a 1.0 s DRN stimulation (10 Hz, 0.2ms, 0.3 mA) was applied 1.0 s before the noxious radiant heat stimulusto the tail. Finally, DRN stimulation was tested during alaproclaterelease.Experiments were also performed to test the effects of DRNstimulation or EA on NPF neurones whose background activity wasmaintained by the constant iontophoretic release of glutamate. Theanimals were prepared as described immediately above but, in addition,EA electrodes were positioned as described in Experiment 1. Once an NPFneurone which responded to a noxious peripheral stimulus was isolated,glutamate was continuously ejected to provide a rapid stable backgroundfor testing the effects of DRN stimulation. The data were collected in10 s bins and visually displayed on a monitor during the course of anexperiment. The effects of DRN stimulation at 1, 2, 4 and 30 s wereexamined using the constant parameters of 10 Hz, 0.2 ms and 0.3 mA. The53Methods and Materialseffect of alaproclate (30 nA), which by itself was ineffective, wastested on a 4 s stimulus.If the recording condition was stable and the neurone was ‘held”sufficiently long, the effect of a 10 mm application of EA wasdetermined on the glutamate-driven cell. Following recovery the EAprocedure was repeated with concurrent iontophoretic release ofalaproclate (30 nA) to the ceTi.Another group of animals was prepared much the same as thatdescribed above. However the five-barrel electrode contained glutamate,alaproclate, GABA and NaC1. After testing DRN stimulation and theeffects of alaproclate on the 4 s stimulus, alaproclate was also testedon the inhibition produced by iontophoretic GABA. Cyproheptadine HC1, a5-HT antagonist, was then slowly administered intravenously in a dose of5.0 mg/kg. The procedure was repeated 5 mm after the cyproheptadinewas administered to determine effects of the drug on DRN stimulation andGABA-induçed inhibition.Finally, in two animals the above procedure was followed exceptthat after testing DRN stimulation and alaproclate on the 4 s DRNstimulus, EA was applied for 10 mm. Following recovery, cyproheptadine(5.0 mg/kg, i.v.) was administered and the procedure repeated.Computer program S41 was used in these experiments.54Methods and MaterialsV. HistologyIn the above experiments the locations of the recording and stimulatingsites as well as the magnitude of the spinal cord tract lesions weremarked by passing a direct negative current through the recording,stimulating (5 #A for 10 mm) or lesion electrodes (0.9 mA, 30 s). Theanimal was killed with an overdose of anaesthetic, the chest opened anda needle, attached to a 50 ml syringe filled with normal saline, wasintroduced into the left ventricle of the heart. The right ventrIclewas then slit and the animal perfused with saline (30 ml over 2 mm)followed by a 10% formaldehyde solution (100 ml over 15 mm). The brainand spinal cord were removed and stored in 10% formaldehyde solution forat least three days. The brain and the spinal cord were cut into 10x8x5mm and 3x2x2 mm sections, respectively, around the lesion and parallelto the tracks of the electrodes. The neuronal tissue was put on thecentre of a holding plate, surrounded with water and frozen at 200 C.The tissue was then cut into 50 #m slices with a freezing microtome(Damon/IEC Division). The slices were mounted onto glass slides usingan ethanol-gelatin solution (2 g gelatin in 2 liters of 40% ethanol).The tissue was stained with cresyl violet using conventional proceduresand viewed under magnification to determine the recording or stimulatingsites as well as the extent of the spinal cord lesions. Micrographswere obtained by taking pictures with a camera attached to a light55Methods and Materialsmicroscope. Black and white as well as color variable contrast film(400 iso) was used.56RESULTS[A on Spinal Cord Wide Dynamic Range (WDR) Dorsal Horn Neurones and theTail-Flick Latency (Experiment 1).Fig. 1 illustrates the inhibitory effect of [A on a WDR neurone recordedextracellularly and the tail-flick reflex. The control record (Fig. 1A)shows that noxious radiant heat to the tail produces a rapid increase indischarge rate which is arrested shortly after the tail-flick reflexwhich occurred after 4.5 s in this case. Immediately following EA theheat-evoked discharge is markedly decreased although the duration isprolonged as is the tail-flick latency (6.5 s; Fig. 1B). A gradualrecovery occurred within 30 mm (Fig. 1C, and D.). [A was found toreproducibly inhibit the nociceptor-driven activity of WOR dorsal hornneurones (Fig. 2A) and to produce a corresponding increase in the tail-flick latency (Fig. 2C). These effects outlasted the period of [Astimulation by several minutes. The spontaneous activity, which waslow, was not affected (Fig, 2B). The anatomical locations of theacupuncture points (Zusanli and Shangjuxu) in the anterior tibial muscleare shown in Fig. 2D.Since a change in the cutaneous temperature is known to alter the tailflick latency (serge, 1988; Tjolsen et al., 1988), the cutaneous tailtemperature was monitored by taping a thermistor to the root of the tail57ResultsAI II I I’ I IDU •11 II •‘i’ ii IFig. 1. EA inhibition of a spinal cord WDR neurone (top) and TF-EMG(middle). Each frame illustrates one sweep of the oscilloscope withthe lower pulse representing the 8.0 s duration of the heatapplication, A: Control. B, C and D were collected 10, 20 and 30 mm.after the onset of EA, respectively.L4.058Resultsr.J>.2.0L5I.00.5C)8CC.)C)>‘C)1)Fig. 2. EA reproducibly inhibits WDR dorsal horn neurones andprolongs the latency of the tail-flick reflex. A: Dorsal horn neuronenociceptor-evoked activity. B: Spontaneous activity of the same cellsimmediately prior to the noxious stimulus. C: Tail-flick reflexlatency. EA was applied during the periods indicated by the bars.Each point in this and subsequent graphs represents the mean ± SEand was compared to it s corresponding control value (*p<O.O5, n=6).D: Anatomical locations of the acupuncture points are shown on rightlimb.ADFemur*PatellaTibiaFibulaZusanliShangj uxu765432 1-10 0 10 20 30 40 50 60 70Time (miii)59Resultsin 4 animals. On no occasion did EA alter the cutaneous or coretemperature of the animals.The Effects of Stimulating Acupuncture vs Non-Acupuncture Points on theInhibition of the Spinal Cord WDR Dorsal Horn Neurones (Experiment 2).To determine whether EA-induced inhibition of WDR neurones in thespinal cord was restricted to acupuncture points, a group of 6 rats wastreated with EA as described previously and then, following recovery, anidentical stimulus was applied bilaterally to non-acupuncture points inthe gastrocnemius muscles in the same animal. EA stimulation of non-acupuncture points failed to alter the evoked WDR activity suggestingthat the effects of EA on WDR neurones are specific for acupuncturepoints (Fig. 3).The above experiments show that EA applied bilaterally for 10 mmin the rat produces a reproducible long-lasting inhibition of the spinalcord nociceptive transmission. This is reflected by an inhibition ofWDR dorsal horn neurones. Thus the evoked activity in these dorsal hornneurones following EA was inhibited even though the receptive field wasexposed to the noxious heat for a longer period of time. The samestimulation applied to non-acupuncture points failed to alternociceptive responses.60Resul tsA30120 1 1HHHLSg’ LO05— I I Il I I I I-10 0 10 20 30 40 50 60 70Time (mm)Fig. 3. The inhibition of WDR neurones is specific to acupuncturepoints. A: WDR nociceptor-evoked activity. B: Spontaneous activityin the same cells immediately prior to the noxious stimulus. Theheavy bar represents EA stimulation on “Zusanli” and “Shangjuxu”points in the anterior tibial muscle. The. thin bar denotes the samestimulation parameters applied to non—acupuncture points in thegastrocnemius muscle. Each point repreSenting the mean ± SE wascompared to it s corresponding control value (*p<OO5, n=6).61ResultsTypes of Afferent Fibres Activated by [A (Experiment 3).Following Experiment 2, the same animals with the EA needles stillin place were used in an attempt to determine what type of fibres wereactivated by EA stimulation. The compound action potentials on thetibial nerve elicited by stimulation through the EA electrodes at theactive sites are illustrated in Fig. 4. At intensities whicheffectively inhibited dorsal horn neurones, potentials representing infibres smaller than AB were not noted (Fig. 4B). A much higherstimulation intensity was required to produce an A6 potential (Fig. 4C).In the six animals tested, the mean stimulation intensities (±SE) whichelicited threshold AB responses, which were effective in producing EAand which elicited threshold A6 potentials were 0.88 (0.24), 0.92 (0.02)and 4.20 (0.10) mA, respectively. The mean conduction velocities (±SE)of the fibres producing the AB and A6 potentials in these experimentswere 41.7 (2.0) rn/s and 22.1 (1.7) m/s, respectively. Table 1 lists thelatencies of compound action potentials recorded from the two sets ofelectrodes placed 20 mm apart on the tibial nerve. Table 1 alsoincludes the required currents to produce AB and A6 potentials in eachexperiment. The set a” electrodes were placed on the tibial nervecloser to the [A stimulation sites than set “b” electrodes.Accordingly, the latency recorded at set hau electrodes was alwayssmaller than the latency recorded at set “b” electrodes.62ResultsAC/1/DI!I1.0 msFig. 4. Compound action potentials recorded from the tihial nerveupon EA stimulation. Each trace is an average of 40 sweeps. A: Aresponse just above threshold at a stimulus intensity of 0.88 mA. B:The Af3 potential recorded at a intensity which was effective inproducing EA (0.92 mA). C: The appearance of an A potential at astimulus intensity of 42 mA. D: As in C hut at a stimulus intensity of4.6 mA. The square wave pu’se indicates the EA pulse duration of1.0 ins (illustration of potentials reported in file 3-26, set a in Table1 .)63Resul tsTai)t C I . I .a(enCieS a 11(1 types o Coin O1J 11(1 Act ott I ot cit a IsRecorded from Two Sets of Electrodes at Various Current Inteiisites.File: 3-Ill Latency Current Potential File: 3-19 Latcncy T Current Potential(ma) (mA) (nis) I (mA)3-Ill-I 0.88 Aflr 3-19-I (1.90Set 3-18-2 2.60 0.93 All Set 3-19-2 2.88 0.92 Alla 3-18-3 3.60 A&r a 3-19-3 3.80 AS1-3- 18-4 3.78 3.90 AS 3-19-4 3.48 3.96 AS3-IS-S Afl-r 3-19-5 All-rSet 3- 18-6 3.08 A Set 3-19-6 3.38 Allb 3-18-7 AS1- b 3-19-7 AS-1-3-18-8 4.70 AS — 3-19-8 4.6(1 ASAl = 20 mrn/(3.08 ms - 2.60 tnt) = 41.7 rn/s A = 20 mrn/(3.38 ms - 2.88 sos) = 36.4 misAS = 20 mm/(470 ms - 3.78 ms) = 21.7 rn/s AS = 20 mm/(4.60 ms - 3.48 ms) = 17.9 rn/SFile: 3-25 Latency Current Potential File: 3-26 Latency Current Potential(ms) (mA) = (ms) (mA)3-25-1 0.88 Al3-i- 3-26-1 0.96 Afr1-Set 3-25-2 2.68 0.92 All Set 3-26-2 27S 1.00 Alla 3-25-3 4.20 AS-1- a 3-26-3 3.80 AS-1-3-25-4 360 4.60 AS 3-26-4 3.78 4.20 AS3-25-5 Al5 — 3-26-5 AI3TSet 3-25-6 3.20 A Set 3-26-6 3.20 Allb 3-25-7 AS-1- b 3-26-7 AS1-3-25-8 4.40 AS — 3-26-8 4.92 ASA3 = 20 mm/(3.20 ms - 2.68 ms) = 38.5 rn/a — All = 20 mm/(3.20 ms - 278 ms) = 47.6 rn/sAS = 20 mm/(4.40 ma - 3.60 ms) = 25.0 m/s AS = 20 mm/(492 ms - 3.78 ma) = 17.5 rn/sFile: 3-27 Latency Current Potential File: 4-8 Latency Current Potential(rns) (mA)=(ms) (mA)3-27-1 0.75 All-t- 4-8-1 0.89Set 3-27-2 2.08 0.84 Al3 Set 4-8-2 2.08 0.94 Alla 3-27-3 3.80 AS-1 a 4-8-3 420 A&r3-27-4 2.78 4.00 AS 4-8-4 2.68 4.40 AS3-27-5 A3- 4-8-5 All-i-Set 3-27-6 2.50 AfI Set 4-8-6 2.60 Allb 3-27-7 AS-1- b 4-8-7 AS1-3-27-8 3.70 AS 4-8-8 338 ASA = 20 mm/(250 ms - 2.08 ma) = 41.6 rn/a — A = 20 mm/(2.60 ms - 208 ms) = 38.5 rn/sAS = 20 mm/(3.70 ms - 2.78 ms) = 21.7 mIs AS = 20 mrn/(338 ma - 268 sot) = 286 nt/sAwr: All potential recorded at a thrcattold current.All: All potential recorded at intensity used to cvokc EA effect.A8t: AS potential recorded at a threshold current.AS: AS potent iii recorded at a auprathrcshold current.64ResultsA Supraspinal Involvement in [A-Evoked Inhibition of the Spinal CordNociceptive Transmission (Experiment 4).To determine whether there is a supraspinal involvement in the EAinduced attenuation of nociceptor-evoked responses, the effects of EAwere determined during a cold-block of the spinal cord. Cumulativehistograms of the responses of a spinal cord WDR neurone to EA beforeand after a cold-block of the spinal cord are shown in Fig. 5. Notethat EA reduced the nociceptor-evoked activity of the cell (Fig. 5A-D)as was previously shown in Fig. 2. The application of a cold-blockresulted in a marked increase in the spontaneous and evoked activity(Fig. 5E). EA applied during the cold block failed to alter theseresponses (Fig. 5F). The grouped data for six animals treated in thismanner are plotted in Fig. 6. Note that there was a reciprocalrelationship between the activity of the dorsal horn neurones and thetail-flick reflex latency (Fig. 6C).Since the failure of EA to have an effect in the presence of thecold-block could have been due to the baseline changes produced by thecold-block, additional control experiments were performed as depicted inFig. 7. Here the protocol was the same as that above except that oncethe effects on the dorsal horn neurone and tail-flick latency weredetermined in the presence of a cold-block, the temperature of thenoxious radiant heat applied to the tail was reduced 2 - 3°C so that the65ResultsFig. 5. The effects of EA and cold-block on the responses of a spinalcord WDR neurone to noxious radiant heat applied to the tail. Eachhistogram is comprised of 4 sweeps with each bin notch indicatingthe cumulative counts per sweep (bin width 0.5 s). The bars abovethe histograms begin with the onset of the heat pulse (SOOC) and endwith the tail—flick reflex. A: Control. 13—D: 10, 20 and 30 mm afterinitialing EA. E: During cold—block of the spinal cord. I”: 10 mm afteri fl! tiati ng FA and d un fl2 cold—block of the Spinal cordA BTI TI aih lit IiillIllhIltTltlluIII_....__DL4-j0c-)020(‘-34)0C-)20(‘-34)00C—)0C——--—-- l.llifTi1j1IhiBI! TIifllfhlldIi..FI II,_ ,Ll llrTT; I ITT!!...,_..,.——,...,——_I._—,...,_....—,-10 0 10 20 S -10 0 10 20 s66A 10Results6050N‘4°3020’B’°NI’l.00.5C7()N 43Time (mm)Fig. 6. Cold-blocking the spinal cord eliminates the effects of EA. Thearrangement is the same as in Fig. 2. The period of cold-block isi nclicaiecl by the thin bar above the time scale. Each pointrepresenting the mean ± SE was compared to it s correspondingcontrol value (*p<oOS **p<o.ool , n6).**67A 70Results6050>‘40a)a.a)a)302010Bx>.. 1.5Li..0.53C7’(JC/)‘ 5’aa)a) 4’3.2-10 0 10 20 30 40 50 60 70Time (mm)Fig. 7. Cold-blocking the spinal cord eliminates the effects of BA evenafter baseline adjustment. As in Fig. 6 except that the intensity ofthe heat lamp is reduced after determining the effects of cold-blockin order to readjust the baseline to equal that in the absence of cold—block (* p<0.05, ** p < 0.001, n6).** ***68Resultsresponses were comparable to control values in the absence of the cold-block. However EA applied at this point also failed to alter the dorsalhorn neuronal activity or the tail-flick latency.Thus, we found that the spinal cord effects of [A were eliminatedif conduction in the spinal cord was blocked rostral to the recordingsite suggesting the irlvolv?ment of a supraspinal loop in the mediationof the spinal cord inhibition.Ascending and Descending Tracts in EA-Evoked Inhibition of Spinal CordNociceptive Transmission (Experiment 5).The effects of unilateral and bilateral lesions of the spinal cordVLT on EA-induced changes in dorsal raphe neuronal activity and thetail-flick latency are illustrated in Fig. 8. EA reproducibly increaseddorsal raphe NCL neurones nociceptor-evoked activity (Fig. 8A), thespontaneous activity of the same cells immediately prior to the noxiousstimulus (Fig. 8B) and the tail-flick latency (Fig. 8C). A unilaterallesion of the VLT failed to alter the above effects of EA. A bilaterallesion, however, blocked all of these effects. Also, nociceptiveresponses of these neurones were eliminated after bilateral lesions ofthe VLT.The results of using the same protocol but making unilateral andbilateral lesions of the DLF rather than the VLT are depicted in Fig. 9.Unilateral lesions failed to alter either the dorsal raphe neuronalfacilitation or the increase in tail-flick latency following EA.6962I I I I I0 20 40 60 80Time (miii)-I — I I100 120 140Fig. 8. Bilateral lesions of the VLT block the effects of EA on dorsalraphe neuronal nociceptor-evoked activity and the tail-flick reflex.A: Nociceptor-evoked activity of DRN cells. B: Spontaneous activity ofthe same cells. C: The tail-flick reflex. EA was applied during timesindicated by the horizontal bars. The first and second vertical barsindicate unilateral and bilateral lesions of the VLT, respectively.Each point representing the mean ± SE was compared to it scorresponding control value (*p<O.05, n=6).4*Results6* *A>t)UUCI,4*87.64.— m_-2070Results?BFig. 9. Bilateral lesions of the DLF block the effect of EA on the tail-flick reflex (C) but do riot alter the enhancement by EA onnociceptor-cvoked activity in dorsal raphe neurones (A). Thearrangements is the same as in Fig. 8 except that the lesions weremade in the I)LF ( p<O.O5. n=6 ).6’A**.16-7.876** *I_7-20 0 20 40 60 80 100 120 140Time (mm)71ResultsBilateral lesions did not alter the effects of EA on nociceptor-evokedand spontaneous activity of dorsal raphe neurones (Fig. 9A and B) butresulted in a decrease in the tail-flick reflex latency (Fig. 9C). EAdid not further influence the tail-flick latency.The size of the lesions in individual animals are sketched in Fig.1OA and 13. Fig. bC shows a micrograph of a lesion in the spinal cordVLT.The Characteristics of the DRN Neurones in Response to NaturalPeripheral Stimuli and EA (Experiment 6).Neurones encountered in the DRN tended to have a slow firing rateand either a very regular or irregular firing pattern. These wereinitially designated as CL or NCL neurones. It soon became apparentthat CL neurones never responded to natural peripheral stimuli but someNCL neurones could be activated (Fig. 11). The neurones were thenclassified as CL, NCL if they responded to peripheral stimuli and NCLNif they failed to respond. A total of 118 neurones were examined in 24animals. These were subdivided into 35 CL, 32 NCL and 51 NCLN neurones.The distribution of the three different types of neurones in the DRN areshown in the three dimensional graphs (Fig. 12). They appear to berandomly distributed. The mean (±SE) spontaneous discharge rates of CL,NCL and NCLN neurones were found to be 1.7 (0.02), 1.4 (0.06) and 1.6(0.05) Hz, respectively.72AResul tsldI1 * -/ -:-... - -(-.;:‘- ?.‘“14: -:.-‘-P—.4S..Fig. 10. The exient of lesions noted in individual animals are shownfor the DLF (A) and VLT (B). C: An example of lesions in the spinalcord VLT.S.C-4473ResultsFig. 11. Oscilloscope traces illustrating the effect of noxious heat on aclock-like (A) and non-clock-like (B) neurone. The pulse in the lowertrace of each pair represents the 8.0 s duration of the heat pulse(50°C) applied to the tail. The tail-flick reflex latency occurred at4.5 s in A and 4.7 s in B.741225Resul tsB55000525Fig. 12. Three dimensional representation of DRN CL (A). NCL (B) andNCLN neurones (C). The axes labelled vertical, posterior, and lateralshow the three stcreotaxic planes of the brain in pin. The posterioraxis indicates distance caudal to I3regma. 0 on the lateral axisrepresents the midline, and the vertical axis indicates the depth fromthe surface of the cortex.75ResultsThe noxious radiant heat applied to the tail produced a relativelysmall but consistent increase in the discharge rate of NCL neurones.Fig. 13 is an example of one such cell. Comparing the evoked activityto spontaneous activity in Fig. 14, one can see that the noxiousstimulus almost doubled the discharge rate. This increase began shortlyafter initiating the noxious stimulus, was maintained during thestimulus and continued for several seconds after completion of thestimulus. This point is not obvious in Fig. 13. Here a tail-flick,although not monitored, would have occurred prior to the end of the heatpulse thus removing the receptive field from the heat source. CLneurones were not activated by noxious radiant heat to the tail as shownin Fig. 15. NCLN neurones, by definition, also did not respond.The majority of the 15 NCL neurones tested were facilitated by EAwith an approximate increase of 1 Hz. Only three cells showed noresponse to [A. None of the eleven CL and only 2 of the NCLN neuronestested responded to [A (Table 2). One NCLN was facilitated and one wasinhibited of the 13 tested. The effects of EA on NCL and CL DRNneurones are illustrated in Figs. 14 and 15. Note the facilitation ofNCL neurones lasted 10 - 20 mm.Test of a Direct Projection from the DRN to the NPF (Experiment 7).Twenty-one NCL neurones of the DRN were tested for antidromicactivation upon NPF stimulation. They all were activated and displayed76ResultsA Bc4z I0c.. L___________________________C D____________________-10 0 10 20 S -10 0 10 20 sFig. 13. EA enhances the nociceptor-evoked activity of a DRNneurone. Noxious radiant heat (500C) indicated by the bar on thetime scale was applied to the tail. Each histogram is comprised of4 sweeps with each bin notch indicating the cumulative counts persweep (bin width 0.5 s). A: Control. B-I): 10. 20 and 30 mm afterinitiating EA.77Resu’tsA40.3.5-3.000zo- 1.5-.1)D 2.5I) * *2.0-1.5-1.0--10Time (mm)l’ig. 14. A: The eilect of EA (bars) on the noxious radiant heat—evoked activity in non-clock-like DRN neurones. B: Spontaneousactivity of the same neurones immediately before the applications ofnOXiOUS radiant heat. Each p01111 representing the mean ± SE wascompared to it s corresponding - control value (*p<0.O5. n1 5 ).***ô ib 2’O 3b 4’O 5b 6b 1078ResultsA302.52.01.5ti)) I)2.01.5-1.0- I I I I-10 0 10 20 30 40 50 60 70Time (miii)Fig. IS. A: The effect of EA (bars) on the nOxiOus radiant heat—evoked activity in clock-like DRN neurones. 13: Spontaneous activityof the same neurones immediately before the applications ol flOXIOUSradiant heat. Each point representing the mean ± SE (n= I I).79ResultsTable 2 Characteristics of DRN Neurones*Non-Clock-Likc-NcuroncClock-Likc-Ncuroncs Non-Clock-Likc-Ncuronc Not Responding to Noxious(CL) (NCL) or Non-noxious Stimuli(NCLN)Facilitated by 0/1 1 1 2/ I 5 / 1 SEAEnhibited by 0/11 0/15 1/13EANo Response to 11/11 3/15 1 1/13EANPF Antidromic 0/1 2 21/21 0/16Activation* Number of neurones responding / number of neurones tested.80Resultsa constant but wide range latency (0.6 ms to 9.0 ms), followed a threepulse train greater than 100 Hz and demonstrated collision betweenspontaneous and evoked action potentials (Fig. 16, Table 2). Theconduction distance between the DRN and the NPF was calculated to be 3.5mm and the conduction velocity of these neurones was found to range from0.40 - 5.8 rn/s with a mean of 1.85 rn/s. None of the CL or NCLN neuronescould be añtidrornically driven (Table 2).Thus it seems that only the NCL neurones of the DRN play a role inEA. These neurones are activated by peripheral natural stimuli,facilitated by EA and project to the NPF.The Characteristics of NPF Neurones in Response to Natural PeripheralStimuli and EA (Experiment 8).When the electrode entered the NPF at a depth of 6 mm below thesurface of the cortex there was generally a high incidence ofspontaneous activity. The discharge pattern could be categorized intothree types: slow firing at a constant rate which varied from 0.8 to 8Hz, bursting neurones (2 - 3 spike bursts every 1 - 2 s) with an overalldischarge rate of 3-10 Hz and, finally, fast firing neurones. Theseneurones were not studied in detail but they seemed to be essentiallythe same as described by Andersen and Dafny (1983a). Most of the slowtype cells as well as a small percentage of the bursting cells could beactivated by a noxious stimulus to the tail. The fast firing81ResultsAlmVSnisAI1•11Jimv- SinsFig. 16. An example of a non-clock-like DRN neurone exhibiting highfrequency following and collision. A: The rieurone follows a three-pulse train applied in the NPF at 200 Hz. The large deflections arestimulus artifacts. B: An example of collision in the same neurone.The stimulator was triggered by spontaneous action potentials (left)in a DRN neurone. The stimulus output (middle large artifact) wasdelivered to the electrode in the NPF after a preselected delay.Varying the delay demonstrates collision in third and fourth traces.82Resultsvariety, however, never responded to a noxious peripheral stimulus. Afew slow firing neurones in the NPF responded to non-noxious mechanicalstimuli but the majority did not. The neurones in the NPF were selectedfor further study only if they responded to noxious radiant heat appliedto the tail.An example of a slow active NPF neurone responding to noxiousradiant heat applied to the tail is shown in Fig. 17. The spontaneousactivity of this neurone was 1.0 Hz. The increased discharge startedshortly after the onset of the heat pulse, continued throughout thestimulation and for several seconds after the completion of thestimulus. The activation of these neurones by noxious radiant heat tothe tail was much greater than that which occurred in DRN neurones(compare the spontaneous to evoked activity in Fig. 19 to that in Fig.14).Fig. 18 shows an example of the EA inhibition of the nociceptorevoked activity of a NPF neurone. Each histogram is comprised of 4sweeps with each bin notch indicating the cumulative counts per sweep.•EA was found to inhibit this evoked activity in NPF neurones (Fig. 19)with a duration of action similar to that for the facilitation of NCLDRN neurones (Fig. 14). The spontaneous activity of the same NPFneurones was also inhibited by EA (Fig. 19B).Fig. 20 shows that EA produces a reciprocal increase in activityof a NCL DRN neurone and a decrease in the firing rate of a glutamatedriven NPF neurone. The average latency for the onset of these effects83ResultsFig. 17. Oscilloscope traces illustrating the spontaneous andnociceptor-evoked activity of an NPF neurone. Lower pulserepresenting the 8.0 s duration of the heat (500C) appliedto the tail. The tail-flick reflex occurred at 5.2 s.84Results111 iJ-,._______________________0flI’hllL—t_______________________________0 10 20sFig. 18. EA inhibits the nociceptor-evoked activity of a NPF neurone.Noxious radiant heat (500C) indicated by the bar on the time scalewas applied to the tail. Each histogram is comprised of 4 sweepswith each bin notch indicating the cumulative counts per sweep (binwidth 0.5 s). A: Control. B—D: 10, 20 and 3() mm after initiating EA.A B140C)014C,)ll!MLC-10 0 10 20s -1085ResultsA30.20C) -z4.3-2-1— I I I I I I-10 0 10 20 30 40 50 60 70Time (mm)Fig. 19. A: The effects of LA (bars) on the noxious radiant heat—evoked activity in NP[ neurones. B: Spontaneous 1C1i Vity of thesame neurones immediately before the application o noxious radiantheat. Each point representing the mean ± SE was compared to it sCorreSpOnding COntrol value (*p<0.05. n= I 9).86ResultsMIIIIIIiIiiIliInhIilIllI[5 mmB IiibFig. 20. The effect of EA on the activity of a NPF and DRN neuronerecorded simultaneously. Each pair of traces (A-D) are from aseparate experiment. The upper trace in each experiment is aglutamate-driven NPF neurone and the lower trace is aspontaneously active NCL cell in the DRN. The bin width is 10 s ineach case. EA was applied for 10 mm with the onset at thedownward arrow and the offset at the upward arrow.87Resultswas 6.2 mm (n=4). Interestingly, the onset and offset of these changeswas usually quite sudden and sometimes the changes appeared to occur insteps. A recovery occurred 20 - 30 mm after initiating EA.The Effects of DRN Stimulation on NPF Neurones and the TF-EMG(Experiment 9).Electrical stimulation (10 Hz, 0.2 rns, 0.3 mA, 30 s) of the ORNhad a marked inhibitory effect upon nociceptive-evoked discharges andspontaneous activity on NPF neurones as well as concomitantly elevatingthe latency of the tail-flick reflex (Fig. 21). The DRN stimulationdecreased the nociceptive responses on NPF neurones to about 50 % ofcontrol. The latency of TF-EMG increased from 5.0 (±0.5) s to 6.8(±0.6) s. The duration of inhibition was approximately 20 mm.Stimulation of the DRN using trains of 1, 2 and 4 s durations werealso studied on glutamate-driven NPF neurones. The data for theseexperiments are described under “Jontophoretic Studies”.The effect of single pulse DRN stimulation (1.0 Hz, 0.2 ms, 0.6mA) on glutamate-driven NPF neurones also showed an inhibitory responsein four of the six neurones tested. The latencies of the inhibitionranged from 6 to 12 ms (Fig. 22) with a mean latency of 8 ms. Thus,with a calcUlated conduction distance of 3.5 mm the range of conductionvelocities is 0.3 to 5.8 rn/s with a mean of 0.43 rn/s. This was asimilar range to that found by antidromic activation of ORN neurones88NC)zC)>,01)cFig. 21. The effects of the DRN stimulus on the noxious radiant heat-evoked activity in NPF neurones and the TF-EMG latency. A: Noxiousheat-evoked activity of NPF neurones. B: The spontaneous activityof the same cells immediately prior to the noxious stimulus. C: Thetail-flick latency. The arrows indicate electrical stimulation of theDRN (10 Hz, 0.2 ms, 0.3 mA) for 30 s. Each point representing themean ± SE was compared to it s corresponding control value (*p<O.O5,n= 8).ResultsA30.2010B54.3.28C7.654,* ** *3.—-10+I I I I0 10 20 30 40 50 60 70 80Time (mm)89ResultsiliJiiIIIiili lIihiIiiii!iih h!iiiiIih1hiIiI5 ins120CountsHIIhIII!hh Ik... 111diiiWiHiffliH Wiiiii1ifl iIii,,iIiIIhFig. 22. Peristimulus histograms of 6 glutamate-driven NPF neuronestested by a single pulse (1 .0 I-Iz, 0.2 ms, 0.6 mA) stimulation in theDRN. The vertical lines indicate the point of stimulation. Bin width= 1 ms. Each histogram is comprised of 400 sweeps.ilihilihitil IhllIlI,IllitIIhIlI1lI liIIh I,I 1111111 iii ii90Resultsfrom the NPF. Although two of these neurones showed no inhibition tosingle pulse DRN stimulation, a 1.0 s train of pulses (10 Hz, 0.2 ms,0.3 mA) did produce an inhibition of 50 - 62 s on these neurones.Serotonergic Involvement in the Inhibitory Projection from the DRN tothe NPF (Experiment 10).A group of 6 rats was treated with bilateral injections of 5,7-DHT, a specific serotonin neurotoxin (Bjorklund et al., 1974), into thelateral ventricles. Within 24 hours of the injection, all six animalsshowed behavioral signs of hyperaggressiveness and hyperactivity. Twoweeks after the 5,7-DHT treatment the animals showed a mean weight lossof 6.8 gin compared to a weight gain of 72.2 gm for control animals overthe same time period (Table 3). The two 5,7-DHT treated animal brainstested using the immunofluorescence technique showed a marked reductionof 5-HT immunofluorescence in the DRN compared to the controls (Fig.23).1) Studies on 5,7-DHT Pretreated AnimalsNPF neurones were also tested in animals treated wi-th 5,7-IJHT aspreviously described. The effects of noxious radiant heat on NPFneurones in these rats are shown in Fig. 24 and should be compared toidentical experiments in non-pretreateci animals shown in Fig. 19. Notethat the noxious heat-evoked response in the pretreated animals was910 r\.)NT=nottested.Table3.ChangesinWeight,BehaviouralActivityandImmunofluorescenceReactionin5,7-DHTTreatedRatsCD (I) -a (I,57.1)1IiIruatadGroup[CornlGroupWeighiHvperaggrcssiveoesstiomuoo-WeighiHvpeuuggressivciiessmmmuAnirneNo.GainorfluorescenceAnimalNo.GaininorTwoWeeksHvpcractivisvReactionTwoWeeksHvper,lel\ilyReeiurnMeanginMeangin1-2YesNoISINoYS2-I SYesNT265NoNT3-3YesNT369NoYes52YesNT453NoNT5-I 3YesNT)545NoNT63YesNT654NoNTMean-6.5Mean72.2/I..‘...‘-?\L-- ‘.Fig. 23. An illustration of serotonergic immunofluorescence in theDRN region of a control rat (top) and a 5,7-DHT treated animal(bottom).ResultsS -• 4.1’-N93Resul tsC.)ii)‘a)Fig. 24.neuronesanimals.prior tomean ±(n=6).A: EA (bars) fails to inhibit the evoked activity of NPFin response to noxious radiant heat in 5,7-DHT treatedB: Spontaneous activity of the same neurones immediatelynoxious radiant heat applications. Each point represents theSE and was compared to it s corresponding control valueA100.80607.65,4 a a a a a a-10 0 10 20 30 40 50 60 70Time (mm)94Resultssignificantly higher (mean: 72.6 Hz ) compared to the non-pretreatedgroup (mean: 25.3 Hz) even though the heat pulse was the sametemperature in the two groups. The spontaneous activity of NPF neuronesin 5,7-DHT treated animals was also greater than that in non-pretreatedanimals. [A, in the 5,7-DHT pretreated animals, failed to produce aninhibition of NPF neurones previously seen in the non-pretreatedanimals. In fact there was a tendency (though not significant) for [Ato enhance the activity of these neurones (Fig. 24).Similarly, stimulating the DRN in 5,7-DI-IT pretreated animals atthe same parameters previously used failed to elicit inhibition of NPFneurones (Fig. 25A) or prolong the tail-flick reflex (Fig. 25C).Compare these results to those obtained in identical experiments. in nonpretreated animals (Fig. 21). Indeed, in the pretreated animals, DRNstimulation produced an increased firing rate in spontaneous andnociceptor-evoked activity of NPF neurones.It is interesting that only NCLN neurones were found in the DRN of5,7-DHT treated animals. Twenty-five of these neurones were found insix 5,7-DI-IT pretreated animals compared to 51 in 24 non-treated animals.The mean frequency of discharge was 1.6 Hz which did not differ from thecontrols. There were 35 CL and 32 NCL neurones in non-treated animals,but none were found in 5,7-DHT treated animals.95Resultsx>-‘C-)1)0C-)0)*1 00 -80-60-1510-5.C.ó ib o 3’O zl’o 5’O 6’0 7’O 8bTime min)Fig. 25. Stimulation of the DRN for 30 s (arrows, 10 I-lz, 0.2 ms, 0.3mA) fails to inhibit (lie noxious radiant heat—evoLed activity in NPFneurones in 5,7—1)1-IT treated animals (A). Spontaneous activity of thesame cells ininiediately prior to the noxious activity (B). The tail—flick latency (C). Each point representing the mean ± SE wascompared it s corresponding control value ( p < 0.05. n=6 ).96Results2) lontophoretic Studieslontophoretic studies were also used to test the idea that 5-I-IT isthe mediator of the inhibition in the NPF produced by DRN stimulation.Alaproclate, a specific 5-HT uptake inhibitor, iontophoretically appliedto NPF neurones which were activated by 10 s pulses of glutamate reducedthe discharge rate at an ejection current of 45 nA (n=7) but not at 15(n=5) or 30 nA (n=7; Fig. 26 and 27; Table 4 ). Tests for currenteffects by iontophoretically applied Na+ at 45 nA from the 0.95 % NaCLfilled barrel, failed to alter the activity of these neurones (n=3,Fig.26; Table 4). 5-HT lontophoretically applied to glutamate-activatedNPF neurones was inhibitory. Fig. 28 illustrates the mean frequency of12 NPF neurones in which the protocol was identical. Preliminary testsin each experiment were done to establish currents to use for each drug.5-HT was applied at a current (usually 30 nA) which reduced theglutamate-evoked activity to about 50 % of control. After severalresponses, alaproclate concomitantly applied at 30 nA, which had beenpreviously shown in these neurones not to alter the discharge rate, wasfound to enhance the inhibitory response (Fig. 28A). Alaprociate wastested alone and failed to alter the glutamate-evoked responses. Alsoalaproclate did not alter the inhibitory effect of GABA (30 - 45 nA) onthese neurones (Fig. 288). An example of the effect of lontophoretic97ResultsALP 3OnA ALP 3OnA ALP 45nA Na+ 45nA30HzL1JL1LRRflJLRRJHLftftJUULRRJUL1L1LflJULRJLrLJLLJ30sFig. 26. An example of the effect of iontophoreticapplication ofalaproclate a 30 and 45 nA on a NPF neurone.Na+ ejected by acurrent of 45 nA applied though the NaCI (0.95%) barrel hadnoeffect. The lower trace represents 10 s tontophoreuc applications ofglutamate. The bars above the (Op trace represent the duralion ofioniophorciic applications of alaproclate andNa+.98Resu’tsNC)30•25’20’15’10’5’0’ L3015 45Cunent (nA)Fig. 27. The effects of iontophoresis of alaproclate on NPF neuronesactivated by glutamate. The open and filled bars represent thedischarge frequency before and during the iontophoresis ofalaproclate at the current specified below, respectively. Each barrepresenting the mean ± SE was compared LO it’s correspondingcontrol value(*p<O.05, n=5-7)99-4 C CTab’e4.lontophoreticApplicationof AlaproclateandNa+onNPFNeuronesActivatedbyGlutamateAveragefrequencybefore/duringiontophorcsis(spike/s)*p<0.05.C,)C,)NumbcrAlaproclateNa+ofNeuronc5nA30nA45nA.15nANo.TesisFILN0.Tesls(1’No.TessI.No.icisf36.6/16.0315.99/18.9136.67/8.542617.18/18.06517.09/16.95615.56/11.473121.15/29.13224.13/23.49422,51/13.644228.08/28,47228.64/29.37227.63/12.07228.48/28.05521 7.90/I7.7021 6.90/18.60119.00/I4,706324.16/25.07224.48/18.63223.43/25.777335.95/35.65238.00/27.33238.30/40.70Mean20.18/21.4723.27/24.2223,41‘/f5.2031)07/31.51(±SE)(2.1)(2.6)(2.8)(2.4)(2.2)(2.4)(4.4)(4.7)>Fig. 28. lontophoretic application of alaproclate enhances theinhibition produced by 5-HT hut not GABi\ on NPF neurones. A: Thefiring rate of NPF neurones activated by glutamate (10 s pulses at 45nA) was reduced by 5-HT appi ication. A greater inhibition wasobserved when alaproclate (30 nA) was applied concurrently. B:Alaproclate did not enhance the inhibition elfect of GABA (45 nA).The period ol application of 5—FIT, alaproclate and GA BA arcindicated by labelled bars. Each point represents the mean ± SE(*1)<Q()5 n 12).101AALP5-HT3025201510•Resultsu 1Time (mm)GABAALPB30’25’>-.C.,C)15’10’I I I I I I I IIu 1 2 3 4 5 6 7Time (mm)ResultsALPA 5-HT1f1i:P9fl1•1 .[IñR!lfl!j7J 20Hz120 sDALP1) GABAp .11 .:‘rrpirii r r1nr;,Fig. 29. An example of the effect of iontophoretic application of5-1-IT, alaproclate and GABA on a NPF neurone. The firing rate of anNPE neurone driven by glutamate was reduced by 5—MT application.A greater inhibition was observed when alaproclate was appliedconcurrently (A). Alaproclate (11(1 not enhance the inhibitory effectof GABA (B). The bars above the trace represent the duration ofion toph ore tic application.102ResultsALP + 43025T TC-,V20Lz 10C-:V 5u 1 ‘2 ‘ 4 S 6 7 ‘10 H 12Time (mm)Fig. 30. The effect of iontophoretically applied alaproclate on noxiousradiant heat—evoked activation of NPF neurones and inhibition byDRN stimulation. Noxious radiant heat-evoked firing of NPF neuroneswas decreased during the application of alaproclate (30 nA, barsabove trace) and further decreased by a combination of alaproclateand DRN stimulation (arrows, 1 0 I-lz, 0.2 ms, 0.3 mA) for 1 .0 s. Eachpoint represents the mean ± SE and was compared to i 1scorresponding control value (*p<O.O5, n= I 2).103Resultsalaproclate application on 5-HT-and GABA-mediated inhibition of an NPFneurone is shown in Fig. 29.Activation of NPF neurones, illustrated in Fig. 30, by noxiousradiant heat to the tail was decreased by the application of alaproclate(30 nA). A combination of alaproclate and DRN stimulation (10 Hz, 0.2ms, 0.3 mA) produced an even greater inhibition of these neurones. Anexample of the effect of iontophoretic application of alaproclate andDRN stimulation on nociceptor-evoked activity in an NPF neurone is shownin Fig. 31.Stimulation (10 Hz, 0.2 ms, 0.3 mA) of the DRN for 1, 2, 4 sproduced a prolonged inhibition in the activity of NPF neurones whosebackground activity was maintained by the constant iontophoretic releaseof glutamate. The inhibitory response had a sudden onset, whichsometimes took several seconds to reach maximum effect, long durationand sudden offset. An example from one experiment is illustrated inFig. 32 A, B. The mean (±SE) inhibitory durations were 51.2 (±7.9),107.7 (±8.8) and 186.4 (±15.3) s after 1, 2 and 4 s stimulation,respectively (Fig. 32, 33 and Table 5). Thus, the duration of theinhibition was correlated with the duration of stimulation. However,the magnitude of the inhibition was the same regardless of the period ofstimulation (Fig. 32, 33, 34 and Table 5).Alaproclate iontophoretically applied to NPF neurones at 30 nA hadno effect on the firing rate of these neurones before the stimulation ofthe DRN. However, a greater inhibitory response in magnitude was104ResultsALP +I I I_LjStj20 Hz120sFig. 31. An example of the effect of iontophoretically appliedalaproclate and DRN stimulation on noxious radiant heat-evokedactivation of an NPF neurone. Noxious radiant heat-evoked firing ofthe NPF neurone was decreased during the application of alaproclate(30 nA, bars above trace) and further decreased by a combination ofalaproclate and DRN stimulation (arrows, 10 Hz, 0.2 ms, 0.3 mA) for1.0 s.I105ResultsABCDEAIFig. 32. The effects of alaproclate on DRN- and EA-evoked inhibitionof a glutamate-driven NPF neurone. The traces are continuous andshow the discharge frequency of an NPF cell (bin width =10 s). Themarks above the trace indicate the points and durations of DRNstimuli. The bars above the trace indicate the periods ofioritophoretic alaproclate application at 30 nA. EA was applied for 10mi n with the onset at the downward arrow and the of Iset at theupward arrow. A and B show the effect of different periods of DRNstimulation and the effect of alaproclate on the inhibition. C and Dsun i larly show the inhibition produced by EA and its enhancementby alaproclatc.is 2sI I IIL5 mi nI I I IEAI106Results25•—s 20’o 15’10’5’0’4+ALP 30Train Duration (S)Fig. 33. The duration of the NPF inhibition produced by DRNstimulation of 1, 2. 4, and 30 s. The 4+i\LP bar represents a 4 sstimulus combined with iontophoresis of alaproclate at 30 nA. Eachbar represents the mean ± SE and was compared to its correspondingcontrol value (n=4 — I I).1 2 4107Table5.TheEffectof DRNStimulationonNPFNeuronesDrivenbyGlutamateNeuroncIS2S4S—4S+ALP3OnA130STesteditEdjln[I_n_Itdn{I,_J11532.0/26.847.0332.0/24.693.3231.0/20.5140.0211 7.3/10.750.1)224.0/13.0100.0123.2/I3.4220,03328.5/24.360.0329.2/23.5103.3227.5/18,2170.04436.7/26.984.3332.2/21.3151.3235.8/19.3285.0534.7/13.4342.0133,7/30.62.65530.5/25.647.8531.6/20.527.5329.5/20.6136.7230.5/)1.6145.06733.1/20.258.21239.2/30.4109.6341.8/34.4I 62.5641.3/14.1208.37928.6/24.558.9630.5/22.6135.0332.5/23,5203.3432.0)13.61 88.0127.9/21.326.08232.5/20.850.0232.5/26.9110.0232.5/24.5210.0232.5/12.2215.0132.2/20.219.39424.7/15.390.0325.1/16.4136.7525.4/16.7243.3224.8/11.5285.0224.0/10.216.710239.0/30.46(1239.2/26.5160.0239,4/19.6I 70.0II239.7/26.7120.0243,0/10.4160.0—29,3/21.751.231.5/23.0107.632.6/22.2186.434,8/13,3[214.229.5/18.)Mean(1.9)(1.9)(7.9)(1.6)(1.8)(8.8)(1.8)(1.7)(15.3)(2.2)(1.0)(23.8(2.2)(2.7)(0.ltSE)—n=9n=10n=11n=8—n=4o=Numhroftests;Averagefrequencybefore/afterDRNstimulation(spikc/s);d=Durationofinhibition(a.);d=Durationofinhibition(stilt.).(tt -IResults403530N2520151050Train Duration (S)Fig. 34. The magnitude of the NPF inhibilion produced by DRNstimulation. The open and filled bars represent the NFP neuronesfrequency of discharge before and after DRN stimulation,respectively. The 4±ALP bar represents a s stimulus combinedwith iontophoresis of alaproclate at 30 nA. Each bar represents themean ± SE. In each case the responses after stimulation weresignificantly greater than the corresponding control values.Alaproclate enhanced the effect of a s stimulus (n=4 - I I , *p<0O5, ).The data are also expressed in Table 5.1 2 4 4+ALP 30109Resultsobserved when alaproclate ejected at this current was combined with a 4s stimulation of the DRN (Fig. 32, 34 and Table 5). However, there wasno difference in duration of the inhibition (Fig. 32, 33 and Table 5).EA on the same neurone illustrated in Fig. 32C and 0 produced aprolonged inhibition which was clearly reached in 2 steps. Repeatingthe procedure in the presence of iontophoretically released alaproclateat a current of 30 nA markedly increased the magnitude of the inhibitionwhile the duration of inhibition was almost identical. The data on the5 neurones tested with alaproclate in this manner are illustrated inFig. 35A and B and Table 6.Another study was performed using cyproheptadine, a serotoninantagonist, to determine whether the ascending pathway from the DRN tothe NPF is serotonergic. Fig. 36 shows that stimulation of the DRN (10Hz, 0.2 ms, 0.3 mA) for 1, 2 and 4 s evoked inhibition of increasingduration on a glutamate-driven NPF neurone. Alaproclate,lontophoretically applied (30 nA), enhanced the inhibition evoked by a 4s stimulation of the DRN. However alaproclate did not alter theinhibitory effect of GABA on this neurone. After the administration ofcyproheptadine (5 mg/kg, i.v.), the inhibitory effect of DRN stimulationusing the same parameters was blocked. Also the spontaneous activity ofthis neurone was increased significantly compared to the control.However, cyproheptadine did not alter the inhibition produced by GABA.The grouped data for 5 neurones tested in this manner are shown in Fig.37 and Table 7. in which the spontaneous activity of the NPF neurones110N00Hg. 35. Alaproclate lontophoretically applied at 30 nA enhances themagnitude (A) but not the duration of inhibition (B) of EA-evokedinhibition of NPE neurones. Each bar representing the mean ± SE wascompared to its corresponding control value (*p<O.05, n=5).ResultsI IA40.35’30’25’20’15’10’B 40’35’30’25’20’15’10’5’C•i-(I*-r5’0’EA EA+ALP0’EA EA+ALP111rzTable6.TheEffectofAlaproclateonofGlutamate-DrivenNPFfAveragefrequencyofBAcontrolgroup,f=AveragefrequencyofBA+alaproclate,L=latency(mm),d=durationofinhibition(mm),*=p<0.05whichwascomparedtocontrolvalueI.n=5.EA-inducedInhibitionNeuronesCD C,) —S,NeuroneEAEA+Alaproclate3OnATestedf’(Ldf”[Ld134.83/26.964.63134.33/14.525.230239.59/29.835.32842.96/13.646.329339.66/26.955.03039.61/18.445.831432.24/24.266.32632.57/13.026.728535.02/23.285.82935.26/12.236,425McaiiJ36.27/26.265.4(0.30)28.8(0.9)*36.94/14.376.08(0.59)28.6(1.0)(±SE)(1.45)(1.05)(1.90)C1.09)is 2s 4sResults30 HzFig. 36. Cyproheptadine (5 mg/kg. i.v.) blocks the DRN-evoked inhibitionof a glutamate—driven NPF’ neurone. The traces are continuous and eachbin equals 10 s. The marks above the trace indicate the points andduranons ol a 1)RN siinulus. The long bars above the trace indicate theperiods of ion[ophorctic alaproclate application at 30 nA. The heavyshort bars above the trace indicate the application of iontophoretic GABAat 30 nA. The arrow represents the adminisiratioii oh cyproheptadinc.I Iis 2s+4s 120 S113AResultsNNC-)C0z01-4B 40’TT1 I30’25’NC)C 20’$22 15’10’5’0 — — — — — — pis 2s 4s 4s+ALP GABA GABA+ALPTreatmentFig. 37. Cyproheptadine blocks the DRN-evoked inhibition ofglutamate—driven NEP neurones. The arrangement is the same asshown by Fig. 34. A: control, B: alter administration ofcyproheptadine (5 mg/kg. i .v. ) Each bar represents the mean ± SE(n 5).is 2s 4s 4s+ALP GABA GABA+ALPTreatment114Results- - -I-C-C-C-C-C<_— -(3 (3C — — — — — —.C-= •0 C = = C 0 CC—•vCC fl —-- -—— :_- ->0--) -r.= -C -=, ÷vu:-,_-_ CC C) - -:H ==ii u fC -----v=0=0=00=00- CCC- -00-D- -00‘5’Or. 0000I “iii’-- _____z C115Resultswas increased after administration of cyproheptadine. Alsocyproheptadine blocks [A-evoked inhibition of a glutamate-driven NPFneurone (Fig. 38).116Results4I:io 38. Cyprohepiadine blocks LA —evoked i nh i hi Lieu oh a glutamate—dr yen NI)!: neuronc.20Hz120 SI I117DISCUSSION[A would seem to activate many systems involving a number ofneurotransmitters. Evidence for this is that antagonists for thefollowing neurotransmitters have been shown to block EA: GABA andglycine (McLennan et al., 1977), substance P (Romita et al., 1992),acetylcholjne and catecholamines (Han, 1986). This study focuses on apathway involving of 5-HT.This study shows that EA applied bilaterally for 10 mm in the ratproduces a reproducible long-lasting inhibition of nociceptivetransmission in the spinal cord . This is reflected by an inhibition ofWDR dorsal horn neurones and a prolongation of the tail-flick reflexlatency which parallel each other in time. Thus the nociceptor-evokedactivity in these dorsal horn neurones was inhibited following EA eventhough the receptive field was exposed to the noxious heat for a longerperiod of time.The data from this study may be compared to a similar studyreported by Pomeranz and Cheng (1979) in the anaesthetized cat. Theintensity of the [A stimulus appears to be comparable in the two studiesalthough Pomeranz and Cheng (1979) stimulated for a period of 30 mmcompared to 10 mm in the present study. I found the magnitude of thenociceptor-evoked activity of WDR spinal cord neurones to be depressedto approximately 50% of the control value whereas Pomeranz and Cheng(1979) reported a mean depression of 16%. They speculated that thisminimal EA effect was due to poor acupuncture point placement of the118Discussionneedles or to excessive depth of chloraiose anaesthesia. It is a commonfinding clinically that complete failure of EA analgesia occurs if theneedles are inappropriately placed. Another possibility for thedifference seen in the two studies is the difference in EA points used:Pomeranz and Cheng (1979) used Futo (Stomach 32) and Yangling(Gallbladder 34) ipsilaterally, whereas I used Zusanli (stomach 36) andShangjuxu (stomach 38) bilaterally. Pomeranz and Cheng (1979) foundthat the depression of these WDR neurones took 20 mm of [A stimulationto reach the peak and 20 mm to wear off. The protocol used in thepresent study did not permit a determination of the onset of the [Ainduced spinal cord inhibition although NPF neurones were inhibitedafter 4 - 8 mm of stimulation. Similarly, the WDR neuronal activityreturned to control levels within about 20 mm of EA offset. Anothersimilarity between the two studies is that [A was found to be effectiveat acupuncture points and ineffective at non-acupuncture points.Wu et al. (1986), in chloralose anaesthetized cats, applied [A at5 Hz for 15-20 mm at various points. Although they demonstrated goodinhibition of spinal cord WDR neurones during [A, the effect waseliminated almost immediately following the cessation of stimulation.This is surprising since they seemed to use a rather intense stimuluswhich they described as “generally below 3 mA”. Their tests for asupraspinal involvement on only 2 neurones were inconclusive.119DiscussionThus it is clear that the results of the present study much moreclosely resemble those reported by Pomeranz and Cheng (1979) than thosedescribed by Wu et al. (1986). In particular, the prolonged time courseof the inhibition following EA was comparable in this study to that ofPomeranz and Cheng (1979). It is also consistent with prolonged effectsobserved in other reported animal and human studies (McLennan et al.,1977; Yee, 1973).Opinions differ as to whether effective EA can be elicited byactivating only AB fibres or whether it is necessary to involve A6 andC-fibres as well. The present results clearly favour the former inagreement with Liu et al. (1986), Pomeranz and Paley (1979) and Toda andIchioka (1978). All these investigators used a relatively mild EAstimulus which was just sufficient to produce muscular contraction. Onthe other hand, at least in experimental animals, it seems clear thathigh intensity stimulation, activating A6 and C-fibres, does produce agreater inhibition (Bing et al., 1990; Chung et al., 1984a,b). Bing etal. (1990) argue, and the evidence supports the notion, that theacupuncture stimulus activates an inhibitory system known as “diffusenoxious inhibitory controls” (DNIC, LeBars et al., 1975, 1976, 1979a,b,1980, 1986; Villanueva, et aL, 1986a,b). Bing et al. (1990) used amanual form of acupuncture in which a needle was inserted at the Zusanlipoint to a depth of 0.5 - 1.0 cm and lifted, thrusted and rotated in aclockwise and anticlockwise fashion. They found a similar magnitude and120Discussiontime-course of the antinociceptive effect when an adjacent non-acupuncture point was stimulated. Thus, whether effective acupunctureis specific to acupuncture points would seem to depend on the intensityof the stimulus applied. However, one could also argue that, due to theclose proximity of the non-acupuncture point to the acupuncture point inthe Bing et al. (1990) study, it is quite possible that the intensemanual stimulus at the non-acupuncture point also stimulated theacupuncture point.Bing et al. (1990) felt that their stimulus was noxious andactivated AB and C-fibres. It is also important to note that the onsetof the inhibition in the studies by Bing et al. (1990) and Chung et al.(1984a,b) occurred almost immediately after the onset of the stimulus.With milder forms of acupuncture stimuli there tends to be a delay ofseveral minutes to reach peak effects (Chan and Fung, 1975; Chapman etal., 1980; Cheng and Pomeranz, 1980; McLennan, et al., 1977; Pomeranzand Cheng, 1979; Takeshige, et al., 1981, 1985), Bao et al. (1989)found C-fibre activation was not important in acupuncture analgesia butwas for DNIC. The stimulation parameters used for clinical acupuncturevary widely but I feel it is safe to say that in the great majority ofcases, and in my personal practices the intensity of the stimulus is notconsidered noxious to the patient.I found that the spinal cord effects of EA were eliminated ifconduction in the spinal cord was blocked suggesting the involvement of121Discussiona supraspinal loop in the mediation of the spinal cord inhibition.These effects are consistent with the results of Pomeranz et al. (1977and 1979) and Shen, et al. (1975) in the cat.Cold-blocking the spinal cord produced a decrease in the latencyOf the TF-EMG and enhanced the nociceptor-evoked activity of dorsal hornneurones. This is consistent with the previous finding of Necker andHellori (1978) and Sinclair et al, (1988) and is thought to be theconsequence of blocking the tonic descending inhibition in the spinalcord. However, this enhanced responsiveness in the cold-blocked statewould not seem to be responsible for blocking the effects of EA since EAwas also ineffective when the responses were reduced by decreasing theintensity of the noxious stimulus.Since the EA-induced increase in NCL dorsal raphe neurones was notblocked by bilateral lesions of the DLF but was eliminated by bilaterallesions of the VLT, it would seem that the ascending limb of thesupraspinal loop was mediated by the VLT. This is in agreement withacupuncture studies by Li, et al. (1983) on spinal cord lesionedpatients. On the other hand, the DLF would appear to be involved in thedescending inhibition since the effects of EA on dorsal horn neuronesand the tail-flick reflex were blocked after bilateral DLF lesions.Bilateral lesions of the ascending or descending pathways were necessaryto block the EA-induced spinal cord inhibition. This was not unexpectedsince EA was also applied bilaterally. My findings on the location of122Discuss ionthe ascending and descending pathways are in agreement with theconclusion reached by Shen, et al. (1975) in studies on cats.The above experiments were difficult in that movement induced bythe onset of the current often resulted in the “loss” of the cell. Thisoccurred in about 40% of the cells studied and the data were discardedin such cases. However, due to the rigid fixing of the animal in thestereotaxic headholder and spinal frame and perhaps due to the use ofcarbon-fibre recording electrodes, I was successful in many cases.The findings obtained in my work in urethane anaesthetized ratsare in agreement with most investigators in that CL DRN neurones areunresponsive to natural peripheral stimuli (Nakahama, et al., 1981;Shima, et al., 1986, 1987). Aghajanian et al. (1978) found that lowintensity stimulation of the sciatic nerve produced a transientinhibition of these CL cells. They found that the response rapidlyadapted with a higher frequency of stimulation which perhaps partiallyaccounts for the failure to observe the effect upon natural stimulation.The slow rate of discharge by these cells also makes it difficult toobserve an inhibition. However, although inconsistent with my work,Sanders et al. (1980) found many DRN units were inhibited by noxiousstimuli in chioralose or pentobarbital anaethetized rats.Agha,janian and his group did not concentrate on the NCL DRNneurones although they reported that some exhibited an excitation to a123Discussionsciatic nerve stimulus which would not be in the noxious range(Aghajanian et al., 1978; Aghajanian and Haigler, 1974; Haigler, 1976).Interestingly, the findings in the work reported here are verysimilar to that reported by Shima and his colleagues in theanaesthetized (Nakahama, et al., 1981) or conscious cat (Shima, et al.,1986). A common finding was that CL neurones were not responsive tonatural stimuli and about half of the irregularly discharging cells wereresponsive to noxious stimuli. They termed the latter NCL neurones.Generally the cells responding to a noxious stimulus also responded to anon-noxious stimulus. Therefore, perhaps it is not surprising that thegreat majority (12/15) of NCL neurones tested in my study werefacilitated by EA. The magnitude of the facilitation was not great,approximately doubling the normally slow discharge rate of the cell.The onset of this facilitation occurred only after a few mm of EAstimulation.The finding that only NCL neurones responded to EA and noxiousstimuli is quite different from the report by the Beijing group (1986).They reported that approximately half of the regularly firing neurones,or in my terminology, the CL neurones, were facilitated by EA. Inaddition, they found that only 2 of the 14 neurones facilitated by EAwere activated by noxious stimuli.Although the present work was not concerned with investigatingwhether NCL DRN neurones are involved in modulating spinal cord sensory124Discuss iontransmission, there is ample evidence for the involvement of DRN indescending systems. One of the major relay sites in this descendingprojection is the nucleus raphe magnus (NRM). Interestingly, the NRMhas been implicated in EA. Liu et al. (1986) reported that EA appliedto the ‘7usanli” point activated NRM neurones and inhibited spinal cordnociceptive responses in dorsal horn neurones. Du and Chao (1976) foundthat a lesion in the medulla which included the NRM significantlydec’eased the inhibitory effect of EA on the viscero-somatic reflex.Therefore it seems quite possible that the NCL neurones in the DRN whichare activated by EA project to brainstem nuclei such as the NRM which,in turn, project to the spinal cord to inhibit nociceptive transmission.Literature reports clearly show that this projection is predominantly inthe DLF. Thus, the finding that the effects of EA on the spinal cordare blocked by DLF lesions fits with the involvement of this descendingsystem in EA.It seems likely that neurones ascending in the SMT participate inthe activation of NCL DRN neurones by noxious stimuli and EA. I haveshown that this activation takes place after lesions in the DLF but isabolished by lesions in the VLT where SMT neurones project. Aspreviously mentioned SMT neurones project directly to the PAG.As mentioned above the DRN of the PAG has been implicated as theupper tier in a descending antinociceptive system (Basbaum and Fields,1979). However the DRN clearly has rostral projections to many areas125Discussionthat have also been implicated in nociception or antinociception. Mywork examined the connection between the DRN and the NPF of t.he medialthalamus and the influences of EA on this pathway. NPF neurorresexhibiting the characteristics described by Andersen and Dafny (1983a)were relatively easy to locate and monitor for long periods of time.Cells were selected for study only if they were excited by noxiousradiant heat applied to the tail. As was the case with Andersen andDafny (1983a) and Benabid et al. (1983), I found a number of NPFneurones responded to a noxious stimulus with an inhibition. Theseneurones, however, were not further examined. While histology revealedthat the recording locations appeared to be within the NPF, theboundaries of this structure are not clearly defined and it is possiblethat some of the cells were in an adjacent medial thalamic nucleus. Inany case the cells under study exhibited a rather vigorous response tonoxious radiant heat of the tail; much more so than the NCL neurones ofthe DRN. Neurones in the STT are likely the major source of excitatoryinput to the NPF upon a noxious peripheral stimulus. Other pathways mayalso indirectly activate the NPF.EA consistently inhibited this nociceptor-evoked activity. Instudies where the background activity of the cells was increased withiontophoretic glutamate release, the onset of inhibition ranged from 4 -8 mm after initiating EA. There appeared to be a reciprocalrelationship between the EA-induced inhibition of these NPF neurones and126Discussionthe excitation of NCL neurories in the DRN suggesting that the activationof the DRN neurones was responsible for the NPF inhibition. The longdelay in the onset of these effects would seem to rule out theactivation of a neuronal pathway to account for the findings. Perhapsprolonged EA stimulation results in the release of a chemical insufficient concentration to produce these effects. In this regard anumber of investigators have found EA to release endogenous opioids(Cheng et al., 1979; Han, et al., 1986; He et al., 1979; He and Dong,1983; He, 1987; Sjolund et al., 1977; Zhang, A. et al., 1986; Zhou, eta1, 1982; Zou, et al., 1986;). Indeed, Shima et al., (1987) found theopiate, morphine, activated NCL DRN neurones. Moss et al. (1981) alsoreported that serotonin—containing neurones in DRN include numerousleucine-enkephlin immunoreactive cells. The finding by several groupsthat the opioid antagonist, naloxone, antagonizes the effects of EA isalso consistent with this idea (Han, et al., 1986; Mayer, et al., 1976,1977; Pomeranz, 1977; Pomeranz and Cheng, 1979; Zhou.et al., 1986).At first glance, it is difficult to reconcile with this proposalthe finding that the NPF inhibition evoked by LA is often abrupt andsometimes appears in steps (eg. Fig 32 C and D). One would expect amore gradual onset and offset if the effect was due to a build up of achemical substance. However, stimulation of the DRN alsG produces aprolonged inhibition of NPF neurones that is correlated with theduration of stimulation (Figs. 32 A,B; 33). Significantly, the onset127Discussionand offset of this inhibition is also abrupt. It is not known whetherstimulation of the DRN produces a prolonged discharge of the activatedcells which then suddenly stop. If this does not occur, the stimulusmust produce a sustained state of activation, either within the NPF orin structures which impinge upon the NPF, where the net effect isinhibition.Dafny and colleagues previously reported that DRN stimulationinfluences the activity of NPF neurones. In an early study theyreported that DRN stimulation activated NPF neurones with a latency of20-25 ms (McClung and Dafny, 1980). Later, Andersen and Dafny (1983a,b)reported an inhibitory effect. However there are differences betweentheir findings and mine. For example, I found that DRN stimulationproduced only inhibition in NPF neurones, whereas, Andersen and Dafny(1983a) found some neurones to be facilitated and that the higher thestimulus intensity the greater the likelihood of producing excitation(Andersen and Dafny, 1983b). They speculated that the excitation wasdue to current spread to the adjacent reticular formation. They used abipolar concentric stimulating electrode with a 500 urn tip separationwhereas I used a monopolar carbon-fibre electrode (8 urn diam, < 15 urnlength) as the stimulating electrode. Another difference is that theyappeared to produce a rather short period of inhibition in NPF neuronesupon DRN stimulation. For example, Fig.8 in Andersen and Dafny (1983b)illustrates an inhibition of less than a 4 s using stimulation128Discussionparameters of a 2.0 s train, 20 Hz, 0.2 ms pulse width and 0.5 mAintensity. I found in glutamate-driven NPF neurones that a 2.0 sstimulus (10 Hz, 1.0 ms, 0.3 mA) produced a mean inhibition of 107 s.In any case the inhibitory effect would appear to be mediated byserotonin. The evidence in my study in support of this is that theinhibitory effect of EA or DRN stimulation on NPF neurones is eliminatedin animals treated with the neurotoxin, 5,7-DHT, or the 5-HT antagonist,cyproheptadine. Also, the inhibitory effect is enhanced by theiontophoretic release of the specific 5-HT neuronal uptake blocker,alaproclate, at the recording sites. Finally, 5-HT iontophoreticallyapplied to NPF neurones is inhibitory.I am confident that the 5,7-DHT treatment was very effective indestroying 5-HT neurones in the DRN. The animals used all exhibited thereported behaviour of 5-HT depletion by 5,7-DHT, namely hyperactivity,hyperaggressiveness and weight loss. In addition, the histochemicalexamination of two animals confirmed the marked 5-HT depletion.The conduction velocity of the DRN neurones projecting to the NPFranged from 0.4 - 5.8 m/s with a mean of 1.85 m/s. Thus, these neuroneswould seem to have similar axonal diameters as the CL neurones whichhave a similar conduction velocity of 1 m/s and are therefore consideredunmyelinated (Wang and Aghajanian, 1977a,b).Alaproclate produced effects consistent with the blockade of 5-NTuptake. It increased the magnitude of the inhibition on NPF neurones by129Discussioniontophoretic 5-HT, IJRN stimulation and EA. It is interesting thatthese inhibitions were enhanced using currents which produced no changesin the background activity. This is perhaps surprising since there isevidence (presented later) that the DRN serotonergic pathway to the NPFis tonically active. However, it must be remembered that the spacialdistribution of alaproclate release differs from that of synapticallyreleased 5-MT. Perhaps the presumably increased 5-I-IT release upon DRNstimulation or EA allowed for an interaction to occur betweenalaproclate and endogenous 5-MT. It would appear that the effect ofalaproclate is specific for 5-HT in that the release at the samecurrents which enhanced the activity of 5-HT failed to alter theinhibitory effect of GABA.Similarly, cyproheptadine, which was very effective in eliminatingthe effect of DRN stimulation or EA on NPF neurones, did not alter theinhibition produced by GABA.Dafny and colleagues have also provided evidence that theprojection from the DRN to the NPF involves 5-MT. Andersen and Dafny(1983b) also found that 5,7-Dill pretreatment enhanced the nociceptorevoked activity of NPF neurones and blocked the DRN-induced inhibition.In addition, Andersen and Dafny (1982) reported that iontophoresis of 5-MT onto NPF neurones was inhibitory. More recently they showed thationtophoretic release of glutamate, morphine or 5-MT into the DRN had aninhibitory effect on nociceptor-evoked activity in NPF neurones. They130Discussionconclude that neurones which are activated by DRN stimulation andinhibit NPF neurones originate in the DRN (and are not axons en passage)and contain opioid and 5-NT receptors (Dafny et al., 1990).Clearly the inhibition produced on NPF neurones by EA and ORNstimulation resemble each other very closely. They both exhibit asudden onset, prolonged duration, sudden offset and respond in anidentical manner to drugs which would be expected to alter 5-FIT neuronaltransmission. They differ in that DRN stimulation produces a rapidinhibition (found to be 6-12 ms in single pulse studies), whereas, EAinduced inhibition is delayed. Therefore, EA must elicit other effectswhich, in turn, activate DRN neurones to inhibit NPF cells.The serotonergic-mediated inhibition of NPF neurones would appearto be tonically active. This is evident from the finding that 5,7-DHTpretreatment results in an increase in the spontaneous activity of NPFneurones and a marked increase to noxious stimuli. In addition,systemically administered cyproheptadine increases the backgroundactivity of these neurones. Finally, alaproclate, appliediontophoretically with high currents, inhibits the activity of thesecells.It would seem that a serotonin-mediated inhibitory pathway is thepredominant pathway from the DRN to the NPF. Pretreating the animalswith 5,7-DHT completely eliminates the inhibition and, in fact, convertsthe inhibition to a facilitation. Thus, there must also be an131Discussionexcitatory pathway from the DRN to the NPF that is normally masked undermy experimental conditions.It is also clear that the NCL neurones of the DRN must be theneurones responsible for the inhibition of NPF neurones. All of the NCLneurones tested were shown electrophysiologically to have a directprojection to the NPF, whereas none of the other DRN neurones testedprojected to this site, The probability of finding neurones which didproject was likely increased by positioning the stimulating electrode inthe NPF at a site where an NPF neurone could be shown to be inhibited byDRN stimulation, and secondly, using a relatively larger stimulatingcurrent (0.3 mA).Since the evidence is good that the pathway is serotonergicallymediated and NCL neurones in the DRN are the projecting neurones, itfollows that there are neurones other than CL neurones in the DRN whichare a serotonergic. However, there seems to be no question that CLneurones are also serotonergic since I was unable to locate theseneurones in 5,7-DHT pretreated rats. This was also the case in earlierwork by Aghajanian (1972) and Andersen and Dafny (1983a) in 5,7-DHT andPCPA pretreated rats, respectively.A recent report by Reichling and Basbaum (1991) is particularlypertinent to the present study. In a double-labeling study using tworetrograde tracers they examined the projections from the rat PAG to thenucleus raphe magnus (NRM) and several forebrain structures. One of132Discussionthese forebrain structures was the medial thalamus with the injectionbeing centred in the NPF. Neurones labeled in the PAG were found in theventrolateral, lateral, dorsal and ventral regions including the DRN.Approximately 75% of the retrogradely labeled neurones were locatedipsilateral to the injection site. Interestingly, about 20 % of the PAGneurones labeled from the medial thalamus also projected to the NRM.Thus, it is possible that the same neurones modulate nociceptivetransmission by both descending and ascending projections. In supportof this idea, any treatment that influenced the nociceptor-drivenneurones in the NPF influenced spinal cord WDR neurones in a likemanner. For example, EA and DRN stimulation inhibit both. Similarly,the effects of stimulating the DRN on the NPF and spinal cord were bothblocked in 5,7-DHT treated animals.It could be argued that the inhibition in the NPF upon DRNstimulation in this study is simply due to activation of a descendinginhibitory system which inhibits nociceptive transmission at the spinalcord level which, in turn, results in a reduced response at the NPF.Although I did not examine this possibility, Qiao and Dafny (1988) andDafny et al. (1990) found the DRN to NPF inhibition still existed aftersectioning the dorsal half the spinal cord.A summary of the findings of this study are depicted schematicallyin Fig. 39. A noxious stimulus activates SIT neurones and perhapsothers which are excitatory to the NPF in the medial thalamus. The same1330 i scuss ion/NPFFig. 39. Schematic arrangement of neurones activated by EA andanoxious stimulus. Filled circle indicates an inhibitory neurone.DRN10EA WDR134Discussionstimulus probably activate SMT neurones which excite NCL neurones in theDRN. EA activates large diameter peripheral fibres. The activatedneurones in the spinal cord project supraspinally through the VLT, anddirectly or indirectly activate NCL neurones in the DRN. These neuronesare serotonergic in nature, project directly to the NPF and exert atonic inhibitory effect on nociceptive NPF neurones. These DRN neuronesalso likely project to brainstem nuclei, such as the NRM, which areinvolved in a descending inhibitory projection to spinal cordnociceptive transmission via the DLF. Thus a supraspinal loop isinvolved in the EA suppression of nociceptive transmission in the spinalcord.The most important discovery from this work is the identificationof a group of neurones in the DRN that participate in the modulation ofnociceptor-driven activity in the medial thalamus. These are neuronesthat exhibit irregular background activity, respond to naturalperipheral stimuli and have been termed NCL neurones. The evidence isthat these NCL DRN neurones are serotonergic, whereas, it was previouslybelieved that only the CL neurones were serotonergic. There isconsiderable evidence in the literature that serotonergic neurones inthe DRN participate in the modulation of nociception and therefore itwas naturally assumed that the CL neurones were responsible. However, Ifound no evidence of CL neurones responding to noxious stimuli or EA. I135Disctissionbelieve it is the NCL serotonergic neurones which are an important linkin these inhibitory modulation system on riociceptive pathways.Secondly, all the evidence from this work points to an importantaspect of the antinociceptive activity of FA being the excitation of NCLDRN neurones. EA was found to be dependent on a serotonergic system andactivated NCL but not CL rteurones. It would also appear that the NCLneurones influence both descending and ascending system to modulatenociceptive pathways. Any treatement which altered NCL neuronalactivity produced a similar effect on the EA-induced inhibition in thespinal cord and NPF.136CONCLUSIONS1. Electroacupuncture (EA) stimulation at sites on Zusanii” (st. 36)and‘tShangjuxu” (St. 38) for 10 mm (10 Hz, 1.0 ms) was found toproduce a long-lasting inhibition of wide dynamic range (WDR)spinal cord dorsal horn neurones and to prolong the latency of thetail-flick reflex in the lightly anaesthetized rat.2. Stimulation of acupuncture points “Zusanhi’ and “Shangjuxu” producedinhibition of WDR neurones. The same stimulation parametersapplied to non-acupuncture points in the gastrocnemius muscleproduced no inhibition.3. This inhibition was effectively produced at an EA stimulationintensity which did not activate fibres smaller than A13.4. There is a supraspinal involvement in the EA-induced inhibition ofspinal cord nociceptive transmission.5. The ascending arm of the supraspinal loop was found to be in theventrolateral tract (VLT) and the descending arm is located in thedorsolateral funiculi (D[F).6. Non-clock-like (NCL) neurones in the dorsal raphe nucleus (DRN)respond to both noxious and non-noxious stimuli but clock-like (Cl)and non-clock-like-non-responding (NCLN) neurones do not respond toeither.137Conclusions7. EA facilitated the discharge of NCL neurones in the dorsal raphenucleus (DRN) but not CL. or NCLN neurones.8. There is a direct projection of only NCL DRN neurones to the nucleusparafascicularis (NPF). All neurones displayed constant but widelyranging latency (0.6 ms to 9.0 ms). The conduction velocity wasestimated to range from 0.40 - 5.8 rn/s.9. There are 3 types of spontaneously active NPF neurones: slow,bursting and fast firing. The slow and some bursting NPF neuroneswere facilitated by noxious stimulation.10. EA inhibited the nociceptor-driven NPF neurones.11. Simultaneously recording from a NCL neurone in the DRN and a NPFneurone revealed that EA produced an increase in the discharge rateof DRN neurones concomitantly with a decrease in the firing rate ofNPF neurones.12. The inhibition of NPF neurones produced by DRN stimulation had asudden onset, offset and a prolonged time course that wascorrelated with the duration of stimulation. However, themagnitude of the inhibition was the same regardless of the periodof stimulation.13. The effect of single pulse URN stimulation on glutamate-driven NPFneurones showed an inhibitory response in most cells tested withlatencies ranging from 6.0 to 12.0 ms. These latencies were in asimilar range to those found by antidromic activation of DRNneurones from the NPF.138Conclusions14. Behavioral signs of hyperaggresiveness, hyperactivity as well asweight loss were observed in 5,7-dihydroxytryptamine (5,7-DHT)treated animals. There was also a marked reduction of 5-HTneurones in the DRN when examined using the indirectimmunofluorescence technique. Only NCLN neurones were recorded in5,7-DHT treated animals; CL and NCL neurones could not be found,15. EA fails to inhibit NPF neurones in response to noxious radiantheat in 5,7-DHT treated animals.16. Stimulation of the DRN fails to inhibit, but rather increases, thenoxious stimulation-evoked activity in NPF neurones in 5,7-DHTtreated animals.17. 5-HI lontophoretically applied to glutamate-driven NPF neurones wasinhibitory. Alaproclate, a 5-HI uptake inhibitor, wheniontophoretically applied at low current (30 nA) failed to alterthe activity of glutamate-driven NPF neurones but enhanced theinhibitory responses of 5-MT but not that of GABA.18. The inhibitory response of NPF neurones to stimulation of the DRNor EA was enhanced by iontophoretically applied alaproclate atcurrents which do not alter the firing rate alone. Alaproclateiontophoretically applied at high current to glutamate-driven NPFneurones reduced the discharge rate.19. Cyproheptadine, a serotonin antagonist, blocked the inhibition ofNPF neurones produced by the stimulation of the DRN, but not theGABA-induced inhibition.139Conclusions20. Cyproheptadine blocked the inhibition of glutamate-driven NPFneurones produced by EA.140SUMMARYElectroacupuncture (EA) appHed to Zusanli” (st. 36) and“Shangjuxu (st. 38) acupuncture points (10 mm, 10 Hz, 1.0 ms) wasfound to produce a long-lasting inhibition of wide dynamic range (WDR)in the spinal cord dorsal horn neurones and to prolong the latency ofthe tail-flick reflex in the lightly anesthetized rat. This inhibitionwas effectively produced at a stimulation intensity which not activatefibres smaller than AB. The effects of EA were eliminated by cold-blocking the spinal cord rostral to the recording site suggesting asupraspinal involvement in the EA-induced inhibition of spinal cordnociceptive transmission. EA also facilitated the discharge of non-clock-like (NCL) dorsal raphe neurones (DRN). Bilateral lesions of theventrolateral tract (VU), but not the dorsolateral funiculi (DLF),blocked this effect suggesting that the ascending arm of the loop is viathe VU. The descending arm is located in the DLF since bilaterallesions of the DLF blocked the spinal cord effects of EA.Evidence in the literature suggests that the DRN may be involvedin the above supraspinal loop as well as in an ascending inhibitorypathway to the nucleus parafascicularis (NPF). Examination of the DRNrevealed there are three types of neurones: clock-like (CL), NCL andnon-clock-like non-responding neurones (NCLN). The NCL neurones wereexcited by noxious and non-noxious natural peripheral stimuli as well asEA. The other neurones were non-responsive to these stimuli. NCL141neurones of the DRN were also antidromically activated by NPFstimulation indicating that the projection from the DRN to the NPF isdirect. Stimulation of the DRN produced an inhibition of NPF neuroneswhich had a sudden onset and offset and a duration that was correlatedwith the length of stimulation. EA also produced long-lastinginhibition of these cells. The inhibitory pathway from the DRN to theNPF, which is activated by EA and presumably mediated by NCL neurones,would appear to be serotonergic. The evidence favoring this idea is thatthe inhibition evoked by ORN stimulation or EA is enhanced byalaproclate (a 5-MT uptake blocker) and blocked by 5,7-dihydroxytryptamine (a 5-HT neurotoxin), or cyproheptadine (a serotoninantagonist).142BIBLIOGRAPHYAbrams, G. H., and Recht, L. Neuropeptides and their role in pain andanalgesia. Acupuncture and Electro-Therapeutics, Res., Tnt. J.7:105-121, 1982.Aghajanian, G. K. Influence of drugs on the firing of serotoniricontaining neurons in brain. Federation Proceedings 31:91-96,1972a.Aghajanian, G. K., Haigler, H. J. and Bloom, F. E. Lysergic aciddiethylamide and serotonin; Direct actions on serotonin-containingneurones in rat brain. Life Sci. 11:615-622, 1972b.Aghajanian, G. K. and Haigler, H. J. L-Tryptophan as a selectivehistochemical marker for serotonergic neurones in single-cellrecording studies. Brain Res. 81:364-372, 1974.Aghajanian, G. K., Wang, R. Y and Baraban, J. Serotonergic and nonserotonergic neuron of the dorsal raphe: reciprocal changes infiring induced by peripheral nerve stimulation. Brain Res. 153;169-175, 1978.Ai, M. and Wu, Z. The influence of electroacupuncture on theultrastructure of synapses in medial region of thalamus. In:Advances in Acupuncture and Acupuncture Anaesthesia.”, Ed., T.Chang, Tiantan Xili Beijing China. The People’s Medical PublishingHouse, pp. 449, 1979.Aimone, L. D. and Gebhart, G. F. Stimulation-produced spinal cordinhibition from the midbrain in the rat is mediated by anexcitatory amino acid neurotransmitter in the medial medulla. J.Neuroscience, 6:1803-1813, 1986.Aimone, L. D., Jones, S. L. and Gebhart, G. F. Stimulation-produceddescending inhibition from the periaqueductal gray and nucleusraphe magnus in the rat: mediation by spinal monoamines but notopioids. Pain 31:123-136, 1987.Albe-Fessard, D. and Kruger, L. Duality of unit discharges from catcentrum medianum in responses to natural and electricalstimulation. Neurophysiol. 25:3-20, 1962.143Bibl iographyAndersen, E and Dafny, N. Microiontophoretically applied 5-HT reducesresponses to noxious stimulation in the thalamus. Brain Res,241:176-178, 1982.Andersen, E. and Dafny, N. An ascending serotonergic pain modulationpathway from the dorsal raphe nucleus to the parafascicularisnucleus of the thalamus. Brain Res. 269:57-67, 1983a.Andersen, IE. and Dafny, N. Dorsal raphe stimulation reduces responsesof parafascicular neurons to noxious stimulation. Pain 15:323-331.1983b,Anderson, S. A. and Holmgren, E. On acupuncture analgesia and themechanism of pain. American J. chinese Medicine. 3:311-334, 1975.Armstrong, J. M. and Millar, J. Carbon fiber microelectrodes. J.Neurosci, Meth. 1:279-287, 1979.Asarch, K. B., Ransom, R. W. and Shih, J. C. 5-HT-la and 5-HT-lbselectivity of two phenylpiperazine derivatives: evidence for 5-[IT-lb heterogeneity. Life Sci, 36:1265-1271, 1985.Bao, H., Zhou, Z., Yu, Y. and Han, J. C fibre is not important inacupuncture analgesia but is for DNIC. Acupuncture Res., 14:139-140, 1989.Basbaum, A., Clanton, C. H., and Fields, H. L. Opiate and stimulus-produced analgesia: functional anatomy of a medullospinal pathway.Proceedings of the National Academy of Sciences of the UnitedStates of America 73:4685-4688, 1976.Basbaum, A. I. and Fields, H. L. Endogenous pain control mechanism:review and hypothesis. Ann. Neurol.4:451-462, 1978.Basbaum, A. I. and Fields, H. L. The Origin of descending pathways inthe dorsolateral funiculus of the spinal cord of the cat and rat:Further studies on the anatomy of pain modulation. J. Comp.Neurol. 187:513-532, 1979.Basbaum, A. I., Clanton, C. H. and Fields, H. L. Three bulbospinalpathways from the rostral medulla of the cat: an autoradiographicstudy of pain modulating systems. J. Comp. Neurol. 178, 209-224,1978.Behbehani, K. M. and Fields, H. L, Evidence that an excitatoryconnection between the periaqueductal gray and nucleus raphemagnus mediates stimulation produced analgesia. Brain Res. 1979.144Bibi iographyBehbehani, M. M. and Pomeroy, S. L. Effect of morphine injected inperiaquduectal gray on the activity of single units in nucleusraphe magnus of the rat. Brain Res. 149:266-269, 1978.Beijing Medical College, Dept. of Biophysics and Physiology. The effectof electro-acupuncture and some exogenous monoamines on unitdischarges of the dorsal nucleus raphe. In: “Research onAcupuncture, Moxibustion, and Acupuncture Anesthesia”, Ed., T.Chang, Science Press, pp. 93-97, 1986.Beitz, A. J. The organization of afferent projections to the midbrainperiaquduectal grey of the rat. Neuroscience, 7:133-159, 1982.Benabid, A. L., Henriksen, S. J., McGinty, J. F. and Bloom, F. E,Thalamic nucleus ventro-postero-lateralis inhibits nucleusparafascicularis response to noxious stimuli through a non-opioidpathway. Brain Res, 280:217-231, 1983.Bennett, G. J. and Mayer, D. J. Inhibition of spinal cord interneuronesby narcotic microinjection and focal electrical stimulation in theperiaqueductal central gray matter. Brain Res. 172:243-257, 1979.Berge, 0. G. Effects of 5-HT receptor agonists and antagonists on areflex response to radiant heat in normal and spinally transectedrats. Pain 13:253-266, 1982.Berge, 0. G. Response latencies in the tail-flick test depend on tailskin temperature. Neurosci. Lett. 86:284-288, 1988.Bing, Z., Villanueva, L. and LeBars, D. Acupuncture and diffuse noxiousinhibitory controls: Naloxone-reversible depression of activitiesof trigeminal convergent neurons. Neuroscience 37:809-818, 1990.Bjorklund, A., Baumgarten, H. G., and Nobin, A. Chemical lesioning ofcentral monoamine axons by means of 5,6-dihydroxytryptamine and5,7-dihydroxytryptamine. Adv. Biochem. Psychopharmacoi. 10:13-33,1974.Blomqvist, A. and Craig, A. D. Organization of spinal and trigeminalinput to the PAG. In “The Midbrain Periaqueductal Gray Matter”,Edited by Depaulis, A. and Bandler, R., Plenum Press, pp.345-363,1990Bobillier, P., Petitjean, S. D., Ligier, M. and Seguin, S. Differentialprojections of the nucleus raphe dorsalis and nucleus raphecentral as revealed by autoradiography. Brain Res. 85:205-210,1975.145BibliographyBoivie, J. and Perl, E. R. Neural substrates of somatic sensation. In:“MTP international review of science, physiology series one.”vol.3, neurophysiology, Ed, Hunt, C.C., University Park Press,Baltimore, pp.303-411, 1975.Bonate, P. L. Serotonin receptor subtypes: functional, physiological,and clinical correlates, Clinical Neuropharcol. 14:1-16, 1991.Bradley, P. 8., Engel, G. and Feniuk, W. Proposals for theclassification and nomenclature of functional receptors for 5-hydroxytryptamine. Neuropharmacol. 25:563-566, 1986.Bradshaw, C. M., Stoker, M. J. and Szabadi, F. Comparison of theneuronal responses to 5-hydroxytryptamine, noradrenaline andphenylephrine in the cerebral cortex: Effects of haloperidol andmethysergide. Neurpharmacol. 22:677-685, 1983.Bramwell, G. J. and Gonye, T. Responses of midbrain neurones tomicroiontophoretically appl led 5-hydroxytryptamine: comparisonwith the response to intravenously administered lysergic aciddiethylamine. Neuropharmacol. 15:457-461, 1976Brockhaus, A. and Elger, C. F. Hypoanalgesic efficacy of acupuncture onexperimental pain in man. Comparison of laser acupuncture andneedle acupuncture. Pain 43:181-185, 1990.Brown, A. 6. Cutaneous afferent fibre collaterals in the dorsal columnsof the cat. Exp. Brain Res. 5:293-305, 1968Brown, A. 6. Effects of descending impulses on transmission through thespinocervical tract. J. Physiol. 219:103-125, 1971.Brown, A. 6. Organization in the spinal cord: the anatomy andphysiology of identified neurones. Springer-Verlag, Berlin, 1981.Brown, A. 6. and Iggo, A. A quantitative study of cutaneous receptorsand afferent fibres in the cat and rabbit. J. Physiol. 193:707-733, 1967.Brown, A. 6. and Franz, 0. N. Responses of spinocervical tract neuronesto natural stimulation of identified cutaneous receptors. Exp.Brain Res. 7:231-249, 1969.Brown, A. 6., Hamann, W. C. and Martin, H. F. Effects of activity innon-myelinated afferent fibres on the spinocervical tract. BrainRes. 98:243-259, 1975.146Bibi iographyBrown, A. G., Fyffe, R. E. W., Noble, R., Rose, P. K. and Snow, P. J.The density, distribution and topographical organization ofspinocervical tract neurones in the cat. J. Physiol. 300:409-428,1980.Burgess, P. R., Petit, D., and Warren, R. M. Receptor types in cathairy skin supplied by myelinated fibers. J. Neurophysiol. 31:833-848, 1968.Burstein, R. Cliffer, K. D. and Giesler, G. J. Direct somatosensoryprojections from the spinal cord to the hypothalamus andtelencephalon. J. Neurosci, 7:4159-4164, 1987.Burstein, R. Cliffer, K. D., and Giesler, 6. J. Cells of origin of thespinohypothalamic tract in the rat. J. Comp. Neurol. 291:329-344,1990.Cadden, S. W. Villanueva, L., Chitour, D. and LeBars, D., Depression ofactiviti@s of dorsal horn convergent neurones by propriospinalmechanisms triggered by noxious inputs; Comparison with diffusenoxious inhibitory controls (DNIC). Brain Res. 275:1-11,1983.Carstens, E. Inhibition of rat spinothalamic tract neuronal responsesto noxious skin heating by stimulation in midbrain periaqueducalgray or lateral reticular formation. Pain 33:215-224, 1988.Carstens, E. and Campbell, I. 6. Parametric and pharmacological studiesof midbrain suppression of the hind limb flexion withdrawal reflexin the rat. Pain 33-201-213, 1988.Cervero, F., Iggo, A. and Molony, V. Responses of spinocervical tractneurones to noxious stimulation of the skin. J. Physiol. 267:537-558, 1977.Cervero, F. Somatic and visceral input to the thoracic spinal cord ofthe cat: Effects of noxious stimulation of the biliary system. J.Physiol. 337:51-67, 1983.Cervero, F. and Woistencroft, J. A positive feedback loop betweenspinal cord nociceptive pathways and antinociceptive areas of thecat’s brain stem. Pain 20:125-138,1984.Chan, S. H. H. and Fung, S. J. Suppression of polysynaptic reflex byelectro-acupuncture and a possible underlying presynapticmechanism in the spinal cord of the cat. Exp. Neurol. 48:336-342,1975.147BibliographyChang, I-f. T. Interactions in thalamus of afferent impulses fromacupuncture point and site of pain. In: “Research on Acupuncture,Moxibustion, and Acupuncture Anesthesia”, Ed., T. Chang, SciencePress, pp.21-30, 1986.Chang, K. L., Wel, C. P., Lin, V. 14., Chou L. P., Chou, L. H., Lo, C.L. and Chang, T. C. Effect of stimulation of midbrain raphenuclei on the discharge on pain sensitive cells in nucleusparafascicularis of the thalamus and its significance inacupuncture analgesia. Acta Physiologica Sinica. 31:209-218, 1979.Chaouch, A., Menetrey, D., Binder, D. and Besson, J. M. Neurons at theorigin of the medial component of the bulbopontine spinoreticulartract in the rat: an anatomical study using horseradich peroxidaseretrograde transport. J. Comp. Neurol. 214:309-320, 1983.Chapman, C. R., Chen, A. C. and Bonica, J. J., Effects ofintrasegmental electrical acupuncture on dental pain: evaluationby threshold estimation and sensory decision theory. Pain 3:213-227, 1977.Chapman, C. R., Colpitts, Y. 14., Benedetti, C., Kitaeff, R. and Gehrig,J. D. Evoked potential assessment of acupuncture analgesia:Attempted reversal with naloxone. Pain 9:183-197, 1980.Cheng, J., Ye, Z. and Ou, S. The effects of naloxone on the inhibitoryaction of electro-acupuncture upon cortical potentials evoked bytooth pulp stimulation in rabbits. In: “Advances in Acupunctureand Acupuncture Anaesthesia.” Tiantan Xili Beijing China. ThePeople’s Medical Publishing House, pp. 497, 1979.Cheng, S. S. R. and Pomeranz, B. Electroacupuncture analgesia could bemedicated by at least two pain-relieving mechanisms: Endorphin andnon-endorphin systems. Life Sciences 25:1957-1962, 1979.Cheng, S. S. R. and Ponieranz, B. Electroacupuncture analgesia ismediated by stereospecific opiate receptors and is reversed byantagonists of type 1 receptors. Life Sciences 26:631-638, 1980.Choi, J. J., Srikantha, K. and Wu, W. H. A comparison ofelectoacupuncture transcutaneous electrical nerve stimulation andlaserphotobiostimulation on pain relief and glucocorticoldexcretion. Acupuncture and Electro-Therapeutics Res., Tnt. J.11:45-51, 1986.Christensen, B. N. and Perl, E. R. Spinal neurons specifically excitedby noxious or thermal stimuli: marginal zone of the dorsal horn.J. Neurophysiol, 33:293-307, 1970.148Bib] lographyChu, L. S. W., Yeh, S. D. J. and Wood, D. D. “Acupuncture manual , awestern approach”, Marcel Dekker, mc, 1979.Chung, J. M., Fang, Z. R., Hori, Y., Lee, K. H. and Willis, W. D.Prolonged inhibition of primate spinothalamic tract cells byperipheral nerve stimulation. Pain 19:259-275, 1984a.Chung, J. M., Lee, K. K., Hori, Y., Endo, K. and Willis, W. D. Factorsinfluencing peripheral nerve stimulation produced inhibition ofprimate spinothalamic tract cell. Pain 19:277-293, 1984b.Chung, K., Kevetter, G. A., Willis, W. D. and Coggeshall, R. E. Anestimate of the ratio of propriospinal to long tract neurons inthe sacral spinal cord of the rat. Neurosci. Lett. 44:173-177,1984.Conrad, L., Christiana, C. A., Leonard, M and Plaff, D. W. Connectionsof the Median and dorsal raphe nuclei in the rat:An autoradiographic and degeneration study. J. Comp. Neurol.156:179-206, 1974.Craig, A. 0. Spinocervical tract cells in cat and dog, labeled by theretrograde transport of horseradish peroxidase. Neurosci. Lett.3:173-177, 1976,Dafny, N. and Gildenberg, P. Morphine effects on spontaneous,nociceptive, antinociceptive and sensory evoked responses ofparafasciculus thalami units in morphine naive and morphinedependent rats. Brain Res. 323:11-20, 1984Dafny, N., Reyes-Vazquez-C and Qiao, J. T. Modification ofnociceptively identified neurons in thalamic parafascicularis bychemical stimulation of dorsal raphe with glutamate, morphine,serotonin and focal dorsal raphe electrical stimulation. BrainRes. Bull. 24:717-23, 1990.Dahlstrom, A. and Fuxe, K. Evidence for the existence of monoaminecontaining neurons in the central nervous system. Acta Physiol.Scand. 62, suppl. 232:1-55, 1964.Davis, H. S. Possible dorsal horn output correlation with acupunctureanesthesia. Proceedings of the NIH Acupuncture Conference.Bethesda, MD., Feb. 28- Mar. 1, 1973.149Bibl iographyDept. of Physiology, Hunan Medical College. Effect of peripheralstimulation and acupuncture on unit discharges in the nucleusraphe magnus of the cat. In: “Advances in Acupuncture andAcupuncture Anaesthesia.” Tiantan Xiii Beijing China. ThePeople’s Medical Publishing House, pp. 352, 1979.Descarries, L., Watkins, K. C., Garcia, S. and Beaudet, A. Theserotonin neurons in nucleus raphe dorsal is of adult rat: a lightand electron microscope radioautographic study, J. Comp. Neuro.207:239-254, 1982.Devore, J. L. Probability and statistics for engineering and thesciences. Brooks/Cole Publishing Company, pp 343-421, 1982.Dong, W. K., Ryu, H. and Wagman, I. H. Nociceptive responses of neuronsin medial thalamus and their relationship tospinothalamic pathways. J. Neurophysiol. 41:1592-1613, 1978.Du, H. J. and Chao, Y. F. Localization of central structures involvedin descending inhibitory effect of acupuncture on viscero-somaticreflex discharges. Scientia Sinica 19:137-148, 1976.Du, H. J., Chao, V. F. and Zheng, R. K. Inhibitory effect of raphestimulation on viscero-somatic reflexes in cat and itsrelationship with acupuncture analgesia. Acta Physiologica Sinica30:1-9, 1978.Du, H. J. and Zhao, Y. F. Relationship between nucleus raphe magnus andhigher brain regions in inhibition of viscerosomatic reflex. In:“Research on Acupuncture, Moxibustion, and AcupunctureAnesthesia”, Ed. T. Chang, Science Press, pp. 70-81, 1986.Duggan, R. M. An overview of completed research on acupuncture. J.Traditional. Acup. 2:43-50, 1978.Eide, P. K. and Hole, K. Acute and chronic treatment with selectiveserotonin uptake inhibitors in mice: effects on nociceptivesensitivity and response to 5-methoxy-N, N-dimethyltryptamine.Pain 32:333-340, 1988.El-Etr, A. A. and Pesch, R. N. Preliminary report on somatosensorycerebral evoked potentials during the stimulation of the Ho Kupoint. Proceedings of the NIH Research Conference. Feb. 28- Mar.1, 1973.150Bibi lographyEnevoldson, T. P. and Gordon, G. Spinocervical neurons and dorsal hornneurons projecting to the dorsal column nuclei through thedorsolateral fascicle: a retrograde HRP study in the cat. Exp.Brain Res. 75:621-630, 1989.Fang, F. G., Moreau, J. L. and Fields, H. 1. Dose-dependentantinociceptive action of neurotensin microinjected into therostroventroniedial medulla of the rat. Brain Res. 420:171-174,1987.Fields, H. L. and Anderson, S.D. Evidence that raphe-spinal neuronsmediate opiate arid midbrain stimulation-produced analgesia. Pain5:333-349, 1978.Fields, H. L., Basbaum, A. I., Clanton, C. H. and Anderson, S. D.Nucleus raphe magnus inhibition of spinal cord dorsal hornneurons. Brain Res. 126:441-453, 1977a.Fields, H. [.., Clanton, C. I-I., and Anderson, S. D. Somatosensoryproperties of spinoreticular neurons in the cat. Brain Res.120:49-66, 1977b.Fields, H. L. and Basbaum, A. I. Brainstem control of spinal pain-transmission neurons. Ann. Rev. Physiol. 40:217-48, 1978.Freeman, T. B.,, Campbell, J. N. and Long, D. M.. Naloxone does notaffect pain relief induced by electrical stimulation in man. Pain17:189-195, 1983.Gaary, A. E. Acupuncture analgesia. Acupuncture and ElectroTherapeutics Res., mt. J. 1:219, 1975.Gallager, D. W. arid Pert, A. Afferents to brain stem nuclei (brain stemraphe, nucleus reticularis pontis caudalis and nucleusgigantocellularis) in the rat as demonstrated bymicroiotophoretically applied horseradish peroxidase. Brain Res.144:257-275, 1978Gasser, H. S. Unmedullated fibers originating in dorsal root ganglia.J. Gen. Physiol. 33:651-690, 1950.Gaw, A. C., Chang, L. W. and Shaw, L. C. Efficacy of acupuncture onosteoarthritic pain. New Eng. J. Med. 293:375-378, 1975.Gebhart, G. F., Sandkuhler, J.., Thaihammer, J. G. and Zimmermann, M.Inhibition in spinal cord of nociceptive information by electricalstimulation and morphine microinjection at identical sites inmidbrain of the cat. J. Neurophysiol. 51:75-89, 1984.151Bibi iographyGhia, J. N., Mao, W., Toomey, T. C. and Gregg, J. N. Acupuncture andchronic pain mechanisms. Pain 2:285-299, 1976.Giesler, 6. J., Menetrey, D., Guilband, 6. and Besson, J. Lumbar cordneurons at the origin of the spinothalamic tract in the rat. BrainRes. 118:320-324, 1976.Giesler, 6. J., Cannon, J. T., Urca, 6. and Liebeskind, J. C. Longascending projections from substantia gelatinosa Rolandi and thesujacent dorsal horn in the rat. Science 202:984-986, 1978.Giesler, 6. J., Menetrey, 0. and Basbaum, A. Differential origins ofspinothalamic tract projections to medial and lateral thalamus inthe rat. J. Comp. Neurol. 184:107-126, 1979.Giesler, G. J., Spiel, I-I. R. and Willis, W. D. Organization ofspinothalamic tract axons within the rat spinal cord. J. Comp.Neurol. 195:243-252, 1981.Giesler, 6. J., Nahin, R. L. and Madsen, A. M. Postsynaptic dorsalcolumn pathway of the rat. I. Anatomical studies. J. Neurophysiol.51:260-275, 1984.Giesler, 6. J. and Cliffer, K. D. Postsynaptic dorsal column pathway ofthe rat. II. evidence against an important role in nociception.Brain Res. 326:347-356, 1985.Glennon, R. Central serotonin receptors as targets for drug research.J. Med. Chem. 30:1-12, 1987.Greguss, P. A model for making acupuncture consistent with easternconcepts of biological information processing. Proceedings fromthe NIH Acupuncture Conference, Feb. 28-Mar. 1, 1973.Grossmann, W. and Jurna, I. Depression by morphine of activity in theventrolateral tract evoked from cutaneous A-fibers. Eur. J.Pharmacol. 29:171-174, 1974.Grossman, t4. L., Basbaum, A. I. and Fields, H. L. Afferent and efferentconnections of the rat tail flick reflex (a model used to analyzepain control mechanisms). J. Comp. Neurol. 206:6-16, 1982.Guilbaud, G., Besson, J. M., Oliveras, J. L. and Liebeskind, J. C.Suppression by LSD of the inhibitory effect exerted by dorsalraphe stimulation on certain spinal cord interneurones in the cat.Brain Res. 61:417-422, 1973.152Bibl iographyHaigler, H. J. Morphine, ability to block neuronal activity evoked by anociceptive stimulus. Life Sciences, 19:841-858, 1976.Hall, J. G., Duggan, A. W., Johnson, S. M. and Morton, C. R. Medullaryraphe lesions do not reduce descending inhibition of dorsal hornneurones of the cat. Neurosci. Lett. 25:25-29, 1981.Hall, J. G., Duggan, A. W. Morton, C. R. and Johnson, S. M. Thelocation of brainstem neurones tonically inhibiting dorsal hornneurones of the cat. Brain Res, 244:215-222, 1982.Han, J. Role of central neuro-transmitters in acupuncture analgesia:In: “Research on Acupuncture, Moxibustion, and AcupunctureAnesthesia”, Ed., T. Chang, Science Press. pp. 241-255, 1986.He, L. Effect of iontophoretic etorphine and electroacupuncture onnociceptive response from nucleus lateral is anterialis of thalamusin rabbits. Acupuncture and Electro-Therapeutics Res.., mt. J.6:151-157, 1981.He, L. Involvement of endogenous opioid peptides in acupunctureanalgesia. Pain 31:99-121, 1987.He, L. and Dong, W. Activity of oploid peptidergic system inacupuncture analgesia. Acupuncture Electro-Therapeutics Res., Tnt.J. 8:257-266, 1983.He, L., Du, 1., Jiang, J., Shi, Z. and Wang, M. Effects of naloxone onanalgesia from caudate nucleus stimulation in rabbits. In:“Advances in Acupuncture and Acupuncture Anaesthesia.” TiantanXiii Beijing China. The People’s Medical Publishing House. pp.480, 1979.He, L., Dong, W. Q. and Wang, M. Effects of lontophoretic etorphine andnaloxone, and electroacupuncture on nociceptive responses fromthalamic neurones in rabbits. Pain 44:89-95, 1991.He, S. Wang, P and Shen, K. Inhibitory effect of cord dorsumstimulation on evoked potential of bulbar reticular formation. In:“Advances in Acupuncture and Acupuncture Anaesthesia.” TiantanXili Beijing China. The People’s Medical Publishing House. pp.361, 1979.Henning, C. W., Steihoff, W. C. and Braughler, S. A. Tonic immobilityand the serotonergic system in chickens: differential blockade ofreceptors by cyproheptadine and cinanserin. Physiology andBehavior, 44:15-20, 1988.153Bibi iographyHoistege, G. Descending pathways from the periaqueductal gray andadjacent areas. In “The Midbrain Periaqueductal Gray Matter”,Edited by Depaulis, A. and Bandler, R., Plenum Press. pp.239-265,1991.Hosobuchi, Y., Rassier, J., Bloom, F. E. and Guillemin, R. Stimulationof human periaqueductal gray for pain relief increasesimmunoreactive B-endorphin in ventricular fluid. Science, 203,279-281, 1977.Hung, C., Li, X., Zhang, C., Liu, S., Pan, X. and Zhou, Z. Effectelectrical stimulation of midbrain periaqueductal gray onacupuncture analgesia in rats. Acta Physiologica Sinica 34:343-357, 1982.Hunskaar, S., Berge, 0. G., Broch, 0. J. and Hole, K. Lesions of theascending serotonergic pathways and antinociceptive effects aftersystemic administration of p-chloroamphetamine in mice. Pharmacol.Biochem. Behav. 24:709-714, 1986.Hynynen, K. H., Ketovouri, [I., Mattila, M. A. K., Mattila, S., NoksoKoivisto, V. M. and Pontinen. P. J. Heart-rate variation as apain indicator in conscious subjects undergoing acupunctureanalgesia. Acupuncture and Electro-Therapeutics Res., mt. J.6:47-55, 1981.Jacob, W. L. A case of severe chronic depression treated withacupuncture. The AAMA Rev. 2:33-34, 1990.Jurna, I. Aminophylline differentiates between the depressant effectsof morphine on the spinal nociceptive reflex and on the spinalascending activity evoked from afferent C fibers. Europ. J.Pharmacol. 71:393-400, 1981.Kevetter, G. A. and Willis, W. D. Spinothalmic cells in the rat lumbarcord with collaterals to the medullary reticular formation. BrainRes. 238:181-185, 1982.Kevetter, G. A. and Willis, W. D. Collaterais of spinothalamic cells inthe rat. J. Comp. Neurol. 215:453-464, 1983.Kneisley, L. W., Biber, M. P. and LaVail, J. H. Brain stem projectionsto the spinal cord using retrograde transport of horseradishperoxidase in Rhesus monkeys. Neurology (Minneap.) 26:350, 1976.Kuypers, H. G. J. M. and Maisky, V. A. Retrograde axonal transport ofhorseradish peroxidase from spinal cord to brain stem cell groupsin the cat. Neurosci. Lett. 1:9-14, 1975.154BibliographyLeBars, D., Menetrey, D., Conseiller, C. and Besson, J. M. Depressiveeffects of morphine upon lamina V cells activities in the dorsalhorn of the spinal cat. Brain Res. 98:261-277, 1975.LeBars, D, Menetrey, 0. and Besson, J. M. effects of morphine upon thelamina V type cells activities in the dorsal horn of thedecerebrate rat. Brain Res. 113:293-310, 1976.LeBars, D., Dickenson, A. H. and Besson, J. M. Diffuse noxiousinhibitory controls (DNIC). I. Effects on dorsal horn convergentneurones in the rat. Pain 6:283-304, 1979a.LeBars, D., Dickenson, A. H. and Besson, J. M. Diffuse noxiousinhibitory controls (DNIC). II. Lack of effect on non-convergentneurones, supraspinal involvement and theoretical implications.Pain 6:305-327, 197gb.LeBars, 0., Dickenson, A. H. and Besson, J. M. t4icroinjection ofmorphine within nucleus raphe magnus and dorsal horn neuronesactivities related to nociception in the rat. Brain Res. 189:467-481, 1980.LeBars, D., Dickenson, A. H., Besson, J. M. and Villanueva, L. Aspectsof sensory processing through convergent neurons. In:”Spinalafferent processing”, Ed., Yaksh. T.L., Plenum Press, New York pp.467-504, 1986.Ledergerber, C. P. Electroacupuncture in obstetrics. Acupuncture andElectro-Therapeutics Res., Tnt. J. 2:105-118, 1976.Lee, P. K., Andersen, T. W. Modell, J. H. and Saga, S. A. Treatment ofchronic pain with acupuncture. JAMA. 232:1133-1135, 1975.Lewis, V. A. and Gebhart, G. F. Evaluation of the periaqueductalcentral gray (PAG) as a morphine-specific locus of action andexamination of morphine-induced and stimulation-produced analgesiaat coincident PAG loci. Brain Res. 124:283-303, 1977.Li, S., Jiang C and Chen, G. The relationship between needlingsensation and acupuncture effects, with special reference to theirascending pathway in the spinal cord. Acupuncture and ElectroTherapeutics Res., Tnt. J. 8:105-110, 1983.Liebeskind, J. C., Guilbaud, G., Besson, J. M. and Oliveras, J. L.Analgesia from electrical stimulation of the periaqueductal graymatter in the cat: behavioral observations and inhibitory effectson spinal cord interneurones. Brain Res. 50:441-446, 1973.155Bibl iographyLiebeskind, J. C., Guilbaud, C., Besson, J. M. and Oliveras, J. L.Analgesia from electrical stimulation of the periaqueductal graymatter in the cat; behavioral observations and inhibitory effectson spinal cord iriterneurons. Brian Res. 269:57-67, 1983.Light, A. R., and Perl, E. R. Central termination of identifiedcutaneous afferent units with fine myelinated fibers. Soc.Neurosci, Abstr. 3, 486, 1977.Light, A. R., and Pen, E. R. Reexamination of the dorsal rootprojection to the spinal dorsal horn including observations on thedifferential termination of coarse and fine fibers. J Comp.Neurol. 186:117-132, 1979.Lindberg, U. H., Thorberg, S. 0., and Bengtsson, S. Inhibitors ofneuronal monoamine uptake. 2. selective inhibition of 5-hydroxytryptamine uptake by a-amino acid esters of phenethylalcohols. J. Med. Chem. 21:448-456, 1978.Lin, B., Lui, Z., Wu, C., Tang, J.., Liu, L. and Zhu, H. The effects ofcauterization of supraoptic nucleus area and the mammillarycomplex of the rats on analgesic effect of electric needling of11Zusanli’. In: 11Advances in Acupuncture and AcupunctureAnaesthesia.’ Tiantan Xili Beijing China. The People’s MedicalPublishing House, pp.335, 1979.Linzer, M. and Atta, L. V. The Electrophysiological assessment ofacupuncture’s effect on single thalamic neuron in the cat: neuralcoding for pain at a thalamic level. Proceedings of the NIHAcupuncture Research Conference, Feb. 28-Mar. 1, 1973.Liii, R. P. Laminar origins of spinal projection to the periaqueductalgray of the rat. Brain Res. 264:118-122, 1983.Liu, X., Zhu, B. and Zhang S. Relationship between electroacupunctureanalgesia and descending pain inhibitory mechanism of nucleusraphe magnus. Pain 24:383-396, 1986.Lloyd, D. P. C. and Chang, H. T. Afferent fibers in muscle nerves. J.Neurophysiol. 11:199-208, 1948.Looney, C. L. Acupuncture for sensorineural hearing impairment.Proceedings of the NIH Acupuncture Research Conference, Feb. 28-Mar. 1, 1973a.Looney, C. L. Effect of acupuncture anesthesia upon peripheral vascularhemodynamics. Proceedings of the NIH Acupuncture ResearchConference, Feb. 28-Mar. 1, 1973b.156I3ibl iographyLooney, C. L. A neurologic basis for acupuncture (autonomic theory)(No.3). Acupuncture and Electro-Therapeutics Res., mt. J. 1:210,1975.Lu, H. C. “The Yellow Emperor”, a translated acupuncture book, Academyof Oriental Heritage, 1973.Lund, R. D. and Webster, K. E. Thalamic afferents from the dorsalcolumn nuclei, an experimental anatomical study in the rat. J.Comp. Neurol. 130:301-311, 1967.Luo, F,, Yuan, J., Yang, S., and Zhang, X. Inhibitory effect onnociceptive discharges of parafascicular nucleus of thalamus byelectrical stimulation. In: “Advances in Acupuncture andAcupuncture Anaesthesia.” Tiantan Xiii Beijing China. ThePeople’s Medical Publishing House. pp. 333, 1979.Matsumato, T., Ambruso, V., Martin, F. and Hayes, Jr. Effect ofacupuncture: I. Castrointestinal atony following vagotomy.Proceedings from the NIH Acupuncture Research Conference, Feb. 28-Mar. 1, 1973.Maunz, R. A., Pitts, N. C. and Peterson, B. W. Cat spinoreticularneurons: locations, responses and changes in responses duringrepetitive stimulation. Brain Res. 148:365-379, 1978.Mayer, D. J. and Liebeskind, J. C. Pain reduction by focal electricalstimulation of the brain: an anatomical and behavioral analysis.Brain Res. 68:73-93, 1974.Mayer, D. J., Price, D. D., Barber, J. and Raf ii, A. Acupunctureanalgesia: Evidence for activation of a pain inhibitory system asa mechanism of action. In:”Advances in Pain Research and Therapy”Ed., Bonica, J. J. and Albe-Fessard, D., Raven Press, New York,Vol.1. pp. 751-754, 1976.Mayer, D. J., Price, D. D. and Raf ii, A. Antagonism of acupunctureanalgesia in man by the narcotic antagonist naloxone. Brain Res.121:368-372, 1977.McClung, R. F. and Dafny, N. The parafascicular nucleus of the thalamusexhibits convergence input from the dorsal raphe and the spinaltract of the trigeminal nerve. Brain Res. 525-531, 1980.McLennan, H., Gilfillan, K. and Heap, Y. Some pharmacologicalobservations on the analgesia induced by acupuncture in rabbits.Pain 3:229-238, 1977.157Bibl iographyMenetrey, D., Chaouch, A. and Besson, J. M. Location and properties ofdorsal horn neurons at origin of spinoreticular tract in lumbarenlargement of the rat. J. Neurophysiol. 44:862-877, 1980.Mosko, S. S. and Jacobs, B. L. Midbrain raphe neurons: spontaneousactivity and response to light. Physiol. and Behav. 13:589-593,1974.Nakahama, H., Shima, K., Yamamoto, M. and Aya, K. Regularity of thespontaneous discharge of neurons in the nucleus raphe dorsal is ofthe cat. Neurosci. Lett. 23: 161-165, 1981.Necker, R. and Hellon, R. F. Noxious thermal input from the rat tail:modulation by descending inhibitory influences. Pain 4:231-242,1978.Numoto, M. N. and Donaghy, R. 11. P. Physiological responses toacupuncture. Proceedings of the NIH Acupuncture ResearchConference, Feb. 28-Mar. 1, 1973.Nyquist, J. K. and Greenhoot, J. H. Unit analysis of nonspecificthalamic responses to high intensity cutaneous input in the cat.Exp. Neurol. 42:609-622, 1974.Ogren, S. 0. and Holm, A. C. Test-specific effects of the 5-HT reuptakeinhibitors alaproclate and zimelidine on pain sensitivity andmorphine analgesia. J. Neural. Trans. 47:253-271, 1980.Ogren, S. 0., Holm, A. C., Hall, H. and Lindberg, U. H. Alaproclate, anew selective 5-HT uptake inhibitor with therapeutic potential indepression and senile dementia. J. Neural. Trans. 59:265-288,1984.Ogren, S. 0. and Berge, 0. G. Evidence for selective serotonergicreceptor involvement in p-chloroamphetamine-inducedantinociception. Naunyn-Schmiedeberg’s Arch. Pharmacol. 329:135-140, 1985.Ogren, S. 0., Nordstrom, 0., Danielsson, E., Peterson, L. L. andBartfai, T. In vivo and in vitro studies on the potentiation ofmuscarinic receptor stimulation by alaproclate, a selective 5-HTuptake blocker. J. Neural. Transm. 61:1-20, 1985.Oleson, T. D. and Kroening, R. J. A comparison of Chinese and Nogierauricular acupuncture points. Amer. J. Acup. 11:3, 1983.158Bibi iographyOleson, T. D., and Liebeskind, J. C. Modification of midbrain andthalamic evoked responses by analgesic brain stimulation in therat. In “Advances in Pain Research and Therapy” ed. Bonica andAlbe-Fessard. Raven Press, New York, 1:487-494, 1976.Oleson, T. D., Twombly, D and Liebeskind, J. D. Effects of pain-attenuating brain stimulation and morphine on electrical activityin the raphe nuclei of the awake rat. Pain 4:211-230, 1978.Oliveras, J. L., Besson, J. M., Guilbaud, 6. and Liebeskind, J. C.Behavioral and electrophysiological evidence of pain inhibitionfrom midbrain stimulation in the cat. Exp. Brain Res. 20:32-44,1974.Oliveras, J. L., Redjemi, F., Guilbaud, 6 and Besson, J. M. Analgesiainduced by electrical stimulation of the inferior centralisnucleus of the raphe in the cat. Pain 1:139-145, 1975.Oliveras, J. L., Guilbaud, 0. and Besson, J. M. A map of serotonergicstructures involved in stimulation producing analgesia inunrestrained freely moving cats. Brain Res. 164: 317-322, 1979.Olpe, H. R. The cortical projection of the dorsal raphe nucleus: someelectrophysiological and pharmacological properties. Brain Res.216:61-71, 1981.Ornstein, D. B. “The Complete Book of Acupuncture”, Celestial Arts,1976.Paxinos, 6 and Watson, C. The rat brain in stereotaxic coordinates,Academic Press, 1982.Pearl, G. S. and Anderson, K. V. Responses patterns of cells in thefeline caudal nucleus reticularis gigantocellularis after noxioustrigeminal and spinal stimulation. Exp. Neurol. 58:231-241, 1978.Pechura, C. and Liu, R. Spinal neurons which project to theperiaqueduectal gray and the medullary reticular formation viaaxon collaterals: a double-label fluorescence study in the rat.Brain Res. 374:357-361, 1986.Peets, J. M. and Pomeranz, B. Studies in suppression of nociceptivereflexes measured with tail flick electromyograms and usingintrathecal drugs in barbituate anesthetized rat. Brain Res.416:301-307, 1987.159BibliographyPeroutka, S. J. Pharmacological differentiation and characterization of5-HT1A, 5-HT1B, and 5-HT1c binding sites in rat frontal cortex. J.Neurochem. 47:529-540, 1986.Peschanski, M., Guilbaud, G. and Gautron, M. Posterior intralaminarregion in rat: neuronal responses to noxious and non-noxiouscutaneus stimuli. Exp. Neurl. 22:226-238, 1981.Peschanski, M., Mantyh, P. W. and Besson, J. N. Spinal afferents to theventrobasal thalamic complex in the rat: an anatomical study usingwheat-germ agglutinin conjugated to horseradish peroxidase. BrainRes. 278:240-244, 1983.Peschanski, M. and Besson, J. M. A spino-reticulo-thalamic pathway inthe rat: an anatomical study with reference to pain transmission.Neuroscience 12:165-178, 1984.Philips, C. M. and Rusy, B. F. Electrical characteristics of twoChinese stimulators. Proceedings of the NIH Acupuncture ResearchConference, Feb. 28- Mar. 1, 1973.Pierce, E. T., Foote, W. E. and Hobson, A. E. The efferent connectionof the nucleus raphe dorsalis. Brain Res. 107:137-144, 1976.Pomeranz, B. Specific nociceptive fibers projecting from spinal cordneurons to the brain: a possible pathway for pain. Brain Res.50:447-451, 1973.Pomeranz, B. Brain’s opiates at work in acupuncture? New Sd. Jan. 6,1977.Pomeranz, B. and Cheng, R. Suppression of noxious responses in singleneurons of cat spinal cord by electroacupuncture and its reversalby the opiate antagonist naloxone. Exp. Neurol. 64:327-341, 1979.Pomeranz, B., Cheng, R. and Law, P. Acupuncture reduceselectrophysiological and behavioral responses to noxious stimuli:pituitary is implicated. Exp. Neurol. 54:172-178, 1977.Pomeranz, B. and Paley, D. Electroacupuncture hypoalgesia is mediatedby afferent nerve impulses: An electrophylsiological study inmice. Exp. Neurol. 66:398-402, 1979.Pomeroy, S. L. and Behbehani, M. M. Physiologic evidence for aprojection from periaqueductal gray to nucleus raphe magnus in therat. Brain Res. 176:143-147, 1979.160Bibi iographyProudfit, H. K. and Anderson, E. 6. Morphine analgesia: blockade byraphe magnus lesions. Brain Res. 98:612-618, 1975.Pycock, C. J.,, Burns, S. and Morris, R. In vitro release of 5-hydroxytryptamine and r-aminobutyric acid from rat periaqueductalgrey and raphe dorsalis region produced by morphine or anenkephalin analogue. Neurosci. Lett. 22:313-317, 1981.Qiao, J. T. and Dafny, N. Dorsal raphe stimulation modulatesnociceptive responses in thalamic parafascicular neurones via anascending pathway: further studies on ascending pain modulationpathways. Pain 34:65-74, 1988.Qiao, J. T., Xie, Y. F. and Chang, H. Y. Supraspinal mechanism involvedin the inhibition of parafascicular neurons by periaqueductal graystimulation in rabbits. Acta Physiol. Sin. 35: 226-230, 1983.Ren, M. and Han, J. Rat tail flick acupuncture analgesia model. ChineseMedical J. 92:576-582, 1979.Reichling, 0. B. and Basbaum, A. I., Collateralization ofperiaqueductal gray neurons to forebrain or diencephalon and tothe medullary nucleus raphe magnus in the rat. Neuroscience42:183-200, 1991.Rexed, B. The cytoarchitectonic organization of the spinal cord in thecat. J. Corn. Neurol. 96:415-466, 1952.Reynolds, D. V. Surgery in the rat during electrical anlgesia inducedby focal brain stimulation. Science 164:444-445, 1969.Richter, J. A., Baum, M., Kunkel, R., Bauerle, A., Heimisch, W.,Amereller, H., Schumacher, K., Erdinann, K. and Bojzewicz, U. V.Clinical experience with electrical acupuncture analgesia in 125patients undergoing open heart surgery. Acupuncture ElectroTherapeutics. Res., mt. J. 1:143-156, 1975.Romita, V V., Yashpal, K., Hui-Chan, C. and Henry, J. L.Electroacupuncture analgesia in the lightly anaesthetized rat isblocked by naloxone and by the novel substance P antagonist, CP96,345. Canadian Pain Soc. Annual Meeting, May 14-16, 1992.Ross, S. B. and Renyi, A. L. Inhibition of the neuronal uptake of 5-hydroxytryptamine and noradrenaline in rat brain by (Z)- and (E)3-(4-bromophenyl )-N, N-dimethyl -3-(3-pyridyl) allylamines andtheir secondary analogues. Neuropharamacol. 16:57-63, 1977.161Bibi iographyRuda, K. A. An autoradiographic study of the efferent connections ofthe mid-brain central gray in the cat. Ph.D. Dissertation,University of Pennsylvania, 1975.Samanin, R., Gumulka, W. and Vaizelli, L. Reduced effect of morphinein midbrain raphe lesioned rats. Eur. J. Pharmacol. 10: 339-343,1970.Samanin, R., Ghezzi, D., Mauron, C. and Vaizelli, L. Effect of midbrainraphe lesion on the antinociceptive action of morphine and otheranalgesics in rats. Psychopharmacol. (Ben.) 33:365-368, 1973.Samanin, R. and Vaizelli, L. Increase of morphine-induced analgesia bystimulation of the nucleus raphe dorsalis. Eur. J. Pharinacol.16:298-302, 1971.Sanders, K. H., Klein, C. E., Mayer, T. E., Heyni, C. H. and Handwerker,H. 0. Differential effects of noxious and non-noxious input onneurones according to location in ventral periaqueductal grey ordorsal raphe nucleus. Brain Res. 186:83-97, 1980.Segal, M. Physiological and pharmacological evidence for a serotonergicprojection to the hippocampus. Brain Res. 94:115-131, 1975.Segal, M. 5-HT antagonists in rat hippocampus, Brain Res. 103:161-166,1976.Shen, E. Participation of Descending Inhibition in acupunctureanalgesia. In: “Research on Acupuncture, Moxibustion, andAcupuncture Anesthesia”, Ed., T. Chang, Science Press. pp.31-38,1986.Shen, E., Tsai, I. T. and Lan, C. Supraspinal participation in theinhibitory effect of acupuncture on viscero-somatic reflexdischarges. chinese Medical J. 1:431-440, 1975.Shen, E., Ma, W. H. and Lan, C. Involvement of descending inhibition inthe effect of acupuncture on the splanchnically evoked potentialin the orbital cortex of cat. Sci. Sinica. 21:677-685, 1978.Shen, K., He, S. and Wang, P. Effect of cord dorsum stimulation on EEGand behavior in unrestrained rabbits. In: “Research onAcupuncture, Moxibustion, and Acupuncture Anesthesia”, Ed., T.Chang, Science Press. pp. 185-194, 1986.162BibliographyShima, K,, Nakahama, H. and Yamamoto, M. Firing properties of two typesof nucleus raphe dorsalis neurons during the sleep-waking cycleand their responses to sensory stimuli. Brain Res. 399:317-326,1986.Shima, K., Nakahama, H., Yamamoto, M., Aya, K. and Inase, M. Effects ofmorphine on two types of nucleus raphe dorsal is neurones in awakecats. Pain 29:376-386, 1987.Sinclair, J. G., Fox, R. E. Mokha, S. S. and Iggo, A. The Effect ofnaioxone on the inhibition of nociceptor driven neurons in the catspinal cord. Quart. J. Exper. Physiol. 65:181-188, 1980.Sjolund, B., Terenius, L. and Eriksson, M. Increased cerebrospinalfluid levels of endorphins after electro-acupuncture. ActaPhysiol. Scand. 100:382-384, 1977.Smith, M. 0. Creating a substance abuse treatment program incorporatingacupuncture. The AAMA rev. 2:29-32, 1990.Soulairac, A., Gottesmann, C. and Charpentier, J. Effects of pain andof several central analgesics on cortex, hippocampus and reticularformation of brain stem. mt. J. Neuropharmacol. 6:71-81, 1967.Steinbusch, H. W. M. Distribution of serotonin immunoreactivity in thecentral nervous system of the rat-cell bodies and terminals.Neuroscience 6:557-618, 1981.Stone, S. A., Wenger, H. C., Ludden, C. T., Stavorski, J. M., and Ross,C. A. Antiserotonin-antihistamine properties of cyproheptadine.J. Pharmacol. Exp. Ther. 131:73-84, 1961.Stux, G. and Pomeranz, B. Acupuncture, textbook and atlas, Springer-Verlag, 1987.Sun, F., Xu, M. and Xu. S The effects of microinjections of naloxoneand p-chloroamphetamine into rabbit central gray on acupuncture.In: “Advances in Acupuncture and Acupuncture Anaesthesia.” TiantanXiii Beijing China. The People’s Medical Publishing House, pp.481, 1979.Swett, J. E., McMahon, S. B. and Wall, P. D. Long ascending projectionsto the midbrain from cells of lamina I and nucleus of thedorsolateral funiculus of the rat spinal cord. J. Comp. Neurol.231:66-77, 1985.Takeshige, C. Neuronal responses to acupuncture. Acupuncture andElectro-Therapeutics Res., Tnt. J. 6:57-741, 1981163Bibi iographyTakeshige, C. Differentiation between acupuncture and non-acupuncturepoints by association with analgesia inhibitory system.Acupuncture and Electro-Therapeutic Res, Tnt. J. 10:195-203, 1985.Takeshige, C., Luo, C. P. and Kamada, Y. Modulation of EEG and unitdischarges of deep structures of brain during acupuncturalstimulation and hypnosis of rabbits. In: “Advance in Pain Researchand Therapy” Ed., Bonica, J.J. and Albe-Fessard, D., Raven Press,New York, Vol. I. pp. 781-785, 1976.Takeshige, C., Sato, T. and Komugi, H. Roll of periaqueductal centralgray in acupuncture analgesia. Acupuncture and ElectroTherapeutics Res, mt. J. 5:323-338, 1980.Takeshige, C., Kamada, Y and Hisamitsu, T. Commonly responsive neuronsin the periaqueductal gray matter and midbrain reticular formationof rabbits to acupuncture stimulation, inversion, pressure on bodyparts and morphine. Acupuncture and Electro-Therapeutics Res.,Tnt. J. 6:57-74, 1981.Tjolsen, A., Berge, 0. G., Eide, P. K., Brock, 0. J. and Hole, K.Apparent hyperanalgesia after lesions of the descendingserotonergic pathways in due to increased tail skin temperature.Pain 33:225-231, 1988.Toda, K. and Ichioka, M. Electroacupuncture: relations between forelimbafferent impulses and suppression of jaw-opening reflex in therat. Exp. Neurol. 61:465-470, 1978.Trevino, D. L., Coulter, J. D. and Willis, W. D. Location of cells oforigin of spinothalamic tract in lumbar enlargement of the monkey.J. Neurophysiol. 36:750-761, 1973.Vanegas, H, Barbaro, N. M., and Fields, H. L. Midbrain stimulationinhibits tail-flick only at currents sufficient to excite rostralmedullary neurons. Brain Res. 321:127-133, 1984.Villanueva, L., Peschanski, M., Calvino, B. and LeBar, D. Ascendingpathways in the spinal cord involved in triggering of diffusenoxious inhibitory controls in the rat. J. Neurophysiol. 55:34-55,1986a.Villanueva, L., Chitour, D. and LeBars, D. Involvement of thedescending spinal projections responsible for diffuse noxiousinhibitory controls in the rat. J. Neurophyisol. 56:1185-1195,1986b.164Bibi iographyWagman, I. H. Propriospinal influences, especially inhibitory on dorsalhorn output. Proceedings from the NIH Acupuncture ResearchConference, Feb. 28-Mar. 1, 1973.Wall, P. D. The laminar organization of dorsal horn and effects ofdescending impulses. J. Physiol. 188:403-423, 1967.Wang, R. Y. and Aghajanian, G. K. Inhibition of neurons in the amygdalaby dorsal raphe stimulation: mediation through a directserotonergic pathway. Brain Res. 120:85-102, 1977a.Wang, R. Y. and Aghajanian, C. K. Antidromically identifiedserotonergic neurones in the rat midbrain raphe: evidence forcollateral inhibition. Brain Res. 132:186-193, 1977b.Willis, W. D. and Coggeshall, R. E. “Sensory Mechanisms of the Spinalcord”, second edition, Plenum Press, 1991.Woolf, C. J., Mitchell, D. and Barrett, C. Antinociceptive effect ofperipheral segmental electrical stimulation in the rat. Pain8:237-252, 1980.Wu, J., Xing, J., Xing, B., Lu, J. and Chen, Y. Inhibitory action ofelectro-acupuncture on dorsal horn neurons of the spinal cord.In: “Research on Acupuncture, Moxibustion, and AcupunctureAnesthesia”, Ed., T. Chang, Science Press, pp. 39-49, 1986.Xia, L. and He, C. Controlling effect of mesencephalon periaqueductalgray matter (PAG) on nociceptive afferent impulse. In: “Advancesin Acupuncture and Acupuncture Anaesthesia.” Tiantan Xili BeijingChina. The People’s Medical Publishing House, pp. 358, 1979.Xiang, L., Zhu, B. and Zhang, S. X. Relationship betweenelectroacupuncture analgesia and descending pain inhibitorymechanism of nucleus raphe magnus. Pain 24:383-396, 1986.Yaksh, I., Robert, L., Plant, L and Rudy, T. A. Studies on theantagonism by raphe lesions of the antinociceptive action ofsystemic morphine. Eur. J. Pharmacol. 41:399-408, 1977.Yee, K. C. Effects of acupuncture on uterine contraction duringdifficult labor. Acta Sci. and Tec. Chekiang, 1:6-13, 1973.Yezierski, R. P. Somatosensory input to the periaqueductal gray: aspinal relay to a descending control center. In “The MidbrainPeriaqueductal Gray Matter”, Ed by Depaulis, A and Bandler, R.,Plenum Press. pp.365-386, 1990.165Bib] iographyYezierski, R. P., Mendez, C. M. and Black, D. J. Cells of origin of therat spinomesencephalic tract: spinal distribution and collateralprojections. Neuroscience, in press. 1991.Zemlan, F. P., Leonard, C. N., Kow, L. M. and Pfaff, D. W. Ascendingtracts of the lateral columns of the rat spinal cord: a studyusing the silver impregnatin and horseradish peroxidasetechniques. Exp. Neurol. 62:298-334, 1978.Zhang, A., Pan, X., Xu, S., Cheng, J. and Mo, H. Endorphin andacupuncture analgesia. In: “Research on Acupuncture, Moxibustion,and Acupuncture Anesthesia”, Ed., I. Chang, Science Press. pp.256-266, 1986.Zhang, G., Wei, J., un, v., Zhou, L., Zhou, L., Lo, Z. and Zhang, 7.Effect of stimulation of midbrain raphe nuclei on discharges ofnociceptive neurons in nucleus parafascicularis of thalamus andits significance in acupuncture analgesia. In: “Research onAcupuncture, Moxibustion, and Acupuncture Anesthesia”, Ed., T.Chang, Science Press. pp. 82-92, 1986.Zhang, X. T. Interaction in the thalamus of afferent impulses fromacupuncture point and site of pain. In: “Advances in Acupunctureand Acupuncture Anaesthesia.” Tiantan Xiii Beijing China. ThePeople’s Medical Publishing House. pp. 26,1979.Zhao, 7., Shao, D. and Yang, 1. Inhibition of evoked discharges ofposterior nuclear group of thalamus induced by stimulation ofcentral gray matter, caudate nucleus and electroacupuncture in theawake rabbits. In: “Research on Acupuncture, Moxibustion, andAcupuncture Anesthesia”, Ed., T. Chang, Science Press. pp, 138-145, 1986.Zhou, G. Z., Chen, M. P. and Xu, S. F. The relationship between theontogenetic development of opiate receptors and acupunctureanalgesia in rabbits. Acupuncture and Electro-Therapeutic Res.,Tnt. J. 7:7-16, 1982.Zhu, L. X., Li, C. Y., Shi, Q. Y. and Ji, C. F. Effect of iontophoretic5-HT on raphe-spinal neurons. Acta Pharmacol. Sinica. 7:394-398,1986.Zou, G., Yi, Q., Wu, S., Lu, V., Wang, F., Yu, Y., Ji, X., Zhang, 7 andZhao, 0. Enkephalin involvement in acupuncture analgesiaradioimmunoassay, In: “Research on Acupuncture, Moxibustion, andAcupuncture Anesthesia”, Ed., T.Chang, Science Press. pp. 267-278,1986.166APPENDIX 1These programs were written or modified by Mr. Roland Burton(Faculty of Pharmaceutical Sciences, UBC) since 1985. They are writtenin Turbo Pascal 5.5 and were run on a PC clone, but typically were usedon an AT. They used a Hercules card to display monochrome graphics anda PCL-720 counter and timer car for counting.The program modules are as follows:S4This has a boxcar integrator in which the bin was 0.5 s. Thereare three channels; one stimulation channel which controls onset oroffset of the radiant heat lamp. The other two channels are used forcounting neurone spikes. It has four cycles per data set and 180 binsper cycle.$41This also has a boxcar integrator, however, the bin width isadjustable in 18 ths of a second. There are two channels forstimulation and one channel for counting the neuronal signal. It hasone cycle per data set and 180 bins per cycle.167AppendixS46This program uses a boxcar integrator in which the bin width isadjustable to 1 ms or larger. An external trigger signal starts thefirst cycle. The #15 bin emits a trigger which delivers a stimulus tothe DRN. There is only one channel for counting. It has 400 cycles perdata set and 50 bins per cycle.L.This program uses a boxcar integrator in which the bin width isadjustable from 0.001 ms to 1.0 ms. An external trigger signal startsthe first cycle. It has 40 cycles per data set.DumpsThis program is used for a screen dump of graphics to a HPLaserjet printer. It is used for all the above programs.GRThis program is used for text and graphics on Hercules and otherscreen and used for all above programs.168APPENDIX 2THE METHODS OF SEROTONIN-IMMIJNOREACTIVITY IN THE CENTRAL NERVOUS SYSTEMTissue preparationThe animals were anaesthetized with sodium pentobarbital(Nembutal, 60 mg/kg bodyweight, i..p.) and perfused through the leftventricle. The blood was washed out with cold, oxygen enriched, Ca2+ -free Tyrode’s buffer (50 ml at 4oC) followed by 500 ml ice-cold 4% (w/v)paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.3 for 30 mm ata pressure of 70 mm Hg. The brains were quickly removed and soaked for90 mm in fresh fixative at 4oC. After fixation, they were rinsed in 5% sucrose dissolved in 0.1 M sodium phosphate buffer (pH 7.3) at 4oC forat least 1 day. Tissue pieces were then frozen on a cryostat withpowdered dry ice. Sections, 10 urn thick, were cut on a cryostat(Dittes, Heidelberg, FRG) at -25oC and subsequently mounted on glassslides coated with chrome alum gelatine to prevent detachment of thesections during the incubation procedure. The sections were immediatelyprocessed for immunohistochemistry.IminunofI uorescence procedureThe indirect immunohistochemical procedure was used. The sectionswere first incubated in 0.1 M phosphate buffered saline (PBS) (pH 7.3)at room temperature for 10-30 mm. They were then incubated at 4°C for18 h with 5-HT antiserum (76 mg/ml) or pre-immune serum that had been169Appendixdiluted 1:500 or 1:1000 with PBS containing 0.1 % Triton X-100. Afterrinsing in PBS for 30 mm at room temperature, the sections were treatedwith fluorescein isothiocyanate conjugated sheep anti-rabbit immuno-qglobulin diluted 1:16 with PBS containing 0.1 % Triton-X-100. Afterthis final incubation of 30 mm at room temperature, the sections wererinsed once again in PBS for 30 mm and mounted under a coverslip in amixture of glycerine-PBS (3:1, v/v). The sections were examined in aZeiss Universal microscope equipped with incident illumination forfluorescence. Kodak Tri-X film was used for photomicrography havingexposure times between 5 s and 20 s. Alternate sections were mounted onseparate slides and stained with cresyl violet.170


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items