Die nicht gekürzte Version dieser Darstellung zur Akupunkturgrundlagenforschung finden Sie in: 
Clinical Acupuncture Scientific Basis
von Stux, G und Hammerschlag, R (Eds.), Springer Verlag Heidelberg

Mechanisms of Acupuncture Analgesia Produced by Low-Frequency
Electrical Stimulation of Acupuncture Points

Chifuyu Takeshige

Three noteworthy phenomena have been recognized in surgical acupuncture analgesia (AA) produced by low-frequency electrical stimulation of acupuncture points (APs): 1) Consciousness is maintained, allowing the patient to talk during surgery; 2) Stimulation of specific acupuncture points is essential to maintain analgesia; and 3) Analgesia persists long after stimulation has been terminated, allowing the patient to move without pain after surgery. The mechanisms by which AA is produced might be clarified by investigating these phenomena. This review will explore possible mechanisms based on results from animal experiments.
Consciousness depends on activation of the brainstem ascending reticular activating system (RAS) that produces widespread stimulation of the cerebral cortex and non-specifically maintains consciousness through the reticular nucleus in the thalamus. The RAS is activated by collateral pathways that diverge from each specific sensory afferent pathway that projects to each sensory cortex. Neurophysiological research has shown that anesthetic drugs used during surgical operations inhibit activity of the RAS. Since consciousness is diminished under this condition, sensory information reaching the sensory cortex is not translated into perception. On the other hand, it is also commonly observed that normally painful stimuli are suppressed on the battlefield of war and on the playing field of aggressive sports such as rugby. Such analgesia is thought to be brought about by activation of the descending pain inhibitory system (DPIS) originating from the limbic system that blocks pain information as it enters the central nervous system. Consciousness can thus be maintained in such a condition. If stimulation of a specific acupuncture point activates the DPIS through a particular pathway connected to the brain system which suppresses pain, it can be assumed that AA is produced by activation of the DPIS. This assumption has been examined in our laboratories by several animal experiments.
  

1. Classification of acupuncture afferent and efferent pathways for producing acupuncture analgesia [11,12,13,16,22,23]

The neuronal structures comprising the AA-producing brain pathway can be identified when microelectrode stimulation induces analgesia in a manner that mimics AA and by tissue ablation that results in subsequent blockage of AA. However, the nature of the analgesia produced depends upon the brain areas stimulated and can be classified into two categories. The first category includes analgesia that 1) is naloxone-reversible, 2) disappears after hypophysectomy, 3) persists long after stimulation of the acupoint is terminated, and 4) exhibits individual variation in effectiveness. These features are similar to those of AA. In this category, brain potentials are evoked by stimulation of acupoints in the same areas that produce analgesia. Stimulation of brain areas associated with the second category produces analgesia that 1) is not naloxone-reversible, 2) is not affected by hypophysectomy, 3) is produced only during stimulation, and 4) exhibits no individual variation in effectiveness. Evoked potentials are not obtained from brain regions producing analgesia of this second category, but non-synchronized neuronal activities are obtained by stimulation of acupoints [16].
Brain regions producing analgesia of the first category appear to comprise an afferent pathway for acupuncture, since the pituitary gland is involved in this analgesia and electrical potentials are evoked in these brain regions by stimulation of acupoints. Similarly, areas producing analgesia related to the second category appear to comprise an efferent pathway for acupuncture, since the pituitary gland is not involved and synchronized electrical potentials are not evoked in these regions by stimulation of acupoints [12,16,19]. All brain regions producing analgesia associated with the second category seem to be connected to the DPIS; AA is produced by activation of the DPIS that is excited by stimulation of specific acupoints through a particular pathway connected to the DPIS. This DPIS-producing analgesia related to the second category is defined as the acupuncture efferent pathway, whereas the particular pathway from specific acupoints to the DPIS is defined as the acupuncture afferent pathway.

1a. Acupuncture efferent pathway [13,16,24]

AA can be abolished by concurrent lesions of the Raphe nucleus and the reticular paragigantocellular nucleus that are known as the origins of the serotonergic and the noradrenergic descending pain-inhibitory systems. Stimulation of these nuclei respectively produces serotonergic and noradrenergic analgesia of the second category. The final production of AA is induced by activation of these descending pain-inhibitory systems. The descending pain-inhibitory pathway serves as the acupuncture efferent pathway from the hypothalamic ventromedian nucleus (HVM); it is divided into two parts that connect to the descending serotonergic and noradrenergic systems. The posterior part of the hypothalamic arcuate nucleus (P-HARN) is anatomically connected to the HVM. Analgesia produced by stimulation of both the HVM and the P-HARN is associated with the second category. Synaptic transmission from the P-HARN to the HVM is apparently dopaminergic, since analgesia produced by stimulation of the P-HARN is blocked by lesions of the HVM or by dopamine antagonists (Fig.1, bottom).

1b. Acupuncture afferent pathway [11,12,23] (Fig.1, top)

The acupuncture afferent pathway starts from an acupoint, ascends through the contralateral anterolateral tract to the dorsal periaqueductal central gray, and reaches the medial part of the hypothalamic arcuate nucleus (M-HARN). Brain regions belonging to the AA afferent pathway can be identified by exhibition of analgesia of the first group related to anatomically known connections. The rostral and caudal relations between these regions have been identified by the loss of stimulation-produced analgesia of the caudal region that follows lesions of the rostral region. These relations are shown in Figures 1 and 2 [23].

1c. Synaptic connections between acupuncture afferent and efferent pathways [25,28].

The final region of the acupuncture afferent pathway is found in the M-HARN, which is anatomically close to the P-HARN, the initial region of the acupuncture efferent pathway (Fig.3). Microinjection of the dopamine antagonist haloperidol antagonizes AA dose-dependently while microinjection of dopamine into the P-HARN induces a dose-dependent analgesia. Dopamine thus seems to serve as the neurotransmitter between the M-HARN and the P-HARN, i.e. as the neurotransmitter at the interface between the acupuncture afferent and efferent pathways. This possibility is further supported by neuronal activity in the P-HARN. Neurons in the P-HARN that respond to acupoint stimulation also respond to iontophoretically administered dopamine, whereas neurons in the M-HARN that do not respond to acupoint stimulation also do not respond to iontophoretically administered dopamine [25] (Fig.4).
A branch of the acupuncture afferent pathway ascending to the M-HARN diverges at the lateral hypothalamus (LH) to reach the pituitary gland. Lesions of brain nuclei near this pathway to the pituitary, e.g. the preoptic area (POA) or the median eminence (ME), abolish AA. Electrical potentials are evoked in these brain areas by stimulation of acupoints, but stimulation of these particular brain structures does not produce analgesia [25,28] (Fig.1, 2 and 5). Since both acupuncture analgesia and pain relief produced by stimulation of the acupuncture afferent pathway to the M-HARN are abolished by hypophysectomy, -endorphin released from the pituitary gland may play an essential role in dopaminergic transmission in the P-HARN [25] (Fig.5). Microinjection of naloxone to the P-HARN antagonizes AA dose-dependently and microinjection of -endorphin or morphine produces analgesia dose-dependently. Analgesia produced by microinjection of -endorphin disappears after denervation of the M-HARN, but analgesia produced by microinjection of dopamine to the P-HARN remains [25] (Fig.6). These findings suggest that -endorphin might act presynaptically at dopaminergic synapses in the P-HARN. This notion is further supported by the activity of P-HARN neurons. Neuronal activity in the P-HARN that occurs in response to acupuncture stimulation is not affected by iontophoretic administration of morphine or by ultramicroinjection of -endorphin via picosprizer [22] (Fig.7).
Since morphine and -endorphin act similarly in the P-HARN, -endorphin released from the pituitary gland might be the neurohumoral factor acting presynaptically on axon terminals of the M-HARN neurons that innervate P-HARN neurons. Although microinjected -endorphin into the P-HARN produces analgesia, electrical stimulation of the POA or ME in the pathway to the pituitary gland does not; therefore, the released amount of -endorphin by such stimulation is not sufficient to activate the P-HARN neurons without afferent impulse from the M-HARN. Morphine and -endorphin might also act in other areas of the AA afferent pathway. This possibility was explored by recording electrical potentials evoked by stimulation of the acupoint in the final station of the AA afferent pathway, the M-HARN. Such potentials are enhanced by intravenously administered morphine (0.5 mg/kg) and are abolished by hypophysectomy. The abolished evoked-potentials are temporarily restored by morphine [12] (Fig. 8). Therefore, sites responsive to -endorphin released from the pituitary gland might be widespread in the AA afferent pathway. Opioid receptors have also been reported in many regions of the acupuncture afferent pathway [1,5,10].
  

2. Stimulation of specific acupoints for production of acupuncture analgesia [23]

Low-frequency (1 Hz) electrical stimulation of the first dorsal finger muscle and the anterior tibial muscle in rats [11, 23] that are the muscles underlying, respectively, the human LI 4 (Hegu) and ST 36 (Zusanli) acupoints, produces behavioral analgesia, as evaluated by tail-flick latency. The intensity of electrical stimulation must be sufficient to cause muscle contraction in order to obtain AA. In contrast to this effect, stimulation of other muscles does not produce behavioral analgesia. Hence, the Hegu and Zusanli acupoints seem uniquely able to activate the DPIS through the particular pathway connected to the DPIS [3].
  

3. Differentiation of acupoints and non-acupoints by responses of central neuronal structures [11,14,15,17,18]

Potentials can be evoked specifically in the bilateral dorsal areas of the periaqueductal central gray (D-PAG) by stimulation of the muscles underlying the Hegu and Zusanli acupoints, but not by stimulation of other muscles (Fig.9). Lesions of the D-PAG abolish AA. Microelectrode stimulation of this region produces analgesia of the first category that can be reversed by either naloxone or hypophysectomy. Stimulation of the auricular levator muscle beneath the X 18 (Chihmo) acupoint in rabbits elicits evokes potentials in the D-PAG [11,23]. Stimulus conditions as stated above which lead to AA were confirmed by potentials in the D-PAG. Therefore, only three acupoints for producing AA have been identified: Hegu, Zunsanli, and Chihmo.
Stimulation of muscles beneath Hegu and Zusanli also produces nonspecific potentials bilaterally in the lateral parts of the periaqueductal central gray (L-PAG) [17] (Fig.9). Potentials in the L-PAG are gradually decreased by 1 Hz repetitive stimulation of these muscles and disappear completely 10 minutes after the onset of stimulation [15,17] (Fig.10). Hence, potentials in the L-PAG are inhibited by such stimulation in a self-inhibiting fashion. Lesions of the L-PAG do not affect AA, but analgesia is produced by stimulation of the rostral L-PAG. This analgesia is largely reversible with dexamethasone and the dexamethasone-insensitive portion is readily blocked by naloxone or hypophysectomy. Hence, acupoints are connected via the D-PAG to the particular pathway that is not self-inhibited during the production of AA. On the other hand, acupoints as well as non-acupoints are connected to the other, self-inhibiting pathway nonspecifically, via the L-PAG. The latter brain region belongs to a pathway distinct from the AA afferent pathway, whose analgesia production is self-inhibiting (Fig.11). These results imply that acupoints and non-acupoints can be differentiated by their connections with different analgesia-producing central pathways [14,17] (Figs.1, 11).
  

Analgesia produced by 0.5 mg/kg morphine is of a similar degree to that produced by low frequency electroacupuncture. In addition, both types of analgesia are abolished by hypophysectomy, by lesions of either the AA afferent or efferent pathways, by naloxone or by antagonists of transmitters involved in the AA efferent pathway. In addition, individual variation in effectiveness between AA and morphine analgesia are highly correlated. Animals can be classified as responders or non-responders by the presence or absence of a significant increase (P < 0.05) in tail-flick latency.
  

5. Activation of the spinal acupuncture analgesia afferent pathway by morphine [8,11,12,14,18]

Potentials evoked in the D-PAG by stimulation of acupoints are blocked by contralateral lesions of the anterolateral tract or by intrathecal administration of the antiserum to methionine-enkephalin (Met-enkephalin). These potentials are also blocked by naloxone, but not by the administration of antisera to leucine-enkephalin or dynorphin [18] (Fig.12) supporting the involvement of a met-enkephalin pathway that is activated by morphine. In AA-responder animals, dose-response curves of analgesia were obtained for both low and high doses of morphine, administered either intraperitoneally or intrathecally. However, in non-responder animals, only a single dose-response curve for higher doses of morphine was obtained. In AA responders, bilateral lesions of the anterolateral tract, or lesions of the D-PAG that is part of the AA afferent pathway abolished dose-dependent responses to low doses of morphine without affecting the dose-response to high doses of morphine. Therefore, morphine analgesia produced by lower doses is probably induced by activation of the AA afferent pathway through Met-enkephalin receptors in the spinal cord [18]. Such receptors in the spinal AA afferent pathway are likely to be those that are activated by intraperitoneal morphine at 0.5 mg/kg or by intrathecal morphine at 0.05 mg/kg, that produce morphine analgesia of a degree similar to that of AA [8,11] (Fig.13). This mechanism may explain the reason for the similarity between AA and morphine analgesia.
  

Acupuncture analgesia is produced by activation of the DPIS through a specific pathway connected to the acupoints while still allowing maintenance of consciousness. The AIS, in contrast, is activated by stimulation of acupoints or non-acupoints, leading to a nonspecific inhibition of different interconnected pathways. Therefore, acupoints and non- acupoints can be distinguished by their anatomically distinct brain pathways. The after-effects of AA might be produced by the actions of an increased amount of -endorphin released from the pituitary gland on components of the AA-producing pathway.
  

References

1. Atweh SF, Kuhar MJ. Autoradiographic localization of opiate receptors in rat brain: the brainstem. Brain Res 1997;129:1-12.

2. Arai T, Guo SY, Takeshige C. Cholecystokinin in the analgesia inhibitory system and its antagonists in this system. J Showa Med Assoc 1992;52:58-67.

3. Huang SF, Luo CP, Takeshige C. Identity of a central analgesia producing mechanism in Ho-ku point stimulation with that in Tsusanli point stimulation. J Showa Med Assoc 1998;48:485-492.

4. Lung CH, Sun AC, Tsao CJ, Chang,YL, Fan L. An observation of the humoral factor in acupuncture analgesia in rats. Amer J Chin Med 1974;2:203-205.

5. Malizia E, Andreucci G, Paolucci D, Grescenzi F, Fabbri A, Fraioli F. Electroacupuncture and peripheral -endorphin and ACTH levels. Lancet 1979; 535-536.

6. Pert CB, Snyder SH. Opiate receptor: demonstration in nervous tissue. Science 1973;179:1011-1014.

7. Sato T, Hishida F, Luo CP, Tsuchiya M, Takeshige C. Relations of adrenal gland and sodium ions in production of acupuncture and non-acupuncture points stimulation-produced analgesia. J Showa Med Assoc 1989;49:286-294.

8. Sato T, Takeshige C, Shimizu S. Morphine analgesia mediated by activation of the acupuncture-analgesia-producing system. Acupunct Electro-Ther Res 1991;16:13-26.

9. Sjolund B, Terenius L, Erikson M. Increased cerebrospinal fluid levels of endorphins after electro-acupuncture. Acta Physiol Scand 1977;100:382-384.

*10. Snyder SH. Opiate receptor in normal and drug altered brain function. Nature 1955;257:185-189.

*11. Takeshige C. Mechanism of acupuncture analgesia based on animal experiments. In: Scientific Bases of Acupuncture, Pomeranz B, Stux G (eds) Springer-Verlag, 1990, pp..

12. Takeshige C. Mechanism of acupuncture analgesia (AA) caused by low frequency stimulation of the acupuncture point based on animal experiments, Part1: Acupuncture afferent and efferent pathways and the nature of AA. Acupunct Scient Int J 1990;1:75-88.

13. Takeshige C, Sato T, Komugi H. Role of peri-aqueductal central gray in acupuncture analgesia. Acupunct Electro-Ther Res 1980;5:323-337.

14. Takeshige C. Differentiation between acupuncture and non-acupuncture points by association with analgesia inhibitory system. Acupunct Electro-Ther Res 1985;10:195-203.

15. Takeshige C. Inhibition associated with acupuncture analgesia: inactivation of hypersensitive neurons. Charlazonits N, Gola M (eds) Alan R. Liss, Inc., 1987, pp.255-262.

16. Takeshige C, Kamada Y, Hisamatu T. Commonly responsive neurons in the periaqueductal gray matter and midbrain reticular formation of rabbits to acupuncture stimulation, inversion, pressure on body parts and morphine. Acupunct Electro-Ther Res 1981;6: 57-74.

17. Takeshige C, Kobori M, Hishida F, Luo CP, Usamai S. Analgesia inhibitory system involvement in nonacupuncture point-stimulation-produced analgesia. Brain Res Bull 1992;28:379-391.

18. Takeshige C, Luo CP, Hishida F, Igarashi O. Differentiation of acupuncture and non-acupuncture points by difference of associated opioids in the spinal cord in production of analgesia by acupuncture and non-acupuncture point stimulation, and relations between sodium and those opioids. Acupunct Electro-Ther Res 1990;15:193-209.

19. Takeshige C, Luo CP, Kamada Y, Oka K, Murai M, Hisamatu T. Relation between midbrain neurons (Periqueductal central gray and midbrain reticular formation) and acupuncture analgesia, animal hypnosis. In Bonica J et al (eds). Advances Pain Res Ther 3: 1979;615-621

20. Takeshige C, Mera H, Hisamatu T, Tanaka M, Hishida F. Inhibition of the analgesia inhibitory system by D-phenylalanine and proglumide. Brain Res Bull 1991;26:385-391.

21. Takeshige C, Murai M, Tanaka M, Hachisu M. Parallel individual variations in effectiveness of acupuncture, morphine analgesia, and dorsal PAG-SPA and their abolition by D-phenylalanine. In Bonica J et al (eds). Adv Pain Res Ther 1983;5:563-569.

22. Takeshige C, Nakanura A, Asamoto S, Arai T. Positive feedback action of pituitary -endorphin on acupuncture analgesia afferent pathway. Brain Res Bull 1992;29: 37-44.

23. Takeshige C, Oka K, Mizuno T, Hisamatu T, Luo CP, Kobori M, Mera H, Fang TQ. The acupuncture point and its connecting central pathway for producing acupuncture analgesia. Brain Res Bull 1993;30:53-67.

24. Takeshige C, Sato T, Mera T, Hisamitsu T, Fang TO. Descending pain inhibitory system involved in acupuncture analgesia. Brain Res Bull 29:1992;617-634.

25. Takeshige C, Tsutiya M, Guo SY, Sato T. Dopaminergic transmission in the hypothalamic arcuate nucleus to produce acupuncture analgesia in correlation with the pituitary gland. Brain Res Bull 1991;26:113-122.

26. Takeshige C, Tsutiya M, Zhao W, Guo S. Analgesia produced by pituitary ACTH and dopaminergic transmission in the arcuate. Brain Res Bull 1991;26:779-788.

27. Takeshige C, Tanaka M, Sato T, Hishida F. Mechanism of individual variation in effectiveness of acupuncture analgesia based on animal experiments. Eur J Pain 1990;11:109-113.

28. Takeshige C, Zhao WH, Guo SY. Convergence from the preoptic area and arcuate nucleus to the median eminence in acupuncture and non-acupuncture point stimulation analgesia. Brain Res Bull 1991;26:771-778.

29. Takeshige C, Luo CP, Kamada Y. Modulation of EGG and unit discharges of deep structure of brain during acupunctureal stimulation and by hypnosis of rabbits. In Bonica JJ, Albe-Fressard D (ed.). Adv Pain Res Ther, Raven Press 1976;1:781-785.

30. Toyoda I, Takeshige C. Changes in characteristics of analgesia produced by difference in duration of stimulation of acupuncture points. J Showa Med Assoc 1992;52:1-7.

31. Xu M, Aiuchi T, Nakaya K, Arakawa H, Maeda M, Tsuji A, Kato T, Takeshige C, Nakamura Y. Effect of low-frequency electric stimulation on in vivo release of cholecystokinin-like immunoreactivity in medial thalamus of conscious rat. Neurosci Let 1990;118: 205-207.