- Email: [email protected]
Long-term depression of the human masseter inhibitory reflex
Neuroscience Letters 329 (2002) 265–268 www.elsevier.com/locate/neulet
Long-term depression of the human masseter inhibitory reflex Jens Ellrich*, Anila Schorr Department of Experimental and Clinical Pharmacology and Toxicology, Emil-Fischer-Center, Friedrich-Alexander-University of ErlangenNuremberg, Fahrstrasse 17, D-91054 Erlangen, Germany Received 18 April 2002; received in revised form 6 June 2002; accepted 10 June 2002
Abstract Long-term depression (LTD) of synaptic transmission is reliably induced by low-frequency stimulation (LFS) of nociceptive afferents in vitro. LTD can only exceptionally be induced in anesthetized animals. In order to fill the gap between the in vitro cell studies and the in vivo situation, the effects of LFS on the masseter inhibitory reflex (MIR) were investigated in man. Noxious LFS of mental nerve afferents caused a significant depression of the early MIR1 and the late MIR2 components. Whereas the onset latency (125%), the duration (242%) and the integral (168%) of the MIR2 were strongly modulated after LFS, only the integral (135%) of the MIR1 significantly changed. The results document a long-term depression of trigeminal somatosensory processing in man. The different effects of LFS on the two components of the MIR may point to central mechanisms of LTD. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Brainstem; Orofacial; Pain; Plasticity; Trigeminal
Low-frequency stimulation (LFS) of presynaptic nerve fibers causes long-term depression (LTD) of synaptic strength in the hippocampus in vitro [8,16]. LTD is an established experimental model of synaptic plasticity and may also play a role in the nociceptive system [1,18]. In rat spinal cord slice with a long dorsal root attached substantia gelatinosa neurons were intracellularly recorded [19]. LFS with stimulus intensities that effectively activated the vast majority of nociceptive Ad-fibers produced a robust LTD in all neurons tested. A similar experiment was conducted in adult rats under general anesthesia [14]. Field potentials evoked by electric stimulation of the sciatic nerve were recorded from the lumbar spinal dorsal horn. LFS at Ad-fiber intensity was tested in 11 rats. In only two rats LTD was induced. Whereas LFS reliably induces LTD in vitro, it only exceptionally induces LTD in vivo. Many experiments suggest that LTD of basal transmission is not readily induced either in awake or anesthetized rats [6,7,11,15]. Thus, the significance of LTD for memory processes has been questioned due to reported failures to induce persistent LTD in vivo. Consequently, LTD can hardly be applied to human beings. In order to fill the gap between the in vitro cell studies and the in vivo situation we investigated the * Corresponding author. Tel.: 149-9131-852-2003; fax: 1499131-852-2774. E-mail address: [email protected] (J. Ellrich).
effects of LFS of trigeminal afferents on somatosensory processing in healthy volunteers by applying the masseter inhibitory reflex. The experiments were conducted in five female and four male healthy volunteers between 22 and 35 years of age, who gave their informed consent prior to their inclusion in the study according to the 1964 Declaration of Helsinki (revised in Edinburgh, 2000). Trigeminal somatosensory processing was measured by the masseter inhibitory reflex (MIR) that consists of an early MIR1 and a late MIR2. The MIR was evoked by electric stimulation of the mental nerve area in close vicinity to the lower lip. Rectangular pulses (500 ms duration) were applied by a small cathode (2 mm diameter) and a large anode (9 mm diameter). The subjects were comfortably sitting on a chair with the eyes open. In three and six volunteers the right and left mental nerve area, respectively, was stimulated. Electromyographic activity (EMG) was recorded from both masseter muscles by surface electrodes with one electrode over the inferior and anterior region of the muscle and another electrode at the temple (bandpass 100–1000 Hz). A ground electrode was fixed at the right forearm. EMG signals were amplified and digitized by a micro CED1401 A/D-Converter (CED, Cambridge, UK) with a 10 000 Hz sampling rate and analyzed by the Signal Software (CED, Cambridge, UK). EMG sweeps were
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 66 9- 9
266
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
recorded from 100 ms before to 200 ms after the electric stimulus. Before applying an electric stimulus the subject was asked to adjust the EMG activity of both masseter muscles to at least 90% of maximum strength by clenching the teeth, controlled by visual and acoustic feedback, because the duration of the suppression periods was inversely related to the magnitude of clenching [12]. The MIR was elicited 1–2 s after having reached this level. The volunteers were requested not to relax masseter muscles when recognizing the stimulus but to maintain the activity for another second. The contractions were serially repeated after 30 s of rest. Individual thresholds for detection (I0) and pain (IP) were determined by applying two series of electric pulses with inand decreasing stimulus intensity according to the method of limits. After each electric stimulus the volunteer was asked to rate the stimulus according to an ordinal scale (0 ¼ no perception, 1–39 ¼ perception but no pricking painful sensation, 40 ¼ slightly pricking painful (pain threshold), 100 ¼ unbearable painful). Based on the threshold detection the test stimulus for the reflex study was adjusted to an intensity clearly above the individual pain threshold and sufficient to reliably evoke the MIR. The MIR was elicited in blocks of 8 test stimuli each with an interstimulus interval of about 30 s. Test stimulation blocks were repeated every 8 min. After the third stimulus block LFS was applied for the next 20 min. LFS consisted of 1200 electric rectangular pulses (500 ms duration) applied with the same painful stimulus intensity as for the reflexes but with a frequency of 1 Hz. A total of 8 min after the end of LFS the test stimulation blocks were continued for about 1 h. Thus, three test stimulus blocks were applied before LFS (time 220, 212 and 24 min) and eight test stimulus blocks were applied after LFS (time 28, 36, 44, 52, 60, 68, 76 and 84 min). The eight trials of each test stimulus block were rectified and averaged off-line. The mean amplitude of the time window from 2100 to 210 ms before the electric stimulus was set as 100% (baseline EMG activity). A MIR was defined as more than 50% depression of baseline EMG activity [3]. Onset latencies and durations of MIR1 and MIR2 were measured at the points where rectified EMG activity crossed a level of 80% of the prestimulus EMG. Integrals of MIR1 and MIR2 were calculated in an individual fixed time window that lasted from the earliest onset latency to the latest end latency of all individual test stimulus block averages. Mean onset latencies, durations, integrals and pain ratings recorded in the three test stimulus blocks before the LFS were set as 100%. Data were statistically analyzed by analysis of variance (one-way analysis of variance, ANOVA, repeated measures design) and by the least significant difference (LSD) post hoc test. Arithmetic mean (mean) and standard error (SE) were calculated. The electric thresholds for detection (I0) and pain (IP) were 0.9 ^ 0.2 and 4.7 ^ 0.8 mA (mean ^ SE). The mean test stimulus intensity of 8.9 ^ 0.3 mA evoked a pinprick
like pain sensation that was rated at 60.9 ^ 4.6 (mean ^ SE). The test stimulus intensity expressed as multiples of the individual IP and I0 was (2.3 ^ 0.3) £ IP and (11.9 ^ 1.6) £ I0 (mean ^ SE). The test stimulus elicited an MIR1 in seven out of nine volunteers and an MIR2 in all nine volunteers. After LFS the MIR significantly changed, whereas MIR1 and MIR2 were differentially affected (Fig. 1). Electric stimulation evoked in the ipsilateral masseter muscle an MIR1 with an onset latency of 12.4 ^ 1.0 ms and a duration of 17.4 ^ 1.7 ms. After LFS, onset latency (12.0 ^ 0.6 ms; F ¼ 0:63, P ¼ 0:78) and duration (16.1 ^ 1.5 ms; F ¼ 0:78, P ¼ 0:65) remained statistically unchanged as calculated by the one-way ANOVA (repeated measures design). The onset latency of MIR2 significantly increased from 45.3 ^ 1.3 to 53.9 ^ 3.6 ms after LFS (F ¼ 2:18, P , 0:05). This change was mainly due to an increase of onset latencies in the test stimulus blocks at time points 60 and 76 min as shown by the LSD test (P , 0:01). After LFS, the MIR2 duration significantly decreased from
Fig. 1. Masseter inhibitory reflex responses of one specimen before and after LFS. EMG activity (arbitrary units) of the ipsilateral masseter muscle was evoked by noxious electric stimulation of right mental nerve afferents. The upper graph shows one test stimulus block with eight sweeps just before LFS (4 min before LFS). The lower graph also demonstrates a test stimulus block of the same volunteer, but just after conditioning LFS (8 min after LFS).
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
32.6 ^ 2.2 to 20.9 ^ 3.9 ms (F ¼ 5:97, P , 0:00001). The LSD test stressed that this effect is mainly caused by a decrease in seven out eight test stimulus blocks after LFS (P , 0:01). MIR1 and MIR2 integrals significantly increased after LFS to 136.0 ^ 6.7% (F ¼ 4:64, P , 0:0001) and 171.6 ^ 5.4% (F ¼ 8:54, P , 0:000001). Whereas the increase of MIR1 integrals was mainly due to a significant change in the last two test stimulus blocks after LFS (P , 0:05), the LSD test documented for the MIR2 that all integrals after LFS were significantly larger than those in the stimulus blocks before LFS (P , 0:05). The baseline EMG activity remained statistically unchanged after LFS (F ¼ 1:37, P ¼ 0:21) (Fig. 2). The MIR1 of the contralateral masseter muscle had an onset latency of 11.7 ^ 1.2 ms and a duration of 17.8 ^ 1.5 ms. After LFS, onset latency (11.9 ^ 0.8 ms; F ¼ 0:75, P ¼ 0:68) and duration (15.9 ^ 1.9 ms; F ¼ 0:70, P ¼ 0:72) remained statistically unchanged as calculated by the one-way ANOVA (repeated measures design). The onset latency of MIR2 significantly increased from 46.2 ^ 1.0 to 59.9 ^ 4.9 ms after LFS (F ¼ 2:30, P , 0:05). The LSD test explained this significant effect by an increase of onset latencies in five out eight test stimu-
Fig. 2. Reflex integrals and baseline EMG activities (top), onset latencies and durations (bottom) of ipsilateral MIR1 and MIR2 components of all volunteers (n ¼ 9). Mean integrals, onset latencies and durations recorded in the three test stimulus blocks before the LFS were set as 100% (dotted line). Data are presented as arithmetic mean ^ SE.
267
lus blocks after LFS (P , 0:05). After LFS, the MIR2 duration significantly decreased from 29.6 ^ 3.0 to 17.1 ^ 4.1 ms (F ¼ 6:85, P , 0:000001). This strong reduction of MIR2 duration was mainly due to a significant decrease in all post LFS stimulus blocks in comparison to all pre LFS blocks (LSD test: P , 0:01). MIR1 and MIR2 integrals significantly increased after LFS to 131.1 ^ 2.8% (F ¼ 2:03, P , 0:05) and 164.5 ^ 3.3% (F ¼ 6:46, P , 0:000001). Whereas the increase of MIR1 integrals was mainly due to a significant change in the first and last test stimulus blocks after LFS (P , 0:05), the LSD test documented for the MIR2 that all integrals after LFS were significantly larger than those in the stimulus blocks before LFS (P , 0:01). The baseline EMG activity remained statistically unchanged after LFS (F ¼ 0:97, P ¼ 0:47). The ratings of stimulus strength significantly decreased after LFS to 90.7 ^ 3.3% compared to the pre LFS stimulations (F ¼ 7:56, P , 0:000001). The LSD test stressed that all ratings after LFS were significantly smaller than those in the stimulus blocks before LFS (P , 0:01). Noxious low-frequency stimulation of mental nerve afferents caused a decrease of the MIR and a decrease of pain ratings. The depressive effect of LFS was stronger on the MIR2 than on the MIR1. This depression of the reflex and the pain sensation lasted for at least 1 h and therefore meets the criteria of long-term depression [1]. The MIR is a trigemino-trigeminal brainstem reflex that is connected with orofacial nociception [9]. Both components, the early MIR1 and the late MIR2, can be elicited by electric stimulation, by mechanical stimulation and by selective activation of nociceptors using laser radiant heat [2,10]. Based on the detection of electric thresholds and on modulation studies both components of the MIR are probably mediated by wide-dynamic-range neurons (WDR) that get input from low-threshold mechanoreceptive and nociceptive afferents [4,10,13]. The MIR1 is probably mediated by pontine WDR neurons within the rostral part of the spinal trigeminal nucleus (STN) or the interstitial nucleus (ISVT), and WDR neurons mediating the medullary MIR2 are probably located in the caudal part of the STN or the medullary ISVT. In a rat transverse spinal cord slice LFS of dorsal root afferents clearly depressed substantia gelatinosa neurons EPSPs amplitudes and initial slopes [19]. The depression effect of LFS on the MIR2 is similar in the present study. The MIR2 integral increased by about 68% of control, corresponding to an inhibition of MIR2, the onset latencies increased by about 25%. The effects are compatible with a modulation of EPSP amplitude and slope in nociceptive sensory neurons. In vivo rat experiments showed a maximum suppression of neuronal activity in the hippocampal CA1 region to about 49% of control after LFS [1]. Many in vivo animal experimental studies did not find any effect of LFS on different populations of central neurons [6,7,11,15]. In contrast, LFS was effective in all investigated volunteers in the present study.
268
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
Whereas the MIR2 was clearly depressed after LFS, the effects on the MIR1 were less pronounced. Onset latency and duration did not change, and the increase of the MIR1 integral was mainly due to an increase in the last test stimulus blocks. These different effects on MIR2 and MIR1 were also true, when analyzing only those experiments where MIR1 and MIR2 were simultaneously evoked. Thus, LFS was differentially effective on MIR1 and MIR2. In animal experimental studies LTD phenomena were investigated at the first synapse between the nociceptive afferent fiber and the secondary sensory neuron in the substantia gelatinosa at the level of the spinal dorsal horn [14,19]. Based on the assumption that LFS exclusively depresses the strength of the first nociceptive synapse, the LTD effect on MIR1 and MIR2 should be very similar. The different effects on MIR1 and MIR2 might be due to a different contribution of nonnociceptive afferents to the reflex arc. In a previous study it was shown, that the electric threshold of the MIR1 is significantly higher than that of the MIR2 [10]. These data contradicts a stronger involvement of non-nociceptive afferents in the MIR1 reflex arc. Furthermore, habituation of MIR1 and MIR2 also might cause an inhibition. The MIR1 does not habituate even with stimulation frequencies of about 3 Hz, the MIR2 slightly habituates with interstimulus intervals of 5 s and never habituates with more than 20 s interstimulus intervals [5]. Alternatively, the different modulations of MIR1 and MIR2 may be due to LFS effects beyond the first nociceptive synapse. Noxious LFS may cause a decrease of presynaptic inhibition, that is involved in generating the MIR2 [17]. The different effectiveness of LFS on nociceptive processing in the spinal cord slice in vitro and in the anesthetized rat in vivo may arouse suspicion that central mechanisms may be involved in LTD [14]. In summary, this study documents a long-term depression of trigeminal somatosensory processing in man. The different effects of LFS on the two components of the MIR may point to central mechanisms of LTD. This study was supported by DFG grant SFB353-B17. [1] Braunewell, K.H. and Manahan-Vaughan, D., Long-term depression: a cellular basis for learning? Rev. Neurosci., 12 (2001) 121–140. [2] Cruccu, G., Agostino, R., Inghilleri, M., Manfredi, M. and Ongerboer de Visser, B.W., The masseter inhibitory reflex is evoked by innocuous stimuli and mediated by A beta afferent fibres, Exp. Brain Res., 77 (1989) 447–450. [3] Cruccu, G., Agostino, R., Lahuerta, L. and Manfredi, M., Inhibition of jaw-closing muscles by electrical stimulation of the ophthalmic division in man, Brain Res., 371 (1986) 298–304.
[4] Cruccu, G. and Romaniello, A., Jaw-opening reflex after CO2 laser stimulation of the perioral region in man, Exp. Brain Res., 118 (1998) 564–568. [5] Desmedt, J.E. and Godaux, E., Habituation of exteroceptive suppression and of exteroceptive reflexes in man as influenced by voluntary contraction, Brain Res., 106 (1976) 21– 29. [6] Doyere, V., Errington, M.L., Laroche, S. and Bliss, T.V., Lowfrequency trains of paired stimuli induce long-term depression in area CA1 but not in dentate gyrus of the intact rat, Hippocampus, 6 (1996) 52–57. [7] Doyle, C.A., Cullen, W.K., Rowan, M.J. and Anwyl, R., Lowfrequency stimulation induces homosynaptic depotentiation but not long-term depression of synaptic transmission in the adult anaesthetized and awake rat hippocampus in vivo, Neuroscience, 77 (1997) 75–85. [8] Dudek, S.M. and Bear, M.F., Homosynaptic long-term depression in area CA1 of hippocampus and effects of Nmethyl-d-aspartate receptor blockade, Proc. Natl. Acad. Sci. USA, 89 (1992) 4363–4367. [9] Ellrich, J., Brainstem reflexes: probing human trigeminal nociception, News Physiol. Sci., 15 (2000) 94–97. [10] Ellrich, J., Hopf, H.C. and Treede, R.D., Nociceptive masseter inhibitory reflexes evoked by laser radiant heat and electrical stimuli, Brain Res., 764 (1997) 214–220. [11] Errington, M.L., Bliss, T.V., Richter-Levin, G., Yenk, K., Doyere, V. and Laroche, S., Stimulation at 1–5 Hz does not produce long-term depression or depotentiation in the hippocampus of the adult rat in vivo, J. Neurophysiol., 74 (1995) 1793–1799. [12] Fung, D.T., Hwang, J.C. and Poon, W.F., Effect of bite force on the masseteric electromyographic silent period in man, Arch. Oral Biol., 27 (1982) 577–580. [13] Hansen, P.O., Svensson, P., Arendt-Nielsen, L. and Jensen, T.S., Human masseter inhibitory reflexes evoked by repetitive electrical stimulation, Clin. Neurophysiol., 113 (2002) 236–242. [14] Liu, X.G., Morton, C.R., Azkue, J.J., Zimmermann, M. and Sandku¨ hler, J., Long-term depression of C-fibre-evoked spinal field potentials by stimulation of primary afferent A delta-fibres in the adult rat, Eur. J. Neurosci., 10 (1998) 3069–3075. [15] Manahan-Vaughan, D., Long-term depression in freely moving rats is dependent upon strain variation, induction protocol and behavioral state, Cereb. Cortex, 10 (2000) 482– 487. [16] Mulkey, R.M. and Malenka, R.C., Mechanisms underlying induction of homosynaptic long-term depression in area CA1 of the hippocampus, Neuron, 9 (1992) 967–975. [17] Rudomin, P. and Schmidt, R.F., Presynaptic inhibition in the vertebrate spinal cord revisited, Exp. Brain Res., 129 (1999) 1–37. [18] Sandku¨ hler, J., Learning and memory in pain pathways, Pain, 88 (2000) 113–118. [19] Sandku¨ hler, J., Chen, J.G., Cheng, G. and Randic, M., Lowfrequency stimulation of afferent Adelta-fibers induces long-term depression at primary afferent synapses with substantia gelatinosa neurons in the rat, J. Neurosci., 17 (1997) 6483–6491.
Long-term depression of the human masseter inhibitory reflex Jens Ellrich*, Anila Schorr Department of Experimental and Clinical Pharmacology and Toxicology, Emil-Fischer-Center, Friedrich-Alexander-University of ErlangenNuremberg, Fahrstrasse 17, D-91054 Erlangen, Germany Received 18 April 2002; received in revised form 6 June 2002; accepted 10 June 2002
Abstract Long-term depression (LTD) of synaptic transmission is reliably induced by low-frequency stimulation (LFS) of nociceptive afferents in vitro. LTD can only exceptionally be induced in anesthetized animals. In order to fill the gap between the in vitro cell studies and the in vivo situation, the effects of LFS on the masseter inhibitory reflex (MIR) were investigated in man. Noxious LFS of mental nerve afferents caused a significant depression of the early MIR1 and the late MIR2 components. Whereas the onset latency (125%), the duration (242%) and the integral (168%) of the MIR2 were strongly modulated after LFS, only the integral (135%) of the MIR1 significantly changed. The results document a long-term depression of trigeminal somatosensory processing in man. The different effects of LFS on the two components of the MIR may point to central mechanisms of LTD. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Brainstem; Orofacial; Pain; Plasticity; Trigeminal
Low-frequency stimulation (LFS) of presynaptic nerve fibers causes long-term depression (LTD) of synaptic strength in the hippocampus in vitro [8,16]. LTD is an established experimental model of synaptic plasticity and may also play a role in the nociceptive system [1,18]. In rat spinal cord slice with a long dorsal root attached substantia gelatinosa neurons were intracellularly recorded [19]. LFS with stimulus intensities that effectively activated the vast majority of nociceptive Ad-fibers produced a robust LTD in all neurons tested. A similar experiment was conducted in adult rats under general anesthesia [14]. Field potentials evoked by electric stimulation of the sciatic nerve were recorded from the lumbar spinal dorsal horn. LFS at Ad-fiber intensity was tested in 11 rats. In only two rats LTD was induced. Whereas LFS reliably induces LTD in vitro, it only exceptionally induces LTD in vivo. Many experiments suggest that LTD of basal transmission is not readily induced either in awake or anesthetized rats [6,7,11,15]. Thus, the significance of LTD for memory processes has been questioned due to reported failures to induce persistent LTD in vivo. Consequently, LTD can hardly be applied to human beings. In order to fill the gap between the in vitro cell studies and the in vivo situation we investigated the * Corresponding author. Tel.: 149-9131-852-2003; fax: 1499131-852-2774. E-mail address: [email protected] (J. Ellrich).
effects of LFS of trigeminal afferents on somatosensory processing in healthy volunteers by applying the masseter inhibitory reflex. The experiments were conducted in five female and four male healthy volunteers between 22 and 35 years of age, who gave their informed consent prior to their inclusion in the study according to the 1964 Declaration of Helsinki (revised in Edinburgh, 2000). Trigeminal somatosensory processing was measured by the masseter inhibitory reflex (MIR) that consists of an early MIR1 and a late MIR2. The MIR was evoked by electric stimulation of the mental nerve area in close vicinity to the lower lip. Rectangular pulses (500 ms duration) were applied by a small cathode (2 mm diameter) and a large anode (9 mm diameter). The subjects were comfortably sitting on a chair with the eyes open. In three and six volunteers the right and left mental nerve area, respectively, was stimulated. Electromyographic activity (EMG) was recorded from both masseter muscles by surface electrodes with one electrode over the inferior and anterior region of the muscle and another electrode at the temple (bandpass 100–1000 Hz). A ground electrode was fixed at the right forearm. EMG signals were amplified and digitized by a micro CED1401 A/D-Converter (CED, Cambridge, UK) with a 10 000 Hz sampling rate and analyzed by the Signal Software (CED, Cambridge, UK). EMG sweeps were
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 66 9- 9
266
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
recorded from 100 ms before to 200 ms after the electric stimulus. Before applying an electric stimulus the subject was asked to adjust the EMG activity of both masseter muscles to at least 90% of maximum strength by clenching the teeth, controlled by visual and acoustic feedback, because the duration of the suppression periods was inversely related to the magnitude of clenching [12]. The MIR was elicited 1–2 s after having reached this level. The volunteers were requested not to relax masseter muscles when recognizing the stimulus but to maintain the activity for another second. The contractions were serially repeated after 30 s of rest. Individual thresholds for detection (I0) and pain (IP) were determined by applying two series of electric pulses with inand decreasing stimulus intensity according to the method of limits. After each electric stimulus the volunteer was asked to rate the stimulus according to an ordinal scale (0 ¼ no perception, 1–39 ¼ perception but no pricking painful sensation, 40 ¼ slightly pricking painful (pain threshold), 100 ¼ unbearable painful). Based on the threshold detection the test stimulus for the reflex study was adjusted to an intensity clearly above the individual pain threshold and sufficient to reliably evoke the MIR. The MIR was elicited in blocks of 8 test stimuli each with an interstimulus interval of about 30 s. Test stimulation blocks were repeated every 8 min. After the third stimulus block LFS was applied for the next 20 min. LFS consisted of 1200 electric rectangular pulses (500 ms duration) applied with the same painful stimulus intensity as for the reflexes but with a frequency of 1 Hz. A total of 8 min after the end of LFS the test stimulation blocks were continued for about 1 h. Thus, three test stimulus blocks were applied before LFS (time 220, 212 and 24 min) and eight test stimulus blocks were applied after LFS (time 28, 36, 44, 52, 60, 68, 76 and 84 min). The eight trials of each test stimulus block were rectified and averaged off-line. The mean amplitude of the time window from 2100 to 210 ms before the electric stimulus was set as 100% (baseline EMG activity). A MIR was defined as more than 50% depression of baseline EMG activity [3]. Onset latencies and durations of MIR1 and MIR2 were measured at the points where rectified EMG activity crossed a level of 80% of the prestimulus EMG. Integrals of MIR1 and MIR2 were calculated in an individual fixed time window that lasted from the earliest onset latency to the latest end latency of all individual test stimulus block averages. Mean onset latencies, durations, integrals and pain ratings recorded in the three test stimulus blocks before the LFS were set as 100%. Data were statistically analyzed by analysis of variance (one-way analysis of variance, ANOVA, repeated measures design) and by the least significant difference (LSD) post hoc test. Arithmetic mean (mean) and standard error (SE) were calculated. The electric thresholds for detection (I0) and pain (IP) were 0.9 ^ 0.2 and 4.7 ^ 0.8 mA (mean ^ SE). The mean test stimulus intensity of 8.9 ^ 0.3 mA evoked a pinprick
like pain sensation that was rated at 60.9 ^ 4.6 (mean ^ SE). The test stimulus intensity expressed as multiples of the individual IP and I0 was (2.3 ^ 0.3) £ IP and (11.9 ^ 1.6) £ I0 (mean ^ SE). The test stimulus elicited an MIR1 in seven out of nine volunteers and an MIR2 in all nine volunteers. After LFS the MIR significantly changed, whereas MIR1 and MIR2 were differentially affected (Fig. 1). Electric stimulation evoked in the ipsilateral masseter muscle an MIR1 with an onset latency of 12.4 ^ 1.0 ms and a duration of 17.4 ^ 1.7 ms. After LFS, onset latency (12.0 ^ 0.6 ms; F ¼ 0:63, P ¼ 0:78) and duration (16.1 ^ 1.5 ms; F ¼ 0:78, P ¼ 0:65) remained statistically unchanged as calculated by the one-way ANOVA (repeated measures design). The onset latency of MIR2 significantly increased from 45.3 ^ 1.3 to 53.9 ^ 3.6 ms after LFS (F ¼ 2:18, P , 0:05). This change was mainly due to an increase of onset latencies in the test stimulus blocks at time points 60 and 76 min as shown by the LSD test (P , 0:01). After LFS, the MIR2 duration significantly decreased from
Fig. 1. Masseter inhibitory reflex responses of one specimen before and after LFS. EMG activity (arbitrary units) of the ipsilateral masseter muscle was evoked by noxious electric stimulation of right mental nerve afferents. The upper graph shows one test stimulus block with eight sweeps just before LFS (4 min before LFS). The lower graph also demonstrates a test stimulus block of the same volunteer, but just after conditioning LFS (8 min after LFS).
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
32.6 ^ 2.2 to 20.9 ^ 3.9 ms (F ¼ 5:97, P , 0:00001). The LSD test stressed that this effect is mainly caused by a decrease in seven out eight test stimulus blocks after LFS (P , 0:01). MIR1 and MIR2 integrals significantly increased after LFS to 136.0 ^ 6.7% (F ¼ 4:64, P , 0:0001) and 171.6 ^ 5.4% (F ¼ 8:54, P , 0:000001). Whereas the increase of MIR1 integrals was mainly due to a significant change in the last two test stimulus blocks after LFS (P , 0:05), the LSD test documented for the MIR2 that all integrals after LFS were significantly larger than those in the stimulus blocks before LFS (P , 0:05). The baseline EMG activity remained statistically unchanged after LFS (F ¼ 1:37, P ¼ 0:21) (Fig. 2). The MIR1 of the contralateral masseter muscle had an onset latency of 11.7 ^ 1.2 ms and a duration of 17.8 ^ 1.5 ms. After LFS, onset latency (11.9 ^ 0.8 ms; F ¼ 0:75, P ¼ 0:68) and duration (15.9 ^ 1.9 ms; F ¼ 0:70, P ¼ 0:72) remained statistically unchanged as calculated by the one-way ANOVA (repeated measures design). The onset latency of MIR2 significantly increased from 46.2 ^ 1.0 to 59.9 ^ 4.9 ms after LFS (F ¼ 2:30, P , 0:05). The LSD test explained this significant effect by an increase of onset latencies in five out eight test stimu-
Fig. 2. Reflex integrals and baseline EMG activities (top), onset latencies and durations (bottom) of ipsilateral MIR1 and MIR2 components of all volunteers (n ¼ 9). Mean integrals, onset latencies and durations recorded in the three test stimulus blocks before the LFS were set as 100% (dotted line). Data are presented as arithmetic mean ^ SE.
267
lus blocks after LFS (P , 0:05). After LFS, the MIR2 duration significantly decreased from 29.6 ^ 3.0 to 17.1 ^ 4.1 ms (F ¼ 6:85, P , 0:000001). This strong reduction of MIR2 duration was mainly due to a significant decrease in all post LFS stimulus blocks in comparison to all pre LFS blocks (LSD test: P , 0:01). MIR1 and MIR2 integrals significantly increased after LFS to 131.1 ^ 2.8% (F ¼ 2:03, P , 0:05) and 164.5 ^ 3.3% (F ¼ 6:46, P , 0:000001). Whereas the increase of MIR1 integrals was mainly due to a significant change in the first and last test stimulus blocks after LFS (P , 0:05), the LSD test documented for the MIR2 that all integrals after LFS were significantly larger than those in the stimulus blocks before LFS (P , 0:01). The baseline EMG activity remained statistically unchanged after LFS (F ¼ 0:97, P ¼ 0:47). The ratings of stimulus strength significantly decreased after LFS to 90.7 ^ 3.3% compared to the pre LFS stimulations (F ¼ 7:56, P , 0:000001). The LSD test stressed that all ratings after LFS were significantly smaller than those in the stimulus blocks before LFS (P , 0:01). Noxious low-frequency stimulation of mental nerve afferents caused a decrease of the MIR and a decrease of pain ratings. The depressive effect of LFS was stronger on the MIR2 than on the MIR1. This depression of the reflex and the pain sensation lasted for at least 1 h and therefore meets the criteria of long-term depression [1]. The MIR is a trigemino-trigeminal brainstem reflex that is connected with orofacial nociception [9]. Both components, the early MIR1 and the late MIR2, can be elicited by electric stimulation, by mechanical stimulation and by selective activation of nociceptors using laser radiant heat [2,10]. Based on the detection of electric thresholds and on modulation studies both components of the MIR are probably mediated by wide-dynamic-range neurons (WDR) that get input from low-threshold mechanoreceptive and nociceptive afferents [4,10,13]. The MIR1 is probably mediated by pontine WDR neurons within the rostral part of the spinal trigeminal nucleus (STN) or the interstitial nucleus (ISVT), and WDR neurons mediating the medullary MIR2 are probably located in the caudal part of the STN or the medullary ISVT. In a rat transverse spinal cord slice LFS of dorsal root afferents clearly depressed substantia gelatinosa neurons EPSPs amplitudes and initial slopes [19]. The depression effect of LFS on the MIR2 is similar in the present study. The MIR2 integral increased by about 68% of control, corresponding to an inhibition of MIR2, the onset latencies increased by about 25%. The effects are compatible with a modulation of EPSP amplitude and slope in nociceptive sensory neurons. In vivo rat experiments showed a maximum suppression of neuronal activity in the hippocampal CA1 region to about 49% of control after LFS [1]. Many in vivo animal experimental studies did not find any effect of LFS on different populations of central neurons [6,7,11,15]. In contrast, LFS was effective in all investigated volunteers in the present study.
268
J. Ellrich, A. Schorr / Neuroscience Letters 329 (2002) 265–268
Whereas the MIR2 was clearly depressed after LFS, the effects on the MIR1 were less pronounced. Onset latency and duration did not change, and the increase of the MIR1 integral was mainly due to an increase in the last test stimulus blocks. These different effects on MIR2 and MIR1 were also true, when analyzing only those experiments where MIR1 and MIR2 were simultaneously evoked. Thus, LFS was differentially effective on MIR1 and MIR2. In animal experimental studies LTD phenomena were investigated at the first synapse between the nociceptive afferent fiber and the secondary sensory neuron in the substantia gelatinosa at the level of the spinal dorsal horn [14,19]. Based on the assumption that LFS exclusively depresses the strength of the first nociceptive synapse, the LTD effect on MIR1 and MIR2 should be very similar. The different effects on MIR1 and MIR2 might be due to a different contribution of nonnociceptive afferents to the reflex arc. In a previous study it was shown, that the electric threshold of the MIR1 is significantly higher than that of the MIR2 [10]. These data contradicts a stronger involvement of non-nociceptive afferents in the MIR1 reflex arc. Furthermore, habituation of MIR1 and MIR2 also might cause an inhibition. The MIR1 does not habituate even with stimulation frequencies of about 3 Hz, the MIR2 slightly habituates with interstimulus intervals of 5 s and never habituates with more than 20 s interstimulus intervals [5]. Alternatively, the different modulations of MIR1 and MIR2 may be due to LFS effects beyond the first nociceptive synapse. Noxious LFS may cause a decrease of presynaptic inhibition, that is involved in generating the MIR2 [17]. The different effectiveness of LFS on nociceptive processing in the spinal cord slice in vitro and in the anesthetized rat in vivo may arouse suspicion that central mechanisms may be involved in LTD [14]. In summary, this study documents a long-term depression of trigeminal somatosensory processing in man. The different effects of LFS on the two components of the MIR may point to central mechanisms of LTD. This study was supported by DFG grant SFB353-B17. [1] Braunewell, K.H. and Manahan-Vaughan, D., Long-term depression: a cellular basis for learning? Rev. Neurosci., 12 (2001) 121–140. [2] Cruccu, G., Agostino, R., Inghilleri, M., Manfredi, M. and Ongerboer de Visser, B.W., The masseter inhibitory reflex is evoked by innocuous stimuli and mediated by A beta afferent fibres, Exp. Brain Res., 77 (1989) 447–450. [3] Cruccu, G., Agostino, R., Lahuerta, L. and Manfredi, M., Inhibition of jaw-closing muscles by electrical stimulation of the ophthalmic division in man, Brain Res., 371 (1986) 298–304.
[4] Cruccu, G. and Romaniello, A., Jaw-opening reflex after CO2 laser stimulation of the perioral region in man, Exp. Brain Res., 118 (1998) 564–568. [5] Desmedt, J.E. and Godaux, E., Habituation of exteroceptive suppression and of exteroceptive reflexes in man as influenced by voluntary contraction, Brain Res., 106 (1976) 21– 29. [6] Doyere, V., Errington, M.L., Laroche, S. and Bliss, T.V., Lowfrequency trains of paired stimuli induce long-term depression in area CA1 but not in dentate gyrus of the intact rat, Hippocampus, 6 (1996) 52–57. [7] Doyle, C.A., Cullen, W.K., Rowan, M.J. and Anwyl, R., Lowfrequency stimulation induces homosynaptic depotentiation but not long-term depression of synaptic transmission in the adult anaesthetized and awake rat hippocampus in vivo, Neuroscience, 77 (1997) 75–85. [8] Dudek, S.M. and Bear, M.F., Homosynaptic long-term depression in area CA1 of hippocampus and effects of Nmethyl-d-aspartate receptor blockade, Proc. Natl. Acad. Sci. USA, 89 (1992) 4363–4367. [9] Ellrich, J., Brainstem reflexes: probing human trigeminal nociception, News Physiol. Sci., 15 (2000) 94–97. [10] Ellrich, J., Hopf, H.C. and Treede, R.D., Nociceptive masseter inhibitory reflexes evoked by laser radiant heat and electrical stimuli, Brain Res., 764 (1997) 214–220. [11] Errington, M.L., Bliss, T.V., Richter-Levin, G., Yenk, K., Doyere, V. and Laroche, S., Stimulation at 1–5 Hz does not produce long-term depression or depotentiation in the hippocampus of the adult rat in vivo, J. Neurophysiol., 74 (1995) 1793–1799. [12] Fung, D.T., Hwang, J.C. and Poon, W.F., Effect of bite force on the masseteric electromyographic silent period in man, Arch. Oral Biol., 27 (1982) 577–580. [13] Hansen, P.O., Svensson, P., Arendt-Nielsen, L. and Jensen, T.S., Human masseter inhibitory reflexes evoked by repetitive electrical stimulation, Clin. Neurophysiol., 113 (2002) 236–242. [14] Liu, X.G., Morton, C.R., Azkue, J.J., Zimmermann, M. and Sandku¨ hler, J., Long-term depression of C-fibre-evoked spinal field potentials by stimulation of primary afferent A delta-fibres in the adult rat, Eur. J. Neurosci., 10 (1998) 3069–3075. [15] Manahan-Vaughan, D., Long-term depression in freely moving rats is dependent upon strain variation, induction protocol and behavioral state, Cereb. Cortex, 10 (2000) 482– 487. [16] Mulkey, R.M. and Malenka, R.C., Mechanisms underlying induction of homosynaptic long-term depression in area CA1 of the hippocampus, Neuron, 9 (1992) 967–975. [17] Rudomin, P. and Schmidt, R.F., Presynaptic inhibition in the vertebrate spinal cord revisited, Exp. Brain Res., 129 (1999) 1–37. [18] Sandku¨ hler, J., Learning and memory in pain pathways, Pain, 88 (2000) 113–118. [19] Sandku¨ hler, J., Chen, J.G., Cheng, G. and Randic, M., Lowfrequency stimulation of afferent Adelta-fibers induces long-term depression at primary afferent synapses with substantia gelatinosa neurons in the rat, J. Neurosci., 17 (1997) 6483–6491.