- Email: [email protected]
Slow repetitive transcranial magnetic stimulation increases somatosensory high-frequency oscillations in humans
Neuroscience Letters 358 (2004) 193–196 www.elsevier.com/locate/neulet
Slow repetitive transcranial magnetic stimulation increases somatosensory high-frequency oscillations in humans Asao Ogawaa,*, Satoshi Ukaia, Kazuhiro Shinosakib, Masakiyo Yamamotoa, Shunsuke Kawaguchia, Ryouhei Ishiia, Masatoshi Takedaa a
Department of Psychiatry and Behavioral Science, Osaka University Graduate School of Medicine, Suita, Osaka 565-0871, Japan b Department of Neuropsychiatry, Wakayama Medical University, Wakayama 641-8509, Japan Received 4 December 2003; received in revised form 16 January 2004; accepted 20 January 2004
Abstract Repetitive transcranial magnetic stimulation (rTMS) has been proposed as a possible treatment for psychiatric and neurological disorders characterized by focal brain excitability, such as major depression and action myoclonus. However, the mechanism of modulating excitability by rTMS is unclear. We examined the changes in high frequency oscillations (HFOs) of somatosensory evoked potentials (SEPs) before and after slow rTMS over the right primary somatosensory cortex (0.5 Hz, 50 pulses, 80% motor threshold intensity). The HFOs, which represent a localized activity of intracortical inhibitory interneurons, were significantly increased after slow rTMS, while the SEPs were not changed. Our results suggest that slow rTMS affects cortical excitability by modulating the activity of the intracortical inhibitory interneurons beyond the time of the stimulation and that rTMS may have therapeutic effects on such disorders. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Repetitive transcranial magnetic stimulation; High-frequency oscillations; Somatosensory evoked potential; Cortical excitability; Inhibitory interneurons; Somatosensory cortex
Repetitive transcranial magnetic stimulation (rTMS) has recently gained considerable attention as a possible treatment for psychiatric and neurological disorders, such as major depression, schizophrenia, action myoclonus, Parkinson’s disease, and epilepsy. Neuroimaging studies have revealed that these disorders are characterized by focal brain excitability. Several reports have shown that rTMS can change cortical excitability beyond the time of the stimulation, and this modulation of cortical excitability might have a therapeutic potential [15]. Slow rTMS uses low frequency (# 1 Hz) stimuli and is assumed to decrease the excitability of the stimulated cortical area, whereas fast rTMS uses high frequency (. 1 Hz) and is assumed to increase the excitability [15]. While these modulations of cortical excitability have been referred to long-term depression and long-term potentiation in animal studies [16], their possible relations are unclear. Slow rTMS delivered to the motor cortex decreased motor evoked potentials (MEPs) [2] and cortical glucose metab* Corresponding author. Tel.: þ 81-6-6879-3051; fax: þ81-6-6879-3059. E-mail address: [email protected] (A. Ogawa).
olism [10]. Slow rTMS delivered to the somatosensory cortex led to an increased tactile threshold [19] and impaired the tactile perception [12]. Slow rTMS applied to the visual cortex led to an increase phosphene threshold [1]. The results of these studies demonstrated that slow rTMS decreases the cortical excitability of the stimulated area. However, those studies did not elucidate the physiological mechanisms responsible for such modulations. Recently, various reports have demonstrated that oscillatory bursts in the range of 300 –900 Hz are superimposed on the primary response (N20) of the human somatosensory cortex following median nerve stimulation [3,7]. Studies using magnetoencephalography revealed that these oscillatory bursts were localized near the N20 source in area 3b of the somatosensory cortex [3,7]. Hashimoto et al. called these oscillatory bursts ‘high-frequency oscillations (HFOs)’, and hypothesized that these bursts represent a localized activity of the GABAergic inhibitory interneurons [7], whereas N20 is generated by excitatory postsynaptic potentials of pyramidal cells. Subsequently, the results of several studies have reinforced this hypothesis [6,20]. The studies described above suggest that further
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.01.038
194
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
investigations into the rTMS effects on HFOs may elucidate the mechanisms employed by rTMS in modulating cortical excitability, especially that observed through the intracortical inhibitory interneurons. In addition, further studies may throw light on the therapeutic mechanisms of rTMS. In this paper, we examined the changes in the HFOs following the application of slow rTMS over the somatosensory cortex and investigated the physiological mechanisms of slow rTMS in altering cortical excitability. Seven healthy volunteers were studied (average age 27.4 years, range 26 –30 years, right handed). All subjects provided written informed consent. Somatosensory evoked potentials (SEPs) were recorded for approximately 30 min before and immediately after slow rTMS using a Nicolet Viking IV (Nicolet Biomedical, Inc, Madison, WI). Subjects lay supine on a bed and were instructed to stay awake with their eyes closed and pay no attention to the stimuli. Electrical stimuli of 0.2 ms duration were delivered to the left median nerve at the wrist with a repetition rate of 3.3 Hz. Intensity was adjusted to 1.2 times the motor threshold so as to induce a small muscular twitch in the thenar muscles. The band-pass filter of the amplifier was set at 0.5–2000 Hz. SEPs were digitized at a 10 kHz sampling rate. Recording electrodes were placed on Fpz and C3 of the international 10–20 systems, and C30 (3 cm posterior to C3) and C300 (3 cm anterior to C3) [14]. The left ear served as a reference. For each run, 300 trials were averaged. In order to maintain the vigilance level as high as possible, the subjects were interviewed about the vigilance level before and after every run, and runs without sufficient vigilance level were rejected. Runs contaminated with artifacts were also rejected. Since averaged signals in each run did not provide sufficient signal-to-noise ratio to estimate HFOs, ten runs were averaged again for obtaining SEPs, corresponding to averaging 3000 trials in approximately 30 min before and after slow rTMS. In order to separate the HFOs from the underlying N20, the recorded signals were digitally filtered through a band-pass of 300–1000 Hz. TMS was generated using a Magstim Super Rapid (Magstim, Whitland, Wales). Stimuli were delivered while the subjects were seated in a chair via a figure-of-eight coil (diameter 70 mm). On the day before the experiment, the optimal site and the motor threshold (MT) for stimulation of the abductor digiti minimi was established. The coil was placed on the scalp at the estimated position of the right motor cortex (7 cm lateral to the vertex). The coil was moved until the position that produced the largest MEPs in the left abductor digiti minimi at supra-threshold intensity was located. MT was determined according to the previous report [18]. This position was marked and used throughout the experiment. For the rTMS, from the marked position, the coil was moved 1 cm posteriorly and 1 cm laterally in order to place it above the somatosensory cortex [8], and then slow rTMS was delivered. A train of rTMS consisted of 50 stimuli at a
frequency of 0.5 Hz. The intensity was set at 80% of the MT of each subject. The band-pass filtered signals from C30 (over the somatosensory area) were used for the data analysis. The oscillations between the onset of N20 and the peak of P27 with an amplitude of twice or more than that of the background noise level were used for the further analysis [13]. The number of negative peaks and the peak-to-peak amplitude of every HFO were measured. Then, the average amplitude and the maximum amplitude and the root mean square (RMS) amplitude of the HFOs, were calculated. The RMS is a statistical measure of the magnitude of a varying quantity and is the square root of the mean of the squares of the values. For analysis of the SEPs, the latency and the amplitude of N20 (bottom-to-peak amplitude) were measured. The statistical differences of these parameters were compared between before and after rTMS using the paired t test. The level of probability selected as significant was a value of P , 0:05: The sham stimulation was performed on a different day from the real stimulation. For the sham stimulation, the coil was placed perpendicular to the scalp, while the coil was tangential for the real stimulation. The other parameters of the rTMS and the analysis remained identical the real stimulation. Before rTMS, the onset latency, the peak latency and the amplitude of N20 were 18.8 ^ 1.20 ms and 3.30 ^ 1.23 mV (mean ^ SD), respectively. These three parameters showed no significant differences before and after rTMS (after rTMS 18.90 ^ 1.27 ms, P ¼ 0:12; 3.03 ^ 1.08 mV, P ¼ 0:39) (Table 1, Figs. 1a and 2). HFOs were obtained from the C30 electrode before and after rTMS in all subjects. The HFOs showed three or four negative peaks before rTMS, and four negative peaks after rTMS in all subjects. Before rTMS, the average amplitude, the maximum amplitude and the RMS amplitude of the HFOs were 0.16 ^ 0.08, 0.29 ^ 0.10 and 0.079 ^ 0.03 mV, respectively. These three parameters were significantly increased after rTMS (after rTMS 0.23 ^ 0.09 mV, P , 0:001; 0.35 ^ 0.09 mV, P ¼ 0:02; 0.095 ^ 0.023 mV, P ¼ 0:04) (Table 1, Figs. 1b and 2). For SEPs, the onset latency, the peak latency and the amplitude of N20 showed no significant change after the sham stimulation. For HFOs, the number of the negative peaks, the average amplitude, the maximum amplitude and the RMS amplitude of the HFOs also showed no significant change after the sham stimulation (Table 1, Fig. 2). The main result of this study indicates that the application of slow rTMS (0.5 Hz, 50 stimuli, 80% of MT intensity) over the primary somatosensory cortex significantly increased the amplitude and the RMS amplitude of the HFOs, but did not alter the amplitude and the latency of the SEPs (N20) (averaged approximately 30 min before and after slow rTMS). To our knowledge, this is the first report that human HFOs are modulated by the slow rTMS beyond the time of the stimulation.
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
195
Table 1 Parameters of HFOs and N20 before and after the real/sham stimulations (mean ^ SD) (n ¼ 7) Real stimulation
Sham stimulation
Before rTMS
After rTMS
P
N20 Peak latency (ms) Peak amplitude (mV)
18.80 ^ 1.20 3.30 ^ 1.23
18.90 ^ 1.27 3.03 ^ 1.08
0.12 0.39
HFOs Amplitude (mV) average Amplitude (mV) max RMS (mV) Number of negative peaks
0.16 ^ 0.08 0.29 ^ 0.10 0.079 ^ 0.03 3.85 ^ 0.38
0.23 ^ 0.09 0.35 ^ 0.09 0.095 ^ 0.023 4.00 ^ 0.00
,0.001 0.02 0.04 0.32
* * *
Before sham
After sham
P
18.90 ^ 1.20 2.99 ^ 1.00
18.90 ^ 0.30 3.06 ^ 1.00
0.18 0.24
0.19 ^ 0.09 0.29 ^ 0.09 0.079 ^ 0.03 3.86 ^ 0.69
0.19 ^ 0.11 0.29 ^ 0.09 0.079 ^ 0.024 3.86 ^ 0.69
0.91 0.29 1 1
The real stimulation significantly increased the average, the maximum and the root mean square (RMS) amplitudes of the HFOs. *P , 0.05 by a paired ttest.
Since the intracortical GABAergic inhibitory interneurons are most likely responsible for generation of the HFOs, the present results suggest that slow rTMS decreases the excitability of the stimulated cortical area through modulating the activity of the intracortical inhibitory interneurons. Hashimoto et al. have hypothesized that HFOs represent a localized activity of GABAergic inhibitory interneurons in layer 4 of area 3b of the somatosensory cortex [7], although some other possible neural origins remain to be considered [3,5]. This hypothesis has been reinforced by other human studies [6,20] and is in line with Jones’ finding that intracellular activity in the fast spiking cells, representing the GABAergic interneurons, showed a
close association with very fast epipial field potentials in the rat barrel cortex [9]. Previous studies examined the effects of slow rTMS on the human primary motor cortex [3,17], somatosensory cortex [12,19] and visual cortex [1] using the changes of their outputs or task performances as indices of cortical excitability. They speculated that the slow rTMS might lead to a decrease in cortical excitability through enhancements of intracortical interneurons. It is notable that the present study has confirmed this speculation clearly and directly using HFOs that reflect the activity of the intracortical inhibitory interneurons. The effect of the slow rTMS on the HFOs persisted after rTMS stimulation. The duration of the enhanced HFOs in this study (approximately 30 min after the 100 s slow rTMS) was remarkably longer than that of previous slow rTMS studies, with the durations lasting less than twice the
Fig. 1. SEPs (N20) (a); and HFOs (b) following stimulation of the left median nerve before and after the real rTMS in a representative subject. The HFOs are significantly increased after rTMS (small arrows), whereas N20 showed no significant changes.
Fig. 2. Changes of the average amplitudes and the root mean square (RMS) amplitudes of the HFOs before and after the real/sham rTMSs. The average and the RMS amplitudes of the HFOs were enhanced by the real rTMS, but showed no significant changes by the sham rTMS.
196
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
durations of the stimulations as a whole. The impairment of a tactile frequency discrimination task lasted 2 min after a 5 min slow rTMS [12]. The decrease of MEPs lasted 10 min after a 10 min 1 Hz rTMS [17]. For the investigation of the therapeutic potential of rTMS in relation to cortical excitability, it is essential to understand how rTMS can make sustained modifications of cortical excitability for days or weeks. Since the HFOs may be an index of longlasting modifications of cortical excitability, it is necessary to examine its long-term changes after rTMS in future studies. SEPs (N20) and HFOs represent parallel and partly independent steps in sensory processing. SEPs also represent a stable somatosensory input while HFOs index variable modes of processing, such as a floating focus of attention [11]. In this study, the amplitudes of the HFOs increased, but the SEPs did not significantly change as well as in the previous study [4]. This result suggests that the slow rTMS might affect the fluctuating information processing and that HFOs may be a promising indices in future investigations of rTMS effects on human cognitive function and cognitive improvement of psychiatric and neurological patients. Since the motor cortex has an effect on somatosensory functions through their interconnections and since the rTMS in this study did not focus on merely the primary somatosensory cortex, it is not clear whether the rTMS acted through the stimulation of the somatosensory cortex alone. However, there was no significant change of the SEPs in the present study, while previous studies demonstrated that rTMS applied to the motor cortex decreased the amplitudes of SEPs [4]. Although the contribution of the motor cortex cannot be excluded completely, it is more likely that the changes of the HFOs were induced by direct effects on the somatosensory cortex.
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Acknowledgements This research was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (15591220).
[17]
[18]
References [1] B. Boroojerdi, A. Prager, W. Muellbacher, L.G. Cohen, Reduction of human visual cortex excitability using 1-Hz transcranial magnetic stimulation, Neurology 54 (2000) 1529–1531. [2] R. Chen, J. Classen, C. Gerloff, Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation, Neurology 48 (1997) 1398–1403. [3] G. Curio, High frequency (600 Hz) bursts of spike-like activities generated in the human cerebral somatosensory system, Electroenceph. clin. Neurophysiol. 49 (1999) 56–61. [4] H. Enomoto, Y. Ugawa, R. Hanajima, K. Yuasa, H. Mochizuki, Y. Terao, Y. Shiio, T. Furubayashi, N.K. Iwata, I. Kanazawa, Decreased
[19]
[20]
sensory cortical excitability after 1 Hz rTMS over the ipsilateral primary motor cortex, Clin. Neurophysiol. 112 (2001) 2154–2158. R. Gobbele, H. Buchner, G. Curio, High-frequency (600 Hz) SEP activities originating in the subcortical and cortical human somatosensory system, Electroenceph. clin. Neurophysiol. 108 (1998) 182 –189. I. Hashimoto, T. Kimura, T. Fukushima, Y. Iguchi, Y. Saito, O. Terasaki, K. Sakuma, Reciprocal modulation of somatosensory evoked N20m primary response and high-frequency oscillations by interference stimulation, Clin. Neurophysiol. 110 (1999) 1445– 1451. I. Hashimoto, T. Mashiko, T. Imada, Somatic evoked high-frequency magnetic oscillations reflect activity of inhibitory interneurons in the human somatosensory cortex, Electroenceph. clin. Neurophysiol. 100 (1996) 189 –203. P. Hlustik, A. Solodkin, R.P. Gullapalli, D.C. Noll, S.L. Small, Somatotopy in human primary motor and somatosensory hand representations revisited, Cerebral Cortex 11 (2001) 312–321. M.S. Jones, K.D. MacConald, B. Choi, Intracellular correlates of fast (. 200 Hz) electrical oscillations in rat somatosensory cortex, J. Neurophysiol. 841 (2000) 1505–1518. T.A. Kimbrell, J.T. Little, R.T. Dunn, M.A. Frye, B.D. Greenberg, E.M. Wassermann, J.D. Repella, A.L. Danielson, M.W. Willis, B.E. Benson, A.M. Speer, E. Osuch, M.S. George, R.M. Post, Frequency dependence of antidepressant response to left prefrontal repetitive transcranial magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism, Biol. Psychiatry 46 (1999) 1603– 1613. F. Klostermann, T. Funk, J. Vesper, R. Siedenberg, G. Curio, Propofol narcosis dissociates human intrathalamic and cortical high-frequency (.400 Hz) SEP components, NeuroReport 11 (2000) 2607–2610. S. Knecht, T. Ellger, C. Breitenstein, E. Bernd Ringelstein, H. Henningsen, Changing cortical excitability with low-frequency transcranial magnetic stimulation can induce sustained disruption of tactile perception, Biol. Psychiatry 53 (2003) 175–179. S. Nakano, I. Hashimoto, The later part of high-frequency oscillations in human somatosensory evoked potentials is enhanced in aged subjects, Neurosci. Lett. 276 (1999) 83 –86. I. Ozaki, C. Suzuki, Y. Yaegashi, M. Baba, M. Matsunaga, I. Hashimoto, High frequency oscillations in early cortical somatosensory evoked potentials, Electroenceph. clin. Neurophysiol. 108 (1998) 536 –542. A. Pascual-Leone, N.J. Davey, J. Rothwell, E.M. Wassermann, Handbook of Transcranial Magnetic Stimulation, Arnold, London, 2002. R.M. Post, T.A. Kimbrell, M. Frye, M.S. George, U. McCann, J. Little, Implications of kindling and quenching for the possible frequency dependence of rTMS, CNS Spectrums 2 (1997) 54–60. J.R. Romero, D. Anschel, R. Sparing, M. Gangitano, A. PascualLeone, Subthreshold low frequency repetitive transcranial magnetic stimulation selectively decreases facilitation in the motor cortex, Clin. Neurophysiol. 113 (2002) 101–107. P.M. Rossini, A.T. Barker, A. Berardelli, M.D. Caramia, G. Caruso, R.Q. Cracco, M.R. Dimitrijevic, M. Hallett, Y. Katayama, C.H. Lucking, A.L. Maertens de Noordhout, C.D. Marsden, N.M.F. Murray, J.C. Rothwell, M. Swash, C. Tomberg, Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application. Report of an IFCN committee, Electroenceph. clin. Neurophysiol. 91 (1994) 79– 92. T. Satow, T. Mima, J. Yamamoto, T. Oga, T. Begum, T. Aso, N. Hashimoto, J.C. Rothwell, H. Shibasaki, Short-lasting impairment of tactile perception by 0.9 Hz – rTMS of the sensorimotor cortex, Neurology 60 (2003) 1045–1047. M. Tanosaki, I. Hashimoto, Y. Iguchi, T. Kimura, R. Takino, Y. Kurobe, Y. Haruta, Y. Hoshi, Specific somatosensory processing in somatosensory area 3b for human thumb: a neuromagnetic study, Clin. Neurophysiol. 112 (2001) 1516–1522.
Slow repetitive transcranial magnetic stimulation increases somatosensory high-frequency oscillations in humans Asao Ogawaa,*, Satoshi Ukaia, Kazuhiro Shinosakib, Masakiyo Yamamotoa, Shunsuke Kawaguchia, Ryouhei Ishiia, Masatoshi Takedaa a
Department of Psychiatry and Behavioral Science, Osaka University Graduate School of Medicine, Suita, Osaka 565-0871, Japan b Department of Neuropsychiatry, Wakayama Medical University, Wakayama 641-8509, Japan Received 4 December 2003; received in revised form 16 January 2004; accepted 20 January 2004
Abstract Repetitive transcranial magnetic stimulation (rTMS) has been proposed as a possible treatment for psychiatric and neurological disorders characterized by focal brain excitability, such as major depression and action myoclonus. However, the mechanism of modulating excitability by rTMS is unclear. We examined the changes in high frequency oscillations (HFOs) of somatosensory evoked potentials (SEPs) before and after slow rTMS over the right primary somatosensory cortex (0.5 Hz, 50 pulses, 80% motor threshold intensity). The HFOs, which represent a localized activity of intracortical inhibitory interneurons, were significantly increased after slow rTMS, while the SEPs were not changed. Our results suggest that slow rTMS affects cortical excitability by modulating the activity of the intracortical inhibitory interneurons beyond the time of the stimulation and that rTMS may have therapeutic effects on such disorders. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Repetitive transcranial magnetic stimulation; High-frequency oscillations; Somatosensory evoked potential; Cortical excitability; Inhibitory interneurons; Somatosensory cortex
Repetitive transcranial magnetic stimulation (rTMS) has recently gained considerable attention as a possible treatment for psychiatric and neurological disorders, such as major depression, schizophrenia, action myoclonus, Parkinson’s disease, and epilepsy. Neuroimaging studies have revealed that these disorders are characterized by focal brain excitability. Several reports have shown that rTMS can change cortical excitability beyond the time of the stimulation, and this modulation of cortical excitability might have a therapeutic potential [15]. Slow rTMS uses low frequency (# 1 Hz) stimuli and is assumed to decrease the excitability of the stimulated cortical area, whereas fast rTMS uses high frequency (. 1 Hz) and is assumed to increase the excitability [15]. While these modulations of cortical excitability have been referred to long-term depression and long-term potentiation in animal studies [16], their possible relations are unclear. Slow rTMS delivered to the motor cortex decreased motor evoked potentials (MEPs) [2] and cortical glucose metab* Corresponding author. Tel.: þ 81-6-6879-3051; fax: þ81-6-6879-3059. E-mail address: [email protected] (A. Ogawa).
olism [10]. Slow rTMS delivered to the somatosensory cortex led to an increased tactile threshold [19] and impaired the tactile perception [12]. Slow rTMS applied to the visual cortex led to an increase phosphene threshold [1]. The results of these studies demonstrated that slow rTMS decreases the cortical excitability of the stimulated area. However, those studies did not elucidate the physiological mechanisms responsible for such modulations. Recently, various reports have demonstrated that oscillatory bursts in the range of 300 –900 Hz are superimposed on the primary response (N20) of the human somatosensory cortex following median nerve stimulation [3,7]. Studies using magnetoencephalography revealed that these oscillatory bursts were localized near the N20 source in area 3b of the somatosensory cortex [3,7]. Hashimoto et al. called these oscillatory bursts ‘high-frequency oscillations (HFOs)’, and hypothesized that these bursts represent a localized activity of the GABAergic inhibitory interneurons [7], whereas N20 is generated by excitatory postsynaptic potentials of pyramidal cells. Subsequently, the results of several studies have reinforced this hypothesis [6,20]. The studies described above suggest that further
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.01.038
194
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
investigations into the rTMS effects on HFOs may elucidate the mechanisms employed by rTMS in modulating cortical excitability, especially that observed through the intracortical inhibitory interneurons. In addition, further studies may throw light on the therapeutic mechanisms of rTMS. In this paper, we examined the changes in the HFOs following the application of slow rTMS over the somatosensory cortex and investigated the physiological mechanisms of slow rTMS in altering cortical excitability. Seven healthy volunteers were studied (average age 27.4 years, range 26 –30 years, right handed). All subjects provided written informed consent. Somatosensory evoked potentials (SEPs) were recorded for approximately 30 min before and immediately after slow rTMS using a Nicolet Viking IV (Nicolet Biomedical, Inc, Madison, WI). Subjects lay supine on a bed and were instructed to stay awake with their eyes closed and pay no attention to the stimuli. Electrical stimuli of 0.2 ms duration were delivered to the left median nerve at the wrist with a repetition rate of 3.3 Hz. Intensity was adjusted to 1.2 times the motor threshold so as to induce a small muscular twitch in the thenar muscles. The band-pass filter of the amplifier was set at 0.5–2000 Hz. SEPs were digitized at a 10 kHz sampling rate. Recording electrodes were placed on Fpz and C3 of the international 10–20 systems, and C30 (3 cm posterior to C3) and C300 (3 cm anterior to C3) [14]. The left ear served as a reference. For each run, 300 trials were averaged. In order to maintain the vigilance level as high as possible, the subjects were interviewed about the vigilance level before and after every run, and runs without sufficient vigilance level were rejected. Runs contaminated with artifacts were also rejected. Since averaged signals in each run did not provide sufficient signal-to-noise ratio to estimate HFOs, ten runs were averaged again for obtaining SEPs, corresponding to averaging 3000 trials in approximately 30 min before and after slow rTMS. In order to separate the HFOs from the underlying N20, the recorded signals were digitally filtered through a band-pass of 300–1000 Hz. TMS was generated using a Magstim Super Rapid (Magstim, Whitland, Wales). Stimuli were delivered while the subjects were seated in a chair via a figure-of-eight coil (diameter 70 mm). On the day before the experiment, the optimal site and the motor threshold (MT) for stimulation of the abductor digiti minimi was established. The coil was placed on the scalp at the estimated position of the right motor cortex (7 cm lateral to the vertex). The coil was moved until the position that produced the largest MEPs in the left abductor digiti minimi at supra-threshold intensity was located. MT was determined according to the previous report [18]. This position was marked and used throughout the experiment. For the rTMS, from the marked position, the coil was moved 1 cm posteriorly and 1 cm laterally in order to place it above the somatosensory cortex [8], and then slow rTMS was delivered. A train of rTMS consisted of 50 stimuli at a
frequency of 0.5 Hz. The intensity was set at 80% of the MT of each subject. The band-pass filtered signals from C30 (over the somatosensory area) were used for the data analysis. The oscillations between the onset of N20 and the peak of P27 with an amplitude of twice or more than that of the background noise level were used for the further analysis [13]. The number of negative peaks and the peak-to-peak amplitude of every HFO were measured. Then, the average amplitude and the maximum amplitude and the root mean square (RMS) amplitude of the HFOs, were calculated. The RMS is a statistical measure of the magnitude of a varying quantity and is the square root of the mean of the squares of the values. For analysis of the SEPs, the latency and the amplitude of N20 (bottom-to-peak amplitude) were measured. The statistical differences of these parameters were compared between before and after rTMS using the paired t test. The level of probability selected as significant was a value of P , 0:05: The sham stimulation was performed on a different day from the real stimulation. For the sham stimulation, the coil was placed perpendicular to the scalp, while the coil was tangential for the real stimulation. The other parameters of the rTMS and the analysis remained identical the real stimulation. Before rTMS, the onset latency, the peak latency and the amplitude of N20 were 18.8 ^ 1.20 ms and 3.30 ^ 1.23 mV (mean ^ SD), respectively. These three parameters showed no significant differences before and after rTMS (after rTMS 18.90 ^ 1.27 ms, P ¼ 0:12; 3.03 ^ 1.08 mV, P ¼ 0:39) (Table 1, Figs. 1a and 2). HFOs were obtained from the C30 electrode before and after rTMS in all subjects. The HFOs showed three or four negative peaks before rTMS, and four negative peaks after rTMS in all subjects. Before rTMS, the average amplitude, the maximum amplitude and the RMS amplitude of the HFOs were 0.16 ^ 0.08, 0.29 ^ 0.10 and 0.079 ^ 0.03 mV, respectively. These three parameters were significantly increased after rTMS (after rTMS 0.23 ^ 0.09 mV, P , 0:001; 0.35 ^ 0.09 mV, P ¼ 0:02; 0.095 ^ 0.023 mV, P ¼ 0:04) (Table 1, Figs. 1b and 2). For SEPs, the onset latency, the peak latency and the amplitude of N20 showed no significant change after the sham stimulation. For HFOs, the number of the negative peaks, the average amplitude, the maximum amplitude and the RMS amplitude of the HFOs also showed no significant change after the sham stimulation (Table 1, Fig. 2). The main result of this study indicates that the application of slow rTMS (0.5 Hz, 50 stimuli, 80% of MT intensity) over the primary somatosensory cortex significantly increased the amplitude and the RMS amplitude of the HFOs, but did not alter the amplitude and the latency of the SEPs (N20) (averaged approximately 30 min before and after slow rTMS). To our knowledge, this is the first report that human HFOs are modulated by the slow rTMS beyond the time of the stimulation.
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
195
Table 1 Parameters of HFOs and N20 before and after the real/sham stimulations (mean ^ SD) (n ¼ 7) Real stimulation
Sham stimulation
Before rTMS
After rTMS
P
N20 Peak latency (ms) Peak amplitude (mV)
18.80 ^ 1.20 3.30 ^ 1.23
18.90 ^ 1.27 3.03 ^ 1.08
0.12 0.39
HFOs Amplitude (mV) average Amplitude (mV) max RMS (mV) Number of negative peaks
0.16 ^ 0.08 0.29 ^ 0.10 0.079 ^ 0.03 3.85 ^ 0.38
0.23 ^ 0.09 0.35 ^ 0.09 0.095 ^ 0.023 4.00 ^ 0.00
,0.001 0.02 0.04 0.32
* * *
Before sham
After sham
P
18.90 ^ 1.20 2.99 ^ 1.00
18.90 ^ 0.30 3.06 ^ 1.00
0.18 0.24
0.19 ^ 0.09 0.29 ^ 0.09 0.079 ^ 0.03 3.86 ^ 0.69
0.19 ^ 0.11 0.29 ^ 0.09 0.079 ^ 0.024 3.86 ^ 0.69
0.91 0.29 1 1
The real stimulation significantly increased the average, the maximum and the root mean square (RMS) amplitudes of the HFOs. *P , 0.05 by a paired ttest.
Since the intracortical GABAergic inhibitory interneurons are most likely responsible for generation of the HFOs, the present results suggest that slow rTMS decreases the excitability of the stimulated cortical area through modulating the activity of the intracortical inhibitory interneurons. Hashimoto et al. have hypothesized that HFOs represent a localized activity of GABAergic inhibitory interneurons in layer 4 of area 3b of the somatosensory cortex [7], although some other possible neural origins remain to be considered [3,5]. This hypothesis has been reinforced by other human studies [6,20] and is in line with Jones’ finding that intracellular activity in the fast spiking cells, representing the GABAergic interneurons, showed a
close association with very fast epipial field potentials in the rat barrel cortex [9]. Previous studies examined the effects of slow rTMS on the human primary motor cortex [3,17], somatosensory cortex [12,19] and visual cortex [1] using the changes of their outputs or task performances as indices of cortical excitability. They speculated that the slow rTMS might lead to a decrease in cortical excitability through enhancements of intracortical interneurons. It is notable that the present study has confirmed this speculation clearly and directly using HFOs that reflect the activity of the intracortical inhibitory interneurons. The effect of the slow rTMS on the HFOs persisted after rTMS stimulation. The duration of the enhanced HFOs in this study (approximately 30 min after the 100 s slow rTMS) was remarkably longer than that of previous slow rTMS studies, with the durations lasting less than twice the
Fig. 1. SEPs (N20) (a); and HFOs (b) following stimulation of the left median nerve before and after the real rTMS in a representative subject. The HFOs are significantly increased after rTMS (small arrows), whereas N20 showed no significant changes.
Fig. 2. Changes of the average amplitudes and the root mean square (RMS) amplitudes of the HFOs before and after the real/sham rTMSs. The average and the RMS amplitudes of the HFOs were enhanced by the real rTMS, but showed no significant changes by the sham rTMS.
196
A. Ogawa et al. / Neuroscience Letters 358 (2004) 193–196
durations of the stimulations as a whole. The impairment of a tactile frequency discrimination task lasted 2 min after a 5 min slow rTMS [12]. The decrease of MEPs lasted 10 min after a 10 min 1 Hz rTMS [17]. For the investigation of the therapeutic potential of rTMS in relation to cortical excitability, it is essential to understand how rTMS can make sustained modifications of cortical excitability for days or weeks. Since the HFOs may be an index of longlasting modifications of cortical excitability, it is necessary to examine its long-term changes after rTMS in future studies. SEPs (N20) and HFOs represent parallel and partly independent steps in sensory processing. SEPs also represent a stable somatosensory input while HFOs index variable modes of processing, such as a floating focus of attention [11]. In this study, the amplitudes of the HFOs increased, but the SEPs did not significantly change as well as in the previous study [4]. This result suggests that the slow rTMS might affect the fluctuating information processing and that HFOs may be a promising indices in future investigations of rTMS effects on human cognitive function and cognitive improvement of psychiatric and neurological patients. Since the motor cortex has an effect on somatosensory functions through their interconnections and since the rTMS in this study did not focus on merely the primary somatosensory cortex, it is not clear whether the rTMS acted through the stimulation of the somatosensory cortex alone. However, there was no significant change of the SEPs in the present study, while previous studies demonstrated that rTMS applied to the motor cortex decreased the amplitudes of SEPs [4]. Although the contribution of the motor cortex cannot be excluded completely, it is more likely that the changes of the HFOs were induced by direct effects on the somatosensory cortex.
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Acknowledgements This research was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (15591220).
[17]
[18]
References [1] B. Boroojerdi, A. Prager, W. Muellbacher, L.G. Cohen, Reduction of human visual cortex excitability using 1-Hz transcranial magnetic stimulation, Neurology 54 (2000) 1529–1531. [2] R. Chen, J. Classen, C. Gerloff, Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation, Neurology 48 (1997) 1398–1403. [3] G. Curio, High frequency (600 Hz) bursts of spike-like activities generated in the human cerebral somatosensory system, Electroenceph. clin. Neurophysiol. 49 (1999) 56–61. [4] H. Enomoto, Y. Ugawa, R. Hanajima, K. Yuasa, H. Mochizuki, Y. Terao, Y. Shiio, T. Furubayashi, N.K. Iwata, I. Kanazawa, Decreased
[19]
[20]
sensory cortical excitability after 1 Hz rTMS over the ipsilateral primary motor cortex, Clin. Neurophysiol. 112 (2001) 2154–2158. R. Gobbele, H. Buchner, G. Curio, High-frequency (600 Hz) SEP activities originating in the subcortical and cortical human somatosensory system, Electroenceph. clin. Neurophysiol. 108 (1998) 182 –189. I. Hashimoto, T. Kimura, T. Fukushima, Y. Iguchi, Y. Saito, O. Terasaki, K. Sakuma, Reciprocal modulation of somatosensory evoked N20m primary response and high-frequency oscillations by interference stimulation, Clin. Neurophysiol. 110 (1999) 1445– 1451. I. Hashimoto, T. Mashiko, T. Imada, Somatic evoked high-frequency magnetic oscillations reflect activity of inhibitory interneurons in the human somatosensory cortex, Electroenceph. clin. Neurophysiol. 100 (1996) 189 –203. P. Hlustik, A. Solodkin, R.P. Gullapalli, D.C. Noll, S.L. Small, Somatotopy in human primary motor and somatosensory hand representations revisited, Cerebral Cortex 11 (2001) 312–321. M.S. Jones, K.D. MacConald, B. Choi, Intracellular correlates of fast (. 200 Hz) electrical oscillations in rat somatosensory cortex, J. Neurophysiol. 841 (2000) 1505–1518. T.A. Kimbrell, J.T. Little, R.T. Dunn, M.A. Frye, B.D. Greenberg, E.M. Wassermann, J.D. Repella, A.L. Danielson, M.W. Willis, B.E. Benson, A.M. Speer, E. Osuch, M.S. George, R.M. Post, Frequency dependence of antidepressant response to left prefrontal repetitive transcranial magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism, Biol. Psychiatry 46 (1999) 1603– 1613. F. Klostermann, T. Funk, J. Vesper, R. Siedenberg, G. Curio, Propofol narcosis dissociates human intrathalamic and cortical high-frequency (.400 Hz) SEP components, NeuroReport 11 (2000) 2607–2610. S. Knecht, T. Ellger, C. Breitenstein, E. Bernd Ringelstein, H. Henningsen, Changing cortical excitability with low-frequency transcranial magnetic stimulation can induce sustained disruption of tactile perception, Biol. Psychiatry 53 (2003) 175–179. S. Nakano, I. Hashimoto, The later part of high-frequency oscillations in human somatosensory evoked potentials is enhanced in aged subjects, Neurosci. Lett. 276 (1999) 83 –86. I. Ozaki, C. Suzuki, Y. Yaegashi, M. Baba, M. Matsunaga, I. Hashimoto, High frequency oscillations in early cortical somatosensory evoked potentials, Electroenceph. clin. Neurophysiol. 108 (1998) 536 –542. A. Pascual-Leone, N.J. Davey, J. Rothwell, E.M. Wassermann, Handbook of Transcranial Magnetic Stimulation, Arnold, London, 2002. R.M. Post, T.A. Kimbrell, M. Frye, M.S. George, U. McCann, J. Little, Implications of kindling and quenching for the possible frequency dependence of rTMS, CNS Spectrums 2 (1997) 54–60. J.R. Romero, D. Anschel, R. Sparing, M. Gangitano, A. PascualLeone, Subthreshold low frequency repetitive transcranial magnetic stimulation selectively decreases facilitation in the motor cortex, Clin. Neurophysiol. 113 (2002) 101–107. P.M. Rossini, A.T. Barker, A. Berardelli, M.D. Caramia, G. Caruso, R.Q. Cracco, M.R. Dimitrijevic, M. Hallett, Y. Katayama, C.H. Lucking, A.L. Maertens de Noordhout, C.D. Marsden, N.M.F. Murray, J.C. Rothwell, M. Swash, C. Tomberg, Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application. Report of an IFCN committee, Electroenceph. clin. Neurophysiol. 91 (1994) 79– 92. T. Satow, T. Mima, J. Yamamoto, T. Oga, T. Begum, T. Aso, N. Hashimoto, J.C. Rothwell, H. Shibasaki, Short-lasting impairment of tactile perception by 0.9 Hz – rTMS of the sensorimotor cortex, Neurology 60 (2003) 1045–1047. M. Tanosaki, I. Hashimoto, Y. Iguchi, T. Kimura, R. Takino, Y. Kurobe, Y. Haruta, Y. Hoshi, Specific somatosensory processing in somatosensory area 3b for human thumb: a neuromagnetic study, Clin. Neurophysiol. 112 (2001) 1516–1522.