Oral structure representation in human somatosensory cortex

NeuroImage 43 (2008) 128–135

Contents lists available at ScienceDirect

NeuroImage j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y n i m g

Oral structure representation in human somatosensory cortex Yohei Tamura a,b,1, Yoshiyuki Shibukawa a,c,⁎,1, Masuro Shintani a,1, Yuzuru Kaneko a,b, Tatsuya Ichinohe a,b a b c

Oral Health Science Center, Laboratory of Brain Research, Tokyo Dental College, Chiba 261-8502, Japan Department of Dental Anesthesiology, Tokyo Dental College, Chiba 261-8502, Japan Department of Physiology, Tokyo Dental College, Chiba 261-8502, Japan

a r t i c l e

i n f o

Article history: Received 28 March 2008 Revised 5 June 2008 Accepted 20 June 2008 Available online 11 July 2008 Keywords: Magnetoencephalography Somatosensory-evoked magnetic field Tactile sensation Intraoral mucosa Primary somatosensory cortex

a b s t r a c t To clarify the topography of the areas representing whole intraoral structures and elucidate bilateral neuronal projection to those areas in the primary somatosensory (S1) cortex, we recorded somatosensoryevoked magnetic fields (SEFs), which reflect the earliest cortical responses to pure tactile stimulation, using magnetoencephalography and a piezo-driven tactile stimulation device. Subjects consisted of 10 healthy male adults. Following tactile stimulation of 6 sites on the oral mucosa (inferior/superior buccal mucosa, posterior/anterior tongue mucosa, and upper/lower lip mucosa), SEFs with a peak latency of 15 ms (1M) were identified bilaterally. In contrast, SEFs with a peak latency of 30 ms following right index finger tactile stimulation were identified only in the contralateral hemisphere. Equivalent current dipoles (ECDs) generating 15 ms components were found along the posterior wall of the central sulcus, bilaterally. The ECD locations for oral mucosa-representing areas were located inferiorly to those for the index finger, with the following pattern of organization from top to bottom along the central sulcus: index finger, upper or lower lip, anterior or posterior tongue and superior or inferior buccal mucosa, with a wide distribution, covering 30% of the S1 cortex. Source strength for 1M in the ipsilateral hemisphere was weaker than that in the contralateral hemisphere. These results clearly indicate that sensory afferents innervating the intraoral region project to both the contralateral and ipsilateral 3b areas via the trigeminothalamic tract, where contralateral projection is predominant. The results clarify the intraoral structure-representing areas in the S1 cortex, adding those areas to the classical “sensory homunculus”. © 2008 Elsevier Inc. All rights reserved.

Introduction Functional mapping of the primary somatosensory (S1) cortex in human was first reported by Penfield and Boldrey (1937). The S1 cortex includes area 3b, known as the “somatosensory homunculus”. The foot-, hand-, and trunk-, as well as orofacial- and intraoralrepresenting areas in the S1 cortex, receive input from peripheral sensory neurons. The results of two earlier studies indicated that cortical representation of orofacial sensation in human was widely distributed over the primary somatosensory cortex, and that this distribution was located more laterally and inferiorly than that for any other body area (Penfield and Boldrey 1937; Penfield and Rasmussen 1950). Using a variety of non-invasive neuroimaging tools such as electroencephalography (EEG), functional-magnetic resonance (functional-MR) imaging and magnetoencephalography (MEG), many authors have reported locations in the S1 cortex for intraoral structures, including the lip (Shöhr and Petruch, 1979; Baumgartner

⁎ Corresponding author. Department of Physiology, Tokyo Dental College, Chiba 261-8502, Japan. Fax: +81 43 270 3771. E-mail address: [email protected] (Y. Shibukawa). 1 Y. Tamura, Y. Shibukawa and M. Shintani contributed equally to this study. 1053-8119/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2008.06.040

et al., 1992; McCarthy et al., 1993; Hoshiyama et al., 1996; Hashimoto, 1988; Nakamura et al., 1998; Nagamatsu et al., 2000; Disbrow et al., 2003; Nakahara et al., 2004; Murayama et al., 2005; Nevalainen et al., 2006), tongue (Picard and Olivier, 1983; Karhu et al., 1991; McCarthy et al., 1993; Nakamura et al., 1998; Disbrow et al., 2003; Nakahara et al., 2004; Murayama et al., 2005), hard palate (McCarthy et al., 1993; Bessho et al., 2007), teeth (dental pulp) (Kubo et al., 2008), and gingiva (Nakahara et al., 2004; Murayama et al., 2005). However, due to electrical noise contamination and/or small amplitudes in EEG/MEG waveforms, these studies were unable to detect early components, which reflect initial cortical neuronal response, excepting studies on the lip (Nagamatsu et al., 2000) and hard palate (Bessho et al., 2007). Initial cortical responses have been identified at around 10–15 ms following stimulation of various orofacial areas in earlier EEG and MEG studies (Shöhr and Petruch, 1979; Hashimoto, 1988; Baumgartner et al., 1992; McCarthy et al., 1993). In MEG studies, initial cortical responses were generated by anterior–superior-oriented current (Nagamatsu et al., 2000; Bessho et al., 2007). Somatosensory stimulation of the trigeminal nerve caused bilateral neural activation of the S1 cortex (Nevalainen et al., 2006). The unilateral somatosensory cortex of the facial area can be resected with no serious facial sensory deficit (Penfield and Rasmussen, 1950; Lehman et al., 1994). In addition, bilateral tongue sensation following

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

unilateral direct cortical S1 stimulation in human has been reported (Penfield and Boldrey, 1937; Penfield and Rasmussen, 1950). Although the ascending pathway, from the trigeminothalamic tract to the ipsilateral cortical representation area, for orofacial and intraoral regions from peripheral sensory neurons is not yet fully understood, two possible projection pathways have been proposed: 1) projection from contralateral cortical activity via the corpus callosum, and 2) direct ipsilateral projection via uncrossed ascending fibers. In monkey studies (Jones et al., 1986; Rausell and Jones, 1991a, b; Manger et al., 1996), it has been reported that ipsilateral cortical representation reflects input from the ipsilateral thalamic ventral posteromedial nucleus innervated by the ipsilateral trigeminal nuclei (Manger et al., 1996). Ipsilateral representation occupied 40% of area 3b trigeminal representation, and 40% of the ventral posteromedial nucleus of the thalamus was devoted to representation of ipsilateral intraoral structures (Manger et al., 1996). Bilateral somatosensoryevoked magnetic field (SEF) responses in human S1 cortex following lip and tongue stimulation have been described (Karhu et al., 1991; Hoshiyama et al., 1996; Disbrow et al., 2003; Nevalainen et al., 2006). These showed peak latencies of approximately 30 to 60 ms, with inferior–posterior orientation of equivalent current dipoles (ECDs) in the bilateral S1 cortices (Karhu et al., 1991; Hoshiyama et al., 1996; Nevalainen et al., 2006). This indicates that these responses were not derived from initial direct cortical activity in the ipsilateral hemisphere, as the initial magnetic component would have shown a shorter latency and anteriorly-oriented currents. In an EEG study, bilateral initial cortical responses of somatosensory-evoked potentials (SEPs) following lip tactile stimulation showed a peak latency of approximately 15 ms in both hemispheres, with no significant differences in latencies (Hashimoto, 1988). However, scalp EEG recordings have low spatial resolution of the current source. In addition, functional-MR imaging technique does not have high

129

enough temporal resolution to distinguish initial neuronal activity in the cortex. Therefore, the aim of the present study was to clarify the functional topography of the areas representing whole intraoral structures, and elucidate bilateral neuronal projection to those areas in the S1 cortex. For this purpose, we recorded SEFs by MEG, which can localize dynamic sources not only with high spatial resolution, but also with very high temporal resolution (Yamamoto et al., 1988; Ribary et al., 1989; Suk et al., 1991; Mogilner et al., 1994). Materials and methods Subjects Subjects consisted of 10 right-handed healthy male adults (mean age, 28 years). All subjects gave written informed consent to participate in the study. The study was approved by the Ethics Committee of Tokyo Dental College in accordance with the Declaration of Helsinki. None of the subjects in our study had any disorder of intraoral sensorimotor function (including acute or chronic pain in the intraoral or orofacial area). Stimulation The stimulation device was modified from Braille cells for the visually impaired (KGS Co., Saitama, Japan) (Fig. 1A). It consisted of a piezoelectric element and stimulus pins pushed out (0.7 mm/0.4 ms) by application of a direct current to the piezoelectric element to produce tactile stimulation to the area targeted. Eight stimulus pins were aligned, with a gap of 2.4 mm between each pin (Fig. 1A). Stimulation was 0.18 N in force and 1 ms in duration, and was applied at randomized intervals of 1– 3 s. The device was applied to each of 6 sites on the right side of the

Fig. 1. (A) Our newly-developed device for stimulation of intraoral mucosa. Stimulus pins were pushed out by application of direct current to piezoelectric element, eliciting tactile stimulation of areas targeted. Inset at right shows higher magnification of eight stimulus pins. Vertical and horizontal bars = 10 mm. (B) Anatomical diagrams showing 6 stimulation sites on intraoral mucosa.

130

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

intraoral mucosa (Fig. 1B): superior or inferior buccal mucosa facing the buccal surface of the upper or lower first molar; anterior or posterior tongue mucosa (anterior position, 1.5 cm from tip of tongue; posterior position, 3.5 cm from tip of tongue); upper or lower lip mucosa, 2 cm lateral to the midline of the lip (Fig. 1B). Stimulation was also applied to the ventral side of the right index finger. All sites in each subject were stimulated in a randomized order in one experimental session. All subjects felt only tactile sensation during stimulation. For electrical stimulation, a constant current of square pulse (SEM-4201, Nihonkohden, Tokyo, Japan) was applied to the right median nerve at the wrist as a control or the ventral side of the right index finger. Stimulus amplitude was set at twice the individual sensory threshold for each subject. The intensity of these electrical stimulations was confirmed to be nonnoxious to the subjects. Stimulations of 0.05 ms in duration were applied at randomized intervals of 1–3 s. Somatosensory-evoked magnetic field recordings A 306-channel superconducting quantum interference device (SQUID) neuromagnetometer (Vectorview, Elekta-Neuromag Co., Helsinki, Finland) was used to record magnetic fields at 102 points on the whole scalp with a pair of gradiometers. To align the MEG data and MR images (1.5 T Symphony Maestro class, Siemens Co., Erlangen, Germany), the positions of 4 head-position indicator coils and 3 anatomical landmarks (the bilateral preauricular points and nasion) were determined with a 3-dimensional digitizer (Isotrak, Polhemus inc., Colchester, VT, U.S.A.). At the beginning of each recording session, weak currents were fed into these coils and the resulting magnetic fields were measured with the sensor array to determine head location. The cortical magnetic signals were digitized at 1 kHz, bandpass-filtered (0.1–100 Hz), and averaged 400 times for tactile and electrical stimulation. Analysis period was set to 500 ms, from 100 ms preceding 400 ms following both electrical and tactile stimuli. Isocontour maps were constructed from the measured data at selected points of time by minimum-norm estimate. The sources of the magnetic fields were modeled as ECDs whose 3dimensional location, orientation and strength were estimated in a spherical conductor model (Source modeling software; Elekta-Neuromag Co). Each ECD was first determined by a least-square search on the basis of magnetic fields at 20–30 points over the response area. Only those ECDs attaining a goodness-of-fit of more than 90% were accepted for further analysis, in which the entire time period and the fields at all the recorded points were taken into account to compute the parameters of a time-varying multi-dipole model (Hämäläinen et al., 1993; Shibukawa et al., 2004, 2007; Bessho et al., 2007). In this model, the strength of the ECDs was allowed to change as a function of time. The locations of ECDs were then superimposed on the brain MR image of each subject to determine the source locations in the brain.

were identified in regions corresponding to subsets of the neuromagnetic sensor array, which was located in the parietal region in the hemispheres both contralateral and ipsilateral to stimulation (circles in Fig. 2). In each waveform showing SEF responses for intraoral mucosa, we consistently identified 3 magnetic components (peaking at approximately 15 ms (1M), 40–60 ms (2M), and 80–120 ms (3M)) in the bilateral hemispheres (Fig. 3). However, in some cases, we could not identify 1M in the ipsilateral hemisphere following intraoral stimulation (Table 1). Table 1 summarizes the latencies for 1M of the contralateral and ipsilateral responses following tactile stimulation of the intraoral mucosa and index finger, as well as electrical stimulation of the median nerve and index finger. There were no significant differences in peak latency for 1M between the contralateral and ipsilateral hemispheres, or among each stimulation site in the intraoral mucosa (Table 1; p N 0.05). Following tactile stimulation of the index finger, we also consistently identified unilaterally 3 magnetic components, the same as for responses following intraoral stimulation. However, the timing of the peak latencies was different (peaking at approximately 25–35 ms (1M), 40–60 ms (2M), and 100–150 ms (3M); Fig. 3). In addition, responses of 1M and 2M were not found in the ipsilateral hemisphere to the tactile-stimulated side (Fig. 3). The mean 1M peak latency was significantly (10 ms) slower than that following median nerve electrical stimulation, and 7 ms slower than that following index finger electrical stimulation (Table 1). Mean 1M peak latencies following index finger electrical/tactile and median nerve stimulation were significantly slower than those following intraoral mucosa stimulation (Table 1). Source distribution and strength Contralateral and ipsilateral ECDs generating 1M components of SEFs following intraoral mucosa stimulation were identified in the posterior banks of the central sulci on each individual head-MR image (Fig. 4; lateral-inferior position of the postcentral gyrus corresponding to the S1 cortex). Bilateral ECDs generating 1M and 2M components of SEFs were located in the same position in the S1 cortex. The ECDs for 1M following intraoral mucosa stimulation were directed anteriorly, whereas those for 2M were directed posteriorly in both hemispheres (not shown). These ECD locations for intraoral mucosa-representing areas were located anteriorly, inferiorly, and laterally to those for the index finger (Fig. 4).

Statistical analysis The results are shown as the mean ± standard error of the mean (S.E.). Differences among latencies in peak responses of SEFs, source strength, and locations of ECDs in 3 axes for each area stimulated were determined using a repeated measures ANOVA. Post hoc comparisons were performed with the Student–Newman–Keuls test. Differences in latencies and source strength between the contralateral and ipsilateral hemispheres were determined using an unpaired t-test. p values of less than 0.05 were considered statistically significant. Results Waveform Using whole-head MEG, we obtained magnetic signals following tactile stimulation of each of 6 sites on the intraoral mucosa (Fig. 1) and one on the index finger. During stimulation, prominent SEF responses

Fig. 2. Typical example of whole-scalp magnetic responses following stimulation of right superior buccal mucosa. Traces area plotted on “flattened head”, as viewed from above, with a subject's nose oriented upwards. Each trace began 100 ms prior to, and ended 400 ms after, stimulus onset. Circles indicate regions showing prominent magnetic signals.

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

131

Fig. 3. Enlarged traces of magnetic signals from parietotemporal region in contralateral (upper traces) and ipsilateral (lower traces) hemispheres during right-side tactile stimulation of each intraoral mucosa site, and index finger. These traces were obtained from four selected channels in region showing prominent magnetic signals (circles in A; Fig. 2). The 3 successive identifiable peak signal components were termed 1M, 2M, and 3M in bilateral hemisphere following intraoral mucosa stimulation. Dashed vertical lines in each trace show peak signal components of 1M (red), 2M (blue), and 3M (black).

The ECD of the initial magnetic component for median nerve electrical stimulation was localized in the upper area of the postcentral gyrus, adjacent to the index finger ECD position (not shown). To further investigate the localization of the intraoral mucosarepresenting area in the S1 cortex, we compared the mean of the 3dimensional ECD locations (Fig. 5). Each ECD following intraoral stimulation was located at an inferior orientation in the S1 cortex, compared with that for the index finger, with the following pattern of organization from top to bottom along the central sulcus: index finger, upper or lower lip, anterior or posterior tongue, and superior or inferior buccal mucosa. In addition, buccal and tongue mucosa ECD locations in the S1 cortex were significantly inferior to the lip ECD locations. However, there were no significant differences in ECD locations between buccal and tongue mucosa. In addition, the ECD locations elicited by stimulation for 2 sites within the same oral structure were not significantly separated (Fig. 5; p N 0.05). We compared differences in source strength (dipole moment) for 1M following tactile stimulation of each site on the intraoral

Table 1 Averaged 1M latencies (ms) of SEFs following tactile stimulation of each site on intraoral mucosa in contralateral and ipsilateral hemispheres Stimulation sites

Inferior buccal Superior buccal Posterior tongue Anterior tongue Lower lip Upper lip Index finger tactile stimulation Index finger electrical stimulation Median nerve electrical stimulation

Latencies for 1M (mean ± SEM (n)) Contralateral

Ipsilateral

16 ± 1 (10)⁎ 15 ± 1 (10)⁎ 14 ± 1 (10)⁎ 14 ± 1 (10)⁎ 15 ± 1 (10)⁎ 16 ± 1 (10)⁎ 30 ± 1 (10)† 23 ± 1 (10)b 20 ± 1 (10)

18 ± 1 (8) 17 ± 2 (8) 15 ± 1 (7) 14 ± 1 (8) 15 ± 1 (8) 15 ± 1 (8) N.D. N.D. N.D.

N.D., not detected; ⁎, p b 0.05 (Comparison with index finger tactile/electrical stimulation and median nerve); †, p b 0.05 (Comparison with median nerve); b, p b 0.05 (Comparison with index finger tactile stimulation).

132

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

Fig. 4. ECD locations from SEFs following tactile stimulation of each region of right intraoral mucosa and right index finger on frontal (left images), sagittal on left (middle images) and coronal (right images) planes of MR images. Anteriorly-oriented dipole ECDs were identified along posterior wall of central sulcus (area 3b) in bilateral hemispheres. Blue circle shows location of equivalent current dipole on MR images; blue bar attached to circle indicates size and direction of equivalent current dipole (end of bar attached to circle represents positive pole).

mucosa and index finger between contralateral and ipsilateral hemispheres to the stimulation site (Fig. 6). There were no significant inter-hemispheric differences in mean values for source strength (p N 0.05), except for with inferior buccal mucosa and posterior tongue mucosa (Fig. 6). However, these source strength values for 1M in the ipsilateral hemisphere were generally smaller (approximately 5.0 nAm vs. 3.0 nAm) than those in the contralateral hemisphere (Fig. 6).

Extent of intraoral region in S1 cortex To determine the extent of the distribution for the intraoral region in the contralateral S1 cortex, we normalized each value in the Z-axis (Z) for ECDs against the height of the postcentral gyrus according to the following formula: Normalized position ¼ ðZ−Zb Þ=ðZt −Zb Þ

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

133

Fig. 5. Averaged ECD locations on each axis for 1M. Comparison of three-dimensional ECD locations for each X-, Y- and Z-axis between each region of intraoral mucosa and index finger-representing area are shown for 1M components. Data points represent mean ± SEM from 10 participants. ⁎p b 0.05; Significant differences in ECD locations between each stimulus area and right index finger.

where Zb is the Z-axis value at the intersection point of the sylvian fissure and central sulcus, and Zt is the value at the top of the central sulcus in the individual head-MR image. As shown in Fig. 7, the intraoral representation area had an area ranging from 27 ± 8% to 54 ± 16% from Zb (where Zt was 100%), and occupied approximately 30% of the entire S1 cortex (Fig. 7). Discussion Tactile sensation is projected to area 3b in the S1 cortex mainly via A-beta fibers, which have rapid conduction velocity. Therefore, it was necessary to record SEFs using MEG with high temporal resolution. However, there are difficulties in recording trigeminal SEFs: 1) artifacts contaminate SEF responses, as SQUID sensors are in contiguity with the oral region; 2) the oral mucosa is exposed to a moist environment by saliva, which interferes with site-specific electrical stimulation due to current leak; and 3) electrical stimulation of the tongue may induce an electrical taste (Bujas et al., 1979; Yamamoto et al., 2003). To avoid these difficulties and record initial cortical neuronal activities induced by pure tactile sensation, we developed a piezo-actuator tactile stimulation device driven by direct current. Using this piezo-actuator tactile stimulation device allowed us to record initial components with a peak latency of 15 ms in the magnetic waveforms, which reflect bilateral responses in the cortex. The ECDs generating 1M components were located along the posterior wall of the central sulcus bilaterally, with anteriorly-oriented dipoles, as the cortical pyramidal neurons in the 3b area lie in an anatomically anterior–posterior orientation. Therefore, the present results clearly indicate that 1M components of SEFs following tactile stimulation of intraoral mucosa reflect evoked activities from neurons in area 3b in the S1 cortex, rather than those in area 1, or even area 2 or 3a. In addition, the contralateral 1M component generated by anteriorly-oriented dipoles following index finger tactile stimulation had a latency of 30 ms (Table 1, and Figs. 3 and 4). This also indicates that this component reflects initial cortical activity in area 3b neurons following index finger tactile stimulation. In contrast, previous MEG studies using air-pressure driven pneumatic tactile stimulation device (s) were unable to record/detect/estimate earlier cortical activities generated by anterior-oriented current dipoles in the S1 cortex following index finger stimulation (Simões et al., 2001; Druschky et al., 2003; Pihko et al., 2004). This was due to the relative weakness of the stimulus and/or indistinct onset of the tactile stimulus, which

resulted in insufficient synchronization of the respective neural population (Pihko et al., 2004). Low amplitude of magnetic field response with low signal-to-noise ratio does not allow for an adequate analysis of source current (ECD) localization (Nevalainen et al., 2006). In the present study, however, we successfully recorded earliest cortical responses, irrespective of stimulation site, using a piezodriven tactile stimulation device that was more temporally precise, with a shorter rising time (0.4 ms) and duration (1 ms) than that available with other pneumatic tactile stimulator device(s) (Pihko et al., 2004; Nevalainen et al., 2006). Therefore, the stimulator we used is optimal for studying earliest cortical somatosensory activity in research, as well as for presurgical MEG evaluation of 3-dimensional functional anatomy of the cortex in a clinical setting (e.g., Mäkelä et al., 2001) following tactile stimulation. Mean 1M peak latency with index finger tactile stimulation was significantly (7 ms) slower than that with electrical stimulation. This difference probably reflects the encoding process from generator potential to action potential generation in the peripheral sensory neurons with tactile stimulation. If so, encoding time at the periphery

Fig. 6. Averaged source strength (dipole moment) for 1M following stimulation of each site on intraoral mucosa in contralateral and ipsilateral hemispheres. Values represent mean ± SEM from 10 participants. Source strength of posterior buccal mucosa and inferior tongue area showed significant differences between contralateral and ipsilateral hemispheres. There were no significant differences in source strength among each site of intraoral mucosa. ⁎p b 0.05; Significant differences in source strength (p b 0.05).

134

Y. Tamura et al. / NeuroImage 43 (2008) 128–135

Fig. 7. Extent of oral region in S1 cortex. To determine extent of distribution of intraoral region in contralateral S1 cortex, we normalized each value of Z-axis for ECD against height of postcentral gyrus (see text). Intraoral representation area (from inferior buccal mucosa to upper lip mucosa) occupied approximately 30% of entire S1 cortex.

would be 7 ms, and actual conduction time from the oral mucosa to the cortex would be 8 ms. Furthermore, actual conduction time from the index finger to the cortex would be 23 ms. The 3-dimensional distance from the cortex to the index finger is approximately 3 times greater than that to the oral cavity. This is in accordance with the difference in conduction time to the cortex from the oral mucosa and hand. In this study, bilateral ECDs generating 1M components were located in area 3b of the bilateral S1 cortex. These contralateral and ipsilateral ECDs were symmetrically located in the same orientation. We found no significant inter-hemispheric differences in initial peak latencies for bilateral 1M in any of the tactile stimulation sites investigated. This indicates that ipsilateral projection from contralateral cortical activity via the corpus callosum is unlikely, as the ipsilateral peak latency for 1M would be slower than contralateral latencies for 1M if ipsilateral responses were propagated through the corpus callosum (Nevalainen et al., 2006). Therefore, our present results suggest ipsilateral direct projection via an uncrossed ascending pathway in human, from the trigeminothalamic tract to the cortical representation area. The source strength of ECDs generating 1M components in the contralateral hemisphere was larger than that in the ipsilateral hemisphere, suggesting that cortical 3b neurons in the contralateral S1 cortex produce larger amplitude excitatory postsynaptic potentials after stimulation. In monkey area 3b, the percentage of neurons with a contralateral receptive field was 52%, that for ipsilateral was 31%, and that for bilateral was 17% (Toda and Taoka, 2004). In the present results, the mean source strength of 1M for oral mucosa was approximately 5.0 nAm for contralateral and approximately 3.0 nAm for ipsilateral to the stimulation site. Thus, our results indicate that sensory afferents innervating the intraoral region project to both contralateral and ipsilateral area 3b in the S1 cortices via crossed and uncrossed ascending pathways, respectively, and that the contralateral ascending pathway for projection is predominant. The 3-dimensional ECD locations for intraoral mucosa-representing areas were positioned according to the following pattern of somatotopic organization from top to bottom along the central sulcus: index finger, upper or lower lip, anterior or posterior tongue, and superior or inferior buccal mucosa, with a wide distribution, occupying 30% of the S1 cortex. It has been reported that the coordinates for dental pulp were located at 80 mm on the Z-axis (Kubo et al., 2008), as

in this study. According to their data, the ECD for dental pulp was suggested to be located between the lip- and tongue mucosarepresentation areas. Our results, together with those of Kubo et al., on the somatotopic organization of the tongue- and lip-representing areas, including dental pulp, in the intraoral cortex agree with the results of intracranial recording (Penfield and Rasmussen, 1950), with organization corresponding precisely with the “sensory homunculus”. The present results showing 3-dimensional sequential localization patterns provide solid evidence for somatotopic representation of intraoral structures in the 3b area. The ECD locations for buccal mucosa in this study were localized at the most inferior position in area 3b of the S1 cortex. Therefore, we are able to add the buccal mucosarepresentation areas to the classical “sensory homunculus”. The ECD locations elicited by stimulation of 2 sites within the same oral structure were not significantly separated in the cortex, and appeared to be relatively contiguous and overlapped. Distribution in the somatosensory cortex with regard to a particular structure is directly related to its innervation density (Mountcastle, 1984). Oral structures are known to be characterized by high neuronal innervation densities (Yamada et al., 1952; Ringel and Ewanowski,1965). Interestingly, a recent neuromagnetic study suggested that ECD locations for the “posterior” area of the upper lip were inferior to those for the “anterior” area of the lower lip (Nakahara et al., 2004). In earlier monkey studies, neurons in area 3b had bimaxillary receptive fields covering the upper and lower lips (Lin et al., 1994; Toda and Taoka, 2002). Those receptive fields included corresponding sites on the lips that would come into contact when the jaws were closed, and were considered to be neural substrates for oral stereognosis (Toda and Taoka, 2004). Neurons with composite receptive fields covering more than one oral structure may be suitable for the detection or manipulation of objects in the oral vestibule or oral cavity proper (Toda and Taoka, 2004). In our results, each ECD location in the same oral structure was contiguous and overlapped. Therefore, sensory information from the oral cavity in the S1 cortex may be integrated, where necessary, as a neural substrate for oral stereognosis to drive orofacial function such as articulation, mastication and oral object exploration. In conclusion, we detected an initial component with a peak latency of 15 ms in magnetic waveforms bilaterally following intraoral mucosa tactile stimulation, which reflects bilateral earliest response in the cortex. The ECDs generating those 1M components were located in bilateral area 3b, with an anterior–superior orientation. The results of this study clearly indicate that sensory afferents innervating the intraoral region project to both contralateral and ipsilateral 3b area in the S1 cortices via the trigeminothalamic tract. A contralateral ascending pathway of projection was predominant. The present results clarify the topography of whole intraoral structure-representing areas in the S1 cortex, adding these areas to the classical “sensory homunculus”. Acknowledgments This study was supported by a grant for High-tech Research Center Projects (HRC6A01 and HRC6A03) from the MEXT (Ministry of Education, culture, Sports, Science and Technology) of Japan. YS is a recipient of Grants-in-Aid (Nos. 18592050 and 20592187) for Scientific Research from the MEXT of Japan, and a Grant from the Dean of Tokyo Dental College. I wish to sincerely thank Drs. H. Bessho and K. Kubo for their continuous support for this work. I would also like to thank Professor Jeremy Williams, Tokyo Dental College, for his assistant with the English of this manuscript.

References Baumgartner, C., Barth, D.S., Levesque, M.F., Sutherling, W.W., 1992. Human hand and lip sensorimotor cortex as studied on electrocorticography. Electroenceph. Clin. Neurophysiol 84, 115–126.

Y. Tamura et al. / NeuroImage 43 (2008) 128–135 Bessho, H., Shibukawa, Y., Shintani, M., Yajima, Y., Suzuki, T., Shibahara, T., 2007. Localization of palatal area in human somatosensory cortex. J. Dent. Res 86, 265–270. Bujas, Z., Frank, M., Pfaffmann, C., 1979. Neural effects of electrical taste stimuli. Sens. Processes 3, 353–365. Disbrow, E.A., Hinkley, L.B., Roberts, T.P., 2003. Ipsilateral representation of oral structures in human anterior parietal somatosensory cortex and integration of inputs across the midline. J. Comp. Neurol 467, 487–495. Druschky, K., Kaltenhäuser, M., Hummel, C., Druschky, A., Huk, W.J., Neundörfer, B., Stefan, H., 2003. Somatosensory evoked magnetic fields following passive movement compared with tactile stimulation of the index finger. Exp. Brain Res 148, 186–195. Hämäläinen, R., Hari, R., Ilmoniemi, R.J., Knuutila, J., Lounasmaa, O.V., 1993. Magnetoencephalography-theory, instrumentation, and applications to noninvasive studies of the working human brain. Rev. Mod. Phys 65, 413–497. Hashimoto, I., 1988. Trigeminal evoked potentials following brief air puff: enhanced signal-to-noise ratio. Ann. Neurol 23, 332–338. Hoshiyama, M., Kakigi, R., Koyama, S., Kitamura, Y., Shimojo, M., Watanabe, S., 1996. Somatosensory evoked magnetic fields following stimulation of the lip in humans. Electroencephalogr. Clin. Neurophysiol 100, 96–104. Jones, E.G., Schwark, H.D., Callahan, P.A., 1986. Extent of the ipsilateral representation in the ventral posterior medial nucleus of the monkey thalamus. Exp. Brain Res 63, 310–320. Karhu, J., Hari, R., Lu, S.T., Paetau, R., Rif, J., 1991. Cerebral magnetic fields to lingual stimulation. Electroencephalogr. Clin. Neurophysiol 80, 459–468. Kubo, K., Shibukawa, Y., Shintani, M., Suzuki, T., Ichinohe, T., Kaneko, Y., 2008. Cortical representation area of human dental pulp. J. Dent. Res 87, 358–362. Lehman, R., Andermann, F., Olivier, A., Tandon, P.N., Quesney, L.F., Rasmussen, T.B., 1994. Seizures with onset in the sensorimotor face area: clinical patterns and results of surgical treatment in 20 patients. Epilepsia 35, 1117–1124. Lin, L.-D., Murray, G.M., Sessle, B.J., 1994. Functional properties of single neurons in the primate face primary somatosensory cortex. I. Relations with trained orofacial motor behaviors. J. Neurophysiol 71, 2377–2400. Mäkelä, J.P., Kirveskari, E., Seppä, M., Hämäläinen, M., Forss, N., Avikainen, S., Salonen, O., Salenius, S., Kovala, T., Randell, T., Jääskeläinen, J., Hari, R., 2001. Threedimensional integration of brain anatomy and function to facilitate intraoperative navigation around the sensorimotor strip. Hum. Brain Mapp 12, 180–192. Manger, P.R., Woods, T.M., Jones, E.G., 1996. Representation of face and intra-oral structures in area 3b of macaque monkey somatosensory cortex. J. Comp. Neurol 371, 513–521. McCarthy, G., Allison, T., Spencer, D.D., 1993. Localization of the face area of human sensorimotor cortex by intracranial recording of somatosensory evoked potentials. J. Neurosurg 79, 874–884. Mogilner, A., Nomura, M., Ribary, U., Jagow, R., Lado, F., Rusinek, H., Llinás, R., 1994. Neuromagnetic studies of the lip area of primary somatosensory cortex in humans: evidence for an oscillotopic organization. Exp. Brain Res 99, 137–147. Mountcastle, V.B., 1984. Central nervous mechanisms in mechanoreceptive sensibility. In: Darian-smith, I. (Ed.), Handbook of Physiology—the Nervous System. American Physiological Society, Bethesda, MD, pp. 789–876. Murayama, S., Nakasato, N., Nakahara, H., Konno, A., itoh, H., 2005. Neuromagnetic evidence that gingival area is adjacent to tongue area in human primary somatosensory cortex. Tohoku J. Exp. Med 207, 191–195. Nagamatsu, K., Nakasato, N., Hatanaka, K., Kanno, A., Iwasaki, M., Yoshimoto, T., 2000. Neuromagnetic localization of N15, the initial cortical response to lip stimulus. Neuroreport 12, 1–5. Nakahara, H., Nakasato, N., Kannno, A., Murayama, S., Hatanaka, K., Itoh, H., Yoshimoto, T., 2004. Somatosensory-evoked fields for gingival, lip and tongue. J. Dent. Res., 83, 307–311.

135

Nakamura, A., Yamada, T., Goto, A., Kato, T., Ito, K., Abe, Y., Kachi, T., Kakigi, R., 1998. Somatosensory homunculus as drawn by MEG. Neuroimage 7, 377–386. Nevalainen, P., Ramstad, R., Isotalo, E., Haapanen, M.L., Lauronen, L., 2006. Trigeminal somatosensory evoked magnetic fields to tactile stimulation. Clin. Neurophysiol 117, 2007–2015. Penfield, W., Boldrey, E., 1937. Somatic motor and somatosensory representation in cerebral cortex of man as studied by electrical stimulation. Brain 60, 389–443. Penfield, W., Rasmussen, T., 1950. A clinical study of localization of function. the cerebral cortex of man. Macmillan, New York. Picard, C., Olivier, A., 1983. Sensory cortical tongue representation in man. J. Neurosurg 59, 781–789. Pihko, E., Lauronen, L., Wikström, H., Taulu, S., Nurminen, J., Kivitie-Kallio, S., Okada, Y., 2004. Somatosensory evoked potentials and magnetic fields elicited by tactile stimulation of the hand during active and quiet sleep in newborns. Clin. Neurophysiol 115, 448–455. Rausell, E., Jones, E.G., 1991a. Chemically distinct compartments of the thalamic VPM nucleus in monkeys relay principal and spinal trigeminal pathways to different layers of the somatosensory cortex. J. Neurosci 11, 226–237. Rausell, E., Jones, E.G., 1991b. Histochemical and immunocytochemical compartments of the thalamic VPM nucleus in monkeys and their relationship to the representational map. J. Neurosci 11, 210–225. Ribary, U., Llinás, R., Kluger, A., Suk, J., Ferris, S.H., 1989. Neuropathological dynamics of magnetic auditory steady-state responses in Alzheimer disease. In: Williamson, S.J., Hoke, M., Stroink, G., Kotani, M. (Eds.), Advances in biomagnetism. Plenum, New York, pp. 311–314. Ringel, R.L., Ewanowski, S.J., 1965. Oral perception. I. Two-point discrimination. J. Speech Hear Res 8, 389–398. Shibukawa, Y., Shintani, M., Kumai, T., Suzuki, T., Nakamura, Y., 2004. Cortical neuromagnetic fields preceding voluntary jaw movements. J. Dent. Res 83, 572–577. Shibukawa, Y., Ishikawa, T., Kato, Y., Zhen, Z.K., Jiang, T., Shintani, M., Shimono, M., Kumai, T., Suzuki, T., Kato, M., Nakamura, Y., 2007. Cerebral cortical dysfunction in patients with temporomandibular disorder in association with jaw movement observation. Pain 128, 180–188. Shöhr, M., Petruch, F., 1979. Somatosensory evoked potentials following stimulation of the trigeminal nerve in man. J. Neurol 220, 95–98. Simões, C., Mertens, M., Forss, N., Jousmäki, V., Lütkenhöner, B., Hari, R., 2001. Functional overlap of finger representations in human SI and SII cortices. J. Neurophysio 86, 1661–1665. Suk, J., Ribary, U., Cappell, J., Yamamoto, T., Llinás, R., 1991. Anatomical localization revealed by MEG recordings of the human somatosensory system. Electroencephalogr. Clin. Neurophysiol 78, 185–196. Toda, T., Taoka, M., 2002. Integration of the upper and lower lips in the postcentral area 2 of conscious macaque monkeys (Macaca fuscata). Arch. Oral Biol 47, 449–456. Toda, T., Taoka, M., 2004. Converging patterns of inputs from oral structures in the postcentral somatosensory cortex of conscious macaque monkeys. Exp. Brain Res 158, 43–49. Yamada, M., Maruhashi, J., Miyake, N., 1952. The distribution of sensory spots on the oral mucous membrane. Jpn. J. Physiol 2, 328–332. Yamamoto, C., Takehara, S., Morikawa, K., Nakagawa, S., Yamaguchi, M., Iwaki, S., Tonoike, M., Yamamoto, T., 2003. Magnetoencephalographic study of cortical activity evoked by electrogustatory stimuli. Chem. Senses 28, 245–251. Yamamoto, T., Williamson, S.J., Kaufman, L., Nicholson, C., Llinás, R., 1988. Magnetic localization of neuronal activity in the human brain. Proc. Natl. Acad. Sci. U. S. A 85, 8732–8736.