Articles in PresS. J Neurophysiol (November 7, 2012). doi:10.1152/jn.01044.2011
1
1
Transcallosal inhibition in patients with callosal infarction
2
3
Jie-Yuan Li,1,3 Ping-Hong Lai,2,3 and Robert Chen4
4
1
5
6
Hospital, Kaohsiung, Taiwan (R.O.C.)
3
7
8
9
Division of Neurology and 2Department of Radiology, Kaohsiung Veterans General
Faculty of Medicine, School of Medicine, National Yang-Ming University, Taipei, Taiwan
(R.O.C.)
4
Division of Neurology, Department of Medicine and Toronto Western Research Institute,
University of Toronto, Toronto, Ontario, Canada
10
Address reprint requests and other correspondence to:
11
12
13
14
15
16
17
18
19
Dr. Robert Chen
Toronto Western Hospital, 7MC411,
399 Bathurst Street,
Toronto, Ontario,
Canada M5T 2S8
Email: [email protected]
Tel.: +1-416-603 5207
Fax: +1-416-603 5004
20
Running head: Transcallosal inhibition in callosal infarction
21
22
23
24
25
26
27
28
29
Author contributions:
Dr. J-Y Li: Conception and design of research, performed experiments, analyzed data,
interpreted results of experiments, prepared figures, drafted and approved final version of
manuscript.
Dr. P-H Lai: Performed experiments, prepared figures, approved final version of
manuscript.
Dr. R. Chen: Conception and design of research, analyzed data, interpreted results of
experiments, edited and revised manuscript, approved final version of manuscript
Copyright © 2012 by the American Physiological Society.
2
30
31
ABSTRACT
Recent studies in normal subjects suggested that callosal motor fibers pass through
32
the posterior body of the corpus callosum (CC) but this has not been tested in patients with
33
callosal infarction. The objective of the study is to define the pathways involved in
34
transcallosal inhibition by examining patients with infarctions in different sub-regions of
35
the CC. We hypothesized that patients with lesions in the posterior half of the CC would
36
have greater reduction in transcallosal inhibition between the motor cortices. Twenty six
37
patients with callosal infarction and 14 healthy subjects were studied. The callosal lesions
38
were localized on sagittal MRI and were attributed to one of five segments of the CC.
39
Transcranial magnetic stimulation was used to assess ipsilateral silent period (iSP), short
40
(SIHI) and long latency (LIHI) interhemispheric inhibition originating from both motor
41
cortices. The results showed that the iSP areas and durations were markedly reduced
42
bilaterally in patients with callosal infarction compared to normal subjects. Patients with
43
callosal infarctions also had less IHI bidirectionally compared to normal subjects. iSP
44
areas and durations were lower in patients with lesions than in patients without lesions in
45
segments 3 (posterior midbody) of the CC. Lesion burden in the posterior half of the CC
46
negatively correlated transcallosal inhibition measured with iSP and SIHI. Our study
47
suggests that callosal infarction led to reduced transcallosal inhibition as measured by iSP,
3
48
SIHI and LIHI. Fibers mediating transcallosal inhibition cross the CC mainly in the
49
posterior half.
50
51
52
53
Keywords: TMS, Interhemispheric inhibition, Ipsilateral silent period, Callosal infarction
4
54
55
INTRODUCTION
The corpus callosum (CC) is the major white matter tract connecting the cerebral
56
hemispheres. It is considered the most important structure for interhemispheric
57
communication of sensory, motor and higher-order information between the hemispheres.
58
The CC is also important for the coordination of bimanual movements. Bimanual
59
coordination deficit can be seen in patients with lesions of the CC or partial callosotomy
60
(Caille et al. 2005; Jeeves et al. 1988). Infarcts of the CC are rare and this is most likely
61
due to the rich blood supply from the main arterial systems, specifically the anterior
62
cerebral, anterior communicating, and posterior cerebral arteries. Lesions of the CC
63
producing disturbances of higher brain functions are often recognized as disconnection
64
syndrome, in which unilateral left hand apraxia, agraphia and tactile anomia are most
65
commonly encountered (Giroud and Dumas 1995; Watson and Heilman 1983; Yamadori et
66
al. 1980). In addition, specific syndromes such as alien hand syndrome (Aboitiz et al. 2003;
67
Feinberg et al. 1992) and an isolated gait disorder (Giroud and Dumas 1995) had been
68
described in relation to lacunas in the anterior portion of the CC. Therefore, patients with
69
callosal infarction may serve as a clinical model to investigate interhemispheric
70
connections.
71
The transcallosal connection between motor cortices (M1) has been studied in human
5
72
with transcranial magnetic stimulation (TMS), including transcallosal inhibition and
73
facilitation. Transcallosal inhibition refers to suppression of one hemisphere by the
74
opposite hemisphere, which may help to maintain hemispheric dominance in cognitive and
75
motor tasks. There are two established TMS methods to evaluate transcallosal inhibition in
76
human M1. The first is ipsilateral silent period (iSP) (Meyer et al. 1999), which involves
77
the interruption of voluntary muscle activities following stimulation of the ipsilateral M1.
78
The second method is interhemispheric inhibition (IHI) and involves a conditioning
79
stimulus given over one M1 followed by the test stimulus over the opposite M1 8-50 ms
80
later. IHI can be divided into short interval IHI (SIHI, interstimulus intervals ~ 10 ms) and
81
long interval IHI (LIHI, ~ 50 ms) with different physiologic properties (Chen et al. 2003a;
82
Ni et al. 2009). Pharmacological studies suggest that LIHI likely involves
83
GABAB-mediated inhibition, while the receptor mediating SIHI remains unknown
84
(Irlbacher et al. 2007). Although TMS is widely used to study neurological and psychiatric
85
diseases, IHI has not been explored in patients with callosal infarction.
86
The topographical organization of the CC has been the focus of several previous
87
studies. Anatomical studies in humans showed the fibers connecting the M1 pass through
88
the anterior midbody of the CC (de Lacoste et al. 1985; van Valkenburg C.T. 1913;
89
Witelson 1989). Recently, diffusion tensor imaging (DTI) in conjunction with
6
90
tractography has been used to visualize major fiber tracts in the human CC in vivo to
91
identify its detailed topographical organization (Abe et al. 2004). These DTI studies and a
92
TMS study suggested a different topographical arrangement of callosal connectivity with
93
the callosal motor fibers crossing more posteriorly, in the posterior body of the CC (Fling et
94
al. 2011; Hofer and Frahm 2006; Meyer et al. 1998; Wahl et al. 2007; Zarei et al. 2006).
95
The aims of this study are to examine how callosal infarction affects transcallosal
96
inhibition and to define the pathways involved in the transcallosal inhibition measured by
97
TMS by examining patients with infarction of different sub-regions of the CC. We
98
hypothesized that the posterior half of the CC plays a greater role than the anterior half in
99
generating transcallosal inhibition to the motor cortex. The reduction of iSP and IHI
100
should be greater in patients with callosal infarction involving the posterior half of the CC.
7
101
MATERIALS AND METHODS
102
Subjects
103
We studied 26 patients (19 men, aged 62.9 ± 11.1 years) with anterior cerebral artery
104
territory infarcts with callosal involvement and 14 age-matched normal subjects (8 men,
105
aged 56.6 ± 8.9 years). There was no involvement of the middle cerebral artery territory in
106
all patients. Intermanual conflict was seen in 5 patients, apraxia in 13, tactile anomia in 6
107
and agraphia in 14. All the patients and normal subjects were right-handed. Handedness of
108
the subjects was evaluated by a modified version of the Edinburgh Handedness Inventory
109
(Oldfield 1971). Table 1 shows the clinical features of the patients. The study was
110
approved by Institutional Review Board-Kaohsiung Veterans General Hospital. Written
111
informed consent was obtained from all subjects.
112
Magnetic Resonance Imaging
113
All MR imaging were performed with a clinical 1.5 T system (General Electric
114
Medical System, Milwaukee, WI). The routine imaging studies included axial and coronal
115
T1-weighted spin-echo (500/30/2 ms [repetition time/echo time/excitations]), T2-weighted
116
fast spin-echo (4000/100/2 ms) with echo train length 8, and axial fast fluid-attenuated
117
inversion recovery (9000/2200/133/1 ms [repetition time/inversion time/echo time/number
118
of excitations]) sequences and sagittal T2-weighted and diffusion-weighted Imaging
8
119
(DWI). The imaging sequence for DWI was a single-shot spin-echo echo-planar imaging
120
(10000/93 ms [repetition time/echo time]) with diffusion sensitivities b=0 s/mm2 and
121
b=1000 s/mm2. An apparent diffusion coefficient (ADC) map was calculated. Sections (5
122
mm thick) with 2.5 mm interslice gaps, 24 cm field of view, and 256 x 192 matrix were
123
used for all scans. The callosal lesions were localized on sagittal MRI (Fig. 1A) and then
124
attributed to one of five segments of the CC numbered from anterior to posterior according
125
to the scheme (Fig. 1B) proposed by Hofer et al. (Hofer and Frahm 2006). A baseline was
126
drawn through the most inferior borders of the splenium and rostrum of the CC. From this
127
line, perpendicular lines were drawn at the anterior edge of the genu and the posterior edge
128
of the splenium and then baseline was divided into five segments. Segment 1 covered the
129
first sixth of the CC. Segment 2 was the rest of the anterior half of the CC. Segment 3 was
130
the posterior half minus the posterior third, segment 4 was the posterior one-third minus
131
posterior one-fourth and segment 5 was the posterior one fourth.
132
EMG recording
133
EMG was monitored on a computer screen and via loudspeakers at high gain to
134
provide feedback on the state of muscle relaxation. Visual feedback was provided by
135
passing the EMG signal through a leaky integrator and the EMG level was displayed on an
136
oscilloscope. The signal was amplified (Digitimer D360, Letchworth Garden, UK),
9
137
filtered (band pass 20 Hz to 2.5 kHz), digitized at 5 kHz (Power 1401, Cambridge
138
Electronics Design, Cambridge, UK) and stored in a laboratory computer for off-line
139
analysis.
140
TMS studies
141
Two 70 mm figure-of-eight coils and two Magstim 200 Stimulators (Magstim
142
Company, Whitland, UK) were used. Both M1 were tested in patients with callosal
143
infarction and in normal subjects. Lesioned side refers to the side with infarction. Surface
144
EMG was recorded from both first dorsal interosseous (FDI) muscles. The optimal coil
145
position over left M1 for eliciting the motor evoked potential (MEP) from the right FDI
146
muscle was established with the handle of the coil held about 45 degrees to the midsagittal
147
line (approximately perpendicular to the presumed direction of the central sulcus). The
148
optimal position was marked on the scalp to ensure identical placement of the coil
149
throughout the experiment. This procedure was then repeated for the right M1 and the left
150
FDI muscle. Resting motor threshold (MT) was the minimum stimulator output that
151
produced MEPs of ≥ 50μV in at least 5 out of 10 trials. Active MT was the minimum
152
stimulator output that produced MEPs of ≥ 100μV in at least 5 out of 10 trials with a
153
constant background contraction of 20% of the maximum integrated EMG.
10
154
Contralateral motor evoked potential and ipsilateral silent period
155
TMS at 50, 75 and 100% of the stimulator output were applied in random order to the
156
M1 with the subject maintaining a 50% maximum contraction of the ipsilateral FDI muscle
157
with visual and auditory feedback (Chen et al. 2003b). Subjects took breaks whenever
158
necessary to avoid muscle fatigue. Each stimulus intensity was repeated 10 times and the
159
stimuli were presented 5 seconds apart.
160
Interhemispheric inhibition (IHI)
161
The protocol was similar to that described by Ferbert et al. (Ferbert et al. 1992a). The
162
subjects relaxed both FDI muscles. The conditioning stimuli were applied to the M1 at
163
75% of the stimulator output with the induced current flowing in the posterior-medial
164
direction (handle of the coil pointed forward and laterally). This stimulus intensity and
165
orientation were chosen because in some subjects it was not possible to place both coils at
166
the optimal positions with the handle pointed backwards and laterally due to the size of the
167
coil. A previous study found no difference in the IHI between 75% and 90% of the
168
stimulator output and between four coil orientations 90 degrees apart (Chen et al. 2003b).
169
The test stimuli were applied to the opposite M1 with the induced current in the
170
anterior-medial direction (handle pointed backward and laterally) and were adjusted to
171
evoke about 1 mV MEP in the contralateral FDI muscle. Test pulse alone and the
11
172
interstimulus intervals (ISIs) of 6, 8, 10, 20 and 50 ms were tested. Each run consisted of
173
10 trials of the test pulse alone and 10 trials of each ISI delivered in random order (60
174
trials).
175
Data Analysis
176
All patients were included in the analysis. Segments 1 and 2 were the anterior half
177
and segments 3, 4 and 5 were the posterior half of the CC. Segment 2 was defined as the
178
anterior midbody and segment 3 as the posterior midbody. For MR images, each CC
179
segment was considered affected if the lesion occupied more than half of the area and was
180
considered unaffected if the lesion occupied less than half of the area. Since the size of
181
each segment is different with size ratios of 2:4:2:1:3 from segments 1 to 5, we assigned
182
lesion burden in proportion to the size of each segment. Therefore, the lesion burden score
183
was assigned as 2 if segment 1 was affected, 4 for segment 2, 2 for segment 3, 1 for
184
segment 4 and 3 for segment 5. Lesion burden for the anterior half of the CC was the sum
185
of scores for segments 1 and 2. Lesion burden for the posterior half of the CC was the sum
186
of each score for segments 3, 4 and 5. For example, a patient with lesions in segments 3
187
and 4 will have a lesion burden of 3 (2 + 1) for the posterior half of the CC. The assessor for
188
MRI was blinded to the TMS results.
189
For TMS studies, the peak-to-peak MEP amplitude for each trial was measured. The
12
190
unconditioned MEP amplitudes and the conditioned MEP amplitudes at each ISI were
191
averaged. The inhibition or facilitation were calculated as a ratio of the conditioned to
192
unconditioned (test pulse alone) MEP amplitude for each subject. Ratios less than one
193
indicate inhibition, and ratios greater than one indicate facilitation. Values are expressed as
194
mean ± standard deviation (SD).
195
Because iSP may be small and their determination may be subjective, the occurrence,
196
onset latencies, areas and durations of iSP were analyzed using automated statistical
197
methods to define their presence (Chen et al. 2003b). For each stimulus intensity, surface
198
EMG from the FDI muscle was rectified and averaged. iSP was deemed significant if the
199
post-stimulus EMG fell below the pre-stimulus mean by at least 1 standard deviation for
200
more than 5 ms (25 consecutive data points based on a 5 KHz sampling rate). iSP onset
201
was defined as last crossing of the mean baseline EMG level and iSP offset the first
202
crossing of the mean baseline EMG level. iSP area was calculated between the iSP onset
203
and offset. The iSP duration was the time between the onset and offset values. The details
204
of the method were described in a previous report (Chen et al. 2003b).
205
Statistical analysis
206
The effects of different stimulus intensities on contralateral MEP, iSP area, iSP
207
duration and IHI were evaluated by two-way repeated measures analysis of variance
13
208
(ANOVA). For contralateral MEP, iSP area and iSP duration, we performed two-way
209
repeated measures ANOVA with side (lesioned/non-dominant vs. non-lesioned/dominant)
210
and stimulus intensity (50%, 75% and 100%) as within subject factors and group (patients
211
and normal subjects) as between subject factor. For IHI, we performed two-way repeated
212
measures ANOVA with side (lesioned/non-dominant vs. non-lesioned/dominant) and ISI
213
(6, 8, 10, 20 and 50 ms) as within subject factors and group (patients and normal subjects)
214
as between subject factor. The effects of lesion compared to no lesion in each callosal
215
segment on iSP area, iSP duration and LIHI (ISI of 50 ms) were analyzed with the unpaired
216
t-test. The effects of lesion in each callosal segment on SIHI (ISI of 8 & 10 ms) were
217
analyzed with repeated measures ANOVA with lesion (presence/absence) as the between
218
subject factor and interstimulus interval as the within subject factor. Fisher’s Protected
219
Least Significant Difference (PLSD) test was used for post hoc testing. Linear regression
220
and multiple regression analyses were used to evaluate the relationship between callosal
221
lesion burden for the anterior half and posterior half of the CC and measures of
222
transcallosal inhibition. For iSP area and iSP duration, the results for stimulation at 100%
223
stimulator output were used for the correlation.
14
224
RESULTS
225
Magnetic Resonance Imaging
226
MRI revealed isolated infarction of CC in 7 patients and CC infarction associated with
227
infarctions in other locations (parasagittal frontal regions, cingulate gyrus or anterior
228
portions of basal ganglia) in 19 patients. The infarcts were in the dominant hemisphere in
229
18 patients and in the non-dominant hemisphere in 8 patients. Most patients had more than
230
one segment of the CC affected and the affected segments in each patient were shown in
231
Table 1. Infarction in the segment 1 was seen in 18 patients, segment 2 in 19, segment 3 in
232
19, segment 4 in 14 and segment 5 in 4 patients (Table 1).
233
Contralateral motor evoked potentials
234
The contralateral MEP amplitude increased with higher stimulus intensity (F(2,
235
35)=45.13, p<0.0001) and the values tended to be lower for the lesioned (2.30±2.85 mV for
236
50%, 3.44±2.55 mV for 75% and 3.86±2.16 mV for 100% of stimulator output) and
237
non-lesioned sides (1.99±2.51 mV for 50%, 3.51±2.75 mV for 75% and 4.16±2.82 mV for
238
100%) in patients and the non-dominant side (1.74±2.73 mV for 50%, 3.30±2.06 mV for
239
75% and 3.95±1.83 mV for 100%) in normal subjects compared to the dominant side
240
(1.99±2.58 mV for 50%, 4.23±2.15 mV for 75% and 4.88±2.24 mV for 100%) in normal
241
subjects, but there was no significant effect of side (lesioned/non-dominant vs.
15
242
non-lesioned/dominant) or group (patients vs. normal subjects) on MEP amplitude.
243
Ipsilateral silent period
244
iSP was detected in all patients and normal subjects. The results for iSP area are
245
shown in Fig. 2A & B. Two-way repeated measures ANOVA showed a significant effects
246
of group (F(1, 37)=11.74, p =0.0015, reduced iSP area in patients, mean difference=-2.87
247
mV*ms) and stimulus intensity (F(2, 37)=41.29, p<0.0001) on iSP area but the effect of
248
side (lesioned/non-dominant vs. non-lesioned/dominant) was not significant. iSP area was
249
significantly lower for stimulus intensity of 50% compared to 75% (p<0.0001, mean
250
difference=-1.66 mV*ms) and 100% (p<0.0001, mean difference=-3.00 mV*ms)
251
stimulator output. The results for iSP duration are shown in Fig. 2A & C. Two-way
252
repeated measures ANOVA showed a significant effects of group (F(1,37)=12.59,
253
p=0.0011, lower iSP area in patients, mean difference=-22.14 ms) and stimulus intensity
254
(F(2, 37)=41.87, p<0.0001) on iSP duration and there was no effect of side
255
(lesioned/non-dominant vs. non-lesioned/dominant). iSP duration was significantly lower
256
for 50% compared to 75% (p<0.0001, mean difference=-16.37 ms) and 100% (p<0.0001,
257
mean difference=-28.14 ms) stimulator output. These findings showed that iSP was
258
markedly reduced in patients compared to controls.
259
Interhemispheric inhibition (IHI)
16
260
The results are shown in Fig. 3. Two-way repeated measures ANOVA showed a
261
significant effect of group (F(1, 36)=25.63, p<0.0001, reduced in patients, mean
262
difference=0.32) and ISI (F(4, 36)=3.90, p=0.0049) on IHI with no significant effect of
263
side (lesioned/non-dominant vs. non-lesioned/dominant).
264
Relationship between transcallosal inhibition and locations of callosal
265
lesions
266
Unpaired t-test showed that patients with lesions in segment 3 had significantly
267
smaller iSP area (t(48)=4.35, p<0.0001, difference between groups=4.14±1.31) and
268
duration (t(48)=3.93, p=0.0003, difference between groups=34.73±9.98) than patients
269
without lesion in this segment. However, iSP area and iSP duration were similar for
270
patients with and without lesions involving segment 1, 2, 4 or 5. For SIHI and LIHI, there
271
was no significant effect of the presence of lesion in all segments of the CC.
272
Linear regression showed that iSP area (p=0.0073, r =0.38, Fig. 4B) and iSP duration
273
(p=0.010, r =0.36, Fig. 4D) negatively correlated with lesion burden in the posterior but not
274
in the anterior half of the CC (Fig. 4A&C). SIHI also correlated with lesion burden in the
275
posterior (p=0.0004, r=0.35) (Fig. 4F) but not in the anterior half of the CC (Fig. 4E). To
276
further analyze the effects of lesion locations within the posterior half of the CC which
277
consists of segments 3, 4 and 5, we performed multiple regression analysis with lesion
17
278
burdens in segments 3, 4 and 5 as three independent variables. Both iSP area and iSP
279
duration significantly correlated with presence of lesion in segment 3 (p=0.0005 and 0.002,
280
respectively) but not with the presence of lesion in segments 4 or 5. However, no
281
significant correlation was found between SIHI and the presence of lesions in segments 3,
282
4 or 5. There was no correlation between lesion burden and LIHI.
283
SIHI correlated with iSP area (p=0.0001, r=0.49) and iSP duration (p=0.0002, r=0.48).
284
LIHI correlated with iSP area (p=0.041, r=0.28) and showed a trend of correlation with iSP
285
duration (p=0.053).
18
286
DISCUSSION
287
Ipsilateral silent period
288
iSP and IHI in patients with callosal infarction have not been previously reported. In
289
patients with partial agenesis (Meyer et al. 1995) or with circumscribed surgical lesions in
290
different parts of the CC (Meyer et al. 1998), iSP was reduced or absent, suggesting it is
291
mediated through a transcallosal pathway. iSP was also found to be delayed or prolonged
292
in neurological disorders such as amyotrophic lateral sclerosis (Wittstock et al. 2007),
293
multiple sclerosis (Schmierer et al. 2000) and writer’s cramp (Niehaus et al. 2001).
294
Patients with corticobasal degeneration or progressive supranuclear palsy had either absent
295
iSP or reduced iSP duration (Wolters et al. 2004). The marked reduction in iSP we
296
observed in patients with callosal infarction further indicates that the iSP is mediated
297
through the corpus callosum.
298
Interhemispheric inhibition
299
IHI is thought to be mediated by excitatory transcallosal fibers that originate from the
300
hand area of the conditioning M1 and project onto local inhibitory interneurons in the
301
homologous area of the contralateral hemisphere (Di Lazzaro et al. 1999; Ferbert et al.
302
1992b). Reduced IHI has been reported in several neurological and psychiatric conditions
303
including schizophrenia (Daskalakis et al. 2002a), corticobasal degeneration (Pal et al.
19
304
2008; Trompetto et al. 2003) and writer’s cramp (Nelson et al. 2010). IHI may help to
305
maintain hemispheric dominance in cognitive and motor tasks by suppressing undesired
306
activities of the opposite hemisphere. An inverse correlation between SIHI and
307
physiological mirror activities during unimanual phasic movements has been reported
308
(Hubers et al. 2008). Moreover, the neural mechanisms underlying iSP may also be
309
involved in the lateralization of voluntary movements (Giovannelli et al. 2009). Reduced
310
LIHI and its effect on intracortical inhibitory circuits has also been reported in Parkinson
311
disease patients with mirror movements (Li et al. 2007).
312
Several previous studies suggested that IHI is due to transcallosal inhibition
313
(Boroojerdi et al. 1996; Daskalakis et al. 2002b; Di Lazzaro et al. 1999; Ferbert et al. 1992b)
314
based on indirect evidence using transcranial electrical stimulation (Ferbert et al. 1992b)
315
and recordings of TMS-evoked descending corticospinal waves in patients with cervical
316
epidural electrodes (Di Lazzaro et al. 1999). They tested SIHI but LIHI was not examined.
317
One report suggested that SIHI may partly be mediated by subcortical circuits (Gerloff et al.
318
1998). We found marked impairment of SIHI and LIHI in patients with callosal infarctions
319
in the absence of significant impairment in corticospinal projection as measured by MEP
320
amplitudes and recruitment curves. These results provide direct evidence that SIHI and
321
LIHI are measures of transcallosal inhibition.
20
322
Reduced transcallosal inhibition and lesion locations in the CC
323
The CC is usually divided into the rostrum, genu, body, isthmus and splenium. Since
324
there is no clear anatomical landmarks to delineate distinct functional sub-regions, several
325
approaches have been used for CC segmentation (Wahl and Ziemann 2008). Most studies
326
rely on Witelson’s scheme (Witelson 1989), which defines five vertical callosal segments
327
based on arithmetic fractions of the anterior-posterior extent. According to studies in
328
monkeys (Pandya et al. 1971) and postmortem studies in humans (Witelson 1989), the
329
callosal fibers connecting the M1 are located in the anterior midbody, defined as the
330
anterior half minus the anterior one-third of the CC. The posterior midbody, defined as the
331
posterior half minus the posterior third region, contains fibers connecting the
332
somatosensory cortex and the posterior parietal area of the two hemispheres. However,
333
recent studies suggested that the human callosal motor fibers are found in the posterior
334
body of the CC. Meyer et al. studied the iSP in one split-brain patient and 13 patients with
335
circumscribed surgical lesions in different parts of the CC (Meyer et al. 1998). They
336
suggested the fibers mediating transcallosal inhibition (iSP) predominantly pass through
337
the posterior half of the trunk of CC. In line with this, DTI tractography studies found that
338
human callosal motor fibers cross the CC at more posterior location than previously
339
indicated (Fling et al. 2011; Hofer and Frahm 2006; Wahl et al. 2007; Zarei et al. 2006).
21
340
Because of the discrepancies between the DTI-based topology in healthy human CC and
341
Witelson’s classification based on post-mortem studies (Witelson 1989), Hofer and Frahm
342
(Hofer and Frahm 2006) suggested subdividing the CC into five segments similar to
343
Witelson’s scheme but with modified proportions of the segments and a different scheme
344
of fiber attribution. For example, callosal motor fibers cross in region 3 in Hofer and
345
Frahm’s scheme instead of region 2 in Witelson’s scheme. We applied Hofer and Frahm’s
346
scheme and found reduced iSP area and duration was mainly due to lesion in segment 3.
347
Thus, our findings suggest that the callosal motor fibers cross the CC through posterior half
348
of the CC, mainly in segment 3 (posterior midbody). This is in agreement with studies of
349
patients with surgical callosal lesion (Meyer et al. 1998) and DTI tractography studies
350
(Hofer and Frahm 2006; Wahl et al. 2007; Zarei et al. 2006). A recent DTI study provided
351
coordinates in a mid-sagittal callosal atlas in normalized Montreal Neurological Institute
352
(MNI) space (Fling et al. 2011). Future studies examining callosal lesions mapped to MNI
353
space may provide further information on the locations of sensorimotor fibers within the
354
CC.
355
In our study, the lesion burden of the posterior half of the CC also correlated with SIHI
356
but not with LIHI. Previous studies suggest that iSP and LIHI may be mediated by
357
overlapping circuits (Chen et al. 2003a). Different neuronal population in ipsilateral M1
22
358
may mediate iSP and IHI, perhaps through different sets of callosal fibers. The lack of
359
correlation between lesion burden of the posterior half of the CC and LIHI may be due to
360
involvement of different callosal fibers or because most patients have mixed lesions of
361
anterior and posterior half of the CC. Studies with larger number of patients and patients
362
with more discrete lesions of the CC will be needed to demonstrate the callosal locations
363
that mediate iSP, SIHI and LIHI.
364
There are limitations to this study. First, most patients have lesions affecting multiple
365
segments rather than lesions restricted to one segment. Second, very few patients had
366
lesions involving segment 5. These factors affect our ability to precisely correlate lesion
367
location and measures of transcallosal inhibition. Moreover, Fig. 4 shows that some
368
patients with no lesion in the posterior half of the CC also had little or no iSP. One possible
369
explanation is areas other than segment 3 may also mediate transcallosal inhibition. For
370
example, Meyer et al. reported absent iSP in patients with abnormalities in the anterior half
371
of the CC (Meyer et al. 1995). In a subsequent study in patients with surgical lesions of the
372
CC, Meyer et al. found reduced iSP predominately in the posterior trunk of the CC, but
373
they also reported abnormally weak iSP in one patient with a lesion in the anterior trunk of
374
the CC (second segment) (Meyer et al. 1998). Therefore, there may be individual
375
variations in the CC segment that mediates iSP, SIHI and LIHI. Other possible
23
376
explanations include areas other than segment 3 may also mediate transcallosal inhibition,
377
some lesions in segment 3 might not be adequately visualized on MRI and infarction of
378
less than half of segment 3 which would be scored as no lesion was sufficient to disrupt
379
transcallosal inhibition. A larger study sample, especially involving patients with more
380
restricted lesions of the CC, will be needed for further evaluation.
381
The present study shows that cerebral infarction of the CC led to reduced transcallosal
382
inhibition as measured by iSP, SIHI and LIHI. Fibers mediating transcallosal inhibition
383
mainly pass through posterior midbody of the CC.
384
385
Grants
386
This work was supported by a grant VGHKS99-033 from Kaohsiung Veterans General
387
Hospital, Taiwan (R.O.C.) to J.-Y. Li.
388
389
390
391
392
393
394
395
396
397
Disclosure
Dr. J-Y Li reports no disclosures.
Dr. P-H Lai reports no disclosures.
Dr. R. Chen reports no relevant disclosures.
24
398
FIGURE LEGENDS
399
400
Figure 1. Illustration of the scheme for localization of lesions in the corpus callosum.
401
The sagittal diffusion-weighted MRI shows acute infarction of the corpus callosum (A) and
402
the lesions were located at segments 1 and 2 on sagittal T2-weighted imaging according to
403
segmentation scheme by Hofer.
404
405
Figure 2. Findings for ipsilateral silent period (iSP). (A) Examples of iSP in a normal
406
subject and on the lesioned side and the non-lesioned side in a patient. Recordings were
407
from the first dorsal interosseous muscle with 50% maximum voluntary contraction and
408
EMG was rectified and averaged from 10 trials. The M1 was stimulated at 100 ms at 100%
409
of stimulator output. (B & C) Effects of stimulus intensities on iSP area and iSP duration.
410
There were significant differences between patients and normal subjects. Significant
411
differences are shown by asterisks. Error bars represent standard errors.
412
413
Figure 3. Interhemispheric inhibition in patients and normal subjects. The conditioning
414
stimuli were applied to the M1 at various ISIs before the test stimuli to the opposite M1.
415
MEP amplitudes from the FDI muscle contralateral to the test stimuli were measured.
25
416
Ratios < 1 represent inhibition and ratios > 1 represent facilitation. Error bars represent
417
standard errors. There was less IHI in patients compared to controls.
418
419
Figure 4. Correlation between transcallosal inhibition and lesion burden of the anterior
420
and posterior half of the corpus callosum. No correlation was found between lesion burden
421
of the anterior half of the corpus callosum and iSP area (A), iSP duration (C) and SIHI (E).
422
There were significant negative correlations between lesion burden of the posterior half of
423
the corpus callosum and iSP area (B), iSP duration (D) and SIHI (F).
424
26
425
426
REFERENCES
427
Abe O, Masutani Y, Aoki S, Yamasue H, Yamada H, Kasai K, Mori H, Hayashi
428
N, Masumoto T and Ohtomo K. Topography of the human corpus callosum using
429
diffusion tensor tractography. J Comput Assist Tomogr 28: 533-539, 2004.
430
Aboitiz F, Carrasco X, Schroter C, Zaidel D, Zaidel E and Lavados M. The alien
431
hand syndrome: classification of forms reported and discussion of a new condition.
432
Neurol Sci 24: 252-257, 2003.
433
Boroojerdi B, Diefenbach K and Ferbert A. Transcallosal inhibition in cortical and
434
subcortical cerebral vascular lesions. J Neurol Sci 144: 160-170, 1996.
435
Caille S, Sauerwein HC, Schiavetto A, Villemure JG and Lassonde M. Sensory
436
and motor interhemispheric integration after section of different portions of the
437
anterior corpus callosum in nonepileptic patients. Neurosurgery 57: 50-59, 2005.
438
Chen R, Yung D and Li JY. Organization of ipsilateral excitatory and inhibitory
439
pathways in the human motor cortex. J Neurophysiol 89: 1256-1264, 2003a.
440
Chen R, Yung D and Li JY. Organization of ipsilateral excitatory and inhibitory
441
pathways in the human motor cortex. J Neurophysiol 89: 1256-1264, 2003b.
27
442
Daskalakis ZJ, Christensen BK, Chen R, Fitzgerald PB, Zipursky RB and
443
Kapur S. Evidence for impaired cortical inhibition in schizophrenia using
444
transcranial magnetic stimulation. Arch Gen Psychiatry 59: 347-354, 2002a.
445
Daskalakis ZJ, Christensen BK, Fitzgerald PB, Roshan L and Chen R. The
446
mechanisms of interhemispheric inhibition in the human motor cortex. J Physiol 543:
447
317-326, 2002b.
448
de Lacoste MC, Kirkpatrick JB and Ross ED. Topography of the human corpus
449
callosum. J Neuropathol Exp Neurol 44: 578-591, 1985.
450
Di Lazzaro V, Oliviero A, Profice P, Insola A, Mazzone P, Tonali P and Rothwell
451
JC. Direct demonstration of interhemispheric inhibition of the human motor cortex
452
produced by transcranial magnetic stimulation. Exp Brain Res 124: 520-524, 1999.
453
Feinberg TE, Schindler RJ, Flanagan NG and Haber LD. Two alien hand
454
syndromes. Neurology 42: 19-24, 1992.
455
Ferbert A, Priori A, Rothwell JC, Day BL, Colebatch JG and Marsden CD.
456
Interhemispheric inhibition of the human motor cortex. J Physiol 453: 525-546,
457
1992a.
28
458
Ferbert A, Priori A, Rothwell JC, Day BL, Colebatch JG and Marsden CD.
459
Interhemispheric inhibition of the human motor cortex. J Physiol 453: 525-546,
460
1992b.
461
Fling BW, Benson BL and Seidler RD. Transcallosal sensorimotor fiber tract
462
structure-function relationships. Hum Brain Mapp 2011.
463
Gerloff C, Cohen LG, Floeter MK, Chen R, Corwell B and Hallett M. Inhibitory
464
influence of the ipsilateral motor cortex on responses to stimulation of the human
465
cortex and pyramidal tract. J Physiol 510: 249-259, 1998.
466
Giovannelli F, Borgheresi A, Balestrieri F, Zaccara G, Viggiano MP, Cincotta M
467
and Ziemann U. Modulation of interhemispheric inhibition by volitional motor
468
activity: an ipsilateral silent period study. J Physiol 587: 5393-5410, 2009.
469
Giroud M and Dumas R. Clinical and topographical range of callosal infarction: a
470
clinical and radiological correlation study. J Neurol Neurosurg Psychiatry 59:
471
238-242, 1995.
472
Hofer S and Frahm J. Topography of the human corpus callosum
473
revisited--comprehensive fiber tractography using diffusion tensor magnetic
29
474
resonance imaging. Neuroimage 32: 989-994, 2006.
475
Hubers A, Orekhov Y and Ziemann U. Interhemispheric motor inhibition: its role
476
in controlling electromyographic mirror activity. Eur J Neurosci 28: 364-371, 2008.
477
Irlbacher K, Brocke J, Mechow JV and Brandt SA. Effects of GABA(A) and
478
GABA(B) agonists on interhemispheric inhibition in man. Clin Neurophysiol 118:
479
308-316, 2007.
480
Jeeves MA, Silver PH and Jacobson I. Bimanual co-ordination in callosal agenesis
481
and partial commissurotomy. Neuropsychologia 26: 833-850, 1988.
482
Li JY, Espay AJ, Gunraj CA, Pal PK, Cunic DI, Lang AE and Chen R.
483
Interhemispheric and ipsilateral connections in Parkinson's disease: relation to mirror
484
movements. Mov Disord 22: 813-821, 2007.
485
Meyer BU, Roricht S, Grafin von EH, Kruggel F and Weindl A. Inhibitory and
486
excitatory interhemispheric transfers between motor cortical areas in normal humans
487
and patients with abnormalities of the corpus callosum. Brain 118: 429-440, 1995.
488
Meyer BU, Roricht S, Schmierer K, Irlbacher K, Meierkord H, Niehaus L and
489
Grosse P. First diagnostic applications of transcallosal inhibition in diseases
30
490
affecting callosal neurones (multiple sclerosis, hydrocephalus, Huntington's disease).
491
Electroencephalogr Clin Neurophysiol Suppl 51: 233-242, 1999.
492
Meyer BU, Roricht S and Woiciechowsky C. Topography of fibers in the human
493
corpus callosum mediating interhemispheric inhibition between the motor cortices.
494
Ann Neurol 43: 360-369, 1998.
495
Nelson AJ, Hoque T, Gunraj C, Ni Z and Chen R. Impaired interhemispheric
496
inhibition in writer's cramp. Neurology 75: 441-447, 2010.
497
Ni Z, Gunraj C, Nelson AJ, Yeh IJ, Castillo G, Hoque T and Chen R. Two phases
498
of interhemispheric inhibition between motor related cortical areas and the primary
499
motor cortex in human. Cereb Cortex 19: 1654-1665, 2009.
500
Niehaus L, von Alt-Stutterheim K, Roricht S and Meyer BU. Abnormal
501
postexcitatory and interhemispheric motor cortex inhibition in writer's cramp. J
502
Neurol 248: 51-56, 2001.
503
Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory.
504
Neuropsychologia 9: 97-113, 1971.
505
Pal PK, Gunraj CA, Li JY, Lang AE and Chen R. Reduced intracortical and
31
506
interhemispheric inhibitions in corticobasal syndrome. J Clin Neurophysiol 25:
507
304-312, 2008.
508
Pandya DN, Karol EA and Heilbronn D. The topographical distribution of
509
interhemispheric projections in the corpus callosum of the rhesus monkey. Brain Res
510
32: 31-43, 1971.
511
Schmierer K, Niehaus L, Roricht S and Meyer BU. Conduction deficits of callosal
512
fibres in early multiple sclerosis. J Neurol Neurosurg Psychiatry 68: 633-638, 2000.
513
Trompetto C, Buccolieri A, Marchese R, Marinelli L, Michelozzi G and
514
Abbruzzese G. Impairment of transcallosal inhibition in patients with corticobasal
515
degeneration. Clin Neurophysiol 114: 2181-2187, 2003.
516
van Valkenburg C.T. Experimental and pathologico-anatomical researches on the
517
corpus callosum. Brain 36: 119-165, 1913.
518
Wahl M, Lauterbach-Soon B, Hattingen E, Jung P, Singer O, Volz S, Klein JC,
519
Steinmetz H and Ziemann U. Human motor corpus callosum: topography,
520
somatotopy, and link between microstructure and function. J Neurosci 27:
521
12132-12138, 2007.
32
522
Wahl M and Ziemann U. The human motor corpus callosum. Rev Neurosci 19:
523
451-466, 2008.
524
Watson RT and Heilman KM. Callosal apraxia. Brain 106: 391-403, 1983.
525
Witelson SF. Hand and sex differences in the isthmus and genu of the human corpus
526
callosum. A postmortem morphological study. Brain 112: 799-835, 1989.
527
Wittstock M, Wolters A and Benecke R. Transcallosal inhibition in amyotrophic
528
lateral sclerosis. Clin Neurophysiol 118: 301-307, 2007.
529
Wolters A, Classen J, Kunesch E, Grossmann A and Benecke R. Measurements
530
of transcallosally mediated cortical inhibition for differentiating parkinsonian
531
syndromes. Mov Disord 19: 518-528, 2004.
532
Yamadori A, Osumi Y, Ikeda H and Kanazawa Y. Left unilateral agraphia and
533
tactile anomia. Disturbances seen after occulusion of the anterior cerebral artery.
534
Arch Neurol 37: 88-91, 1980.
535
Zarei M, Johansen-Berg H, Smith S, Ciccarelli O, Thompson AJ and Matthews
536
PM. Functional anatomy of interhemispheric cortical connections in the human brain.
537
J Anat 209: 311-320, 2006.
1
Table 1. Clinical and MRI features of patients with callosal infarction
Patient
no.
Age
Sex
Duration
Lesion location
(segments)
1
69
M
21D
1
2
58
M
18D
1
3
50
M
12D
1
4
47
M
7D
1,2
5
71
F
5D
1,2
6
69
M
10D
1,2
7
64
F
8D
1,2,3
8
81
M
17D
1,2,3
9
78
M
15D
1,2,3
10
80
M
9.1M
2,3
11
71
M
28D
2,3
12
63
F
8D
2,3,4
13
48
M
7D
2,3,4
14
73
F
18D
1,2,3,4
15
72
M
13D
1,2,3,4
16
64
M
23D
1,2,3,4
17
39
M
7D
1,2,3,4
18
67
F
17D
1,2,3,4
19
55
F
5D
1,2,3,4
20
56
M
19D
1,2,3,4
21
51
M
2.5M
3
22
74
M
22.7M
3,4
23
50
M
6M
4,5
24
61
M
87.3M
3,4,5
25
64
M
85.5M
1,2,3,4,5
26
59
F
11D
1,2,3,4,5
D: days, M: months
MRI features
Figure 1
A
B
1 2 34 5
Figure2
A
B
Lesioned side
Non-lesioned side
Normal, non-dominant side
Normal, dominant side
0.5
iSP area (mV • ms
s)
MEP Amplitude (m
mV)
Normal subject
TMS artifact
0.4
Mean Pre-stimulus
EMG Level
0.3
Ipsilateral
silent period
0.2
0.1
0.0
0
50
100
150
200
250
300
5
4
3
2
1
0
*
*
*
50
75
100
50
75
100
Stimulus intensity (% of stimulator output)
80
0.5
0.4
0.3
0.2
0.1
60
30
*
*
50
40
*
*
70
iSP
P duration (ms)
MEP
P Amplitude (mV)
*
C
Lesioned side
*
*
20
10
0.0
0
50
100
150
200
250
300
250
300
Non-lesioned side
0.5
Ipsilateral
silent
period
0.4
0.3
0.2
0.1
0
50
100
150
200
Time (ms)
0
50
75
100
50
75
100
Stimulus intensity (% of stimulator output)
Time (ms)
MEP Am
mplitude (mV)
*
7
6
Time (ms)
0.0
*
9
8
Figure3
From non-lesioned to lesioned side
From lesioned to non-lesioned side
Normal, from dominant to non-dominant
Normal, from non-dominant to dominant
de (ratio to conttrol)
MEP Amplitud
1.2
1.0
*
*
*
*
*
*
*
8
10
*
*
*
0.8
0.6
0.4
0.2
0
6
8
10
20
50
6
Interstimulus Interval (ms)
20
50
Figure 4
A
B
14
iSP area (mV • ms)
iSP area (mV • ms)
16
12
10
8
6
4
2
0
0
1
2
3
4
5
6
16
14
12
10
8
6
4
2
0
7
0
1
2
3
0
1
2
3
4
5
6
7
Lesion burden in the posterior half
Lesion burden in the anterior half
D
140
140
120
120
duration (ms)
iSP d
duration (ms)
iSP d
C
100
80
60
40
20
0
0
1
2
3
4
5
6
100
80
60
40
20
0
7
Lesion burden in the anterior half
4
5
6
7
Lesion burden in the posterior half
F
MEP Amplitude (ratio to control)
MEP Amplitude (ratiio to control)
E
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
1
2
3
4
5
6
7
Lesion burden in the anterior half
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
1
2
3
4
5
6
7
Lesion burden in the p
posterior half