Abstract

Studies in non-human primates have shown that medial premotor projections to the striatum are characterized as a set of distinct circuits conveying different type of information. This study assesses the anatomical projections from the supplementary motor area (SMA), pre-SMA and motor cortex (MC) to the human striatum using diffusion tensor imaging (DTI) axonal tracking. Eight right-handed volunteers were studied at 1.5 T using DTI axonal tracking. A connectivity matrix was computed, which tested for connections between cortical areas (MC, SMA and pre-SMA) and subcortical areas (posterior, middle and anterior putamen and the head of the caudate nucleus) in each hemisphere. Pre-SMA projections to the striatum were located rostral to SMA projections to the striatum. The SMA and the MC were similarly connected to the posterior and middle putamen and not to the anterior striatum. These data show that the MC and SMA have connections with similar parts of the sensorimotor compartment of the human striatum, whereas the pre-SMA sends connections to more rostral parts of the striatum, including the associative compartment.

Introduction

Over the past 15 years, anatomical studies in animals have shown that cortical areas project to the striatum as multiple discrete circuits (Alexander et al., 1986), conveying either sensorimotor (skeletomotor and oculomotor circuits), association (dorsolateral prefrontal and lateral orbitofrontal circuit) or limbic (anterior cingulate and medial orbitofrontal circuit) information (Alexander et al., 1986). In monkeys, the sensorimotor compartment of the striatum corresponds to the postcommissural portion of the putamen (Kunzle, 1975; Selemon and Goldman-Rakic, 1985; Flaherty and Graybiel, 1991; Parent and Hazrati, 1995a,b; Takada et al., 1998). This region receives mainly projections from executive somatotopically organized regions of the cortex including the primary motor cortex (MC) and posterior part of the supplementary motor area (SMA). Thus, homologous body regions of the SMA and MC send widespread and substantially overlapping projections, to portions of the striatum (Inase et al., 1996). In contrast, more rostral regions of the presupplementary motor cortex projects toward more rostral parts of the putamen (Selemon and Goldman-Rakic, 1985; Inase et al., 1999). Thus, the anterior premotor and the motor/posterior premotor compartments of the striatum are differentially organized along the rostro-caudal axis in the putamen of monkeys.

Based on anatomical and physiological evidence in animals, the SMA has been subdivided in two distinct areas, the SMA-proper or (SMA) and the pre-SMA. Anatomically, the limit between the SMA-proper and the pre-SMA is usually defined by the VAC line, a vertical line that intersects the anterior commissure and that is perpendicular to the anterior commissure–posterior commissure plane (Picard and Strick, 1996, 2001; Rizzolatti et al., 1996). The SMA has many similarities with executive motor areas. It receives precentral and postcentral afferents from the primary motor cortex, caudal premotor areas and primary and secondary sensory areas, sends direct corticospinal efferences and projects to the primary motor cortex (Picard and Strick, 1996). The SMA has a somatotopic organization (Picard and Strick, 1996, 2001; Rizzolatti et al., 1996). Comparatively, the pre-SMA is organized as an associative area. It receives its afferents from associative frontal and parietal areas, does not project to motor areas and the spinal cord and is not somatotopically organized.

Invasive tracing studies used in animals are not applicable to human studies. The anatomical cortico-basal ganglia connectivity is largely based on primate-human extrapolations and has not been demonstrated directly until now. Diffusion tensor imaging (DTI) is a new technique that allows demonstration of fiber tracts in vivo in humans (Conturo et al., 1999; Mori et al., 1999; Basser et al., 2000; Poupon et al., 2000). Water diffusion in brain tissue is directionally dependent in the white matter, a property called anisotropy. Anisotropy in the white matter was attributed to the organization of axons and myelin sheaths. Diffusion is higher following the direction of axons and fiber bundles. A previous study has shown that DTI fiber tracking was able to differentiate the connections of the sensorimotor, associative and limbic compartments of the human striatum with the frontal lobes (Lehericy et al., 2004). This study aimed to compare cortico-striatal projections from the pre-SMA, SMA and MC using diffusion tensor imaging (DTI) axonal tracking.

Materials and Methods

Subjects

Eight healthy right-handed volunteers were studied (three males, five females, age range 24–33 years, mean 29 years). The National Ethics Committee approved the study. All subjects gave informed consent.

Imaging

DTI was performed using echo-planar imaging (EPI) at 1.5 T (GE) with standard head coil for signal reception. DTI axial slices were obtained using the following parameters: repetition time = 10 s, echo time = 88 ms, flip angle = 90°, matrix = 128 × 128, field of view = 380 × 380 mm2, slice thickness = 3 mm, no gap (3 mm isotropic voxels), acquisition time = 320 s. Four averages were used with signal averaging in the scanner buffer. Diffusion weighting was performed along six independent directions, with a b-value of 900 s/mm2. A reference image with no diffusion weighting was also obtained. High-resolution 3-D anatomical images were used for display and anatomical localization (110 axial contiguous inversion recovery three dimensional fast SPGR images, 1.5 mm thick, TI = 400 ms, FOV = 240 × 240 mm2, matrix size = 256 × 256).

Image Processing

Raw diffusion-weighted data were corrected for geometric distortion secondary to eddy currents using a registration technique based upon the geometric model of distortions (Haselgrove and Moore, 1996; Mangin et al., 2002). Each diffusion-weighted slice was aligned with the corresponding T2-weighted image of the series. For convenience, the image was in the XY plane of the magnet axes with the phase-encoding direction along Y. The following distortion model was used: (i) residual gradient in the slice-encoding direction Z produced uniform translation along Y; (ii) residual gradient in the frequency-encoding direction X produced a shear parallel to Y (i.e. a translation linearly related to X); and (iii) residual gradient in the phase-encoding direction Y produced a uniform scaling in the Y direction. Hence, the geometric model was written for each column X as Y′ = S.Y + T0 + T1.X, which amounted to a simple slice-dependent affine transformation (S is the scale factor, T0 a global translation and T1 a shear). An additional global multiplicative correction by 1/S was applied to the slice intensities, which was done after estimation of (S, T0, T1). This latest correction stemmed from an energy-conservation-based MR principle. To estimate the affine transformation, which brought a diffusion-weighted slice into spatial alignment with the standard T2-weighted slice, a similarity measure taking into account the complex dependence between intensities was chosen, the mutual information which came from information theory (Maes et al., 1997). The optimal affine transformation (S, T0, T1) corresponded to the maximization of the mutual information between the transformed diffusion-weighted image and the reference T2-weighted image. Diffusion tensors, fractional anisotropy (FA) and fiber tracks were calculated using in-house software. The tracking algorithm employed is based on the method described in Basser et al. (2000). At any position along the track trajectory, a diffusion tensor was interpolated and eigenvectors were computed. The eigenvector associated with the greatest eigenvalue indicated the principal direction of water diffusion. The track was propagated along this direction over a small distance (<0.5 mm) to the next point where a new diffusion tensor was interpolated. Tracking terminated when the angle between two consecutive eigenvectors was >60°, or when the FA value was <0.1, indicating a region of low diffusion anisotropy. A spherical shape marker indicated the fiber termination points. High-resolution anatomical images and DTI images were spatially co-registered to allow for the simultaneous display of tracks and anatomical images. DTI-anatomical images co-registration was performed by matching reproducible landmarks (such as the hand area in the motor cortex, the anterior commissure, the corpus callosum, the borders of the striatum) between the two images.

Anatomical Localization of Tracks

Regions-of-interests (ROIs) were determined using reproducible anatomical landmarks on high-resolution 3-D T1-weighted images (Fig. 1). ROIs were hand drawn by one experimenter (S.L.) on DTI images to avoid mis-registration effects between DTI and T1-weighted images. Three ROIs were defined in the cortex. The motor cortex (MC) included the precentral gyrus (mean ROI volume ± SD: 2422 ± 608 and 2403 ± 693 mm3 in the right and the left hemispheres, respectively). The central sulcus was located using the characteristic shape of the hand motor area or hand knob. This ROI probably included part of the adjacent posterior premotor cortex. The supplementary motor area (SMA) extended from the brain vertex to the cingulate sulcus in the supero-inferior direction and from the precentral sulcus caudally to the VAC line rostrally (mean ROI volume ± SD: 8563 ± 970 and 9084 ± 1116 mm3 in the right and the left hemispheres, respectively). The pre-SMA extended rostral to the VAC line to a virtual line passing through the genu of the corpus callosum parallel to the VAC line (Picard and Strick, 1996, 2001) (mean ROI volume ± SD: 9901 ± 1711 and 10310 ± 1237 mm3 in the right and the left hemispheres, respectively). The volumes of cortical ROIs were different (Friedman test, P < 0.001). The difference was not related to differences in volumes between the two hemispheres (Wilcoxon test). Then, right and left values were averaged. Paired comparisons (Wilcoxon test) showed that preSMA was larger than SMA (P < 0.05) and MC (P < 0.02) and SMA was larger than MC (P < 0.02). Four ROIs were defined in the striatum. These ROIs included the posterior, middle and anterior putamen and the head of the caudate nucleus (Fig. 1). In the supero-inferior direction, ROIs in the putamen extended from the superior border of the structure to the AC–PC plane. The ROI corresponding to the anterior putamen included all putamen rostral to the VAC line (mean ROI volume ± SD: 783 ± 216 and 775 ± 180 mm3 in the right and the left hemispheres, respectively). The putamen caudal to the VAC line was further subdivided in two parts (middle and anterior) by a virtual line parallel to the VAC line, located at mid-distance between the VAC line and the posterior border of the putamen. The ROI corresponding to the middle putamen included the anterior half of the putamen caudal to the VAC line (mean ROI volume ± SD: 771 ± 157 and 737 ± 107 mm3 in the right and the left hemispheres, respectively). The ROI corresponding to the posterior putamen included the posterior half of the putamen caudal to the VAC line (mean ROI volume ± SD: 702 ± 175 and 675 ± 228 mm3 in the right and the left hemispheres, respectively). The ROI in the head of the caudate nucleus was located rostral to the thalamus and was limited ventrally by the ventral part of the internal capsule (mean ROI volume ± SD: 790 ± 88 and 775 ± 114 mm3 in the right and the left hemispheres, respectively). There was no significant difference between ROIs volumes in the striatum (Friedman test, P = 0.58). One ROI in the cortex (left SMA) and one ROI in the striatum (left anterior putamen) were randomly selected to assess intra-rater reliability. These ROIs were traced twice by the same experimenter (S.L.). There was no difference between the two measurements (anterior putamen: 775 ± 180 and 802 ± 165 mm3 for the first and the second measurements, respectively, P = 0.18; SMA: 9084 ± 1116 and 9219 ± 1249 mm3 for the first and the second measurements, respectively, P = 0.34). Mean FA in the subcortical ROIs ranged from 0.20 ± 0.04 in the left posterior putamen to 0.23 ± 0.03 in the left anterior putamen. There was no statistical difference in FA values between ROIs. All ROIs overlapped with adjacent white matter because of partial voluming due to the relatively low resolution of DTI images. Tracking was initiated from all seeding points in each cortical ROI. A connectivity matrix was computed, which tested for the number of tracks connecting each of the cortical ROIs to the subcortical ROIs (Conturo et al., 1999).

Statistical Analysis

Analyses were conducted using statistical software (v. 11.0; SPSS Inc., Chicago, IL). The dependant variable was the number (n) of tracks between each of the three cortical ROIs and each of the four striatal ROIs in both hemispheres, yielding 48 different possible connections (24 possible connections between the three cortical ROIs and the four subcortical ROIs in each hemisphere and 24 possible contralateral connections). All values are expressed as mean ± SD. Statistical comparisons were performed using non-parametric procedures because of repeated violations of the assumption of distribution normality: k-related samples test (Friedman test) and when these results were significant follow-up pairwise comparisons (Wilcoxon test). Correlation analyses were performed using Spearman’s r′ correlation coefficient. The significance level was set at P < 0.05.

Results

Tracks reconstructed from the cortical ROIs are displayed in Figures 2 and 3. Overall, projections from the MC and SMA were mainly directed to the posterior putamen whereas pre-SMA projections were directed to more rostral parts of the striatum (Figures 2 and 3).

A mean of 2107 ± 1124 seeds were placed in each of the cortical ROIs (MC = 669 ± 146 seeds, SMA = 2660 ± 458, pre-SMA = 2992 ± 431, both hemispheres together) yielding 627 ± 139 tracks for the MC, 2247 ± 434 tracks for the SMA and 2504 ± 445 tracks for the pre-SMA. Of all these tracks, 2.6 ± 1.9% were selected by the striatal ROIs (4.8 ± 1.9% for MC tracks, 1.0 ± 0.2% for SMA and 1.4 ± 1.2% pre-SMA tracks). The mean number of tracks linking cortical and striatal ROIs ranged from 0 (between MC and contralateral striatum) to 33.4 ± 24.3 (between right MC and right posterior putamen; Fig. 4). The relationships between ROI volume and number of tracks were tested for each connection. There was no systematic effect of ROI volume on the distribution of striatal projections. MC ROIs were smaller than SMA and pre-SMA ROIs, but the mean number of tracks originating from the three cortical ROIs were not significantly different although the pre-SMA (mean of the two hemispheres = 44 ± 34 tracks) and MC (mean of the two hemispheres = 30 ± 25 tracks) had a larger number of tracks that intersected the striatum than the SMA (mean of the two hemispheres = 22 ± 24 tracks). Overall, there was no significant correlation between ROI volume and track number except between the volume of the pre-SMA and the number of tracks linking the pre-SMA to the contralateral anterior putamen (r′ = 0.96, P < 0.01) and between the volume of the posterior putamen and the number of tracks linking the posterior putamen to the ipsilateral (r′ = 0.86, P < 0.02) and contralateral SMA (r′ = 0.92, P < 0.01). Some tracks terminated in the striatum (when FA values were <0.1), whereas others crossed the striatum to terminate in another structure (such as the mesencephalon in the area of the substantia nigra for the posterior putamen). Thus, the location of tracks in the striatum was determined partly by their location where the white matter pathways approached the putamen (for tracks terminating in the striatum).

Statistical analyses using a ranking procedure were conducted to determine whether the number of tracks between ipsilateral and contralateral cortical (MC, SMA, pre-SMA) and striatal ROIs (posterior, middle and anterior putamen and caudate nucleus) for both hemispheres, were similar (null hypothesis) or not. Friedman test for multiple related-samples was first conducted for all 48 connections to determine whether the number of tracks between regions was equally distributed or not. Friedman test between all connections was significant (P < 0.001). Then, pairwise comparisons using Wilcoxon test were conducted to determine whether a right or left hemispheric dominance could explain the difference in the number of tracks between regions. Wilcoxon tests on each connection showed no difference between right and left hemispheres. Thus, the numbers of tracks were averaged in 12 ipsilateral and 12 contralateral connections. Data are summarized in Table 1 and Figure 4.

To determine the effect of the distribution in ipsilateral and contralateral connections, Friedman test for all 24 ipsilateral and contralateral connections was conducted and was significant (P < 0.001). Pairwise comparisons using Wilcoxon test were conducted to determine which connections were predominantly ipsi- or contralateral. Connections between the MC and the posterior (P < 0.02) and anterior putamen (P < 0.02), between the SMA and the posterior (P < 0.02) and middle putamen (P < 0.02) and between the pre-SMA and the middle putamen (P < 0.03) were more important ipsilaterally than contralaterally (Table 1). There was also a trend between the pre-SMA and anterior putamen (P = 0.05).

To determine whether the number of tracks among ipsi- or contralateral connections were equally distributed or not, Friedman tests were conducted, which were significant both for ipsilateral connections (P < 0.001) and contralateral connections (P < 0.001). Friedman tests on individual ROIs showed significant differences for connections of the MC (P < 0.001), SMA (P < 0.001), pre-SMA (P < 0.01), posterior putamen (P < 0.001), middle putamen (P < 0.001), anterior putamen (P < 0.01) and caudate nucleus (P < 0.01).

Motor Cortex

The MC was only connected to the ipsilateral striatum with unequal distribution of tracks to striatal ROIs (P < 0.001). More tracks were directed to the posterior putamen than to the middle and anterior putamen and the caudate nucleus (P < 0.02 for each comparison).

SMA

The distribution of tracks were unequal among both ipsilateral (P < 0.001) and contralateral connections (P < 0.05). The number of SMA tracks directed to the ipsilateral striatum was larger for the posterior than the middle and anterior putamen and the caudate nucleus (P < 0.02 for each comparison) and larger for the middle than the anterior putamen and the caudate nucleus (P < 0.02 for each comparison). The number of SMA tracks directed to the contralateral striatum was more important for the posterior than the middle and anterior putamen (P < 0.03 for each comparison).

Pre-SMA

Friedman test was significant (P < 0.01) only for ipsilateral connections. The number of pre-SMA tracks directed to the ipsilateral striatum was larger for the middle than the posterior and anterior putamen (P < 0.02 for each comparison) and the caudate nucleus (P < 0.03).

Posterior Putamen

Friedman tests showed differences among the numbers contralateral tracks (P < 0.01) and a trend for ipsilateral tracks (P = 0.05). Ipsilateral connections of the posterior putamen to SMA and MC were similar (P = 0.26). The posterior putamen was more connected with the SMA and MC than with the pre-SMA (P < 0.03 for each comparison). The number of contralateral connections was higher with the pre-SMA than the SMA (P < 0.03).

Middle Putamen

Distribution of tracks were unequal among both ipsilateral (P < 0.01) and contralateral connections (P < 0.01). The middle putamen was more connected to the pre-SMA than to the MC and SMA for both ipsilateral (P < 0.02 for each comparison) and contralateral hemispheres (P < 0.05 for each comparison).

Anterior Putamen

Difference in the distribution of tracks to both ipsilateral (P = 0.09) and contralateral hemispheres (P = 0.06) did not reach significance, thus no further statistical analysis was conducted.

Caudate Nucleus

Distribution of tracks were unequal among both ipsilateral (P < 0.05) and contralateral connections (P < 0.03). The caudate nucleus was more connected to the ipsilateral pre-SMA than SMA (P < 0.05) and MC (P < 0.06) and was more connected to the contralateral pre-SMA than MC (P < 0.05).

For all striatal regions, absence of contralateral connection to the MC yielded significant differences when compared to pre-SMA (P < 0.05 for each comparison). When compared to SMA, a significant difference was only detected for the posterior putamen.

Discussion

Our findings show that the MC and the SMA have connections with similar parts of the posterior putamen, whereas the pre-SMA sends connections to more rostral parts of the striatum, including the associative compartment. This study provides the first demonstration in humans that motor/posterior premotor areas and anterior premotor areas projections to the striatum are organized in distinct circuits along the caudo-rostral axis.

Pre-SMA projections were mainly directed to anterior parts of the striatum extending caudal to the AC in the middle portion of the putamen and almost no track was found in the caudal tier of the putamen. In contrast, SMA projections predominated in the caudal tier of the putamen, fewer tracks were found in the middle portion of the putamen and almost no fiber was found in the anterior striatum. Similarly to SMA projections, the MC projected to the posterior putamen with only few tracks to the middle portion of the putamen. This pattern was highly concordant with monkey studies (Kunzle, 1978; Selemon and Goldman-Rakic, 1985; Parthasarathy et al., 1992; Inase et al., 1996, 1999; Takada et al., 1998). In the macaque monkey, cortico-striatal inputs from pre-SMA were distributed mainly in the striatal cell bridges connecting the rostral aspects of the caudate nucleus and the putamen and in the neighboring striatal areas (Inase et al., 1999). In this study, although SMA and pre-SMA projections were found on the same coronal slices at the level of the AC and on the slices immediately caudal to the AC, they did not overlap (Inase et al., 1999). In the present study, spatial resolution was not high enough to determine whether these tracks are separated in the middle portion of the putamen. In monkeys, MC projections were located in comparable portions of the posterior putamen, distributing from the level of the AC to the most posterior extent of the putamen (Inase et al., 1996). The present study also confirms that SMA and MC tracks end in territories in the striatum that are largely overlapped, as already reported in monkeys (Inase et al., 1996).

Results show that SMA projections to the striatum were bilateral but largely predominated in the ipsilateral hemisphere. Pre-SMA projections to the contralateral striatum were larger than SMA projections to the contralateral striatum. No fiber originating from the MC was directed to the contralateral striatum. Overall, these results are in line with monkey studies. In monkeys, both the SMA and pre-SMA send bilateral projections to the striatum (McGuire et al., 1991a,b; Wiesendanger et al., 1996; Inase et al., 1999), although these projections predominate in the ipsilateral hemisphere (McGuire et al., 1991a,b; Inase et al., 1996, 1999; Wiesendanger et al., 1996). Similarly to the present findings, the pre-SMA had more transcallosal connections than the SMA (Liu et al., 2002). In contrast to the present findings, a small fraction of motor cortical areas projected contralaterally via the corpus callosum although most connections projected to ipsilateral subcortical structures (Flaherty and Graybiel, 1993; Wiesendanger et al., 1996). This difference with animal studies may reflect an anatomical variability in cortico-striatal connections in humans as compared to monkeys or more likely it may result from a failure of DTI fiber tracking to detect crossed tracks coming from the motor cortex. Fiber direction estimated with DTI is the average of all fiber directions within each voxel. When fiber tracts cross, ‘kiss’, kink, merge or diverge within one voxel, then the eigenvector associated with the largest eigenvalue of the diffusion tensor corresponds to the average of these fiber directions in the voxel. The complexity of fiber directions in the frontal lobe may prevent the detection of crossed motor cortex tracks. Increased spatial resolution or improved fiber tracking method may help overcome this limitation.

Overall, there was no correlation between ROI volumes and the number of cortical tracks projecting to striatal ROIs, except that larger pre-SMA volume was associated with a larger number of tracks to the anterior putamen and that larger posterior putamen volume was associated with a larger number of tracks to the SMA. There was no significant difference in the number tracks that intersected the striatum between the three cortical ROIs in line with monkey studies (Inase et al., 1996), although sizes of these ROIs were different. Thus, ROI volume was not a factor explaining the general pattern of cortico-subcortical connectivity, e.g. the predominant connections of the MC and SMA to the posterior putamen and of the pre-SMA to more rostral parts of the putamen.

This anatomic organization of SMA and pre-SMA connections is paralleled by functional differences (Picard and Strick, 1996, 2001; Rizzolatti et al., 1996; Tanji, 1996). In monkeys, the pre-SMA is involved in higher-order aspects of motor control (Picard and Strick, 1996; Tanji, 1996). Motor-related activation occurred mainly in the SMA, whereas neurons with preparatory activity are located more rostrally in the pre-SMA (Alexander and Crutcher, 1990; Romo and Schultz, 1992). Similarly, motor-related activation occurred mainly in the putamen caudal to the anterior commissure, whereas neurons with working-memory and preparatory activity are located more rostrally in the anterior part of the striatum (Alexander and Crutcher, 1990). In humans, functional imaging studies have shown that the pre-SMA has an important role in supramotor activities such as visuo-motor associations, learning, selection, preparation and sequencing of movements (Colebatch et al., 1991; Deiber et al., 1991, 1996; Rao et al., 1993; Humberstone et al., 1997; Hikosaka et al., 1999; Sakai et al., 1999), whereas the SMA along with the motor cortex is more closely related to movement execution (Deiber et al., 1996; Jenkins et al., 2000; Krainik et al., 2001). For example, SMA activation observed during movement preparation was located more rostrally than during movement execution (Deiber et al., 1996; Humberstone et al., 1997; Toni et al., 1999). Similarly, activation in the putamen during preparation was rostral to the one observed during movement execution (Jenkins et al., 2000; Lehericy and Gerardin, 2002). Our data provide anatomical support for the observation that preparatory and executive aspects of movements are conveyed in distinct cortico-striatal circuits.

This study was supported by grants from NIH (NS44825), BTRR P41-RR008079, the Keck Foundation, the Human Frontiers Science Program, PHRC AOR01109 and ACI 2001.

Figure 1. Location of the regions-of-interests (ROIs). ROIs are superimposed on T1-weighted sagittal (upper row left, parasagittal view of the left hemisphere; upper row right, parasagittal view of the right hemisphere), coronal (lower left) and axial (lower middle and right) views of one volunteer. Abbreviations: ant, anterior; CN, caudate nucleus; L, left; MC, motor cortex; post, posterior; Pu, putamen; R, right; SMA, supplementary motor area;

Figure 1. Location of the regions-of-interests (ROIs). ROIs are superimposed on T1-weighted sagittal (upper row left, parasagittal view of the left hemisphere; upper row right, parasagittal view of the right hemisphere), coronal (lower left) and axial (lower middle and right) views of one volunteer. Abbreviations: ant, anterior; CN, caudate nucleus; L, left; MC, motor cortex; post, posterior; Pu, putamen; R, right; SMA, supplementary motor area;

Figure 2. Cortico-striatal tracks. Tracks were calculated from all seeding points in the cortical ROIs in the left hemisphere and superimposed on a multiplanar 3D view of the brain (axial, coronal and parasagittal view of the left hemisphere). Left: comparison of tracks coming from the SMA (green) and pre-SMA (yellow); right: comparison of tracks coming from the SMA (green) and MC (red). Abbreviations are as in Figure 1 legend.

Figure 2. Cortico-striatal tracks. Tracks were calculated from all seeding points in the cortical ROIs in the left hemisphere and superimposed on a multiplanar 3D view of the brain (axial, coronal and parasagittal view of the left hemisphere). Left: comparison of tracks coming from the SMA (green) and pre-SMA (yellow); right: comparison of tracks coming from the SMA (green) and MC (red). Abbreviations are as in Figure 1 legend.

Figure 3. Cortico-striatal tracks in the striatum of individual subjects. Tracking was initiated from the motor cortex (red), SMA (green) and pre-SMA (yellow) in both hemispheres. Tracks are superimposed on axial (left and middle) and coronal (right) T1-weighted images. Dots represent a cross-section through all the striatal tracks at the slice level (only tracks crossing the striatum are displayed). Tracks originating from the motor cortex (red) and SMA (green) were directed to the posterior part of the putamen. Tracks originating from the pre-SMA (yellow) were located rostral to SMA and MC tracks. Abbreviations: GP, globus pallidus; IC, internal capsule; Th, thalamus; other abbreviations are as in Figure 1 legend.

Figure 3. Cortico-striatal tracks in the striatum of individual subjects. Tracking was initiated from the motor cortex (red), SMA (green) and pre-SMA (yellow) in both hemispheres. Tracks are superimposed on axial (left and middle) and coronal (right) T1-weighted images. Dots represent a cross-section through all the striatal tracks at the slice level (only tracks crossing the striatum are displayed). Tracks originating from the motor cortex (red) and SMA (green) were directed to the posterior part of the putamen. Tracks originating from the pre-SMA (yellow) were located rostral to SMA and MC tracks. Abbreviations: GP, globus pallidus; IC, internal capsule; Th, thalamus; other abbreviations are as in Figure 1 legend.

Figure 4. Connections of the pre-SMA, SMA and MC. Bar graphs represent the number of tracks connecting the pre-SMA, SMA and MC with each one of the four striatal regions in the left and the right hemispheres.

Figure 4. Connections of the pre-SMA, SMA and MC. Bar graphs represent the number of tracks connecting the pre-SMA, SMA and MC with each one of the four striatal regions in the left and the right hemispheres.

Table 1


 Number of tracks (±SD) between cortical and striatal ROIs

 Ipsilateral     Contralateral    
 Posterior putamen Middle putamen Anterior putamen Caudate nucleus  Posterior putamen Middle putamen Anterior putamen Caudate nucleus 
Motor cortex 51.5 ± 32.6  5.9 ± 7.3  3.1 ± 5.4  0.4 ± 0.7  
SMA 34.0 ± 25.0 10.3 ± 15.7  0.4 ± 0.5  0.4 ± 1.1  5.9 ± 6.1 1.0 ± 1.8 0.4 ± 0.5 0.9 ± 1.4 
Pre-SMA  7.4 ± 9.4 52.1 ± 40.6 13.0 ± 12.3 15.4 ± 20.3  1.1 ± 1.4 11.6 ± 16.1 2.9 ± 3.6 5.3 ± 8.9 
 Ipsilateral     Contralateral    
 Posterior putamen Middle putamen Anterior putamen Caudate nucleus  Posterior putamen Middle putamen Anterior putamen Caudate nucleus 
Motor cortex 51.5 ± 32.6  5.9 ± 7.3  3.1 ± 5.4  0.4 ± 0.7  
SMA 34.0 ± 25.0 10.3 ± 15.7  0.4 ± 0.5  0.4 ± 1.1  5.9 ± 6.1 1.0 ± 1.8 0.4 ± 0.5 0.9 ± 1.4 
Pre-SMA  7.4 ± 9.4 52.1 ± 40.6 13.0 ± 12.3 15.4 ± 20.3  1.1 ± 1.4 11.6 ± 16.1 2.9 ± 3.6 5.3 ± 8.9 

Values are the mean of both hemispheres as there was no statistical difference between the two hemispheres; SD are across the 16 hemispheres of the eight subjects).

References

Alexander GE, Crutcher MD (
1990
) Preparation for movement: neural representations of intended direction in three motor areas of the monkey.
J Neurophysiol
 
64
:
133
–150.
Alexander GE, DeLong, MR, Strick PL (
1986
) Parallel organization of functionally segregated circuits linking basal ganglia and cortex.
Annu Rev Neurosci
 
9
:
357
–381.
Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A (
2000
) In vivo fiber tractography using DT-MRI data.
Magn Reson Med
 
44
:
625
–632.
Colebatch JG, Deiber MP, Passingham RE, Friston KJ, Frackowiak RS (
1991
) Regional cerebral blood flow during voluntary arm and hand movements in human subjects.
J Neurophysiol
 
65
:
1392
–1401.
Conturo TE, Lori NF, Cull TS, Akbudak E, Snyder AZ, Shimony JS, McKinstry RC, Burton H, Raichle ME (
1999
) Tracking neuronal fiber pathways in the living human brain.
Proc Natl Acad Sci USA
 
96
:
10422
–10427.
Deiber MP, Passingham RE, Colebatch JG, Friston KJ, Nixon PD, Frackowiak RS (
1991
) Cortical areas and the selection of movement: a study with positron emission tomography.
Exp Brain Res
 
84
:
393
–402.
Deiber MP, Ibanez V, Sadato N, Hallett M (
1996
) Cerebral structures participating in motor preparation in humans: a positron emission tomography study.
J Neurophysiol
 
75
:
233
–247.
Flaherty AW, Graybiel AM (
1991
) Corticostriatal transformations in the primate somatosensory system. Projections from physiologically mapped body-part representations.
J Neurophysiol
 
66
:
1249
–1263.
Flaherty AW, Graybiel AM (
1993
) Two input systems for body representations in the primate striatal matrix: experimental evidence in the squirrel monkey.
J Neurosci
 
13
:
1120
–1137.
Haselgrove JC, Moore JR (
1996
) Correction for distortion of echo-planar images used to calculate the apparent diffusion coefficient.
Magn Reson Med
 
36
:
960
–964.
Hikosaka O, Nakahara H, Rand MK, Sakai K, Lu X, Nakamura K, Miyachi S, Doya K (
1999
) Parallel neural networks for learning sequential procedures.
Trends Neurosci
 
22
:
464
–471.
Humberstone M, Sawle GV, Clare S, Hykin J, Coxon R, Bowtell R, Macdonald IA, Morris PG (
1997
) Functional magnetic resonance imaging of single motor events reveals human presupplementary motor area.
Ann Neurol
 
42
:
632
–637.
Inase M, Sakai ST, Tanji J (
1996
) Overlapping corticostriatal projections from the supplementary motor area and the primary motor cortex in the macaque monkey: an anterograde double labeling study.
J Comp Neurol
 
373
:
283
–296.
Inase M, Tokuno H, Nambu A, Akazawa T, Takada M (
1999
) Corticostriatal and corticosubthalamic input zones from the presupplementary motor area in the macaque monkey: comparison with the input zones from the supplementary motor area.
Brain Res
 
833
:
191
–201.
Jenkins IH, Jahanshahi M, Jueptner M, Passingham RE, Brooks DJ (
2000
) Self-initiated versus externally triggered movements. II. The effect of movement predictability on regional cerebral blood flow.
Brain
 
123
:
1216
–1228.
Krainik A, Lehericy S, Duffau H, Vlaicu M, Poupon F, Capelle L, Cornu P, Clemenceau S, Sahel M, Valery CA, et al. (
2001
) Role of the supplementary motor area in motor deficit following medial frontal lobe surgery.
Neurology
 
57
:
871
–878.
Kunzle H (
1975
) Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in Macaca fascicularis.
Brain Res
 
88
:
195
–209.
Kunzle H (
1978
) An autoradiographic analysis of the efferent connections from premotor and adjacent prefrontal regions (areas 6 and 9) in Macaca fascicularis.
Brain Behav Evol
 
15
:
185
–234.
Lehericy S, Gerardin E (
2002
) Normal functional imaging of the basal ganglia.
Epileptic Disord
 
4
:S
23
–30.
Lehericy S, Ducros M, van de Moortele PF, Francois C, Thivard L, Poupon C, Swindale N, Ugurbil K, Kim DS (
2004
) Diffusion tensor fiber tracking shows distinct corticostriatal circuits in humans.
Ann Neurol
 
55
:
522
–529.
Liu J, Morel A, Wannier T, Rouiller EM (
2002
) Origins of callosal projections to the supplementary motor area (SMA): a direct comparison between pre-SMA and SMA-proper in macaque monkeys.
J Comp Neurol
 
443
:
71
–85.
McGuire PK, Bates JF, Goldman-Rakic PS (
1991
) Interhemispheric integration: I. Symmetry and convergence of the corticocortical connections of the left and the right principal sulcus (PS) and the left and the right supplementary motor area (SMA) in the rhesus monkey.
Cereb Cortex
 
1
:
390
–407.
McGuire PK, Bates JF, Goldman-Rakic PS (
1991
) Interhemispheric integration: II. Symmetry and convergence of the corticostriatal projections of the left and the right principal sulcus (PS) and the left and the right supplementary motor area (SMA) of the rhesus monkey.
Cereb Cortex
 
1
:
408
–417.
Maes F, Collignon A, Vandermeulen D, Marchal G, Suetens P (
1997
) Multimodality image registration by maximization of mutual information.
IEEE Trans Med Imaging
 
16
:
187
–198.
Mangin JF, Poupon C, Clark C, Le Bihan D, Bloch I (
2002
) Distortion correction and robust tensor estimation for MR diffusion imaging.
Med Image Anal
 
6
:
191
–198.
Mori S, Crain BJ, Chacko VP, van Zijl PC (
1999
) Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging.
Ann Neurol
 
45
:
265
–269.
Parent A, Hazrati LN (
1995
) Functional anatomy of the basal ganglia. I. The cortico-basal ganglia–thalamo-cortical loop.
Brain Res
 
Brain Res Rev
 
20
:
91
–127.
Parent A, Hazrati LN (
1995
) Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external pallidum in basal ganglia circuitry.
Brain Res
 
Brain Res Rev
 
20
:
128
–154.
Parthasarathy HB, Schall JD, Graybiel AM (
1992
) Distributed but convergent ordering of corticostriatal projections: analysis of the frontal eye field and the supplementary eye field in the macaque monkey.
J Neurosci
 
12
:
4468
–4488.
Picard N, Strick PL (
1996
) Motor areas of the medial wall: a review of their location and functional activation.
Cereb Cortex
 
6
:
342
–353.
Picard N, Strick PL (
2001
) Imaging the premotor areas.
Curr Opin Neurobiol
 
11
:
663
–672.
Poupon C, Clark CA, Frouin V, Regis J, Bloch I, Le Bihan D, Mangin J (
2000
) Regularization of diffusion-based direction maps for the tracking of brain white matter fascicles.
Neuroimage
 
12
:
184
–195.
Rao SM, Binder JR, Bandettini PA, Hammeke TA, Yetkin FZ, Jesmanowicz A, Lisk LM, Morris GL, Mueller WM, Estkowski LD, et al. (
1993
) Functional magnetic resonance imaging of complex human movements.
Neurology
 
43
:
2311
–2318.
Rizzolatti G, Luppino G, Matelli M (
1996
) The classic supplementary motor area is formed by two independent areas.
Adv Neurol
 
70
:
45
–56.
Romo R, Schultz W (
1992
) Role of primate basal ganglia and frontal cortex in the internal generation of movements. III. Neuronal activity in the supplementary motor area.
Exp Brain Res
 
91
:
396
–407.
Sakai K, Hikosaka O, Miyauchi S, Sasaki Y, Fujimaki N, Putz B (
1999
) Presupplementary motor area activation during sequence learning reflects visuo-motor association.
J Neurosci
 
19
:
RC1
.
Selemon LD, Goldman-Rakic PS (
1985
) Longitudinal topography and interdigitation of corticostriatal projections in the rhesus monkey.
J Neurosci
 
5
:
776
–794.
Takada M, Tokuno H, Nambu A, Inase M (
1998
) Corticostriatal input zones from the supplementary motor area overlap those from the contra- rather than ipsilateral primary motor cortex.
Brain Res
 
791
:
335
–340.
Tanji J (
1996
) New concepts of the supplementary motor area.
Curr Opin Neurobiol
 
6
:
782
–787.
Toni I, Schluter ND, Josephs O, Friston K, Passingham RE (
1999
) Signal-, set- and movement-related activity in the human brain: an event-related fMRI study.
Cereb Cortex
 
9
:
35
–49.
Wiesendanger M, Rouiller EM, Kazennikov O, Perrig S (
1996
) Is the supplementary motor area a bilaterally organized system?
Adv Neurol
 
70
:
85
–93.