Behavioural Brain Research 293 (2015) 19

Contents lists available at ScienceDirect

Behavioural Brain Research

journa l homepage: www.e lsev ier .com/ locate /bbr

Research report

Abnorm s incallosa

Erhan Ge ,e, Oa Ruhr Universib Department o 528 Frc Brain Imaging Center Frankfurt, Schleusenweg 216, D-60528 Frankfurt am Main, Germanyd Ernst Strngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society, Deutschordenstr. 46, Frankfurt am Main D-60528, Germanye Frankfurt Institute for Advanced Studies, Goethe University, Ruth-Moufang-Str. 1, D-60438 Frankfurt am Main, Germany

h i g h l

In controls BOLD conn In controls In acallosa

a r t i c l

Article history:Received 2 JunReceived in reAccepted 4 JulAvailable onlin

Keywords:Motor cortexInterhemispheCorpus callosuDTIfMRI

1. Introdu

The corpconnecting [1]. Callosarons of the h

CorresponBiopsychology

E-mail add

http://dx.doi.o0166-4328/ i g h t s

, unilateral hand movements evoked asymmetric BOLD activity for both M1.ectivity show that this is induced by interhemispheric motor suppression., this suppression was mediated by microstructure of specic CC bers.l patients interhemispheric motor suppression was absent.

e i n f o

e 2015vised form 3 July 2015y 2015e 14 July 2015

ric inhibitionm agenesis

a b s t r a c t

During unilateral hand movements the activity of the contralateral primary motor cortex (cM1) isincreased while the activity of the ipsilateral M1 (iM1) is decreased. A potential explanation for thisasymmetric activity pattern is transcallosal cM1-to-iM1 inhibitory control. To test this hypothesis, weexamined interhemispheric motor inhibition in acallosal patients. We measured fMRI activity in iM1 andcM1 in acallosal patients during unilateral hand movements and compared their motor activity pattern tothat of healthy controls. In controls, fMRI activation in cM1 was signicantly higher than in iM1, reectinga normal differential task-related M1 activity. Additional functional connectivity analysis revealed thatiM1 activity was strongly suppressed by cM1. Furthermore, DTI analysis indicated that this contralaterallyinduced suppression was mediated by microstructural properties of specic callosal bers interconnect-ing both M1s. In contrast, acallosal patients did not show a clear differential activity pattern betweencM1 and iM1. The more symmetric pattern was due to an elevated task-related iM1 activity in acallosalpatients, which was signicantly higher than iM1 activity in a subgroup of gender and age-matchedcontrols. Also, interhemispheric motor suppression was completely absent in acallosal patients. Thesendings suggest that absence of callosal connections reduces inhibitory interhemispheric motor interac-tions between left and right M1. This effect may reveal novel aspects of mechanisms in communicationof two hemispheres and establishment of brain asymmetries in general.

2015 Published by Elsevier B.V.

ction

us callosum (CC) is the largest commissure in humans,the two hemispheres via 200350 million axon bersl bers mostly project excitatory on pyramidal neu-omotopic contralateral area, which then often activate

ding author at: Ruhr-University Bochum, Faculty of Psychology,, Room: GAFO 05/620, D-44780 Bochum, Germany.ress: erhan.genc@rub.de (E. Genc ).

GABAergic interneurons in the contralateral hemisphere [2]. Thus,the ultimate effect of callosal activity on the contralateral hemi-sphere is assumed to be mostly inhibitory [2]. A domain wheretranscallosal inhibition is important is motor control of hand move-ments. During intended unilateral hand movements there is anasymmetric involvement of the primary motor cortex (M1) in thetwo hemispheres. Due to the architecture of the motor system thereis a predominant role of the contralateral M1 (cM1) in controllingthe hand [3,4]. The contributions of the ipsilateral M1 (iM1) aremore complex and show only an initial involvement [5]. PreviousTMS/MEP studies indicate that iM1 can control the ipsilateral hand,

rg/10.1016/j.bbr.2015.07.0162015 Published by Elsevier B.V.al interhemispheric motor interactionl agenesis

nca,b,c,d,, Sebastian Ocklenburga, Wolf Singerb,c,d

ty Bochum, Biopsychology, GAFO 05/620, D-44780 Bochum, Germanyf Neurophysiology, Max Planck Institute for Brain Research, Deutschordenstr. 46, D-60 patients with

nur Gntrkna

ankfurt am Main, Germany

2 E. Genc et al. / Behavioural Brain Research 293 (2015) 19

most likely through uncrossed corticospinal projections [36]. Thisinitial ipsilateral control is normally inhibited soon through tran-callosal inhibition by the cM1 in conditions where crosstalk isobstructive [4,68]. Functional magnetic resonance imaging (fMRI)studies havboth M1s foblood oxygincreased wpared to a basymmetrictralateral tostudy in heaassumes this caused bytranscallosain healthy pus callosusuppressionapproach. UOReilly et BOLD activia novel apptralaterallymicrostructwith diffusitigating theinhibition athat a stronbe associatecal experimcontralaterAgCC. AgCCitally compenvironmenAs a conseinterhemispent higher is importandemonstratthese patienthat the lacmotor BOLDfact would in AgCC pacates that cfor establisbrain asymhemispheri

2. Materia

2.1. Particip

Four pattial AgCC (s27.4 years; normal schlies. They wpatients wetrous as meAgCC patienhealthy conyears). Indegroup diffetest showed

(p = 0.99). In order to balance both groups in regard to handedness,four out of seven healthy participants were right-handed and threeleft-handed as measured by the Edinburgh Handedness Inventory[25]. In order to replicate the ndings by Hayashi et al. [10] in a big-

plepantdditiart inl visit was we

Soci of th

quis

datain, Gen, Gtreng

Anato co-data

1 mmle: 9was

Moto theg (E: 392 mize =as 9 m

Diffu diffcho ess =

sized alnall

ed inr comce. Tans tr dif

otor

usedhl et aVersws Xdcryh a mhile

rnatingeot, a

thamilent ed e revealed concordant asymmetric involvements ofr hand movements. During unilateral hand movementsenation level dependent (BOLD) activation of cM1 ishile the activation of iM1 is decreased or reduced, com-aseline period with no motor preparation [911]. This

BOLD activation pattern is potentially caused by a con- ipsilateral M1 motor suppression. Based on an fMRIlthy subjects, Hayashi et al. [10] proposed a model that

at contralaterally induced ipsilateral BOLD suppression transcallosal inhibition. In order to test this postulatedl effect we used a multimethod neuroimaging approachparticipants and in patients with agenesis of the cor-m (AgCC). We rst examined interhemispheric motor

in healthy participants using a functional connectivitysing the psychophysiological interactions (PPI) methodal. [12], we determined whether suppression of iM1ty is dependent on the activity of cM1. Next, by usingroach, we then tested for the rst time, whether con-

induced BOLD suppression of iM1 was related to theural properties of specic callosal bres as determinedon tensor imaging (DTI). Since previous studies inves-

link between TMS induced interhemispheric motornd callosal microstructure [13,14], we hypothesizedg iM1 BOLD suppression in a healthy individual shouldd with an efcient callosal microstructure. In a criti-ent, we than tested in a second approach, whether thisally induced iM1 suppression is absent in patients with

is a rare condition, in which callosal bers are congen-letely or partially absent, caused by different genetic ortal factors during prenatal callosal development [15].

quence AgCC patients usually show prolonged visualheric transfer times [16,17] and deciencies in differ-

cognitive abilities, where interhemispheric integrationt [18,19]. In the motor domain, early TMS studies haveed that interhemispheric motor inhibition is affected ints [20,21], therefore, we hypothesized for our patientsk of callosal bers should diminish interhemispheric

suppression during unilateral hand movements. Thisresult in a more symmetric involvement of both M1stients. Beyond motor functions, previous work indi-allosal suppression in healthy individuals is importanthment of brain asymmetries [22,23]. In AgCC patients,metries are reduced [19] or abnormal [24], reectingc independence for sensory and cognitive functions.

l and methods

ants

ients with complete, isolated and one patient with par-ee Fig. 1) took part in the study (three males; mean age,range, 1855 years). All AgCC patients had undergoneooling and came from German middle class fami-ere recruited via internet advertisements. Two AgCCre right-handed, one left-handed and two ambidex-asured by the Edinburgh Handedness Inventory [25].ts were compared to seven age-, and gender-matchedtrols (5 males; mean age, 29.8 years; range 1864pendent t tests revealed that there was no signicantrence in age (p = 0.80) and a two-sample Fishers exact

that there was no signicant group difference in gender

ger samparticiwere atook pnormaconsencedurePlanckmittee

2.2. Ac

All am MaErlangdient s

2.2.1. For

tional 1 1 ip angimage

2.2.2. For

imaginappliedFoV = 1voxel stask w

2.2.3. The

spin-ethicknmatrixtributeAdditioacquirand fosequentive sctime fo

2.3. M

Weby Waware (Windoa liquithrougscans wof alteof the right foperformwere fmovemconrm of healthy controls, thirty-eight right-handed healthys (17 males; mean age 26.63 years; range 2231 years)onally recruited. In total forty-ve healthy participants

our study. All participants had normal or corrected-to-on and were paid for participation. Written informeds obtained from all participants. The experimental pro-re in accordance with the ethical regulations of the Maxety and the study was approved by the local ethics com-e University Hospital Frankfurt am Main.

ition of imaging data

were acquired at the Brain Imaging Center Frankfurtermany, using a Siemens 3-Tesla Trio scanner (Siemens,ermany) with a 8-channel head coil and maximum gra-th of 40 mT/m.

mical imagingregistration and anatomical localization of func-, a T1-weighted high-resolution anatomical image of

3 was acquired (MP-RAGE, TR = 2250 ms, TE = 2.6 ms,, FoV: 256 mm). The acquisition time for the anatomical10 min.

r task motor task, a gradient-recalled echo-planar-PI) sequence with the following parameters was6 slices, TR = 2640 ms, TE = 30 ms, ip angle = 90,m, slice thickness = 3 mm, gap thickness = 0.75 mm,

3.0 3.0 3.0 mm. The acquisition time for the motorin.

sion tensor imagingusion-weighted data were acquired using single-shotecho-planar imaging (TR = 8200 ms, TE = 99 ms, slice

2 mm, FoV = 192 mm, voxel size = 2.0 2.0 2.0 mm, = 96 96). Diffusion weighting was isotropically dis-ong 60 directions using a b-value of 1000 s/mm2.y, ten data sets with no diffusion weighting wereitially as anatomical reference for motion correctionputation of diffusion coefcients during the diffusiono increase signal-to-noise, we acquired three consecu-hat were subsequently averaged [26]. Total acquisitionfusion imaging was 30 min.

task and experimental procedure

a simple visually guided motor task similar as describedl. [13]. Stimuli were generated with Presentation soft-

ion 10.3, www.neurobs.com) running under MicrosoftP and were back-projected onto a frosted screen withstal-display projector. Participants viewed the screenirror xed on the head coil. We acquired a total of 192

participants performed four blocks (21 s per condition)ng rest, pursing movements of lips, exion movementsrs of the left hand, ngers of the right hand, toes of thend toes of the left foot. Participants were instructed toe movements at a self-paced rate of 1 Hz. Participantsiarized with the instructions and practiced the motorbefore scanning. After the experiment, all participantsthat they had been able to carry out the task correctly.

E. Genc et al. / Behavioural Brain Research 293 (2015) 19 3

Fig. 1. Sagitta s T1-wdirections: rig llosumimages of SOH 1 has intact. (For int o the w

Throughoutoutside the

2.4. Analysi

2.4.1. PreprFunction

FSL toolboxusing a numtial smoothcutoff in sehigh-resolustereotaxic (MNI).

2.4.2. AtlasIn a

were denAtlas (left GM PMC 4pformed usinBOLD spaceto extract tfrom each ptransformadictions forthe conditiofurther anaQUERY to cin the left ment compwere extracdata as inpwith SPSS America). Fwe computhemispherehand (domFor the comseven healtrepeated mcontralaterawithin-partticipants) aperformed

Funct usedine t du

handlaterilistiure l line

andsupptivitiffere1 acent nnecion >

rightivit-leveted ) 1 + nnecage- resh

withddititativl (top) and axial (bottom) view of the DTI-based images overlaid onto the individualhtleft (red), anteriorposterior (green) and superiorinferior (blue). The corpus ca10, ARA04, MMZ11 and RHE26 demonstrating complete callosal agenesis and JGR1erpretation of the references to colour in this gure legend, the reader is referred t

the sessions, the motor movements were monitored scanner room by the experimenter.

s of functional data

ocessingal motor data were analyzed using FEAT, part of the

(www.fmrib.ox.ac.uk/fsl). Images were pre-processedber of steps: motion- and slice-timing correction, spa-

ing with a 5 mm FWHM Gaussian kernel, high-pass lterconds (150), linear coregistration to the individualstion T1-weighted anatomical image and to the standardspace template of the Montreal Neurological Institute

-based ROI analysesrst step, cortical regions for ROI-based analysesed probabilistically using the Jlich Histologicalhand M1 = Area GM PMC 4p L; right hand M1 = Area

L) in MNI space [27]. Second, both M1s were trans-g FLIRT [28] from standard MNI space into the native, via the individuals high-resolution anatomical image,he mean amplitude of the BOLD activation in M1. Dataarticipant were visually inspected to conrm that the

tion procedure was successful. Since we had only pre- the interactions of both M1s of the hand area, onlyns of nger movements for the hands were used forlyses. Based on z-statistic procedures, we used FEAT-

2.4.3. We

determintereslateralcontraprobabprocedgeneraregionmore connecfour deral) Mmovemtive coactivatment connechigherconducEffectsPPI coseven tical thoption

In aquantiompute the mean amplitude of the BOLD activationand right hand M1 for each unilateral hand move-ared to baseline. Two z-values per hand movementted (ipsilateral M1, contralateral M1). We used theseut to our second-level statistics, which we performed(version 20, SPSS Inc., Chicago, IL, United States ofor the large sample of healthy participants (N = 45)ed a two-way (2 2) repeated measures ANOVA with

(ipsilateral M1 = iM1; contralateral M1 = cM1) andinant; non-dominant) as within-participants factors.parisons between AgCC patients and the subgroup ofhy participants we performed a three-way (2 2 2)easures ANOVA with hemisphere (ipsilateral M1 = iM1;l M1 = cM1) and hand (dominant; non-dominant) asicipants factors and group (AgCC patients; healthy par-s between-participants factors. Statistical tests wereusing an -level of 0.05.

cM1-to-iM1value was busing the tFor each mM1 area. A iM1 motor hand moveor in multip

2.5. Analys

2.5.1. PreprDiffusion

fusion Tool(i) correctiothe gradieneters from eighted anatomy. The colors represent ber orientations in different of the healthy participant SDA01 (right) is clearly visible in red. The

partial agenesis where the anterior part of the corpus callosum is stilleb version of this article.)

ional connectivity the psychophysiological interactions (PPI) approach towhich voxels in the brain co-vary with a seed region ofring a particular behavioral task [12], which were uni-

movements in our study. As seed regions we used theal M1 hand activity clusters, which we again denedcally from the Jlich Histological Atlas by using the sameas mentioned above. Depending on the design of thear model (GLM), these covariations between the seed

all other voxels in the brain could be a negative (aression-like) or a positive (a more facilitation-like)y pattern. For each individual participant we creatednt PPI maps (right hand movement left (contralat-tivation > whole brain negative connectivity; right hand left (contralateral) M1 activation > whole brain posi-tivity; left hand movement right (contralateral) M1

whole brain negative connectivity; left hand move-t (contralateral) M1 activation > whole brain positivey). We used these maps as input to random effectsl analyses in FEAT. The random effects analyses wereusing the FLAME (FMRIBs Local Analysis of Mixed2 option. This analysis was conducted to compare thetivity pattern of the AgCC patients with those of theand gender-matched healthy participants. The statis-old was set at a FWE-corrected cluster thresholding

a p value < .05 and Z value >2.3.on to calculating group statistics we also determined ae connectivity value representing the interhemispheric motor suppression for each individual control. Thisased on z-statistic procedures and was computed by

wo negative connectivity PPI maps mentioned above.ap we extracted a mean z-value for the ipsilateral handhigher z-value was associated with a stronger cM1-to-suppression. The two z-values were averaged for eachment and entered as dependent variable in correlationle-regression analyses.

is of diffusion data

ocessing tensor images were analyzed using FDT (FMRIBs Dif-

box) implemented in FSL. Preprocessing steps includedn for eddy current and head motion, (ii) correction oft direction for each volume using the rotation param-the head motion. For the evaluation of white-matter

4 E. Genc et al. / Behavioural Brain Research 293 (2015) 19

microstructure, we calculated fractional anisotropy (FA) maps byusing the DTIFIT tool. By using FNIRT all diffusion images were non-linearly aligned to the standard space template of the MNI brain viathe individuals high-resolution T1-weighted anatomical image.

2.5.2. Geom(CC)

In ordethe BOLD iindividual cstandardizecontrols. Pring the CC a parcellatitracking in specic bewith the scsagittal lenwhole CC iwhole CC wMNI standaterior alongposterior mFA images standard spthe callosalFinally, we ues for the analyses of motor suppaverage z-vpression methe FA valuindependenanalyses.

3. Results

3.1. ROI An

To test described bmeasures AM1 = iM1; dominant) signicant 2 = 0.87), inity for iM1 (movementsthis effect wmoved theiCohens d =hand, the apared to theof hand emthat particimoved theinon-domininteraction However, Bments withthe iM1 ac(p < 0.001; level in the

strong trend for increased activation during the non-dominanthand movements (Fig. 2a).

3.2. ROI Analyses for AgCC patients and controls

rders inha thhere

domi(AgC

The A= 43.canto cMant ing tLD acentserge

ts hals (0.d anls (p ion l. Theo rea

nctio

Negarderral hap entsssion). Inteund mot

by t and) anatienssion).

Posited on

andt explthy pheri), neatieove

ed inhereer, th

cM1o hemach o

nctio

resunearetry-based tract segmentation in the corpus callosum

r to test whether the inter-individual variation ofnterhemispheric motor suppression is related to theallosal white-matter microstructure we performed ad geometrical parcellation of the CC for all healthyevious studies used anatomical markers for parcellat-into different subregions [29]. In our study, we usedon method that was validated using in-vivo DTI berhumans [30]. Here, each callosal segment representsrs projecting to distinct cortical areas. In accordance

heme of Hofer and Frahm [30], we measured the mid-gth of the maximum anteriorposterior extent of then MNI standard space. As dened by the scheme, theas divided once into ve sub-segments (see Fig. 4a) inrd space. These segments were from anterior to pos-

the y-axis: anterior third (I), anterior midbody (II),idbody (III), isthmus (IV) and the splenium (V). Sinceof all participants were non-linearly aligned to MNIace, we used an automatic procedure in transforming

segments back to the FA images of each participant.computed for each participants FA map mean FA val-ve corresponding callosal sub-segments. For regressionthe relationship between cM1-to-iM1 interhemisphericression and callosal white-matter microstructure, thealue for the cM1-to-iM1 interhemispheric motor sup-ntioned above was entered as dependent variable andes of the different callosal segments were entered ast variables, either individually or in multiple-regression

alyses in the larger control group

the inhibitory interhemispheric motor control asy Hayashi et al. [10], a two-way (2 2) repeatedNOVA was computed with hemisphere (ipsilateral

contralateral M1 = cM1) and hand (dominant; non-as within-participants factors. The ANOVA revealed amain effect of hemisphere (F(1,44) = 288.08; p < 0.001;dicating that participants showed reduced BOLD activ-

0.10 0.08) compared to cM1 (0.84 0.08) during hand (Fig. 2a). Bonferroni corrected post-hoc tests showedas slightly different for both hands: when participants

r dominant hand, iM1 activity was decreased (p < 0.001; 1.35), whereas when they moved their non-dominantctivity was reduced (p < 0.001; Cohens d = 1.15) com-

baseline activity. Additionally, a signicant main effecterged (F(1, 44) = 38.60; p < 0.001; 2 = 0.47), indicatingpants had reduced BOLD activity in M1 when theyr dominant hand (0.35 0.08) in comparison to theirant hand (0.60 0.08). The effect of hemisphere handwas not signicant (F(1,44) = 0.68; p = 0.41; 2 = 0.02).onferroni corrected post-hoc tests showed that move-

the dominant hand evoked stronger reduction intivity than movements with the non-dominant handCohens d = 0.52), whereas the asymmetric activation

cM1s was not signicant (p = 0.054), but showed a

In oreduceputed hemisphand (group factor.(F(1,10)a signipared signicindicatM1 BOmovemtion empatiencontroshowecontroactivatgroupsfailed t

3.3. Fu

3.3.1. In o

unilatetivity mmovemsuppreFig. 3aalso fothe pre(SMA)to-iM1(Fig. 3bAgCC psuppreFig. 3c

3.3.2. Bas

actionsdid noin heahemispFig. 3dAgCC phand minvolvhemispHowevitatorythe twfrom e

3.4. Fu

Thelosum to test whether the absence of the corpus callosumibitory interhemispheric motor interactions we com-ree-way (2 2 2) repeated measures ANOVA with

(ipsilateral M1 = iM1; contralateral M1 = cM1) andnant; non-dominant) as within-participants factors andC patients; healthy controls) as between-participantsNOVA revealed a signicant main effect of hemisphere

61; p < 0.001; 2 = 0.81), indicating that participants hadtly reduced BOLD activity for iM1 (0.82 0.21) com-1 (1.44 0.21) during hand movements as well as a

main effect of group (F(1,10) = 7.18; p = 0.02; 2 = 0.42),hat AgCC patients (1.68 0.32) had an overall highertivity than controls (0.58 0.27) during unilateral hand. Moreover, a signicant hemisphere group interac-d (F(1,10) = 6.23; p = 0.03; 2 = 0.38), indicating that AgCCd a ten times higher iM1 activity (1.49 0.33) than15 0.28, Fig. 2b). Bonferroni-corrected post-hoc tests

increased iM1 activity for AgCC patients compared to= 0.02; Cohens d = 1.85), whereas the difference in theevel of cM1 was not signicant (p = 0.13) between the

main effect of hand and all other interaction effectsch signicance (p > .49).

nal connectivity for AgCC patients and controls

tive connectivity to test whether there is a suppression of iM1 duringand movements we computed a negative PPI connec-using cM1 as seed region. For both left and right hand

we found a strong interhemispheric cM1-to-iM1 motor in healthy participants (p < .05, Z > 2.3, FWE corrected,restingly, in addition to the iM1 motor suppression wea strong bilateral suppression of the medial portion ofor cortex also known as the supplementary motor areahe cM1. In contrast, both suppression patterns (cM1-

premotor) were completely absent in AgCC patientsd a group difference between healthy participants andts revealed only a cM1-to-iM1 interhemispheric motor

for healthy participants (p < .05, Z > 2.3, uncorrected,

ive connectivity previous studies about interhemispheric motor inter-

the results of the ROI analyses described above weect a positive PPI connectivity between cM1 and iM1articipants. Analyses showed that there was no inter-c cM1-to-iM1 facilitation (p < .05, Z > 2.3, FWE corrected,ither for left nor for right hand movements. Since thents showed increased iM1 activation during unilateralments, indicating that in addition to cM1, iM1 is also

executing the movement, one could expect that boths be functionally coupled during hand movements.e PPI connectivity analyses did not indicate such a facil--to-iM1 relation, suggesting that also in AgCC patientsispheres execute the hand movements independentlyther (see Fig. 3ef).

nal connectivity and callosal microstructure

lts above indicate that the absence of the corpus cal-ly completely abolishes the interhemispheric motor

E. Genc et al. / Behavioural Brain Research 293 (2015) 19 5

Fig. 2. Ipsilatean asymmetricof seven age ana more symmerror.

Fig. 3. Functiovoxels of the cseed region. (Anear M1 of thecorrected). (C)(DF) For healorange; p < .05

suppressionspheric mobers, we cmotor suppdifferent caanalysis wicM1-to-iM1variable, FAments werecM1-to-iM1ral and contralateral BOLD responses in M1 during unilateral hand movements. (A) Region BOLD response for the contralateral (cM1) and ipsilateral M1 (iM1). Here the iM1 activity id gender matched participants reects a similar asymmetric response pattern (four bars

etric activity pattern, in which particularly the iM1 activity is signicantly elevated (fou

nal connectivity maps during unilateral hand movement as estimated by the means ofontralateral M1 hand activity (green) and estimated which of the voxels in the brain sho) In the subgroup of seven-matched healthy participant we found strong suppression of v

other hemisphere (purple). (B) This interhemispheric contralateral to ipsilateral M1 (cM A group comparison demonstrates a stronger interhemispheric cM1-to-iM1 suppression ithy participants as well as for agenesis patients an interhemispheric cM1-to-iM1 facilitat, FWE corrected). (For interpretation of the references to colour in this gure legend, the

for AgCC patients. To examine whether the interhemi-tor suppression is directly modulated by the callosalorrelated the individual cM1-to-iM1 interhemisphericression z-value with the individual FA values for thellosal segments. In a combined multiple-regressionth callosal segment FAs as independent variables and

interhemispheric motor suppression as dependent of the posterior midbody and the isthmus seg-

the only variables providing unique contribution to interhemispheric motor suppression (posterior mid-

body segme = .79, t(39

separate bishowed thato-iM1 inter(43) = .38, found betwand FA of r(43) = .02, rior midbods-of-interest (ROI) analysis in forty-ve healthy participants indicatess decreased or reduced and cM1 is increased. (B) The healthy subgroup

on the left). In comparison, the group of ve agenesis patients showsr bars on the right; see Section 3). Error bars represent the standard

psychophysiological interactions (PPI). As seed region we used thewed a negative (left panel) or a positive (right panel) coupling to thisoxels (activation map in blue; p < .05, FWE corrected) covering voxels1-to-iM1) suppression was absent in agenesis patients (p < .05, FWEn healthy participants than in agenesis patients (p < .05, uncorrected).ion during unilateral hand movements was absent (activation map in

reader is referred to the web version of this article.)

nt: = .66, t(39) = 3.21, p = 0.003; isthmus segment:) = 3.57, p = 0.001; other predictors: p > 0.22). However,variate correlation analyses for the callosal segmentst only FA of the isthmus segment correlated with cM1-rhemispheric motor suppression (isthmus segment:p = 0.01, see Fig. 4b). No signicant correlations wereeen cM1-to-iM1 interhemispheric motor suppressionthe other CC segments (posterior midbody segment:p = 0.26; splenium segment: r(43) = .18, p = 0.26; ante-y segment: r(43) = .28, p = 0.06; anterior third segment:

6 E. Genc et al. / Behavioural Brain Research 293 (2015) 19

Fig. 4. Schemsuppression fodened by theanterior midbisthmus signiand FA valuesarticle.)

r(43) = .23, dent variabvariable, bumultiple-repression iin other indpower of thterior midbnot related its own. Theciated with

4. Discussi

In our stBOLD activreduced ipsAdditional fhemispheriactivities, ipressed by that differeinterconnecerally induindividualsacallosal pamotor funcatic description of the geometry-based tract segmentation in the corpus callosum and cr forty-ve healthy participants. (A) Geometry-based tract segmentation in the corpus ca

scheme of Hofer and Frahm [30], the whole CC was divided manually into ve sub-segmody (orange), posterior midbody (blue), isthmus (light-blue) and the splenium (green). (Bcantly predicted the variability of interhemispheric cM1-to-iM1 suppression. No relati

of bers in other callosal sub-segments. (For interpretation of the references to colour i

p = 0.13, see Fig. 4b). A situation in which an indepen-le shows no bivariate correlation with the dependentt makes a signicant contribution in the context of agression analysis with other variables, is called sup-n statistics. The variable suppresses noise varianceependent variables and thereby enhances predictivee variable set as a whole [31]. In our data set, the pos-ody FA seems to act as a suppressor variable, since it isto cM1-to-iM1 interhemispheric motor suppression onrefore, only FA of the isthmus segment is directly asso-

the cM1-to-iM1 interhemispheric motor suppression.

on

udy, healthy individuals showed an expected increasedity of the contralateral M1 and a decreased orilateral M1 activity during unilateral hand movements.unctional connectivity analysis revealed negative inter-c coupling between contralateral und ipsilateral M1ndicating that ipsilateral activity was primarily sup-activity of the contralateral M1. Furthermore, we foundnces in the microstructural properties of callosal bersting both M1 correlated with variations of contralat-ced ipsilateral (cM1-to-iM1) suppression in healthy. In contrast, cM1-to-iM1 suppression was absent intients (AgCC), reecting hemispheric independence fortions.

The expmovementsings [911,with the ceof the undesured by loshown thatrelated to ethe prominments.

Howevetrol in heaiM1 can concorticospininhibited thditions wheresolution represent thipsilateral hhigher actinant hand. in accordanby asymmecM1iM1 inplay an impin motor codetect asymorrelations between transcallosal bers and interhemispheric BOLDllosum overlaid onto the FMRIB58 fractional anisotropy template. Asents from anterior to posterior along the y-axis: anterior third (red),) Only the fractional anisotropy (FA) of bers projecting through theons were found between interhemispheric cM1-to-iM1 suppressionn this gure legend, the reader is referred to the web version of this

ected increase in cM1 BOLD activity for unilateral hand relative to baseline is in accordance with previous nd-32]. Increased or positive BOLD activity is associatedrebral blood ow and therefore with the metabolismrlying structure [33,34] and the neural activity mea-cal eld potentials [35,36]. For the motor cortex it was

increased BOLD responses in M1 were predominantlyxcitatory synaptic activity [37]. Our results reconrment role of the cM1 in controlling unilateral hand move-

r, there is also an involvement of iM1 in motor con-lthy individuals. Early TMS/MEP studies indicate thattrol the ipsilateral hand, most likely through uncrossedal projections [35]. This ipsilateral control is normallyrough trancallosal inhibition [4,6] by the cM1 in con-re crosstalk should be avoided [7,8]. As the temporal

of fMRI is relatively poor, BOLD responses of iM1 onlye net effect of transcallosal inhibition and activation byand movements [911]. Our results indicate a slightly

vation of iM1 of the non-dominant than of the domi-This signicant asymmetric pattern of iM1 activation isce with Hayashi et al. [10] and is potentially inducedtric iM1 involvement in hand control and asymmetricterhemispheric inhibition [38]. These two factors alsoortant role in the generation of hemispheric dominancentrol [8]. In addition to studies using TMS and fMRI tometric interhemispheric effects in the motor domain,

E. Genc et al. / Behavioural Brain Research 293 (2015) 19 7

earlier studies using motor training in humans detected similar dif-ferential effects from training in one hand to performance of theother untrained hand. In some studies it is argued that the domi-nant hemisphere is the only procient system for motor engrams.This assumpets more faround, sinof training iafterwards studies it isprocient sysphere. Heruntrained hsphere storehemispherebenets moaround [41benet fromnon-dominstandards twith existinnegative inthis unwana strong innon-dominshowing a sthe domina

By usingshow that tspheric supfor whole biM1 and in results suggindirect mondings by ing approaca strong netralateral SMhand moveboth handsand reciprodynamicallybition to fastudy only cM1-to-iM1

By usinghemispherimicrostructthe projectiiological inbecause diftors, includIncreased mdiffusion ofvalues in a glosal ber cthe posterioregions conpacked bedict that inddensely pachemispheriin showingmus and z vThis is in acbetween TM

losal microstructure [13,14]. They found the same relationshipthat the stronger the interhemispheric inhibition the higher theFA values in specic callosal regions covering mostly the pos-terior midbody and parts of the isthmus. Another recent study

d thas is aheri

of thheri

ts. Mete [2periomispral mral mgnal,by a

is alsion iete aCC it

cove1], inial fo

thint two

(ii) absenutiogligibant iealthiatedppreberus th

showy of i

for tountd momisp. Wehat tal phd res

dematie-to-i

intacTMS pprerongthy ithey llosauctiLD a

nddica

respoilatessions anand

the tion leads to an effect in which the dominant hand ben-rom non-dominant hand training than the other wayce for the latter, additional interhemispheric transfernformation is needed when performing the movementwith the untrained non-dominant hand [39,40]. In other

assumed that the dominant hemisphere is the morestem for motor engrams, than the non-dominant hemi-e training information is already communicated to theemisphere during learning. Since the dominant hemi-d superior movement standards and the non-dominant

inferior movement standards, the non-dominant handre from dominant hand training than the other way43]. Another reason why the dominant hand does not

the non-dominant hand training is the fact, that theant hemisphere probably transfers inferior movemento the dominant hemisphere, which in turn interfereg superior movement standards, leading to a morehibitory transfer effect [41]. One possibility to avoidted crosstalk from the non-dominant hemisphere isterhemispheric inhibition from the dominant to theant hemisphere. Our data supports this assumption intronger decreased activation for iM1, by movements ofnt than non-dominant hand.

a functional connectivity approach we were able tohe iM1 activity was strongly related to the interhemi-pression by the cM1. When cM1 was used as seed regionrain functional connectivity analyses, mainly voxels inthe SMAs of both hemispheres were suppressed. Theseest that there is a direct cM1-to-iM1 and an additionaltor suppression via the SMAs. This is in accordance withGrefkes et al. [44] applying the dynamic causal model-h (DCM) for unilateral hand movements. They foundgative effective connectivity by the cM1 and the con-A to the iM1 when participants conducted unilateral

ments. Interestingly, when participants had to move at the same time, both M1s showed strong positivecal effective connectivity. Thus, both M1s are able to

switch their interhemispheric interactions from inhi-cilitation as measured with the BOLD signal. Since ourincludes unimanual hand movements, the facilitatory

interaction is absent in our connectivity analyses. a novel approach we were able to show that inter-c cM1-to-iM1 BOLD suppression was related to theural properties of the isthmus, which is likely to beng zone of callosal motor bers in humans [45]. Phys-terpretation of diffusion properties are challengingfusion anisotropy can be inuenced by a number of fac-ing myelination, ber density or axon diameter [46].yelin thickness and high ber density hinders radial

water molecules and are therefore related to high FAiven voxel. Light microscopic analysis of the human cal-omposition revealed clear regional differences [1]. Forr midbody and the isthmus it was shown that thesesist of larger-diameter, myelinated and less denselyrs. The myelin or ber density hypothesis would pre-ividuals with increased myelin thickness and/or moreked callosal motor bers, would show stronger inter-c motor suppression. Our data supports this hypothesis

a positive correlation between FA values in the isth-alues representing the cM1-to-iM1 BOLD suppression.cordance with previous studies investigating the linkS induced interhemispheric motor inhibition and cal-

showeisthmuhemispbers hemisppatiencomplsilent one hetralateipsilateMEP silowed periodinhibitcompltial AglesionsiSPs [2essent

Weat leasand/orto the acontribare neimport

In his medthis sumotor and thal. [49]activitreasonvoxel cfocuseinterheby cM1show tof neurings anclearlythese pof cM1sum isUsing iM1 suvery stin healwhen the acathe rediM1 BO

Thedata inBOLD ing unsuppremetriedistal hduringt also macroscopic properties (callosal thickness) of thessociated to unilateral hand performance where inter-c inhibition is important [47]. Further evidence thate isthmus and posterior third are important for inter-c motor suppression comes from TMS studies in AgCCeyer and colleagues investigated whether patients with0] or partial AgCC [21] show TMS induced ipsilateralds (iSPs). Transcranial magnetic stimulation of M1 inhere produces motor evoked potentials (MEPs) in con-uscles transmitted through corticospinal tracts [3]. Foruscles there is an initial short period of an increased

induced by uncrossed corticospinal projections [5], fol- longer period of a MEP signal suppression. The lattero known as iSP, which is related to transcallosal motornduced by the contralateral M1 [3]. In patients withgenesis, the iSP is absent [20]. For patients with par-

was shown that only the group of patients who hadring the isthmus/posterior third of the CC did not showdicating that only bers projecting to this region arer interhemispheric motor suppression.k that the abnormal increase in iM1 activity in AgCC has

causes: (i) an enhanced ipsilateral motor projection [5] lack of interhemispheric cM1-to-iM1 suppression duece of callosal bers. Previous research indicates that then of ipsilateral motor projections to hand movementsle [48]. Therefore, we assume that the latter factor isn explaining our ndings.y participants, we found that cM1-to-iM1 suppression

by specic callosal bers, whereas in AgCC patientsssion is absent. This suggests that the absence of callosals had a major impact on the cM1-to-iM1 disinhibitione elevated iM1 activity. In a previous study Reddyeted that acallosal patients demonstrate a similar BOLD

M1 like controls, which is contrast to our ndings. Onehis discrepancy could by that Reddy et al. [49] used a

procedure to detect iM1 BOLD activity. Our approachre on the amplitude of the BOLD response in iM1 and theheric functional negative connectivity in iM1 induced

used these procedures, since several other researchershis kind of analyses is more suited to detect this typeenomenon [9,10,44]. Therefore we think that our nd-ults from previous TMS studies in AgCC patients [20,21]onstrate that interhemispheric motor suppression innts is massively affected. Interestingly, this reductionM1 inhibition is also apparent when the corpus callo-t but functions of motor areas are affected after stroke.Murase et al. [50] show that interhemispheric cM1-to-ssion is important for accurate motor control, since it is

immediately preceding the unilateral hand movementndividuals. In stroke patients, this suppression is absentmove their affected hand. Similarly to our ndings inl subjects, the study in stroke patients also showed thaton of cM1-to-iM1 inhibition also lead to an increasedctivity when the affected hand was moved [51].ings and proposed model by [10] and results of ourte that callosal suppression shapes the asymmetricnses in cM1 (high) and iM1 (low or negative) dur-

ral hand movements. Research indicates that callosal is also important in the establishment of brain asym-d handedness [8,52,53]. It has been proposed that

movements are initially generated bilaterally, and onlynal preparation phase the movement becomes uni-

8 E. Genc et al. / Behavioural Brain Research 293 (2015) 19

lateral through transcallosal inhibition [50,54]. The healthy andadult human population consists of 90% right-handed and 10%left-handed or ambidextrous individuals [55,56]. In adults withcomplete AgCC, this distribution is different in that only 70% areright-handeOne signicAgCC patieduring the could triggphase, whichemisphereinfants showpared to hewere right-ambidextrothere is a coout childhotranscallosaing (from 7hemispheretigated in thAgCC. Similpatients the35% left-hanedness distthe absence

5. Conclus

In concluduring intenof specic tion, we fouis diminishfurther evidpus callosuinto the nehemisphere

Conict of

The auth

Acknowled

This worthank Axel helpful discments, Matand Ralf DeThomas Sat

References

[1] F. Aboitizcorpus ca

[2] J.S. Bloomtransfer o5971.

[3] E.M. Wasmagnetic

[4] E.M. Wasof the ips

[5] U. ZiemanDissociatmotor-ev3) (1999)

[6] A. Ferbert, A. Priori, J.C. Rothwell, B.L. Day, J.G. Colebatch, C.D. Marsden,Interhemispheric inhibition of the human motor cortex, J. Physiol. 453 (1992)525546.

[7] R. Nass, Mirror movement asymmetries in congenital hemiparesis: theinhibition hypothesis revisited, Neurology 35 (1985) 10591062.

iemaivationinan

Allisoctionrolog. Hayaisph

ilateragneticuan,

positieuros

OReihe tra. CognWahl,man mrostru. Voin

role orhem

. Paul,l., Ageects oerluc

uomotassingratioects, B. Brow

evoklosal acklenralizanesis,. Meyibitoras in nlosum. Meypus catices, . Gazzmun

in 123. Hervisph80.

ancke,h stro97) 13. Oldentoryenc , Jpe sub414. Eickhew SPctionaJenkinust anroim

Petersision 72.ofer, isitedonancohen, ressioaum,

. Kim,ctionmmet. Raicn with123

Smithebral I, Prod and 30% are left-handed or ambidextrous [5760].ant factor for this shift in handedness distribution innts could be the absence of transcallosal suppressionnal preparation phase of hand control. This in turner bilateral involvement during the nal preparationh would weaken the strong motor dominance of one. Further support for this hypothesis is the nding that

a less pronounced lateralization of handedness com-althy adults. It was shown that only 70% of the infantshanded and the remaining 30% were left-handed orus [61]. This nding could be explained by the fact thatntinued development of the corpus callosum through-od and adolescence [62,63]. The late maturation of thel inhibitory system [64,65] could be one aspect in boost-0% to 90%) and sustaining the motor dominance of one

at a later age. Interestingly, Sacco et al. [61] also inves-e same study the handedness in infants with completear to the healthy infants and to the adult complete AgCCir handedness distribution was 65% right-handed andded or ambidextrous. The unchanged pattern of hand-

ribution in older AgCC could potentially be caused by of callosal maturation.

ions

sion, we have shown that interhemispheric inhibitionded unilateral hand movements is related to propertiesbers in the corpus callosum of healthy humans. In addi-nd that the inhibitory interaction between motor areased when the corpus callosum is absent. This providesence in humans for the inhibitory function of the cor-m in the motor system and may provide novel aspectsurobiological underpinnings of communication of twos and establishment of brain asymmetries in general.

interest

ors declare no conict of interest.

gments

k was supported by the Max Planck Society. The authorsKohler, Johanna Bergmann and Caspar Schwiedrzik forussions on design and interpretation of the experi-hias Wahl for the support with the motor experimentichmann, Sandra Anti, Steffen Volz, Ulrike Nth, andtler for support with the MRI measurements.

, A.B. Scheibel, R.S. Fisher, E. Zaidel, Fiber composition of the humanllosum, Brain Res. 598 (1992) 143153., G.W. Hynd, The role of the corpus callosum in interhemisphericf information: excitation or inhibition, Neuropsychol. Rev. 15 (2005)

sermann, P. Fuhr, L.G. Cohen, M. Hallett, Effects of transcranial stimulation on ipsilateral muscles, Neurology 41 (1991) 17951799.sermann, A. Pascual-Leone, M. Hallett, Cortical motor representationilateral hand and arm, Exp. Brain Res. 100 (1994) 121132.n, K. Ishii, A. Borgheresi, Z. Yaseen, F. Battaglia, M. Hallett, et al.,

ion of the pathways mediating ipsilateral and contralateraloked potentials in human hand and arm muscles, J. Physiol. 518 (Pt

895906.

[8] U. Zactdom

[9] J.D.FunNeu

[10] M.JHemipsma

[11] H. YforJ. N

[12] J.X.of tSoc

[13] M. Humic

[14] A.NTheinte

[15] L.Ket aasp

[16] G. Bvis

[17] G. Tintedef

[18] W.Sandcal

[19] S. Olateage

[20] B.UInharecal

[21] B.Ucorcor

[22] M.ScomBra

[23] P.Yhem69

[24] L. Jwit(19

[25] R.Cinv

[26] E. Gsha149

[27] S.BA nfun

[28] M. robNeu

[29] M. Div62

[30] S. Hrevres

[31] J. CRegErlb

[32] S.GFunasy

[33] M.Ema233

[34] A.J.CerfMRnn, M. Hallett, Hemispheric asymmetry of ipsilateral motor cortex during unimanual motor tasks: further evidence for motor

ce, Clin. Neurophysiol. 112 (2001) 107113.n, K.J. Meador, D.W. Loring, R.E. Figueroa, J.C. Wright, M.R.I.al, Cerebral activation and deactivation during nger movement,y 54 (2000) 135142.shi, D.N. Saito, Y. Aramaki, T. Asai, Y. Fujibayashi, N. Sadato,eric asymmetry of frequency-dependent suppression in thel primary motor cortex during nger movement: a functional

resonance imaging study, Cereb. Cortex 18 (2008) 29322940.C. Perdoni, L. Yang, B. He, Differential electrophysiological couplingve and negative BOLD responses during unilateral hand movements,ci. 31 (2011) 95859593.lly, M.W. Woolrich, T.E. Behrens, S.M. Smith, H. Johansen-Berg, Toolsde: psychophysiological interactions and functional connectivity,it. Affect. Neurosci. 7 (2012) 604609.

B. Lauterbach-Soon, E. Hattingen, P. Jung, O. Singer, S. Volz, et al.,otor corpus callosum: topography, somatotopy, and link betweencture and function, J. Neurosci. 27 (2007) 1213212138.eskos, F. Farzan, M.S. Barr, N.J. Lobaugh, B.H. Mulsant, R. Chen, et al.,f the corpus callosum in transcranial magnetic stimulation inducedispheric signal propagation, Biol. Psychiatry 68 (2010) 825831.

W.S. Brown, R. Adolphs, J.M. Tyszka, L.J. Richards, P. Mukherjee,nesis of the corpus callosum: genetic, developmental and functionalf connectivity, Nat. Rev. Neurosci. 8 (2007) 287299.chi, S. Aglioti, C.A. Marzi, G. Tassinari, Corpus-callosum and simpleor integration, Neuropsychologia 33 (1995) 923936.ari, S. Aglioti, R. Pallini, G. Berlucchi, G.F. Rossi, Interhemisphericn of simple visuomotor responses in patients with partial callosalehav. Brain Res. 64 (1994) 141149.n, M.A. Jeeves, R. Dietrich, D.S. Burnison, Bilateral eld advantage

ed potential interhemispheric transmission in commissurotomy andgenesis, Neuropsychologia 37 (1999) 11651180.burg, A. Ball, C.C. Wolf, E. Genc, O. Gunturkun, Functional cerebraltion and interhemispheric interaction in patients with callosal

Neuropsychology (2015).er, S. Roricht, H. Gran von Einsiedel, F. Kruggel, A. Weindl,y and excitatory interhemispheric transfers between motor corticalormal humans and patients with abnormalities of the corpus, Brain 118 (Pt 2) (1995) 429440.er, S. Roricht, C. Woiciechowsky, Topography of bers in the humanllosum mediating interhemispheric inhibition between the motorAnn. Neurol. 43 (1998) 360369.aniga, Cerebral specialization and interhemisphericication: does the corpus callosum enable the human condition?

(2000) 12931326.e, L. Zago, L. Petit, B. Mazoyer, N. Tzourio-Mazoyer, Revisiting humaneric specialization with neuroimaging, Trends Cognit. Sci. 17 (2013)

G. Wunderlich, G. Schlaug, H. Steinmetz, A case of callosal agenesisng anatomical and functional asymmetries, Neuropsychologia 35891394.eld, The assessment and analysis of handedness: the Edinburgh, Neuropsychologia 9 (1971) 97113.. Bergmann, W. Singer, A. Kohler, Interhemispheric connectionsjective experience of bistable motion, Curr. Biol. 21 (2011)

99.off, K.E. Stephan, H. Mohlberg, C. Grefkes, G.R. Fink, K. Amunts, et al.,M toolbox for combining probabilistic cytoarchitectonic maps andl imaging data, Neuroimage 25 (2005) 13251335.son, P. Bannister, M. Brady, S. Smith, Improved optimization for thed accurate linear registration and motion correction of brain images,age 17 (2002) 825841., S. Oeltze, D. Seminowicz, H. Steinmetz, S. Koeneke, L. Jancke,of the corpus callosum into subregions, Brain Cognit. 50 (2002)

J. Frahm, Topography of the human corpus callosumcomprehensive ber tractography using diffusion tensor magnetice imaging, Neuroimage 32 (2006) 989994.P. Cohen, S.G. West, L.S. Aiken, Applied Multiplen/correlation Analysis for the Behavioral Sciences, Lawrence

Mahwah, NJ, 2003. J. Ashe, K. Hendrich, J.M. Ellermann, H. Merkle, K. Ugurbil, et al.,al magnetic resonance imaging of motor cortex: hemisphericry and handedness, Science 261 (1993) 615617.hle, Measurement of local cerebral blood ow and metabolism in

positron emission tomography, Fed. Proc. 40 (1981)34., H. Blumenfeld, K.L. Behar, D.L. Rothman, R.G. Shulman, F. Hyder,energetics and spiking frequency: the neurophysiological basis ofc. Natl. Acad. Sci. U. S. A. 99 (2002) 1076510770.

E. Genc et al. / Behavioural Brain Research 293 (2015) 19 9

[35] N.K. Logothetis, The neural basis of the bloodoxygen-level-dependentfunctional magnetic resonance imaging signal, Philos. Trans. R. Soc. Lond. BBiol. Sci. 357 (2002) 10031037.

[36] R. Mukamel, H. Gelbard, A. Arieli, U. Hasson, I. Fried, R. Malach, Couplingbetween neuronal ring, eld potentials, and FMRI in human auditory cortex,Science 309 (2005) 951954.

[37] D. Waldvogel, P. van Gelderen, W. Muellbacher, U. Ziemann, I. Immisch, M.Hallett, The relative metabolic demand of inhibition and excitation, Nature406 (2000) 995998.

[38] J. Netz, U. Ziemann, V. Homberg, Hemispheric asymmetry of transcallosalinhibition in man, Exp. Brain Res. 104 (1995) 527533.

[39] H.G. Taylor, K.M. Heilman, Left-hemisphere motor dominance inrighthanders, Cortex 16 (1980) 587603.

[40] K. Schulze, E. Luders, L. Jancke, Intermanual transfer in a simple motor task,Cortex 38 (2002) 805815.

[41] G. Thut, N.D. Cook, M. Regard, K.L. Leenders, U. Halsband, T. Landis,Intermanual transfer of proximal and distal motor engrams in humans, Exp.Brain Res. 108 (1996) 321327.

[42] J.I. Laszlo, R.A. Baguley, P.J. Bairstow, Bilateral transfer in tapping skill in theabsence of peripheral information, J. Mot. Behav. 2 (1970) 261271.

[43] R. Millisen, C. Van Riper, Differential transfer of training in a rotary activity, J.Exp. Psychol. 24 (1939) 640.

[44] C. Grefkes, S.B. Eickhoff, D.A. Nowak, M. Dafotakis, G.R. Fink, Dynamic intra-and interhemispheric interactions during unilateral and bilateral handmovements assessed with fMRI and DCM, Neuroimage 41 (2008) 13821394.

[45] M. Zarei, H. Johansen-Berg, S. Smith, O. Ciccarelli, A.J. Thompson, P.M.Matthews, Functional anatomy of interhemispheric cortical connections inthe human brain, J. Anat. (2006) 311320.

[46] C. Beaulieu, The basis of anisotropic water diffusion in the nervous systematechnical review, NMR Biomed. 15 (2002) 435455.

[47] F. Kurth, E.A. Mayer, A.W. Toga, P.M. Thompson, E. Luders, The rightinhibition? Callosal correlates of hand performance in healthy children andadolescents callosal correlates of hand performance, Hum. Brain Mapp. 34(2013) 22592265.

[48] B. Zaaimi, S.A. Edgley, D.S. Soteropoulos, S.N. Baker, Changes in descendingmotor pathway connectivity after corticospinal tract lesion in macaquemonkey, Brain 135 (2012) 22772289.

[49] H. Reddy, M. Lassonde, N. Bemasconi, A. Bemasconi, P.M. Matthews, F.Andermann, et al., An fMRI study of the lateralization of motor cortexactivation in acallosal patients, Neuroreport 11 (2000) 24092413.

[50] N. Murase, J. Duque, R. Mazzocchio, L.G. Cohen, Inuence of interhemisphericinteractions on motor function in chronic stroke, Ann. Neurol. 55 (2004)400409.

[51] C. Grefkes, D.A. Nowak, S.B. Eickhoff, M. Dafotakis, J. Kust, H. Karbe, et al.,Cortical connectivity after subcortical stroke assessed with functionalmagnetic resonance imaging, Ann. Neurol. 63 (2008) 236246.

[52] M.S. Gazzaniga, Cerebral specialization and interhemisphericcommunication: does the corpus callosum enable the human condition?Brain 123 (Pt 7) (2000) 12931326.

[53] P.Y. Herv, L. Zago, L. Petit, B. Mazoyer, N. Tzourio-Mazoyer, Revisiting humanhemispheric specialization with neuroimaging, Trends Cognit. Sci. 17 (2013)6980.

[54] P.M. Rossini, F. Zarola, E. Stalberg, M. Caramia, Pre-movement facilitation ofmotor-evoked potentials in man during transcranial stimulation of thecentral motor pathways, Brain Res. 458 (1988) 2030.

[55] M.C. Corballis, The evolution and genetics of cerebral asymmetry, Philos.Trans. R. Soc. Lond. B Biol. Sci. 364 (2009) 867879.

[56] M. Raymond, D. Pontier, Is there geographical variation in humanhandedness? Laterality 9 (2004) 3551.

[57] C. Chiarello, A house divided? Cognitive functioning with callosal agenesis,Brain Lang. 11 (1980) 128158.

[58] H.C. Sauerwein, M. Lassonde, Cognitive and sensori-motor functioning in theabsence of the corpus callosum: neuropsychological studies in callosalagenesis and callosotomized patients, Behav. Brain Res. 64 (1994) 229240.

[59] L.B.N. Hinkley, E.J. Marco, A.M. Findlay, S. Honma, R.J. Jeremy, Z. Strominger,et al., The role of corpus callosum development in functional connectivity andcognitive processing, PLoS One 7 (2012) e39804.

[60] J.P. Owen, Y.O. Li, E. Ziv, Z. Strominger, J. Gold, P. Bukhpun, et al., Thestructural connectome of the human brain in agenesis of the corpus callosum,Neuroimage 70 (2013) 340355.

[61] S. Sacco, M.L. Moutard, J. Fagard, Agenesis of the corpus callosum and theestablishment of handedness, Dev. Psychobiol. 48 (2006) 472481.

[62] R.A. Rauch, J.R. Jinkins, Analysis of cross-sectional area measurements of thecorpus callosum adjusted for brain size in male and female subjects fromchildhood to adulthood, Behav. Brain Res. 64 (1994) 6578.

[63] R. Westerhausen, E. Luders, K. Specht, S.H. Ofte, A.W. Toga, P.M. Thompson,et al., Structural and functional reorganization of the corpus callosumbetween the age of 6 and 8 years, Cereb. Cortex 21 (2011) 10121017.

[64] A. Danek, B. Heye, R. Schroedter, Cortically evoked motor responses inpatients with Xp22.3-linked Kallmanns syndrome and in female genecarriers, Ann. Neurol. 31 (1992) 299304.

[65] K. Muller, F. Kass-Iliyya, M. Reitz, Ontogeny of ipsilateral corticospinalprojections: a developmental study with transcranial magnetic stimulation,Ann. Neurol. 42 (1997) 705711.

Abnormal interhemispheric motor interactions in patients with callosal agenesis1 Introduction2 Material and methods2.1 Participants2.2 Acquisition of imaging data2.2.1 Anatomical imaging2.2.2 Motor task2.2.3 Diffusion tensor imaging

2.3 Motor task and experimental procedure2.4 Analysis of functional data2.4.1 Preprocessing2.4.2 Atlas-based ROI analyses2.4.3 Functional connectivity

2.5 Analysis of diffusion data2.5.1 Preprocessing2.5.2 Geometry-based tract segmentation in the corpus callosum (CC)

3 Results3.1 ROI Analyses in the larger control group3.2 ROI Analyses for AgCC patients and controls3.3 Functional connectivity for AgCC patients and controls3.3.1 Negative connectivity3.3.2 Positive connectivity

3.4 Functional connectivity and callosal microstructure

4 Discussion5 ConclusionsConflict of interestAcknowledgmentsReferences