Comparison of representational maps using functional magnetic resonance imaging and transcranial magnetic stimulation.

OBJECTIVE: Comparison of functional magnetic resonance imaging (fMRI) representational maps, that were generated during voluntary thumb abduction, hand dorsiflexion and foot elevation to amplitude maps of motor-evoked potentials (MEPs) elicited by single transcranial magnetic stimulation (TMS) administered to cortical motor representation areas of the muscles of the thenar eminence, extensor carpi radialis and tibialis anterior muscles. METHODS: Stimulus locations that produced maximal motor-evoked potential amplitudes were compared to fMRI activation maxima in three-dimensional (3D)-space and in a 2D-projection using a novel technique that allowed fMRI activation sites to be projected onto the surface of the brain. RESULTS AND CONCLUSIONS: When analyzing pooled data from all target muscles, the location of projected fMRI and TMS activation maxima on the cortical surface differed by an average 13.9 mm. The differences in 3D distances were particularly large for representation areas of lower leg muscles. 3D distances between fMRI activation maxima and highest MEP site in TMS correlated significantly with higher TMS thresholds. These observations strongly suggest that higher TMS excitation thresholds and lower MEP amplitudes are largely due to the absolute distance between the stimulation site and the excitable cortical tissue targeting this muscle. After the projection 4 out of 5 representation sites as evaluated by TMS were located anterior to the fMRI activation maxima, an observation which may due to the orientation of the magnetic field induced by the current in the coil. The representation sites as evaluated with both methods were specific for the type of movement: distances between representation maxima of the same movements were significantly smaller than those within different movements. Nevertheless, fMRI and TMS provide complementary information, which is discussed on the basis of the functional map observed with both methods.

Dynamics of process formation during differentiation of tectal neurons in embryonic zebrafish.

Neurons acquire their distinct shapes after passing through many transitional stages in early development. To reveal the dynamics and spatiotemporal sequence of process formation in situ, the growth of neurons in the optic tectum of live zebrafish embryos (54 to > 100 h old) was monitored using time-lapse videorecordings. Neurons were labeled by injecting the fluorescent vital dye DiO into the cell-rich layer of the developing tectum in 50- to 70-h-old embryos. In phase 1, tectal neurons possess an apical "primary process" which reaches to the ventral aspect of the tectal neuropil. The primary process produces at its tip short transitory branches, some with growth cones, over a period of roughly 6 h. One of the growth cones then elongates rapidly and grows toward the caudal tectum via a route characteristic of efferent axons. After retraction of excess branches and growth cones, branching activity resumes at the tip of the primary process to form the dendritic tree (phase 2). The dendritic tree develops in the tectal neuropil through emission and retraction of many branches during a period of > 20 h (our longest continuous time-lapse movie). The tectal territory "explored" in this way is larger than the area finally covered by the tree resulting from growth and loss of branches. The dynamics observed here directly are probably characteristic for dendrite formation in vertebrates. Moreover, consistent with the sequence of neuronal differentiation observed in vitro, the growth of the axon precedes that of the dendrites, although both emerge from a common primary process in this type of tectal neuron.

Growth behavior of retinotectal axons in live zebrafish embryos under TTX-induced neural impulse blockade.

The growth dynamics of individual DiO-labeled retinal axons deprived of normal neural impulse activity by TTX was monitored in the tectum of living zebrafish embryos with time-lapse video microscopy and compared with normal active axons. Growth cones of TTX-blocked axons advance intermittently with an average velocity similar to normal axons. While exploring their local environment, they are broadened and bear ruffling lamellipodia and filopodia, but become streamlined when advancing. The activity-deprived axons grow directly towards their retinotopic target sites in the tectum as do their normal counterparts and very rarely extend branches en route. Much like normal axons, TTX-blocked axons begin to branch and develop their terminal arbors only at their retinotopic target area. They emit and retract numerous short side branches over a period of several hours. The area they contact (the "exploration field") is of similar dimension as that of active axons, covering from 1% to 7.4% of the tectal neuropil surface, but the final arbors cover an area only one-half to one-sixth as large. TTX arbors are as small as arbors of normal active axons and retinotopically correct. Thus, the typical exploratory growth behavior of developing retinal axons in the tectum, the dynamics of terminal arbor formation at retinotopically correct sites, the dimension of the exploration field, and the shaping of the arbors in zebrafish embryos are unaffected by TTX-induced neural impulse blockade.

Dynamics of terminal arbor formation and target approach of retinotectal axons in living zebrafish embryos: a time-lapse study of single axons.

In a variety of species, developing retinal axons branch initially more widely in their visual target centers and only gradually restrict their terminal arbors to smaller and defined territories. Retinotectal axons in fish, however, appeared to grow in a directed manner and to arborize only at their retinotopic target sites. To visualize the dynamics of retinal axon growth and arbor formation in fish, time-lapse recordings were made of individual retinal ganglion cell axons in the tectum in live zebrafish embryos. Axons were labeled with the fluorescent carbocyanine dyes Dil or DiO inserted as crystals into defined regions of the retina, viewed with 40x and 100x objectives with an SIT camera, and recorded, with exposure times of 200 msec at 30 or 60 sec intervals, over time periods of up to 13 hr. (1) Growth cones advanced rapidly, but the advance was punctuated by periods of rest. During the rest periods, the growth cones broadened and developed filopodia, but during extension they were more streamlined. (2) Growth cones traveled unerringly into the direction of their retinotopic targets without branching en route. At their target and only there, the axons began to form terminal arborizations, a process that involved the emission and retraction of numerous short side branches. The area that was permanently occupied or touched by transient branches of the terminal arbor--"the exploration field"--was small and almost circular and covered not more than 5.3% of the entire tectal surface area, but represented up to six times the size of the arbor at any one time. These findings are consistent with the idea that retinal axons are guided to their retinotopic target sites by sets of positional markers, with a graded distribution over the axes of the tectum.