In order to study migration of neurons in vitro, we cultured microexplants of the newborn mouse cerebellum outer layer, which is rich in immature granule cells, on a substratum double-coated with poly-L-lysine and laminin. The granule cells first migrated away from the explant along radially oriented parallel bundles of their neurites, thus displaying typical contact guidance. Then, in almost all explants, they changed their orientation by 90° to extend cell processes and translocate perpendicular to the radial neurites. Orientation and migration of neurons perpendicular to the aligned parallel structure is a novel type of contact-guided cell behavior, and may have interesting implications in migration of neurons in the cerebellum and other parts of the nervous system.

Contact guidance is a phenomenon by which cell shape and movement are aligned along the topographical features of the substratum, such as parallel fibers or grooves (Dunn, 1982). It is believed to play important roles during embryogenesis (Nakatsuji, 1984; Nakatsuji & Johnson, 1984; Wood & Thorogood, 1984), and development of nervous system such as migration of the newly born neurons along the radial glial cell processes in the cortex (Rakic, 1988) and pathfinding by the growing axon (Singer et al. 1979; Silver & Sidman, 1980; Silver & Ogawa, 1983).

During prenatal development, granule cell precursors first migrate beneath the pial surface into the cerebellum region to form the external germinal layer. They probably utilize the pre-existing axon bundle for the substratum and guidance (Hynes et al. 1986). After the proliferation in an outer layer of the external granular layer, the immature granule cells extend their bipolar long axons parallel to the pial surface. These axons form the horizontal parallel fibers. During this period, the granule cells have spindle-shaped cell bodies aligned along the parallel fibers. Then, during the period of postnatal 5–15 days, the granule cells in the cerebellum change their polarity by orienting the longer axis of their cell bodies perpendicular to their axons, and now move vertically inward across the bundle of parallel fibers. Many studies have suggested that this later vertical migration phase is guided by vertically aligned processes of the Bergmann glia (Rakic, 1971; Edmondson & Hatten, 1987; Gregory et al. 1988).

In the present study, we tried to analyze behavior of the granule cells in the microexplant culture on a laminin-coated substratum in a serum-free, hormone-supplemented culture medium (Fisher, 1982). It showed that the granule cells drastically changed their orientation and started to extend cell processes and migrate perpendicular to their parallel neurites bundles. TTterefore, we propose that the granule cells may start inward movement in situ by acquiring a tendency to orientate themselves perpendicular to the parallel fibers. Such behavior could provide another mechanism to ensure vertical migration of the granule cells in addition to the guidance by glia cells.

Culture substratum

Glass coverslips were first coated with poly-L-lysine (PL) as described by Schnitzer & Schachner (1981). Sterile coverslips (16 mm in diameter), were immersed in a 100μg ml−1 PL (poly-L-lysine hydrobromide, Sigma) solution, dried under a sterile air flow, rinsed with distilled water, and used for further coating with laminin. About 50 μl of a solution of affinity-purified laminin (20 μg ml−1, E-Y Laboratories) were applied to the dried PL-coated surface for 1 to 2 h at 37 °C, rinsed twice with the culture medium, and used immediately for the microexplant culture.

Culture medium

A serum-free, hormone-supplemented medium was prepared according to Fisher (1982). Briefly, it contained bovine serum albumin (1 mg ml−1, Biomedical Technology), insulin (10 μg ml−1, Sigma), transferrin (100 μg ml−1, Sigma), aprotinin (lμml, Sigma), sodium selenate (30 nM Sigma) and thyroxine (0·1 nM, Sigma). Glucose, glutamine and antibiotics were added as described in the previous paper (Schnitzer & Schachner, 1981).

Microexplant culture

We modified the methods of cerebellar microexplant cultures which have been reported by Nagata et al. (1986) and Fisher et al. (1986). Vermal regions of cerebellar tissues were dissected out from early postnatal (2–-6 days) mice (BALB/c strain) and freed from meninges and choroid plexus. Then, slices were made with a razor blade, from which white matter and deep cerebellar nuclei were removed. Rectangular pieces (300–400pμm) were dissected out from the remaining tissue, which consisted of the external granular and Purkinje cell layers, by using a razor blade. Such prepared microexplants were rinsed with the culture medium and placed evenly on the PL/laminin double-coated glass coverslips (10–20 microexplants per coverslip) with 50 μl of the culture medium. 1 or 2 h after the plating, they were transferred into a Petri dish (35 mm in diameter, Nunc), added with 1ml of the culture medium, and put in a CO2 incubator. A previous study (Nagata et al. 1986) showed that about 90% of cells were granule cells in these explants.

Scanning electron microscopy

Explants on coverslips were rinsed with 0·1 m-sodium cacodylate buffer (PH 7·3), fixed with 2·5% glutaraldehyde and subsequently with 1 % OsO4 both in the same buffer solution, dehydrated through ethanol series, and critical-point dried through liquid CO2. The samples were examined with a Hitachi S-800 scanning electron microscope after sputtercoating with platinum/palladium.

Time-lapse video recording

Cell behavior was recorded using a time-lapse video system (Professional Editing Recorder BR-8600, JVC controlled by Time-lapse Video Controller SIV, Sankei, Tokyo) equipped with a video camera (C1965, Hamamatsu Photonics) which was attached to an inverted microscope (Nikon Diaphoto TMD) equipped with a heated (37 °C) box using ×20 or ×40 DIC objective lenses. Single recording for 0·1 s was repeated at 30 s intervals.

Small granule cell neurons started migration out from the microexplants by one day of culture. Their neurites formed radially oriented parallel bundles after 2 days (Fig. 1A). Their cell bodies were attached to these radial neurites (Fig. IB, C). Time-lapse video recording showed that they moved along the radial neurites apparently propelled by an active leading end with many motile filopodia (Fig. 2). Direction of the migration was mostly away from the explant, but occasional cells moved toward the explant (Fig. 3A).

Fig. 1.

Phase contrast light micrographs (A,B,D,E) and scanning electron micrographs (C,F) of the granule cells and radial neurite bundles growing from the microexplant (postnatal day 2), which is located in the right-bottom direction of each photograph. A-C show earlier period, after 2 days of culture, when the granule cells are migrating away from the explant along the radial neurites. D-F show later period, after 4 (D,E) or 5 (F) days of culture, when the granule cells have changed their orientation by extending cell processes and moving perpendicular to the radial neurites. Arrows in F indicate growthcone-like structures with short filopodia, which are present not only at tips but also middle of the cell processes. Scale bars indicate 50 μm (A,B,D,E) or 10 μm (C,F).

Fig. 1.

Phase contrast light micrographs (A,B,D,E) and scanning electron micrographs (C,F) of the granule cells and radial neurite bundles growing from the microexplant (postnatal day 2), which is located in the right-bottom direction of each photograph. A-C show earlier period, after 2 days of culture, when the granule cells are migrating away from the explant along the radial neurites. D-F show later period, after 4 (D,E) or 5 (F) days of culture, when the granule cells have changed their orientation by extending cell processes and moving perpendicular to the radial neurites. Arrows in F indicate growthcone-like structures with short filopodia, which are present not only at tips but also middle of the cell processes. Scale bars indicate 50 μm (A,B,D,E) or 10 μm (C,F).

Fig. 2.

Photographic prints made from time-lapse video recording of the radial migration of the granule cell neurons. The explant is located beyond the left-hand edge of the frame. A few cells are labelled with numbers. They show movement to the right direction, which is away from the explant. Most cells migrate along the preexisting radial neurites of other cells. The actively locomoting cell possesses a growth-cone-like leading end with many motile filopodia (arrows). Time is shown in the upper-right corner of each print in hour, minute and second. A scale bar indicates 20 μm.

Fig. 2.

Photographic prints made from time-lapse video recording of the radial migration of the granule cell neurons. The explant is located beyond the left-hand edge of the frame. A few cells are labelled with numbers. They show movement to the right direction, which is away from the explant. Most cells migrate along the preexisting radial neurites of other cells. The actively locomoting cell possesses a growth-cone-like leading end with many motile filopodia (arrows). Time is shown in the upper-right corner of each print in hour, minute and second. A scale bar indicates 20 μm.

Fig. 3.

Trajectory of the granule cells traced from timelapse video recording by marking approximate center of the cell body at every hour for 20 h. Solid circles indicate starting points. Broken lines in the upper-right corner indicate profile of the explant, which was dissected out from a mouse on postnatal day 5. Dotted lines indicate orientation of the radial neurite bundles. A shows cell migration during earlier period, after 1–2 days of culture, when the granule cells move along the radial neurite bundle. B shows later period of the same explant, after 4 days of culture, when the granule cells show translocation perpendicular to the radial neurites. A cell in the upper-left corner made 3 perpendicular turns during its migration, thus took an alternative radial or perpendicular direction relative to the radial neurites. Such change of direction was frequently observed. Scale bars indicate 100 μm.

Fig. 3.

Trajectory of the granule cells traced from timelapse video recording by marking approximate center of the cell body at every hour for 20 h. Solid circles indicate starting points. Broken lines in the upper-right corner indicate profile of the explant, which was dissected out from a mouse on postnatal day 5. Dotted lines indicate orientation of the radial neurite bundles. A shows cell migration during earlier period, after 1–2 days of culture, when the granule cells move along the radial neurite bundle. B shows later period of the same explant, after 4 days of culture, when the granule cells show translocation perpendicular to the radial neurites. A cell in the upper-left corner made 3 perpendicular turns during its migration, thus took an alternative radial or perpendicular direction relative to the radial neurites. Such change of direction was frequently observed. Scale bars indicate 100 μm.

After 3 or 4 days of culture, the small granule cell neurons started to extend new cell processes at right angles and, soon, most cells assumed perpendicular orientation to the radial neurite bundle (Fig. 1D-F). Time-lapse video recording of this later culture period showed that the cell body frequently translocated in the direction perpendicular to the radial neurites (Fig. 3B). Video recording and scanning electron microscopy showed that the perpendicular cell processes were apparently in direct contact to the radial neurites (Figs 1F, 4). Many active filopodia are apparently making contact to the radial neurites not only at tips of the perpendicular cell processes but also along their shafts (Figs 1F, 4). We found no other cellular or extracellular structures arranged perpendicular to the radial neurites (Fig. 1F), which would have provided conventional contact guidance to the granule cells. We measured the angle made by the granule cell process and the radial neurites around the cell to estimate how strict is the perpendicularity. Results (Fig. 5) indicate strikingly narrow distribution whose peak is very close to 90°.

Fig. 4.

Photographic prints made from the time-lapse video recording of the perpendicular migration of a granule cell (A-E) and the perpendicular extension of cell processes (F-K). In both cases, the tip of cell processes have a growth-conelike structure with many motile filopodia (arrows). Frequently, active filopodia appeared not only at tips but also along the shaft of the cell process (arrowheads), and they make contact with the radial neurites bundle beneath them. The explants are located to the upper direction. Scale bars indicate 20 μm. Time-lapse of B-E from A are 48 min, 4h 39 min, 8h 44min, and 9h 4min. Time-lapse of G-K from F are 37min, 1 h 54min, 3h 10min, 3h 56min, and 4h 31 min.

Fig. 4.

Photographic prints made from the time-lapse video recording of the perpendicular migration of a granule cell (A-E) and the perpendicular extension of cell processes (F-K). In both cases, the tip of cell processes have a growth-conelike structure with many motile filopodia (arrows). Frequently, active filopodia appeared not only at tips but also along the shaft of the cell process (arrowheads), and they make contact with the radial neurites bundle beneath them. The explants are located to the upper direction. Scale bars indicate 20 μm. Time-lapse of B-E from A are 48 min, 4h 39 min, 8h 44min, and 9h 4min. Time-lapse of G-K from F are 37min, 1 h 54min, 3h 10min, 3h 56min, and 4h 31 min.

Fig. 5.

A histogram showing distribution of angles made by the granule cell process and the radial neurite bundle around the cell. The angle was measured clockwise from the axis of the neurite bundle away from the explant. This histogram shows data obtained from the explant shown in Fig. 1D. The distribution is strikingly narrow and centered at 90° (mean and standard deviation are 90·4 ± 16·5 degree, n = 98).

Fig. 5.

A histogram showing distribution of angles made by the granule cell process and the radial neurite bundle around the cell. The angle was measured clockwise from the axis of the neurite bundle away from the explant. This histogram shows data obtained from the explant shown in Fig. 1D. The distribution is strikingly narrow and centered at 90° (mean and standard deviation are 90·4 ± 16·5 degree, n = 98).

The following observations strongly suggest that the granule cell neurons are not reacting to a gradient of diffusible factors produced by the explant. Instead, they are reacting to the topographical arrangement of the radial neurites in the vicinity of the cell. First, even if the explant was removed from the culture before the orientation change, a perpendicular orientation was created if parallel neurite bundles remained. Second, in the case where the radial neurite bundle was bent in the middle of outgrowth, granule cell neurons assumed perpendicularity relative to the bent parallel neurites around the cell, but not relative to the direction toward the explant. Finally, orientation of the granule cell neurons was random in the area where the radial neurites were disordered and did not make a parallel bundle, even if such area was small and surrounded by areas of the parallel neurites and strict perpendicularity of the granule cells.

The change from typical contact guidance into paradoxical perpendicular orientation cannot be explained by simple loss of adhesiveness of the granule cells to the radial neurites, because it would have produced disordered random orientation. Also, we think it is very unlikely that the granule cell neurons are reacting to a gradient of diffusible factors produced by the explant as described before. Therefore, we propose that the perpendicular orientation described in this study is a novel type of the contact-guided phenomenon. In addition, immunostaining with anti-glial fibrillary acidic protein antibodies showed that glial cell processes remained in close vicinity to the explant in such cultures (Nagata & Nakatsuji, 1989), and we are certain that there was no such glial process in the area of the orientation change analyzed here.

It is not clear how such paradoxical cell behavior is brought about by motile machinery of the cell. Drastic change of the behavior suggests that there are changes in the cytoskeletons and their regulation. Our only observation that might be related to such change is that the growth cone at the tip of neurites seems to be dominating during the earlier radial migration, while many active filopodia appear along the neurite shaft and attach to the parallel neurite bundle during the later perpendicular movement. It would be interesting to examine distribution pattern of the cytoskeletal protein molecules in these cells before and after the orientation change.

The perpendicular process formation and migration by the granule cells in vitro seem to correspond to the vertical inward migration of the granule cells across the parallel fiber bundle in the cerebellum. The present study suggests that the granule cells acquire a tendency to orientate themselves perpendicular to the parallel fibers during histogenesis of the cerebellum. Such behaviour could provide another mechanism to ensure vertical migration of the granule cells in addition to the guidance by the Bergmann glia (Rakic, 1971; Edmondson & Hatten, 1987; Gregory et al. 1988).

Guidance by the radial glia is also believed to play the central role in radial migration of the newly born neurons to form columnar structure in cortices of the central nervous system (Rakic, 1988; Gray et al. 1988). However, some of the very recent studies on the chick and rat forebrain produced rather surprising evidence of tangential migration of the neurons in addition to the radial migration (Balaban et al. 1988; Walsh & Cepko, 1988). The perpendicular contact guidance described in this study may be working in such cases by providing a novel type of cell behavior in histogenesis of the central nervous system.

The authors are grateful to Dr M. Schachner for helpful discussion.

Balaban
,
E.
,
Teillet
,
M.-A.
&
Le Douarin
,
N.
(
1988
).
Application of the quail-chick chimera system to the study of brain development and behavior
.
Science
241
,
1339
1342
.
Dunn
,
G. A.
(
1982
).
Contact guidance of cultured tissue cells: a survey of potentially relevant properties of the substratum
.
In Cell Behaviour
(ed.
R.
Bellairs
,
A.
Curtis
&
G.
Dunn
), pp.
247
280
.
Cambridge
:
Cambridge Univ. Press
.
Edmondson
,
J. C.
&
Hatten
,
M. E.
(
1987
).
Glial-guided granule neuron migration in vitro: A high-resolution time-lapse video microscopic study
.
J. Neurosci
.
7
,
1928
1934
.
Fischer
,
G.
(
1982
).
Selection of mouse cerebellar cells in serum free, hormonally defined media: Survival of neurons
.
Neurosci. Lett
.
28
,
325
329
.
Fisher
,
G.
,
Kunemund
,
V.
&
Schachner
,
M.
(
1986
).
Neurite outgrowth patterns in cerebellar microexplant cultures are affected by antibodies to the cell surface glycoprotein LI
.
J. Neurosci
.
6
,
605
612
.
Gray
,
G. E.
,
Glover
,
J. C.
,
Majors
,
J.
&
Sanes
,
J. R.
(
1988
).
Radial arrangement of clonally related cells in the chicken optic tectum: Lineage analysis with a recombinant retrovirus
.
Proc, natn. Acad. Sci. U.S.A
.
85
,
7356
7360
.
Gregory
,
W. A.
,
Edmondson
,
J. C.
,
Hatten
,
M. E.
&
Mason
,
C. A.
(
1988
).
Cytology and neuron-glial apposition of migrating cerebellar granule cells in vitro
.
J. Neurosci
.
8
,
1728
1738
.
Hynes
,
R. O.
,
Patel
,
R.
&
Miller
,
R. H.
(
1986
).
Migration of neuroblasts along preexisting axonal tracts during prenatal cerebellar development
.
J. Neurosci.
6
,
867
876
.
Nagata
,
I.
,
Keilhauer
,
G.
&
Schachner
,
M.
(
1986
).
Neuronal influence on antigenic marker profile, cell shape and proliferation of cultured astrocytes obtained by microdissection of distinct layers from the early postnatal mouse cerebellum
.
Dev. Brain Res
.
24
,
217
232
.
Nagata
,
I.
&
Nakatsuji
,
N.
(
1989
).
A novel granule cell behavior on laminin in cerebellar microexplant cultures
.
Submitted for publication
.
Nakatsuji
,
N.
(
1984
).
Cell locomotion and contact guidance in amphibian gastrulation
.
Am. Zool
.
24
,
615
627
.
Nakatsuji
,
N.
&
Johnson
,
K. E.
(
1984
).
Experimental manipulation of a contact guidance system in amphibian gastrulation by mechanical tension
.
Nature, Lond
.
307
,
453
455
.
Rakic
,
P.
(
1971
).
Neuron-glia relationship during granule cell migration in developing cerebellar cortex. A Golgi and electronmicroscopic study in Macacus rhesus
.
J. comp. Neurol
.
141
,
283
312
.
Rakic
,
P.
(
1988
).
Specification of cerebral cortical areas
.
Science
241
,
170
176
.
Schnitzer
,
J.
&
Schachner
,
M.
(
1981
).
Expression of Thy-1, H-2 and NS-4 cell surface antigens and tetanus toxin receptors in early postnatal and adult mouse cerebellum
.
J. Neuroimmunol
.
1
,
429
456
.
Silver
,
J.
&
Ogawa
,
M. Y.
(
1983
).
Postnatally induced formation of the corpus callosum in acallosal mice on glia-coated cellulose bridges
.
Science
220
,
1067
1069
.
Silver
,
J.
&
Sidman
,
R. L.
(
1980
).
A mechanism for the guidance and topographic patterning of retinal ganglion cell axons
.
J. comp. Neurol
.
189
,
101
111
.
Singer
,
M.
,
Nordlander
,
R. H.
&
Egar
,
M.
(
1979
).
Axonal guidance during embryogenesis and regeneration in the spinal cord of the newt: The blueprint hypothesis of neuronal pathway patterning
.
J. comp. Neurol
.
185
,
1
22
.
Walsh
,
C.
&
Cepko
,
C. L.
(
1988
).
Clonally related cortical cells show several migration patterns
.
Science
241
,
1342
1345
.
Wood
,
A. T.
&
Thorogood
,
P.
(
1984
).
An analysis of in vivo cell migration during teleost fin morphogenesis
.
J. Cell Sci
.
66
,
205
222
.