QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Polleux, F. Articles by Ghosh, A. Search for Related Content PubMed PubMed Citation Articles by Polleux, F. Articles by Ghosh, A.

Control of cortical interneuron migration by neurotrophins and PI3-kinase signaling

¹ INSERM U.371, 18 avenue Doyen Lépine, 69675 Bron, France
² Department of Neuroscience, Johns Hopkins University School of Medicine, 725 N Wolfe St, Baltimore, MD 21205, USA
³ Department of Biomedical Sciences, University of Edinburgh, Hugh Robson Building, Georges Square, Edinburgh EH8 9XD, UK

View larger version (96K): [in a new window]   Fig. 1. Tangentially migrating cells from the ganglionic eminence invade the neocortex and CA fields of the hippocampus, but not the dentate gyrus. (A) A coronal section through the developing telencephalon showing the locations of the medial (M) and lateral (L) ganglionic eminences, cortex and hippocampus (Hipp.) during embryonic development. (B) Low magnification photomicrograph of the relative positions of the GE from GFP-expressing mice (GE-GFP), cortex and hippocampus in a co-culture assay. The white label in the cortex indicates the location of GE-GFP cells that have migrated from the GE to the cortex. Note the extensive invasion of the cortex and hippocampus by the tangentially migrating cells. (C,D) Analysis of the spatial distribution of E15 GE-GFP cells that have migrated into the cortex after 20 hours in vitro indicates that GFP cells are found mainly in the intermediate zone (IZ) and the marginal zone (MZ) of co-cultured cortical slices. In these zones, cells are mainly oriented tangentially. Note that some GFP cells have already invaded the cortical plate (VI, CP). Cells that enter the CP from the IZ typically have leading processes directed towards the pial surface (arrowhead, C), while those entering the CP from the MZ have leading processes directed away from the pial surface (arrowhead, D), indicating that cell bodies follow the leading process as the cells migrate into the CP. The dorsal (D) and lateral (L) aspects of the cortical slice are indicated in C. In (C,D), the GE explants are located to the right of the cortical explant shown. (E) By 36 hours in vitro, large numbers of E16 GE-GFP cells have migrated into the cortex, most of which travel through the IZ. Progressively more cells are found in the CP but they no longer show a clear radial orientation in the CP, suggesting that cells alter their trajectories of migration after entering the CP. The broken line in E indicates the edge of the slice on the ventricular side. In C-E, the red channel correspond to MAP2 immunofluorescence. (F,G) Examples of the morphology of tangentially migrating cells. The cells typically have a leading process that is 10-15 times the cell body diameter. The leading process is always tipped by a prominent growth cone (arrowheads) and often contains multiple filopodia (arrow in F). Many of the migrating cells also have branched leading processes (G), which may participate in the mechanics of altering trajectories of migration (see also Fig. 5). (H) In GE-cortex co-cultures, E16 GE-GFP cells migrate tangentially up to the most medial aspect of the telencephalon by 3 days in vitro where they accumulate in the developing hippocampus but avoid the developing dentate gyrus (DG). Cells in the marginal zone (star) of the putative CA regions (pCA) stop migrating sharply at the interface between the pCA and DG (arrow) identified as being TuJ1 negative (red channel). (I-K) The region of the developing hippocampus avoided by E16 GE-GFP cells (green channel, I) is a highly proliferative zone (delineated by the broken line in K; red channel, BrdU), which is typical of the DG anlage. (L-N) The avoidance of the developing DG by GE-GFP cells can be observed even when an explant of E16 MGE-GFP (to the right in L) is placed directly adjacent to the DG for 3 days in vitro. Higher magnification images show a marked difference in GFP cell density in DG (M) and pCA (N) regions located 250 µm away from the interface with the MGE explant. (O) GE-GFP cells do not migrate into wild-type dorsal thalamus (WT-DT) slices in a co-culture assay, indicating that GE cells are selective about their target zones of migration. All panels in this and other figures are from isochronic co-cultures, unless otherwise indicated. CP: cortical plate; SP: subplate; IZ: intermediate zone; VZ: ventricular zone; DG: developing dentate gyrus; pCA: putative CA regions; VI: cortical layer VI. Scale bars: 150 µm in C-E; 30 µm in F,G; 250 µm in H; 350 µm in I-K; 300 µm in L; 75 µm in M,N; 200 µm in O.

View larger version (110K): [in a new window]   Fig. 2. The majority of cells migrating tangentially into the cortex are generated in the MGE. (A,B) Comparison of the migration of GFP-positive cells from LGE-GFP (A) and MGE-GFP (B) explants into co-cultured P2 cortical slices at 3 days in vitro. Note that many more GFP-positive cells migrate into the cortex from the MGE (B) compared with the LGE (A). (C-H) E14 MGE explants (C-E) or E14 LGE explants (F-H) were exposed to BrdU in vitro for 8 hours before being co-cultured with P2 cortex for 3 days without BrdU. GFP-positive cells migrating from the MGE were five times more frequently double labeled for BrdU than were GFP-positive cells migrating from the LGE (red arrowheads in C-H; see text for quantification), indicating that the majority of GE-derived cells that migrate into the cortex originate in the MGE. (I) MGE-derived cells do not proliferate after migrating to the cortex. E14 GFP-expressing MGE explants were co-cultured with P2 cortex for 48 hours and pulsed labeled with BrdU for 4 hours just before fixation. No GFP-positive cells in the cortex were labeled with BrdU, indicating that MGE-derived cells migrate into the cortex after their final mitosis. Scale bars: 600 µm in A,B; 100 µm in C-H; 120 µm in I.

View larger version (164K): [in a new window]   Fig. 3. Fate of MGE-derived tangentially migrating cells. (A-C) After 2 days in vitro, about 85% of E14 MGE-GFP cells (green, B) that migrate into the intermediate zone (IZ) express the neuronal marker MAP2 (A, red in C). The arrow in C delineates the border between the cortical plate (CP) and the IZ. (D-F) After 3 days in vitro, about 65% of E14 MGE-GFP cells (E, green in F) migrating in the IZ express the nuclear neuronal marker NeuN (D, red in F). (G-I) After 7 days in vitro, about 35% of E14 MGE-GFP cells (H, green in I) found in the cortex express the neurotransmitter GABA (G, red in I). Blue arrowheads indicate examples of double-labeled cells. Scale bars: 150 µm in A-C,G-I; 60 µm in D-F.

View larger version (172K): [in a new window]   Fig. 4. Cellular interactions of MGE-derived tangentially migrating neurons. (A,B) Relationship between the leading process of E14 GE-GFP cells and the radial glial fibers in the intermediate zone (red, nestin). Cells migrating tangentially in the intermediate zone make several contacts with radial glial fibers, whereas some of the cells invading the cortical plate that migrate radially (arrow in A, shown at higher magnification in B) make multiple close contacts with the nestin+ glial fibers (arrowheads in B). In each image, the dorsal (D) and lateral (L) aspects are indicated. (C) Example of a migrating E14 GE-GFP cell making a few close contacts with axons (red, neurofilament 165 kDa) within the cortical intermediate zone. The insets numbered 1 (red, NF 165 kDa), 2 (green, GFP) and 3 (1 and 2 merged) represent a X-Z orthogonal section taken at the level indicated by the arrowhead in C. The blue arrows (3) indicate the location of close contact between an axon (red) and the leading process (green) of the tangentially migrating cell shown in C. (D) Example of an E14 GE-GFP cell migrating radially in the cortical plate towards the pia. This GE-GFP neuron makes contacts with cortical apical dendrites, revealed by MAP2 immunofluorescence (red). The insets numbered 1 (red, MAP 2), 2 (green, GFP) and 3 (1 and 2 merged) represent a X-Z orthogonal section taken at the level indicated by the arrowhead in D. The blue arrows (3) indicate the location of a close contact between an apical dendrite and the leading process of the GFP+ cell shown in D. Scale bars: 80 µm in A; 20 µm in B,C; 30 µm in D.

View larger version (30K): [in a new window]   Fig. 5. Dynamics of tangentially migration MGE-GFP neurons. (A,B) Time-lapse analysis of MGE-GFP cells in isochronic co-cultures moving radially down from the MZ to the CP (A) and from the CP to the MZ (B). The position and morphology of the migrating cells is shown at 15 minute time intervals. Note that in each case, the cells pause at the interface between the MZ and the CP before crossing into the new zone. (C) Illustration of the kinetics of migration of an E15 GE-GFP cell within the MZ. Each circle represents the position of the cell body every 15 minutes and reveals that the cell body translocates according to a saltatory mode of migration. (D) Trajectories of 23 MGE-GFP cells migrating within the MZ. Twenty out of these 23 cells migrate coherently from the lateral (left-hand side) to medial (right-hand side) aspect of the cortex. One cell (#3) migrates mediolaterally, one cell (#1) migrates down into the CP and two cells (#13 and #2) migrate from the CP to the MZ. (E,F) Morphological dynamics of the leading process during a change in the direction of migration. These cells, migrating within the IZ, make 90° (E) or 180° (F) changes in their trajectory in less than 2 hours (time interval is indicated below each drawings). Note that in each case, the turn is not initiated by the growth cone of the leading process, but instead by a second leading process that emerges from the cell body. (G) Example of an MGE-GFP cells crossing from the IZ to the CP. Note that the cell initially migrating in the IZ pauses for a prolonged period (arrowhead) before changing its trajectory by 90° and entering the CP. The arrows indicate the direction of migration. Conventions as in C. (H) Trajectories of six GE-GFP cells migrating within the IZ. Note that most cells migrate from a lateral (left) to medial (right) direction. One cell can be seen changing trajectories to enter the CP. Scale bar: 50 µm in A-D,G-H; 100 µm in E,F.

View larger version (53K): [in a new window]   Fig. 6. BDNF and NT-4 regulate tangential migration of MGE-derived cells to the cortex. (A-C) E14 MGE-GFP explants were co-cultured with P2-P4 cortical slices for 1 day in the presence of control vehicle (0.1 M PBS, A), recombinant BDNF (50 ng/ml, B), or recombinant NT-4 (50 ng/ml, C) in the slice culture medium. Both BDNF and NT4 treatment lead to an increase the total number of MGE-GFP cells that migrate into the cortex. (D) Nuclear counterstaining (Hoechst 33258) reveals the cellular organization of the cortex in MGE-cortex co-cultures. (E) Quantification of the effects of NT4 on migration of MGE neurons into the cortex in MGE-cortex co-cultures (experiment shown in A,C). The migration index represents the density of GFP+ cells (number of cells per 104 µm2) found in 100 µm wide stripes of intermediate zone at indicated distances from the MGE-cortex interface (arrowhead on the inset). Note that NT4 treatment (blue triangles) leads to an increase in the number of cells that migrate into the cortex and in the maximum distance of migration. Quantification was performed on nine explants taken from three independent experiments. NT4 treatment showed a statistically significant increase of the migration index compared with the control according to an ANOVA test (P<0.0001). (F-K) Expression pattern of NT4 in the developing telencephalon. Coronal sections were processed for NT4 immunofluorescence (F,I) and counterstained with Hoechst (G,J). The double immunofluorescence images are shown in (H,K). (F-H) At low magnification, NT4 can be detected in the ventricular zone of the GE (double arrows in F), as well as in the cortical plate and marginal zone of the cortex (single arrow in F). (I-K) With a higher magnification image through the cortical intermediate zone, NT4 immunofluorescence can be detected in radially oriented processes (arrows in I), which are likely to be radial glial processes. (L) NT4 can acutely induce the motility of MGE-derived cells. Time-lapse video images of an E15 MGE-GFP cell located in the IZ of a P2 cortical slice. The cell was not moving for the first 60 minutes (red arrowhead) before the application of NT4. Application of NT4 (green arrowhead; 20 µl at 20 ng/ml) induces migration in the MGE-GFP cell. The arrowhead in each panel indicates the initial position of the cell body. Numbers at the bottom right indicate time in minutes. (M) Recombinant NT4 can acutely increase the rate of migration of MGE-derived cells. E14 MGE-GFP cells migrating in the IZ of P2 cortical slices were imaged by time-lapse video-microscopy for 70 minutes in the presence of control vehicle (PBS) and for 80 minutes after application of recombinant NT4 (20 ng/ml). The instantaneous rate of migration (µm per hour) of MGE-GFP cells (n=77) is plotted as a function of time. Quantification was performed on eight explants taken from two independent time-lapse video-microscopy experiments. Scale bar: 250 µm in A-C; 400 µm in D; 150 µm in F; 40 µm in L. *, P<0.05; **, P<0.003.

View larger version (59K): [in a new window]   Fig. 7. Localization and function of TrkB receptors in tangential migration. (A-C) Double immunofluorescence for GFP (A) and TrkB (B) indicates that E14 MGE-GFP cells migrating in the cortical intermediate zone express the TrkB receptor (merged image in C). Note that the TrkB receptor is localized to the leading processes (arrowheads in C) of MGE-GFP cells. (D-F) Effect of inhibiting Trk receptors on migration of GE-GFP cells. (D,E) Confocal images of E14 MGE-GFP cells migrating tangentially into the IZ of early postnatal cortex in the presence of control vehicle (DMSO 1:1000; D) or 50 nM of the Trk tyrosine kinase inhibitor K252a (E). (F) Quantification of the experiments shown in D,E from six slices taken from three independent experiments. Conventions are the same as in Fig. 6E. K252a treatment (red) significantly inhibits migration according to an ANOVA test (F=28.53; P<0.0001). (G-J) Analysis of tangential migration in wild-type and TrkB null mice. (G-H) Low magnification photographs of coronal sections of E15 wild-type (G) and TrkB–/– mice (H) stained for calbindin 28K. At this rostrocaudal level of section, the global cytoarchitecture of the telencephalon in the two genotypes is indistinguishable. (I-J) High magnification photographs of the lateral cortex of sections shown in G,H. Tangentially migrating calbindin+ cells are found mainly in the subventricular zone (SVZ), the upper part of the intermediate zone (IZ), the subplate (SP) as well as in the marginal zone (MZ) in both wild-type (I) and TrkB–/– (J) cortex. (K) Quantification of the number of calbindin+ cells per 200 µm wide radial column (indicated by double arrows in G,H). This analysis reveals a significant decrease in the number of tangentially migrating calbindin+ cells in TrkB–/– cortex compared with TrkB+/+ cortex at E15 [–32%; **P<0.005 – Mann-Whitney test; 12 sections from three wild-type and three knockout mice]. Scale bars: 15 µm in A-C; 250 µm in D,E, 400 µm in G,H; 120 µm in I,J.

View larger version (30K): [in a new window]   Fig. 8. Activation of PI3-kinase by neurotrophins and its role in tangential migration. (A) AKT phosphorylation in GE explants induced by BDNF or NT4 and visualized by western blotting with anti-phospho-AKT antibodies. Note that both BDNF and NT4 induce AKT phosphorylation (top panel), which is suppressed by preincubation with LY294002 (50 µM), a specific PI3-kinase inhibitor (lanes 1, 4 and 6). Lower panel indicates western analysis of the same blot with anti-AKT antibodies as a loading control. (B-E) Inhibition of PI3-kinase suppresses tangential migration of MGE-GFP cells. E15 MGE-GFP explants were co-cultured with P2-3 cortex for 24 hours in presence of control (B, DMSO), DMSO plus recombinant NT4 (50 ng/ml, C), or recombinant NT4 (50 ng/ml) plus 50 µM LY294002 (a PI3-kinase inhibitor, D). (E) Quantification of the results shown in B-D using the migration index as described in Fig. 6E (n=8 explants in each condition from four independent experiments). Note that inhibition of PI3-kinase with LY294002 treatment suppresses the NT4 effect, indicating that the effect of NT4 on migration requires PI3-kinase activation. Also indicated in green (full squares) is the quantification for treatment with LY294002 alone (ANOVA; F=0.623; not significantly different from LY294002+NT-4, see text). Comparison with Fig. 7F indicates that in LY294002-treated slices, migration is reduced to levels seen with K252a, consistent with the interpretation that endogenous neurotrophins regulate tangential migration via TrkB and PI3-kinase activation. (F) Quantification of the effect of inhibiting MEK1/2 activity (U0126, 10 µM) and inhibiting phospholipase C (U73122, 1 µM) on tangential migration of MGE-derived cells. Conventions are as in Fig. 8E. The migration index curves obtained for U0126 and U73122 do not differ significantly for the control (DMSO) curves [ANOVA test; F=1.021 and F=0.875, respectively]. Scale bar: 500 µm in B-D.