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

First published online May 11, 2006
doi: 10.1242/10.1242/dev.02379

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 Andrews, W. Articles by Richards, L. J. Search for Related Content PubMed PubMed Citation Articles by Andrews, W. Articles by Richards, L. J. Social Bookmarking                         What's this?

Robo1 regulates the development of major axon tracts and interneuron migration in the forebrain

William Andrews^1,2,*, Anastasia Liapi^1,*, Céline Plachez^3,*, Laura Camurri², Jiangyang Zhang⁴, Susumu Mori^4,5, Fujio Murakami⁶, John G. Parnavelas¹, Vasi Sundaresan^7, and Linda J. Richards^3,8,

¹ Department of Anatomy and Developmental Biology, University College London, London WC1E 6BT, UK.
² MRC Centre for Developmental Neurobiology, King's College London, London SE1 1UL, UK.
³ The University of Maryland School of Medicine, Department of Anatomy and Neurobiology, and The Program in Neuroscience, Baltimore, MD 21201, USA.
⁴ Johns Hopkins University School of Medicine, Department of Radiology, Division of NMR Research and Department of Biomedical Engineering, Baltimore, MD 21205, USA.
⁵ F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD 21205, USA.
⁶ Osaka University, Osaka, Japan.
⁷ Department of Cellular Pathology, St Margarets Hospital, The Plain, Epping CM16 6TN, UK.
⁸ The University of Queensland, School of Biomedical Sciences and The Queensland Brain Institute, Brisbane, Queensland, Australia.

View larger version (73K): [in a new window]   Fig. 1. Generation of conditional Robo1 knockout mice. (A) Exon structure (depicted by a square box) corresponding to Ig domains 2b, 3a and 3b (exons 4 to 6, respectively, indicated in red). The position of the terminal nucleotides within an amino acid codon is indicated above the exon in red. The yellow triangles shown in the upper panel represent lox P sites cloned into unique XbaI and KpnI sites (loxP sites flanking a neo cassette, not shown in diagram). (B,C) Genotype analysis of tail DNA obtained from one F2 litter. Allele sizes are discriminated by the XbaI site that was `silenced' by the cloning of the loxP site between exons 4 and 5 and probed with a genomic probe including exon 4 (B) or by PCR across exons 4 and 6 (C). (D) RT-PCR across exons 3 and 7 on mRNA isolated from wild-type and homozygous brains also confirmed that the deletion had occurred in the mutants. (E) The mRNA sequence of the RT-PCR products that were cloned into a TA vector. The lower panel (mutant) indicates the presence of multiple premature stop sites within the protein. (F) In situ hybridization was performed using probes against Robo1, Robo2 and Robo3. Robo1 mRNA was absent, but Robo2 and Robo3 mRNA were still expressed in the mutant. (G) Similarly, Robo1 protein was not expressed in the mutant as indicated by western blot analysis.

View larger version (125K): [in a new window]   Fig. 2. Robo1 and Robo2 expression in the forebrain. Robo1 and Robo2 expression was examined using antibodies specific for each receptor. (A-D) In rostral sections, Robo1 protein was highly expressed on callosal axons at E17 (A, coronal view; arrow in C; C is a higher power view of A), whereas Robo2 staining is very faint (B, coronal view at E17; arrow in D; D is a higher power view of B). (E,F) In more caudal regions, Robo1 is expressed on the hippocampal commissure (HC) and on the corpus callosum (CC) (E, horizontal view at E17); Robo2 is expressed on the nigrostriatal pathway (NSP), the internal capsule (IC) and the optic tract (OT) (F, coronal view at E17). Scale bar in A: 200 µm for A,B,E,F; in C, 100 µm for C,D.

View larger version (102K): [in a new window]   Fig. 3. Magnetic resonance diffusion tensor imaging of Robo1 knockout brains. To survey the entire brain for axonal tract defects, we scanned Robo1 and control littermate brains in a 9.4 Tesla magnet. In these images, orientations of axonal tracts are pseudo-coloured according to their orientation within the brain. Axonal tracts projecting mediolaterally, such as commissures, are green; tracts projecting in the dorsoventral axis are blue; and tracts in the rostrocaudal axis are red. Brains from wild-type (A-C), heterozygote (D-F) and Robo1 knockout (G-I) littermates were analyzed at E17. (A'-I') Higher power views of the boxed regions in A-I, respectively. Three planes of section were extracted from the 3D data and are shown at the level of the rostral corpus callosum: horizontal (A,D,G); coronal (B,E,H) and sagittal (C,F,I). The corpus callosum is indicated by the arrows in the wild-type brain in A'-C' and in the knockout in G'-I'. The anterior commissure is observed as a green spot in cross-section and indicated by the arrowheads in C',F',I'. From this analysis, it is evident that the corpus callosum (arrow in A') and hippocampal commissure (arrowhead in A') are greatly reduced in the Robo1 knockout (compare A' with G', and C' with I') and that the orientation of the fibres has changed from mediolaterally (green) to dorsoventrally (blue) projecting. Scale bar: 650 µm for A,D,G; 400 µm for B,E,F; 500 µm for C,F,I; 250 µm for B',C',E',F',H',I'; 350 µm for A',D',G'.

View larger version (154K): [in a new window]   Fig. 4. Robo1 knockout mice display aberrant axonal pathfinding in the corpus callosum and hippocampal commissure. (A-H) Coronal brain sections taken at the rostral level of littermates of Robo1 mice were labelled with L1-CAM immunohistochemistry to reveal axonal projections in the corpus callosum (CC) and fornix/hippocampal commissure at E17.5. In wild-type (A,B,E,F) and heterozygous (C,D,G,H) mice, both callosal and hippocampal projections appeared normal. However, in knockout littermates, medially projecting axons formed tight fascicles that failed to cross the midline (white arrows in K,K',L,L'). Sections of the same brain from rostral (I,K) to caudal (J,L) indicated that axons along the rostrocaudal extent of these commissures were disrupted (I-L). Some axons still crossed the midline (asterisks in L,L'), indicating that not all callosal axons were affected. E-H and K,L are higher power images of A-D and I,J, respectively. K' and L' are higher power views of the boxed regions in K and L, respectively. Scale bars: in J, 200 µm for A-D,I,J; in L, 100 µm for E-H,K,L and 50 µm for K',L'.

View larger version (64K): [in a new window]   Fig. 5. Tract tracing analysis reveals major pathfinding defects in callosal and hippocampal axons. (A-O) Dual tract tracing was performed at E17.5 by labelling the callosal projection with DiI in the medial cortex (red label in the figure) and labelling the hippocampal projection through the fornix and hippocampal commissure with DiA in the dentate gyrus (green label in the figure). Coronal sections were counterstained with DAPI (A,F,K). Brains were analysed from rostral to caudal at the level of the corpus callosum and hippocampal commissure. (There is no green labelling in C as this section is rostral to the fornix and HC.) In Robo1 knockout mice, callosal (B, arrowhead; G and L, arrows) and hippocampal (arrowheads in H and M) axons formed tight fascicles that did not cross the midline. Furthermore, fascicles from both the cortex and hippocampal projections overlapped at the midline (knockout overlay images are in I and N), whereas in wild-type mice these projections remained completely separate (E,J,O). As observed by immunohistochemical analysis, tract tracing analysis revealed axons crossing the midline in the rostral region of this Robo1 knockout brain (arrow in B; D, overlay). Scale bar: 200 µm.

View larger version (119K): [in a new window]   Fig. 6. Midline glial populations are present in Robo1 knockout mice. (A-D) To assess the development of midline glial structures in Robo1 knockout (C,D) or control (A,B) mice, E17.5 brains were sectioned coronally and labelled by glial fibrillary acidic protein (GFAP) immunohistochemistry. Three midline glial populations were present: the glial wedge (GW), glia within the indusium griseum (arrow labelled IGG in B and D) and the midline zipper glia (MZG) in both wild-type and Robo1 knockout brains. Some abnormalities were noted in the MZG but this is probably due to the morphological disruption within the area by the formation of large axon fascicles and the lack of a definitive corpus callosum (arrow labelled MZG in D). Scale bar: 400 µm in A,C; 200 µm in B,D.

View larger version (84K): [in a new window]   Fig. 7. Thalamocortical and corticofugal axons advance faster towards their respective targets in Robo1 knockout brains. DiI was placed in the dorsal thalamus to label thalamocortical axons (red in all panels and in schematic diagram) and DiA was placed in the presumptive somatosensory cortex of the same brains to label corticofugal axons (green in all panels and in schematic diagram) at either E14.5 (A-E) or E18.5 (F-M). Robo1 knockout brains (A-C,F-I) were compared with controls (D,E,J-M) in coronal sections of middle and caudal regions. At E14.5, thalamocortical axons in the knockout (A,B; arrow in B indicates retrograde labelling of cortical cells) advanced further into the cortex than in wild-type littermates (D,E) which had not yet passed the corticostriatal notch. Similarly, corticofugal axons had already reached the thalamus in E14.5 Robo1 knockout brains (A), but had not yet entered the striatum in controls (D,E). (B,C) Higher power views of the boxed regions in A (red and blue channels only). At E14.5 and E18.5, we observed an aberrant projection of DiI-labelled axons coursing transversely over the axons of the internal capsule in Robo1 knockout brains in a `knot'-like structure (arrows in C and H), but not in controls (D and arrow in L). As observed in Fig. 5, most callosal axons (labelled here with DiA) projected aberrantly into the septum of Robo1 knockouts (F, arrow) compared with controls (J, arrow). (G,K) Higher power views of the boxed regions in F,J (red channel only). The advance of thalamocortical axons persisted at E18.5 in the Robo1 knockout, where these axons projected further medially into the cortex (G, arrow) compared with controls (K, arrow). Furthermore, back-labelled cells appeared in greater numbers in the thalamus of mutants following placement of dye in the cortex (compare I with M, arrow), further confirming that thalamocortical axons had arrived earlier in these brains. Scale bars: in A, 400 µm in A,D; in B, 200 µm in B,C; in F, 400 µm in F,H,I,J,L,M; in G, 200 µm in G,K.

View larger version (102K): [in a new window]   Fig. 8. Robo1, calbindin and GAD65 immunohistochemistry of cells from the ganglionic eminence. (A) Image of a wild-type coronal section showing robust Robo1 staining in the mantle zone of the ganglionic eminence (GE) and in two bands in the marginal zone (MZ) and lower intermediate zone (IZ) of the developing cortex. At high magnification, the receptor was clearly localized in some individual cells, especially near the corticostriatal notch, that showed features of migrating neurons (boxed area is shown at higher magnification in A'). (B) Similar to Robo1 staining, calbindin immunoreactivity was localized chiefly in the GE, and in the corridors within the cortex used by migrating cortical interneurons, the MZ and lower IZ. (C-E) Dissociated GE cell cultures prepared from an E15 wild-type animals were co-stained with Robo1 (C) and GAD65 (D), showing that GABAergic cells in this part of the ventral telencephalon express the receptor (E, a composite of C and D). Scale bars: 200 µm in A,B; in E, 10 µm in A' and 35 µm in C-E.

View larger version (81K): [in a new window]   Fig. 9. Cell migration defects in the forebrain of Robo1 knockout brains. (A) Quantification of the total number of calbindin labelled cells that entered the cortex of control and Robo1 knockout brains at E12.5 showed that the latter contained roughly twice as many cells (*P<0.001, Student's t-test; error bars represent s.e.m.). (B) Histogram depicts the average number of calbindin cells in a 200 µm wide radial strip of dorsomedial cortex (DMC) in control and Robo1 knockout brain sections at E18.5. More calbindin-positive cells were observed in the knockout cortex (*P<0.01, Student's t-test; error bars represent s.e.m.). (C) At E18.5, the average number of calbindin cells in rostral and middle regions (200 µm wide strip) of DMC is higher in Robo1 knockout brains compared with control (*P<0.01 for both rostral and middle regions, Student's t-test; error bars represent s.e.m.), whereas no difference was observed in caudal regions. Calbindin staining in representative coronal sections taken from rostral cortex at E12.5 (D,E) and middle (parietal) cortex at E18.5 (F-G') of control (D,F) and Robo1 knockout (E,G) mouse brains. The cortex of the knockout brains appeared to contain a greater number of calbindin-labelled cells. The arrows in F,G indicate the striatal region, shown at higher magnification at F' and G', which is populated by calbindin cells in Robo1 knockout brains (G,G'), but not in controls (F,F'). Scale bars: in E, 200 µm in D,E; in G', 400 µm in F,G and 200 µm in F',G'.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter