Role of EphA/ephrin--a signaling in the development of topographic maps in mouse corticothalamic projections

Abstract

Corticothalamic (CT) feedback outnumbers thalamocortical projections and regulates sensory information processing at the level of the thalamus. It is well established that EphA7, a member of EphA receptor family, is involved in the topographic mapping of CT projections. The aim of the present study was to dissect the precise impact of EphA7 on each step of CT growth. We used in utero electroporation-mediated EphA7 overexpression in developing somatosensory CT axons to dissect EphA7/ephrin-A-dependent mechanisms involved in regulating both initial targeting and postnatal growth of the CT projections. Our data revealed that topographic maps of cortical afferents in the ventrobasal complex and medial part of the posterior complex in the thalamus become discernible shortly after birth and are fully established by the second postnatal week. This process starts with the direct ingrowth of the CT axons to the designated areas within target thalamic nuclei and by progressive increase of axonal processes in the terminal zones. Large-scale overproduction and elimination of exuberant widespread axonal branches outside the target zone was not observed. Each developmental event was coordinated by spatially and temporally different responsiveness of CT axons to the ephrin-A gradient in thalamic nuclei, as well as by the matching levels of EphA7 in CT axons and ephrin-As in thalamic nuclei. These results support the concept that the topographic connections between the maps in the cerebral cortex and corresponding thalamic nuclei are genetically prespecified to a large extent, and established by precise spatiotemporal molecular mechanisms that involve the Eph family of genes.

Figure 1

Shift of CT projections in VB by the overexpression of EphA7 using in utero electroporation. A–D: Immunohistochemistry for EGFP (green, counterstained with DAPI [magenta]) at P1 in a brain that was electroporated with the EGFP-expression plasmid at E12.5. CT projections are labeled by EGFP from the somatosensory (SS) cortex, through the internal capsule (ic) to POm and VB in the thalamus. E, F: Images at POm and VB of control- (E) and EphA7-electroporated (F) brains at P12. CT axons in the control brain (E) distribute in discrete areas of VB and POm (arrows), whereas CT axons that overexpress EphA7 accumulate densely at the VB/POm boundary region (F). G: Fc-tagged EphA7 receptor binding at POm and VB at P4, revealing minimal binding of EphA7 at the POm/VB border region¹. H: Plot profile of the gray value quantified in POm and VB along the axis indicated by the arrow in G. Scale bars = 500 μm. ¹The image was taken from the same slide that was previously shown in Torii and Levitt, (2005).

Figure 2

Ingrowth of CT axons into VB and POm at P0. A, B: Immunohistochemistry for EGFP (counterstained with DAPI) at VB and POm in control-(A) and EphA7-electroporated (B) brains. Several EGFP-labeled CT axons with few side branches cross the VB/POm border (broken line) in both cases (arrows). Scale bar = 500 μm. C: High magnification view of the EGPF labeling in the area boxed in B, showing little branching from the main shaft of these axons. Arrows point the growth cones at the tip of growing axons. Scale bar = 50 μm.

Figure 3

Increase of CT axonal processes in VB and POm at P1–P2. A–F: Immunohistochemistry for EGFP (counterstained with DAPI in A, B, D, E) at VB and POm in control- (A–C) and EphA7-electroporated (D–F) brains at P1 (A, D) and P2 (B, C, E, F). C and F are higher magnification views of the yellow-boxed areas in B and E, respectively. At P1, in contrast to further growth of CT axons into the body of POm in the control-electroporated brain (A, arrow), EphA7-overexpressing axons appear to be stalled in close proximity of the POm/VB border region (D, arrow). The insets in A and D are higher magnification views of the square areas in each panel, showing the branching of CT axons. At P2, some EGFP⁺ CT axons in control-electroporated brains reach the dorsomedial border of POm (arrow in B), and the punctuate labeling in the body of POm and VB is increased (B, C). The inset in B is a higher magnification view of the square area, showing the puncta of labeled CT axons in the body of VB. In the EphA7-electroporated brain, EGFP⁺ axonal puncta were significantly increased in the vicinity of the POm/VB border (arrow in E and F), but not in the body of the nuclei (arrowheads in F). The bundles of main axon shafts in VB are similarly observed in both control- (C, brackets) and EphA7-electroporated brains (F, brackets). Scale bars = 500 μm (A, B, D, E), 100 μm (C, F).

Figure 4

Robust increase of CT axonal processes at the terminal zones in VB and POm at P12–P40. A-J: Immunohistochemistry for EGFP (counterstained with DAPI in E, J) at VB and POm in control- (A–E) and EphA7-electroporated (F–J) brains at P12 (A–D, F–I) and P40 (E, J). B, C and G, H are higher magnification views of the red- and yellow-boxed areas in A and F, respectively. D and I are higher magnification views of the boxed areas in C and H, respectively. At P12, labeling of EGFP⁺ axons is extensive at their specific terminal zones in VB and POm in the control-electroporated brain (arrows in A), whereas it is restricted to the VB/POm border region in the EphA7-electroporated brain (F). Punctate labeling of CT axonal processes accumulates densely around the VB/POm border region (arrow in G), with much less labeling in the body of VB and POm (arrowheads in G, and H) in the EphA7-electroporated brain. A wider distribution of CT axonal processes is observed in the control-electroporated brains (B, C). In the EphA7-electroporated brain, bundles of main axonal shafts appeared similar to control (brackets in D and I). However many fewer axonal branches and terminals surrounded the shafts (I, compare with D). Similar patterns of CT innervation in discrete areas of VB and POm in the control (E, arrows) and at the VB/POm border region in the EphA7-electroporated brain (J) were observed in more mature brains at P40. Scale bars = 500 μm (A, E, F, J), 100 μm (B, C, G, H), 20 μm (D, I).

Figure 5

Continuous increase of CT axonal processes at the terminal zones in VB and POm. A: Schema to quantify the distribution of CT axons in VB and POm. B: The percentage distribution of area occupied by EGFP⁺ CT axons (y axis) is represented against relative position across POm and VB (x axis, as indicated by the arrow in A). In contrast to the distribution in control-electroporated brains, the accumulation of EGFP⁺ axons at the POm/VB border region is evident as early as P2. This restricted distribution is maintained during the increasing of CT axonal processes through the postnatal period. n = 4 and 4 (P2), 5 and 6 (P12), and 4 and 6 (P40) for control- and EphA7-electroporated brains, respectively. Error bars represent SEM.

Figure 6

Models of CT axon mapping in VB and POm. CT axons (green) in both control- and EphA7-electroporated brains invade VB around P0 from the ventrolateral margin without showing the expected repulsion to the high level of ephrin-As (gradients in orange). Within POm, initial targeting of CT axons is regulated by the interaction of EphAs in CT axons and the ephrin-A gradient in the nucleus. After P1, CT axonal processes (primary axons and their branches) are progressively increased in their number and/or length at specific terminal zones in VB and POm, determined by the levels of EphA expression in the CT axons and ephrin-As within these nuclei. The main axon shafts are not eliminated at any time of CT development despite being exposed to ephrin-As at a high level.