EphA4 has distinct functionality from EphA7 in the corticothalamic system during mouse brain development

Abstract

Deciphering the molecular basis for guiding specific aspects of neocortical development remains a challenge because of the complexity of histogenic events and the vast array of protein interactions mediating these events. The Eph family of receptor tyrosine kinases is implicated in a number of neurodevelopmental activities. Eph receptors have been known to be capable of responding to several ephrin ligands within their subgroups, often eliciting similar downstream effects. However, several recent studies have indicated specificity between receptor-ligand pairs within each subfamily, the functional relevance of which is not defined. Here we show that a receptor of the EphA subfamily, EphA4, has effects distinct from those of its close relative, EphA7, in the developing brain. Both EphA4 and EphA7 interact similarly with corresponding ligands expressed in the developing neocortex. However, only EphA7 shows strong interaction with ligands in the somatosensory thalamic nuclei; EphA4 affects only cortical neuronal migration, with no visible effects on the guidance of corticothalamic (CT) axons, whereas EphA7 affects both cortical neuronal migration and CT axon guidance. Our data provide new evidence that Eph receptors in the same subfamily are not simply interchangeable but are functionally specified through selective interactions with distinct ligands in vivo. J. Comp. Neurol. 524:2080-2092, 2016. © 2015 Wiley Periodicals, Inc.

Keywords: AB_10015282; AB_221569; AB_2313608; AB_514496; AB_777699; Eph receptor; cortex; corticothalamic projections; ephrin; nif-0000-30467; rid_000042; thalamus.

Figure 1. Patterns of EphA receptor binding in the cortex of the perinatal brain

(A–H) Binding patterns of Fc-tagged EphA receptors (EphA1 – EphA8) within the cortex at P4. All EphA-Fc receptor displayed binding in the cortex, with significant binding observed in the upper cortical layers. V = Visual Cortex, S = Somatosensory Cortex, Au = Auditory Cortex. Scale bar = 500 μm.

Figure 2. Patterns of EphA receptor binding in the VB and POm thalamic nuclei in the perinatal brain

(A–H) Binding patterns of Fc-tagged EphA receptors (EphA1 – EphA8) within the VB and POm at P4. Note that the labeling by EphA3-Fc, A5-Fc, A7-Fc and A8-Fc binding shows a distinct gradient in the VB with the strongest labeling at the ventro-lateral region (C, E, G, H), whereas EphA1-Fc, A2-Fc, EphA4-Fc, and EphA6-Fc binding in the VB (A, B, D, F) does not. Scale bar = 500 μm.

Figure 3. Patterns of ephrin-A ligand expression in the VB/POm and cortex of the perinatal brain

(A–E) In situ hybridization of ephrin-A ligands within the VB/POm at P4. Expression of ephrin-A2 and -A3 (B and C) is observed throughout the VB/POm in a uniform fashion. In contrast, the expression of ephrin-A5 is observed in a distinct ventro-lateral gradient within the VB/POm (E). Expression of ephrin-A1 and –A4 was low/not expressed (A and D). (F–J) In situ hybridization of ephrin-A ligands within the cortex. Strong expression of ephrin-A2, -A3, and –A5 (G, H and J) are observed in the cortex, while ephrin-A1 and –A4 expression was low/not expressed. V = Visual Cortex, S = Somatosensory Cortex, Au = Auditory Cortex. Scale bar = 500 μm.

Figure 4. Distinct expression patterns of EphA4 and EphA7, in comparison to ephrin-A5 binding in the early postnatal brain

(A–D) In situ hybridization of EphA4 (A and B) and EphA7 (C and D) in anterior (A, C) and posterior (B, D) sections of the P4 mouse brain. In both anterior and posterior sections, EphA4 is observed predominantly in upper cortical layers, with lower expression in deeper layers (A and B, brackets). In contrast, EphA7 labeling is evident in distinct gradients within deeper layers, including layer VI (C and D, brackets). Both EphA4 and EphA7 have distinct expression patterns in the striatum (A and C) and thalamus (B and D). (E and F) Binding patterns of ephrin-A5-Fc at anterior (E) and posterior (F) levels. The relative labeling intensity of ephrin-A5-Fc binding displays an overall similarity to EphA7 expression, including the gradients in layer VI (see brackets), striatum, and thalamus. Scale bar = 500 μm.

Figure 5. Columnar segregation of EphA4-overexpressing neurons in the neocortex

(A–D) Effects of EphA overexpression in cortical neurons. E12.5 mice were co-electroporated (EP) with EYFP and an overexpression or control vector, with brains collected at P4. Brains were immunostained for EYFP and counterstained with DAPI (blue, C). Overexpression of EphA7 (B) or EphA4 (C and D) in cortical neurons resulted in distinct columnar segregation, which was not observed in the control (A). (E-H) EphA4-Fc and EphA7-Fc binding of E14.5 brains. Binding of both EphA4-Fc (E) and EphA7-Fc (F) receptors are observed in the developing cortex in similar patterns. G and H are higher magnification views of the boxed areas in E and F, respectively, showing EphA4-Fc and EphA7-Fc binding in the ventricular/subventricular zones (VZ/SVZ), intermediate zone (IZ) and cortical plate (CP). Scale bars = 200 μm (A, B, D), 500 μm (C) and 500 μm (E, F).

Figure 6. Overexpression of EphA4 in the somatosensory cortex does not result in a shift of CT projections in the VB/POm at P4

(A) Schema for in utero electroporation with two fluorescent markers at different concentrations. The control, EphA4-, or EphA7-expression plasmid together with 0.5 mg/ml (low concentration) of EYFP-expression plasmid and 3 mg/ml (high concentration) of DsRed2-expression plasmid was injected into the lateral ventricle of E12.5 brains, followed by electroporation. (B) P4 cortical neurons in the somatosensory cortex electroporated with low concentration of EYFP (green) and high concentration of DsRed2 (magenta). EYFP⁺ cells (white; EYFP⁺/DsRed2⁺) displays neurons with higher levels of ectopic EphA expression, while EYFP⁻ cells (magenta; EYFP⁻/DsRed2⁺) shows cells with lower levels of ectopic EphA expression. (C–E) The VB/POm of control electroporated (EP) brains. EYFP⁺ and EYFP⁻ CT axons are distributed similarly in the center of the VB (insets show higher magnification views of the boxed areas) as well as at the border region between the VB and POm (arrowheads). (F–H) The VB/POm of EphA7 electroporated brain. CT axons with high levels of EphA7 expression (EYFP⁺) show a restricted distribution at the border region between the VB and POm (arrowheads) and are segregated from axons with low levels of ectopic EphA7 expression (EYFP⁻) that distribute also in the body of the VB (insets show higher magnification views of the boxed areas). (I–K) The VB/POm of EphA4 electroporated brain. Unlike EphA7 overexpression, EphA4 has no effect on CT axons in this region similar to that of controls, with high (EYFP⁺) and low (EYFP⁻) levels of expression distributed similarly at the VB and POm (insets show higher magnification views of the boxed areas). Scale bar = 100 μm (for B), 200 μm (for C–K).

Figure 7. Centroid analysis of control, EphA7- and EphA4-overexpressing CT projections within the VB

(A–C) The VB/POm of control electroporated brains. The centroids of the area covered by DsRed2⁺ and EYFP⁺ CT axons (dots in magenta and green, respectively) are closely positioned. (D–F) The VB/POm of EphA7 electroporated brains. The centroid of EYFP⁺ CT axons are shifted dorsally from that of DsRed2⁺ CT axons compared to controls. (G–I) The VB/POm of EphA4 electroporated brains. The centroids of DsRed2⁺ and EYFP⁺ CT axons are closely positioned.

Figure 8. No significant shifts in CT projections within the VB by overexpression of EphA4

The relative position of the centroid of EYFP⁺ axons from that of DsRed2⁺ axons along the ventro-medial to dorso-lateral (VM-DL) axis (X-axis) and the ventro-lateral to dorso-medial (VL-DM) axis (Y-axis) (arrows in Fig. 7I). The data for EphA4-, EphA7- and control-electroporated brains are graphed by setting the relative position of the centroid of EYFP⁺ projections in control-electroporated brains at (0.0). The differences only along the VL-DM axis (Y-axis) are significant between control- and EphA7-electroporated brains, and between EphA4- and EphA7-electroporated brains (p < 0.01, ANOVA/Tukey-Kramer’s multiple comparison). Error bars show the SEM, and n = the number of brains analyzed in each condition.