Segregation of ipsilateral retinal ganglion cell axons at the optic chiasm requires the Shh receptor Boc

Abstract

The pattern of contralaterally and ipsilaterally projecting retinal ganglion cell (RGC) axons at the optic chiasm is essential for the establishment of binocular vision. Contralateral axons cross the chiasm midline as they progress from the optic nerve to the optic tract. In contrast, ipsilateral axons deviate from the chiasm and continue in the ipsilateral optic tract, avoiding the chiasm midline. The molecular mechanism underlying this phenomenon is not completely understood. Here we show that the Sonic Hedgehog (Shh) receptor Boc is enriched in ipsilateral RGCs of the developing retina. Together with the presence of Shh at the midline, this complementary expression pattern led us to hypothesize that Shh might repel ipsilateral RGC axons at the chiasm. Consistent with this hypothesis, we found that only Boc-positive RGC axons retract in vitro in response to Shh and that this response is lost in Boc mutant RGCs. In vivo, we show that Boc is required for the normal segregation of ipsilateral axons at the optic chiasm and, conversely, that Boc expression in contralateral RGCs prevents their axons from crossing the optic chiasm. Together, these results suggest that Shh repels ipsilateral RGC axons at the optic chiasm via its receptor Boc. This work identifies a novel molecular pathway required for the segregation of axons at the optic chiasm.

Figure 1.

Boc is preferentially expressed in ipsilateral RGCs. A–B″, Boc expression detected by β-gal histochemistry in Boc ^+/lacZ mice (A–A″) and immunohistochemistry with an antibody against Boc (goat anti-Boc) in E15 horizontal section of the mouse retina (B–B″). Both methods show a strong enrichment of Boc in the VT retina as shown in A″ and B″. C–C‴, Boc (green, rabbit anti-Boc antibody) is expressed by RGC cell bodies in the GCL and is present in the RAL of the VT. Boc colocalizes with neurofilament (NF) on the axons of RGCs in the RAL (arrow; C–C‴) (horizontal sections of E17 rat retina). D–D″, Zic2-positive cells (red) colocalize with Boc (green, goat anti-Boc antibody) in the VT crescent of E15 mouse retina (horizontal sections). Higher magnification of the boxed regions from D and D″ shows the nuclear localization of Zic2 surrounded by membranous expression of Boc in the GCL (E–E″). CB, Ciliary body. Scale bars: A, 100 μm; A′, A″, 50 μm; C–C‴, 20 μm.

Figure 2.

Shh induces the rapid retraction of a subset of RGC axons. E17 rat retinal explants were cultured for 24 h, which allows RGC axons to elongate long distances out of the explant. A–C, Time-lapse images showing the retraction of an RGC axon following addition of 10 nm ShhN. A, Morphology of an RGC axon and growth cone at the time of ShhN addition. The black arrow points to the initial position of the growth cone. B, The growth cone starts to collapse. C, The growth cone has completely collapsed and the axon is strongly retracting. D, Time for a growth cone to initiate retraction in response to ShhN. The plot shows a sample of 33 RGC axons from four different experiments. The bar indicates the median time to initiate retraction (15 min). The dashed line indicates the mean time to initiate retraction (22 min). E, Quantification of the percentage of axons that exhibit net retraction 90 min after ShhN treatment. ShhN induces an increase in the percentage of axons with net retraction from 14.0 to 38.5% (p < 0.01). This effect is completely blocked by adding an inhibitor of Smoothened (SANT-1, 137 nm) 1 h before ShhN treatment (12.2% of RGC axons show net retraction; p < 0.01) (one-way ANOVA with Bonferronni's post-test, **p < 0.01). The mean percentage of retracted axons was quantified from at least three independent experiments for each condition, with at least four retinal explants per condition per experiment. Error bars indicate SEM. Scale bar, 10 μm.

Figure 3.

Shh specifically regulates axon and growth cone dynamics of Boc-expressing RGCs. A, Immunofluorescent labeling of RGC axons from E17 rat retinal explants shows the presence of high Boc (rabbit anti-Boc antibody) levels in a subpopulation of RGC axons (arrow), while most of the axons exhibit no or low Boc levels (arrowhead). B, Immunofluorescent labeling shows that the Shh signaling mediator Smoothened (Smo) is present on RGC axons. C, ShhN induces the growth cone collapse of RGC axons exhibiting high levels of Boc. Immunofluorescent labeling of Boc protein (green, rabbit anti-Boc antibody) in RGC growth cones (arrow), colabeled with the F-actin marker phalloidin (red) to delineate the morphology of the growth cone. Top, Growth cones of RGC axons expressing high levels of Boc remain uncollapsed 20 min after control treatment. Bottom, In contrast, 20 min after ShhN addition, RGC axons expressing a high level of Boc (arrow) have a collapsed growth cone, while the growth cones of RGC axons expressing no or low levels of Boc (arrowhead) are not sensitive to ShhN and do not collapse. Addition of control medium does not affect the size of RGC growth cones with a high level of Boc (top). D, Quantification of the collapse response from C. Measurement of the size of the RGC growth cone area (in square micrometers) 20 min after treatment with control medium (white bars) or ShhN (gray bars) shows a significant reduction in the area of RGC growth cones that express a high level of Boc when treated with ShhN. In contrast, RGC growth cones with no/low Boc levels do not have a significant change in area in response to ShhN. n = 87 growth cones. Error bars indicate SEM (one-way ANOVA with p < 0.05). E, Boc is required to mediate Shh-induced retraction of axons. Retinal explants isolated from Boc ^+/− and Boc ^−/− E15.5 embryos were cultured for 24 h. RGC axon dynamics were analyzed by time-lapse microscopy after addition of 10 nm ShhN. The percentage of RGC axons exhibiting net retraction 90 min after ShhN treatment was significantly higher for Boc ^+/− RGCs treated with ShhN (30.6%) than for those treated with control medium (7.5%, p = 0.0002), while Boc ^−/− RGC axons treated with ShhN were not significantly different from Boc ^−/− RGCs treated with control medium (p > 0.05) (one-way ANOVA with Bonferonni's post-test, ns = p > 0.05, **p < 0.01, ***p < 0.001). Error bars indicate SEM. Scale bar, 10 μm.

Figure 4.

Boc mutant mice have normal retinal cell fate. A, Horizontal sections of mouse E15 ventral retina from Boc ^+/− and Boc ^−/− mice were stained with various RGC markers. Zic2 is an ipsilateral RGC nuclear marker. EphB1 is an axonal marker of ipsilateral RGCs. Brn3b is a general RGC nuclear marker. Cdon labels RGCs and their axons. The lowest panels showcross-sections of the whole ventral retina (with the temporal side up) stained with L1, a marker of all RGC axons. None of these markers appears to be different in Boc ^−/− retina compared with Boc ^+/− retina. Insets represent a zoom in of the GCL. B, Quantification of the number of Zic2-positive RGCs shows that the number of ipsilateral RGCs is similar between Boc ^+/+, Boc ^+/−, and Boc ^−/− retinas. C, Quantification of the number of Brn3b-positive RGCs shows that the numbers of RGCs in the VT region are similar between Boc ^+/− and Boc ^−/− retinas. D, E, The number of cells in the VT region of the GCL (D) and the developing INL (E) is not significantly different between wild-type, Boc ^+/−, and Boc ^−/− retinas. n ≥ 3 embryos for each genotype. Scale bars, 50 μm; insets, 10 μm.

Figure 5.

Normal segregation of ipsilateral axons at the optic chiasm requires Boc. A, Schematic of the rostral part of the E18.5 mouse visual system after mono-ocular injection of DiI crystals (red) in the whole retina. B, Coronal sections of E18.5 Boc ^+/− and Boc ^−/− optic tracts after mono-ocular injection of DiI crystals (red) and counterstaining with DAPI (blue). The optic tracts are circled with yellow dashed lines. The DiI intensity is lower in the ipsilateral tract of Boc ^−/− compared with Boc ^+/− mice. Note that although these images show saturated pixels for ease of viewing the faint ipsilateral signal, the quantification was performed on unsaturated images. C, Quantification of the percentage of DiI signal in the ipsilateral optic tract as a proportion of the total DiI signal from the contralateral and ipsilateral optic tract reveals a significant decrease in ipsilateral projections in Boc ^−/− mice compared with control (Boc ^+/− or Boc ^+/+) mice [18.5 ± 2.5% (n = 14 embryos) and 10.0 ± 1.8% (n = 6), respectively; p < 0.05 using Student's unpaired t test]. Scale bar, 100 μm.

Figure 6.

Ectopic Boc expression in contralateral RGCs prevents their axons from crossing the optic chiasm. A, Schematic of the rostral part of the E16.5 mouse visual system, 3 d after in utero electroporation of the retina at E13.5. The positions of the retina (B) and the optic chiasm (C) are indicated by the boxes. The green line indicates the normal path of YFP-electroporated contralateral RGC axons. B, Horizontal section of an E16.5 mouse retina 3 d after electroporation. Due to electrode positioning, the VT region never contained YFP-positive cells. C, Control axons expressing YFP alone exhibit normal axon pathfinding, crossing the optic chiasm and reaching the contralateral optic tract (D). They very rarely enter the ipsilateral optic tract (E) (1.8 ± 0.4%, E″) or the contralateral optic nerve (F) (2.5 ± 1.0%, F″). D, D′, Semiperpendicular sections of the contralateral optic tract. E, E′, Semiperpendicular sections of the ipsilateral optic tract. F, F′, Parallel sections of the contralateral optic nerve. When Boc is coelectroporated with YFP (Boc+YFP), Boc+YFP-expressing axons enter the ipsilateral optic tract (E′) and contralateral optic nerve (F′) at a higher frequency. E″, F″, Quantification of the percentage of misguided RGC axons over the total number of electroporated axons. E″, Histogram showing that the percentage of ipsilateral axons is significantly increased (by fourfold, p = 0.0262 using Student's unpaired t test) when RGCs express Boc+YFP compared with YFP alone (YFP, n = 13; Boc, n = 19 embryos). F″, Histogram showing that the percentage of axons re-entering the contralateral optic nerve is significantly higher (by 3.7-fold, p = 0.0085 using Student's unpaired t test) in RGCs expressing Boc+YFP compared with YFP alone (YFP, n = 11; Boc+YFP, n = 16 embryos). on, Optic nerve; oc, optic chiasm; ot, optic tract. Scale bar, 100 μm.

Figure 7.

Summary of the Boc gain- and loss-of-function phenotypes at the chiasm. A, In wild-type conditions, axons expressing Boc (red) originate from the VT retina and avoid Shh at the optic chiasm. B, In Boc mutant mice, VT axons do not express Boc and fail to detect Shh at the midline. Instead of being repelled by Shh, they cross the chiasm and project contralaterally (represented by the dashed line). C, Ectopic expression of Boc in presumptive contralateral RGCs prevents their axons from crossing the optic chiasm. Two types of defects were observed: the conversion of contralateral projections into ipsilateral projections and the misrouting of axons to the contralateral optic nerve (represented by the dashed lines). V, Ventral; D, dorsal; N, nasal; T, temporal; on, optic nerve; oc, optic chiasm; ot, optic tract.