Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 28.
Published in final edited form as: Dev Biol. 2026 Jul 21;538:214–225. doi: 10.1016/j.ydbio.2026.07.012

The extracellular matrix in selective decussation of retinal ganglion cell axons: β2 laminins regulate the ipsilateral projection

Alanis Hernandez-Arce 1, Madeline Turo 1,*, Adam N Robinson 1,&, Danny Yeo 1, Skylyn McNamara 1, Reyna I Martínez-De Luna 1
PMCID: PMC13404271  NIHMSID: NIHMS2196640  PMID: 42480735

Abstract

In binocular animals, retinal ganglion cell (RGC) axons selectively decussate at the optic chiasm. Selective decussation is directed by the radial glia and midline neurons in the optic chiasm. Radial glia attach to the underlying pial basement membrane (PBM) that is rich in β2 laminins. Here, we asked whether, and if so, how, β2 laminins in the PBM impact the selective decussation of RGC axons. β2 laminins are expressed in the PBM during the peak and late phases of RGC axon growth, and their deletion results in the fragmentation of the PBM and dysmorphic radial glia that lack properly elaborated endfeet. Genetic deletion of the β2 laminin subunit increased the proportion of non-decussating RGC axons in the ipsilateral tract. Consistent with the increase in the proportion of ipsilateral axons, we found persistent expression of the ipsilateral guidance cue EphrinB2 in radial glia during the late phase of axonal decussation and an increased number of ipsilateral RGCs in the ventrotemporal retina. These results suggest that in the optic chiasm β2 laminins are required for the integrity of the PBM and radial glia organization, and that this in turn may influence the production of EphrinB2. Our results also suggest that β2 laminins in the retina are necessary for neurogenesis and the laterality fate of ipsilateral RGCs, a function possibly mediated by their role in maintaining the attachment of retinal progenitor cells (RPCs) to the inner limiting membrane (ILM).

Introduction

Retinal ganglion cells (RGCs) transmit visual information from the retina to the brain through their axons that form the optic nerves and tracts. RGCs extend axons that exit the eye and course to the ventral mid-diencephalon where they establish the optic chiasm. At this decision point, RGC axons are selectively routed to either cross the midline into the contralateral optic tract or not cross and remain ipsilateral, a process that is known as selective decussation. This selective routing of RGC axons ensures the orderly establishment of the circuits required for binocularity.

The selective decussation of RGC axons is orchestrated by the radial glia and midline neurons of the ventral mid-diencephalon as the optic chiasm forms (Marcus et al., 1995; Sretavan et al., 1994). These cells express axon guidance cues that direct axon routing across distinct developmental phases defined by which RGCs are extending axons to the chiasm at the time (Petros et al., 2008). During the peak phase (E14–E17 in mice), RGCs born in the dorsonasal and ventrotemporal retina reach the mid-diencephalon. The concerted action of PlexinA1 expressed in the midline neurons and Sema6D and NrCAM in the radial glia promote the crossing of the dorsonasal axons (Kuwajima et al., 2012; Williams et al., 2006). Simultaneously, axons from the non-crossing ventrotemporal RGCs are guided into the ipsilateral tract by the transient expression of EphrinB2 in the radial glia. In the late phase (E17–P0 in mice), EphrinB2 expression is downregulated, and axons from the last RGCs born in the ventrotemporal retina are routed contralaterally. Thus, optic chiasm formation is coordinated by a program of guidance cue expression that dynamically regulates axon routing over developmental time.

The selective decussation of RGC axons is also controlled by the laterality identity of the RGCs. RGCs acquire an ipsilateral or contralateral fate based on the timing of their neurogenesis and birthplace. Contralateral RGCs are generated earlier in development and are specified by the transcription factors Pou3f1 and Isl2 (Fries et al., 2023; Kuwajima et al., 2012; Marcucci et al., 2019; Pak et al., 2004). Genetic deletion of either Pou3f1 or Isl2 genes results in an increase in the number ipsilateral RGCs, suggesting that these transcription factors promote contralateral fate by suppressing ipsilateral fate. The SoxC transcription factors (Sox 4, 11 and 12) do not specify but promote the differentiation of contralateral RGCs and the expression of the axon guidance cues PlexinA1 and NrCam (Kuwajima et al., 2017). In contrast, ipsilateral RGCs are generated later and are specified by the transcription factor Zic2 (Herrera et al., 2003). Zic2 also promotes the expression the tyrosine kinase receptor EphB1 that mediates the midline avoidance of ipsilateral RGCs (Lee et al., 2008).

RGC selective axon decussation is also influenced by cues provided by the extracellular matrix (ECM). The abundant ECM proteins like hyaluronan and the heparan and chondroitin sulfate proteoglycans regulate axonal decussation (Brittis et al., 1992; Conway et al., 2011; Lin et al., 2007; Pratt et al., 2006). Other ECM proteins involved in axon guidance are laminins. Laminins are heterotrimeric molecules composed of an α, β, and γ chain that are integral proteins of basement membranes, specialized sheets of ECM that function as signaling platforms to regulate cell adhesion, proliferation and differentiation (Yurchenco, 2011). In zebrafish, which lack an ipsilateral projection, mutations in the genes Lama1, Lamb2 and Lamc1 that encode for the laminin α1, β2 and γ1 chains, respectively, resulted in misrouting of axons to the ipsilateral side of the brain (Barresi et al., 2010; Karlstrom et al., 1996; Paulus and Halloran, 2006). Further supporting the role of laminin in axon guidance, dystroglycan (DG), a canonical laminin receptor was in implicated in RGC axon decussation (Clements and Wright, 2018).

In mouse, laminins containing the β2 subunit are abundant in the inner limiting membrane (ILM) of the retina and pial basement membrane (PBM) of the brain (Hunter et al., 1992; Libby et al., 2000; Radner et al., 2013). Here we hypothesize that β2 laminins in the PBM of the mouse optic chiasm function in the selective decussation of RGC axons. We found that genetic deletion of β2 laminins increased the proportion of non-decussating axons routed into the ipsilateral tract. Loss of β2 laminins caused fragmentation of the PBM and changes to the radial glia basal processes and endfeet. These findings on the specialized niche of the optic chiasm suggested that the expression of axon guidance cues might also be altered. We found that the protein expression of the ipsilateral axon guidance cue EphrinB2 persists beyond its normal developmental window, likely extending the time window for ipsilateral guidance into the late phase of RGC axon-selective decussation. We also found that the number of ipsilateral RGCs is increased during this late phase of axonal decussation. Together, these results suggest that β2 laminins maintain the integrity of the PBM and radial glia organization, and these functions may be necessary for the timely restriction of EphrinB2 protein expression to the peak phase of RGC axon guidance. In the retina, β2 laminins may similarly, through their role in RPC adhesion to the ILM, affect neurogenesis and cell fate of ipsilateral RGCs.

Materials and methods

Animals

C57BL/6J wild-type, Lamb2−/− (Lamb2tm1Jrs), and Lamc3−/− (Lamc3tm1.1Wjbr) mice were used in the studies. Deletion of the Lamb2 and Lamc3 genes and the generation of the knockout lines were previously described (Li et al., 2012; Noakes et al., 1995). The Lamb2 line was maintained as heterozygotes, and the Lamc3 line as homozygotes. The sex of the embryos was not assessed for the experiments. The morning that the vaginal plug was observed was considered E0.5. For embryo collection, pregnant dams were euthanized by CO2 asphyxiation. Embryos at a specified age were collected and genotyped, with wild-type (WT) littermates serving as controls. Postnatal animals were euthanized by transcardial perfusion. All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Upstate Medical University and adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Anterograde labeling of retinal projections and measurement

Heads from P4 animals were fixed in 4% paraformaldehyde (PFA) overnight at 4°C. The fixative was then removed by washing the heads in 1X phosphate-buffered saline (PBS, 137 mM sodium chloride, 10 mM sodium phosphate, pH 7.4) 3 times for 5 min each. A small incision was made at the ora serrata, and the cornea and iris were cut, circling the ora serrata, leaving a piece intact to create a cornea/iris flap. The flap was lifted, the lens removed, and crystals of DiI (DiIC18(3), Invitrogen) were placed to cover the optic nerve head. The lens was placed back in the eye, and the cornea/iris flap was placed back over the lens. The DiI-labeled heads were then incubated at 37°C for 14 days in 1X PBS containing 0.02% sodium azide. After 14 days, the heads were washed in 1X PBS 3 × 10 min, and the brains were removed from the skull with care not to damage the optic chiasm or tracts. The optic chiasm and projections in the intact brain were imaged and photographed in a Leica MZ16 fluorescent dissection microscope.

The proportion of DiI-labeled ipsilateral and contralateral projections was determined by image analysis in FIJI (Schindelin et al., 2012). Each projection was traced from the most proximal to the chiasm to the edge of the thalamus to measure the area. The optic chiasm area was excluded from the measurements. The area of each projection was then used to calculate its proportional area dividing the area of each projection by the sum of the areas of both projections. The ipsilateral coefficient (Cip) and contralateral optic nerve co-efficient (Con) were obtained as previously described (Williams et al., 2003). Briefly, squares were overlaid covering the labeled optic nerve, the ipsilateral projection in the region proximal to the optic chiasm and the contralateral optic nerve. The fluorescence intensities within each overlaid square were measured using FIJI. Each fluorescent measurement was normalized to the area and background intensity from an area without signal in the same image was subtracted from each measurement. The co-efficients were then calculated for each animal as follows: Cip=Ipsilateral optic tract fluorescence/(Ipsilateral optic tract fluorescence + Optic nerve fluorescence) and Con= Contralateral optic nerve fluorescence/(Contralateral optic nerve fluorescence + Ipsilateral optic nerve fluorescence).

Immunohistochemistry

The antibodies used are detailed in the Key Resources Table (Supplementary Table 1). Embryo heads were fixed in 4% PFA overnight at 4°C. Heads were then washed in 1X PBS 3 times for 5 minutes each, and cryoprotected using sucrose solutions of increasing concentration. Heads were placed in 10%, 20%, and 30% sucrose in 1X PBS overnight at 4°C. Half of the 30% sucrose solution was replaced with OCT medium, mixed thoroughly by mixing in a nutator for an hour, and the heads were incubated in this mixture overnight at 4°C on a shaker. The heads were then embedded in OCT and kept at −80°C until sectioning. Cryostat sections (20 μm) of the optic chiasm or retina were obtained. The sections were washed in 1X PBS, pH 7.5, 3 times for 10 min each, and then incubated in blocking solution (5% Normal Donkey Serum (NDS) and 0.2% Triton X-100 in 1X PBS) for 1 hour at room temperature. The sections were then incubated in primary antibodies overnight at 4°C in antibody diluting solution (5% NDS, 0.2 % Triton X-100, 1:15,000 Hoechst), washed in 1X PBS 3 × 10 min, and incubated with secondary antibody in antibody diluting solution for 2 hours at room temperature. The slides were then washed in 1X PBS 3 times for 10 min each and coverslipped using Prolong Gold or Glass antifade mounting medium (Thermo Fisher). Antigen retrieval with sodium Citrate buffer (pH 6.0) was performed before the blocking step for immunolabeling with Zic2 and Brn3a antibodies (Slavi et al., 2023).

To detect basement membrane proteins and visualize radial glial endfeet, embryo heads were fixed in 0.05% PFA for 1 hour on ice. The sections obtained from these lightly fixed heads were kept at −20°C during collection and until staining. For staining, the sections were fixed in ice-cold methanol for 2 minutes, washed in 1X PBS pH 7.5 3 times for 10 min each, and then incubated in blocking solution (5% Normal Donkey Serum (NDS) and 0.05% Triton X-100 in 1X PBS) for 1 hour at room temperature. The sections were then incubated in primary antibodies overnight at 4°C in antibody diluting solution (5% NDS, 0.01% Triton X-100, 1:15,000 Hoechst), washed in 1X PBS 3 × 10 min, and incubated with secondary antibody in antibody diluting solution for 2 hours at room temperature. The slides were then washed in 1X PBS 3 times for 10 min each and coverslipped using Prolong Gold or Glass antifade mounting medium (Thermo Fisher).

In situ hybridization

Hybridization chain reaction in situ hybridization was performed using the HCR Gold RNA FISH system (Molecular Instruments; Los Angeles, CA). EphrinB2 probes and hairpin amplifiers were purchased from the manufacturer. Heads from E17.5 embryos were fixed overnight in 4% PFA, cryoprotected in increasing concentrations of sucrose in 1X PBS (10%, 20%, and 30%) and then cryostat sectioned at 20 μm. Sections were pre-treated following manufacturer’s instructions for tissue sections. Samples were pre-hybridized at 37°C for 30 minutes and then incubated with probe sets (2 nM each) in hybridization buffer overnight at 37°C. Following the probe wash steps, hairpin amplifiers (60 nM each, snap-cooled separately) were added in amplification buffer and incubated at room temperature for 4 hours in the dark. Samples were washed, counterstained with DAPI, and mounted for confocal imaging. The negative control was incubated without probe in the hybridization step and amplification was done as described.

For standard in situ hybridization, the EphrinB2 coding sequence was PCR-amplified from the cDNA clone BC057009 (Horizon Discovery; Lafayette, CO) after sequence verification. The first 905 bp of the coding sequence was amplified using primers 5’-ATGGCCATGGCCCGGTC-3’ and 5’-CGCTGTCTGCAGTCCTTAGTGG-3’ to obtain a similar probe as described previously (Birgbauer et al., 2000) but excluded the last 107 bp of the coding sequence that has an 84% identity (Clustal Omega alignment) to the 3’end of EphrinB1 and EphrinB3. PCR amplicons were TA-cloned into pGEMTEZ (Promega; Fitchburg, WI) and sequence-verified. Digoxigenin-labeled riboprobes were transcribed in vitro from pGEMTEZ.EphrinB2 (SpeI, T7 polymerase for antisense and PspMOI, SP6 polymerase for sense), using the T7 or SP6 HiScribe High Yield RNA Synthesis kits (New England Biolabs; Ipswich, MA). For in situ hybridization on tissue sections, embryonic heads were fixed in 4% PFA overnight, washed 3 times in 1X PBS for 5 minutes each, cryoprotected in 10%, 20%, and 30% sucrose, and embedded in OCT. Optic chiasm sections (20 μm) were obtained and stored at −20°C until use. To minimize signal variability of the colorimetric reaction, WT and Lamb2−/− sections were placed next to each other on the same slide. In situ hybridization was performed as previously described (Martinez-De Luna et al., 2013). All sequences and plasmids are available upon request.

Quantitative Reverse Transcription PCR (qRT-PCR)

Total RNA was extracted from optic chiasm/mid-diencephalon tissues dissected from three E17.5 wild-type and Lamb2−/− embryos using the Monarch Spin RNA kit (New England Biolabs). RNA (250 ng) was reverse transcribed using the iScript Reverse Transcription Supermix (BioRad). 10 ng of cDNA were amplified in quadruplicates using the Power Track SYBR Green Master Mix (Thermo Fisher Scientific) in BioRad CFX 384 Real Time System (BioRad). The EphrinB2 primers were designed using Primer3 (Undergasser et al 2012; Koressar and Remm 2007; and Koressar et al 2018) and their sequences were 5’- ATAACCAGGAGGGAGGGGTGT-3’ and 5’- CTGGACGTCTTGTTGGACCG-3’. PCR products were TA cloned (pGEM-T-Easy sytem, Promega) and sequence verified to confirm EphrinB2 amplification. The PCR was run at 95°C for 2 min, then cycled through 40 cycles of 95°C for 15 seconds and 60°C for 1 minute per manufacturer’s instructions. Melting curves were obtained for each PCR run at 60° to 95°C. Transcript levels were normalized to β-actin (Actb) and mRNA fold change was calculated relative to mean control values using the comparative Ct method (Schmittgen and Livak, 2008).

Confocal microscopy and fluorescence intensity measurements

Confocal images were acquired in the Neuroscience Program Microscopy Core at Upstate Medical University. Optic chiasm and retina sections were imaged on a Zeiss LSM780 laser scanning confocal microscope, equipped with an Argon laser and a He/Ne 633nm laser, using a Zeiss Plan Apochromat 20X/0.8 NA dry and 63X/1.4 NA oil lens. Pinhole was set to 1AU with a resolution of 512 × 512 pixels. A line average of 16 was applied. Confocal Z-series were acquired using identical gain and laser power to image WT and Lamb2−/− sections. Fluorescent intensity measurements of β1-Integrin, β-Dag, and EphrinB2 were performed using Fiji software. Measurements were acquired from WT and Lamb2−/− optic chiasm (3–9 sections) with equal image threshold and background subtraction. The chiasm area was defined using Nestin staining to identify the anterior, middle, and posterior sections. The fluorescence of the blood vessels and axons were subtracted from the β1-Integrin and the β-Dag intensity measurements. Images of ipsilateral and contralateral RGC counts were acquired at 20X with a digital zoom of 1.5 and a 0.92 step size to differentiate cells from one another carefully. High-magnification, high-resolution images of the PBM and radial glia endfeet were obtained on a Leica SP8 at 100X and 1012 × 1012 pixels with a digital zoom of 1.5 using a HC PL APO 100x/1.4 Oil STED white lens.

Blinding of the sample genotype for analysis

Fluorescence intensity measurements and cell counts were performed in a blinded manner with respect to the animals’ genotypes. In most cases, images were acquired by someone other than the person analyzing them. After image acquisition, the Z-stacks were randomized using a Python script that assigned random labels to each file and generated a de-blinded list of the samples. The blinded Z-stacks were given to the person analyzing the images. The original identities of the images were revealed after the counts were performed. The Z-stack scrambler code can be found at https://github.com/Martinez-De-Luna-Lab/Sample-Randomizer.

Statistical Analysis

The sample sizes are stated in the figures and figure legends. Comparisons between groups were performed using unpaired Student’s t-test and One-Way Analysis of Variance (ANOVA) with Tukey’s multiple comparisons post-test. Significance was set at p<0.05. Graph Pad Prism software (GraphPad Software, La Jolla, CA) was used for all analyses.

Results

The β2 and γ3 laminin subunits are expressed in the optic chiasm and the retina at the time of selective axon decussation

The optic chiasm in the ventral brain is underlaid by the pial basement membrane (PBM). Laminins containing the β2 and γ3 chains are abundant in the PBM (Hunter et al., 1992; Libby et al., 2000; Radner et al., 2013). Therefore, we decided to examine their distribution in the optic chiasm. We immunolabeled sections from E15.5 and E17.5 embryos using previously characterized laminin β2- and γ3-chain-specific antibodies (Li et al., 2012; Sasaki et al., 2002). These antibodies detect the laminin isoforms that contain the β2 and γ3 chains, that from hereon will be referred as β2 and γ3 laminins. We selected E15.5 and E17.5 because these ages correspond to the peak and late phases of axon growth through the optic chiasm, respectively (Petros et al., 2008).

At E15.5 and E17.5, immunoreactivity for β2 laminins is robust in the PBM underlying the axons of the optic chiasm (Fig. 1A, B, D, and E), where they are co-expressed with Perlecan, another core basement membrane protein (Fig. 1C). We also find strong β2 laminin immunoreactivity in the vascular basement membrane of blood vessels that display CD31 immunoreactivity (Fig. 1A, B, D, E, and F; arrowheads). In addition, localization of the β2 chain is also observed in the ventricular zone of the third ventricle where the radial glial cell bodies are located, the pia mater, the dura that surrounds the RGC axons, and cells within Rathke’s pouch, the presumptive anterior pituitary (Fig. 1A, B, D, and E, asterisks and arrowheads). The immunoreactivity observed with the β2-chain specific antibody is absent in the optic chiasm of the knockout mouse (Lamb2−/−) lacking β2 laminins (Supp. Fig. 1).

Figure 1. The β2 and γ3 laminin subunits are expressed in the optic chiasm and retina at the time of selective axon decussation.

Figure 1.

Open in a new tab

(A, B, D, and E) Protein expression of the β2 laminin subunit in the optic chiasm. Asterisk: pia mater; arrowhead: dura mater that covers the optic nerves and tracts. (G, H, J, and K) Protein expression of the γ3 laminin subunit in the optic chiasm. Asterisk: pia mater; arrowhead: dura mater that covers the optic nerves and tracts. (C, F, I, and L) Confirmation of β2 and γ3 expression in the PBM and microvasculature of the optic chiasm by co-staining with the PBM marker Perlecan and the endothelial cell marker CD31. (M-X) β2 and γ3 subunit protein expression in the retina with magnified views of the dorsal and ventral retina. D: dorsal; V: ventral. Asterisk: Retina periphery; L: lens; ILM: inner limiting membrane; Bruch’s: Bruch’s membrane. N=3 embryos analyzed for each panel.

We next investigated the protein distribution of the laminin γ3 chain using a previously validated antibody (Li et al., 2012). The protein distribution of the γ3 laminins is more restricted than that of the β2 laminins. At E15.5 and E17.5, γ3 laminins are detected in the PBM where Perlecan is also present (Fig. 1G, H, J, K and I). γ3 laminins are a known component of the CNS microvasculature (Li et al., 2012). Consistent with previous findings, we detected robust immunoreactivity of γ3 laminins in the basement membrane of the diencephalic microvasculature and the pial vasculature, positive for CD31 (Fig. 1G, H, J, K and L). We also found strong immunoreactivity of γ3 laminins in the dura that envelops the RGC axons and at the edges of the chiasm (Fig. 1G, H, J, and K; arrowheads).

We also determined the protein distribution for the β2 and γ3 laminins in the retina. At E15.5, β2 laminins are abundant in the basement membrane of the retina, the inner limiting membrane (ILM) (Fig. 1M–O). The ILM expression of β2 laminins extends into the dorsal and ventral ciliary marginal zone (CMZ) with some expression in the peripheral dorsal and ventral retina (Fig. 1N and O). In addition to the ILM and CMZ, β2 laminins are observed in Bruch’s membrane abutting the retinal pigment epithelium, the lens epithelium, and fragments of the hyaloid vasculature attached to the lens (Fig. 1M–O). At E17.5, the distribution of β2 laminins is very similar to E15.5 with the most robust expression in the ILM and some expression in the peripheral dorsal and ventral retina and CMZ (Fig. 1S–U). In contrast to the expression of β2 laminins, at E15.5 and E17.5, the expression of γ3 laminins is less robust in the ILM and high in Bruch’s membrane, a finding that is consistent with previous work (Fig. 1P and V) (Li et al., 2012). We did observe very faint expression of γ3 laminins in the dorsal and ventral CMZ at E15.5 (Fig. 1Q and R), that was absent at E17.5 (Fig 1W and X). We also observed expression of γ3 laminins in the fragments of the hyaloid vasculature attached to the ILM and lens at both ages (Fig. 1P–R and V–X).

Together, these results show that β2 and γ3 laminins are expressed in the developing optic chiasm. In the retina, β2 laminins are predominantly expressed in the ILM with some expression in the peripheral retina. In contrast, γ3 laminins are predominantly expressed in Bruch’s membrane at the ages examined.

β2 laminins influence the selective decussation of ipsilateral RGC axons

Laminins have been implicated in the guidance of axons in commissures of the central nervous system and the optic chiasm in zebrafish (Barresi et al., 2010; Paulus and Halloran, 2006). Based on these studies and the expression patterns we observed in the optic chiasm, we hypothesized that β2 and γ3 laminins (encoded by the Lamb2 and Lamc3 genes, respectively) in the PBM are positioned to provide cues that influence the selective decussation of retinal axons. To test this hypothesis, we asked if the selective decussation of RGC axons was altered in Lamb2−/− and Lamc3−/− mice compared to WT controls. We anterogradely labeled RGC axons from the right eye with DiI to quantify the ratio of ipsilateral to contralateral axons in the optic tracts (Figure 2A). We measured the area of each projection and used these measurements to calculate the proportional area of the ipsilateral and contralateral optic tracts in WT, Lamb2−/−, and Lamc3−/− mice (Fig. 2B).

Figure 2. The proportion of the ipsilateral projection is increased in Lamc3−/− and Lamb2−/− mice.

Figure 2.

Open in a new tab

(A) Experimental schematic showing placement of DiI crystals and anterograde labeling of retinal projections. The proportional areas of axons in the optic tracts were calculated excluding the area of the optic chiasm using the formula shown. (B-D) Labeled RGC projections in the WT (N=7), Lamb2−/− (N=4), and Lamc3−/− (N=3) mice. Dashed lines trace the area of each projection. (E) The proportional area of the ipsilateral projection. (F) The proportional area of the contralateral projection. (G) The co-efficient of the ipsilateral projection. (H) The co-efficient of the contralateral optic nerve (ON). Statistical significance was calculated using a One-way ANOVA with Tukey’s multiple comparisons post-test. Error bars represent the S.E.M.

In P4 WT mice, the proportional area of the ipsilateral projection was 0.38 ± 0.01 of the total size of the projections, while the proportional area of the contralateral projection was 0.61 ± 0.004 (Fig. 2B, E, and F). We found that in the Lamb2−/− and Lamc3−/− mice, the proportion of the ipsilateral projection was increased compared to wild-type (Lamb2−/−: 0.44 ± .008, p=0.0001; Lamc3−/−: 0.42 ± 0.009, p=0.0002) (Fig. 2C, D, and E). With this analysis, we also found that the size of the contralateral projection was reduced in the Lamb2−/− and Lamc3−/− mice compared to the WT (Lamb2−/−: 0.56 ± 0.008, p=0.0001; Lamc3−/−: 0.58 ± 0.009, p=0.056) (Fig. 2C, D and F). We next verified the increase in the ipsilateral projections of Lamb2 and Lamc3 animals by determining the co-efficient of the ipsilateral projection (CIP) (Herrera et al., 2003). We measured the mean fluorescent intensity of the ipsilateral optic tract and compared it to the mean fluorescent intensity of the labeled optic nerve. In the WT, the ipsilateral co-efficient was 0.15 ± 0.009 (Fig. 2G). In contrast, the ipsilateral co-efficient in the Lamb2−/− and Lamc3−/− mice was significantly increased compared to the WT (Lamb2−/−: 0.21 ± 0.01, p=0.02; Lamc3−/−: 0.26 ± 0.02, p=0.0003) (Fig. 2G). However, the ipsilateral co-efficient was not significantly different between Lamb2−/− and Lamc3−/− animals (p=0.05) (Fig. 2G).

We also determined if there was an increase in the number of axons routed to the contralateral optic nerve. For this we obtained a co-efficient of the contralateral optic nerve (CCON), by comparing the fluorescence of the contralateral optic nerve to the fluorescence of the labeled optic nerve. We found that the co-efficient of the contralateral optic nerve was not significantly different in Lamb2−/− and Lamc3−/− animals compared to the WT (WT: 0.07 ± 0.01; Lamb2−/−: 0.12 ± 0.01, p=0.07; Lamc3−/−: 0.09 ± 0.02, p=0.62) (Fig. 2H).

These results demonstrate that β2 and γ3 laminins present in the PBM contribute to the regulation of the selective decussation of RGC axons into the correct optic tract. These findings also suggest that loss of β2 and γ3 laminins does not result in an increased number of RGC axons misrouted into the contralateral optic nerve.

β2 laminins are necessary for the integrity of the PBM and the normal morphology of radial glia basal processes and endfeet

The midline radial glia that control RGC axonal decussation have endfeet that contact the PBM (Colello and Guillery, 1992; Marcus et al., 1995). In the developing cortex, loss of β2 laminins resulted in dysmorphic radial glia with misaligned, highly branched basal processes and poorly formed endfeet (Radner et al., 2013). Together with these defects in the radial glia, the absence of β2 laminins resulted in fragmentation of the inner limiting membrane (ILM) in the retina and the PBM in the cortex (Pinzon-Duarte et al., 2010; Radner et al., 2013).

Based on these findings, we determined if similar defects were present in the PBM and radial glia of the optic chiasm. We asked if loss of β2 laminins compromised the integrity of the PBM and altered the morphology of the midline radial glia. We immunolabeled optic chiasm sections from WT and Lamb2−/− embryos at E15.5 and E17.5 to examine the integrity of the PBM and the shape of the radial glia basal processes and endfeet. At E15.5, the WT PBM is a continuous sheet labeled by Nidogen immunoreactivity at the pial surface (Fig. 3A and C). The radial glia basal processes extend, parallel to each other, to contact the PBM with their endfeet (Fig. 3B and C). In contrast, in the Lamb2−/− optic chiasm, the PBM is fractured and only apparent in patches (Fig. 3G). Below the fragmented PBM, the pial blood vessels are prominent, as shown by Nidogen immunostaining (Fig. 3G). The radial glial basal processes are disorganized and do not run parallel to one another toward the PBM. Instead, the basal processes are undulated and overlap with neighboring basal processes (Fig. 3H). The basal processes also appear to lack defined endfeet and attach to fragments of the PBM and the pial blood vessels (Fig. 3H and I).

Figure 3. Loss of β2 laminins leads to fragmentation of the pial basement membrane and radial glia with abnormal basal processes and endfeet.

Figure 3.

Open in a new tab

(A-T) Organization of the PBM (red) and radial glia basal processes (green) at E15.5 and E17.5. The PBM fragmentation at E17.5 is shown using Nidogen and pan-Laminin antibodies. (P, Q, U and V) Quantification of the width in Nidogen staining and pan-laminin staining (W-BB) Magnified and higher resolution view of the dashed square in panels N and U showing the fragmentation of the PBM and disrupted elaboration of the radial glia endfeet in the Lamb2−/− chiasm. (X’, Y’, AA’, BB’) 3D-volume reconstruction from confocal Z-stacks shown in T, U, Y and Z. N=3 embryos per experiment at each age.

At E17.5, the WT PBM is a continuous, thicker sheet of Nidogen and pan-Laminin immunoreactivity at the pial surface (Fig. 3D and M). The WT radial glia basal processes are elongated and organized nearly parallel to one another (Fig. 3E and N). The basal processes terminate in mature, mallet-shaped endfeet that attach to the PBM (Fig. 3E, F, N, and O). In contrast, in the Lamb2−/− optic chiasm, the PBM is fragmented, and lacks the continuity of Nidogen and pan-Laminin immunoreactivity as in the WT (Fig. 3J and R). Instead, the pia mater under the remnants of the fragmented PBM exhibit increased Nidogen and pan-Laminin immunoreactivity, giving the fragmented PBM and the pia a thickened and ruffled appearance (Fig. 3J and R). We confirmed this observation by measuring the width of the Nidogen and pan-Laminin staining comprising the fragmented PBM and pia at E17.5. The layers staining with Nidogen and pan-Laminin antibodies are thicker in the Lamb2−/− optic chiasm than in the wild-type (Fig. 3P: Lamb2−/− 22.84 μm ± 1.5 μm vs WT 7.48 μm ± 0.54 μm; Fig. 3Q: Lamb2−/− 0.07 μm2 ± 0.006 μm2 vs WT 0.02 μm ± 0.001 μm2 for pan-Laminin; and Fig. 3U: Lamb2−/− 23.83 ± 0.83 μm vs WT 6.72 μm ± 1.29 μm and V: Lamb2−/− 0.07 μm2 ± 0.001 μm2 vs WT 0.02 ± 0.003 μm2 for Nidogen). In addition to the changes in the PBM, the radial glial basal processes appear less parallel to one another, with underdeveloped endfeet that lack their typical mallet-shaped morphology (Fig. 3K, L, S and T). We developed a classification of the defects observed in the radial glia and PBM to summarize our observations (Table 1).

Table 1.

Classification of optic chiasm defects observed in Lamb2−/− embryos.

Classification Chiasm defect Phenotype WT KO
Category 1 Normal Radial glia basal processes are parallel to one another and PBM is a continuous sheet of Nidogen/EHS Laminin 6/6
100%		 0/6
0%
Category 2 Abnormal radial glia Basal processes less parallel to one another 0/6
0%		 6/6
100%
Category 3 Abnormal radial glia endfeet Endfoot termination lacks typical mallet shape 0/6
0%		 6/6
100%
Category 4 Fragmented/expanded PBM PBM is not a continuous layer and appears ruffled and thickened 0/6
0%		 6/6
100%
Open in a new tab

The morphological defects in the radial glia and PBM were scored in 6 WT and 6 Lamb2−/− embryos at E17.5 stained for Nestin, and either Nidogen or Laminin as a PBM marker.

Next, we further investigated the fragmentation of the PBM and the elaboration of endfeet. We acquired higher-magnification, higher-resolution images of the radial glial endfeet and PBM in the demarcated area (dashed white square) of the images shown in panels M-O and R-T (Fig. 3W, X, Y, Z, AA and BB). We then used these images to generate 3D-reconstructions of the images to better visualize the endfeet and PBM (Fig. 3X’, Y’, AA’ and BB’). At higher magnification and resolution, the PBM in the WT is a continuous sheet in which the mallet-shaped endfeet of the radial glia are embedded (Fig. 3W, X, Y and Y’). The endfeet attach to the PBM in close apposition to one another, forming the glia limitans (Fig. 3X and X’). The PBM in the Lamb2−/− chiasm is fragmented and nearly absent with prominent pan-Laminin staining observed in the pia (3Z, BB, and BB’). The radial glia endfeet in the Lamb2−/− chiasm terminate in abnormally shaped endfeet that fasciculate with each other to form clumps (Fig. 3AA, BB and AA’). These clumped endfeet do not form a continuous glia limitans as in the WT (Fig.3AA and AA’). Despite their abnormal morphology, the endfeet of the Lamb2−/− radial glia attach to fragments of the PBM and what appear to be blood vessels the pia mater (Fig. 3BB and BB’).

In summary, our results demonstrate that β2 laminins maintain the organization of the PBM and disruption of this organization results in abnormal radial glia basal processes that overlap each other and have poorly elaborated endfeet. Our findings suggest that the organization of the PBM and the radial glia that direct axon guidance is dependent on the presence of β2 laminins.

The expression of laminin receptors in the radial glia is reduced in the Lamb2−/− optic chiasm

The disruption of the PBM and changes to the radial glia basal processes and endfeet led us to investigate if expression of the laminin receptors β-Dystroglycan (β-Dag) and β1-Integrin was reduced in the Lamb2−/− optic chiasm compared to the WT. Radial glia express β-Dag and β1-integrin (Loulier et al., 2009; Myshrall et al., 2012). Additionally, conditional deletion of these receptors in radial glia results in fragmentation of the PBM and dysmorphic radial glial basal processes that detach from the PBM (Graus-Porta et al., 2001; Myshrall et al., 2012; Satz et al., 2010). Given the similar changes in the radial glia and PBM in Lamb2−/− embryos, and because laminins regulate the expression of their own receptors (Condic and Letourneau, 1997), we asked if β1-Integrin and β-Dag expression was altered in the radial glia of the Lamb2−/− optic chiasm.

The radial glia in the WT optic chiasm expressed β-Dag and β1-Integrin with strong immunolabel detected in the ventricular zone surrounding the radial glia cell bodies and in the glial endfeet that abut the PBM (Fig. 4A–D). We also detected β-Dag and β1-integrin expression in the blood vessels and axons, which were excluded from our measurements (Fig. 4A–D). In the Lamb2−/− optic chiasm, we observed reduced expression of β-Dag and β1-integrin expression. To confirm this observation, we measured the fluorescent intensity of the β-Dag and β1-integrin immunolabel and normalized it to the expression of Nestin in the radial glia. We found that β-Dag and β1-integrin expression was reduced in the radial glia cell bodies and processes (Fig. 4F–I; N=4). We found that the reduction is a modest 15% for β1-Integrin and 13% for β-Dystroglycan, but both are significantly different from the WT (Fig. 4E and J; N=4; WT: 100% and Lamb2−/−: 85.46 ± 4.64% for β1-Integrin and WT: 100% and Lamb2−/−: 87.49 ± 4.43% for β-Dag). These results show that genetic deletion of β2 laminins modestly reduces the expression of two canonical laminin receptors in the radial glia of the optic chiasm. These findings suggest that the structural changes in the radial glia of the Lamb2−/− optic chiasm may be due, at least in part, to the partial reduction of the expression of these receptors.

Figure 4. β1-Integrin and β-Dag expression is reduced in the absence of β2 laminins.

Figure 4.

Open in a new tab

(A, B, and F, G) Comparison of β-Dag protein expression (red) in WT and Lamb2−/− radial glia (green). (C, D, and H, I) Comparison of β1-Integrin expression (red) in WT and Lamb2−/− radial glia (green). (E) Quantification of the percent of β-Dag fluorescence intensity in the radial glia. (J) Quantification of the percent of β1-Integrin fluorescence intensity in the radial glia. N= 4 mice per genotype.

EphrinB2 expression persists into the late phase of retinal ganglion cell axon decussation in the absence of β2 laminins

EphrinB2 is expressed in radial glia, and its peak of expression is at E15.5 when most of the ipsilateral and contralateral RGCs are extending axons to the optic chiasm (Williams et al., 2003). EphrinB2 expression is reduced by E17.5, the late phase of axon decussation, when the remaining RGC axons entering the optic chiasm project contralaterally (Williams et al., 2003). We found that the size of the ipsilateral projection is increased in the Lamb2−/− mice at P4. The increase in the ipsilateral projection is accompanied by changes in the radial glia basal processes and endfeet. We hypothesized that these changes in the optic chiasm could impact the production of the axon guidance cues in the radial glia. This prompted us to ask if the expression of the axon guidance cue EphrinB2 was changed in the Lamb2−/− radial glia.

We found that EphrinB2 protein expression was minimal in the E17.5 RC2-positive radial glia of the WT chiasm, with low levels in the basal processes as expected (Fig. 5A–C and I). This result is consistent with previous findings that EphrinB2 expression diminishes in the chiasm during the late phase of RGC axonal decussation (Williams et al., 2003). Surprisingly, in the Lamb2−/− chiasm, we found that high EphrinB2 expression persisted in the Nestin-positive radial glia into the late phase of axon decussation and that it was 70% higher than in the WT (WT: 100% and Lamb2−/−: 170 ± 9.86%; Fig. 5E–G and I). These results show that β2 laminins, likely through their role in radial glia endfoot adhesion, play a role in the downregulation of EphrinB2 protein during the shift from the peak to the late phase of RGC axon guidance.

Figure 5. EphrinB2 protein expression persists in the radial glia during the late phase of RGC axon decussation.

Figure 5.

Open in a new tab

(A-C and E-G) EphrinB2 protein expression (red) in WT and Lamb2−/− radial glia (green) at E17.5, which corresponds to the late phase of RGC axonal decussation. N=4 mice used per genotype. (D and H) EphrinB2 in situ fluorescent hybridization (magenta) in optic chiasm sections with nuclei labeled with Hoechst (blue). N=4 mice used per genotype. (I) Quantification of EphrinB2 fluorescence intensity in WT and Lamb2−/− radial glia. (J) Quantification of mRNA single molecule spots in the radial glia cell bodies. (K) Quantification of mRNA single molecule spots in the radial glia basal processes. N=4 mice used per genotype. (L) EphrinB2 quantitative reverse transcription PCR (qRT-PCR) from optic chiasm tissues at E17.5. N=3 mice per genotype.

Next, we determined if the observed persistence of EphrinB2 protein expression was at the transcript level. We performed Hybridization Chain Reaction Fluorescent In Situ Hybridization (HCR-FISH) in anterior, mid and posterior optic chiasm sections. In the WT anterior chiasm, we detected EphrinB2 transcripts in the ventricular zone of the third ventricle where the radial glia nuclei are located, and in the processes of the radial glia (Fig. 5D) as previously described (Williams et al., 2003). We did not find a significant difference in the number of single mRNA molecule spots in the area containing the radial glia cell bodies or the basal processes of the anterior WT and Lamb2−/− optic chiasm (Fig. 5D, H, J and K). We also quantified the number of EphrinB2 transcripts in the middle and posterior chiasm and found no significant difference in the ventricular zone, basal processes or pia (Supp. Fig. 2A and B, D–G, and not shown). We did not detect EphrinB2 transcripts in sections that were not incubated with HCR probes (Supp. Fig. 2C) Standard in situ hybridization seemed to agree with the HCR-FISH results (Supp. Fig. 2 H–N).

We further verified the HCR-FISH results with quantitative reverse transcription PCR (qRT-PCR) from the mid-diencephalic region containing the optic chiasm at E17.5 (Fig. 5L). We found that EphrinB2 expression was not significantly different between the WT and the Lamb2−/− chiasm, consistent with our HCR-FISH findings. These results demonstrate that the observed persistence of EphrinB2 expression is due to persistence of protein expression that is not correlated with increased transcript levels. Together, these results suggest that β2 laminins influence EphrinB2 expression at the protein level. β2 laminins, likely indirectly through the regulation of radial glia endfoot attachment, may regulate the translation, stability or degradation of EphrinB2.

The number of Zic2-positive retinal ganglion cells is increased in the absence of β2 laminins

Zic2 specifies the ventrotemporal RGCs that project ipsilaterally, and its expression is restricted to the peak phase of axon segregation (Herrera et al., 2003). We considered that the increased proportion of ipsilateral axons in the Lamb2−/− mutants could result from an increase in the number of ipsilateral RGCs. We tested this possibility by immunolabeling and counting the number of Zic2-positive ipsilateral RGCs, Brn3a-positive contralateral RGCs, and Zic2 plus Brn3a double-positive cells in the ventrotemporal retina at E15.5 and E17.5 (Quina et al., 2005; Xiang et al., 1995). The counts were performed double-blind to the genotype of the animals.

At E15.5, the peak phase of RGC birth and axon segregation, the number of Zic2 (WT: 29.13±2.242 vs Lamb2−/−: 26.2±3.322; p=0.3670), Brn3a (WT: 13.62±1.578 vs Lamb2−/−: 10.87±0.8338; p=0.1355) and Zic2 plus Brn3a positive cells (WT: 3.43±0.6160 vs Lamb2−/−: 2.94±0.3025; p=0.4814) in the Lamb2−/− ventrotemporal retina is the same as in WT (Fig. 6A–I). We did not find a significant difference in the number of nuclei in the differentiated cell layer of the ventrotemporal retina (Fig. 6J) (WT: 95.68±2.75 vs Lamb2−/−: 99.42±1.528; p=0.2478).

Figure 6. The number of ipsilateral RGCs is increased in the Lamb2−/− ventrotemporal retina.

Figure 6.

Open in a new tab

(A-C and F-H) Zic2 positive ipsilateral (green) and Brn3a positive contralateral RGCs (red) in the ventrotemporal retina at E15.5. Nuclei are labeled with Hoechst (blue). (K-M and P-R) Zic2 (green) and Brn3a (red) RGCs in the ventrotemporal retina at E17.5. (D and N) Quantification of Zic2-positive cells. (E and O) Quantification of Brn3a-positive cells. (I and S) Quantification of Zic2 and Brn3a double-positive cells. (J and T) Quantification of Hoechst-labeled nuclei in the differentiated cell layer of the ventrotemporal retina. N=4 mice per genotype.

Next, we examined the ventrotemporal retina at E17.5, when the last RGCs born in this region project contralaterally. Surprisingly, the number of Zic2 positive RGCs is significantly increased in the Lamb2−/− ventrotemporal retina compared to the WT (WT: 35±1.861 vs Lamb2−/−: 46±1.68; p<0.0001) (Fig. 6K, N, and P). In contrast, the percentage of Brn3a (WT: 34.87±2.230 vs Lamb2−/−: 30.49±1.241; p=0.093) or Brn3a/Zic2 (WT: 6.835±0.6604 vs Lamb2−/−: 6.849 ±0.5542; p=0.9875) double positive cells was not increased (Fig. 6L, M, O, Q, R and S). Additionally, as at E15.5, the total number of nuclei in the ventrotemporal retina did not change in the Lamb2−/− retinas compared to the WT (WT: 110.3±2.547 vs Lamb2−/−: 104.2±2.615; p=0.0999) (Fig. 6T). These results suggest that, in the ventrotemporal retina, β2 laminins are necessary for the generation of the correct number of ipsilateral RGCs. Like in the optic chiasm, in the retina, the maintenance of RPC attachment to the ILM by β2 laminins, may impact the laterality fate and neurogenesis of ipsilateral RGCs.

Discussion

We provide evidence that β2 laminins in the developing optic chiasm and retina play a role in regulating the size of the ipsilateral projection. In the optic chiasm, β2 laminins maintain the integrity of the PBM, the organization of radial glia basal processes, and proper elaboration of radial glia endfeet that attach to the PBM. There is also a persistence of EphrinB2 protein in radial glia during the late phase of axonal decussation. Our findings suggest that the restriction of EphrinB2 expression to the peak phase of axon guidance may be regulated by β2 laminins indirectly through the attachment of the midline glia endfeet to the PBM. We also show that, in the retina, β2 laminins may indirectly regulate the generation of the correct number ipsilateral RGCs, possibly by mediating the adhesion of RPCs to the ILM. Together, our findings uncover a role for β2 laminins, a key component of basement membranes, in the regulation of the ipsilateral projection.

β2 laminins in the regulation of EphrinB2 expression

β2 laminins have roles in maintaining the structural integrity of the PBM and adhesion of the radial glia scaffold, as well as in regulation of temporal expression of EphrinB2. Persistence of EphrinB2 protein into the late phase of axonal decussation suggests that that β2 laminins are part of a mechanism that terminates EphrinB2 expression. β2 laminins could regulate EphrinB2 protein expression indirectly by mediating endfoot adhesion or directly by signaling through its receptors in the radial glia. The regulation of EphrinB2 protein levels by β2 laminins may be at the translational or post-translational level.

At the level of translation, β2 laminins, either indirectly or directly, could interfere with local translation of EphrinB2 at the radial glia endfeet. Radial glia actively translate mRNAs at their endfeet (Pilaz et al 2016; Pilaz et al 2023) and β2 laminins are known to produce extracellular cues that regulate local translation in growth cones (Rhatod et al 2012). β2 laminin-mediated attachment of the radial glia endfeet to the PBM may localize other factors that stop EphrinB2 translation at the endfeet. However, we did not observe accumulation of EphrinB2 mRNA at the endfeet by HCR-FISH (Fig. 5). This argues against the modification of local translation by β2 laminins in regulating EphrinB2 protein levels.

EphrinB2 is regulated post-translationally by proteolysis of the extracellular domain, endocytosis, or by ubiquitin mediated degradation (Reviewed in Sullivan and Bashaw, 2023). The EphrinB2 extracellular domain is cleaved by ADAM10 (Hattori et al 2000; Janes et al 2005; Ji et al 2014) and ADAM10 interacts directly with the β2 laminin chain (Gilpin et al., 1998) and indirectly through binding of the tetraspanin CD151 that interacts with the laminin-binding integrins α3β1 and α6β1 (Andre et al., 2006; Le Naour et al., 2006).

Loss of β2 laminins may impair localization of ADAM10 and cleavage of EphrinB2, resulting in its persistence during the late phase of axonal decussation. Future experiments will determine if β2 laminins, through modulation of endfoot adhesion or laminin signaling regulate EphrinB2 protein levels.

β2 laminins in the generation of ipsilateral RGCs

RPCs have basal processes that attach to the ILM (Saito et al., 2003; Serjanov et al., 2018). The observed increase in the number of ipsilateral RGCs in the Lamb2−/− retina may be a consequence of detachment of these processes from the ILM. Loss of β2 laminins leads to fragmentation of the ILM, and subsequent detachment of RPC basal processes (Pinzon-Duarte et al., 2010; Serjanov et al., 2018). Detachment of RPCs from the ILM results in a shift in the mitotic axis, accompanied by decreased RPC proliferation, and a change in cell fate that produces an excess of rods at the expense of other late-born cell types in the post-natal mouse retina (Serjanov et al., 2018).

Ipsilateral RGCs are generated from progenitors in the neural retina and the CMZ (Herrera et al., 2003; Marcucci et al., 2016; Slavi et al., 2023). In the Lamb2−/− retina, detachment of RPC basal processes from the ILM and the resulting shift in mitotic axis might directly alter the laterality fate of daughter RGCs. The shift to asymmetric divisions would lead to more progenitors acquiring an ipsilateral fate. In the CMZ, attachment of RPCs to the ILM may increase the generation of ipsilateral RGCs from CyclinD2-expressing progenitors. The loss of attachment in the Lamb2−/− retina may increase the number of CyclinD2 progenitors and the proportion of ipsilateral RGCs produced from this population. We are currently testing for potential links between ILM integrity and RPC attachment to the ILM to the acquisition of laterality fate.

Zic2 is a transcription factor that determines ipsilateral fate in RGCs (Herrera et al., 2003), and its expression must be downregulated at the correct time in development to limit the number of RPCs that adopt an ipsilateral fate. The attachment of progenitor basal processes to the ILM may control the timing of the expression of Zic2 expression to produce the correct number of ipsilateral RGCs. Loss of β2 laminin would thereby impair this downregulation resulting in overproduction of ipsilateral RGCs. Loss of β2 laminins may result in persistence of Zic2 expression in RGCs without driving the production of additional ipsilateral RGCs. Future studies at later developmental time points will determine whether Zic2 expression persists without a change in the number of ipsilateral RGCs.

Requirement of β2 laminin function

β2 laminins are expressed in the retina and the optic chiasm. Our experiments did not test whether the function of β2 laminins in RGC axon decussation lies predominantly in the retina or the optic chiasm. Control of RGC laterality fate acquisition may be the driver of increased ipsilateral projection as a modulation of ipsilateral fate affects the entire population of ipsilateral RGCs. Future experiments using conditional knockouts of β2 laminins specifically in the retina or optic chiasm will further dissect the function of these laminins at each site.

Supplementary Material

Supplementary Figure 1

Supplementary Figure 1. β2 laminin is not detected in Lamb2−/− embryos. (A and B) Lamb2 protein (green) is detected in the PBM underlining the axons (red), ventricular zone, and blood vessels of the WT optic chiasm. (C and D) Lamb2 protein is not detected in the optic chiasm of Lamb2−/− embryos. Nuclei were counterstained with Hoechst. N=2 animals per genotype.

Supplementary Figure 2

Supplementary Figure 2. EphrinB2 mRNA expression in the mid and posterior optic chiasm. (A, B, D and E) EphrinB2 hybridization chain reaction fluorescent in situ hybridization (HCR-FISH) (magenta). Nuclei are labeled with Hoechst (blue). N=4 mice per genotype. (C) Negative control HCR-FISH incubated with the EphrinB2 probe but without any hairpin amplification. N=2 mice per genotype. (F and G) Quantification of mRNA single molecules in the radial glia cell bodies and basal processes in mid and posterior chiasm. (H-M) EphrinB2 standard in situ hybridization using an anti-sense probe. N=4 mice per genotype. (N) Control standard in situ hybridization using a sense EphrinB2 probe. N=2 embryos of each genotype.

Supplementary Movie 1

Supplementary movie 1. The PBM and radial glia endfeet in the WT optic chiasm. A 15° inclining view of the PBM and radial glia endfeet 3D reconstruction from Z-stacks of Nestin and pan-Laminin immunostained sections of the WT optic chiasm. The PBM is continuous, and the mallet-shaped radial glia endfeet connect to form the glia limitans.

Supplementary Movie 2

Supplementary movie 2. The PBM and radial glia endfeet in the Lamb2−/− optic chiasm. A 15° inclining view of the PBM and radial glia endfeet 3D reconstruction from Z-stacks of Nestin and pan-Laminin immunostained sections of the Lamb2−/− optic chiasm. The PBM is fragmented and nearly absent. The radial glia endfeet have lost their characteristic mallet shape and are clumped into small groups.

Supplementary Table 1

Supplementary Table 1. Key Resources Table.

Acknowledgements

We thank members of the Brunken, Viczian, and Zuber labs for helpful discussions and their insights. We also thank Dr. William Brunken for providing the Lamb2−/− and Lamc3−/− mice, reagents, and his laboratory to get this project started. Drs. Dale Hunter and Michael E. Zuber for critical reading of this manuscript, and Francisco J. Agosto-Perez for the Python script to blind samples for analysis. This study was supported by a NIH Research Supplement to Promote Diversity for RMDL to NEI R01 EY12676 (PI William J. Brunken), Research to Prevent Blindness Career Development Award to RIMDL, Fight for Sight Grant-in-Aid to RIMDL, a NEI R01 EY034653 to RIMDL, and an Unrestricted Grant to the Department of Ophthalmology from Research to Prevent Blindness.

References:

  1. Barresi MJ, Burton S, Dipietrantonio K, Amsterdam A, Hopkins N, Karlstrom RO, 2010. Essential genes for astroglial development and axon pathfinding during zebrafish embryogenesis. Dev Dyn 239, 2603–2618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Birgbauer E, Cowan CA, Sretavan DW, Henkemeyer M, 2000. Kinase independent function of EphB receptors in retinal axon pathfinding to the optic disc from dorsal but not ventral retina. Development 127, 1231–1241. [DOI] [PubMed] [Google Scholar]
  3. Brittis PA, Canning DR, Silver J, 1992. Chondroitin sulfate as a regulator of neuronal patterning in the retina. Science 255, 733–736. [DOI] [PubMed] [Google Scholar]
  4. Clements R, Wright KM, 2018. Retinal ganglion cell axon sorting at the optic chiasm requires dystroglycan. Dev Biol 442, 210–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Colello RJ, Guillery RW, 1992. Observations on the early development of the optic nerve and tract of the mouse. J Comp Neurol 317, 357–378. [DOI] [PubMed] [Google Scholar]
  6. Condic ML, Letourneau PC, 1997. Ligand-induced changes in integrin expression regulate neuronal adhesion and neurite outgrowth. Nature 389, 852–856. [DOI] [PubMed] [Google Scholar]
  7. Conway CD, Howe KM, Nettleton NK, Price DJ, Mason JO, Pratt T, 2011. Heparan sulfate sugar modifications mediate the functions of slits and other factors needed for mouse forebrain commissure development. J Neurosci 31, 1955–1970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fries M, Brown TW, Jolicoeur C, Boulan B, Boudreau-Pinsonneault C, Javed A, Abram P, Cayouette M, 2023. Pou3f1 orchestrates a gene regulatory network controlling contralateral retinogeniculate projections. Cell Rep 42, 112985. [DOI] [PubMed] [Google Scholar]
  9. Graus-Porta D, Blaess S, Senften M, Littlewood-Evans A, Damsky C, Huang Z, Orban P, Klein R, Schittny JC, Muller U, 2001. Beta1-class integrins regulate the development of laminae and folia in the cerebral and cerebellar cortex. Neuron 31, 367–379. [DOI] [PubMed] [Google Scholar]
  10. Herrera E, Brown L, Aruga J, Rachel RA, Dolen G, Mikoshiba K, Brown S, Mason CA, 2003. Zic2 patterns binocular vision by specifying the uncrossed retinal projection. Cell 114, 545–557. [DOI] [PubMed] [Google Scholar]
  11. Hunter DD, Llinas R, Ard M, Merlie JP, Sanes JR, 1992. Expression of s-laminin and laminin in the developing rat central nervous system. J Comp Neurol 323, 238–251. [DOI] [PubMed] [Google Scholar]
  12. Karlstrom RO, Trowe T, Klostermann S, Baier H, Brand M, Crawford AD, Grunewald B, Haffter P, Hoffmann H, Meyer SU, Muller BK, Richter S, van Eeden FJ, Nusslein-Volhard C, Bonhoeffer F, 1996. Zebrafish mutations affecting retinotectal axon pathfinding. Development 123, 427–438. [DOI] [PubMed] [Google Scholar]
  13. Kuwajima T, Soares CA, Sitko AA, Lefebvre V, Mason C, 2017. SoxC Transcription Factors Promote Contralateral Retinal Ganglion Cell Differentiation and Axon Guidance in the Mouse Visual System. Neuron 93, 1110–1125 e1115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kuwajima T, Yoshida Y, Takegahara N, Petros TJ, Kumanogoh A, Jessell TM, Sakurai T, Mason C, 2012. Optic chiasm presentation of Semaphorin6D in the context of Plexin-A1 and Nr-CAM promotes retinal axon midline crossing. Neuron 74, 676–690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lee R, Petros TJ, Mason CA, 2008. Zic2 regulates retinal ganglion cell axon avoidance of ephrinB2 through inducing expression of the guidance receptor EphB1. J Neurosci 28, 5910–5919. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Li YN, Radner S, French MM, Pinzon-Duarte G, Daly GH, Burgeson RE, Koch M, Brunken WJ, 2012. The gamma3 chain of laminin is widely but differentially expressed in murine basement membranes: expression and functional studies. Matrix Biol 31, 120–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Libby RT, Champliaud MF, Claudepierre T, Xu Y, Gibbons EP, Koch M, Burgeson RE, Hunter DD, Brunken WJ, 2000. Laminin expression in adult and developing retinae: evidence of two novel CNS laminins. J Neurosci 20, 6517–6528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lin L, Wang J, Chan CK, Chan SO, 2007. Effects of exogenous hyaluronan on midline crossing and axon divergence in the optic chiasm of mouse embryos. Eur J Neurosci 26, 1–11. [DOI] [PubMed] [Google Scholar]
  19. Loulier K, Lathia JD, Marthiens V, Relucio J, Mughal MR, Tang SC, Coksaygan T, Hall PE, Chigurupati S, Patton B, Colognato H, Rao MS, Mattson MP, Haydar TF, Ffrench-Constant C, 2009. beta1 integrin maintains integrity of the embryonic neocortical stem cell niche. PLoS Biol 7, e1000176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Marcucci F, Murcia-Belmonte V, Wang Q, Coca Y, Ferreiro-Galve S, Kuwajima T, Khalid S, Ross ME, Mason C, Herrera E, 2016. The Ciliary Margin Zone of the Mammalian Retina Generates Retinal Ganglion Cells. Cell Rep 17, 3153–3164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Marcucci F, Soares CA, Mason C, 2019. Distinct timing of neurogenesis of ipsilateral and contralateral retinal ganglion cells. J Comp Neurol 527, 212–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Marcus RC, Blazeski R, Godement P, Mason CA, 1995. Retinal axon divergence in the optic chiasm: uncrossed axons diverge from crossed axons within a midline glial specialization. J Neurosci 15, 3716–3729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Martinez-De Luna RI, Ku RY, Lyou Y, Zuber ME, 2013. Maturin is a novel protein required for differentiation during primary neurogenesis. Dev Biol 384, 26–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Myshrall TD, Moore SA, Ostendorf AP, Satz JS, Kowalczyk T, Nguyen H, Daza RA, Lau C, Campbell KP, Hevner RF, 2012. Dystroglycan on radial glia end feet is required for pial basement membrane integrity and columnar organization of the developing cerebral cortex. J Neuropathol Exp Neurol 71, 1047–1063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Noakes PG, Gautam M, Mudd J, Sanes JR, Merlie JP, 1995. Aberrant differentiation of neuromuscular junctions in mice lacking s-laminin/laminin beta 2. Nature 374, 258–262. [DOI] [PubMed] [Google Scholar]
  26. Pak W, Hindges R, Lim YS, Pfaff SL, OĽeary DD, 2004. Magnitude of binocular vision controlled by islet-2 repression of a genetic program that specifies laterality of retinal axon pathfinding. Cell 119, 567–578. [DOI] [PubMed] [Google Scholar]
  27. Paulus JD, Halloran MC, 2006. Zebrafish bashful/laminin-alpha 1 mutants exhibit multiple axon guidance defects. Dev Dyn 235, 213–224. [DOI] [PubMed] [Google Scholar]
  28. Petros TJ, Rebsam A, Mason CA, 2008. Retinal axon growth at the optic chiasm: to cross or not to cross. Annu Rev Neurosci 31, 295–315. [DOI] [PubMed] [Google Scholar]
  29. Pinzon-Duarte G, Daly G, Li YN, Koch M, Brunken WJ, 2010. Defective formation of the inner limiting membrane in laminin beta2- and gamma3-null mice produces retinal dysplasia. Invest Ophthalmol Vis Sci 51, 1773–1782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pratt T, Conway CD, Tian NM, Price DJ, Mason JO, 2006. Heparan sulphation patterns generated by specific heparan sulfotransferase enzymes direct distinct aspects of retinal axon guidance at the optic chiasm. J Neurosci 26, 6911–6923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Quina LA, Pak W, Lanier J, Banwait P, Gratwick K, Liu Y, Velasquez T, OĽeary DD, Goulding M, Turner EE, 2005. Brn3a-expressing retinal ganglion cells project specifically to thalamocortical and collicular visual pathways. J Neurosci 25, 11595–11604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Radner S, Banos C, Bachay G, Li YN, Hunter DD, Brunken WJ, Yee KT, 2013. beta2 and gamma3 laminins are critical cortical basement membrane components: ablation of Lamb2 and Lamc3 genes disrupts cortical lamination and produces dysplasia. Dev Neurobiol 73, 209–229. [DOI] [PubMed] [Google Scholar]
  33. Saito K, Kawaguchi A, Kashiwagi S, Yasugi S, Ogawa M, Miyata T, 2003. Morphological asymmetry in dividing retinal progenitor cells. Dev Growth Differ 45, 219–229. [DOI] [PubMed] [Google Scholar]
  34. Sasaki T, Mann K, Miner JH, Miosge N, Timpl R, 2002. Domain IV of mouse laminin beta1 and beta2 chains. Eur J Biochem 269, 431–442. [DOI] [PubMed] [Google Scholar]
  35. Satz JS, Ostendorf AP, Hou S, Turner A, Kusano H, Lee JC, Turk R, Nguyen H, Ross-Barta SE, Westra S, Hoshi T, Moore SA, Campbell KP, 2010. Distinct functions of glial and neuronal dystroglycan in the developing and adult mouse brain. J Neurosci 30, 14560–14572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A, 2012. Fiji: an open-source platform for biological-image analysis. Nat Methods 9, 676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Schmittgen TD, Livak KJ, 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 3, 1101–1108. [DOI] [PubMed] [Google Scholar]
  38. Serjanov D, Bachay G, Hunter DD, Brunken WJ, 2018. Laminin beta2 Chain Regulates Retinal Progenitor Cell Mitotic Spindle Orientation via Dystroglycan. J Neurosci 38, 5996–6010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Slavi N, Balasubramanian R, Lee MA, Liapin M, Oaks-Leaf R, Peregrin J, Potenski A, Troy CM, Ross ME, Herrera E, Kosmidis S, John SWM, Mason CA, 2023. CyclinD2-mediated regulation of neurogenic output from the retinal ciliary margin is perturbed in albinism. Neuron 111, 49–64 e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sretavan DW, Feng L, Pure E, Reichardt LF, 1994. Embryonic neurons of the developing optic chiasm express L1 and CD44, cell surface molecules with opposing effects on retinal axon growth. Neuron 12, 957–975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Williams SE, Grumet M, Colman DR, Henkemeyer M, Mason CA, Sakurai T, 2006. A role for Nr-CAM in the patterning of binocular visual pathways. Neuron 50, 535–547. [DOI] [PubMed] [Google Scholar]
  42. Williams SE, Mann F, Erskine L, Sakurai T, Wei S, Rossi DJ, Gale NW, Holt CE, Mason CA, Henkemeyer M, 2003. Ephrin-B2 and EphB1 mediate retinal axon divergence at the optic chiasm. Neuron 39, 919–935. [DOI] [PubMed] [Google Scholar]
  43. Xiang M, Zhou L, Macke JP, Yoshioka T, Hendry SH, Eddy RL, Shows TB, Nathans J, 1995. The Brn-3 family of POU-domain factors: primary structure, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons. J Neurosci 15, 4762–4785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yurchenco PD, 2011. Basement membranes: cell scaffoldings and signaling platforms. Cold Spring Harb Perspect Biol 3. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figure 1

Supplementary Figure 1. β2 laminin is not detected in Lamb2−/− embryos. (A and B) Lamb2 protein (green) is detected in the PBM underlining the axons (red), ventricular zone, and blood vessels of the WT optic chiasm. (C and D) Lamb2 protein is not detected in the optic chiasm of Lamb2−/− embryos. Nuclei were counterstained with Hoechst. N=2 animals per genotype.

NIHMS2196640-supplement-Supplementary_Figure_1.tif (12.1MB, tif)
Supplementary Figure 2

Supplementary Figure 2. EphrinB2 mRNA expression in the mid and posterior optic chiasm. (A, B, D and E) EphrinB2 hybridization chain reaction fluorescent in situ hybridization (HCR-FISH) (magenta). Nuclei are labeled with Hoechst (blue). N=4 mice per genotype. (C) Negative control HCR-FISH incubated with the EphrinB2 probe but without any hairpin amplification. N=2 mice per genotype. (F and G) Quantification of mRNA single molecules in the radial glia cell bodies and basal processes in mid and posterior chiasm. (H-M) EphrinB2 standard in situ hybridization using an anti-sense probe. N=4 mice per genotype. (N) Control standard in situ hybridization using a sense EphrinB2 probe. N=2 embryos of each genotype.

NIHMS2196640-supplement-Supplementary_Figure_2.tif (20.2MB, tif)
Supplementary Movie 1

Supplementary movie 1. The PBM and radial glia endfeet in the WT optic chiasm. A 15° inclining view of the PBM and radial glia endfeet 3D reconstruction from Z-stacks of Nestin and pan-Laminin immunostained sections of the WT optic chiasm. The PBM is continuous, and the mallet-shaped radial glia endfeet connect to form the glia limitans.

NIHMS2196640-supplement-Supplementary_Movie_1.mov (137.3MB, mov)
Supplementary Movie 2

Supplementary movie 2. The PBM and radial glia endfeet in the Lamb2−/− optic chiasm. A 15° inclining view of the PBM and radial glia endfeet 3D reconstruction from Z-stacks of Nestin and pan-Laminin immunostained sections of the Lamb2−/− optic chiasm. The PBM is fragmented and nearly absent. The radial glia endfeet have lost their characteristic mallet shape and are clumped into small groups.

NIHMS2196640-supplement-Supplementary_Movie_2.mov (82.4MB, mov)
Supplementary Table 1

Supplementary Table 1. Key Resources Table.

NIHMS2196640-supplement-Supplementary_Table_1.docx (24.5KB, docx)

RESOURCES