EphrinB2 sharpens lateral motor column division in the developing spinal cord
- 1.1k Downloads
During sensori-motor circuit development, the somas of motoneurons (MN) are distributed in a topographic manner in the ventral horn of the neural tube. Indeed, their position within the lateral motor columns (LMC) correlates with axonal trajectories and identity of target limb muscles. The mechanisms by which this topographic distribution is established remains poorly understood. To address this issue, we assessed the role of ephrinB2 in MN topographic organization in the developing mouse spinal cord.
First, we used a reporter mouse line to establish the spatio-temporal expression pattern of EfnB2 in the developing LMC. We show that early in LMC development, ephrinB2 is differentially expressed in MN of the lateral versus medial LMC, suggesting a possible role in MN sorting and/or migration. We demonstrate that while MN-specific excision of EfnB2 did not perturb specification or migration of MN, conditional loss of ephrinB2 led to the blurring of the LMC divisional boundary and to errors in the selection of LMC axon trajectory in the limb.
Altogether, our study uncovered a novel cell autonomous role for ephrinB2 in LMC MN thus emphasizing the prevalent role of this ephrin member in maintaining cell population boundaries.
KeywordsSpinal cord Motor neurons Motor columns Eph/ephrin Cell sorting Mouse
Green fluorescent protein
Lateral motor column
Lateral lateral motor column
Medial lateral motor column
A recurring theme in the organization of the central nervous system is the grouping of neurons innervating the same target. Because neurons are often born at a distance from their final settling position, the establishment of this topography requires complex migration and clustering processes [1, 2, 3]. In the ventral spinal cord, motoneurons (MN) are grouped in motor columns according to their identity and to their target muscle. MN innervating the limb settle in the lateral motor column (LMC) which is further divided into two divisions: lateral LMC (LMCl) composed of MN innervating the dorsal part of the limb and medial LMC (LMCm) formed by MN innervating the ventral part of the limb. Both LMCl and LMCm occupy stereotypical positions within the LMC [4, 5].
Shortly after exiting the cell cycle at the basal side of the ventricular zone of the spinal cord, MN migrate radially toward the marginal zone. A second phase of tangential migration followed by coalescence of same-identity MN soma gives rise to the stereotypical organization of motor columns in the ventral horn of the spinal cord. The different motor columns are characterized by the expression of different sets of transcription factors. For instance, all somatic MN express HB9, whereas Foxp1 is expressed in all LMC MN at high level. Lastly, LMCl and LMCm MN express Lim1 (Lhx1) and Islet1 respectively . All these transcription factors have been shown to contribute to the establishment of MN organization in columns. Indeed, gain and loss of function of Foxp1, Lim1, Islet1 and HB9 lead to important defects of MN positioning within the spinal cord along with a range of axon pathfinding defects [7, 8, 9, 10, 11, 12]. Although the role of these transcription factors in specifying the identity and position of MN within the spinal cord is well established, little is known on their potential effector genes. A handful of molecular players involved in MN soma migration have been identified in the last few years, for instance, Reelin, an extracellular protein well known for its role in controlling radial migration of cortical neurons, was shown to control tangential migration of LMC MN . Moreover, members of the cadherin family, especially type II cadherins, have also been involved in this migratory process  and a recent study identified axon guidance molecules of the Slit/Robo and Netrin/DCC pathways as repulsive and attractive cues, respectively, for MN cell bodies in the ventral spinal cord . Of note, cadherins are the only molecular players identified to date in the control of MN soma clustering [15, 16].
The Eph/ephrin family has been widely involved in mechanisms of cell sorting, cell migration and axon guidance during development . In the sensory-motor circuit innervating the limb, this family of proteins has been shown to control guidance of motor axons, sorting between motor and sensory axons and synaptogenesis [18, 19]. Amongst all members of the Eph/ephrin family, ephrin-B2 and EphA4 seem to play a particularly important role in cell sorting. For instance in the zebrafish hindbrain, the ephrinB2/EphA4 pair is responsible for the formation and maintenance of rhombomeres boundaries [20, 21] and this same pair was shown to be involved in maintaining the anteroposterior patterning of somites [22, 23]. Interestingly, rostrocaudal displacement of a MN pool innervating the hindlimb has been reported in EphA4 deficient mice  as well as LMC axon guidance defects . Concerning ephrinB2, previous work has shown that it plays a dual role in controlling LMC MN axon guidance. As a ligand expressed in the limb mesenchyme, it activates Eph signaling in growing LMCm axons thus repelling them from the dorsal limb . In addition, experiments in the chick showed that expression of ephrinB2 in LMCl attenuates Eph signaling in these axons thus allowing them to invade the dorsal limb .
Herein, we asked whether ephrinB2 regulates the migration, position and/or grouping of MN soma in the mouse spinal cord. Using the EfnB2:H2BGFP reporter mouse line, we established the spatial and temporal expression pattern of ephrinB2 in MN of the LMC. We show that ephrinB2 is differentially expressed in LMCl vs. LMCm MN at the time these two populations coalesce, with a higher expression in LMCl MN. We confirm that ephrinB2 cell autonomously controls the guidance of LMCl axons in the mouse and we provide evidence that conditional loss of EfnB2 in MN impairs clustering of LMC MN soma without affecting their migration.
Results and discussion
Dynamic expression of ephrinB2 in the ventral spinal cord
EphrinB2 is required for guidance of LMCl axons
EphrinB2 does not control LMC soma migration
EphrinB2 is required to maintain sharp segregation between LMC MN
Ordered topography of neuronal soma requires the delicate orchestration of various processes including neuronal specification, radial and tangential migration as well as soma coalescence. While mechanisms controlling specification of LMC MN are fairly well characterized, the actual molecular effectors –cytoplasmic factors, cytoskeleton proteins and cell surface receptors- involved in MN migration and grouping remain elusive. We show here that in addition to its well characterized role in guiding motor axons , Eph:ephrin signalling plays a role in setting up the topography of MN soma, lending support to the notion that identical factors control several steps of myotopic organisation.
Efnb2 +/GFP , Efnb2 lox/lox and Olig2-Cre mice were as described [23, 28, 34]. Efnb2 cKO (EfnB2 lox/H2BGFP ; Olig2-Cre) and control embryos were collected from the same litters. Control genotypes included Efnb2 lox/H2BGFP , Efnb2 +/H2BGFP , Efnb2 +/H2BGFP ; Olig2-Cre. All animal procedures were pre-approved by the “Comité d’éthique Régional” (protocol number: MP/07/21/04/11).
REDExtract-N-Amp™ Tissue PCR kit was used for all genotyping PCR. Yolk sac was used to genotype embryos while tail tissue was used to genotype pups. The following primers were used for the EfnB2 floxed allele: Cs1 5′-CTTCAGCAATATACACAGGATG-3′and Cas1 5′-TGCTTGATTGATTGAAACGAAGCCCGA-3′; for the Cre allele: Cre-S 5′- ACGGAAATCCATCGCTCGACCA-3′ and Cre-AS 5′-GTCCGGGCTGCCACGACCAA-3′. H2BGFP was detected by epifluorescence on whole embryos.
In situ hybridization
ISH was performed on transversal thick vibratome sections at brachial level of E12.5 embryos. Briefly, embryos were fixed in 4 % paraformaldehyde (PFA) and dehydrated in ethanol. Following rehydration, embryos were sectioned and 70 μm sections at brachial level were treated with proteinase K (10 μg/ml in PBS/0.1 % Tween-20) for 7 min at room temperature and subsequently post-fixed in PFA/glutaraldehyde solution. Embryos were incubated overnight at 65 °C in hybridization buffer (50 % formamide, 5x SSC (pH 6), 0.1 % SDS, 50 μg/ml heparin, 500 μg/ml yeast RNA) containing the labelled probe. Embryos were washed twice with solution I (5x SSC, 50 % formamide, 0.1 % SDS) at 65 °C and 3 times in solution III (2x SSC, 50 % formamide, 0.1 % SDS) at 65 °C, rinsed in TBS/0.1 % Tween-20 and incubated overnight in blocking buffer (TBS with 2 % goat serum, 0.1 % blocking reagent (Roche), 0.1 % Tween-20) containing an AP-labelled anti-DIG antibody (1/2000) (Roche). NBT/BCIP was used as a substrate for the Alkaline Phosphatase.
Embryos were fixed overnight at 4 °C in 4 % PFA (or Ethanol/Acetic acid for ephrinB2 staining). Thick vibratome sections (70 μm) at brachial level were collected in PBS, washed in PBS/ Triton 0.5 % and blocked in PBS containing 1 % BSA/0.1 % Triton. Sections were incubated with primary antibodies against Foxp1 (gift from Dr. Novitch: 1/2000), Islet1 (gift from Dr. Jessell: 1/1000), Islet1/2 (Hybridoma Bank: 1/100), Lim1 (gift from Dr. Jessell: 1/1000), HRP (Jackon Immuno Research:1/2000) or ephrinB2 (R&D systems: 1/50) overnight at 4 °C. The ephrinB2 antibody was validated previously by us and others [35, 36, 37, 38]. Secondary antibodies were applied for 1 h at room temperature. Confocal microscopy was carried out on a Leica SP5 confocal.
HRP retrograde labeling of motor neurons
Retrograde labeling of mouse motor neurons using HRP (Roche) as tracer was performed as described . HRP was injected into either dorsal or ventral hindlimb shank musculature of E12.5 mouse embryos. LMC MN nuclei were visualized using H2BGFP expressed from the EfnB2 promoter (not shown in the figure). MN were considered HRP positive when the nucleus was surrounded by HRP labelling.
Heatmap representation was obtained using the MatLab software. For each gray value in an 8-bit image, pixels were counted in the defined interval and then transformed in a heatmap with calibration bar representing the relative abundance of GFP (0 to 100 %). GFPhigh cells are cells with a 100 % relative abundance of GFP (red cells on heatmaps). Calculation of the distance to the nearest neighbor was done using the software R (see Additional file 2 for details). The measurement of surface overlap was done as described in (Laussu et al. Antagonistic Eph:ephrin signaling patterns the ventral neural tube, Submitted).
We thank Dr Novitch for sharing the Olig2-Cre mice and Ralf Adams for the EfnB2 flox mice. The Lim1 antibody was a kind gift from Dr Jessell. We are indebted to Artur Kania and Tzu-Jen Kao for their expertise on back fill experiments and for insightful discussions. We are greatful to our laboratory colleagues for critical reading of the manuscript. Confocal images were acquired at the Rio Imaging Plateform (Toulouse) with the help of Brice Ronsin. Mice were housed at the Anexplo ABC facility. This work was supported by a grant from Association Française contre les Myopathies to AD. ML was the recipient of a studentship from CNRS/Région Midi-Pyrénées. JL was a recipient of a MESR studentship. Both ML and JL obtained funding from Fondation pour la Recherche Médicale.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.