Sprague-Dawley rats (Depré, St Doulchard, France) and C57BL/6J mice (Elevage Janvier, Le Genest Saint Isle) were housed and cared for, following the animal welfare guidelines established by the INSERM and the Montpellier University. All animal procedures were performed in strict adherence to the European Union directive 2010/63 and the ARRIVE guidelines (Kilkenny et al., 2010) upon CEEA-LR-12031 authorization.

Total RNA was extracted from neuro-spheres and brain using the NucleoSpin^® 8/96 RNA kit (Macherey-Nagel, Germany) according to the manufacturer’s instructions. RNA concentration and purity were evaluated with the Agilent RNA 6000 Nano Kit (Agilent Technologies, Waldbronn, Germany). Only RNAs with an RNA Integrity Number (RIN) > 8 were used. cDNA was synthesized in a 20 μl reaction mixture. One μg of total RNA was mixed with 1 μl random primer hexamer (0.5 μg/μl) (Amersham, Orsay, France) and incubated at 70°C for 10 min. After cooling on ice, the solution was mixed with 4 μl of 5X first strand buffer, 2 μl of 0.1 M DTT, 1 μl each of dATP, dTTP, dCTP, dGTP (each 10 mM) and 1 μl MuMLV reverse transcriptase (200 U; Promega, Charbonnieres, France) and incubated at 37°C for 60 min and then at 70°C for 15 min. Parallel reactions for each RNA sample were run in the absence of MuMLV to assess the degree of genomic DNA contamination. Obtained amplicons were stored at −20°C until use.

Deleted versions of Reg-1α were obtained with a PCR-based mutagenesis protocol. The coding sequence of human Reg-1α (GenBank Accession Number 002909.5; OriGene Technologies) was subcloned in the pCAGGS/ES vector (Addgene). The CDS of human Reg-1α in pCAGGS/ES was used as template to generate the deleted versions Reg-1α^Δ23–33 lacking 11 amino acids between the signal peptide and the proteolytic site (33 AA), and Reg-1α^Δ34–166, lacking the C-terminal domain of the protein.

Reg-1α^{Δ 23–33} was generated combining PCR reactions. The upstream region of human Reg-1α encompassing the region for amino acid 1-22 was amplified by the primer 5′-GGGGTACCATGGCTCAGGCCAGCTCA-3′ including translation initiation start (ATG) and KpnI site and the primer 5′-GCCTTCTGGGCAGCTGATGCCTTGGCTCTGAGACAG-3′. The downstream region of the protein (from amino acid 34) was amplified using the 5′-CTGTCTCAGAGCCA AGGCATAGCTGCCCAGAAGGC-3′ and the 5′-GGGAGC TCTAGTTTTTGAACTTGCAGAC-3′ primers including translation initiation stop and SacI site. The two PCR products were mixed and amplified in a third PCR reaction using primers 5′-GGGGTACCATGGCTCAGGCCAGCTCA-3′ and 5′-GGGAGCTCTAGTTTTTGAACTTGCAGAC-3′. The final PCR product encoding CDS human Reg-1α^{Δ 23–33} was cloned in the pCAGGS/ES vector. Human Reg-1α^{Δ 34–166} was constructed by PCR amplification using the forward 5′-GGGGTACCATGGCTCAGGCCAGCTCA-3′ and reverse 5′-GGGAGCTCTA GTTTTTGAACTTGCAGAC-3′ primers, including translation initiation stop and SacI site. Full length mouse Reg-1α cDNA (BC-028761, openBIOSYSTEMS) was sub-cloned into the pCAGGS/ES vector. All plasmids were sequenced and contained the signal peptide that allows protein secretion. The pCAG-GFP plasmid, allowing expression of green fluorescent protein (GFP), was used as a positive transfection control.

Real-time PCR was performed using a LightCycler^® 480 Instrument (Roche Diagnostics, Mannheim, Germany) according to the manufacturer’s instructions. PCR reactions were performed in a final volume of 6 μl with 1 × LightCycler 480 SYBR Green I Master mix, 1 μM of each primer and 1/12.5 diluted RT mixture as PCR template (or water as negative control). The amplification conditions were: (i) cDNA denaturation (1 cycle: 95°C for 10 min); (ii) amplification (45 cycles: denaturation at 95°C for 10 s, annealing at 60°C for 10 s, elongation at 72°C for 10 s); (iii) melting curve analysis (1 cycle from 65 to 95°C for 60 s); (iv) cooling (1 cycle: 40°C for 60 s). Real-time detection of the SYBR Green I fluorescence intensity, indicating the amount of PCR product formed, was measured at the end of each elongation phase. To confirm the amplification specificity, PCR products from each primer pair were subjected to a melting curve analysis and then separated by agarose gel electrophoresis. The amounts of target genes were normalized relative to cyclophilin used as reference gene in the corresponding samples. Primers previously described (Jing et al., 2010) were used with some modifications. The sequences of the primers specific for the examined genes and the predicted product sizes are shown in Table 1.

Frontal cortex was dissected from male rats at different ages and disrupted with a Dounce homogenizer in cold lysis buffer (20 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Triton, 1 mM EGTA) with Complete Mini protease inhibitor mixture (Roche Applied Science). Protein concentration was determined with the BCA assay (Pierce), and 20 μg of proteins were loaded in 12.5% SDS-PAGE. For protein/peptide detection, nitrocellulose membranes (0.22 μm, Bio-Rad) were incubated with primary antibodies at 4°C overnight and with secondary antibodies at RT for 2 h. Immunolabeling was revealed by chemiluminescence reaction using the ECL Western blot detection reagents (Luminata crescendo western HRP substrate, Millipore) and measured using the Licor detection system. Primary antibodies were: rabbit anti-Reg-1α [1:500 (Duplan et al., 2001)], goat anti-EXTL3 (1:1000, #AF2635, R&D Systems), rabbit anti-p-GSK-3β Ser⁹ (1:1000,#9336S, Cell Signaling); mouse anti-GSK-3β (1:5000, # 610201, BD Transduction laboratory), mouse anti-p-GSK-3β Tyr²¹⁶ (1:1000, # 612313, BD Transduction laboratory) anti-β-catenin (1:5000,#610153, BD Transduction laboratory) and mouse anti-β-actin (1:5000,#A5441, Sigma-Aldrich). Generation and specificity of the anti-Reg-1α rabbit polyclonal antibody has been previously reported (Duplan et al., 2001; Laurine et al., 2003; Acquatella-Tran Van Ba et al., 2012; Moussaed et al., 2018).

Neural Precursor Cells were obtained from telencephalon of E13.5 Sprague-Dawley rat embryos. Cells were dissociated enzymatically by incubation in a solution of trypsin and 0.05% EDTA, and then mechanically by repeated pipetting. After filtration through a 70 μm cell strainer, cells were resuspended in proliferation medium constituted by DMEM/F12 medium supplemented with 1% N2 (Invitrogen), 1% antibiotics (penicillin/streptomycin), 2 mM glutamine, 20 ng/mL bFGF and 10 ng/mL EGF (Invitrogen). Cells were grown in non-adherent culture conditions in flasks coated with polyhydroxyethyl metacrylate (Sigma) at 37°C and 84% hygrometry. Small floating cell clones were observed from day 2 after plating and formed neurospheres by clonal expansion (Armando et al., 2007). Neurospheres were passaged every week. At passage 3 (P3), neurospheres were equally divided among the different experiments, centrifuged at 300 g for 5 min and each pellet was resuspended in the appropriate medium (with or without growth factors) and plated on an adherent support. Neurospheres were seeded on 10 μg/mL poly-D-lysine-coated 12 mm glass coverslips in 24-wells dishes for immunocytochemical studies, or directly on poly-D-lysine-coated 35 mm dishes or 6-wells plates for RNA extraction, western blot and FACS analysis. Cells were cultured at 37°C in humidified air with 5% CO₂ for 48 h.

NSCs were washed twice with PBS and re-suspended in lysis buffer (10 mM HEPES, pH 7.5, 10 mM NaCl, 1 mM EDTA, 1 mM DTT) with complete mini protease inhibitor cocktail (Roche Biomedicals). Cells were allowed to swell on ice for 15 min and then they were homogenized with a Dounce homogenizer. Nuclei integrity was verified under a light microscope and homogenates were layered on 40% sucrose. The nuclei were collected by centrifugation at 2000 rpm for 15 min and the nuclear pellets were re-suspended in 100 μl of the same buffer. The supernatants were then centrifuged at 13,000 rpm at 4°C for 30 min to separate the cytosol (supernatant) and membrane (pellet) fractions. Protein concentration was determined with the BCA assay (Pierce). Cytosolic and nuclear fractions (20 μg) were loaded on 12.5% SDS-polyacrylamide gels (SDS-PAGE).

Pregnant mice were ethically anesthetized with isoflurane combined with oxygen (TEM SEGA). The mice were coinjected with pCAG-GFP and pCAG expressing different constructs of Reg-1α (1.5 μg/μl) in a 1:3 ratio, were mixed with 0.5 μl Fast Green (0.03%, Sigma) and injected (2 μl) into the lateral ventricle of E14.5 embryonic brains using a pulled glass micropipette. Electrode pair (5 mm in diameter; CUY650-P5, NEPA Gene) attached to an ECM 830 electroporator (BTX Harvard Apparatus) transmitted 5 squares 40 V electric pulses for 50 ms at 950 ms intervals through the uterine wall. Each condition was applied to at least three independent litters, targeting the same dorso-lateral region of the neocortex. For 5-bromo-2-deoxyuridine labeling (BrdU, Sigma), electroporated pregnant dams were intraperitoneally (i.p) injected with BrdU (100 μg/g) 1 day after electroporation and sacrificed 24 h later. The sections were stained for rabbit anti-T-box brain 1 (Tbr1) to delineate the cortical plate. The percentage of targeted cells was determined by dividing the number of GFP positive cells in each layer (cortical plate and intermediate zone/ventricular zone) by the total number of GFP-positive cells in the entire section.

Neural progenitor cultures were fixed in formaldehyde/PBS for 30 min followed by three washes in PBS. Cells were blocked and permeabilized in PBS supplemented with 2% BSA and 0.1% Triton and then incubated at 4°C with primary antibodies. Animals were anesthetized with ketamine (Imalgen 1000) and xylazine (Rompun 2%) (75 and 25 mg/kg IP, respectively) and perfused with 40 ml of phosphate buffered saline (PBS) and 40 ml of 4% paraformaldehyde in 0.1 M PBS. Adult brains were post-fixed for 48 h in 4% paraformaldehyde and rinsed in PBS. Embryonic brains were post-fixed for 4 h at 4°C. Samples were cryoprotected in 30% sucrose solution and embedded in O.C.T. compound (Sakura). Embedded tissues were sectioned on a cryostat (14 μm thickness) in the coronal plane. For analyses of in utero electroporation, brains were dissected after intracardiac perfusion of the embryos with antigenfix solution (Microm Microtech), 2 or 3 days following electroporation. Brains were post-fixed for 4 h at 4°C, rinsed in 0.1M phosphate buffer (pH 7.4, Sigma) and immersed in a 10% sucrose solution diluted in phosphate buffer overnight at 4°C for cryoprotection. Brains were sectioned at 20 μm in the coronal plane. The following primary antibodies were used: mouse anti-Nestin (1:200, #4760, Cell Signaling), mouse anti-βIII Tubulin (1:400, #T8660, Sigma), mouse-NeuN (1:200, clone A60, #MAB377, Millipore), rabbit anti-CalB (1:1000, #CN-38a, Swant), rat anti-Brdu (1:250, #ab6326, Abcam) rabbit anti-T-box brain 1 (Tbr1, 1:300, #AB10554, Millipore), rabbit anti-Ki67 (1:100, #ab16667, Abcam) and rabbit anti-Reg-1α (1:500, Duplan et al., 2001). Electroporated cells were identified by GFP expression. For BrdU labeling, sections were incubated in 2N hydrochloric acid for 20 min at RT followed by neutralization with 100 mM sodium borate (Sigma), pH 8.9, for 15 min before the permeabilization step. Secondary antibodies directed against rabbit or mouse and rat IgGs conjugated to Alexa Fluor 488 (1:1000, Invitrogen) and Cy3 or Cy5 (1:400, Jackson Immuno-Research Laboratories) were incubated 2 hours at room temperature and then the nuclei were counterstained with 4′, 6′-diamino-2-phenylindole (DAPI, Molecular Probes). Sections were imaged with confocal microscopy with a Leica TCS4D confocal laser-scanning microscope (Leica Microsystems, Heidelberg, Germany). Images were processed using Adobe Photoshop (Adobe Systems Inc., Mountain View, CA, United States).

To quantify neural precursor cells migration and differentiation, we measured the shape of 10 neurospheres, the distance covered by migrating cells and βIII-Tubulin expression level in cells outside neurospheres. The perimeter of each neurosphere after DAPI staining was calculated with ImageJ software. The number of migrating cells was defined as the total number of cells outside each neurosphere, determined using the cell counter plug-in. The distance between the border of each neurosphere and the 10 furthest cells was evaluated and taken as an estimate of the migratory capacity of the cells. Additional quantifications using the NeuronJ plug-in included: (i) the intensity of βIII-Tubulin staining around each neurosphere and (ii) the longest neurite calculated by sampling 10 cells, the soma of which was located at the edge of the neurosphere. Images were obtained by imaging neurospheres using a Zeiss Axiovert microscope and a MRm2 camera system (Supplementary Figure 1).

Neural precursor cells were washed with PBS and then incubated with a trypsin/0.05% EDTA solution at 37°C for 5 min. Dissociated cells were centrifuged at 300 g for 5 min (and also after each of the following steps). Pellets were washed once with PBS and fixed in formaldehyde/PBS solution for 30 min followed by another wash in PBS. Cells were blocked and permeabilized in PBS supplemented with 2% BSA and 0.1% Triton and then incubated at 4°C with mouse anti-βIII-Tubulin or anti-Nestin antibodies in the same buffer overnight. The next day, cells were incubated with a secondary antibody directed against mouse IgGs coupled to Alexa Fluor 488 (Invitrogen). Cells were analyzed using a BD FACSCanto^TM cell analyzer (BD Biosciences) with the running FACS Diva v6.1.3 software for acquisition. Data were processed with the FlowJo software.

Hippocampal neurons were isolated from 17.5-days-old embryos of timed pregnant Sprague-Dawley rat (CE Depré, Saint Doulchard, France) as previously described (Acquatella-Tran Van Ba et al., 2012). In brief, the hippocampi from embryonic brains were dissected in PBS supplemented with 6% glucose, 10 units/ml penicillin/streptomycin and digested with 0.05% trypsin-EDTA. The cell suspensions were plated on 18-mm glass coverslips (25.10⁴ cells/well for transfection), or 12-mm glass coverslips (5.10⁴ cells/well for treatment). Cells were re-suspended in Neurobasal medium with 0.5% B27, 50 μm L-Glutamine, 0.25% Glutamax, 1% penicillin/streptomycin and 10% FBS. 5 × 10⁵ freshly dissociated hippocampal cells were transfected using 4D-Nucleofector^R System Nucleofector^TM (Amaxa Biosystems, Cologne, Germany), using the program N/Rat Hipp-cortical HV, pulse code CU 133. The transfected cells were then seeded on glass coverslips in 24-well plates coated with 10 μg/ml poly-D-lysine to study neurite elongation. Cells were then cultured at 37°C for 48 h in humidified air with 5% of CO_2. For the analysis of neuronal morphology, axonal like processes (Tau-positive/MAP2-negative) from isolated transfected and non-transfected neurons were visualized and photographed using a Axioimager Leica fluorescent microscope equipped with an Axiovision LE 4.4 software. The length of the axons was measured using the NeuronJ software (ImageJ), tracing β3 Tubulin-positive fibers.

All bar graphs are plotted as mean ± SEM. The values obtained in electroporation conditions were compared using GraphPad InStat (San Diego, CA, United States). Significance of the percentage of GFP-positive cells in the CP and in IZ/VZ for each condition was determined using Mann Whitney test. One-way ANOVA of cell cycle exit and neurite outgrowth revealed significant differences among experimental groups, multiple comparisons among experimental groups were performed using Dunnett’s or Bonferroni’s post hoc test.

We have previously demonstrated that secreted Reg-1α stimulates axon outgrowth from cultured neurons via Extl3 (Acquatella-Tran Van Ba et al., 2012), suggesting a possible physiological role during brain development. In developing brain, Reg-1α is known to be expressed (de la Monte et al., 1990), but its exact distribution is still poorly explored. To address whether Reg-1α is involved in cortical neurogenesis, we thus first investigated its expression pattern in the developing rat cortex. Western blot analysis of brain extracts at different developmental stages (from E13.5 to adulthood) revealed the presence of a specific 18 kDa band (Figure 1A), corresponding to the reported Reg-1α MW found in the adult small intestine (ileum) (Namikawa et al., 2005) and here used as a control. Quantification of identified band revealed a progressive developmental decrease, so that 1 month old brains had threefold less protein that E13.5 brains, with further decrease with aging (Figure 1A). However, the quantitative RT-PCR of Reg-1α transcript expression normalized to the levels of cyclophilin mRNA did not show any significant difference between the different stages and did not corroborate with expression kinetics (Figure 1B compare to Figure 1A). Quantitative RT-PCR of Extl3 expression, normalized to the levels of cyclophilin mRNA, showed a high expression at early stages (between E17.5 and P30) of cortical development with decreased expression in the adult tissue (Figure 1B).

Double immunostaining for Reg-1α and either the pan-neuronal NeuN or the immature GABAergic interneuron Calbindin (CalB) markers identified positive neuronal subtypes in the cortex at E17.5 (Figure 1C). Reg-1α colocalized with NeuN in the cortical plate and piriform cortex (respectively box1 and 2 in Figure 1Ca; higher magnifications in Figure 1Cb), whereas Reg-1α and Calbindin were co-expressed in a restricted interneuron subpopulation concentrated in superficial layer (Figure 1C, boxes 3–4; higher magnifications in Figure 1Cc). Reg-1α was still detectable in NeuN positive cells in all layers of the adult parietal cortex (Figure 1D, boxes 5–7; higher magnifications in Figure 1Db).

The abundant expression of Reg-1α at early stages (E13.5) of telencephalic development suggested a role in migration and differentiation of neural progenitors. To assess this possibility, we used neurosphere (NS) cultures from rat embryonic telencephalon and tested their adhesion and proliferation properties. We plated free-floating NS after 3 passages on plates with or without poly-D-lysine coating and cultured them in proliferation medium for 2 and 48 h. Measure of the NS size showed no significative differences between non-adherent and adherent neurospheres (Figure 2A) and in both conditions the NS size increased about 1.5 times between 2 and 48 h. We thus selected to use adherent neurospheres grown in proliferation medium for 48 h to be able to measure their differentiation and migration processes (Supplementary Figure 1). NS contained cells at various stages of differentiation including from progenitors to post-mitotic neurons as illustrated by Nestin and β3-Tubulin immunostaining. Nestin-positive cells were localized inside and at the periphery of the NS whereas β3-Tubulin positive neurons were found mostly on the outside (Figure 2B), indicating a gradient of differentiation toward the periphery.

Notably, quantitative RT-PCR and western blot analysis of the NS detected mRNA and protein expression for both Reg-1α and Extl3 (Figure 2C). Reg-1α was also expressed in culture medium of NS (Figure 2C, right panel) indicating that the protein is secreted by neural cells, as previously shown for hippocampal neurons (Acquatella-Tran Van Ba et al., 2012). As NS contain cells at various stages of differentiation, including Nestin-positive proliferating progenitors and β3-Tubulin positive post-mitotic neurons, we next asked if Reg-1α is present in both populations. Confocal microscopy analysis showed that this was the case (Figure 2D).

We next asked whether Reg-1α could regulate neural precursor migration and differentiation. To this end, we analyzed adherent NS grown in proliferation medium in the presence of 10^–7 M recombinant Reg-1α for 48 h (Figure 3), which are conditions previously shown to increase neuritic outgrowth (Acquatella-Tran Van Ba et al., 2012). β3-Tubulin and DAPI staining showed a significant decrease in NS size and an increase in the number of differentiated cells present outside of the NS upon Reg-1α (Figures 3A,B, panels a–c, black bars vs. white bars). Reg-1α also increased the intensity of β3-Tubulin staining and neurite length compared to control, suggesting that Reg-1α promotes neural precursor cell differentiation (Figures 3Bd,e). We next switch the NS culture medium to a previously described differentiation medium (Armando et al., 2007) and show that the results obtained with recombinant Reg-1α treatment are comparable with those obtained with the differentiation medium (Supplementary Figure 2).

To assess if Reg-1α had a direct effect on neurite outgrowth, we tested the effects of Reg-1α on dissociated adherent cells derived from telencephalic NS (Figure 3C). β3-Tubulin staining showed that Reg-1α increases neuritic elongation: the total neuritic length and the length of the longest neurite were increased in treated vs. untreated cells (Figures 3Da,b) but no difference in the number of processes per cell was observed (Figure 3Dc). These results are in agreement with those previously reported for embryonic hippocampal neurons (Acquatella-Tran Van Ba et al., 2012).

To confirm the specificity of Reg-1α effects on neural precursor migration and differentiation, we blocked the effect of Reg-1α with the addition to the culture medium of a previously described anti-Reg-1α polyclonal antibody (Acquatella-Tran Van Ba et al., 2012). The antibody partially inhibited Reg-1α effect on precursor cells migration (Figures 4Aa–c, gray bars vs. black bars) and differentiation (Figures 4Bd,e, gray bars vs. black bars) determined as above.

All in all these data demonstrate that Reg-1α specifically promotes migration and differentiation of neural precursor cells from the telencephalon.

Reg-1α regulates GSK-3β (Moussaed et al., 2018) and GSK-3β/β-catenin signaling is known to play a key role in stem cell migration and differentiation. We thus asked whether the observed Reg-1α effects were mediated via this pathway. Notably, incubation of adherent NS with Reg-1α significantly increased p^Ser9GSK-3β and the cytosolic level of β-catenin (Figures 5A,C, black bars vs. white bars), both of which are indicators of pathway activation (Cohen and Frame, 2001). The levels of p^Y216GSK-3β were instead decreased (Figure 5B, black bars vs. white bars), as expected given that this modification leads to an opposite effect on GSK-3β activity. These data suggest that Reg-1α acts, at least in part, on neural precursor cells migration and differentiation in a GSK-3β dependent manner.

All the results obtained in vitro on neural precursor cells led us to test in vivo the effect of Reg-1α on cortical development in mouse.

We next asked if the effects of Reg-1α in NS were also occurring in vivo. To this end we overexpressed the human or mouse Reg-1α gene together with GFP (as tracer) in the cortex of E14.5 mouse embryos using in utero electroporation (Figure 6A). Three days after electroporation, brains were harvested, sectioned in the coronal plane and immunostained for Tbr1, an early post-mitotic neuronal marker, which is strongly expressed in the subplate at this developmental stage. In the developing telencephalon, progenitor cells that exit the cell cycle in the VZ migrate toward the cortical plate, providing an excellent model to study the effect Reg-1α on both proliferation and migration.

To determine the effect of Reg-1α, we divided the cortical neuroepithelium in two regions: the strongly Tbr1-positive area up to the cortical plate (CP), and the Tbr1-negative ventricular zone/intermediate zone (VZ/IZ; Figure 6B). In each one of them we compared the distribution of GFP-positive cells in control and Reg-1α electroporated embryos. In E17.5 control embryos, less than 50% of the GFP-positive cells had entered the CP (43.3 ± 3.7%; Figure 6B, gray bars) whereas the remaining GFP-cells were still found in the intermediate zone (56.6 ± 3.7%, white bars). In contrast, an increased fraction of progenitors electroporated with human Reg-1α were found in the CP (62 ± 3.1%, p < 0.0026; Figure 6B, gray bars) with a concomitant decrease in GFP-positive cells in the VZ/IZ (38 ± 3.1%, white bars). A similar effect was observed when cells were co-electroporated with mouse Reg-1α(mReg-1α) and GFP. A significantly higher percentage of mReg-1α–positive cells (70 ± 6.7%, p < 0.0137) were distributed throughout the CP with a decrease of cells in the VZ/IZ (29.9 ± 6.7%; Figure 6B) compared to control. Using GFP and Reg-1α labeling, we observed that Reg-1α localized mainly to the cell body and nerve processes of the electroporated cells (Figure 6C), which acquired a polarized and radially oriented morphology, typical of migrating cells in the developing cortical plate.

Altogether these data indicate that the overexpression of Reg-1α has an evolutionary conserved role in accelerating neuronal migration in the cortical plate.

The Reg-1α-mediated increase of cell migration into the CP could be the result of the progenitors’ premature exit from the cell cycle associated with their differentiation. To test this possibility, we coupled electroporations of E14.5 embryos with subsequent BrdU injections at E15.5. Embryos harvested at E16.5 were immunostained for BrdU (Figure 7A). Reg-1α transfected cells were rarely BrdU-positive with a significantly lower percentage of co-labeled cells in the VZ (14.14 ± 1.7%, Figure 7B, bottom panel, arrows) when compared with controls (23.2 ± 1.8%, Figure 7B, top panel, arrows). To test if this reduction was associated with premature differentiation into neurons, we measured the cell cycle exit index by immunostaining for BrdU and Ki67, a cell proliferation marker (Figure 7C). GFP-positive cells labeled with both BrdU and Ki67 should reflect actively proliferating cells, whereas cells positive only for GFP and BrdU were considered as cells that already left the cell cycle. An increased proportion of cells electroporated with Reg-1α had left the cell cycle (Reg-1α : 79.7 ± 3.3% compared with GFP controls: 57.3 ± 5.7%), explaining the reduction of proliferating progenitors (Figure 7C). In agreement with an increased differentiation we also observed that at P0, Reg-1α overexpression increased the length of labeled callosal axons projecting from layer II/III cortical neurons (1058 ± 137 μm versus Gfp 707.7 ± 25 μm, p < 0.002, Mann–Whitney test; data not shown), further confirming the effect of Reg-1α observed in vitro.

Altogether our data demonstrate that overexpression of Reg-1α in vivo promotes cell cycle progression exit and promotes premature neuronal differentiation and migration to the upper layers further enhancing axonal elongation.

Reg-α is cleaved into two domains (Nter and Cter). We thus asked whether the observed developmental activities of Reg-α required the participation of both domains or one of them was sufficient to mimic the effect of the entire protein. To address this question, we generated two mutant versions of human Reg-1α in the pCAG expression plasmid: Reg-1α^{ΔN (23–33)}, which lacks the N-terminal domain and Reg-1α^{ΔC (34–166)}, lacking the C-type lectin domain (Figure 8A). We then overexpressed the resulting human mutated forms by means of in utero electroporation following the same protocol we used for the intact protein (Figure 6). Comparison of the distribution of GFP-positive cells across the thickness of the cortical neuroepithelium showed that Reg-1α^ΔN did not promote progenitor migration. Rather, GFP-positive cells were found preferentially in the VZ/IZ (Reg-1α^ΔN: 61.7 ± 6.1%) compared with GFP controls (56.6 ± 3.7%; Figure 8B) and accordingly a smaller proportion of cells ended up in the CP (Reg-1α^ΔN 38.4 ± 6%) as compared to GFP controls (43.3 ± 3.7%). In contrast, Reg-1α^ΔC was sufficient to promote the accumulation of larger fraction of GFP-positive cells in the CP (58.1 ± 4.5%, p < 0.0317, Figure 8B, gray bars) with a consequent decrease of GFP-positive cells in the VZ/IZ (Reg-1α^ΔC: 41.9 ± 4.5%, white bars). Similarly to what observed for the intact protein, Reg-1α^ΔC promoted a premature exit of progenitor cells from the cell cycle. Reg-1α^ΔC caused a slight but not significant reduction of BrdU/GFP-positive cells in the ventricular zone compared to control (Figure 9A) but measure of the index for cell cycle exit highlighted a significant increase (Reg-1α^ΔC: 80.04 ± 3.8%, comparable to that observed with Reg-1α (Reg-1α : 79.7 ± 3.3% compared with controls (57.3 ± 5.7%, Figure 9B).

Thus, Reg-1α^ΔC alone can mimic Reg-1α activity in the cortical plate, whereas the C-terminal domain has no effect, demonstrating a critical role for the N-terminal undecapeptide domain of Reg-1α in neuronal differentiation.