Neuropilins and their ligands are important in the migration of gonadotropin-releasing hormone neurons

Abstract

Gonadotropin-releasing hormone (GnRH) neurons in the hypothalamus play an important role in reproductive function. These cells originate in the nasal compartment and migrate into the basal forebrain in association with olfactory/vomeronasal nerves in embryonic life in rodents. Here, we studied the role of neuropilins and their ligands, semaphorins, in the development of the olfactory-GnRH system. We focused on Neuropilin-2 knock-out (Npn-2(-/-)) mice, because they are known to display defasciculation of olfactory nerves and reduced fertility. We found a significant decrease in the number of GnRH neurons in the hypothalamus and a marked reduction in their gonadal size. We then observed an abnormal increase of GnRH neurons in the noses of Npn-2(-/-) mice, indicating that these cells failed to migrate into the forebrain. However, because neuropilins and semaphorins are involved in events of neuronal migration in the brain, we asked whether the observed reduction in GnRH neurons was directly attributable to the action of these molecules. Using fluorescence-activated cell sorting and reverse transcription-PCR on mRNA derived from embryonic green fluorescent protein (GFP)-GnRH transgenic mice, we found expression of class 3 semaphorins and their receptors (neuropilin-1/2 and plexin-A1) in GnRH neurons. Furthermore, double-immunofluorescence experiments showed that migrating GnRH neurons, as well as associated olfactory fibers, express Npn-2 in the nasal region. We then used a line of immortalized GnRH neurons (GN11 cells) that display the same expression patterns for semaphorins and their receptors as GFP-GnRH cells and found that class 3 semaphorins and vascular endothelial growth factors modulate their migratory activity. These studies provide support for the direct involvement of neuropilins and their ligands in the establishment of the GnRH neuroendocrine system.

Figure 1.

Analysis of GnRH neurons and gonads in Npn-2^−/− mice. a, b, Coronal sections through the noses of Npn-2^+/+ (a) and Npn-2^−/− (b) mice at P0 were stained with an anti-GnRH antibody. Those taken from Npn-2^−/− animals showed an excess of GnRH-positive cells, often appearing in clusters (arrowheads). c, d, Sagittal sections through the basal forebrain of Npn-2^−/− (d) mice at P0 showed a significantly reduced number of GnRH neurons (arrows) and fibers projecting toward the median eminence of the hypothalamus (arrowheads) compared with Npn-2^+/+ littermates (c). e, Gross morphological phenotypes of testes derived from adult Npn-2^+/+ and Npn-2^−/− mice, showing a marked reduction in size of the latter. f, g, Hematoxylin and eosin staining of seminiferous tubules from WT and Npn-2^−/− mice. Seminiferous tubules in WT (f) have well developed lumens (arrows), whereas lumen formation is reduced in Npn-2^−/− mice (g, arrowheads), with fewer spermatozoa and an accumulation of spermatogenic immature cells. CP, cribriform plate; OT, optic tract. Scale bars: a (for a, b), c (for c, d), 250 μm; e, 10 mm; (in f) f, g, 100 μm.

Figure 2.

Innervation of the olfactory bulb by peripherin-positive olfactory fibers. a, b, Sagittal sections through the nasal regions and the rostral forebrain at E15 showed defasciculation of the fibers stained for peripherin (green) (b, arrowheads) in the nasal compartment (NC) and improper innervation of the olfactory bulb (OB) (b, arrows) of Npn-2^−/− mice compared with Npn-2^+/+ (a). Scale bar, 100 μm.

Figure 3.

Expression of semaphorin and neuropilin molecules in GnRH neurons. a, Schematic drawing of the experimental paradigm used to isolate mRNA from GFP–GnRH neurons. Fluorescence-activated cell sorting was applied to cells dissected from GFP–GnRH mice at different embryonic stages; mRNA extraction and RT-PCR were performed on GFP-positive samples. b, RT-PCR analysis at E13, using GFP–GnRH cell mRNA, showed the presence of Sema3A, Sema3F, Npn-1, Npn-2, and Plxna1 transcripts (330, 262, 190, 260, and 199 bp, respectively). c, Schematic drawing of the anatomy of the olfactory system in embryonic rodents. Olfactory fibers originating from the olfactory placode (OP) are depicted as black and green lines. The latter correspond to the VNNs, specifically labeled for Npn-2. GnRH neurons are typically seen in association with these bundles of fibers. d, Immunofluorescence, applied to sagittal sections taken from E15 mouse, showing Npn-2-immunoreactive olfactory fibers (green) originating from the OP. Nuclei were counterstained with DAPI (blue). e, Multipanel composition of contiguous sections taken from the nose of an E15 mouse and stained for GnRH and Npn-2 (green) and peripherin (red). As shown in the merged panels resulting from the overlays between GnRH/Npn-2 and GnRH/peripherin staining, GnRH neurons are in close association with peripherin-positive fibers, which are specifically stained also for Npn-2. f, Double immunofluorescence for GnRH (green) and Npn-2 (red) on sagittal sections taken from E13.5 mouse showing colocalization of the two antigens in the same cells, shown in the small inset, derived from single high-magnification confocal scans. FB, Forebrain; +ve, positive; −ve, negative; M, marker. Scale bars: d, 100 μm; e, 250 μm; f, 100 μm; f, inset, 50 μm.

Figure 4.

GN11 cells show the same expression for neuropilins and semaphorins as GFP–GnRH neurons. a, RT-PCR analysis performed on mRNA extracted from GN11 cells showed the presence of specific transcripts for Sema3A, Sema3F, Npn-1, Npn-2, and Plxna1, with amplicons of 330, 262, 190, 260, and 199 bp, respectively. b–g, Multipanel composition of immunofluorescence stainings of GN11 cells for Npn-1, Npn-2, and Plxna1. GN11 cells were immunoreactive for Npn-1/2 and Plxna1 (c, e, f, green), with greater signal for Npn-2; b shows the negative control processed in the absence of the primary antibody (−Ab). In addition, supplement of CM-Sema3A and -Sema3F induced Npn-1 and Npn-2 internalization, respectively, with formation of immunopositive aggregates within the cytoplasm (d, g, arrowheads). Scale bars: (in b) b, c, e, f, 50 μm; (in d) d, g, 25 μm. M, Marker.

Figure 5.

Effects of neuropilin ligands on migration of GN11. a, Migration of GN11 cells was inhibited by Sema3A and Sema3F in three-dimensional collagen matrices. GN11 cells moved away from the aggregate in the presence of 1% FBS (positive control) and CM-COS but not in the presence of CM-Sema3A and -Sema3F. Scale bar, 250 μm. b, In a chemotactic assay, using a 48-well Boyden's chamber, CM-Sema3A and -Sema3F induced a significant reduction (p < 0.05 and 0.01, respectively) of cells that migrated compared with control CM (CM-myc and CM-FLAG respectively). OptiMEM was a negative control (first column). c, Immunoneutralization of CM-Sema3A and -Sema3F with anti-c-myc and anti-FLAG antibodies significantly restored the number of cells that migrated, whereas it did not affect migration induced by control CMs, suggesting specificity of the effects observed. d, Simultaneous exposure of GN11 cells to CM-Semas and VEGFs (+VEGFs) induced a reversal in the inhibitory effects of semaphorins, suggesting antagonistic effects of VEGFs on the migration of GN11 cells. *p < 0.05; **p < 0.01. Error bars represent SEM. Ab, Antibody; CM-3A, CM-Sema3A; CM-3F, CM-Sema3F; sqmm, mm².