Ganglioside GD3 regulates neural stem cell quiescence and controls postnatal neurogenesis

Abstract

The postnatal neural stem cell (NSC) pool hosts quiescent and activated radial glia-like NSCs contributing to neurogenesis throughout adulthood. However, the underlying regulatory mechanism during the transition from quiescent NSCs to activated NSCs in the postnatal NSC niche is not fully understood. Lipid metabolism and lipid composition play important roles in regulating NSC fate determination. Biological lipid membranes define the individual cellular shape and help maintain cellular organization and are highly heterogenous in structure and there exist diverse microdomains (also known as lipid rafts), which are enriched with sugar molecules, such as glycosphingolipids. An often overlooked but key aspect is that the functional activities of proteins and genes are highly dependent upon their molecular environments. We previously reported that ganglioside GD3 is the predominant species in NSCs and that the reduced postnatal NSC pools are observed in global GD3-synthase knockout (GD3S-KO) mouse brains. The specific roles of GD3 in determining the stage and cell-lineage determination of NSCs remain unclear, since global GD3S-KO mice cannot distinguish if GD3 regulates postnatal neurogenesis or developmental impacts. Here we show that inducible GD3 deletion in postnatal radial glia-like NSCs promotes the NSC activation, resulting in the loss of the long-term maintenance of the adult NSC pools. The reduced neurogenesis in the subventricular zone (SVZ) and the dentate gyrus (DG) of GD3S-conditional-knockout mice led to impaired olfactory and memory functions. Thus, our results provide convincing evidence that postnatal GD3 maintains the quiescent state of radial glia-like NSCs in the adult NSC niche.

Figure 1.. Generation of radial glia-like neural stem cell specific GD3S-conditional knockout mice.

(a) Metabolic pathways and structure of ganglioside GD3. Glycosyltransferases (underlined) catalyze the biosynthesis of gangliosides. GD3S is a critical enzyme for the synthesis of GD3 from GM3. GlcCerS: Glucosylceramide synthase, Ugcg; LacCerS: Lactosylceramide synthase, B4galt5; GM3S: GM3 synthase, St3gal5, GD3S: GD3 synthase, St8Sia1; NeuAc: N-Acetylneuraminic Acid. (b) Schematic of generation of radial glia-like neural stem cell (RGL) specific GD3S-conditional knockout, Gli1-CreERT2;GD3Sf/f (GD3S-cKO) mice. First panel illustrates that the L1L2_Bact_P cassette was inserted at position 142914589 of chromosome 6 upstream of the Exon2 of GD3S in the ES cell line, The cassette is composed of an FRT site followed by lacZ sequence and a loxP site. This first loxP site is followed by a neomycin resistance gene under the control of the human beta-actin promoter, SV40 polyA, a second FRT site and a second loxP site. A third loxP site was inserted downstream of the Exon2 at position 142913744. The exon2 of GD3S gene is thus flanked by loxP sites. In the second step, the floxed allele was created by crossing the F1 mice with flp recombinase expressing mice (Actin-Flp). Subsequent cre expression resulted in the generation of knockout mice. At the third step, we generated GD3S-cKO mice. Gli1-CreERT2;tdTomato;GD3Sf/f (GD3S-cKO) mice were generated by crossing floxed allele of GD3S (GD3Sf/f) with Gli1-CreERT2 and TdTomato (Ai9) mice. (c) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 5 more days later. (d) Immunohistochemistry for tdTomato (Tmt, red) and GD3S (green) combined with DAPI (blue) on coronal 20 μm cryosections. GD3S signal was depleted in the SVZ of cKO mouse brain. LV: lateral ventricle. Scale bar, 20 μm.

Figure 2.. GD3-deficient RGLs markedly decreased quiescence and promoted to proliferate and differentiate to the neural precursor cells in the SVZ of GD3S-cKO mice at 5 days after tamoxifen injection.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 5 more days later. (b) Immunohistochemistry for Tmt (red), Sox2 (magenta), and BrdU (green) combined with DAPI (blue) on coronal 20 μm cryosections. Scale bar, 20 μm. LV: lateral ventricle. (c) Quantification of BrdU-incorporated Tmt+Sox2+ cells after recombination and its statistical analysis. *; p<0.05. (d) Immunohistochemistry for Tmt (red), Sox2 (magenta), and DCX (green) combined with DAPI (blue) on coronal 20 μm cryosections. Scale bar, 10 μm. (e–g) Quantification of (e) Tmt+Sox2+DCX-, (f) Tmt+Sox2-DCX+, and (g) Tmt+Sox2+DCX+ cells and their statistical analysis. *; p<0.05. (h, i) Quantitative analysis for intensities of Sox2 (h) and DCX (i) within Tmt+ cells. Mean ± SEM. *; p<0.05. (j) Intensity ratio of DCX to Sox2. Mean ± SEM. *; p<0.05. n= 5 mice/group.

Figure 3.. Nestin+GFAP+ radial glia-like quiescent neural stem cells decreased in GD3-deficient RGL at the SVZ of GD3S-cKO mice at 1 month after tamoxifen injection.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 1 more month later. (b) Immunohistochemistry for Tmt (red), Nestin (green), and GFAP (magenta) combined with DAPI (blue) on coronal 20 μm cryosections. Scale bar, 20 μm. LV: lateral ventricle. Quantification of (c) Nestin+GFAP+ and (d) Nestin+GFAP- in RGL-derived cells (Tmt+) from imaging data of (b). *; p<0.05. n= 5 mice/group.

Figure 4.. GD3 deletion in radial glia-like neural stem cells (RGLs) reduced the NSC pool in the SVZ of the GD3S-cKO mice at 3 month after tamoxifen injection.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis after 3 more months. (b) Immunohistochemistry for Tmt (red), Sox2 (magenta), and PCNA (green) combined with DAPI staining (blue) on coronal 20 μm cryosections. Scale bar, 20 μm. LV: lateral ventricle. (c–f) Quantification of (c) Tmt+ RGL-derived cells, (d) Tmt+PCNA+ dividing cells among Tmt+ cells, (e) Tmt+Sox2+PCNA+ activated NSCs among Tmt+ RGL-derived cells, and (f) Tmt+Sox2+PCNA+ activated NSCs in Tmt+Sox2+ RGLderived NSCs. *; p<0.05. (g) Quantification of the intensity of Sox2 within Tmt+ cells. Mean ± SEM. *; p<0.05. n= 5 mice/group.

Figure 5.. Reduced neurogenesis from GD3-deficient radial glia-like neural stem cells (RGLs) was found in the RMS-OB of GD3S-cKO mice.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 3 more months later. (b) Immunohistochemistry for Tmt (red), DCX (green), GFAP (magenta) combined with DAPI staining (blue) on sagittal 20 μm cryosections (RMS: rostral migratory stream). Scale bar, 100 μm. (c) Immunohistochemistry for Tmt (red) and DCX (green) combined with DAPI (blue) in the granule cell layer (GCL) of the OB. Scale bar, 200 μm. (d) Immunohistochemistry for Tmt (red) and NeuN (magenta) combined with DAPI (blue) in the periglomerular layer (PGL) of the OB. Scale bar, 50 μm. (e) Quantification of Tmt+ cells among NeuN+ cells from imaging data of (d). Tmt+NeuN+ cells were decreased in GD3S-cKO mice.. *; p < 0.05. n= 5 mice/group.

Figure 6.. Induced ablation of postnatal GD3 in radial glia-like neural stem cells (RGLs) impairs olfaction in GD3-cKO mice.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 3 more months later. (b) Schematic diagram showing the buried pellet test. The buried pellet test, which relies on the animal’s natural tendency to use olfactory cues for foraging, is used to confirm the ability to smell. The time until mice uncovered the food pellet in bedding was measured as an average of 4 trials. (c) Quantification of the latency (seconds) to uncover the buried pellet. The box range represents the upper and lower quartiles, and the end of the whiskers represents the minimum and maximum values. The median values are represented by bars in the boxes. *; p < 0.05. n= 12 for the control and n= 7 for GD3S-cKO.

Figure 7.. Reduced neurogenesis from GD3-deficient radial glia-like neural stem cells (RGLs) was found in the DG of GD3S-cKO mice.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 3 more months later. (b) Immunohistochemistry for Tmt (red), PCNA (green), and Sox2 (magenta) combined with DAPI (blue) in the dentate gyrus (DG). Scale bar, 100 μm. (c) Quantification of Sox2+PCNA+ proliferating-active NSCs among Tmt+ RGL-derived cells. *; p < 0.05. (d) Immunohistochemistry for Tmt (red) and NeuN (magenta) combined with DAPI (blue) in the DG. Scale bar, 100 μm. (e) Quantification of Tmt+NeuN+ newborn neurons in the DG area. *; p < 0.05. n= 5 mice/group.

Figure 8.. Induced ablation of postnatal GD3 in radial glia-like neural stem cells (RGLs) impairs hippocampus-dependent memory in GD3-cKO mice.

(a) Summary of the experimental design. 1-month-old mice were treated with Tamoxifen for 5 days and followed by analysis 3 more months later. (b) Experimental procedure of the novel object recognition test. (c, d) The proportion of the exploring time for two familiar objects in the training session. (e, f) The proportion of the exploring time for the novel object to the familiar object 1h (e) or 24h (f) after the training. The box range represents the upper and lower quartiles, and the end of the whiskers represents the minimum and maximum values. The median values are represented by bars in the boxes. *; p < 0.05. n=12 for the control and n= 6 for GD3S-cKO.