Partial Deletion of Cxcl12 from Hippocampal Cajal–Retzius Cells Does Not Disrupt Dentate Gyrus Development or Neurobehaviors

Partial deletion of CXCL12 in CR cells does not affect their survival in the postnatal hippocampus. A, Representative confocal images showing coimmunostaining of CXCL12 and the CR cell marker reelin (RELN) in P5 control (Cxcl12^flox/flox) and knock-out (ΔNp73Cre; Cxcl12^flox/flox) hippocampus. Arrows indicate double-positive cells. Scale bars, 20 µm. Quantification of the percentage of CXCL12-positive CR cells is shown at right. B, Representative confocal images showing coimmunostaining of CR cell markers RELN and TRP73 at P20. CA1, cornu ammonis 1; HF, hippocampal fissure; DG, dentate gyrus. Scale bars, 200 µm. C, Quantifications of RELN⁺ TRP73⁺ CR cell density in the hippocampal fissure at P5, P20, and 12 weeks. Data are shown as scatterplots with individual data points (each representing one animal); error bars indicate ±SD. Three sections were analyzed per animal. Solid and open symbols represent female and male mice, respectively; half-solid symbols represent P5 mice with undetermined sex. Statistical analyses were performed using nested t test or nested one-way ANOVA with Tukey's post hoc test. ***p < 0.001. Extended Data Figure 2-1 supports Figure 2.

Partial deletion of CXCL12 in CR cells does not disrupt early postnatal hippocampal morphogenesis or neurogenesis. A, Representative confocal images showing coimmunostaining of CTIP2 and PROX1 in the P5 hippocampus. CTIP2 labels CA1 pyramidal neurons and mature dentate gyrus (DG) granule neurons. PROX1 labels granule neuron precursors and mature granule neurons. CA1, cornu ammonis 1. The boxed region indicates the DG area shown in C. Scale bars, 200 µm. B, Quantification of CA1 width based on CTIP2 immunostaining. C, High-magnification images of CTIP2 and PROX1 coimmunostaining in the DG. Scale bar, 20 µm. D, E, Quantification of PROX1⁺ (D) and CTIP2⁺ (E) granule neuron densities in the DG. F, Representative overview images of GFAP and SOX2 coimmunostaining in the P5 hippocampus. The boxed region indicates the DG area shown in G. Scale bars, 200 µm. G, High-magnification images of GFAP and SOX2 coimmunostaining in the DG. Scale bar, 10 µm. H, Quantification of GFAP⁺ SOX2⁺ progenitor cell density in the subgranular zone. Data are presented as scatterplots, with each data point representing one animal. Error bars indicate ±SD. Three sections were analyzed per animal. Half-solid symbols represent P5 mice with undetermined sex. Statistical analyses were performed using nested one-way ANOVA followed by Tukey’s post hoc test.

Interneuron density in the late postnatal hippocampus is unchanged following partial deletion of CXCL12 from CR cells. A, Representative DAPI-stained hippocampal section showing the regions analyzed for interneuron quantification. Scale bar, 100 µm. B–D, Representative confocal images showing coimmunostaining for GAD67 and GABA in the CA1 pyramidal layer (B), hippocampal fissure (HF; C), and dentate gyrus (DG) granule cell layer (D) at P20. Dashed lines indicate the quantified regions, and arrowheads mark GAD67⁺ GABA⁺ double-positive interneurons. Scale bar, 50 µm. Quantification of total GAD67⁺ GABA⁺ interneurons is shown at the right. Error bars indicate mean ± SD. Two sections were analyzed per animal; solid and open symbols represent female and male mice, respectively. Statistical analyses were performed using nested one-way ANOVA with Tukey's post hoc test.

Selective loss of CXCL12 in CR cells does not impact astrocyte density in the postnatal hippocampus. A, Representative DAPI-stained hippocampal section showing the regions analyzed for astrocyte quantification. Scale bar, 100 µm. B–D, Representative confocal images showing coimmunostaining for NFIA, GFAP, and SOX2 in the CA1 pyramidal layer (B), hippocampal fissure (HF; C), and dentate gyrus (DG) granule cell layer (D) at P20. White arrowheads indicate NFIA⁺ GFAP⁺ SOX2⁺ astrocytes. Scale bar, 50 µm. Quantification of total NFIA⁺ GFAP⁺ SOX2⁺ astrocytes is shown at the right. Error bars indicate mean ± SD. Two sections were analyzed per animal; solid and open symbols represent female and male mice, respectively. Statistical analyses were performed using nested one-way ANOVA with Tukey's post hoc test.

Partial deletion of CXCL12 in CR cells preserves adult dentate gyrus neurogenesis. A, Representative DAPI staining showing the overall morphology of the dentate gyrus (DG) at P20. Quantification of the DG area is shown at the right. Scale bars, 200 µm. B, Quantifications of CA1 width based on CTIP2 immunostaining and DG area based on DAPI staining in 12-week-old mice. C, Representative high-magnification images of CTIP2 and PROX1 coimmunostaining in the 12-week-old control DG. Quantification of PROX1⁺ and CTIP2⁺ granule neuron densities is shown at right. Scale bar, 20 µm. SGZ, subgranular zone; GL, granular layer; ML, molecular layer. D, Representative confocal images showing coimmunostaining for GFAP, SOX2, and Ki67 in the DG of 12-week-old mice. White arrowheads indicate proliferating (Ki67⁺) GFAP⁺ SOX2⁺ radial glia-like cells. Scale bar, 20 µm. Quantification of total and proliferating (Ki67⁺) GFAP⁺ SOX2⁺ radial glia-like cells is shown at the right. E, Representative low-magnification images of DCX immunostaining in the DG. Scale bar, 200 µm. Quantification of DCX⁺ neuroblasts and immature neurons is shown at the right. Data are presented as scatterplots, with each data point representing one animal. Error bars indicate ±SD. Three sections were analyzed per animal. Solid and open symbols represent female and male mice, respectively. Statistical analyses were performed using nested one-way ANOVA with Tukey's post hoc test.