Erythropoietin Stimulates GABAergic Maturation in the Mouse Hippocampus

Increased number of GABA-immunoreactive neurons in the hippocampus from Tg21 mice. A, Representative images of double immunofluorescence staining against GABA (green) and NeuN (magenta) for CA3 and CA1 areas of WT and Tg21 mice at P21. so: stratum oriens, sp: stratum pyramidale, sr: stratum radiatum. B, Unbiased quantification of NeuN+ (upper graph) and GABA+ (middle graph) cells in total CA3 and CA1 area of WT and Tg21 mice, showing age-specific differences between genotypes. Ratio of GABA+/NeuN+ cells (lower graph) in WT and Tg21 mice, shows a significant increase of GABA+ cells at P14. Data are given as mean ± SD of total neuronal numbers in the CA1 and CA3 area of the hippocampus; N = 6 animals per age and genotype. Two-way ANOVA test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Scale bar: 50 μm. C, Representative images of PV+, SST+, and NPY+ immunoperoxidase staining in hippocampus of WT and Tg21 mice at P14 illustrate the stronger immunoreactivity in PV, SST, and NPY in Tg21 mice. DG: dentate gyrus. D, Unbiased quantification of the total cell numbers in CA3 and CA1 areas, revealing increased numbers of PV+ cell in CA3, increased number of SST+ cells in CA1, and increased number of NPY+ cells in CA3 and CA1 in Tg21 mice. Data are given as mean ± SD, N = 6 animals per genotype for PV and NPY and N = 4 animals per genotype for SST staining. Multiple t test; *p < 0.05, **p < 0.01. Scale bar: 200 μm.

Early onset of PV expression in the hippocampus from Tg21 mice. A, Representative images of PV immunoperoxidase staining in CA3 and CA1 areas of WT and Tg21 mice at P7, P11, and P21. Note the earlier appearance of PV+ cells in the CA3 area (P7). Scale bar: 50 μm. B, Unbiased quantification of the total PV+ cell numbers at each represented postnatal time point for WT and Tg21 mice. Multiple t test; *p < 0.05, **p < 0.01. C, Significant differences in PV+ cell numbers between genotypes occur in CA3 area at P7, P11, and P14. Grofit interference analysis, *p < 0.05, **p < 0.01. D, Significant difference in PV+ intensity is to observe in the CA3 area at postnatal ages: 7, 11, 14, and 21; and in the CA1 area at postnatal ages: 14 and 21. Data are given as scatter dot plots and mean ± SD, N = 6 animals per genotype. Two-way ANOVA test; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. E, Representative images of cleaved caspase-3 (green) and DAPI (blue) staining in CA1 areas of WT and Tg21 mice at P7. Scale bar: 50 μm. F, Cleaved caspase-3 quantification in WT and Tg21in CA3 and CA1 area at P7 and P11. A significant reduction in apoptosis at P7 is observed in Tg21 mice. Two-way ANOVA; *p < 0.05.

EPO influences PNN formation around PV+ cells in the hippocampal CA1 and CA3 areas. A, Representative images of double labeling for PV immunoreactivity (magenta) and WFA fluorescence that identify PNNs (green) and merged images in CA3 and CA1 areas at P11. Scale bar: 100 μm. B, Quantification of PV+ cells surrounded by PNN (WFA+) in CA3 and CA1 across postnatal ages. No increase in WFA+ cells nor change in the onset is observed between genotypes. C, WFA intensity in CA3 and CA1 across postnatal ages. WFA fluorescence intensity is stronger in Tg21 mice CA3 area at P11 and in CA1 area at P11 and P14. D, PV intensity in WFA+ cells in CA3 and CA1 area across postnatal ages. PV intensity is stronger in CA3 and CA1 areas at P11 in Tg21 mice. E, Correlation analysis of PV and WFA fluorescence intensity in WT and Tg21 mice CA1 area at P11. r: Pearson correlation. F, PV+ cells soma size at P11. PV+ cells covered by WFA have larger cell somas in CA3 and CA1 area from WT and Tg21 mice. WFA+/PV+ cells are larger in Tg21 mice CA1 area. G, PV immunofluorescence intensity at P11. PV intensity is stronger in WFA+ cells, and the intensity is even higher in Tg21 mice in CA3 and CA1 areas. Graphs (B–F) show scatter dot plots and mean bars ± SD, N = 6 animals per age and genotype. Two-way ANOVA; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Increased GABA_AR cluster density in CA1 pyramidal cells of Tg21 mice at P14. A, Representative images of immunofluorescence staining against VGAT: vesicular GABA transporter (blue), gephyrin (green), and γ2-GABA_AR subunit (red) in WT and Tg21 mice. B, Quantification of presynaptic (VGAT) terminals in WT and Tg21 mice showing increased terminal size (cumulative plot) in Tg21 mice. C, Quantification of postsynaptic clusters of gephyrin showing increased cluster number (bar graph) and size (cumulative plot). D, Quantification of γ₂-GABA_AR subunits in WT and Tg21 mice showing increased density (bar graph) in Tg21 mice. Density of postsynaptic (triple labeled) clusters in WT and Tg21 mice at P14 (E), P21(F), and P60 (G). Bar graphs show mean ± SD, data points represent individual mice, N = 5 animals per age and genotype, Student’s t test (B–J). Cumulative frequency plots show total number of quantified clusters in WT and Tg21 animals; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Scale bar: 10 μm.

Enhancement of GABAergic synaptic transmission in CA1 area of Tg21 mice at P13–P15. A, Representative raw and averaged mean traces of the mIPSCs in WT and Tg21 mice at P13–P15. B–E, Cumulative frequency distribution plots and data points graphs at P13–P15 of (B) IEIs and frequency, (C) amplitude, (D) rise-time constant, and (E) decay-time constant of the mIPSCs from WT (black bars) and Tg21 (red bars). An increase in frequency, amplitude, rise time, and decay time is observed in Tg21mice at P13–P15. F, Representative raw and averaged mean traces of the mIPSCs for WT and Tg21 mice at P19–P22. G–J, Cumulative frequency distribution plots at P19–P22 of (G) IEIs and frequency, (H) amplitude, (I) rise-time constant, and (J) decay-time constant. No differences are observed in Tg21 mice at P19–P22. Graphs show mean ± SD, data points represent individual mice, N = 9 animals per genotype (P13–P15) and N = 8 animals (P19–P22), Student’s t test. Cumulative frequency plots show total number of events in WT and Tg21 animals, KS tests; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

EPORs expression in the CA1 area is restricted to principal cells. Representative images of fISH for mRNA EPORs (green dots) colabeled with DAPI (gray) in WT and Tg21 mice at P7, P11, and P21 in (A) stratum pyramidale (SP) and (C) stratum radiatum (SR). Negative control (P7): fluorophore; positive control (P7): housekeeping gene. Scale bar: 10 μm. Quantification of EPOR mRNA dots in CA1 SP (B) and CA1 SR (D) at different postnatal ages, showing its selective presence in SP throughout postnatal development. More EPOR mRNA dots are quantified at P7 in Tg21 mice. Data are given as mean with individual values ± SD, N = 6 animals per genotype and age. Two-way ANOVA test; CA1 SP, ***p < 0.001.

Constitutive deletion of EPORs from Gad65 cells has no impact on the GABAergic system. A, Representative images of fISH for EPORs in GAD65-cre^+/+, EPOR^fx/fx mice and GAD65-creT^g/+, EPOR^fx/fx (GAD65: Glutamat decarboxylase isoform 65) mice at P11 in CA1 stratum pyramidale area. B, Quantification of EPOR mRNA dots in CA1 area, showing no effect of the targeted mutation. C, Representative images of PV immunoperoxidase staining in GAD65-cre^+/+, EPOR^fx/fx mice and GAD65-creT^g/+, EPOR^fx/fx mice at P11 showing no change in PV immunostaining. D, Unbiased quantification of PV⁺ cell numbers in CA1 area shows no alteration in cell numbers on deletion of EPOR in interneurons. Data are given as mean ± SD, N = 3 animals and 3 hippocampi per genotype for fISH and N = 6 animals per genotype for PV stereology. Two-way ANOVA test. Scale bars: 10 μm (A) and 200 μm (C).