Figure 2.

TrkB Signaling Regulates CREB Activation in Immature DGCs and the Integration and Maturation of These Cells into the GCL

(A–F) Immature DGCs accumulate at the hilar border in the absence of TrkB signaling. (A, C, and E) and (B, D, and F) Representative images of hippocampal coronal sections from controls (BAC-Gad1-Cre^tg/+; Ai9^T/+) and mutants (Trkb^Gad1-KO; Ai9^T/+), respectively, at P8, P21, and 2M immunostained for doublecortin (DCX) (green) and superimposed to tdTomato signal (red). At P8, DCX immunostaining shows mainly colocalization with endogenous tdTomato (red) indicating migrating DGCs and a similar pattern between controls and mutants (A and B and insets). DAPI staining (blue). At P21, however, DCX immunostaining shows accumulation of immature DGCs in the subgranular zone (SGZ) of mutants compared with controls (C and D and respective insets). At 2M, as expected, very few DCX-positive DGCs are present in the SGZ of controls (E and insets), whereas in mutants the SGZ appears still abundant in DCX-positive cells, although reduced compared with P21 (F and insets). Scale bars: 100 μm in (A and B) and respective insets; 100 μm in (C–F) and 50 μm in respective insets. White arrowheads indicate examples of colocalization between tdTomato and DCX occurring only in immature neuron DGCs beginning to migrate and mature in the GCL, but not in neural precursors (yellow arrows).

(G–L) Representative images of hippocampal coronal sections and insets from controls (G, I, and K, BAC-Gad1-Cre^tg/+; Ai9^T/+) and mutants (H, J, and L, Trkb^Gad1-KO; Ai9^T/+), respectively, at P8, P21, and 2M immunostained for phosphoCREB (pCREB). The time course analysis shows pCREB peak of activation at P21 in control mice appearing in the lower third of the granule cell layer (GCL) at the hilar border (Hlb) where newborn immature neurons reside (I and inset). pCREB immunoreactivity declined by 2M age (K and inset). Trkb deletion in immature DGCs induces reduced activation of CREB at P21 compared with controls (J and I) followed by a decline at 2M (L). Scale bars: 100 μm in (G and H) and 50 μm in insets; 200 μm in (I–L) and 100 μm in insets. MF, mossy fibers; ∗examples of autofluorescence in structures like blood vessels.

(M–O) Golgi staining of newly formed DGCs to analyze dendritic density in the presence and absence of TrkB signaling. (M) Representative image of 2M-old Golgi-stained granule cells in the GCL and drawing of the indicated (arrow) cell showing branching ramification. (N) A similar number of dendritic spines was found between genotypes in proximal dendrites (second to third order, Trkb^Gad−1-WT, 0.68 ± 0.05; n = 33 cells from 4 mice; Trkb^Gad−1-KO, 0.71 ± 0.05; n = 34 cells from 3 mice; p = 0.99) of DGCs, whereas a significant reduction in dendritic spines was observed along distal dendrites in mutant mice compared with control littermate (fourth order, Trkb^Gad1-WT, 1.15 ± 0.08; n = 5 cells from 4 mice; Trkb^Gad1-KO, 0.88 ± 0.07; n = 19 cells from 3 mice; ∗p = 0.02). Data are number of spines per 1-μm dendritic segment ± SEM. (O) Representative images of 10-μm dendritic segment of second and fourth order from mutants and control mice.

See also Figure S2.