Molecular Characterization of Superficial Layers of the Presubiculum During Development

Abstract

The presubiculum, a subarea of the parahippocampal region, plays a critical role in spatial navigation and spatial representation. An outstanding aspect of presubicular spatial codes is head-direction selectivity of the firing of excitatory neurons, called head-direction cells. Head-direction selectivity emerges before eye-opening in rodents and is maintained in adulthood through neurophysiological interactions between excitatory and inhibitory neurons. Although the presubiculum has been physiologically profiled in terms of spatial representation during development, the histological characteristics of the developing presubiculum are poorly understood. We found that the expression of vesicular glutamate transporter 2 (VGluT2) could be used to delimit the superficial layers of the presubiculum, which was identified using an anterograde tracer injected into the anterior thalamic nucleus (ATN). Thus, we immunostained slices from mice ranging in age from neonates to adults using an antibody against VGluT2 to evaluate the VGluT2-positive area, which was identified as the superficial layers of the presubiculum, during development. We also immunostained the slices using antibodies against parvalbumin (PV) and somatostatin (SOM) and found that in the presubicular superficial layers, PV-positive neurons progressively increased in number during development, whereas SOM-positive neurons exhibited no increasing trend. In addition, we observed repeating patch structures in presubicular layer III from postnatal days 12. The abundant expression of VGluT2 suggests that the presubicular superficial layers are regulated primarily by VGluT2-mediated excitatory neurotransmission. Moreover, developmental changes in the densities of PV- and SOM-positive interneurons and the emergence of the VGluT2-positive patch structures during adolescence may be associated with the functional development of spatial codes in the superficial layers of the presubiculum.

Keywords: adeno-associated virus; anterior thalamus; calbindin; development; interneuron; mouse; presubiculum; vesicular glutamate transporter 2.

Figure 1

Anterograde tracing of the presubiculum by injection of adeno-associated viral (AAV) vectors into the ATN. (A) Experimental diagram. AAV (red) was injected into the ATN (light blue). Two-week after the injection, the PrS (purple) was anterogradely traced. After perfusion and postfixation, the brain was first coronally sectioned from the rostral end to confirm the injection site (i.e., ATN), followed by horizontal sectioning from the dorsal end for the presubiculum. (B) Representative immunohistochemical image (10×) of the injection site. Nissl, anti-vesicular glutamate transporter 2 (VGluT2), and mCherry signals are represented by blue, green, and red, respectively. Note that in the current study, the ATN is subdivided into the anterior dorsal thalamus (AD) and the anterior ventral thalamus (AV), each of which is encircled by white dashed lines. (C) Low-magnification image (4×) of the hippocampus and parahippocampal region in a slice of an AAV-injected mouse. The AAV-mediated anterograde tracing signal was confined to a region downstream of the subiculum. (D) Fluorescent image (10×) of the boxed area in (C). (E) Magnified image (40×) of the boxed area in (D). (F) Magnified image (60× objective and 2× electronic zoom) of the boxed area in (E). Note that synaptic terminals or boutons from ATN (red) include VGluT2 (green), suggesting that ATN axons target presubicular neurons. ATN, anterior thalamic nucleus; PrS, presubiculum; DG, dentate gyrus; Sub, subiculum.

Figure 2

Representative photographs of VGluT2 immunoreactivity and AAV-mediated anterograde tracing in superficial layers of the presubiculum from an adult mouse along the dorsoventral axis. (A) Superficial layers of the presubiculum were stained for Nissl substances (blue, leftmost (first)) and VGluT2 (green, second) and simultaneously visualized with AAV-mediated anterograde tracing (red, third) in a slice at 600 μm depth. A merged image is displayed in the fourth panel. The dorsoventral level was indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–H) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, 1,800 μm, 2,100 μm, 2,400 μm, and 2,700 μm, respectively. VGluT2, vesicular glutamate transporter 2; AAV, adeno-associated virus.

Figure 3

Spatial correlations between all possible pairs of three signals at different dorsoventral levels. Spatial correlations were calculated between all possible pairs of signals (i.e., Nissl vs. VGluT2, Nissl vs. mCherry (as a result of AAV-mediated anterograde tracing), and VGluT2 vs. mCherry). At either dorsoventral level, where the most dorsal section is defined as 0 μm, the spatial correlations between VGluT2 and mCherry were significantly higher than those between the other two pairs (n = 5–6 mice, *P < 0.05, paired t-test with post hoc Bonferroni correction). VGluT2, vesicular glutamate transporter 2; AAV, adeno-associated virus.

Figure 4

Dorsal-to-ventral decrease in the area of the superficial layers of the presubiculum across all ages. The superficial layers of the presubiculum were determined based on VGluT2 immunofluorescence. The area of the superficial layers was estimated at 600 μm, 1,200 μm, 1,800 μm, and 2,400 μm (at ages older than P7) across all ages tested (i.e., P5, P6, P7, P8, P10, P12, P14, P21, and 6 week old). At all ages, the area tended to decrease from the dorsal to ventral end (n = 3–4 mice, *P < 0.05, Tukey–Kramer test).

Figure 5

Complementary pattern of AAV-mediated tracing and calbindin immunosignals in the superficial layers of the presubiculum. (A) Presubiculum (at 700 μm from the most dorsal section) was imaged by staining Nissl substances (gray, leftmost (first)), AAV-mediated fluorescent tracing (red, second), and immunostaining of calbindin (green, third). The merged signal (except for Nissl) is shown in the fourth panel. Calbindin is enriched in layer II in the presubiculum, whereas mCherry signals are evident in layers I and III. The presubicular superficial layers are indicated by a white loop in the first panel. A magnified image of the white boxed area in the fourth panel is displayed in the fifth panel. (B,C) The same as (A) but at 1,600 μm and 2,500 μm, respectively. PrS, presubiculum; AAV, adeno-associated virus.

Figure 6

Representative photographs of the presubicular superficial layers of a postnatal 5-day-old mouse. (A) Superficial layers of the presubiculum of a postnatal 5-day-old mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–E) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, and 1,800 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 7

Representative photographs of the presubicular superficial layers of a postnatal 6-day-old mouse. (A) Superficial layers of the presubiculum of a postnatal 6-day-old mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm).(B–E) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, and 1,800 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 8

Representative photographs of the presubicular superficial layers of a postnatal 10-day-old mouse. (A) Superficial layers of the presubiculum of a postnatal 10-day-old mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–G) The same as (A), but at 900 μm, 1,200 μm, 1,500 μm, 1,800 μm, 2,100 μm, and 2,400 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 9

Representative photographs of the presubicular superficial layers of a postnatal 12-day-old mouse. (A) Superficial layers of the presubiculum of a postnatal 12-day-old mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–H) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, 1,800 μm, 2,100 μm, 2,400 μm, and 2,700 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 10

Representative photographs of the presubicular superficial layers of a postnatal 21-day-old mouse. (A) Superficial layers of the presubiculum of a postnatal 21-day-old mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–H) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, 1,800 μm, 2,100 μm, 2,400 μm, and 2,700 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 11

Representative photographs of the presubicular superficial layers of a 6-week-old mouse. (A) Superficial layers of the presubiculum of a 6-week-old (adult) mouse were stained for Nissl substances (gray, leftmost (first)) and immunostained for VGluT2 (green, second), parvalbumin (PV; blue, third), and somatostatin (SOM; red, fourth) in a slice at 600 μm depth. A merged image (except for Nissl) is displayed in the fifth panel. The presubicular superficial layers are delineated by a white loop in the first panel. The dorsoventral level is indicated as the distance (μm) from the most dorsal section (i.e., 0 μm). (B–H) The same as (A) but at 900 μm, 1,200 μm, 1,500 μm, 1,800 μm, 2,100 μm, 2,400 μm, and 2,700 μm, respectively. PrS, presubiculum; VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 12

Presubicular neurons coexpressing PV and SOM. (A) Representative image (20×) of the superficial layers of the adult (6-week-old) presubiculum immunostained for VGluT2 (green), PV (blue), and SOM (red). (B) High-magnification (60×) image of the white boxed area in (A). The section was stained for Nissl substances (gray, leftmost (first)) and immunostained for PV (blue, second) and SOM (red, third). The merged image is displayed in the fourth panel, which shows (i) neurons coexpressing both PV and SOM and (ii and iii) neurons expressing PV alone and SOM alone, respectively. VGluT2, vesicular glutamate transporter 2; PV, parvalbumin; SOM, somatostatin.

Figure 13

Progressive increase in the number of PV-expressing (not SOM-expressing) interneurons in the superficial layers of the presubiculum during development. (A) Density of PV-expressing neurons (blue, left) and SOM-expressing neurons (red, right) in the superficial layers of the presubiculum (at 600 μm from the dorsomost level) during development. Note that the number of PV-expressing neurons increases progressively with age, whereas the number of SOM-expressing neurons does not. (B–D) The same as (A) but at 1,200 μm, 1,800 μm, and 2,400 μm, respectively. PV, parvalbumin; SOM, somatostatin.

Figure 14

Emergence of VGluT2-positive repeating patch structures in layer III of the presubiculum from 12 day postnatal mouse. (A) Representative immunohistochemical image of the presubiculum of a postnatal 10-day-old mouse (top). The section was immunostained for VGluT2 (green). The top image is shown in a pseudocolored manner (middle). The VGluT2 immunosignals are almost uniform. The schema of the presubicular superficial layers is presented in the bottom panel. (B) The same as (A) but for a postnatal 12-day-old mouse. At this age, VGluT2 immunosignals in layer III are heterogeneous, as indicated by the white arrows. The VGluT2-positive repeating structure is depicted in the schema in the bottom panel (green). (C–E) The same as (B) but for postnatal 14-day-old, 21-day-old, and 6-week-old mice, respectively. VGluT2, vesicular glutamate transporter 2; PrS, presubiculum.

Figure 15

Summary of the current study. (A) Bird’s-eye view of the mouse brain. We injected AAV-hSyn-Syn-mCherry (red), an anterograde tracer, into the ATN (pale blue) and visualized the superficial layers of the presubiculum (purple) based on mCherry signals. (B) mCherry signals (red) in the presubicular superficial layers totally overlapped with the VGluT2 immunosignals (green). We then considered that the VGluT2 immunosignals could be used to delimit the presubicular superficial layers during development. (C) We coimmunostained VGluT2, PV, and SOM during development and found that the number of PV-positive interneurons gradually increased after P5. We further accidentally found that the VGluT2 immunopositive area in layer III in the presubiculum formed a repeating patch structure from P12. VGluT2, vesicular glutamate transporter 2; PrS, presubiculum; ATN, anterior thalamic nucleus; DG, dentate gyrus; PV, parvalbumin; SOM, somatostatin; AAV, adeno-associated virus.