Transgenic 5$\times$FAD mice were obtained from Jackson Laboratories (strain: 034848JAX; Bar Harbor, ME, USA) and were maintained through regular breeding with C57BL/6 mice (Damool Science, Daejeon, Korea). Female C57BL/6 wild-type (WT) and 5$\times$FAD mice at 3, 6, or 12 months of age were included in this study. Genotyping confirmed the presence of five familial AD mutations: App KM670/671NL (Swedish), App I716V (Florida), App V717I (London), Psen1 M146L (A $>$ C), and Psen1 L286V. Mice were anesthetized with isoflurane (66794-013-10, Piramal Critical Care Inc, Bethlehem, PA, USA; 3% for induction, 1.5–2% for maintenance) and euthanized by cervical dislocation under deep anesthesia, in accordance with institutional guidelines. Experimental timelines are presented in Fig. 1A.

Fig. 1.

Schematic of the experimental design and behavioral changes in 5$\mathrm{\times }$FAD mice. (A) Pole and T-maze tests were performed on 5$\times$FAD and WT mice aged 3, 6, and 12 months. Subsequently, the animals were sacrificed to collect samples for Golgi staining, Western blot, and IHC. (B) Images representative of the pole tests are provided (left panel). The bar graphs illustrate time to turn and time to descend in the pole test in 5$\times$FAD mice compared with the WT (right panels). (C) Representative illustrations of the T-maze test are provided (left panel). The bar graphs illustrate the spontaneous alternation ratio observed in the T-maze test (right panel). Fig. 1A and Fig. 1C were generated using BioRender.com (https://www.biorender.com/). The data represent the combined mean $\pm$ standard error (SE) from two distinct experiments (n = 10/group). * p 0.05, ** p 0.01, *** p 0.001, **** p 0.0001. IHC, immunohistochemistry; WT, wild-type; 5$\times$FAD, five familial Alzheimer’s disease.

The pole test was performed based on a previously established method, incorporating minor modifications [18]. In brief, a metal pole (0.5 cm diameter, 50 cm height) was covered with gauze to prevent slippage. Each mouse was placed facing upward at the top of the pole, and the time taken to descend to the base was measured. Mice completed three trials, separated by 20-minute intervals, and the average times (s) were assessed.

The spontaneous alternation T-maze test was carried out using a previously reported method with minor changes [19]. A T-maze was constructed from black acrylic plastic, featured arms measuring 30 $\times$ 10 $\times$ 20 cm. Each mouse was placed at the start arm for 10 min, after which the central divider was removed to begin the trial. The animal was allowed to move freely toward either the left or right goal arm. Once the tail fully entered the chosen arm, it was closed off with a divider. After 30 s, the divider was lifted, and the mouse was returned to the start arm for another trial. This procedure was performed twice daily for three days, with 3-hour intervals between sessions. Scoring was performed as follows: a score of 0 was assigned if the mouse chose the same arm consecutively, and 1 if it alternated between arms within the trial.

To examine dendritic architecture, including spine morphology and density, neurons from the cornu ammonis (CA) 1 subregion and dentate gyrus (DG) of the hippocampus were subjected to Golgi staining. Following previously described procedures [20], we used the FD Rapid Golgistain™ kit (#PK401, FD Neurotechnologies, Columbia, MD, USA), following the instructions provided by the manufacturer. Brain hemispheres (n = 5/group) were rinsed with 0.1 M phosphate buffer saline (PBS, 1X; diluted from 10X stock, 502B828, Samchun Pure Chemical Co., Ltd., Gyeonggi-do, Korea) and immersed in Golgi–Cox solution for 2 weeks. Subsequently, the samples were placed in a sucrose-based solution at room temperature (RT; 22 $\pm$ 2 °C) for an additional 3 to 7 days. Coronal brain sections, cut at 200 µm thickness, were mounted on gelatin-coated slides with a small amount of sucrose solution. After air-drying in the dark for 3 days, the slides were processed according to the staining kit’s standard protocol.

Quantification of dendritic complexity, length, and branching was performed according to established methods [20]. Neuronal structures were imaged at 200$\times$ magnification using a Leica DM750 microscope (Leica Microsystems, Wetzlar, Germany), and neuronal traces were obtained with Leica Application Suite (v4.12) alongside Adobe Photoshop CS6 (Adobe Systems, San Jose, CA, USA). Sholl analysis was carried out with ImageJ software (version 1.53f51, NIH, Bethesda, MD, USA) using a concentric circle overlay technique [21]. For each animal (n = 5/group), 10 neurons were selected per hippocampal subregion following defined criteria [22]: The cell body was situated in the designated subregion, exhibiting thorough and effective branch staining along its entirety while remaining isolated from adjacent cells. Concentric circles were drawn from the center of the soma to the most distal dendritic end, spaced 10 µm apart. The number of dendritic intersections at each radius was recorded to assess arborization. Ten neurons per subregion were analyzed per mouse, and the mean value from each mouse was treated as n = 1. Group results were calculated as the mean $\pm$ standard error (SE), and sample details are provided in the respective figure legends.

The density and morphology of dendritic spines were measured according to an established methodology [23]. For this analysis, we enumerated all identifiable dendritic spines along a 30 µm segment of the distal dendrite of each neuron using a magnification of 1000$\times$. Only segments that were intact, clearly stained, and unbranched were selected. Spines were categorized into three morphological types: thin spine, featuring a small head and elongated neck; mushroom spine, distinguished by a large head and defined neck; stubby spine, characterized by a compact appearance and absence of a discernible neck. A total of 10 dendritic segments were analyzed per animal (n = 5/group), and spine density was calculated as the number of spines per 10 µm of dendritic length.

Western blotting was conducted in accordance with established protocols [24]. Each sample was sonicated in buffer H for 10 s, which contained 50 mM $\beta$-glycerophosphate (#G6376, Sigma-Aldrich, St. Louis, MO, USA), 1.5 mM ethylene glycol tetra-acetic acid (#E3889, Sigma-Aldrich), 1 mM dithiothreitol (#09779, Sigma-Aldrich), 10 µg/mL aprotinin (#A1153, Sigma-Aldrich), 1 mM phenylmethanesulfonyl fluoride (#P7626, Sigma-Aldrich), 0.1 mM Na₃VO₄ (#S6508, Sigma-Aldrich), 10 µg/mL leupeptin (#L2884, Sigma-Aldrich), and 2 µg/mL pepstatin (#EI10, Sigma-Aldrich) at pH 7.4. Subsequently, 6$\times$ sodium dodecyl sulfate (SDS) loading buffer was added, and samples were boiled at 100 °C for 10 min. Protein samples were separated on 10–15% SDS-polyacrylamide gel electrophoresis (#1658033, Bio-Rad; Hercules, CA, USA), followed by transfer onto polyvinylidene difluoride membranes (#10600023, Amersham Hybond-P; GE Healthcare Life Sciences, Pittsburgh, PA, USA). To block nonspecific binding, membranes were incubated at RT for 1 h in PBS (1X; diluted from 10X stock, 502B828, Samchun Pure Chemical Co., Ltd.) with 0.1% (v/v) Tween-20 (#P2287, Sigma-Aldrich; PBS-T; pH 7.4) containing 1% (v/v) normal goat serum (NGS; #S-1000, Vector Laboratories, Burlingame, CA, USA) and 0.5% (v/v) bovine serum albumin (BSA; #A9647, Sigma-Aldrich). Membranes were then incubated overnight at 4 °C with primary antibodies diluted in blocking solution and then left at RT for 1 h. Primary antibodies included mouse anti-A$\beta$ (1-16) (clone 6E10, 1:1000; #803001, BioLegend, San Diego, CA, USA), rabbit anti-p-Tau (Thr181) (1:1000; #12885S, Cell Signaling Technology, Danvers, MA, USA), mouse anti-glial fibrillary acidic protein (GFAP, 1:1000; #3670, Cell Signaling Technology), rabbit anti-ionized calcium-binding adapter molecule 1 (Iba-1, 1:1000; #016-20001, Fujifilm Wako, Osaka, Japan), rabbit anti-PSD-95 (1:1000; #2507, Cell Signaling Technology), and rabbit anti-activity-regulated cytoskeleton-associated protein (Arc; 1:1000; #ab183183, Abcam, Cambridge, UK). Subsequent to the primary antibody incubation, membranes were incubated for 2 h with either a horseradish peroxidase-conjugated anti-rabbit secondary antibody (1:5000; #31460, Thermo Fisher Scientific, Waltham, MA, USA) or mouse secondary antibody (1:5000; #31181, Thermo Fisher Scientific). Signals were visualized using the EZ-Western Lumi Femto kit (#DG-WF200, DoGenBio, Seoul, South Korea). Following stripping, membranes were re-probed with mouse anti-$\beta$-actin antibody (1:5000; #A5441, Sigma-Aldrich) at RT for 2 h. The optical density (OD) of each band was evaluated utilizing an iBright CL750 Imaging System (Thermo Fisher Scientific). The mean band intensity of the WT group was designated as ‘1’ for each blot, and relative expression levels were calculated as fold changes. Group means were calculated and reported as mean $\pm$ SE.

Immunohistochemistry (IHC) was performed in accordance with the protocol previously established [20]. Brain hemispheres were fixed and sectioned sagittally into 3 µm slices. The paraffin-embedded tissues were processed using the Vectastain® Elite ABC kit (Rabbit IgG, #PK-6101; Mouse IgG, #PK-6102, Vector Laboratories) in accordance with standard immunohistochemical procedures. To inhibit endogenous peroxidase activity, sections were incubated in 0.3% hydrogen peroxide (#4104-4400, Daejung Chemicals & Metals Co., Ltd., Siheung, South Korea) for 20 min. Subsequently, tissues were blocked with either 5% NGS or 5% normal horse serum (Vectastain® Elite ABC kit) in PBS-T at RT for 1 h to minimize non-specific binding. The sections were then incubated overnight at 4 °C with the following primary antibodies: mouse anti-A$\beta$ (1-16) (clone 6E10, 1:2000; #803001, BioLegend), rabbit anti-p-Tau (Thr181) (1:100; #12885S, Cell Signaling Technology), mouse anti-GFAP (1:4000; #3670, Cell Signaling Technology), and rabbit anti-Iba-1 (1:2000; #019-19741, Fujifilm Wako). Following three washes with PBS, the slides were incubated with the appropriate biotinylated secondary antibodies and subsequently treated with ABC peroxidase in accordance with the kit protocol (Vectastain® Elite ABC kit; Vector Laboratories). Finally, peroxidase labeling was developed using the 3,3^′-Diaminobenzidine (DAB) Substrate kit (Vector Laboratories), followed by hematoxylin counterstaining.

Cytokine levels, specifically tumor necrosis factor alpha (TNF$\alpha$, #DY410), interferon-gamma (IFN$\gamma$, #DY485-05), interleukin (IL)-1$\beta$ (#DY401), IL-6 (#DY406), and IL-10 (#DY417), were quantified in hippocampal lysates from 5$\times$FAD mice using commercial Enzyme-linked Immunosorbent Assay (ELISA) kits (R&D Systems, Minneapolis, MN, USA) following the manufacturer’s instructions. The hippocampal tissues were mechanically disrupted in lysis buffer, and the concentrations of total proteins were subsequently determined. Lysates were applied to ELISA plates, and cytokine concentrations were assessed by comparing sample absorbance to standard curves created with recombinant cytokines. Absorbance measurements were conducted using an Epoch Microplate Spectrophotometer (Bio-Tek Instruments, Winooski, VT, USA).

Statistical analyses were performed using GraphPad software (version 9.3.1, GraphPad Software, San Diego, CA, USA). To evaluate group differences in behavioral performance, dendritic complexity, and spine density between WT and 5$\times$FAD mice, a two-way ANOVA followed by Sidak’s post hoc test was used. Additional pairwise comparisons were conducted using an unpaired Student’s t-test. A significance threshold of p 0.05 was considered for all analyses. All numerical values are presented as mean $\pm$ SE. The number of samples included in each experiment is indicated in the corresponding Results sections and figure legends.

The 5$\times$FAD mouse model exhibits age-related deficits in motor function [25]; however, limited information exists regarding the temporal profile of motor dysfunction in these mice. Therefore, this study conducted a pole test to validate the fine motor dysfunctions in 5$\times$FAD mice (n = 10 mice/group; the time to turn down: F_interaction (2,54) = 1.946, p = 0.1527, the time to descend: F_interaction (2,54) = 8.825, p = 0.0005; Fig. 1B). Notably, no significant differences were observed at 3 months in the time it took the mice to turn down from the top of the pole (WT vs. 5$\times$FAD: p = 0.4608) or to descend to the base of the pole (WT vs. 5$\times$FAD: p = 0.7966). Comparatively, the 5$\times$FAD mice displayed a significantly prolonged time to turn (p = 0.0357) and descend (p = 0.0006) at 6 months compared to the WT mice. At 12 months, these impairments were significantly more pronounced, with 5$\times$FAD mice showing a significantly longer time to turn (p = 0.0004) and to descend (p 0.0001). Collectively, these findings indicate that motor dysfunction is evident in 5$\times$FAD mice and exacerbates with increased age.

Next, the T-maze test was conducted in 5$\times$FAD and WT mice at 3, 6, and 12 months of age to assess the working memory through spontaneous alternation behavior (n = 10/group; F_interaction (2,54) = 2.943, p = 0.0612; Fig. 1C). This assessment evaluates the innate exploratory behavior of rodents in novel environments, indicative of hippocampal-dependent cognitive function [26] (Fig. 1C, left scheme). No significant differences were noted in the spontaneous alternation ratio at 3 months between WT and 5$\times$FAD mice (p = 0.9955), indicating comparable cognitive performance during the early stage of AD pathology. However, by 6 months, the 5$\times$FAD mice exhibited a significant reduction in the spontaneous alternation ratio relative to WT mice (p = 0.0141), suggesting the emergence of hippocampus-dependent cognitive impairment is linked to AD pathology. Meanwhile, cognitive deficits in 5$\times$FAD mice were significantly more pronounced at 12 months—indicated by a further decrease in the spontaneous alternation ratio (p = 0.0042). Statistical analysis demonstrated an age-related cognitive decline in 5$\times$FAD mice, with significant differences observed at 6 months, which were exacerbated by 12 months (Fig. 1C, right bar graphs).

This study investigated the age-dependent changes in A$\beta$ and p-Tau (Thr181) expression in the hippocampi of 5$\times$FAD mice compared to WT controls using Western blot and IHC (n = 3 mice/group). Western blot analysis demonstrated a significant and progressive increase in A$\beta$_1-42 levels in 5$\times$FAD mice with age compared to WT controls (t(4) = 3.311, p = 0.0296 at 3 months; t(4) = 3.467, p = 0.0257 at 6 months; t(4) = 5.260, p = 0.0063 at 12 months) (Fig. 2A). Moreover, an age-dependent increase in p-Tau expression levels was observed, achieving statistical significance at 12 months (Fig. 2A). At 3 months, p-Tau levels did not significantly differ from WT controls (t(4) = 0.8643, p = 0.4362). At 6 months, p-Tau levels were elevated in 5$\times$FAD mice, although this difference did not reach statistical significance (t(4) = 2.166, p = 0.0963). At 12 months, hippocampal p-Tau levels in 5$\times$FAD mice significantly increased (t(4) = 3.854, p = 0.0182).

Fig. 2.

Expression levels of A$\beta$_1-42 and p-Tau (Thr181) in the hippocampus. (A) Representative Western blot images and bar graphs of the relative levels of A$\beta$_1-42 (~4 kDa) and p-Tau (Thr181) in the hippocampi of WT and 5$\times$FAD mice (n = 3 mice/group). The full-length blot images are presented in Supplementary Fig. 1. (B) Representative immunohistochemical images and bar graphs of the relative immunoreactivities of A$\beta$ and p-Tau (Thr181) in the hippocampus of WT and 5$\times$FAD mice (n = 3 mice/group). High-magnification versions of the insets are presented in Supplementary Fig. 2. The data represent the combined mean $\pm$ SE from two distinct experiments. * p 0.05, ** p 0.01, *** p 0.001. Scale bars represent 500 µm (low magnification) and 100 µm (inset) in panel B. A$\beta$, amyloid $\beta$; OD, optical density; p-Tau, phosphorylated Tau.

Next, we performed IHC to characterize further the spatial distribution of A$\beta$ and p-Tau (Thr181) expression across hippocampal subregions. IHC analysis demonstrated a significantly progressive increase in A$\beta$ immunoreactivity with age in 5$\times$FAD mice (t(4) = 5.879, p = 0.0042 at 3 months; t(4) = 8.966, p = 0.0009 at 6 months; t(4) = 4.443, p = 0.0113 at 12 months) (Fig. 2B, left panels). At 3 months, A$\beta$ exhibited extensive staining in the stratum pyramidale of CA1 and the distal CA1 subregion adjacent to the subiculum, with slight staining observed in the molecular layer of the DG. Immunoreactivity was more pronounced at 6 months than at 3 months of age, with A$\beta$ also detected in the stratum oriens of CA1, stratum lucidum of CA3, and the subgranular zone as well as in the hilus of the DG. At 12 months, the staining pattern was significantly enhanced across the hippocampus of 5$\times$FAD mice.

IHC analysis of p-Tau demonstrated a delayed yet progressive accumulation in the hippocampi of 5$\times$FAD mice (Fig. 2B, right panels). However, the p-Tau levels in 5$\times$FAD mice did not significantly differ from those in WT controls at 3 months (t(4) = 1.483, p = 0.2122). Conversely, the difference in p-Tau expression reached statistical significance at 6 and 12 months (t(4) = 4.275, p = 0.0129 at 6 months; t(4) = 3.822, p = 0.0187 at 12 months). In the spatial pattern analysis, the immunoreactivity of p-Tau was faintly detected at 3 months; however, p-Tau was observed in the stratum lacunosum-moleculare of CA1, distal CA1 area, and hilus at 6 months of age. Moreover, enhanced immunoreactivity was noted across the hippocampus of 5$\times$FAD mice at 12 months of age. Collectively, these findings indicate a notable age-related escalation in early A$\beta$ and a subsequent increase in p-Tau expression within the hippocampi of 5$\times$FAD mice.

We assessed the age-related alterations in astrocytic and microglial activation and cytokine levels in the hippocampi of 5$\times$FAD mice. GFAP and Iba-1, which serve as astrocytes and microglia markers, respectively, were assessed through Western immunoblotting and IHC (n = 3 mice/group). Western blot analysis revealed a progressive increase in GFAP and Iba-1 expression with age in 5$\times$FAD mice relative to WT controls (Fig. 3A). At 3 months, GFAP levels were marginally elevated in 5$\times$FAD mice, though this difference did not achieve statistical significance (t(4) = 1.986, p = 0.1180). Comparatively, GFAP expression was significantly elevated in 5$\times$FAD mice at 6 months and 12 months relative to WT controls (6 months: t(4) = 2.904, p = 0.0439; 12 months: t(4) = 4.792, p = 0.0087). Iba-1 levels were significantly elevated in 5$\times$FAD mice across all examined age groups: 3 months: (t(4) = 6.161, p = 0.0035), 6 months: (t(4) = 4.351, p = 0.0121), and 12 months: (t(4) = 5.931, p = 0.0041). In the IHC analysis, the left panels in Fig. 3B demonstrate no significant differences in the GFAP intensity in the hippocampus between groups at 3 months (t(4) = 1.054, p = 0.3515). However, a slight increase was observed in the 5$\times$FAD mice in the distal CA1 area. Elevated GFAP expression was significantly noted at 6 months (t(4) = 4.722, p = 0.0092) and 12 months (t(4) = 7.026, p = 0.0022). Notably, the intensities were markedly pronounced throughout the hippocampus of 5$\times$FAD at 6 and 12 months of age, particularly in the distal CA1, stratum oriens of CA1, and hilus. Iba-1 immunoreactivity in microglia (Fig. 3B, right panels) showed significant differences with age-dependent increases (t(4) = 3.320, p = 0.0294 at 3 months; t(4) = 4.742, p = 0.0090 at 6 months; t(4) = 4.031, p = 0.0157 at 12 months). Furthermore, the Iba-1 expression was primarily noted in the distal CA1 region at 3 months. In contrast, Iba-1 expression was present throughout the hippocampus at 6 and 12 months of age, especially in the distal CA1, stratum oriens of CA1, and molecular layer and hilus of the DG.

Fig. 3.

GFAP and Iba-1 expression and cytokine levels in the hippocampus. (A) Representative Western blot images and bar graphs of the relative levels of GFAP and Iba-1 in the hippocampi of WT and 5$\times$FAD mice (n = 3 mice/group). The full-length blot images are presented in Supplementary Fig. 3. (B) Representative immunohistochemical images and bar graphs of the relative immunoreactivities of GFAP and Iba-1 in the hippocampi of WT and 5$\times$FAD mice (n = 3 mice/group). High-magnification versions of the insets are presented in Supplementary Fig. 4. (C) The bar graphs illustrate the cytokine levels (TNF$\alpha$, IFN$\gamma$, IL-1$\beta$, IL-6, and IL-10) in the hippocampi of WT (n = 5) and 5$\times$FAD mice (n = 7) at 3, 6, and 12 months of age, as determined by ELISA. The data represent the combined mean $\pm$ SE from two distinct experiments. * p 0.05, ** p 0.01. Scale bars represent 500 µm (low magnification) and 100 µm (inset) in panel B. ELISA, enzyme-linked immunosorbent assay; GFAP, glial fibrillary acidic protein; Iba-1, ionized calcium-binding adapter molecule 1; IFN$\gamma$, interferon-gamma; IL, interleukin; TNF$\alpha$, tumor necrosis factor alpha.

Additionally, ELISA was employed to assess the inflammatory cytokine profile in the hippocampi of 5$\times$FAD mice for TNF$\alpha$, IFN$\gamma$, IL-1$\beta$, IL-6, and IL-10 at 3, 6, and 12 months of age (n = 5 for WT and n = 7 for 5$\times$FAD; Fig. 3C). The inflammatory cytokines levels in the hippocampi of 5$\times$FAD mice showed an overall upward trend across ages. At 3 months, the IL-1$\beta$ (t(10) = 3.042, p = 0.0124), IL-6 (t(10) = 2.632, p = 0.0251), and IL-10 (t(10) = 3.084, p = 0.0116) levels were significantly increased in 5$\times$FAD mice compared to WT controls (Fig. 3C, left bar graphs). At 6 months, the TNF$\alpha$ (t(10) = 2.375, p = 0.0390), IFN$\gamma$ (t(10) = 2.329, p = 0.0422), IL-1$\beta$ (t(10) = 3.016, p = 0.0130), IL-6 (t(10) = 2.472, p = 0.0330), and IL-10 (t(10) = 2.946, p = 0.0146) levels were significantly elevated (Fig. 3C, middle bar graphs). At 12 months, the elevated level of IL-1$\beta$ (t(10) = 2.567, p = 0.0280) was statistically significant (Fig. 3C, right bar graphs). Thus, these findings demonstrate a progressive and age-dependent increase in astrocytic and microglial activation, along with alterations in cytokine expression, in the hippocampi of 5$\times$FAD mice.

We assessed hippocampal structural plasticity alterations in 5$\times$FAD mice by analyzing the dendritic complexity of neurons in the CA1 and DG subregions, focusing on the number of crossing dendrites, total dendritic length, and branch points per neuron. Fig. 4 illustrates the counting of dendritic intersections at various radial distances from the neuronal soma in the CA1 and DG subregions, conducted through Sholl analysis (n = 5 mice/group). In the CA1 basal subregion (Fig. 4A, upper panels), the 5$\times$FAD group showed fewer dendritic intersections compared to the WT group at a Sholl radius of 50 µm from the soma (F_interaction (20,1960) = 1.944, p = 0.0073; left-upper line graphs) at 3 months. Dendritic intersections exhibited a significant reduction at Sholl radii of 10–80 µm from the soma (F_interaction (20,1960) = 4.581, p 0.0001; middle-upper line graphs) at 6 months, with a more pronounced effect observed at 12 months, particularly at Sholl radii of 10–90 µm (F_interaction (20,1960) = 5.652, p 0.0001; right-upper line graphs). The total dendritic length in the CA1 basal subregion was reduced in 5$\times$FAD mice relative to WT controls across all ages; however, these differences did not reach statistical significance (F_interaction (2,24) = 0.2901, p = 0.7507; Fig. 4A, left-upper bar graphs). Analysis of neuronal branch points (F_interaction (2,24) = 3.405, p = 0.0499; Fig. 4A, right-upper bar graphs) revealed no significant differences at 3 months (p = 0.8217). However, a significant reduction was observed in 5$\times$FAD mice at 6 months compared to WT controls (p = 0.0126), with a more pronounced decrease at 12 months (p = 0.0005).

Fig. 4.

Dendritic complexity of hippocampal neurons in WT and 5$\mathrm{\times }$FAD mice. Photographs illustrate representative neurons used for the Sholl analysis in the CA1 (A) and DG (B) subregions. Line graphs illustrate the mean number of intersections per 10 µm radial unit distance from the soma (0 µm) in the neuronal dendrites of CA1 basal (A, upper panels) and apical (A, lower panels), as well as in the DG (B) subregions. The bar graphs illustrate the total dendritic length (left panels) and the number of dendritic branch points (right panels) per neuron within each subregion. The data represent the combined mean $\pm$ SE from two distinct experiments involving 10 neurons per mouse and five mice per group. The actual images of the neurons are presented in Supplementary Fig. 5. The gray areas illustrate the SE of the means presented in the line graphs. * p 0.05, ** p 0.01, *** p 0.001, **** p 0.0001. Scale bars shown in the illustrated neuron images indicate 100 µm. CA1, cornu ammonis 1; DG, dentate gyrus.

No significant differences in dendritic intersections were noted in the CA1 apical subregion at 3 months between the 5$\times$FAD and WT groups (F_interaction (30,2940) = 1.199, p = 0.2103; Fig. 4A, left-lower line graphs). The dendritic intersections at 6 months showed a significant reduction at Sholl radii of 90–200 µm from the soma (F_interaction (30,2940) = 6.308, p 0.0001; Fig. 4A, middle-lower line graphs). This gap became more pronounced at 12 months, particularly at Sholl radii of 30, 50–70, and 90–190 µm (F_interaction (30,2940) = 10.070, p 0.0001; Fig. 4A, right-lower line graphs). The total dendritic length in the apical subregion (F_interaction (2,24) = 4.403, p = 0.0235; Fig. 4A, left-lower bar graphs) did not show a significant reduction in 5$\times$FAD mice compared to WT controls at 3 months (p = 0.2714). However, the total dendritic length in the apical subregion was statistically significant at 6 months (p = 0.0002) and 12 months (p 0.0001). Analysis of neuronal branch points (F_interaction (2,24) = 4.125, p = 0.0289; Fig. 4A, left-lower bar graphs) revealed no significant differences at 3 months. Meanwhile, a significant reduction was observed in 5$\times$FAD mice at 6 months compared to WT controls (p = 0.0001); this reduction persisted at 12 months (p = 0.0012).

No significant differences were observed in the dendritic intersections in the DG subregion between 5$\times$FAD and WT mice at 3 months (F_interaction (25,2450) = 0.8327, p = 0.7022; Fig. 4B, left line graphs). The dendritic intersections in 5$\times$FAD mice at 6 months were significantly diminished at Sholl radii of 80–140 µm from the soma (F_interaction (25,2450) = 5.886, p 0.0001; Fig. 4B, middle line graphs). This reduction became more pronounced at 12 months, particularly at Sholl radii of 100–160 µm (F_interaction (25,2450) = 4.635, p 0.0001; Fig. 4B, right line graphs). The total dendritic length (F_interaction (2,24) = 0.5119, p = 0.6057; Fig. 4B, left bar graphs) did not show a significant reduction in 5$\times$FAD mice, although a trend towards a decrease was observed across all ages in 5$\times$FAD group. Neuronal branch points in the DG (F_interaction (2,24) = 1.460, p = 0.2520; Fig. 4B, right bar graphs) exhibited a slight decrease at 3 months, with no significant difference observed. Alternatively, significant reductions were observed at 6 and 12 months in 5$\times$FAD mice compared to WT controls (p = 0.0049; p = 0.0103, respectively). Collectively, these findings suggest that dendritic complexity decreases with age in the CA1 and DG subregions of 5$\times$FAD mice.

Dendritic spines serve as the main components of synapses in hippocampal neurons and undergo rapid modifications in response to particular microenvironments, including neurodegenerative conditions [27]. This study investigated the temporal variations in dendritic spine density and morphology within the hippocampal CA1 and DG subregions of 5$\times$FAD and WT mice aged 3, 6, and 12 months (n = 5 mice/group; Fig. 5). Fig. 5A shows representative photomicrographs of neuronal dendrites across the various hippocampal subregions. No significant differences in spine density were observed in the CA1 basal subregion between 5$\times$FAD and WT mice at 3 months (F_interaction (2,24) = 3.295, p = 0.0544; Fig. 5B, left bar graphs). At 6 months, 5$\times$FAD mice exhibited a significant reduction in spine density (p 0.0001), which persisted at 12 months (p 0.0001). In the CA1 apical subregion of 3-month-old 5$\times$FAD mice (F_interaction (2,24) = 5.820, p = 0.0087; Fig. 5B, middle bar graphs), spine density exhibited a slight but significant decrease (p = 0.0450); a notable reduction was observed at 6 months (p = 0.0114), with an additional decrease at 12 months (p 0.0001). The spine density at 3 months in the DG subregion (F_interaction (2,24) = 9.058, p = 0.0012; Fig. 5B, right bar graphs) presented no significant difference between groups (p = 0.9738). At 6 months, 5$\times$FAD mice exhibited a significant reduction (p = 0.0018), which remained evident at 12 months (p 0.0001).

Fig. 5.

Dendritic spine density and morphology of hippocampal neurons in WT and 5$\mathrm{\times }$FAD mice. (A) Representative images of dendritic spines in the CA1 basal, CA1 apical, and DG regions of 5$\times$FAD and WT mice at 3, 6, and 12 months of age. (B) The bar graphs illustrate spine density per 10 µm of dendrite length across hippocampal subregions in both WT and 5$\times$FAD mice. (C) The bar graphs exhibit proportional variations in spine morphology categories (thin, stubby, and mushroom) per 10 µm of dendrite. The data represent the combined mean $\pm$ SE from two distinct experiments involving 10 neurons per mouse and five mice per group. * p 0.05, ** p 0.01, *** p 0.001, **** p 0.0001. Scale bars in panel A represent 5 µm.

Additionally, differences in the proportion of dendritic spine morphology (% in 10 µm dendrite) were observed in the subregions of the hippocampi of 5$\times$FAD and WT mice (Fig. 5C). The proportion of thin spines (Fig. 5C, left panels) was significantly elevated in the CA1 basal subregion (F_interaction (2,24) = 1.933, p = 0.1667) of 6-month-old 5$\times$FAD mice (p = 0.0385), the CA1 apical subregion (F_interaction (2,24) = 4.453, p = 0.0227) of both 6- and 12-month-old 5$\times$FAD mice (p = 0.009 and p = 0.0309, respectively), and the DG subregion (F_interaction (2,24) = 0.5795, p = 0.5678) of 6- and 12-month-old 5$\times$FAD mice (p = 0.0070 and p = 0.0063, respectively). In contrast, the proportion of mushroom spines (Fig. 5C, right panels) was significantly decreased in the CA1 basal subregion (F_interaction (2,24) = 0.7758, p = 0.4715) of 6-month-old 5$\times$FAD mice (p = 0.0306), the CA1 apical subregion (F_interaction (2,24) = 6.364, p = 0.0061) of both 6- and 12-month-old 5$\times$FAD mice (p = 0.0017 and p 0.0001, respectively), and the DG subregions (F_interaction (2,24) = 1.003, p = 0.3818) of 6- and 12-month-old 5$\times$FAD mice (p = 0.0179 and p = 0.0026, respectively). No significant difference was exhibited between WT and 5$\times$FAD mice in the proportion of the stubby spines across all hippocampal subregions and age (Fig. 5C, middle panels). Collectively, these findings indicate that changes in spine density and morphology in the hippocampal neurons of 5$\times$FAD mice significantly worsen with advancing age.

The subsequent investigation focused on the molecular alterations associated with postsynaptic dysfunction in the hippocampi of 5$\times$FAD mice, specifically analyzing the protein expression levels of PSD-95 and Arc, which serve as markers of neuroplasticity [28, 29]. Western blot analysis was performed on hippocampal lysates from 3-, 6-, and 12-month-old WT and 5$\times$FAD mice (n = 3 mice/group; Fig. 6). At 3 months, no significant differences in PSD-95 or Arc expression were noted between WT and 5$\times$FAD mice (t(4) = 1.005, p = 0.3719; t(4) = 0.2712, p = 0.7997, respectively). Conversely, both PSD-95 and Arc expression were significantly reduced in 5$\times$FAD mice at 6 months relative to WT controls (PSD-95: t(4) = 3.169, p = 0.0339; Arc: t(4) = 3.746, p = 0.0200). While Arc levels remained consistently reduced at 12 months (t(4) = 2.875, p = 0.0452), PSD-95 expression showed a more pronounced decline (t(4) = 4.639, p = 0.0097). The results indicate an age-related decrease in the expression of PSD-95 and Arc in the hippocampi of 5$\times$FAD mice.

Fig. 6.

Expression levels of PSD-95 and Arc in the hippocampi of 5$\mathrm{\times }$FAD and WT mice. Representative Western blot images and bar graphs of the relative levels of PSD-95 and Arc in the hippocampi of WT and 5$\times$FAD mice (n = 3 mice/group). The full-length blot images are presented in Supplementary Fig. 6. The data represent the combined mean $\pm$ SE from two distinct experiments. * p 0.05, ** p 0.01. Arc, activity-regulated cytoskeleton-associated protein; PSD-95, postsynaptic density protein-95.