Twenty APP/PS1 mice (8-month-old, 44–50 g) and ten wild-type (WT) littermate mice (8-month-old, 38–44 g) were provided by Liaoning Changsheng Biotechnology Co., Ltd. (Benxi, Liaoning, China). All mice stayed in the SPF-level laboratory, where each mouse lived in a single cage with sufficient water and food. 10 WT mice were included in the WT group (n = 10). All APP/PS1 mice were randomly and equally divided into the APP/PS1 group and RB-treated APP/PS1 group (n = 10). Mice in the WT and APP/PS1 group were given normal saline (0.9%) orally every day, while the RB-treated APP/PS1 mice were treated with 20 mg/kg RB (117-02-2, Chengdu Herbpurify Co., Ltd., Chengdu, Sichuan, China) through the same way. The intragastric administration dose was 0.1 mL/10 g, which means 0.1 mL was given for every 10 g of mouse body weight. The whole agent treatment lasted for 8 weeks and behavioral tests began since the 6th week. Subsequently, the mice were intraperitoneally injected with 150 mg/kg of pentobarbital (P3761, Sigma-Aldrich, Shanghai, China) sodium and their brain samples were procured for subsequent experimental analysis. Animal experiments were approved by Institutional Animal Care, use Committee of Jilin University (SY202103007), and the date (10-03-2021) of this approval. All procedures were in accordance with ARRIVE guidelines.

The MWM test was conducted on the 36th day. MWM test included 2 parts, the 6-day navigation test and 1-day probe trial. In the navigation test, mice were put into the muddy water containing TiO₂ in the tank which is 120 cm in diameter. Their goal was to find the platform, which was hidden 2 cm below the water surface, within a 60 s time limit. If they were unsuccessful in finding the platform, they would be guided to it and required to stay there for 30 s to become familiar with its location. All mice were trained for 5 days and the formal test was carried out on the 6th day (day 41st since the beginning of agent treatment). In the probe trial, the platform was taken away, and the mice were permitted to swim freely in the opaque water for 60 s. A video tracking system (XR-XM101, Shanghai XinRuan Information Technology Co., Ltd., Shanghai, China) was utilized to record the every-day escape latency, the crossing numbers in the platform area and time staying in this area in the probe trail, and their trajectories. Data analysis was conducted by an observer unaware of the treatment conditions.

Since the 44th day, the step-down test was initiated. The equipment consisted of was a square box (30 cm $\times$ 30 cm $\times$ 30 cm), with a circular platform (8 cm in diameter) positioned at its center (XR-3TB, Shanghai XinRuan Information Technology Co., Ltd., Shanghai, China). This platform was elevated 4 cm above a grid floor made of stainless steel. As previously mentioned, the mice were initially placed on the grid floor and subjected to an electric shock, prompting them to jump onto the platform to evade the stimulus. Following a 24-hour interval, the mice were placed on the platform again. The step-down latency was determined by measuring the time it took for each mouse to jump back down to the floor for the first time. A maximum step-through latency of 300 s was established. During this period, the number of times each mouse jumped down was recorded as error counts.

The step-through test [17] was performed since day 47th. The apparatus comprised two identical compartments (each measuring 20 cm $\times$ 13 cm $\times$ 12 cm), one illuminated and the other dark (XR-Med, Shanghai XinRuan Information Technology Co., Ltd.). An automated doorway separated these two compartments. The floor in the dark compartment was charged with electricity. In the training session, the mice were first put in the light compartment for 10 s of free exploration. Later, the door was opened, and driven by their natural curiosity, the mice were inclined to enter the dark compartment. Upon chamber entry, the door was shut, and subjects received a 0.2 mA electric foot shock for 5 seconds. After a 48-hour interval, mice were placed back into the light compartment with the open-door configuration. The latency to enter the dark compartment was measured as step-through latency, with a 300-second ceiling. Blinded evaluation was conducted by an observer unaware of treatment assignments.

On day 50th, NOR test was performed referring to Wan et al. [18]. Initially, the mice were permitted to explore an empty apparatus (50 cm $\times$ 50 cm) for a period of 5 min to become accustomed to the environment. Subsequently, two identical objects, labeled A and B, were introduced into the apparatus, and the mice were given an additional 5 min to explore. Following the exploration phase, the mice were taken out, and the apparatus was meticulously cleaned with 75% ethyl alcohol to remove any remaining odors. Then, object B was substituted with a new object C, which was identical in size and material but featured a distinct shape. The time spent by the mice exploring each object was meticulously recorded by a video tracking system (XR-XM101). The recognition index (%) was calculated using the formula: New object exploration time / (new object exploration time + old object exploration time) $\times$ 100%. Data analysis was carried out by an observer unaware of the treatment groups to maintain objectivity.

The general health status of the mice was closely monitored throughout the entire experimental period. Following the completion of behavioral tests, all mice were euthanized using CO₂ anesthesia. CO₂ was displaced into the euthanasia vessel at a flow rate of 40% of the chamber’s volume per minute. Like Wan et al. [18], after mice were euthanized, the intact brain samples were fixed with 4% paraformaldehyde (BL539A, biosharp, Hefei, Anhui, China) instantly followed by dehydration and embedding. The paraffin-embedded blocks were cut into slices with a thickness of 5 µm. After re-hydrated, slices were subjected to antigen retrieval at 96 °C. Then slices were sequentially blocked with 5% bull serum albumin (abs9157, Absin (Shanghai) Bioscience Inc., Ltd., Shanghai, China), incubated with anti-A$\beta$_1–42 (A24422, 1:500, ABclonal, Wuhan, China) and secondary antibody (E-AB-1003, 1:5000, Elabscience, Wuhan, Hubei, China), and finally stained with DAB and hematoxylin. All agents were from IHC detection kit (RK05872, ABclonal). Images were taken using microscope (BX51, Olympus, Beijing, China). For all antibodies’ information see Supplementary Material-antibodies.

N2a cells are a neuroblastoma cell line with neuronal and amoeboid stem cell morphology isolated from brain tissue. Mouse neuroblastoma N2a cells (CL-0168, Procell, Wuhan, Hubei, China) were maintained in MEM medium (11575032) supplemented with 10% fetal bovine serum (16140071) and 1% penicillin-streptomycin solution (15140148). Cells were cultured at 37 °C in a moist environment with 5% CO₂. All reagents were offered by Thermo (Waltham, MA, USA). The N2a cells used in the experiment were morphologically characterized and confirmed to be free of mycoplasma contamination.

For the agent treatment in the following experiments, in the 3 h pretreatment period, N2a cells in the RB-treated groups were treated with 5 µM RB and 20 µM RB, respectively. In both the control and model groups, the medium was replaced with an equal volume of basic MEM. Then in the 24 h treatment period, the control group was still treated with basic MEM. The model group was treated with 5 µM A$\beta$_1–42 (107P64, Taigu Biology, Nanjing, Jiangsu, China). The RB-treated groups were firstly stimulated with 5 µM A$\beta$_1–42, then treated with 5 µM RB and 20 µM RB, respectively.

This part refers to the previous method [19]. 100 µL of N2a cell suspension were added in the 96-well plate (5 $\times$ 10⁴/mL) for cell viability test. Following the pretreatment for 3 h and treatment for 24 h as described in the “Cell Culture and Agent Treatment” section, 20 µL of a 5 mg/kg MTT (1334GR001, Guangzhou saiquo biotech Co., Ltd., Guangzhou, Guangdong, China) solution was added to the cells. Following incubation at 37 °C for 4 hours. The crystals were dissolved using dimethyl sulfoxide (D670381, Aladdin, Shanghai, China) and the absorbance was measured at 490 nm using microplate reader (Synergy 4, Omega Bio-tek, Inc., Norcross, GA, USA).

N2a cells were plated at 5 $\times$ 10⁴ cells/mL on glass coverslips to evaluate nuclear translocation of NF-$\kappa$B following RB treatment. After completing the pre-treatment and drug administration protocols described in section 2.6, culture medium was aspirated, and cells were washed with phosphate buffer. Sample preparation followed the manufacturer’s instructions (SN368, Beyotime, Beijing, China). The slides were incubated with NF-$\kappa$B (A2547, 1:300, ABclonal) antibodies, followed by secondary antibody (AS058, 1:500, ABclonal), and ultimately with DAPI (D1306, Thermofisher, Waltham, MA, USA). Confocal imaging was performed using a Zeiss LSM710 laser scanning microscope (Shanghai, China). For all antibodies’ information see Supplementary Material-antibodies.

Brain sample were collected to homogenize (50 Hz, 60 s, twice) and centrifuge (13,000 rpm, 5 min, twice) to extract protein solution. After treatment was finished as shown in 2.6, N2a cells were collected to lyse (20 min) and centrifuge (13,000 rpm, 5 min, twice) to extract protein sample. The whole process of protein extraction was done under 4 °C. The solution was qualified using BCA kit (P0012S, beyotime, Shanghai, China). Proteins (30–40 µg) were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to PVDF membranes. Primary antibodies anti-IL-1$\beta$ (A16288, 1:1000, ABclonal), anti-IL-4 (A14660, 1:1000, ABclonal), anti-IL-6 (A0286, 1:1000, ABclonal), anti-TNF-$\alpha$ (A11534, 1:1000, ABclonal), anti-p-inhibitor of kappa B kinase (IKK) (AP0546, 1:1000, ABclonal), anti-IKK (A2062, 1:3000, ABclonal), anti-p-inhibitor of NF-$\kappa$B (I$\kappa$B) (AP0614, 1:500, ABclonal), anti-I$\kappa$B (A11397, 1:1000, ABclonal), anti-p-nuclear factor kappa-B (NF-$\kappa$B) (AP0123, 1:400, ABclonal), anti-NF-$\kappa$B (A2547, 1:1000, ABclonal), anti-p-Tau (bs-2392R, 1:2000, Bioss, Beijing, China) and anti-GAPDH (A19056, 1:5000, ABclonal) were used to incubate membranes at 4 °C overnight. The membranes were subsequently washed and incubated with goat anti-rabbit immunoglobulin G (IgG) (AS058, 1:10,000, ABclonal) at 4 °C for 4 hours. After incubation, membranes were washed again and analyzed by gel imager (GenoSens 2150, Beijing, China). For all antibodies’ information see Supplementary Material-antibodies.

Semi-quantitative analysis of immunohistochemical staining and western blot results was performed using ImageJ software (1.8.0.345, Bethesda, MD, USA). GraphPad Prism 9 (GraphPad Software, Inc., San Diego, CA, USA) was utilized to generate bar charts. Statistical comparisons among WT, APP/PS1, and RB-treated APP/PS1 groups were conducted with BONC DSS Statistics 25 software (Business-intelligence of Oriental Nations Co., Ltd., Beijing, China), presenting data as mean $\pm$ (standard error of the mean) SEM. The normality of the behavioral data was assessed using the Shapiro-Wilk test, and non-parametric statistical analysis was conducted via the Kruskal-Wallis H test due to potential deviations from normal distribution. For the comparison of multiple groups in the other data, one-way ANOVA was performed, followed by Least Significant Difference (LSD) post hoc analysis. Statistical significance was determined at p 0.05. For original data see Supplementary Material-Original data.

MWM is widely utilized to evaluate spatial memory and learning capacities in rodent models [20]. Representative swimming trajectories on day 6 are presented in Fig. 1A. During the initial navigation session, no significant differences in platform localization time were observed among WT, APP/PS1, and APP/PS1 + RB groups (Fig. 1B). Following repeated training, RB-treated APP/PS1 mice demonstrated reduced search times compared to untreated counterparts (Fig. 1B). Notably, APP/PS1 mice displayed more disorganized swimming patterns compared to RB-treated animals. Statistical analysis revealed that RB administration significantly decreased escape latency during formal navigation testing in APP/PS1 mice (p 0.01) (Fig. 1C). Probe trial results showed distinct trajectory patterns (Fig. 1D), with APP/PS1 mice spending significantly less time in the target quadrant compared to WT controls (p 0.01) (Fig. 1E). Additionally, platform crossings were reduced in APP/PS1 mice (p 0.001) (Fig. 1F). Significantly, RB intervention enhanced both target area occupancy and platform crossing frequency in APP/PS1 mice (p 0.05, Fig. 1E,F). Passive avoidance testing (step-down/step-through) evaluated memory function following RB administration. Step-down test results showed APP/PS1 mice had shorter latency to step down compared to WT controls (p 0.001, Fig. 1G). Additionally, APP/PS1 mice made more jump-down attempts within 5 minutes despite repeated shocks (p 0.001, Fig. 1H). RB treatment reversed these impairments in APP/PS1 mice, increasing step-down latency (p 0.001, Fig. 1G) and reducing jump-down frequency (p 0.01, Fig. 1H). Step-through test results revealed RB-treated mice had significantly longer latency to enter the dark compartment compared to untreated APP/PS1 mice (p 0.001, Fig. 1I). NOR test results showed APP/PS1 mice had impaired non-spatial recognition memory, reflected by reduced recognition index, which was improved by RB treatment (p 0.05, Fig. 1J). These data indicate RB improves both spatial and non-spatial memory in APP/PS1 mice.

Fig. 1.

RB ameliorated memory loss and learning ability of APP/PS1 mice. (A) Represent trajectories on day 6 in the navigation test (n = 10). The escape latency (B) from day 1 to day 6 and (C) on day 6 in the navigation part of MWM test (n = 10). (D) Represent trajectories of mice in the probe trial (n = 10). RB increased (E) time that mice spent in the platform area and (F) numbers of crossing the platform area (n = 10). (G,H) RB improved the memory performance of APP/PS1 mice in the step-down passive avoidance test (n = 10). (I) RB extended the step-through latency of APP/PS1 mice in the step-through passive avoidance test (n = 10). (J) RB enhanced the recognition index of APP/PS1 mice in the NOR test (n = 10). ^#⁢#p 0.01, ^#⁢#⁢#p 0.001 compared with WT mice; *p 0.05, **p 0.01, and ***p 0.001 compared with APP/PS1 mice. RB, rubiadin; APP/PS1, Mo/HuAPP695swe/PS1-dE9; MWM, Morris water maze; NOR, novel object recognition; WT, wild-type.

To investigate the pathological effects of RB treatment in APP/PS1 mice, immunohistochemical staining was performed to assess A$\beta$ plaque distribution in cortical and hippocampal areas. Fig. 2 demonstrates that WT mice had almost no A$\beta$ plaques, whereas APP/PS1 mice exhibited substantial A$\beta$ accumulation in these regions (p 0.01). Strikingly, RB-treated APP/PS1 mice showed significant reductions in A$\beta$ plaque formation in both the cortex and hippocampus (p 0.05, Fig. 2A–C).

Fig. 2.

RB inhibited deposition of A$\beta$ plaques in the brains of APP/PS1 mice. Example images showing A$\beta$ deposition in the (A) cortex and (C) hippocampus of APP/PS1 mice (n = 3). Scale bar is equal to 50 µm for 40$\times$ and 200$\times$ magnification, respectively. The arrows indicate the A$\beta$ plaques in the cortex and hippocampus. Semi-quantitative analysis of A$\beta$ deposition in the (B) cortex and (C) hippocampus of APP/PS1 mice, expressed as the fold change relative to the WT group (n = 3). Results were presented as means $\pm$ SEM. ^#⁢#p 0.01 and ^#⁢#⁢#p 0.001 compared with WT mice; *p 0.05 and **p 0.01 compared with APP/PS1 mice. SEM, standard error of the mean.

Subsequently, western blot analysis was performed to assess inflammatory markers in APP/PS1 mouse brains. Compared to WT mice, APP/PS1 mice showed increased pro-inflammatory cytokines (IL-1$\beta$, IL-6, TNF-$\alpha$) and decreased anti-inflammatory cytokine IL-4 (p 0.001, Fig. 3A–E). RB treatment significantly reversed these alterations, reducing pro-inflammatory markers and increasing IL-4 levels (p 0.05, Fig. 3A–E), confirming its anti-inflammatory properties. Phosphorylation of NF-$\kappa$B signaling components was also evaluated. RB treatment significantly suppressed phosphorylation of IKK, I$\kappa$B, and NF-$\kappa$B in APP/PS1 mice compared to untreated controls (p 0.001, Fig. 3F–I), further supporting its role in mitigating neuroinflammation. For all original WB figures in Fig. 3A,F, see the Supplementary Material-original images of WB.

Fig. 3.

RB suppressed neuroinflammation in the brains of APP/PS1 mice. (A) RB blocked the production of IL-1$\beta$, IL-6, and TNF-$\alpha$ and promoted that of IL-4 in the brains of APP/PS1 mice (n = 3). Semi-quantitative analysis of (B) IL-1$\beta$, (C) IL-4, (D) IL-6, and (E) TNF-$\alpha$ which was normalized to GAPDH and presented as the fold of the WT group (n = 3). (F) RB blunted the activation of IKK/I$\kappa$B/NF-$\kappa$B in the brains of APP/PS1 mice (n = 3). Semi-quantitative analysis of (G) p-IKK, (H) p-I$\kappa$B, and (I) p-NF-$\kappa$B which was normalized to their total proteins and presented as the fold of the WT group (n = 3). Results were presented as means $\pm$ SEM. ^#p 0.05 and ^#⁢#⁢#p 0.001 compared with WT mice; *p 0.05 and ***p 0.001 compared with APP/PS1 mice. IKK, inhibitor of kappa B kinase; I$\kappa$B, inhibitor of NF-$\kappa$B; NF-$\kappa$B, p-nuclear factor kappa-B.

To further validate RB’s neuroprotective effects in vitro, N2a cells were challenged with A$\beta$_1–42. RB treatment significantly improved cell viability in A$\beta$_1–42-exposed N2a cells (p 0.001, Fig. 4A), indicating its cytoprotective potential. Compared to control cells, A$\beta$_1–42 exposure induced marked increases in pro-inflammatory cytokines (IL-1$\beta$, IL-6, TNF-$\alpha$) and decreases in anti-inflammatory IL-4 (p 0.01, Fig. 4B–F), reflecting inflammatory activation. For all original WB figures in Fig. 4B, see the Supplementary Material-original images of WB. Concurrently, A$\beta$_1–42 treatment significantly upregulated phosphorylated Tau protein expression in N2a cells (p 0.001, Fig. 4G). RB administration effectively restored cytokine balance and reduced p-Tau levels in A$\beta$_1–42-treated cells (p 0.01, Fig. 4B–G), demonstrating its inhibitory effects on inflammatory responses.

Fig. 4.

RB protected N2a cells against A$\beta$_1–42-induced cell damage and inflammation. (A) RB improved cell viability of A$\beta$_1–42-induced N2a cells (n = 8). (B) RB suppressed the production of IL-1$\beta$, IL-6, TNF-$\alpha$ and p-Tau and increased that of IL-4 in the A$\beta$_1–42-exposed N2a cells (n = 3). Semi-quantitative analysis of (C) IL-1$\beta$, (D) IL-4, (E) IL-6, (F) TNF-$\alpha$ and (G) p-Tau was conducted, with expression levels normalized to GAPDH and expressed as fold change relative to the CTRL N2a group (n = 3). Results were presented as means $\pm$ SEM. ^#⁢#p 0.01 and ^#⁢#⁢#p 0.001 compared with CTRL N2a cells; **p 0.01 and ***p 0.001 compared with A$\beta$_1–42-exposed N2a cells.

To further confirm whether NF-$\kappa$B was regulated in A$\beta$_1–42-treated N2a cells, the phosphorylation of NF-$\kappa$B and its upstream IKK and I$\kappa$B was detected through western blot. Once the N2a cells were treated with 5 µM A$\beta$_1–42, the expression levels of phosphorylated IKK, I$\kappa$B, and NF-$\kappa$B were all augmented (p 0.001) (Fig. 5A–D). For all original WB figures in Fig. 5A, see the Supplementary Material-original images of WB. Both 5 µM and 20 µM RB, significantly inhibited the activation of IKK, I$\kappa$B, and NF-$\kappa$B (p 0.05) (Fig. 5A–D). The red fluorescence–labeled NF-$\kappa$B and the blue fluorescence–labeled nucleus overlapped much more in the A$\beta$_1–42-exposed N2a cells than in the CTRL N2a cells, a phenomenon that was finally reversed by RB treatment (Fig. 5E). Consequently, these findings indicate that RB suppresses inflammation in A$\beta$_1–42-treated N2a cells, which may partially involve the IKK/I$\kappa$B/NF-$\kappa$B pathway (Fig. 5F).

Fig. 5.

RB inhibited the activation of IKK/I$\kappa$B/NF-$\kappa$B pathway in the A$\beta$_1–42-induced N2a cells. (A) RB down-regulated the phosphorylation of IKK, I$\kappa$B, and NF-$\kappa$B in the A$\beta$_1–42-treated N2a cells (n = 3). Semi-quantitative analysis of (B) p-IKK, (C) p-I$\kappa$B, and (D) p-NF-$\kappa$B which was normalized to their total proteins and presented as the fold of the control (CTRL) N2a group (n = 3). (E) RB prevented the nuclear translocation of NF-$\kappa$B in N2a cells treated with A$\beta$_1–42 (n = 3). (F) The quantitative results of the nuclear fluorescence intensity of NF-$\kappa$B were presented (n = 3). Scale bar is equal to 20 µm for 400$\times$ magnification. Results were presented as means $\pm$ SEM. ^#⁢#p 0.01 and ^#⁢#⁢#p 0.001 compared with CTRL N2a cells; *p 0.05, **p 0.01 and ***p 0.001 compared with A$\beta$_1–42-exposed N2a cells.