Liver X receptor-β improves autism symptoms via downregulation of β-amyloid expression in cortical neurons

Animal experiments were conducted in strict accordance with the approved animal protocols and guidelines established by Medicine Ethics Review Committee for animal experiments of Linyi People’s Hospital.

A total of 32 adult BTBR mice aged 8–10 week and 16 B6 mice (C57BL/6) (BTBR mice were mice model with autistic behavior; B6 mice were normal healthy mice, purchased from Beijing Experimental Animal Center, China), which were 50 % male and weighted 20–25 g, were divided into: (1) the experimental group: 16 BTBR mice fed with agonist T0901317 (20 mg/kg/day, purchased form Cayman Company) which dissolved in dimethyl sulfoxide (DMSO); (2) the control group: 16 BTBR mice fed with DMSO of the same volume sans T0901317; and (3) the normal control group: 16 B6 mice fed with DMSO with the same volume sans T0901317. After seven consecutive days of gavage followed by fasting overnight, 8 mice in each group were randomly selected and executed the next day between 8:00 am to 4:00 pm. After intraperitoneal injection of 10 % water and chloral (0.4 g/kg, the lower left i.p.) anesthesia, the mice were then decapitated, and brains were harvested into liquid nitrogen and stored at −70 °C for measurement of Aβ. The remaining mice were placed into the conduct laboratory for one hour of acclimatization, and then used for autistic behavioral test.

About 80 mg brain tissue was weighed and homogenates were prepared according to the methods by Howland DS et al. The Aβ radioimmunoassay kit (Bo Yan Biological Technology Co., Ltd. Shanghai) was used to determine the content of Aβ in the brain tissue samples. A 100 μL of homogenate was diluted by 5-fold with 1 mol/L Tris–HCl (PH 8.0) and mixed well, from which a 100 μL sample was taken for measurement. The standard product is diluted with PBS buffer to various concentrations and added antiserum and ¹²⁵ I-Aβ protein, mixed well, stayed at 4 °C for 24 h and added separating agent at room temperature for 20 min, then centrifuged at 4 °C 3500 rpm for 20 min with the supernatant discarded. The cpm value was measured using γ counter and plotted for standard curves. The same protocol was then applied on the sample to calculate the positivity rates, which was used to determine the content of the Aβ in picograms per milliliter of sample volume (pg/ml) according to the standard curve.

Brain tissue protein of LXRβ agonist T0901317 treated mice was extracted and tested for the changes in Aβ precursors APP, the Aβ-degrading enzymes (neutral endopeptidase, NEP, insulin-degrading enzyme, IDE), the β-secretase enzyme (BACE1) that cut the APP to generate Aβ, and autophagy-related protein (LC3-I, LC3-II, autophagy substrate protein P62, etc.), and to test the phosphorylated Ras/Raf/ERK1/2 protein expression. Western blots were performed according to the method described previously [18, 19]. The prepared protein sample was separated by SDS-PAGE electrophoresis (for BACE1, LC3-I, LC3-II, P62, we used 15 % gel; for APP, NEP, IDE, we used 8 % gel; for Ras, Raf, MEK1/2, ERK1/2 we used 10 % gel). The test was run with a sample concentration of 20–60 μg/L and under 80-120 V, the electrophoresis terminated at the first appearance of the bromophenol blue. The gel was removed, and placed in the order of filter, nitric acid cellulose membrane, gel, filter from bottom to top, and expelled air bubbles in between, and placed in electrophoretic transfer membrane device at 100 V for 1.5 h of transmembrane. Then, the nitrocellulose filter was removed and washed with PBST, added 5 % skim milk solution, shaken at room temperature and blocked for 2 h. The primary antibody was diluted with 5 % skim milk or 5 % BSA to the appropriate concentration (Anti-β-actin (Santa cruz, sc-47778, 1: 400); anti- IDE (Abeam, ab32216, l: 5000); anti-NEP (R & D, AF1126, 0.2 μg/ml); anti-APP (Sigma, A8717, 1: 20000); anti-BACEl (Abeam, ab2077, l: 1000); phospho-Mekl/2, phospho-C-Raf, Ras, phospho-ERKl/2 (Cell Signaling Technology, USA)). The membrane was washed with PBST for 10 min × 4 times and the secondary antibody was diluted in blocking buffer (1: 4000) and shaken at room temperature for 2 h, following, the membrane was washed with PBST, 10 min × 4 times, colored by chemiluminescence (ECL). Then the molecular weight of the target band and net optical density value of the sample was quantitatively analyzed using gel image processing system Image J (National Institutes of Health, Bethesda, MD, USA), and standardized with β-actin concentrations.

Stereotyped behavior was assessed using blinded evaluation using 60 min video recording of mouse activity. Mice were placed into a round Plexiglas cylinder (12 inch tall and 8 inch in diameter) individually, and all the mice were admitted to explore freely. Mice ID, status, and body weight were recorded, and changes of behaviors including rotation, rears, sniffing, grooming and exploratory activity were observed. Two raters watched these recording and independently rated stereotyped behavior based on the animal stereotyped behavior score table and data summary table mentioned by Kelley AE [20].

Experiment mice were transported to the laboratory around one hour in advance to adapt to the experimental environment. Experiment staff recorded ID, date and status of the mice, and cleaned experimental device with 75 % ethanol and water, and dried with paper towels, then released the mice (back facing the experimenter) from the cage into the center of the OF device, and used video recording system to record the activity of mice in the OF for a total of 15 min. The recorded indicators include total distance (total activity), average speed, rest time, activity time, the number of activities, regional distribution index (edges, corners, surrounding, center, grooming, grooming frequency and time).

The experiment was divided into two parts. (1) Navigation test: experiment mice were allowed to free swimming for 2 min during the first day to adapt to the environment. Starting from the second day, the water maze was artificially divided into four quadrants. The platform was placed in a random quadrant and standing 1 cm above the water level and decorated with bright-color flags in order to attract the attention of mice. Mice were placed into the four quadrants facing the wall. Computer was used to monitor and record the route, distance and time the mice used to find and climb the platform from the fourth day. For mice that failed to find the platform in 2 min, they were drawn to the platform, and the incubation period was recorded as 2 min. (2) Space search experiment: the platform was removed in the fifth day, mice were placed into the water at a random point facing the wall, computer recorded the times of swimming in each quadrant and number of crossing the platform within 90s, and searched platform trajectory.

We used SPSS 21.0 statistical software (SPSS Inc., Chicago, IL) for statistical analyses. Data were expressed as mean ± standard deviation (\( \overline{\mathrm{x}} \) ± s). The one-way analysis of variance (ANOVA) was used to compare among multiple groups, t-test was used to compare between two groups. P < 0.05 was considered statistically significant.

Figure 1 and Table 1 show that, compared with the control group not treated with T0901317 (6.51 ± 0.73 pg/mg), the normal control group (5.37 ± 0.59 pg/mg) had significantly lower Aβ expressions (P = 0.004), suggesting Aβ is higher in the brain of autistic mice than normal B6 mice. After LXRβ agonist T0901317 processing, Aβ in brain tissue of mice in the experimental group (5.66 ± 0.60 pg/mg) was significantly lower compared with the control group not treated with T0901317 (6.51 ± 0.73 pg/mg).

Figure 2 shows that, compared with the normal control group, the BTBR mice treated with the same volume of DMSO sans T0901317 in control group had significantly increased APP and BACE1 proteins, and significantly decreased Aβ degradation enzyme (NEP, IDE) protein levels (all P < 0.05). Compared with the control group, the autistic mice treated with T0901317 in experimental group had no significant difference in APP levels (P > 0.05), but increased Aβ degrading enzyme (NEP, IDE) levels (both P < 0.05), and lower BACE1 protein levels (P < 0.05).

Figure 3 shows that, compared with the normal control group, the control group not treated with T0901317 showed significantly decreased expression of P62 (P < 0.05), and increased LC3-I, LC3-II levels (all P < 0.05). These indicate that in the absence of T0901317 processing, the autophagy level was significantly lower in BTBR than the normal B6 mice. After T0901317 treatment, the brain of autistic mice had significantly higher P62 protein than the control group (P < 0.05), and significantly lower LC3-I and LC3-II levels than the control group (all P < 0.05).

Compared with the normal control group, control group without T0901317 treatment had significantly increased levels of related proteins (P < 0.05). Compared with the control group, BTBR mouse treated with agonist showed significantly weakened Ras, PA- Raf, PB-Raf, PC-Raf, C-Raf, P-Mekl/2, P-Erkl/2 bars, and reduced average expression (all P < 0.05), as shown in Fig. 4.

Autistic mice showed sign of repeated combing, sustained grasping, licking, biting and other stereotypic movements. In order to observe the behaviors of mice with autism, we observed the activities of autistic mice within one hour via videos. The results are shown in Table 2. Compared with the normal control group, control group without T0901317 treatment showed significantly higher stereotyped behavior (rotation, rears, sniffing, grooming and exploratory activity) (all P < 0.05), suggesting that when not treated with T0901317, BTBR mice showed more severe repeated stereotyped behavior than normal B6 mice. After treatment of LXRβ agonist T0901317, repeated stereotyped behaviors was significantly improved in autistic mice; significant differences were found in comparisons of stereotyped behaviors between the experimental group and the control group (all P < 0.05), while no such difference was found between the experimental group and the normal control group (all P > 0.05).

OP test was used to examine spontaneous activity and anxiety. Autism mice exhibited significant behavioral disorders and narrowed interests. Compared with the normal control group, mice in control group without T0901317 treatment showed significantly increased combing time, and decreased percentage of time in central stay and wall standing (all P < 0.05). Compared with control group without T0901317 treatment, mice treated with LXRβ agonist T0901317 showed significantly reduced inactivity time and self-combing time, and significantly increased percentage of time in central stay and wall standing (all P < 0.05). These results show that BTBR mice treated with T0901317 showed less autism symptoms: improvements in spontaneous activities, narrowed interests, stereotyped repetitive movements, lack of aversion and other behaviors (all P < 0.05) (Fig. 5).

In the navigation test, as shown in Fig. 6, normal B6 mice (C) used more straight path and relatively shorter time to reach the platform, while the control group (A) used more tortuous path and used longer time as compared with the normal control group (P < 0.05). BTBR mice treated with T0901317 (B) used a path similar with that of the control group (C), and their difference was not statistically significant (P > 0.05).

The space search results (Table 3) showed that the T0901317 treated mice stayed longer in the platform quadrant and had higher number of platform crossing during the of 90 s space search experiment as compared with the control group (all P < 0.05), but similar to that of the normal control group (all P > 0.05). Figure 7 shows the space search paths for all groups, which suggests that autism mice treated with T0901317 (B) stayed mostly in the inner quadrant platform and crossed the platform more frequently, which was similar to that of the normal control group (C); the autistic mice (A) showed more scattered paths and fewer platform crossing.