Dendrobium alkaloids prevent Aβ_25–35-induced neuronal and synaptic loss via promoting neurotrophic factors expression in mice

Drugs and reagents

Dendrobium was collected from Dendrobium planting regions of Xintian Traditional Chinese Medicine Industry Development co., LTD of Guizhou Province in 2014. The dried stems of the herb (10 kg) were extracted by 95% ethanol solution. DNLA was isolated from the extracts, and analyzed by LC-MS/MS. Alkaloids accounted for 79.8% of DNLA, and mainly contained Dendrobine (C₁₆H₂₅O₂N, 92.6%), Dendrobine-N-oxide (C₁₆H₂₅O₃N, 3.3%), Nobilonine (C₁₇H₂₇O₃N, 2.0%), Dendroxine (C₁₇H₂₅O₃N, 0.9%), 6-Hydroxy-nobilonine (C₁₇H₂₇O₄N, 0.32%), and 13-Hydroxy-14-oxodendrobine (C₁₆H₂₃O₄N, 0.07%). The chemical structures of these ingredients were shown in the Fig. 1A, and the chromatograms of the sample solutions were shown in the Fig. 1B.

Aβ_25–35 (Sigma, St Louis, MO, USA) was dissolved in physiological saline to a final concentration of 2.5 g/L, and incubated for one week at 37 °C to reach a state of aggregation. CNTF antibodies (ab190985) and GDNF antibodies (ab18956) were obtained from Abcam (Cambridge, England). BDNF (sc-546) was purchased from Santa Cruz Biotechnologies (Santa Cruz, CA, USA). TUNEL kits (In Situ Cell Death Detection Kit and POD) were purchased from Roche (Switzerland).

Animals

Male Kunming mice (20–25 g) were purchased from the Laboratory Animal Center, Chongqing, China (Grade: specific pathogen-free [SPF], Certificate no.: SCXK 2012-0005). Mice were housed in SPF-grade animal facilities (Certificate no.: SYXK 2014-003), with 22–23 °C, and a 12-hour light/dark cycle. Mice were given food and water freely. All animal procedures were approved by the animal experimental ethical committee of Zunyi Medical College.

Experimental designs

The animals were randomly divided into three groups (n = 6): sham surgery, model, and DNLA (40 mg/kg) treatment group. Mice were anesthetized with 7% chloral hydrate (35–45 mg/kg, i.p.) and fixed on a stereotaxic instrument (RWD Life Science, Guangdong, China). Then Aβ_25–35 (2.5 µg/µL, 2 µL) was injected into both lateral cerebral ventricle in the model and DNLA groups via a 5-µl microinjector. The injection site was posterior from the bregma (AP) = −0.4, mediolateral from the midline (MR) = 1.2, and dorsoventral from the skull (DV) = 2.7. Mice in the sham group were treated the same as the model group except for the injection of Aβ_25–35 instead of sterile normal saline (Nie et al., 2010). Mice in the DNLA group were administered intragastrically with DNLA (40 mg/kg) on the first day after the injection for 19 consecutive days, while mice in the sham and model groups were administered with distilled water.

Morris water maze test

The ability of spatial learning and memory of the mice was evaluated through the Morris Water Maze (MWM) test (Edwards et al., 2014; Milner et al., 2014). The test was performed on the 14th day after drug administration. The pool of MWM was 120 cm in diameter, filled with water (22–25 °C) and divided into four quadrants. A hidden platform was placed in the center of the target quadrant 1.5–2.0 cm under the surface of the water. MWM included the place navigation test and the spatial probe test. In the place navigation test, all mice were trained three times per day for four days. Mice were initially placed in the three quadrants which did not have a platform. Each trail lasted for 60 s or ended as soon as the mouse climbed on the platform. The time was recorded as the escape latency, if the mouse climbed on the platform within 60 s. If the mouse failed to find the platform within 60 s, its escape latency was recorded as 60 s. The spatial probe test was performed following the end of the place navigation test. In this test, the platform was removed, and each mouse was allowed to swim for 60 s. The searching distance in the target area and total area were measured. Data was recorded and analyzed by Top View Animal Behavior Analyzing System (Version 3.00).

Morphometric analysis

After the MWM test, three mice were selected randomly from each group and anesthetized with 7% chloral hydrate. Then these mice were perfused with phosphate-buffered saline (0.1 M, 4 °C) via the ascending aorta, followed by 4% paraformaldehyde until the tail and limbs were rigid. Thereafter, the brains were removed and cut in half. Half of the brain was fixed with 4% paraformaldehyde for seven days and cut into coronal sections (4-µm thick) for hematoxylin-eosin (H&E), Nissl and TUNEL staining (Yang et al., 2014; Liu et al., 2015). The other half was fixed with 2.5% glutaraldehyde for electron microscope detection.

Western blot analysis

Western blot analysis was performed as previously reported (Liu et al., 2015; Li et al., 2015). Three mice from each group were sacrificed after WMW test, and the hippocampal and cortical tissues were collected and homogenized in RIPA lysis buffer (1:5, w/v). Protein concentrations were determined by BCA protein assay kit. An aliquot of 45 µg of protein was applied for electrophoresis followed by the transferring of protein to the polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% nonfat dry milk in TBST buffer for one hour at room temperature, and incubated with the primary antibody: anti-CNTF (1:1,000), BDNF (1:1,000), and GDNF (1:1,000) at 4 °C overnight. Next, HRP-labeled goat anti-rabbit IgG (Beyotime Biotechnology, Jiangsu, China; A0208, 1:2,000) was incubated with the membrane at room temperature for one hour. The blots were visualized using the enhanced ECL Western blot detection kit (7Sea Biotech, Shanghai, China) and scanned to Gel Imaging. The band intensity was quantified using Quantity One 1-D analysis software v4.52 (BioRad, Hercules, California, USA).

Statistical analysis

All data were presented as mean ± standard error. Data were analyzed by SPSS 13.0 statistics software by one-way ANOVA or student’s t-test. P < 0.05 was considered as statistically significant.

Protective effects of DNLA on learning and memory deficits induced by Aβ_25–35

Compared with the sham group, the escape latency of the model group was significantly prolonged in the navigation test, and the explored distance in the target area was decreased in the spatial probe test. After treatment for 14 days, the escape latency of the DNLA group was decreased and the explored distance in the target area was increased as compared with the model group. The results indicated that DNLA could protect the mice against learning and memory deficits which were induced by Aβ_25–35 injection (Fig. 2).

Morphological changes of cortex and hippocampus tissues

H&E staining was performed to observe the morphological changes of cortex and hippocampus tissues in mice brain. The results showed that the morphology of brain tissues in the sham group was normal, and the cell density was higher. However, loose structures in hippocampal and cortical neurons with disordered hippocampal pyramidal cell layers occurred in the model group. Furthermore, the images showed that not only the number of neurons was reduced, but also that neurons staining were abnormal with nuclear condensation in the model group. DNLA treatment significantly decreased the number of abnormally stained neurons. Also, the morphology of brain tissues was generally normal in the DNLA group (Fig. 3).

Effect of DNLA on apoptosis in the cortex and hippocampus

The effect of DNLA on neuronal apoptosis induced by Aβ_25–35 was observed using the TUNEL method. The results revealed that the number of apoptotic cells in the hippocampus and cortex were not affected in the sham group, but were increased in the model group. DNLA treatment decreased the number of apoptotic cells (Fig. 4).

Changes in Nissl bodies in cortex and hippocampus tissues

As a neural characteristic structure, the number of Nissl bodies reflects the state of neurons. In physiological conditions, the Nissl bodies were big and abundant, showing that the function of neuronal protein synthesis was strong; on the other hand, when nerve cells were damaged, the number of Nissl bodies will be reduced or even disappear. Intraneural Nissl bodies of the cortex and hippocampus were lightly stained and appeared to be sparsely arranged in the model group. However, deeper stained Nissl bodies with higher density in cortex and hippocampus neurons were found in the sham and DNLA groups (Fig. 5).

Effects of DNLA on the ultrastructures of neurons in Aβ_25–35-injected mice

The ultrastructures of neurons were observed under electron microscope. The results revealed that the nuclear morphology of mice in the sham group was normal, and the chromatin was normally distributed. In the cytoplasm, mitochondrial morphology was normal with clear and regularly-arranged cristae. Furthermore, a number of ribosomes could be observed in these regularly-arranged rough endoplasmic reticulum. However, in the model group, the ultrastructures of neurons were abnormal. It could be found that nuclear membrane had folded or had obscure boundaries, mitochondria were swelling even without the cristae, and the endoplasmic reticulum was swelling. DNLA treatment could reduce the injury. Compared with the model group, nuclear membrane was more smooth, swelling of mitochondrial was reduced, and partial rupture of cristae were observed in the DNLA group (Fig. 6). The neuronal synapses were also observed under electron microscope. Injection of Aβ_25–35 would reduce the number of synapses and synaptic vesicles, and cause the shape and size of synaptic vesicles uneven. The structure of synapses was obscured, and the structure of the post-synaptic lattice became thin. DNLA inhibited the loss of synapses significantly (P < 0.05). Also, in the DNLA group, the generally normal synaptic structures were maintained (Fig. 7).

Effects of DNLA on the protein expressions of BDNF, CNTF and GDNF in the cortex and hippocampus

The protein expressions were analyzed by Western blot. Compared with the sham group, the expressions of BDNF, CNTF and GDNF significantly decreased in the model group. DNLA group had significantly increased expressions of BDNF, CNTF and GDNF in the cortex and hippocamous (Fig. 8).