Geniposide (Zhao et al., 2016b; purity > 98%) was purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) and was free of endotoxins. 1,1,1,3,3,3-Hexafluoro-2-propanal (HFIP) and penicillin/streptomycin were obtained from Sigma (St. Louis, MO, USA). Fetal bovine serum (FBS), B27, and Neurobasal-A medium were purchased from Gibco (Waltham, MA, USA).

Experimental mPrP-APPswe/PS1dE9 (APP/PS1) doubly transgenic mice and C57BL/6 mice were purchased from Beijing HFK Bio-Technology Co., Ltd. Nine-month-old male mice were individually housed under a 12 h light/dark cycle at an ambient temperature of 23 ± 1°C and relative humidity of 55 ± 5% and were given food and water ad libitum. Before the experiment, the mice were housed under these conditions for 2–3 days to allow them to adapt to the environment. APP/PS1 mice were randomly categorized into the treatment groups or the vehicle group and treated with either geniposide (12.5, 25, or 50 mg/kg/days; n = 15) or water (n = 15), respectively, for 3 months via intragastric administration. Age-matched C57BL/6 mice were fed water as the vehicle control (n = 15). Geniposide was dissolved in water within 24 h before use. Equal volumes of liquid were given to each group daily for 3 months before the mice were sacrificed.

All animal procedures performed in this study were approved by the Beijing Normal University Laboratory Animal Care and Use Committee in accordance with the National Institute of Health “Guidelines for the Care and Use of Laboratory Animals” (NIH Publications No. 8023, revised 1996). The mice were sacrificed by cervical dislocation after being anesthetized. All efforts were made to minimize the number of animals used and their suffering.

Oligomeric Aβ_1–42 was prepared from commercially available synthetic peptides (Sigma Chemical Co., St. Louis, MO, USA), as previously described (Dahlgren et al., 2002; Yin et al., 2011). The lyophilized peptide was resuspended in cold HFIP at a concentration of 1 mg/mL and aliquoted into microcentrifuge tubes to quickly obtain 0.1 mg stocks. The stocks were stored at room temperature and protected from light for 2–4 h before the removal of HFIP under gentle vacuum, thereby leaving a thin transparent film of peptides on the internal surface of the tube. The stocks were stored at −20°C. For the aggregation protocols, HFIP-treated peptides were dissolved in anhydrous dimethyl sulfoxide at 5 mM and diluted to 100 μM in Ham's F12 Nutrient Mixture (Thermo Fisher Scientific, Waltham, MA, USA). The diluted peptides were incubated at 4°C for 24 h to obtain oligomeric Aβ_1–42.

One-day-old male C57BL/6 mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd., and transported within a specific pathogen-free, air permeable, and bacteria shield shipping box. The mice were sacrificed by cervical dislocation.

Primary hippocampal neurons were prepared from the hippocampi of 1 day-old (newborn) pups. The hippocampi were dissected in cold D-Hanks solution. Tissues were collected and washed in D-Hanks, and 0.05% (v/v) trypsin was added for digestion at 37°C for 20 min. Digestion was terminated by adding FBS to a final concentration of 10% (v/v). Cells were collected by centrifugation at 800 × g for 10 min to remove the D-Hanks solution and were then resuspended in Neurobasal-A medium (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 2% (v/v) B27 (Thermo Fisher Scientific).

For the various analyses, cells were plated onto 6-, 12-, and 96-well-plates or glass-bottom dishes with four chambers (CELLview, Greiner, Germany) (~5 × 10⁴ cells/mL) pre-coated with poly-D-lysine (10 μg/mL). The cells were cultured at 37°C and 5% CO₂ until use. The initial medium was removed after 4 h and replaced with fresh medium. After 14 days, the medium was replaced with Neurobasal-A medium without serum and phenol red (which affect the aggregation of Aβ). The primary cultured hippocampal neurons were pre-incubated for 24 h in the absence or presence of geniposide (2.5, 5, or 10 μM) before adding oligomeric Aβ_1–42 (200 nM) for 24 h to assess the protective effect of geniposide on the Aβ_1–42-treated neurons.

The brain tissue or neuron samples were lysed in 10 volumes (w/v) of radio-immunoprecipitation assay buffer containing a cocktail of complete protease and phosphatase inhibitors and were centrifuged at 15,000 × g for 10 min at 4°C. The protein concentration of the supernatant was determined using the BCA method. Proteins were examined by Western blot analysis using standard protocols. Equal amounts of proteins were loaded and resolved by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA). The membranes were incubated in blocking solution (5% skim milk in PBST, 20 mM Tris-HCl, 150 mM NaCl, 0.1% Tween-20) at room temperature for 1.5 h. The membranes were incubated and gently shaken overnight (at 4°C) in PBST containing 5% skim milk and the indicated primary antibodies from among the following: monoclonal rabbit antibodies against c-AMP response element binding protein (CREB, 1:400, Cell Signaling, Beverley, MA, USA), and GAPDH (1:3000, Cell Signaling); polyclonal rabbit antibodies against phosphorylated Ca²⁺/calmodulin dependent protein kinase II α (p-CaMKIIα, 1:8000, Santa Cruz, CA, USA); and monoclonal mouse antibodies against synaptophysin (1:500, Abcam, Cambridge, UK), CaMKIIα (1:250, Santa Cruz), phosphorylated CREB (p-CREB, 1:400, Cell Signaling), postsynaptic density protein-95 (PSD-95, 1:500, Abcam), and β-actin (1:2000, Santa Cruz). After four washes with PBST, the membranes were incubated for 1.5 h at room temperature with the corresponding secondary antibodies. The membranes were washed four times with PBST and detected with an infrared imaging system (Odyssey). The intensity of the blots was analyzed and compared using NIH ImageJ program.

ROS were measured with 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA). The cell-permeable DCFH-DA can be oxidized to dichlorofluorescein (DCF) by ROS in the cytoplasm and emit intensely fluorescence. After incubation with Aβ_1–42 in the presence and absence of geniposide for 24 h, cultured neurons were washed with PBS and incubated with 10 μM DCFH-DA for 30 min at 37°C in an incubator with 5% CO₂. Fluorescence images were captured using a laser-scanning confocal microscope (TCS-SPE, Leica, Germany). Fluorescence intensity was analyzed using NIH ImageJ program.

Hippocampal neurons at 14 day in vitro (DIV) were used for axonal mitochondrial density assay, and the length of these neurons was measured after the treatments. Neurons cultured in glass-bottom dishes with four chambers (CELLview, Greiner, Germany) were incubated with MitoTracker Red (Thermo Fisher Scientific, Waltham, MA, USA) at 100 nM for 20 min at 37°C and 5% CO₂. Neurons were fixed in 4% paraformaldehyde for 20 min after being permeabilized with 0.5% Triton X-100 in 0.1% sodium citrate and blocked with 10% goat serum. Up to 0.25 mL of anti-Tau antibody (Abcam, 1:500) was added to each chamber, followed by incubation with goat anti-rabbit secondary antibody (Alexa Fluor 488, Abcam, UK). Axons were identified based on morphological characteristics. In live images, branches that are longer and with thin and uniform diameter, as well as sparse branching, were considered to be axons. Particles with strong red fluorescence (compared with background) and clear edges colocalized with axons were considered to be mitochondria. Fluorescence images were captured using a laser-scanning confocal microscope (TCS-SPE, Leica, Germany) and analyzed using NIH ImageJ program.

MitoTracker Green (Thermo Fisher Scientific, Waltham, MA, USA) was used to label the mitochondria in living neurons. The axons and mitochondria were identified using the procedure described in Section Axonal Mitochondrial Density Assay and Length Measurement. The proximal region of axons was selected for time-lapse imaging analysis. Time-lapse images were captured under an inverted laser-scanning confocal microscope (TCS-SPE, Leica, Germany) with a stage-based chamber under 5% CO₂ and 37°C. Time-lapse image stacks composed of five images (512 × 512 pixel) were taken every 10 s for 5 min for a total of 150 images under 40× magnification. Kymographs were generated using the kymograph tool for ImageJ program under maximum intensity projection. The width of the kymographs represents the length (μm) of the imaged axon, and the height represents the recording time. A mitochondrion was considered to be stationary if it remained non-mobile for the entire recording period. A mitochondrion was considered movable only if the displacement was more than 2 μm. The percentages of stationary and movable mitochondria and of anterograde and retrograde mitochondrial movement were measured separately from the corresponding kymographs by using NIH ImageJ program. A mitochondrion that moves from the soma toward the distal end of an axon is considered anterograde, whereas movement from the distal end toward the soma is considered retrograde. The average velocity (μm/s) of all anterograde and retrograde axonal mitochondria was calculated using the displacement of each mitochondrion.

Neurons cultured in glass-bottom dishes with four chambers (CELLview, Greiner, Germany) were fixed in 4% paraformaldehyde for 20 min and washed with PBS after the neurons were incubated with 2 μM preheated CellTracker CM-DiI (Life Technologies, Waltham, MA, USA) for 20 min at 37°C. LAS X software was used to control the laser-scanning confocal microscope (TCS-SPE, Leica, Germany) equipped with a 63× objective and excitation at 543 nm. Images were taken under the confocal microscope, and Z-stacks were gathered at increments of 0.67 μm. To increase the accuracy of the identification of the surfaces of the dendrite shaft and spines, sequence images were deconvolved to reduce point-spread functions by using the adaptive blind 3D deconvolution method (AutoQuant X, Media Cybermedics, Inc., Bethesda, MD) prior to analysis. The output images were obtained with Build Neurites and Build Spines in NeuronStudio software (CNIC, Mt. Sinai School of Medicine, New York, USA) to mark and record all the spines on the selected dendrites. Dendritic protrusions with a clearly identifiable neck attached to the branch of the dendrite composed a spine. The output data included length, neck diameter, and head diameter of each spine, based on which the types of spines were recognized. Spines with a spine length-to-neck diameter ratio <2.0 were defined as of the stubby type. Spines were categorized as mushroom type if they presented a length-to-neck diameter ratio of more than 2.0 and a head-to-neck diameter ratio of more than 1.3. Spines were categorized as thin type if the spines exhibited a length-to-neck diameter ratio of more than 2.0 and a head-to-neck diameter ratio of <1.3 (Du et al., 2014). Spine density and morphology were measured separately for portions of three to five selected secondary dendrites per neuron.

Cultured hippocampal neurons were fixed in 4% paraformaldehyde for 20 min after the neurons were permeabilized with 0.5% Triton X-100 in 0.1% sodium citrate and blocked with 10% goat serum. Up to 0.25 mL each of anti-MAP2 antibody (Abcam, 1:100) and anti-synaptophysin (Abcam, 1:200) was added to each chamber, followed by incubation with goat anti-rabbit and mouse secondary antibodies (Alexa Fluor 488, Alexa Fluor 594, Abcam, UK) for 30 min at 37°C. Images were taken under a laser-scanning confocal microscope (TCS-SPE, Leica, Germany) by using a 40× objective and excitation of 488 and 543 nm, and Z-stacks were gathered at increments of 0.67 μm. The z-stack images were compressed to a single image via the max projection method and analyzed using NIH ImageJ program.

The results were processed for statistical analysis using SPSS (version 20.0 for Windows). The results are presented as the mean ± standard error of the mean (SEM). Statistical analyses were performed with one-way analysis of variance, followed by Fisher's protected least significant difference test for post-hoc comparisons. A value of P < 0.05 was considered significant.

Mitochondria are vital to the function of synapses because they supply energy for maintenance, calcium buffering, synaptic transmission, and vesicle release. Mitochondria are thought to be synthesized perinuclearly. Thus, they must be transferred from the soma to distal synapses through axons and dendrites via mitochondrial transport and constantly reconfigured to meet synaptic needs. Mitochondrial morphology is also dynamic and can be regulated through fusion and fission. An elongated morphology may confer bioenergetic advantages for ATP generation and dispersal (Skulachev, 2001). Therefore, the effects of geniposide on Aβ-induced abnormal axonal mitochondrial trafficking, distribution, and morphology were analyzed. A primary cultured hippocampal neuronal model was examined as previously reported because of the technical limitations on observing mitochondrial trafficking in vivo (Du et al., 2010). Axonal processes were selected for the quantitative analysis of mitochondrial length, density, distribution, and mobility because of their known morphologic and dynamic characteristics (Banker and Cowan, 1979; Du et al., 2010) and because of the significantly synaptic pathology of AD.

Axonal mitochondria are dynamic organelles, and their dynamics, trafficking, and docking are critical to maintain synaptic function and plasticity. The impairment of axonal mitochondrial may lead to synaptic loss and dysfunction. Various lines of evidence indicate that synaptic loss and deactivation are the biological bases of AD, and the accumulation of Aβ is an early event associated with synaptic and mitochondrial damage in AD. Thus, the density of synapses, the morphology of dendritic spines, and the levels of synapse-related proteins were analyzed to assess the contributions of abnormal mitochondrial fragmentation and trafficking to synaptic loss and dysfunction.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.