Bone marrow-derived mesenchymal stem cells improve cognitive impairment in an Alzheimer’s disease model by increasing the expression of microRNA-146a in hippocampus

Alzheimer’s disease (AD) is characterized by the accumulation of amyloid-β and tau. We previously reported that administration of bone marrow mesenchymal stem cells (BM-MSCs) ameliorates diabetes-induced cognitive impairment by transferring exosomes derived from these cells into astrocytes. Here, we show that intracerebroventricularly injected BM-MSCs improve cognitive impairment in AD model mice by ameliorating astrocytic inflammation as well as synaptogenesis. Although AD model mice showed an increase in NF-κB in the hippocampus, BM-MSC-treated AD model mice did not show this increase but showed an increase in levels of microRNA (miR)-146a in the hippocampus. Intracerebroventricularly injected BM-MSCs were attached to the choroid plexus in the lateral ventricle, and thus, BM-MSCs may secrete exosomes into the cerebrospinal fluid. In vitro experiments showed that exosomal miR-146a secreted from BM-MSCs was taken up into astrocytes, and an increased level of miR-146a and a decreased level of NF-κB were observed in astrocytes. Astrocytes are key cells for the formation of synapses, and thus, restoration of astrocytic function may have led to synaptogenesis and correction of cognitive impairment. The present study indicates that exosomal transfer of miR-146a is involved in the correction of cognitive impairment in AD model mice.

T a r g e t q u a d r a n t O th e r q u a d r a n ts T a r g e t q u a d r a n t O th e r q u a d r a n ts T a r g e t q u a d r a n t O th e r q u a d r a n ts T a r g e t q u a d r a n t O th e r q u a d r a n ts T a r g e t q u a d r a n t O th e r q u a d r a n ts        Fig. 1c). After intravenous injection four times at 2-week intervals, the MWM test was conducted.

Electron microscopic observations.
Electron microscopic observations were done as described previously 1) . In brief, the left hemispheres were cut into 100-µm thick sagittal sections and fixed with 2.5% glutaraldehyde (WAKO, Osaka, Japan). Then, sections were post-fixed with 2% osmium tetroxide and embedded in an epoxy resin (EPON812, TAAB Laboratories Equipment, Berks, UK). Sections were cut at 70 nm thick and stained with uranyl acetate and lead citrate. Then, the subiculum area was observed with a transmission electron microscope (H7650, HITACHI, Tokyo, Japan) at 30,000× magnification. The synapses were identified by the presence of a post-synaptic density. The average of the number of synapses that were counted in 18 randomly selected fields for each mouse was analyzed.

Rat cytokine antibody array.
BM-MSCs were cultured until 90-100% confluent in 15-cm dishes in alpha-minimal essential medium with 15% FBS and 1% PS. Then, the conventional medium was changed to FBS-free alpha-minimal essential medium with 1% PS.
After 24 h, the media were collected and centrifuged at 2600 ×g for 10 min at 4°C. The supernatant was loaded onto a Rat Cytokine Antibody Array (RayBiotech Life, Norcross, GA, USA), according to the manufacturer's protocol. Digital images were obtained with the Las-3000 imaging system (Fujifilm, Tokyo, Japan). miRNA and mRNA isolation from the hippocampus and quantitation.
The mirVana miRNA isolation kit was used to extract miRNA from the frozen right hippocampus. The TaqMan MicroRNA Assay protocol was used to synthesize cDNA from targeted miRNA. The primers that targeted miR-146a, miR-133b, and snoRNA 135 are listed in Supplementary Table 3

. Applied Biosystems 7500 and TaqMan Universal Master
Mix II were used for real-time PCR. Using snoRNA 135 as an endogenous control, relative expression of miR-146a as well as miR-133b was calculated with the 2 −ΔΔCt comparative method.
Subicular mRNA from APP/PS1 and WT mice that were injected with BM-MSCs or vehicle was isolated from 4% paraformaldehyde-fixed 100-µm thick sections. The subiculum area was cut out from the sections with a scalpel, and mRNA was isolated using the RecoverAll Total Nucleic Acid Isolation Kit (Thermo Fisher Scientific). Then, mRNA was converted into cDNA using the Sensiscript RT Kit (QIAGEN). Real-time PCR was performed using SYBR green and Applied Biosystems 7500. Using GAPDH as an endogenous control, relative expression of mRNA was calculated with the 2 −ΔΔCt comparative method. The primers for IRAK1, TRAF6, NF-κB, RhoA, and GAPDH are listed in Supplementary Table 4.
Isolation and culture of primary rat astrocytes.
Hippocampal tissues of neonatal Sprague Dawley rats (1 or 2 days old) were isolated after inhalation of excess isoflurane, as described previously 2) . Briefly, cells of hippocampal tissues were cultured in DMEM/F-12 with 10% FBS and 1% PS in poly-Llysine-coated dishes. After cells of hippocampal tissues were seeded at a density of 1.5 × 10 5 cells/cm 2 , the medium was changed twice weekly. At 7-8 days after seeding, the culture dishes were shaken at 200 rpm overnight to remove unattached cells. Then, attached cells were trypsinized and seeded at 5.0 × 10 4 cells/cm 2 in poly-L-lysine-coated 24-well plates or at 1.2 × 10 4 cells/cm 2 in poly-L-lysine-coated four-well chambers. More than 95% of cultured cells were confirmed to be GFAP-positive astrocytes.

CM of BM-MSCs or that of miR-146a-transfected BM-MSCs.
BM-MSCs of passage three were used for culturing. BM-MSCs were seeded at a density of 1.5 × 10 4 cells/cm 2 in two 15-cm dishes. BM-MSCs were cultured in alpha-minimal essential medium with 15% FBS and 1% PS. When BM-MSCs reached 90% confluence, one dish was transfected with 10 nM miR-146a using Hiperfect transfection reagent, and the other dish was not transfected. At 4 h after transfection, each dish was washed with PBS, and the conventional medium was replaced with 10 ml FBS-free alpha-minimal essential medium. After 24 h, the medium was collected, and exosomes were isolated. ExoQuick-TC (System Biosciences) was used to isolate exosomes, according to the manufacturer's protocol. After isolation of exosomes, the residual supernatant was used for the analysis of free-floating miRNA.
Using 2 nM exogenous synthetic cel-miR-39 as an external control, the mirVana PARIS Kit (Thermo Fisher Scientific) was used for the extraction of miRNA. For reverse transcription, the TaqMan Advanced miRNA cDNA Synthesis kit (Thermo Fisher Scientific) was used. Real-time PCR was performed using the TaqMan Fast Advanced Master Mix (Thermo Fisher Scientific) and Applied Biosystems 7500. The primers that targeted miR-146a and cel-miR-39 are listed in Supplementary Table 3. Using cel-miR-39 as an external control, relative expression of miR-146a was calculated with the 2 −ΔΔCt comparative method.

Adding miR-146a mimic to astrocytes.
At 24 h after astrocytes were seeded at 5.0 × 10 4 cells/cm 2 in poly-L-lysine-coated 24well plates, 10 nM miR-146a mimic without Hiperfect transfection reagent was added to the medium to examine whether free-floating miR-146a affected the expression of miR-146a in astrocytes. At 24 h after adding the miR-146a mimic, astrocytes were collected by trypsinization. The miRNA of astrocytes was extracted with a mirVana miRNA isolation kit. cDNA synthesis and real-time PCR were performed as described above. The primers for miRNA are listed in Supplementary Table 3. Using snoRNA135 as an endogenous control, relative expression of miRNA-146a was calculated with the 2 −ΔΔCt comparative method.