Intracerebral Transplantation of Adipose-Derived Mesenchymal Stem Cells Alternatively Activates Microglia and Ameliorates Neuropathological Deficits in Alzheimer's Disease Mice

Abstract

Recent studies suggest that transplantation of mesenchymal stem cells might have therapeutic effects in preventing pathogenesis of several neurodegenerative disorders. Adipose-derived mesenchymal stem cells (ADSCs) are a promising new cell source for regenerative therapy. However, whether transplantation of ADSCs could actually ameliorate the neuropathological deficits in Alzheimer's disease (AD) and the mechanisms involved has not yet been established. Here, we evaluated the therapeutic effects of intracerebral ADSC transplantation on AD pathology and spatial learning/memory of APP/PS1 double transgenic AD model mice. Results showed that ADSC transplantation dramatically reduced β-amyloid (Aβ) peptide deposition and significantly restored the learning/memory function in APP/PS1 transgenic mice. It was observed that in both regions of the hippocampus and the cortex there were more activated microglia, which preferentially surrounded and infiltrated into plaques after ADSC transplantation. The activated microglia exhibited an alternatively activated phenotype, as indicated by their decreased expression levels of proinflammatory factors and elevated expression levels of alternative activation markers, as well as Aβ-degrading enzymes. In conclusion, ADSC transplantation could modulate microglial activation in AD mice, mitigate AD symptoms, and alleviate cognitive decline, all of which suggest ADSC transplantation as a promising choice for AD therapy. This manuscript is published as part of the International Association of Neurorestoratology (IANR) supplement issue of Cell Transplantation.

Keywords

Alzheimer's disease (AD)Adipose-derived mesenchymal stem cells (ADSCs)Transplantation Microglia

Introduction

Alzheimer's disease (AD), one of the most common types of dementia, affects more than 35 million people throughout the world, and this number will continue to go up with increasing life expectancy (39). AD is pathologically characterized by hallmark brain lesions, including senile plaques (SPs), as well as intracellular neurofibrillary tangles (NFTs) (12). SPs are primarily composed of β-amyloid (Aβ) peptide, which has been considered as a key factor in AD pathogenesis. Recent studies have also connected Aβ deposits with the activation of microglia and astrocytes, which respond to cerebral amyloidosis by inducing chronic inflammation, another causative factor of the disease (38). Several laboratories are aiming at modulation of microglial activation to manipulate this inflammatory process and therefore reduce the disease pathology and restore learning/memory in transgenic AD animal models (44).

Developments in stem cell technology in recent years have given rise to new therapeutic strategies for neurodegenerative disorders. Adipose-derived mesenchymal stem cells (ADSCs) can be extracted from adipose tissue easily, are capable of expansion in vitro, have the capacity to differentiate into multiple cell lineages, and can produce beneficial effects in several neurological disorders (1,30,34). Application of ADSCs has been shown to exhibit dopaminergic neuroprotective effects in animal models with Parkinson's disease-like pathology (34) as well as rescue Purkinje neurons and alleviate neuroinflammatory responses in Niemann–Pick disease type C model mice (1). In addition, ADSC treatment also improved cognitive performance in an animal model of traumatic brain injury (30). Although ADSC transplantation seems to represent a promising therapeutic strategy for these central nervous system (CNS) disorders (1,30,34), the efficacy of ADSC treatment on neuropathology in AD models, especially concerning the cognitive function and the underlying mechanism, remains to be established.

Microglia serve as the principal immune effector cells providing immune surveillance and tissue maintenance in the CNS (5). Studies have provided evidence that microglia surround amyloid plaques and infiltrate the deposits with their processes not only in AD transgenic animal models but also in human AD postmortem brains (8). Some research results have indicated that microglia could be activated by Aβ peptides to secrete neurotoxins and cytokines, leading to further neurodegeneration (16,42). In contrast, other studies have demonstrated that microglia actually provide neuroprotection by producing neurotrophic factors, diminishing Aβ deposition, and restricting SP formation (20,22). Jimenez et al. (22) have even observed a massive infiltration of microglia within the core of SPs in AD transgenic mouse models. Interestingly, Ohtaki and coworkers (36) have also found that adult stem cell transplantation could promote microglial activation in an ischemia rodent model and switch the phenotype of these microglial cells to an alternatively activated status exerting anti-inflammatory effects in response to injury. Moreover, results from Lee et al. (27,28) have suggested that intracerebral transplantation of bone marrow mesenchymal stem cells (BMSCs) could enhance microglial activation and decrease Aβ burden in AD animal models.

In this article, we investigated whether intracerebral ADSC transplantation could exert beneficial effects via microglial activation in APP/PS1 double transgenic AD model mice and whether these microglia could display an alternatively activated phenotype induced by the immunomodulatory action of the transplanted ADSCs. We found that ADSC transplantation promoted alternative microglial activation, inhibited proinflammatory responses, enhanced anti-inflammatory pathways, alleviated Aβ pathology in the brains of APP/PS1 transgenic AD model mice, and mitigated cognitive decline of the model animals.

Materials and Methods

Transgenic Mice

Double-transgenic mouse lines of Swedish mutation of amyloid precursor protein (APPswe; K595N/M596L) and presenilin 1 (PS1ΔE9) mice, the strain of B6C3-Tg (APPswe, PSEN1dE9) 85Dbo/J (stock number 004462), were from the Jackson Laboratories (Bar Harbor, ME, USA). The mice were maintained in vivarium facilities on their original genetic background until the desired age. These APP/PS1 double transgenic mice develop Aβ deposits and plaques in brains at the age of 6–7 months (4) and exhibit significant spatial learning/memory decline at about 8 months (40). In the present study, we used only male mice to exclude the sex-specific differences in AD pathology. All experimental procedures were conducted under the guidelines of the Institutional Animal Care and Use Committee and Animal Welfare and Ethics Committee of Tsinghua University. Research animals were kept in an air-conditioned room under a controlled temperature (22 ± 1°C) with a 12-h light–12-h dark cycle.

Cell Isolation and Culture

Adipose tissues were dissected from testicular fat pads of three male Sprague–Dawley rats (6- to 8-week; provided by the Animal Center of Haidian, Beijing, China). Isolation and culture of ADSCs were performed according to previously published protocols from our laboratory (47,48). Briefly, the harvested adipose tissue was washed with ice-cold Hank's balanced salt solution (HBSS; Invitrogen, Carlsbad, CA, USA) and dissociated mechanically. The tissue was then digested using collagenase type II (Gibco, Gaithersburg, MD, USA) at 37°C for 40 min and dissociated softly with pipettes. After filtering with a 70-μm cell strainer (BD Biosciences, San Jose, CA, USA), the suspension was centrifuged to separate the stromal vascular fraction from the floating adipocytes. The precipitation was resuspended and plated onto a cell culture dish (Corning, Corning, NY, USA) at a density of 10⁵ cells/ml in fresh Dulbecco's modified Eagle's medium (DMEM)/F-12 (Invitrogen) supplemented with 1% penicillin/streptomycin (Sigma-Aldrich, St. Louis, MO, USA) and 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA). Cell cultures were maintained in a standard humidified incubator, saturated by 5% CO₂, at 37°C. Twenty-four hours later, the nonadherent cells were eliminated by changing the culture medium. The adherent cells were expanded by serial passages, and ADSCs from three to five passages were used for subsequent experiments. These ADSCs were positive for CD44 and CD90 and negative for CD11b, CD31, and CD45 (48). They also displayed the capacity for multilineage differentiation into adipocytes and osteocytes (47).

Transplantation of ADSCs in APP/PS1 Transgenic AD Model Mice

ADSCs suspended in HBSS or HBSS alone were transplanted into the hippocampi of APP/PS1 transgenic mice at the age of 8 months (n = 10 per group). After anesthetization with 5% chloral hydrate (8 ml/kg body weight; Beijing Chemical Works, Beijing, China), the mice were fixed on a stereotaxic apparatus (RWD Life Science, Shenzhen, China). ADSC suspension of 3 ml (approximately 10⁵ cells) was injected bilaterally into the hippocampus via an automated infusion pump (PHD 2000; Harvard Apparatus, Cambridge, MA) at the rate of 1 μl/min, using the following stereotaxic coordinates: 2 mm posterior to bregma, 2 mm bilateral from midline, and 2 mm below surface. For HBSS control groups, HBSS alone was infused. After surgery, the mice were kept warm until they had recovered from the anesthesia and then returned to their cages.

Behavioral Experiments

Morris water maze (MWM) and novel object recognition (NOR) tasks were designed and performed to examine the spatial learning/memory function of APP/PS1 transgenic mice transplanted with HBSS or ADSCs, and their wild-type (WT) littermates (n = 5 per group). In the MWM test, we applied the method previously described with some modifications (10). A white tank (1.2 m in diameter; Beijing Xintiandi Technologies, Beijing, China) was filled with water (20 cm in depth), and the water was made opaque with nontoxic white paint (Van Aken Int., Cucamonga, CA, USA) before tests. A white circular platform (10 cm in diameter; Beijing Xintiandi Technologies) was located in the center of the northwest quadrant of the tank, approximately 1 cm below the water surface. Signs of distinct colors were pasted on each wall of the four quadrants to serve as visual position clues. The animals were trained with four trials per day for 5 consecutive days, with different starting locations between trials. Each animal was placed in water against the wall at the starting point, and it was given 60 s to swim to locate the hidden platform beneath the water surface. If the platform was not located after 60 s, the mouse would be gently guided to the location and was allowed 10 s on the platform. Escape latency was measured as the time mice took to locate the platform. In the last trial, the navigation paths of the mice from the starting point to the platform were recorded with a video camera. The probe test was performed on day 5, 6 hours after the navigation path recording. The platform was taken away from the pool, and each animal was allowed to swim freely in the water for 120 s; the times of crossing the exact location where the platform used to be placed was recorded.

The NOR task was carried out according to the previously described procedure with some modifications (2). The experiments were performed in a black open-field box of 25 cmx25 cmx40 cm (Beijing Xintiandi Technologies). Each mouse was subjected to a 10-min training session into the box, where two identical nontoxic objects were put symmetrically and equidistantly from the corners and the side walls, and then the mouse was returned to its home cage. After a 30-min retention interval, the mouse was placed back into the arena, with one of the familiar objects (used in the training session) replaced by a novel one with similar texture and size, but a different shape. Each mouse was allowed to explore for 5 min. Exploration occurred when the animal sniffed or touched the objects with nose and/or forepaws. The times of exploration were manually recorded. The recognition index was evaluated by the ratio of T_N/(T_F+T_N), where T_F represented times of exploring the familiar object and T_N represented times of interacting with the novel one.

Tissue Preparation

After behavioral experiments of cognitive function assessment, mice were anesthetized with 5% chloral hydrate (8 ml/kg body weight) and perfused intracardially with PBS (pH 7.4; YTHX Biotechnology, Beijing, China), which was followed by 4% paraformaldehyde (Beijing Chemical Works) at 37°C. The brain was removed and postfixed overnight at 4°C before being placed in 30% PBS-buffered sucrose solution (Beijing Chemical Works) until equilibration. Brains were then serially cut into 40-μm coronal sections with a freezing microtome (CM 1900; Leica, Wetzlar, Germany) and stored at 4°C until processed.

Plaque Staining with Thioflavin S

The sections were stained in PBS containing 0.5% Thioflavin S (Sigma) at room temperature for 10 min. After three washes with PBS, the sections were observed with a fluorescence microscope (BX41; Olympus, Tokyo, Japan).

Immunohistochemistry

Brain sections were incubated with primary antibodies overnight. The primary antibodies used included ionized calcium-binding adaptor molecule 1 (Iba-1; rabbit, 1:500 dilution; Wako Chemical, Osaka, Japan), interleukin 4 (IL-4; goat, 1:250 dilution; Santa Cruz, Santa Cruz, CA), and Aβ amino acids 1–16 antibody (6E10; mouse, 1:200 dilution; Covance, Princeton, NJ). For double or triple labeling, the primary antibodies of concern were used simultaneously to incubate the sections. The sections were then incubated with secondary antibodies conjugated to fluorescein isothiocyanate (FITC; ZSGB-BIO, Beijing, China), tetramethyl rhodamine isothiocyanate (TRITC; ZSGB-BIO), or cyanine 5 (Cy 5; ZSGB-BIO) for another 2 h. Nuclear staining was conducted by incubating the sections in PBST solution [PBS (YTHX Biotechnology) with 0.1% Tween-20 (Amresco, Solon, OH, USA)] with 4′,6-diamidino-2-phenylindole (DAPI; 1:1000 dilution; Dojindo, Kumamoto, Japan). Finally, the sections were visualized using a laser scanning confocal microscope (FV10i, Olympus).

Western Blot Analysis

Hippocampal and cortical tissues were isolated from the mice and homogenized in ice-cold lysis buffer (YTHX Biotechnology) containing protease inhibitors (YTHX Biotechnology) and 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF; YTHX Biotechnology), followed by a 10-s sonication on ice. The lysate was centrifuged at 11,000xg at 4°C for 10 min. The bicinchoninic (BCA) kit (Pierce, Rockford, IL, USA) was used to determine protein content. Equal amounts of protein were further diluted by 4x gel-loading buffer (60 mM Tris-HCl, 10% glycerol, 2% sodium dodecyl sulfate, 10% β-mercaptoethanol, and 0.005% bromophenol blue, pH 6.8; Beijing Chemical Works) before boiling for 5 min. These protein samples were separated using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE; reagents from YTHX Biotechnology) and then transferred from the gel onto a nitrocellulose membrane (0.45 μm; Amersham, Little Chalfont, Buckinghamshire, UK). The membrane was treated with blocking solution (YTHX Biotechnology) at room temperature for 2 h before incubation with 6E10 primary antibody (1:1000 dilution) at 4°C overnight to recognize APP. After incubation with secondary antibodies conjugated to alkaline phosphatase, protein detection was performed using BCIP/NBT substrates (Sigma). β-Actin (antibody from ZSGB-BIO) was used as the sample loading control.

Ab 40/42 ELISA Assay

Tissue extracts from transgenic mouse hippocampi and cortex were collected. Protease inhibitors and AEBSF were added to prevent the degradation of Aβ peptides. The concentration of Aβ40/42 was detected with human Aβ40 or Aβ42 ELISA kits (Invitrogen), respectively, following the manufacturer's instructions.

Quantitative Real-Time PCR

Total RNA was extracted from the hippocampus and the cortex of APP/PS1 transgenic mice with Trizol reagent (Invitrogen). The cDNA was generated from 2 to 4 μg of total RNA with the First-Strand cDNA Synthesis Kit (GeneCopoeia, Rockville, MD, USA) using oligo(dT)_12–18 primers. Quantitative real-time PCR was then performed in a system containing SYBR Green qPCR mix (GeneCopoeia) on the Stratagene Mx3000P machine (La Jolla, CA, USA). The ΔΔCT method was used to evaluate the fold changes in gene expression level between glyceraldehyde 3-phosphate dehydrogenase (GAPDH; control gene) and target genes. Sequences for qPCR primers used here have been previously published (11,23,26,45).

Statistical Analysis

Data were presented as mean ± SD. Analysis of variance with Bonferroni correction was applied to perform statistical comparisons. Values of p < 0.05 were considered statistically significant.

Results

ADSC Transplantation Alleviates Cognitive Impairment in APP/PS1 Transgenic Mice

To evaluate whether ADSC transplantation could alleviate cognitive impairment of these AD model mice, HBSS- and ADSC-transplanted APP/PS1 transgenic mice, as well as the WT littermates, were tested 25 days after the transplantation with the MWM and NOR tasks (2,10). The MWM task is hippocampus dependent and indicates the spatial learning/memory function of the animals. As indicated by escape latencies to find the hidden platform, HBSS-treated APP/PS1 transgenic mice exhibited significant spatial learning/memory deficits compared with the WT counterparts, whereas ADSC-transplanted APP/ PS1 transgenic mice performed much better than these HBSS-treated controls (Fig. 1A). At day 5, the representative navigation path recording suggested spatial learning/memory decline in APP/PS1 transgenic mice treated with HBSS, whereas the navigation pattern of ADSC-transplanted mice was comparable with that of the WT littermates (Fig. 1B). In the following probe tests, the spatial learning/memory impairment was significantly mitigated in ADSC-transplanted mice (Fig. 1C).

Figure 1.

ADSC transplantation alleviates cognitive impairment of APP/PS1 transgenic mice. (A) Escape latencies of Hank's balanced salt solution (HBSS; vehicle)- or adipose-derived stem cell (ADSC)-treated amyloid precursor protein-presenilin 1 (APP/PS1) transgenic mice and the WT littermates in MWM task (n = 5). (B) Representative navigation path recording on day 5 of the behavioral performance test. (C) Single probe test of spatial learning assessment on the final day (n = 5). (D) Learning/memory capability evaluation of APP/PS1 transgenic mice treated with HBSS or ADSCs and the WT counterparts with the NOR task (n = 5). Data shown are presented as mean ± SD. *p < 0.05, and **p < 0.01, compared with HBSS-treated mice.

To further compare cognitive capability of the transgenic mice treated with ADSCs or HBSS and that of their WT controls, we examined the recall memory of these mice with the NOR test, a nonaversive task based on the natural exploratory behavior of mice. WT mice preferentially explored novel objects. However, APP/PS1 transgenic mice treated with HBSS exhibited no discrimination between the novel and the familiar objects, whereas ADSC-transplanted mice showed more curiosity about the novel one (Fig. 1D). These results demonstrated that ADSC transplantation could alleviate cognitive impairment in APP/PS1 transgenic AD model mice.

ADSC Transplantation Reduces Aβ Burden in APP/PS1 Transgenic Mice

To investigate if ADSC transplantation could delay or diminish Aβ pathology, mice were killed and evaluated for Aβ plaque formation with Thioflavin S staining. We observed a significant reduction in Aβ depositions in ADSC-transplanted model mice compared with HBSS-treated controls (Fig. 2A) in both the hippocampus (Fig. 2B) and the cortex (Fig. 2C). SPs are primarily composed of Aβ peptides, which originate from sequential cleavage of amyloid precursor proteins (APPs) by β- and γ-secretases. To find out whether the decrease in Aβ aggregation was due to the reduced level of APP [molecular weight of APP (MW_APP) ~ 110 kDa, MW_sAPPα ~ 90 kDa, MW_Aβ ~ 4 kDa], Western blot analysis with 6E10 antibody was performed. There were no obvious differences in APP expression levels between model mice treated with HBSS and those treated with ADSCs (Fig. 2D), indicating that the reducing effects in Aβ burden of ADSC transplantation involves other mechanisms rather than influencing APP expression.

Figure 2.

ADSC transplantation diminishes Aβ plaques in the brains of APP/PS1 transgenic mice. (A) Thioflavin S staining detection of β-amyloid (Aβ) plaques in brain sections from mice treated with HBSS or ADSCs. Plaques were observed by fluorescence microscopy. Scale bar:100 μm. (B, C) Quantitative analysis of Aβ burden in hippocampus (B) and cortex (C) from mice treated with HBSS or ADSCs (n = 5). (D) Western blot analysis with 6E10 primary antibody indicating APP expression levels only (distinguished from Aβ by molecular weight of ~110 kDa) in the brains of APP/PS1 transgenic mice treated with HBSS or ADSCs. Data shown are presented as mean ± SD. *p < 0.05, and **p < 0.01, compared with HBSS-treated mice.

To further confirm that total Aβ load in the hippocampus and the cortex was reduced by ADSC transplantation, the brain sections were examined with immunofluorescence using 6E10 primary antibody. In ADSC-transplanted model mice, total 6E10 positive area was significantly smaller compared with that in HBSS-infused controls (Fig. 3A and B). Moreover, Aβ sandwich ELISA experiments indicated that the hippocampus from ADSC-transplanted mice contained significantly lower Aβ 40 and 42 than that from HBSS-infused animals (Fig. 3C and D). There were slight differences in Aβ 40 and 42 contents between the cortex regions of the two groups; however, the differences were not statistically significant. Nevertheless, we observed a nonsignificant decrease in Aβ 42 levels in the cortex of ADSC-treated mouse models compared with HBSS-treated controls (Fig. 3D). Taken together, these data provided clear evidence that ADSC transplantation inhibits Aβ deposition and plaque formation in APP/PS1 transgenic AD model mice.

Figure 3.

ADSC transplantation decreases Aβ burden in the brains of APP/PS1 transgenic mice. (A) Aβ plaques in brain sections of APP/PS1 transgenic mice immunostained with 6E10 antibody. 4′,6-Diamidino-2-phenylindole (DAPI) was used as a nuclear stain. Scale bar: 100 μm. (B) Quantitative analysis of area occupied by 6E10-positive Aβ plaques in HBSS- or ADSC-treated APP/PS1 transgenic mice (n = 5). (C, D) ELISA assessment of Aβ 40 (C) and Aβ 42 (D) expression levels in the brains of APP/PS1 transgenic mice (n = 3). Data shown are presented as mean ± SD. **p < 0.01 compared with HBSS-treated mice.

ADSC Transplantation Promotes Microglial Activation in the Brains of APP/PS1 Transgenic Mice

ADSCs have recently been suggested to alleviate the inflammatory responses in several neurological and other kinds of diseases (1,6,13). Microglia serve as the immune cells in the CNS. Recent studies have shown that microglia could be activated by BMSC treatment, and active microglia reduced Aβ deposits in AD animal models (27,28). Considering these results, we next investigated whether microglial activity was influenced by ADSC transplantation in APP/PS1 transgenic mice. Immunofluorescence was then performed with Iba-1 primary antibody, which could label activated microglia. The area positive for Iba-1 was significantly enlarged in ADSC-treated APP/ PS1 transgenic mice, in both regions of the hippocampus and the cortex, compared with HBSS-treated animals (Fig. 4A and B). These results suggested that microglial activation could be enhanced by ADSC transplantation in the brains of APP/PS1 transgenic mice.

Figure 4.

Activated microglia adopt alternative phenotypes in the brains of ADSC-transplanted APP/PS1 transgenic mice. (A) Immunofluorescence pattern of ionized calcium-binding adaptor molecule 1 (Iba-1) primary antibody in brain sections of HBSS-or ADSC-treated APP/PS1 transgenic mice. Scale bar: 20 μm. (B) Quantitative analysis of area covered by activated microglia in the regions of hippocampus and cortex in APP/PS1 transgenic mice treated with HBSS or ADSCs (n = 5). (C, D) mRNA expression levels of tumor necrosis factor-α (TNF-α) (C) and interleukin (IL)-1β (D) detected by quantitative real-time PCR in APP/PS1 transgenic mice treated with HBSS or ADSCs (n = 3). (E, F) Quantitative real-time PCR assessment detecting mRNA expressions of the alternatively activated microglia markers, IL-4 (E) and arginase (ARG1) (F), in HBSS- or ADSC-treated APP/PS1 transgenic mice (n = 3). Data shown are presented as mean ± SD. *p < 0.05, and **p < 0.01, compared with HBSS-treated mice.

ADSC Transplantation-Activated Microglia Adopt an Alternative Phenotype in APP/PS1 Transgenic Mice

Our results indicated that ADSC transplantation could promote microglial activation in APP/PS1 transgenic mice. Recent studies suggest that microglia could be activated by Aβ accumulation in AD progression to increase the expression of proinflammatory cytokines, tumor necrosis factor-α (TNF-α), and interleukin-1β (IL-1β), which are related to the neuronal malfunction and cognitive decline in AD models (18). In fact, downregulation of these toxic factors has also been demonstrated to reduce the disease pathology (17,18). Accordingly, we investigated the mRNA expression levels of TNF-α and IL-1β in our models. Results revealed that ADSC transplantation dramatically downregulated the mRNA expression levels of these cytokines in both brain regions of the hippocampus (Fig. 4C) and the cortex (Fig. 4D) in APP/PS1 transgenic mice.

It is interesting to note that ADSC transplantation promoted microglial activation but the expression of these proinflammatory cytokines was reduced. Recent studies have brought clear evidence that microglia could exhibit different phenotypes, that is, the classic and the alternative forms. Alternatively activated microglia participate in repair processes during or next to the first steps of acute innate immune responses (15,16). Thus, we wondered whether the activated microglia in ADSC-transplanted mice did adopt the alternatively activated phenotype. It is widely accepted that the alternative activation of macrophage/microglia involves strong expression of IL-4 and arginase 1 (ARG1) (5,7). To find out whether ADSC transplantation-activated microglia expressed these alternatively activation markers, quantitative real-time PCR was performed to examine the mRNA expression levels of IL-4 and ARG1 genes. Obvious increases in the expressions of IL-4 and ARG1 were observed in both regions of the hippocampus (Fig. 4E) and cortex (Fig. 4F) of APP/ PS1 transgenic mice transplanted with ADSCs compared with HBSS-treated controls.

ADSC Transplantation Enhances Aβ Clearing Capacity of Microglia

Results showed that ADSC transplantation promoted microglial activation in the brains of APP/PS1 transgenic mice, and the microglia expressing anti-inflammatory cytokines adopted an alternatively activated phenotype (Fig. 4). Therefore, we next tried to find out if there was a close connection between the enhanced activation of microglia and the diminished deposition of Aβ peptides in ADSC-transplanted APP/PS1 transgenic mice. The coimmunostaining with Iba-1 and 6E10 primary antibodies was performed to simultaneously detect activated microglia and Aβ deposits. Confocal microscopy demonstrated that, in the brains of ADSC-transplanted APP/PS1 transgenic mice, many Iba-1-positive microglia were near Aβ plaques (Fig. 5A).

Figure 5.

ADSC transplantation recruits microglia around Aβ plaques and enhances their Aβ clearing capacity in APP/PS1 transgenic mice. (A) Individual and coimmunostaining of microglia and Aβ deposition with Iba-1 and 6E10 antibodies in brain sections of APP/ PS1 transgenic mice treated with HBSS or ADSCs. DAPI was used as a nuclear stain. Scale bar: 20 μm. (B, C, D) mRNA expression levels of the Aβ-degrading enzymes, insulin-degrading enzyme (IDE) (B), neprilysin (NEP) (C), and matrix metalloproteinase 9 (MMP-9) (D), detected by quantitative real-time PCR analysis (n = 3). Data shown are presented as mean ± SD. **p < 0.01 compared with HBSS-treated mice.

It has been reported that alternatively activated microglia could secrete Aβ-degrading enzymes, including insulin-degrading enzyme (IDE), neprilysin (NEP), and matrix metalloproteinase 9 (MMP-9) (9,20,33,49). Previous studies have shown that microglia become impaired in their expression of Aβ-degrading enzymes in AD model mice as they age (48). On the basis of these reports, we believe that Aβ clearance after ADSC transplantation in APP/PS1 transgenic mice is mediated, at least partly, by the modulation of microglial secretion of Aβ-degrading enzymes. We investigated the expression of these Aβ-related proteolytic enzymes with quantitative real-time PCR analysis. The results showed that ADSC transplantation dramatically increased the mRNA expression levels of these microglia-secreted Aβ-degrading enzymes in both the hippocampus and the cortex in AD model animals (Fig. 5B, C, and D). These results suggest that ADSC transplantation could enhance microglial secretion of Aβ-degrading enzymes in APP/PS1 transgenic mice.

Furthermore, triple-immunofluorescence staining with 6E10, Iba-1, and IL-4 antibodies was performed to confirm if microglia near Aβ plaques in ADSC-transplanted APP/PS1 transgenic mice were actually alternatively activated. Results indicated that microglia surrounding Aβ plaques in ADSC-treated mouse brains expressed IL-4; that is, they did exhibit an alternatively activating phenotype. The immunofluorescence staining also demonstrated that IL-4 expression level was dramatically upregulated in ADSC-transplanted AAP/PS1 transgenic mice, as compared with the HBSS-treated group (Fig. 6). The Iba-1-positive microglia expressing IL-4 were abundantly found to locate near Aβ depositions in the brains of ADSC-transplanted APP/PS1 transgenic mice but not in HBSS-infused animals. These results suggested that microglia surrounding Aβ plaques were alternatively activated, eliciting neuroprotective effects in ADSC-transplanted APP/PS1 transgenic mice.

Figure 6.

Microglia surrounding Aβ deposits express IL-4 in ADSC-transplanted APP/PS1 transgenic mice. Triple labeling of 6E10, Iba-1, and IL-4 in APP/PS1 mice treated with HBSS (upper panel) or ADSCs (lower panel). IL-4-positive microglia are indicated by arrows. Scale bar: 20 μm.

Discussion

The present study evaluated the therapeutic potential of intracerebral ADSC transplantation and the underlying mechanism in AD model mice overexpressing familial AD-linked APP and PS1. Results from behavioral experiments indicated that ADSC transplantation significantly improved the cognitive function in AD model mice. Furthermore, ADSC transplantation promoted microglial activation, which enhanced Aβ clearing capacity in the hippocampus and the cortex. During microglial activation, ADSC-transplanted mice exhibited not only decreased neurotoxic cytokine expression levels but also increased expression levels of anti-inflammatory cytokines, indicative of alternative microglial activation. The alternatively activated microglia preferentially surrounded and infiltrated into Aβ deposits in the brains of APP/PS1 transgenic mice after ADSC transplantation. Collectively, our results demonstrate the feasibility and efficiency for the application of ADSCs in AD therapy.

Although activated microglia have long been associated with neuroinflammation, which contributes to AD pathogenesis (5), it is recently accepted that microglia exhibit more complex and heterogeneous phenotypes, depending on the pathological circumstances. In some CNS disease cases, they could release neuroprotective cytokines and exert beneficial effects. For example, recent studies indicate that activated microglia could constitutively mediate the clearance of Aβ peptides and restrict SP formation in AD (20,43). Besides, in an excitotoxicity-induced neurodegenerative model, ramified microglia exerted beneficial effects in providing neuronal resistance toward excitotoxic insults (46). Herein, we found that microglial activation was dramatically enhanced after ADSC transplantation in the brains of APP/PS1 transgenic model mice (Fig. 4A and B). The underlying mechanism could be complicated. Brain-resident microglia undergo low-level proliferation throughout the organism's lifetime. On the other hand, circulating monocytes in the periphery could infiltrate into the brain and contribute to microglial accumulation, particularly after blood–brain barrier breakdown under certain pathological conditions such as AD (5). The exact mechanism of the enhanced microglial activation in APP/PS1 transgenic mice after ADSC transplantation observed in the present study merits further investigation.

Microglia could play a coordinated and sophisticated role to deactivate the initial immune response and lead to processes for tissue repair (15,16). They could be transformed from the classic proinflammatory phenotype to an alternatively activating form. Given that microglial functions were mainly coordinated through cytokines (3), we measured the expression levels of microglia-related cytokines to identify microglial status. Encouragingly, the mRNA expression levels of proinflammatory cytokines (TNF-α and IL-1β) were dramatically reduced (Fig. 4C and D), whereas the mRNA expression levels of anti-inflammatory cytokines (IL-4 and ARG1) were significantly elevated (Fig. 4E and F) after ADSC transplantation in APP/PS1 transgenic mice. IL-4 is one of the widely acknowledged immune regulatory cytokines suppressing inflammation (31). ARG1 is also an inflammatory modulator accounting for wound healing (19). Especially, both IL-4 and ARG1 have been recognized as markers for the alternative activation of macrophage (5,7,14). Therefore, we believe that the activated microglia in ADSC-transplanted APP/PS1 transgenic mice mainly adopted an alternatively activated phenotype, which would be beneficial in our AD models.

It has been recently shown that microglia can exert neuroprotective effects in AD via mediating Aβ clearance (43). Endogenous microglial activation induced by the BMSC transplantation participated in eliminating Aβ deposits in AD animal models (28). On the other hand, the dysfunction of microglia was associated with defective Aβ clearance pathway (20), and the impairment of microglia accumulation led to higher brain Aβ levels and then accelerated AD progression (8,25). Accordingly, we examined whether the alternatively activated microglia after ADSC transplantation in APP/PS1 transgenic mice could eliminate Aβ deposition. Our results showed that the enhanced microglial activation was accompanied by a significant decrease in Aβ deposition in the brains of ADSC-transplanted APP/PS1 transgenic mice (Figs. 2–4). ELISA assessments also indicated a decrease in Aβ 40 and Aβ 42 peptides in the brains of these mice after ADSC transplantation (Fig. 3). It is also observed that activated microglia migrated to Aβ plaques in response to ADSC transplantation (Fig. 5). Furthermore, ADSC transplantation dramatically increased the mRNA expression levels of several microglia-secreted Aβ-degrading enzymes, including IDE, NEP, and MMP-9, which are major regulators of Aβ metabolism within the brain (9,33). IDE expression is lower, and its enzymatic activity is reduced in AD brains, increasing disease risk (9). NEP is dramatically decreased in high plaque areas of AD brains, and the deficient degradation of Aβ due to NEP deficiency may contribute to AD pathogenesis (33). On the basis of these results, we propose that the Aβ deposition-reducing effect of ADSC transplantation in APP/PS1 transgenic mice can be attributed, at least partly, to increased Aβ-degrading enzymatic activity produced by the alternatively activated microglia. In addition to Aβ-degrading enzymes, microglia may also express Aβ-binding receptors under physiological and pathological conditions (20,41), which could be another pathway facilitating Aβ clearance through phagocytosis that merits further investigation.

Our triple-immunofluorescent analysis further indicated that microglia surrounding Aβ plaques were predominantly expressing IL-4 in ADSC-transplanted APP/ PS1 transgenic mice, whereas there were few IL-4-positive microglial cells anywhere near Aβ deposits in the AD mice treated with HBSS (Fig. 6). It was reported that microglia expressing IL-4 could secrete growth factors (5), reduce Aβ toxicity, and enhance the clearance of Aβ deposits (24,31). This result is an additional support for our supposition that the alternative microglial activation is an important mechanism by which ADSC transplantation exerts neuroprotective effects on APP/PS1 transgenic AD models.

The MWM and NOR tests herein demonstrated that ADSC transplantation significantly improved the cognitive and behavioral performance in APP/PS1 transgenic mice (Fig. 1). It is widely accepted that the cognitive impairment in AD models is linked to increased Aβ deposition, one of the most important pathological signatures of AD (12,29,38,40). Accordingly, we propose that ADSC transplantation enhances alternative activation of microglia, increases the secretion of Aβ-degrading enzymes and anti-inflammatory cytokines, suppresses proinflammatory cytokine expression, and therefore eliminates Aβ deposition and alleviates cognitive decline in APP/PS1 transgenic AD model mice.

In this work, we focused on the feasibility and efficacy of ADSC transplantation in AD model mice therapy and the underlying mechanism from the angle of immunomodulatory effect via microglial activation. There are still many other questions concerning the mechanisms, such as the fate of the transplanted ADSCs and the way through which they affect microglial activation. Recent findings suggest that the persistence of the transplanted mesenchymal stem cells (MSCs) might not be necessary for functional recovery (21,37). It was also reported that most of these cells disappear quickly after transplantation. For example, human bone marrow MSCs transplanted into the hippocampus of adult mice did not proliferate, and no more than 26% of these cells survived 3 days after transplantation, whereas their expression of neurotrophins could greatly increase the proliferation of adjacent cells positive for stem cell markers to achieve tissue repair (35). Moreover, transdifferentiation and/or spontaneous cell fusion could be alternative mechanisms through which MSCs elicit their beneficial effects in the damaged nervous system (32). Considering that these two phenomena take place at extremely low frequency in vivo, the contribution of these processes to MSC-mediated neural repair is probably limited (35). To this end, the paracrine activities of MSCs through the secretion of neurotrophic factors may play an important role in MSC-mediated neural recovery.

Conclusion

In the present study, we showed that intracerebral ADSC transplantation could efficiently mitigate cognitive decline and spatial learning/memory impairment in APP/PS1 transgenic AD mouse models. Aβ deposition was significantly reduced, and microglial activation was predominantly enhanced after ADSC transplantation. We propose that ADSC transplantation exerts neuroprotective effects by an alternative microglial activation characterized by secretion of anti-inflammatory cytokines and Aβ-degrading enzymes. These findings provide a basis for the immunomodulatory strategies of AD therapy using ADSCs.

Footnotes

Acknowledgments

This work was supported by the National High Technology Research and Development Program of China (“863” Program, 2012AA020905), the State Key Laboratory of Biomembrane and Membrane Biotechnology, the National Natural Science Foundation of China (NSFC)/Research Grants Council of Hong Kong (RGC) Joint Research Scheme (81161160570), and Tsinghua-Yue-Yuen Medical Sciences Fund (20240000514). The authors declare no conflict of interest.

References

Bae

J. S.

; Carter

J. E.

; Jin

H. K.

Adipose tissue-derived stem cells rescue Purkinje neurons and alleviate inflammatory responses in Niemann–Pick disease type C mice. Cell Tissue Res. 340(2): 357–369; 2010.

Bevins

R. A.

; Besheer

Object recognition in rats and mice: A one-trial non-matching-to-sample learning task to study “recognition memory”. Nat. Protoc. 1(3): 1306–1311; 2006.

Bodles

A. M.

; Barger

S. W.

Cytokines and the aging brain—What we don't know might help us. Trends Neurosci. 27(10): 621–626; 2004.

Borchelt

D. R.

; Ratovitski

; van Lare

; Lee

M. K.

; Gonzales

; Jenkins

N. A.

; Copeland

N. G.

; Price

D. L.

; Sisodia

S. S.

Accelerated amyloid deposition in the brains of transgenic mice coexpressing mutant presenilin 1 and amyloid precursor proteins. Neuron 19(4): 939–945; 1997.

Cameron

; Landreth

G. E.

Inflammation, microglia, and Alzheimer's disease. Neurobiol. Dis. 37(3): 503–509; 2010.

Chen

Y. T.

; Sun

C. K.

; Lin

Y. C.

; Chang

L. T.

; Chen

Y. L.

; Tsai

T. H.

; Chung

S. Y.

; Chua

; Kao

Y. H.

; Yen

C. H.

; Shao

P. L.

; Chang

K. C.

; Leu

; Yip

H. K.

Adipose-derived mesenchymal stem cell protects kidneys against ischemia-reperfusion injury through suppressing oxidative stress and inflammatory reaction. J. Transl. Med. 9: 51; 2011.

Colton

C. A.

; Mott

R. T.

; Sharpe

; Xu

; Van Nostrand

W. E.

; Vitek

M. P.

Expression profiles for macrophage alternative activation genes in AD and in mouse models of AD. J. Neuroinflammation 3: 27; 2006.

El Khoury

; Luster

A. D.

Mechanisms of microglia accumulation in Alzheimer's disease: Therapeutic implications. Trends Pharmacol. Sci. 29(12): 626–632; 2008.

Farris

; Mansourian

; Chang

; Lindsley

; Eckman

E. A.

; Frosch

M. P.

; Eckman

C. B.

; Tanzi

R. E.

; Selkoe

D. J.

; Guenette

Insukin-degrading enzyme regulates the levels of insulin, amyloid beta protein, the beta-amyloid precursor protein intracellular domain in vivo. Proc. Natl. Acad. Sci. USA 100(7): 4162–4167; 2003.

10.

Fiorentini

; Rosi

M. C.

; Grossi

; Luccarini

; Casamenti

Lithium improves hippocampal neurogenesis, neuropathology and cognitive functions in APP mutant mice. PLoS One 5(12): e14382; 2010.

11.

Gangadharan

; Hoeve

M. A.

; Allen

J. E.

; Ebrahimi

; Rhind

S. M.

; Dutia

B. M.

; Nash

A. A.

Murine gammaherpesvirus-induced fibrosis is associated with the development of alternatively activated macrophage. J. Leukoc. Biol. 84(1): 50–58; 2008.

12.

Goedert

; Spillantini

M. G.

A century of Alzheimer's disease. Science 31(5800): 777–781; 2006.

13.

Gonzalez-Rey

; Gonzalez

M. A.

; Varela

; O'Valle

; Hernandez-Cortes

; Rico

; Buscher

; Delgado

Human adipose-derived mesenchymal stem cells reduce inflammatory and T cell responses and induce regulatory T cells in vitro in rheumatoid arthritis. Ann. Rheum. Dis. 69(1): 241–248; 2010.

14.

Gordon

Alternative activation of macrophage. Nat. Rev. Immunol. 3(1): 23–35; 2003.

15.

Gordon

; Martinez

F. O.

Alternative activation of macrophages: Mechanism and functions. Immunity 32(5): 593–604; 2010.

16.

Graeber

M. B.

; Streit

W. J.

Microglia: Biology and pathology. Acta Neuropathol. 119(1): 89–105; 2010.

17.

; Zhong

; Lindholm

; Berning

; Lee

; Lemere

; Staufenbiel

; Li

; Shen

Deletion of tumor necrosis factor death receptor inhibits amyloid beta generation and prevents learning and memory deficits in Alzheimer's mice. J. Cell Biol. 178(5): 829–841; 2007.

18.

Heneka

M. T.

; O'Banion

M. K.

Inflammatory processes in Alzheimer's disease. J. Neuroimmunol. 184(1-2): 69–91; 2007.

19.

Hesse

; Modolell

; La Flamme

A. C.

; Schito

; Fuentes

J. M.

; Cheever

A. W.

; Pearce

E. J.

; Wynn

T. A.

Differential regulation of nitric oxide synthase-2 and arginase-1 by type 1/type 2 cytokines in vivo: Granulomatous pathology is shaped by the pattern of L-arginine metabolism. J. Immunol. 167(11): 6533–6544; 2001.

20.

Hickman

S. E.

; Allison

E. K.

; El Khoury

Microglial dysfunction and defective beta-amyloid clearance pathways in aging Alzheimer's disease mice. J. Neurosci. 28(33): 8354–8360; 2008.

21.

Himes

B. T.

; Neuhuber

; Coleman

; Kushner

; Swanger

S. A.

; Kopen

G. C.

; Wagner

; Shumsky

J. S.

; Fischer

Recovery of function following grafting of human bone marrow-derived stromal cells into the injured spinal cord. Neurorehabil. Neural Repair 20(2): 278–296; 2006.

22.

Jimenez

; Baglietto-Vargas

; Caballero

; Moreno-Gonzalez

; Torres

; Sanchez-Varo

; Ruano

; Vizuete

; Gutierrez

; Vitorica

Inflammatory response in the hippocampus of PS1M146L/APP751SL mouse model of Alzheimer's disease: Age-dependent switch in the microglial phenotype from alternative to classic. J. Neurosci. 28(45): 11650–11661; 2008.

23.

Kalinin

; Gavrilyuk

; Polak

P. E.

; Vasser

; Zhao

; Heneka

M. T.

; Feinstein

D. L.

Noradrenaline deficiency in brain increases beta-amyloid plaque burden in an animal model of Alzheimer's disease. Neurobiol. Aging 28(8): 1206–1214; 2007.

24.

Koenigsknecht-Talboo

; Landreth

G. E.

Microglial phagocytosis induced by fibrillar beta-amyloid and IgGs are differentially regulated by pro-inflammatory cytokines. J. Neurosci. 25(36): 8240–8249; 2005.

25.

Lee

C. Y. D.

; Landreth

G. E.

The role of microglia in amyloid clearance from the AD brain. J. Neural Transm. 117(8): 949–960; 2010.

26.

Lee

H. J.

; Lee

J. K.

; Lee

; Carter

J. E.

; Chang

J. W.

; Oh

; Yang

Y. S.

; Suh

J. G.

; Lee

B. H.

; Jin

H. K.

; Bae

J. S.

Human umbilical cord blood-derived mesenchymal stem cells improve neuropathology and cognitive impairment in an Alzheimer's disease mouse model through modulation of neuroinflammation. Neurobiol. Aging 33(3): 588–602; 2012.

27.

Lee

J. K.

; Jin

H. K.

; Bae

J. S.

Bone marrow-derived mesenchymal stem cells reduce brain amyloid-beta deposition and accelerate the activation of microglia in an acutely induced Alzheimer's disease mouse model. Neurosci. Lett. 450(2): 136–141; 2009.

28.

Lee

J. K.

; Jin

H. K.

; Endo

; Schuchman

E. H.

; Carter

J. E.

; Bae

J. S.

Intracerebral transplantation of bone marrow-derived mesenchymal stem cells reduces amyloid-beta deposition and rescues memory deficits in Alzheimer's disease mice by modulation of immune responses. Stem Cells 28(2): 329–343; 2010.

29.

Lesne

; Koh

M. T.

; Kotilinek

; Kayed

; Glabe

C. G.

; Yang

; Gallagher

; Ashe

K. H.

A specific amyloid-beta protein assembly in the brain impairs memory. Nature 440(7082): 352–357; 2006.

30.

D. F.

; Yang

; Qu

R. M.

; Yang

H. Y.

; Yu

M. C.

; Tao

; Dai

J. X.

; Yuan

Cognitive improvement following transvenous adipose-derived mesenchymal stem cell transplantation in a rat model of traumatic brain injury. Neural Regen. Res. 6(10): 732–737; 2011.

31.

Lyons

; Griffin

R. J.

; Costelloe

C. E.

; Clarke

R. M.

; Lynch

M. A.

IL-4 attenuates the neuroinflammation induced by amyloid-beta in vivo and in vitro. J. Neurochem. 101(3): 771–781; 2007.

32.

Maltman

D. J.

; Hardy

S. A.

; Przyborski

S. A.

Role of mesenchymal stem cells in neurogenesis and nervous system repair. Neurochem. Int. 59(3): 347–356; 2011.

33.

Marr

R. A.

; Rockenstein

; Mukherjee

; Kindy

M. S.

; Hersh

L. B.

; Gage

F. H.

; Verma

I. M.

; Masliah

Neprilysin gene transfer reduces human amyloid pathology in transgenic mice. J. Neurosci. 23(6): 1992–1996; 2003.

34.

McCoy

M. K.

; Martinez

T. N.

; Ruhn

K. A.

; Wrage

P. C.

; Keefer

E. W.

; Botterman

B. R.

; Tansey

K. E.

; Tansey

M. G.

Autologous transplants of adipose-derived adult stromal (ADAS) cells afford dopaminergic neuroprotection in a model of Parkinson's disease. Exp. Neurol. 210(1): 14–29; 2008.

35.

Munoz

J. R.

; Stoutenger

B. R.

; Robinson

A. P.

; Spees

J. L.

; Prockop

D. J.

Human stem/progenitor cells from bone marrow promote neurogenesis of endogenous neural stem cells in the hippocampus of mice. Proc. Natl. Acad. Sci. USA 102(50): 18171–18176; 2005.

36.

Ohtaki

; Ylostalo

J. H.

; Foraker

J. E.

; Robinson

A. P.

; Reger

R. L.

; Shioda

; Prockop

D. J.

Stem/progenitor cells from bone marrow decrease neuronal death in global ischemia by modulation of inflammatory/immune response. Proc. Natl. Acad. Sci. USA 105(38): 14638–14643; 2008.

37.

Parr

A. M.

; Tator

C. H.

; Keating

Bone marrow-derived mesenchymal stromal cells for the repair of central nervous system injury. Bone Marrow Transplant. 40(7): 609–619; 2007.

38.

Praticò

; Trojanowski

J. Q.

Inflammatory hypotheses: Novel mechanisms of Alzheimer's neurodegeneration and new therapeutic targets? Neurobiol. Aging 21(3): 441–445; 2000.

39.

Querfurth

H. W.

; La Ferla

F. M.

Mechanisms of disease Alzheimer's disease. New Engl. J. Med. 362(4): 329–344; 2010.

40.

Reiserer

R. S.

; Harrison

F. E.

; Syverud

D. C.

; McDonald

M. P.

Impaired spatial learning in the APP+PSEN1 DeltaE9 bigenic mouse model of Alzheimer's disease. Genes Brain Behav. 6(1): 54–65; 2007.

41.

Schlachetzki

J. C.

; Hüll

Microglial activation in Alzheimer's disease. Curr. Alzheimer Res. 6(6): 554–563; 2009.

42.

Schwab

; McGeer

P. L.

Inflammatory aspects of Alzheimer disease and other neurodegenerative disorders. J. Alzheimers Dis. 13(4): 359–369; 2008.

43.

Simard

A. R.

; Soulet

; Gowing

; Julien

J. P.

; Rivest

Bone marrow-derived microglia play a critical role in restricting senile plaque formation in Alzheimer's disease. Neuron 49(4): 489–502; 2006.

44.

Tan

; Town

; Paris

; Mori

; Suo

; Crawford

; Mattson

M.P.

; Flavell

R.A.

; Mullan

Microglial activation resulting from CD40-CD40L interaction after beta-amyloid stimulation. Science 286(5448): 2352–2355; 1999.

45.

Umemura

; Saio

; Suwa

; Kitoh

; Bai

; Nonaka

; Ouyang

G. F.

; Okada

; Balazs

; Adany

; Shibata

; Takami

Tumor-infiltrating myeloid-derived suppressor cells are pleiotropic-inflamed monocytes/ macrophages that bear M1- and M2-type characteristics. J. Leukoc. Biol. 83(5): 1136–1144; 2008.

46.

Vinet

; van Weering

H. R. J.

; Heinrich

; Kalin

R. E.

; Wegner

; Brouwer

; Heppner

F. L.

; van Rooijen

; Boddeke

H. W. G. M.

; Biber

Neuroprotective function for ramified microglia in hippocampal excitotoxicity. J. Neuroinflammation 9: 27; 2012.

47.

Wei

; Gong

; Zheng

; Liu

; Wang

; Zhang

; Ao

; Gong

; Zhang

Schwann-like cell differentiation of rat adipose-derived stem cells by indirect co-culture with Schwann cells in vitro. Cell Prolif. 43(6): 606–616; 2010.

48.

Wei

; Gong

; Zheng

; Wang

; Ao

; Gong

; Zhang

Chitosan/silk fibroin-based tissue-engineered graft seeded with adipose-derived stem cells enhances nerve regeneration in a rat model. J. Mater. Sci. Mater. Med. 22(8): 1947–1964; 2011.

49.

Yan

; Hu

; Song

; Yin

; Bateman

R. J.

; Cirrito

J. R.

; Xiao

; Hsu

F. F.

; Turk

J. W.

; Xu

; Hsu

C. Y.

; Holtzman

D. M.

; Lee

J. M.

Matrix metalloproteinase-nine degrades amyloid-beta fibrils in vitro and compact plaques in situ. J. Biol. Chem. 281(34): 24566–24574; 2006.