Metformin Preconditioning of Human Induced Pluripotent Stem Cell-Derived Neural Stem Cells Promotes Their Engraftment and Improves Post-Stroke Regeneration and Recovery

Abstract

While transplantation of human induced pluripotent stem cell-derived neural stem cells (hiPSC-NSCs) shows therapeutic potential in animal stroke models, major concerns for translating hiPSC therapy to the clinic are efficacy and safety. Therefore, there is a demand to develop an optimal strategy to enhance the engraftment and regenerative capacity of transplanted hiPSC-NSCs to produce fully differentiated neural cells to replace lost brain tissues. Metformin, an FDA-approved drug, is an optimal neuroregenerative agent that not only promotes NSC proliferation but also drives NSCs toward differentiation. In this regard, we hypothesize that preconditioning of hiPSC-NSCs with metformin before transplantation into the stroke-damaged brain will improve engraftment and regenerative capabilities of hiPSC-NSCs, ultimately enhancing functional recovery. In this study, we show that pretreatment of hiPSC-NSCs with metformin enhances the proliferation and differentiation of hiPSC-NSCs in culture. Furthermore, metformin-preconditioned hiPSC-NSCs show increased engraftment 1 week post-transplantation in a rat endothelin-1 focal ischemic stroke model. In addition, metformin-preconditioned cell grafts exhibit increased survival compared to naive cell grafts at 7 weeks post-transplantation. Analysis of the grafts demonstrates that metformin preconditioning enhances the differentiation of hiPSC-NSCs at the expense of their proliferation. As an outcome, rats receiving metformin-preconditioned cells display accelerated gross motor recovery and reduced infarct volume. These studies represent a vital step forward in the optimization of hiPSC-NSC-based transplantation to promote post-stroke recovery.

Introduction

Stroke is an enormous public health issue with 40% of patients having permanent neurological disabilities [1]. Novel poststroke interventions oriented to improve functional recovery are urgently needed due to the rising incidence of stroke impairment as a result of increased post-stroke survival and growing and aging population [1]. Stem cell-based therapies have been extensively studied in the last 15 years as a treatment for ischemic stroke. Particularly, the generation of human induced pluripotent stem cells (iPSCs) through reprogramming of somatic cells by a set of core pluripotent transcription factors has revolutionized cell therapy by providing a source of autologous cells for transplantation [2]. Earlier studies showed that transplantation of iPSCs promotes functional recovery in animal stroke models, and that these iPSCs can successfully differentiate into cells expressing markers of glial and neuronal lineages in vivo [3,4]. However, the pluripotent feature of iPSCs increases the likelihood of teratoma formation after their transplantation into stroke model [3 –6]. To limit tumorigenesis and maximize neural differentiation lineage, extensive research has focused on transplanting human iPSC-derived neural stem cells (hiPSC-NSCs) to promote functional recovery in rodent stroke models [7 –11]. While these studies showed promising therapeutic potential, major concerns for translating hiPSC-based cell therapy to the clinic are efficacy and safety. Therefore, there is a demand to develop an optimal strategy to enhance the engraftment and regenerative capacity of transplanted hiPSC-NSCs to generate fully differentiated neural cells.

Several different methods to optimize NSC transplants have been explored with varying success, including transplantation with hydrogels, hypoxic preconditioning, and genetic modifications [12,13]. These treatments focus on increasing initial engraftment with limited capacity to promote NSC differentiation into mature neurons and limited studies showing long-term cell survival accompanied by behavioral evaluation. Previous work shows that FDA-approved drug metformin is an optimal neuroregenerative agent, acting on multiple stages of adult neural precursor development, including proliferation, differentiation, and cell survival [14 –16]. Intriguingly, metformin-induced NSC proliferation persistently occurred in secondary and tertiary neurospheres in the absence of metformin in culture [14]. This persistent increase in NSC proliferation suggests a potential to use metformin as a preconditioning agent before transplantation to enhance regenerative capabilities of hiPSC-NSCs.

To date, most transplantation studies with hiPSC (or human embryonic stem cell [hESC])-derived NSCs utilize a middle cerebral artery occlusion (MCAO) stroke model. While this model resembles human ischemia as it reflects the most common stroke occlusion in humans, it produces large and variable infarcted regions and complications that interfere with poststroke neurological behavior [17 –21]. This may pose challenges when translating research into the clinic [22]. While endothelin-1 (ET-1) focal ischemic stroke with gradual reperfusion over 16–22 h more closely mimics human stroke, little research has assessed the effects of hiPSC-NSC transplantation in ET-1 focal ischemic stroke [23]. Consequently, we used the ET-1 focal ischemic stroke model in this study.

There are many mechanisms that may contribute to functional recovery following neural stem cell transplantation for stroke. While mechanisms involving support of endogenous neurogenesis, angiogenesis, reduction of inflammation, and reduced neural cell death have received much attention, differentiation and integration of exogenous cells require further investigation for its long-term effect on functional recovery [24]. These processes may occur in parallel and likely support each other.

In this study, we show that metformin preconditioning enhances the proliferation and differentiation of hiPSC-NSCs in culture. Furthermore, transplantation of preconditioned hiPSC-NSCs into a rat ET-1 stroke-damaged brain increased their engraftment. Notably, rats receiving metformin-preconditioned cell grafts showed accelerated recovery of gross motor function and a significant reduction in infarct volume. These studies represent a vital step in the optimization of hiPSC-NSC-based transplantation to promote post-stroke recovery.

Materials and Methods

Animal handling and housing

All animal use was approved by the Animal Care Committees of the University of Ottawa in accordance with the Canadian Council of Animal Care policies. Sprague Dawley rats from Charles River (200–250 g) were maintained on reverse light cycle (12-h dark/12-h light) with ad libitum access to food and water. Rats were housed in pairs in Sealsafe green line caging. All rats were habituated for 1 week followed by 1 week of handling (∼5 min/rat/day) before commencing behavioral testing.

Human iPSC-NSC induction and culturing

The hiPSC line, WLS1C, derived at the Ottawa Hospital Research Institute and cultured at Ottawa hiPSC core facility, was used in this study [25]. Use of hiPSCs was approved by Ottawa Hospital Ethics Committee. hiPSCs were maintained in E8 media containing DMEM/F-12 (Life Technologies, 11330-057), ascorbic acid (64 mg/L, Sigma, A8960-5G), sodium bicarbonate (543 mg/L, Sigma, S-5761), human recombinant insulin (20 mg/L, Wisten, 511-016-CM), sodium selenite (4.2 μg/L, Sigma, S5261-10G), human holo-transferrin (10.7 mg/L, Sigma, T0665-1G), bFGF (100 μg/L, Life Technologies, PHG0263), recombinant human transforming growth factor β1 (2 μg/L, Life Technologies, PHG9202), and gentamicin (Wisent, 450-135-XL) on Matrigel™-coated plastic plates (BD Biosciences, 354230) in a 10% CO₂, 5.0% O₂ incubator. hiPSCs were passaged approximately every 5 days or at 80% confluency using EDTA (BioBasic, EB0185).

hiPSCs were induced into neural stem cells using the commercially available induction media (Stem cell Technologies, MT21031CV). At 80% confluency, hiPSC-NSCs were mechanically dissociated into a single cell suspension by trituration followed by diluting the suspension with DMEM/F12. The cell pellet was resuspended in neural induction media supplemented with 10 μM ROCK inhibitor Y27632 (Stem cell Technologies, 72302), and plated on poly-L-Ornithine (Sigma, 72302)- and laminin (Thermo Fisher Scientific, CB 40232)-coated 24-well plates at 200,000 cells/cm² in a 5% CO₂ and room O₂ incubator. Medium changes were conducted daily with induction media for 7–9 days. After three passages, hiPSC-NSCs were then plated on Matrigel-coated 6-well plates in complete neural progenitor media (Stem cell Technologies, 05833). hiPSC-NSCs were maintained with daily medium changes and were passaged every week at 80,000 cells/cm² on Matrigel-coated plates.

Metformin and 5-bromo-2′-deoxyuridine in vitro treatment

To assess proliferation, hiPSC-NSCs were passaged and plated into neural progenitor media and the media were supplemented with 0, 50, or 200 μM metformin (Sigma, D150959-5G). Daily medium changes were performed. After 3 days, cells were incubated with 10 μM 5-bromo-2′-deoxyuridine, BrdU (Sigma, B9285-1G), in neural progenitor media for 8 h and fixed.

To assess differentiation, cells were passaged and plated into neural progenitor media and the media were changed to neuronal differentiation media (NDM) (DMEM/F12 supplemented with 1% N2 supplement (Thermo Fisher, 17502048), 20 ng/mL BDNF (PeproTech, 450-02), 1% B27 supplement (Thermo Fisher, 17504-044), and10% fetal bovine serum (Life Technologies, 12484010), and treated with 0, 50, or 200 μM metformin. A half medium change was performed 2 days later and the culture was fixed at 5 days following differentiation.

To determine the effects of metformin preconditioning in culture, cells were treated with metformin at the concentration of 50 μM with daily medium changes with neural progenitor media for one passage (5–6 days). Following metformin treatment, hiPSC-NSCs were passaged in the absence of metformin according to the proliferation or differentiation experiments described above.

ET-1 surgery

ET-1 (1 μg/μL, Abcam, AB120471-100UG) diluted in phosphate-buffered saline (PBS) was sonicated in a 4°C water bath and left on ice for the period of the surgery. After baseline behavioral testing, rats were anesthetized using 4%–5% isoflurane and 2% oxygen, and mounted to a stereotaxic frame. Total surgery time was ∼1 h. Body temperatures were monitored using a rectal thermometer and maintained at 36.5°C on a heating blanket. Bur holes were made at all three injection sites. The three injection coordinates were unilateral and opposite to the rat's dominant forelimb, which was determined using staircase baseline measurements. Injections were performed using a 10 μL syringe with a 26G cemented needle (Hamilton, 80366). The needle was positioned at following injection sites:

0.0 mm AP, ±3.0 mm ML and −1.7 mm DV

+2.3 mm AP, ±3.0 mm ML, and −1.7 mm DV

+0.7 mm AP, ±3.8 mm ML and −7.0 mm DV

AP: Anterior-Posterior, ML: Medial-Lateral, DV: Dorsal-Ventral

Before injection of ET-1, 1 μL/site, the needle was lowered an additional −0.1 mm DV for 1 min and raised back to position to create a pocket to receive the ET-1. The ET-1 was then injected at 0.25 μL/min over 4 min. Incision sites were closed by sutures and treated with bupivacaine immediately after surgery and 4 h postsurgery. Subcutaneous buprenorphine was administered as an analgesic agent.

Transplantation of hiPSC-NSCs

hiPSC-NSCs from P5 to P7 passaging were treated with 50 μM metformin or sterile water for 48 h before surgery. hiPSC-NSCs were lifted from plate with accutase and resuspended in 1 mL PBS, and counted using trypan blue. Cells were then centrifuged at 300 g for 5 min and resuspended in PBS at 100,000 cell/μL. For the sham transplant group, the above protocol was conducted on an empty Matrigel-coated six-well plate and brought to an equal amount of PBS. The surgery was performed over 45 min and was conducted similar to the ET-1 protocol with few adjustments. Immediately before the first injection, cells were resuspended using a pipette and 4 μL was drawn into a 10 μL syringe with a 26G cemented needle. Cells were injected into the two cortical sites (0.0 mm AP, ±3.0 mm ML, and −1.7 mm DV +2.3 mm AP, ±3.0 mm ML, and −1.7 mm DV) 7 days following stroke. This time point falls within the critical time window for optimal poststroke recovery [24,26]. One hundred fifty thousand cells were delivered per injection site, and 300,000 cells were injected in total per animal. To limit back flow, the needle was lowered −0.1 mm DV and left in place for 1 min. The needle was then raised to position and the injection was initiated at 200 nL/min for 3 min. The needle was raised +0.3 mm DV and left there for 4.5 more minutes of the injection at 200 nL/min. The needle was left in place for 2 min and removed over a 1-min period. For sham groups, the same surgery was performed with PBS treated as described above.

Cyclosporin A and BrdU injections

Cyclosporin A (BioShop Canada, CYC002.5) was used as an immunosuppressive agent to avoid immune rejection. Cyclosporin A was prepared in cremophor EL (Sigma, C5135) at 100 mg/mL. Dissolved cyclosporin A was then mixed with sterile PBS at a 1:5 dilution and administered the same day. Rats received subcutaneous injections of cyclosporin A 20 mg/kg daily starting 2 days before transplantation and administered every 2 days starting 1 week post-transplantation.

Tissue preparation

At 1 and 7 weeks post-transplantation, rats were anesthetized with 1 mL i.p. injection of sodium pentobarbital (65 mg/mL). Rats were perfused with 120 mL of 4°C PBS (20 mL/min for 5 min) and 120 mL of sterile filtered 4°C 4% paraformaldehyde (PFA, Sigma, 159127-500G) in PBS. Brains were then dissected and incubated in 4% PFA in PBS for 16 h, followed by storage in 30% sucrose solution containing 1% sodium azide (Fisher, 19038-1000) for at least 72 h. Samples were covered in optimum cutting temperature solution (VWR, 95057-838) and submerged into 2-methylbutane (Thermo Fisher Scientific, O3551-4) at −40°C to −50°C. Serial 30 μm sections were obtained using a cryostat (Leica Biosystems, CM1850) to encompass the entire stroke over 32 gelatin-coated slides. Samples were left to dry over an hour and stored at −80°C.

Western blot analysis

Total lysates from culture were obtained by lysing cells with lysis buffer (25 mM Tris, pH = 7.4, 10 mM NaCl, 2 mM EDTA, 1 mM EGTA 0.5% Triton-100, 10% glycerol containing 1 mM PMSF, 1 mM sodium orthovanadate, 20 mM sodium fluoride, 10 μg/mL aprotinin, and 10 μg/mL leupeptin) followed by sonication (3 strokes with 5-s length and 1-min interval) and centrifugation at 13,000 g for 15 min. Protein samples were mixed with sample buffer, boiled for 5 min at 85°C, and run on a 12% SDS gel. Primary antibodies: anti-Sox2 (EMD Millipore, ABE5603MI), anti-Oct-3/4 (1:2,000, Cell Signaling, 2890S), anti-HLA-A (1:1,000, Acris, SM2012P), and anti-GAPDH (1:50,000, EMD Millipore, CB1001) were suspended in 3% bovine serum albumin (BSA) in TBS-T. Membranes were incubated overnight at 4°C in primary antibody followed by five washes for 5 min in TBS-T and a 1-h incubation in secondary antibody: anti-rabbit HRP (Cell Signaling technologies, 7074), and anti-mouse-HRP (Cell Signaling Technologies, 7076).

Immunocytochemistry and immunohistochemistry

Cells and brain sections were fixed in 4% PFA followed by three 5-min washes with PBS, and blocked and permeabilized in 10% normal goat serum (Jackson Immunoresearch, 008-000-121) or 3% BSA in PBS with 0.3% triton X-100 (Fisher Scientific, 9002-93-1) for at least 1 h. Samples were incubated in primary and secondary antibodies as described below.

Primary antibodies anti-Ku80 (1:100, Cedarlanes Labs, Y40400), anti-β III tubulin (1:500, Covance, 802001 and EMD 1:500, Millipore, MAB1637), mouse anti-Ki67 (1:200, BD Pharmingen, 550609), rabbit anti-Ki67 (1:200, Millipore, AB9260), anti-BrdU (1:400, AbD Serotec, OBT0030G), anti-Nestin (1:1,000, EMC Millipore, MAB5326), anti-Pax6 (1:500, BioLegend, 901301), anti-Oct-4 (1:1,000, Cell Signaling, 2890S), anti-GFAP (1:1,000, abcam, ab7260), anti- Sox2 (1:100, EDM Millipore, AB5603MI), and anti-Olig2 (1:1,000, EMD Millipore, AB9610) were incubated with brain sections overnight at 4°C. Three 5-min washes in PBS were performed followed by a 1-h incubation in Alexa Fluor-conjugated secondary antibodies: goat or donkey anti-rabbit Alexa Fluor^® 555, (1:500, Cell signaling Technology, 4413S), goat or donkey anti-mouse Alexa Fluor^® 555 (1:500, Cell signaling Technology, 4409), donkey anti-rabbit Alex Fluor 488 (1:500, Thermo Fisher Scientific, a31573), goat or donkey anti-mouse Alexa Fluor^® 488 (1:500, Cell signaling Technology, 4408S), and anti-mouse Alexa Fluor^® 647 (1:500, Cell signaling Technology, 4410). Samples were counterstained with Hoechst 33342 (Cell Signaling, 4082) at 1ug/mL in PBS for 5 min followed by three 5-min washes in PBS. Samples were mounted and imaged.

Imaging and quantification

Fluorescent images were taken on a Zeiss Axioplan 2 fluorescent microscope using Zeiss Axiovision software. Images for in vitro quantification were taken at random with at least four images per treatment with Z-stacking of 10–20 images at 1 μm apart. Counting was performed blinded to experimental groups using FIJI software. For in vivo quantification, all hKu80-positive cells were imaged. A total of hKu80⁺ cells per rat were calculated by quantifying number of cells per section and multiplying by the number of slides per series. Images were captured using an LSM800 confocal microscope with Zeiss Zen Pro software.

Staircase test

Rats were food restricted to 12 g/rat the night before beginning staircase training, which continued throughout training and testing days. To habituate rats to the staircase pellets, they received ∼50 sucrose pellets/cage/night for the first week of training. To measure unilateral forelimb coordination, rats were placed in clear Plexiglass staircase boxes with two descending staircases baited with 45 mg sucrose pellets (TestDiet, 1811155-5TUL) in 3 mm deep wells for 15 min twice a day, once in the morning and once in the afternoon, in a dark room with a fan to provide white noise. The staircases consist of seven steps with three sucrose pellets on each step for a total of 21 pellets per staircase. The plexiglass box is narrow enough to prevent the rats from reaching the staircase with the opposite paw. Poor sensorimotor function is reflected in the inability to reach, grasp, and eat the pellets when food deprived. Rats were habituated for 30 min in the staircase room before testing. The training period spanned 26 trials and rats that did not reach and consume at least 15 pellets with a standard deviation of ±2 pellets in the last 4 trials were excluded from the staircase analysis. Testing intervals included 3 days of testing where only the last four trials would be scored. Rats were scored on the number of pellets eaten at 1, 2, 4, 6, and 8 weeks post-stroke.

Beam walking test

Rats were food deprived as described for the staircase test and left to habituate for 30 min in the testing room under room lighting. One day of training was performed before baseline testing, where rats would be incrementally placed at a greater distance from a dark box containing sucrose pellets along a gradually tapered 1 m long two-tiered beam. This was performed until rats could run across the beam toward the box without pausing or falling off the beam. On testing days, which took place before stroke and 1, 2, 4, 6, and 8 weeks post-stroke, at least three trials were recorded and three trials with no or limited pausing were analyzed. Missed steps were defined as any time the forelimb or hindlimb of the rat did not contact the top of the beam during a step.

Cresyl violet staining

Sections were submerged in 0.015% cresyl violet solution (pH 3.5) for 20–30 min. Samples were then dehydrated in sequential baths of 70%, 95%, and 100% ethanol baths and submerged in clearing agent, citrisolv (Fisher, 22-143-975). Slides were mounted with permount (Fisher, FL-10-0505) and imaged.

Infarct volume measurements

Cresyl violet-stained sections were imaged using Aperio digital pathology slide scanner (Leica Biosystems) at 20 × resolution. Images were analyzed using Imagescope software (Leica Biosystems) and FIJI. Infarct area per section was measured and total infarct volume was calculated using the following equation: Infarct Volume = Σ(Intact contralateral cortex/striatum − Intact ipsilateral contex/striatum) × 0.03 mm (thickness of each section) × 7 (number of sections)

Statistical analysis

Statistical analysis was conducted using GraphPad Software. Behavioral data were analyzed using a two-way analysis of variance (ANOVA) and Bonferonni post-hoc analysis. Single variable data were analyzed using a one-way ANOVA with Tukey multiple comparisons or two-tailed paired t-test. All data were analyzed using a significance level of α = 0.05 and all values are expressed as mean ± standard error of the mean.

Results

Generation of human NSCs from a human iPSC line

We generated human NSCs from an hiPSC line (WLS-1C) using a Neural Induction Kit (Stem Cell Technology). hiPSCs were successfully differentiated to Pax6⁺ cortical precursors 1 week after neural induction with 97% of cells expressing Pax6 and small clusters of Oct-3/4-positive undifferentiated cells (Fig. 1a–c). Pax6-positive cells often formed groups resembling neural rosettes, which persisted until passage 8 and 9 (Fig. 1a, b). Further immunocytochemistry at later passages identified these cells largely expressing neural stem cell markers, Pax6, Sox2, and Nestin (Fig. 1d, e). Sox2 was expressed in both hiPSCs and hiPSC-NSCs. To discriminate Sox2 expression levels in hiPSCs and hiPSC-NSCs, we performed a western protein analysis, which qualitatively showed that the expression of Sox2 in NSCs was much lower than those in hiPSCs (Fig. 1f). Oct-3/4 protein expression levels were only detected in hiPSCs, but not in hiPSC-NSCs (Fig. 1f and Supplementary Fig. S1a; Supplementary Data available online at www.liebertpub.com/scd).

FIG. 1.

Generation of human neural stem cells (NSCs) from human iPSCs. (a) Photographs of live hiPSCs, hiPSCs undergoing induction in P1, and hiPSC-NSCs after induction in P5. (b, c) Immunocytochemistry and quantification of hiPSC-NSCs at passage 1 for neural stem marker, Pax6 (red), and pluripotent marker, Oct3/4 (green), counterstained with Hoechst (blue). Scale bar: 20 μm. (d) Immunocytochemistry of hiPSC-NSCs from passage 5 to 9 for neural stem markers, Pax6 (green), Nestin (red), and Sox2 (red). Scale bar: 20 μm. (e) Quantification of neural and pluripotent markers as the percentage of total live cells counterstained with Hoechst (blue) (n = 3 independent experiments). (f) Western blot analysis of cultured hiPSCs and hiPSC-NSCs for Oct3/4 and Sox2, using GAPDH as a loading control. *p < 0.05. hNSCs, human induced pluripotent stem cell-derived neural stem cells.

Metformin enhances proliferation and differentiation of hiPSC-NSCs in vitro

Previous studies have already demonstrated that metformin treatment is able to increase the proliferation and differentiation of adult murine SVZ neural stem and progenitor cells (NPCs) [14,16]. In this study, we proposed to determine whether similar functional roles of metformin also occur in hiPSC-NSCs. To assess proliferation, hiPSC-NSCs were plated into neural proliferation medium in the absence and presence of metformin (50 or 200 μM) followed by BrdU pulse-labeling (8 h) incubation (Fig. 2a). Our quantification analysis showed a significant increase in the proportion of Ki67⁺/BrdU⁺ double-labeled proliferating cells in 50 μM metformin-treated group compared to control (Fig. 2b, c). To determine whether reduced apoptosis could contribute to increased proliferation rate caused by metformin treatment, we immunostained hiPSC-NSCs with apoptotic marker, cleaved caspase 3 (CC3), following metformin treatment (Supplementary Fig. S2a, b). Quantification of CC3⁺ cells with condensed nuclei showed no alteration in apoptosis in metformin-treated groups (50 and 200 μM). To assess neuronal differentiation, hiPSC-NSCs were plated into NDM and were treated in the absence and presence of metformin (50 or 200 μM) (Fig. 2a). At 5 days following differentiation, the proportion of β III tubulin⁺ newborn neurons were quantified in cultured hiPSC-NSCs. Both 50 and 200 μM metformin treatment significantly increased the population of β III tubulin⁺ neurons compared to control group (Fig. 2d, e).

FIG. 2.

Metformin enhances proliferation and differentiation of hiPSC-NSCs in vitro. (a) Experimental outline for in vitro analysis. (b) Photographs of hiPSC-NSCs immunostained for Ki-67 and BrdU after 3 days of metformin treatment (control and 50 and 200 μM) under a proliferation condition. Arrows denote Brdu and Ki67 double-labeled human NSCs inculture. (c) Quantification of the percentage of Ki-67⁺/BrdU⁺ cells over total live cells in culture. (d) Photographs of cultured hiPSC-NSCs immunostained for β III tubulin 5 days after metformin treatment (control and 50 and 200 μM) under a neuronal differentiation condition. (e) Quantification of the percentage of β-III tubulin⁺ neurons over total live cells in culture. (f) Experimental outline for in vitro preconditioning paradigm analysis. (g) Photographs of hiPSC-NSCs immunostained for Ki-67 and BrdU and counterstained with Hoechst, under the proliferation condition 3 days after passaging from control/metformin-preconditioned (50 μM metformin for 5 days) groups. (h) Quantitative analysis of the percentage of Ki-67⁺/BrdU⁺ proliferating cells out of total live cells. (i) Photographs of cultured hiPSC-NSCs under the neuronal differentiation condition for 5 days, immunostained for β III tubulin and counterstained with Hoechst, after passaging from control/metformin-preconditioned (50 μM metformin for 5 days) groups. (j) Quantitative analysis of the percentage of β III tubulin⁺ newborn neurons out of total live cells. *P < 0.05; **P < 0.01 (n = 3–4 independent experiments). Scale bar: 20 μm.

Since we found that metformin at the concentration of 50 μM can significantly promote both proliferation and differentiation of hiPSC-NSCs (Fig. 2a–e), we then selected 50 μM as the optimal preconditioning concentration to pretreat hiPSC-NSCs before assessing their proliferation and differentiation (Fig. 2f). Thus, hiPSC-NSCs were treated in the absence and presence of metformin for one passage after plating (5–6 days), and then were passaged and cultured in proliferation medium or differentiation medium in the absence of metformin (Fig. 2f). Quantitative analysis showed that hiPSC-NSCs preconditioned with 50uM metformin exhibited a significant increase in the number of Ki67⁺/BrdU⁺ co-labeled proliferating cells upon proliferation, as well as in the population of β III tubulin⁺ neurons upon differentiation (Fig. 2f–j). To determine whether metformin preconditioning would induce hiPSC-NSC differentiation under the proliferation condition, we quantified the percentage of β III tubulin⁺ neurons from cultured hiPSC-NSCs under the proliferation condition in the absence and presence of metformin. The analysis showed no difference in the number of β III tubulin⁺ newborn neurons between control and 50 μM metformin (Supplementary Fig. S2c, d). Thus, metformin is able to enhance the proliferation and differentiation of hiPSC-NSCs through direct and preconditioning experimental paradigms.

Metformin preconditioning before transplantation improves hiPSC-NSC survival 1 week post-transplantation

To determine whether metformin preconditioning of hiPSC-NSCs in culture before transplantation improves their initial engraftment in a rat ET-1 stroke model, we transplanted hiPSC-NSCs, which were pretreated with water (naive group) or 50 μM metformin (preconditioned group) for 48 h, into the stroke-damaged cortex. The 48-h preconditioning time window was chosen because cultured cells reached their 100% confluence 2 days following metformin treatment, which was an optimal condition to harvest hiPSC-NSCs for transplantation. To generate the ET-1 focal ischemic stroke, rats were subjected to two cortical and one striatal injection of ET-1 to induce focal ischemia in the forelimb motor cortex and dorsolateral striatum. To avoid peak of cell death following ischemia, rats were scheduled for transplantation 1 week post-stroke. Starting from 2 days before transplant, we administered daily 20 mg/kg cyclosporin A (subcutaneous injections) to stroke rats until the end of experiment. Rats were sacrificed 1 week post-transplantation (2 weeks post-stroke) (Fig. 3a). We did not observe the altered infarct volume between naive and preconditioned groups (Fig. 3b, c).

FIG. 3.

Metformin preconditioning before transplantation improves hiPSC-NSC survival 1 week post-transplantation in a rat ET-1 stroke model. (a) Time line for in vivo transplantation of hiPSC-NSCs into a rat focal ischemic ET-1 stroke model. (b) Serial brain sections stained by cresyl violet. Scale bar: 2 mm. (c) Quantitative analysis of cortical and striatal infarct volume using cresyl violet staining (naive n = 6 and preconditioned n = 6). (d) Photographs of brain sections immunostained for Ku80, a human nuclear marker, in both the naive and preconditioned cell groups 1 week post-transplantation. Scale bar: 40 μm. Arrows denote engrafted human cells that are positive for hKu80. (e) Quantitative analysis of Ku80⁺ cells per rat in naive and preconditioned cell groups 1 week post-transplantation (n = 6), *P < 0.05. (f) Photographs of brain sections immunostained for Ku80 and Ki67, a marker for proliferating cells, in both the naive and preconditioned cell groups 1 week post-transplantation. Scale bar: 40 μm. Arrows denote proliferating human cells that are positive for both Ki67 and hKu80. (g) Quantitative analysis of the percentage of Ki67 out of total Ku80⁺ cells in naive and preconditioned cell groups 1 week post-transplantation (n = 5 naive and n = 6 preconditioned). (h) Representative western blot gel images of total lysates from control and metformin-preconditioned (50 μM metformin for 5 days) hiPSC-NSCs, probed for HLA-A, using GAPDH a loading control. (i) Quantitative analysis of HLA-A expression levels, normalized to control *P < 0.05 (n = 3 independent experiments). ET-1, endothelin-1.

Transplanted human cells were detected using anti-human Ku80 antibody (also called STEM101). Quantification analysis revealed that metformin preconditioning led to a significant increase in the number of grafted cells from 364 ± 152 (naive group) to 1,037 ± 228 (preconditioned group), suggesting improved initial cell survival following transplantation (Fig. 3d, e). Human Ku80⁺ cells were always located within the infarct core. While one of six rats from the naive group did not have any Ku80⁺ cells, all rats from the preconditioned group showed detectable grafts. We further performed immunohistochemistry for Ki67, a marker for proliferating cells, in the two groups of brain sections. 19% ± 6% of total naive Ku80⁺cells expressed Ki67, whereas 33% ± 8% of preconditioned Ku80⁺cells expressed Ki67. These data showed a nonsignificant trend toward an increase in proliferation (Fig. 3f, g). To test whether other underlying mechanisms may contribute to enhanced initial engraftment of preconditioned cells, we examined HLA-A protein expression in hiPSC-NSCs treated with metformin in culture. Metformin treatment significantly reduced HLA-A levels (Fig. 3h, i; Supplementary Fig. S1b), which may lead to less immune rejection of preconditioned cells following transplantation, contributing to increased initial engraftment of preconditioned cells.

Metformin preconditioning before transplantation improves hiPSC-NSC survival and neural regeneration 7 weeks post-transplantation

To assess the long-term engraftment of hiPSC-NSCs after stroke, we sacrificed a cohort of rats from sham transplantation (PBS), naive, and preconditioned groups at 7 weeks post-transplantation (Fig. 4a). First, we assessed cortical and striatal infarct volumes from three groups of rats through cresyl violet staining. A trend toward reduced cortical infarct size in the preconditioned group was observed compared to the sham transplant group (Fig. 4b, c). Interestingly, when we compared stroke volumes between 1 and 7 weeks post-transplantation (2 and 8 weeks post-stroke), a significant decrease in cortical stroke volume was observed in the preconditioned group, but not in the naive group (Fig. 4d).

FIG. 4.

Metformin preconditioning before transplantation improves hiPSC-NSC survival and neural regeneration 7 weeks post-transplantation in a rat ET-1 stroke model. (a) Time line for in vivo transplantation of hiPSC-NSCs in rat focal ischemic ET-1 stroke model. Rats were acclimatized for 1 week and handled for 1 week before commencing behavioral training. (b) Cresyl violet staining of representative cortical infarcts of sham, naive, and preconditioned groups. Scale bar: 0.5 mm. (c) Quantitative analysis of cortical and striatal infarct volumes using cresyl violet staining. No difference among groups in cortical (sham n = 9, naive n = 10, and preconditioned n = 10) or striatal volumes (sham n = 5, naive n = 8, and preconditioned n = 6). Red highlighted markers correspond to rats with grafts at 7 weeks post-transplantation. (d) Quantitative analysis of cortical and striatal infarct volumes for samples from 1 week and 7 weeks post-transplantation. *P < 0.05. (e) Photographs of brain sections immunostained for Ku80, a human nuclear marker, in a preconditioned group 7 weeks post-transplantation. (f) Quantitative analysis of Ku80⁺cells per rat in naive and preconditioned cell groups 7 weeks post-transplantation. Scale bar: 40 μm. (g) Photographs of brain sections immunostained for β III tubulin, GFAP, and Olig2 7 weeks post-transplantation. Arrows denote differentiated human neuron (top), astrocyte (middle), and oligodendrocyte (bottom). Scale bar: 50 μm (h) Quantitative analysis of the percentage of β III tubulin⁺, GFAP⁺, and Olig2⁺ cells over total human Ku80⁺ cells 7 weeks post-transplantation for naive and preconditioned groups (n = 3–6 sections).

To determine whether surviving transplants contribute to reduced stroke volumes, we performed hKu80 immunohistochemistry on all sections spanning the cortical infarct. Successful cell grafts at 7 weeks post-transplantation were observed in 20% of rats (2 out of 10) from the preconditioned group compared to 10% of rats (1 out of 10) from the naive group. All three grafts appeared as an aggregate of cells filling the infarct core, bordering the peri-infarct area (Fig. 4e; Supplementary Fig. S3a). Interestingly, the number of grafted cells from the naive group was much higher than those from the preconditioned group (Fig. 4f). However, the grafted cells from the preconditioned group exhibited a higher extent of differentiation into three neural cell lineages compared to those from the naive group (Fig. 4g, h; Supplementary Fig. S3b). Following immunohistochemistry for neuronal, astrocytic, and oligodendrocytic markers, β III tubulin, GFAP, and Olig2 respectively, we observed that grafts from the preconditioned group contained an average of 36% β III tubulin, 24% GFAP, and 7% Olig2, while grafts from the naive group contained dramatically fewer differentiated cells (4.88% β III tubulin, 3.9% GFAP, and 0% Olig2) (Fig. 4g, h). Associated with the enhanced neural differentiation, we also observed that grafted cells from the preconditioned group contained a lower percentage of Ki67-positive proliferating precursors when compared to grafted naive cells (Supplementary Fig. S3c, d). These results suggest that metformin preconditioning pushes grafted human NPCs toward differentiation at the expense of their proliferation at a later time point following transplantation (7 weeks post-transplantation).

Cell transplantation accelerates gross motor functional recovery, but not fine motor recovery

Several studies showed that transplantation of hiPSC-NSCs into the striatum following MCAO ischemic stroke promotes functional recovery in the beam walking and staircase task [10,27,28]. To determine whether transplantation of naive or metformin-preconditioned hiPSC-NSCs promotes functional recovery during the chronic phase of a focal ischemic stroke, we performed behavioral assessment of sensorimotor function up to 8 weeks post-stroke using two tests, beam walking and Montoya staircase tasks (Fig. 4a).

The beam walking task focused on both forelimb and hindlimb gross motor function. We found that rats that were transplanted with metformin-preconditioned cells resulted in rapid recovery on contralateral forelimb (injured forelimb) at 4 weeks post-stroke, while the sham-transplanted group showed persistent deficits on contralateral forelimb, and the group receiving naive cells exhibited a more gradual recovery curve (Fig. 5a, b). Similar to contralateral forelimb, the preconditioned group also had faster recovery on contralateral hindlimb than both sham and naive groups.

FIG. 5.

Cell transplantation accelerates recovery of gross, but not fine, motor skills following an ET-1 ischemic stroke. (a, b) Quantitative analysis of percent successful steps reached, normalized to baseline on the contralateral forelimbs (a) and hindlimbs (b). Two-way ANOVA, contralateral forelimb: time x treatment interaction, F (10,130) = 1.015, P = 0.43; post-hoc multiple comparison, ***P < 0.001, **P < 0.01, *P < 0.05 compared to baseline for each treatment, n = 9–10; contralateral hindlimb time x treatment interaction, F (10,130) = 1.085, P = 0.37; post-hoc multiple comparison, ***P < 0.001, **P < 0.01, *P < 0.05 compared to baseline for each treatment, n = 9–10. (c) Quantitative analysis of average pellets eaten on the contralateral side. Significant difference by time (***P < 0.001, F (5,120) = 32), but not between groups. ANOVA, analysis of variance; PBS, phosphate-buffered saline.

The staircase task was used to measure fine forelimb motor skills. All three cohorts of rats were trained for 13 days (2 trials per day) to acquire a consistent reaching ability (Supplementary Fig. S4). No difference in acquisition was observed among groups. Of rats that successfully learned the staircase task, a deficit of ∼60% in pellets eaten was observed in all groups, indicating consistent and significant behavioral deficits among groups (Fig. 5c). There was no significant difference between groups after cell transplantation in pellets eaten (Fig. 5c). No spontaneous recovery was observed in pellets eaten between 1 week post-stroke and 8 weeks post-stroke, suggesting a major impairment of fine motor function. These results suggest that metformin preconditioning of hiPSC-NSCs accelerates the gross motor function recovery, but not fine motor skills.

Discussion

In this study, we report multiple beneficial effects of FDA-approved drug metformin preconditioning of hiPSC-NSCs before cell transplantation in a rat ET-1 focal ischemic stroke model. First, we show that metformin treatment promotes proliferation and differentiation of hiPSC-NSCs through both direct stimulation and a preconditioning paradigm in culture. Furthermore, metformin preconditioning of hiPSC-NSCs before transplantation increases cell engraftment 1 week post-transplantation and improves the rate of surviving grafts 7 weeks post-transplantation. In addition, rats transplanted with preconditioned cells show reduced stroke infarct sizes during the chronic phase of stroke. Importantly, rats receiving preconditioned cells show more rapid recovery of gross motor function. However, there was no benefit of the transplants on recovery of fine motor skills in the Montoya staircase task. In summary, our studies revealed that metformin preconditioning is an effective and vital approach to optimize cell transplantation for stroke and provides limited functional benefit on motor function.

Metformin promotes the proliferation and differentiation of human iPSC-NSCs in culture

Previous studies have shown that metformin promotes both proliferation and differentiation of primary murine adult NSCs [14,16]. We have extended these findings on the neuroregenerative capability of metformin in hiPSC-NSCs. It is interesting to note that metformin concentrations we used in this experiment for hiPSC-NSCs and other studies for hESC-NSCs and hiPSCs were much higher than those used in primary rodent NSCs. This could potentially be due to differential expression levels of metformin's cell surface transporter organic cation transport 1 (OCT1) [29], or account for the different metabolic regulation in both rodent and human cells since the direct target of metformin in the cells is to inhibit the mitochondrial complex 1.

Metformin preconditioning promotes cell engraftment 1 week post-transplantation

Using a focal ET-1 stroke model, we showed that metformin could be used as a preconditioning agent to enhance cell engraftment shortly following transplants. Since over 90% of cells die 4 days following transplantation [30], the remaining cell population at 1 week post-transplantation best represents an early time point of the surviving graft. The increase in cell engraftment 1 week post-transplantation could be due to a multitude of factors. First, the increase could be attributable to the increase in proliferation rate of hiPSC-NSCs. Although we could not detect a statistical significance, there is a strong trend toward increased proliferation in the metformin-preconditioned group. In addition, the immunomodulatory effect of metformin on hiPSC-NSCs by reducing their HLA-A protein levels may reduce immune rejection of hiPSC-NSCs engrafting in the injured rat brain. A previous study reported that preconditioning of rats with metformin before ischemia reduced apoptotic cells in the peri-infarct region [31]. Other studies also show that metformin increases cell survival in cytotoxic environments through stimulation of autophagy flux [32]. Thus, metformin likely promotes the engraftment of hiPSC-NSCs through a combination of these mechanisms.

Although metformin preconditioning promoted cell engraftment, the engrafted cell number observed in both the naive and preconditioned groups was much less than other neural stem cell transplant studies [10,11,33]. A possible explanation is that cyclosporin A, an immunosuppressant agent, may not reach optimal concentration in rats to fully inhibit the adaptive immune system through inactivation and inhibition of proliferation of T cells [34]. Partial inactivation of T cells could result in extensive rejection of xenografted cells, which indeed provides a valid model for assessing metformin preconditioning as an optimized approach that enhances cell engraftment for brain regeneration following ischemic stroke.

Since the size of stroke may influence cell engraftment rates, we examined stroke volumes between naive and preconditioned groups 1 week post-transplantation and found no significant difference between groups. These data suggest that metformin preconditioning-increased cell engraftment is independent of stroke size. On the other hand, it also suggests that increased cell engraftments by metformin preconditioning were not able to change stroke infarct volume within the short term (1 week post-transplantation).

Metformin preconditioning promotes cell regeneration 7 weeks post-transplantation

Although paracrine effects can trigger remodeling and promote early recovery, the ultimate goal of cell-based therapy is to allow long-term survival of grafts, which can be differentiated and integrated into new and existing neural circuits. In the previous studies, many reports showed that only a fraction of rats contained graft cells months after transplantation of hiPSC-NSCs [9,10]. We observed the same phenomenon. Intriguingly, we discovered that metformin preconditioning increases the proportion of rats that contained long-term cell grafts compared to the naive group. In addition, grafted cells that survived until at least 7 weeks from the preconditioned group represent a 23-fold increase in cell numbers compared to those at 1 week post-transplantation (1,037 cells/rat at 1 week post-transplantation vs 23,023 cells/rat at 7 weeks post-transplantation), suggesting that exogenous hiPSC-NSCs have expanded since initial engraftment. Importantly, metformin-preconditioned cell grafts at 7 weeks post-transplantation showed higher percentages of differentiated neural cells, including newborn neurons, astrocytes, and oligodendrocytes, but lower percentage of proliferating cells than naive cell grafts, associated with less hiPSC-NSC expansion. Thus, the higher rate of successful engraftment and differentiation from metformin-preconditioned grafts suggests that metformin preconditioning could be used to optimize neural stem cell transplantation.

Interestingly, at the functional level, metformin-preconditioned grafts were able to accelerate functional recovery of a gross motor skill measured by a beam walking task, but not a fine motor skill measured by a staircase task. This limited recovery on the fine motor skill from metformin-preconditioned group may be due to (1) low cell engraftment at 1 week post-transplantation, which could account for little early recovery through the paracrine effect; (2) the lack of integration of grafted cells into existing healthy brain tissue and neural circuits at a later time point may result in little potential for cell replacement mechanisms to promote recovery; and (3) typically, recovery of skilled reaching requires some form of rehabilitation (eg, post-stroke reach training) that specifically targets the forelimb impairments [24,35]. In contrast, rats use all four limbs to locomote and this provides a form of “self-rehabilitation,” which may have contributed to recovery on the beam task. To optimize future transplantation approaches to improve recovery, rehabilitation should be included as part of the stem cell therapy. In addition, there is need to optimize that transplantation approaches will focus on allowing exogenous hiPSC-NSCs to migrate and fully differentiate into mature neural cells in the injured brain, ultimately forming a neural network integrated with existing neural circuits.

Overall, our studies revealed that metformin preconditioning is an effective and promising approach to enhance long-term cell engraftment in the stroke-damaged brain.

Footnotes

Acknowledgments

This work was supported by HSF Canadian Partnership for Stroke Recovery (CPSR) catalyst grants to J.W., W.L.S., and D.C., and CPSR Trainee Award to F.O.-B. We thank Nicolay Hristozov and Sudhir Karthik for their technical assistance in behavioral analysis and brain tissue preparation.

Author Disclosure Statement

No competing financial interests exist.

References

Krueger

, Koot

, Hall

, O'Callaghan

, Bayley

and Corbett

. (2015). Prevalence of individuals experiencing the effects of stroke in Canada: trends and projections. Stroke, 46:2226–2231.

Takahashi

, Ohnuki

, Narita

, Ichisaka

, Tomoda

and Yamanaka

. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131:861–872.

Wang

, Chao

, Han

, Zhang

, Xi

, Li

, Jiang

, Wang

, Yu

, Tian

and Zhang

. (2013). PET demonstrates functional recovery after transplantation of induced pluripotent stem cells in a rat model of cerebral ischemic injury. J Nucl Med, 54:785–792.

Chen

, Chang

, Tsai

, Chang

, Chou

, Huang

, Tai

, Chen

, Ku

, Li

and Chiou

. (2010). Functional improvement of focal cerebral ischemia injury by subdural transplantation of induced pluripotent stem cells with fibrin glue. Stem Cells Dev, 19:1757–1767.

Kawai

, Yamashita

, Ohta

, Deguchi

, Nagotani

, Zhang

, Ikeda

, Matsuura

and Abe

. (2010). Tridermal tumorigenesis of induced pluripotent stem cells transplanted in ischemic brain. J Cereb Blood Flow Metab, 30:1487–1493.

Yamashita

, Kawai

, Tian

, Ohta

and Abe

. (2011). Tumorigenic development of induced pluripotent stem cells in ischemic mouse brain. Cell Transplant, 20:883–891.

Chang

, Lee

, Park

, Choi

, Jeon

, Kwon

, Oh

, Shin

, Do

, et al. (2013). Therapeutic potential of human induced pluripotent stem cells in experimental stroke. Cell Transplant, 22:1427–1440.

Chau

, Deveau

, Song

, Gu

, Chen

and Wei

. (2014). iPSC Transplantation increases regeneration and functional recovery after ischemic stroke in neonatal rats. Stem Cells, 32:3075–3087.

Jensen

, Yan

, Krishnaney-Davison

, Al Sawaf

and Zhang

. (2013). Survival and differentiation of transplanted neural stem cells derived from human induced pluripotent stem cells in a rat stroke model. J Stroke Cerebrovasc Dis, 22:304–308.

10.

Oki

, Tatarishvili

, Wood

, Koch

, Wattananit

, Mine

, Monni

, Tornero

, Ahlenius

, et al. (2012). Human-induced pluripotent stem cells form functional neurons and improve recovery after grafting in stroke-damaged brain. Stem Cells, 30:1120–1133.

11.

Tornero

, Wattananit

, Gronning

, Koch

, Wood

, Tatarishvili

, Mine

, Ge

, Monni

, et al. (2013). Human induced pluripotent stem cell-derived cortical neurons integrate in stroke-injured cortex and improve functional recovery. Brain, 136:3561–3577.

12.

Moshayedi

, Nih

, Llorente

, Berg

, Cinkornpumin

, Lowry

, Segura

and Carmichael

. (2016). Systematic optimization of an engineered hydrogel allows for selective control of human neural stem cell survival and differentiation after transplantation in the stroke brain. Biomaterials, 105:145–155.

13.

Wei

, Wei

, Jiang

, Mohamad

and Yu

. (2017). Stem cell transplantation therapy for multifaceted therapeutic benefits after stroke. Prog Neurobiol, 157:49–78.

14.

Fatt

, Hsu

, He

, Wondisford

, Miller

, Kaplan

and Wang

. (2015). Metformin acts on two different molecular pathways to enhance adult neural precursor proliferation/self-renewal and differentiation. Stem Cell Rep, 5:988–995.

15.

Liu

, Tang

, Zhang

, Wang

and Yang

G-Y

. (2014). Metformin promotes focal angiogenesis and neurogenesis in mice following middle cerebral artery occlusion. Neurosci Lett, 579:46–51.

16.

Wang

, Gallagher

, DeVito

, Cancino

, Tsui

, He

, Keller

, Frankland

, Kaplan

and Miller

. (2012). Metformin activates an atypical PKC-CBP pathway to promote neurogenesis and enhance spatial memory formation. Cell Stem Cell, 11:23–35.

17.

Koizumi

, Yoshida

, Nakazawa

and Ooneda

. (1986). Experimental studies of ischemic brain edema. Nosotchu, 8:1–8.

18.

Kanemitsu

, Nakagomi

, Tamura

, Tsuchiya

, Kono

and Sanoet

. (2002). Differences in the extent of primary ischemic damage between middle cerebral artery coagulation and intraluminal occlusion models. J Cereb Blood Flow Metab, 22:1196–1204.

19.

Hossmann

. (2012). The two pathophysiologies of focal brain ischemia: implications for translational stroke research. J Cereb Blood Flow Metab, 32:1310–1316.

20.

Dittmar

, Spruss

, Schuierer

and Horn

. (2003). External carotid artery territory ischemia impairs outcome in the endovascular filament model of middle cerebral artery occlusion in rats. Stroke, 34:2252–2257.

21.

, Omae

and Fisher

. (1999). Spontaneous hyperthermia and its mechanism in the intraluminal suture middle cerebral artery occlusion model of rats. Stroke, 30:2464–2470.

22.

Corbett

, Carmichael

, Murphy

, Jones

, Schwab

, Jolkkonen

, Clarkson

, Dancause

, Weiloch

, et al. (2017). Enhancing the alignment of the preclinical and clinical stroke recovery research pipeline: consensus-based core recommendations from the Stroke Recovery and Rehabilitation Roundtable translational working group. Int J Stroke, 12:462–471.

23.

Somaa

, Wang

, Niclis

, Bruggeman

, Kauhausen

, Guo

, McDougall

, Williams

, Nisbet

, Thompson

and Parish

. (2017). Peptide-based scaffolds support human cortical progenitor graft integration to reduce atrophy and promote functional repair in a model of stroke. Cell Rep, 20:1964–1977.

24.

Murphy

and Corbett

. (2009). Plasticity during stroke recovery: from synapse to behaviour. Nat Rev Neurosci, 10:861–872.

25.

Chang

, Lavoie

, Kwon

, Chen

, Manias

, Behbahani

, Ling

, Kandel

, Stewart

and Stanford

. (2013). Feeder-independent derivation of induced-pluripotent stem cells from endothelial progenitor cells. Stem Cell Res, 10:195–202.

26.

Biernaskie

, Chernenko

and Corbett

. (2004). Efficacy of rehabilitative experience declines with time after focal ischemic brain injury. J Neurosci, 24:1245–1254.

27.

Eckert

, Huang

, Gonzalez

, Kim

, Hamblin

and Lee

. (2015). Bystander effect fuels human induced pluripotent stem cell-derived neural stem cells to quickly attenuate early stage neurological deficits after stroke. Stem Cells Transl Med, 4:841–851.

28.

Yuan

, Liao

, Feng

, Lou

, Niu

, Zhang

, Wang

and Deng

. (2013). Human induced pluripotent stem cell-derived neural stem cells survive, migrate, differentiate, and improve neurologic function in a rat model of middle cerebral artery occlusion. Stem Cell Res Ther, 4:73.

29.

Apáti

, Szebényi

, Erdei

, Várady

, Orbán

and Sarkadi

. (2016). The importance of drug transporters in human pluripotent stem cells and in early tissue differentiation. J Expert Opin Drug Metab Toxicol, 12:77–92.

30.

Sortwell

, Pitzer

and Collier

. (2000). Time course of apoptotic cell death within mesencephalic cell suspension grafts: implications for improving grafted dopamine neuron survival. Exp Neurol, 165:268–277.

31.

Jiang

, Yu

, Zhu

, Wang

, Tan

, Cao

, Zhang

, Gao

, Shi

, Zhang

and Tan

. (2014). Acute metformin preconditioning confers neuroprotection against focal cerebral ischaemia by pre-activation of AMPK-dependent autophagy. Br J Pharmacol, 171:3146–3157.

32.

Zhang

, Xuan

, Zheng

, Zhou

, Lin

, Wu

, Zhou

, Huang

, Wang

, et al. (2017). Metformin improves functional recovery after spinal cord injury via autophagy flux stimulation. Mol Neurobiol, 54:3327–3341.

33.

Lam

, Lowry

, Carmichael

and Segura

. (2014). Delivery of iPS-NPCs to the stroke cavity within a hyaluronic acid matrix promotes the differentiation of transplanted cells. Adv Funct Mater, 24:7053–7062.

34.

Taylor

, Watson

CJE

and Bradley

. (2005). Immunosuppressive agents in solid organ transplantation: mechanisms of action and therapeutic efficacy. Crit Rev Oncol Hematol, 56:23–46.

35.

Jeffers

, Hoyles

, Morshead

and Corbett

. (2014). Epidermal growth factor and erythropoietin infusion accelerate functional recovery in combination with rehabilitation. Stroke, 45:1856–1858.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.10 MB

0.17 MB

0.08 MB