Tissue Engineering and Regenerative Medicine
TISSUE ENGINEERING AND REGENERATIVE MEDICINE
a
Center for Stem Cell
Research and Regenerative
Medicine, Department of
Neurology, Tulane University
School of Medicine, New
Orleans, Louisiana, USA;
b
Sanford-Burnham Institute
for Medical Research,
Neuroscience, Aging and
Stem Cell Research, La Jolla,
California, USA; cSeoul
National University, College
of Medicine, Department of
Pharmacology, Seoul,
Republic of Korea
Correspondence: Jean-Pyo Lee,
Ph.D., Tulane University School of
Medicine, Department of
Neurology, Center for Stem Cell
Research and Regenerative
Medicine, 1430 Tulane Avenue,
SL99, New Orleans, Louisiana
70112, USA. Telephone: 504-988-
7074; E-Mail: jeanpyol@tulane.
edu
Received September 1, 2014;
accepted for publication April 6,
2015; published Online First on
May 29, 2015.
©AlphaMed Press
1066-5099/2015/$20.00/0
http://dx.doi.org/
10.5966/sctm.2014-0184
STEM CELLS TRANSLATIONAL MEDICINE 2015;4:841–851 www.StemCellsTM.com
Bystander Effect Fuels Human Induced Pluripotent
Stem Cell-Derived Neural Stem Cells to Quickly
Attenuate Early Stage Neurological Deficits After
Stroke
AUSTON ECKERT,a LEI HUANG,a RODOLFO GONZALEZ,b HYE-SUN KIM,c MILTON H. HAMBLIN,a
JEAN-PYO LEEa,b
Key Words. Inflammation x Blood-brain barrier x Cellular therapy x Stem cell transplantation x
Induced pluripotent stem cells x Stroke x Neural stem cell
ABSTRACT
Present therapies for stroke rest with tissue plasminogen activator (tPA), the sole licensed antithrom-
botic on the market; however, tPA’s effectiveness is limited in that the drug not only must be admin-
istered less than 3–5 hours after stroke but often exacerbates blood-brain barrier (BBB) leakage and
increases hemorrhagic incidence. A potentially promising therapy for stroke is transplantation of hu-
man induced pluripotent stem cell-derived neural stem cells (hiPSC-NSCs). To date, the effects of iPSCs
on injuries that take place during early stage ischemic stroke have not been well studied. Conse-
quently, we engrafted iPSC-NSCs into the ipsilesional hippocampus, a natural niche of NSCs, at
24 hours after stroke (prior to secondary BBB opening and when inflammatory signature is abundant).
At 48 hours after stroke (24 hours after transplant), hiPSC-NSCs had migrated to the stroke lesion and
quickly improved neurological function. Transplanted mice showed reduced expression of proinflam-
matory factors (tumor necrosis factor-a, interleukin 6 [IL-6], IL-1b, monocyte chemotactic protein 1,
macrophage inflammatory protein 1a), microglial activation, and adhesion molecules (intercellular
adhesion molecule 1, vascular cell adhesion molecule 1) and attenuated BBB damage. We are the first
to report that engrafted hiPSC-NSCs rapidly improved neurological function (less than 24 hours after
transplant). Rapid hiPSC-NSC therapeutic activity is mainly due to a bystander effect that elicits re-
duced inflammation and BBB damage. STEM CELLS TRANSLATIONAL MEDICINE 2015;4:841–851
SIGNIFICANCE
Clinically, cerebral vessel occlusion is rarely permanent because of spontaneous or thrombolytic
therapy-mediated reperfusion. These results have clinical implications indicating a much extended
therapeutic window for transplantation of human induced pluripotent stem cell-derived neural stem
cells (hiPSC-NSCs; 24 hours after stroke as opposed to the 5-hour window with tissue plasminogen
activator [tPA]). In addition, there is potential for a synergistic effect by combining hiPSC-NSC trans-
plantation with tPA to attenuate stroke’s adverse effects.
Reducing these early stage injuries could ex-
INTRODUCTION
tend the therapeutic window and lead to oppor-
Stroke causes long-term neurological disability. tunities for better clinical outcome. Presently, the
Transient ischemic reperfusion (IR), not permanent only available drug approved by the U.S. Food and
occlusion, is more relevant because, from a clinical Drug Administration is thrombolytic recombinant
standpoint, a good deal of stroke injury results from tissue plasminogen activator (tPA), which dis-
reperfusion following ischemia. The blood-brain solves clots and salvages cells in the ischemic pen-
barrier (BBB), which separates the brain from the umbra. The downside of tPA is its short (5-hour)
circulatory system, is a key target of the ischemic therapeutic window and greater risk of hemor-
reperfusion insult. IR causes biphasic opening of rhage due to exacerbated damage to the BBB [2].
the BBB [1] that initiates within several hours. If un- Human induced pluripotent stem cell-derived
checked, an irreversible second opening occurs neural stem cells (hiPSC-NSCs) could be an ideal
within the following 24–72 hours. This second ep- transplantation cell type for stroke therapy [3].
isode contributes significantly to cell death. These cells potentially possess multiple therapeutic
©AlphaMed Press 2015
842
actions, including functional neural replacement and bystander
effects (e.g., anti-inflammatory action, enhancement of endoge-
nous repair mechanisms by delivery of genetically engineered or
inherently synthesized gene products) [4, 5]. We hypothesized
that reducing inflammation and curtailing BBB leakage during
stroke’s initial stage could mitigate further neuronal damage
and hemorrhage.
We previously reported that human NSCs derived from plu-
ripotent embryonic stem cells have both a bystander effect (e.g.,
anti-inflammatory effect) and a neural cell replacement func-
tion in vivo without tumor formation [4]. Others have reported
engraftment of exogenous iPSC-derived NSCs after ischemia in-
to various regions of the brain at various time points in rodent
stroke models [6–11]. However, we and others demonstrated in
previous studies that when stem cells are transplanted into both
intracranial and intravascular regions 4–12 hours after middle ce-
rebral artery occlusion (MCAO), cell migration into the ischemic ter-
ritory of parenchyma is severely limited.
Early stage BBB damage is linked with increased endothelial-
leukocyte adhesion molecules, inflammation, matrix metalloprotei-
nases, and disruption of tight junctions [12–17]. In this study, we
tested the bystander or “chaperone” effect of iPSC-NSCs on
repairing the BBB between 24 and 48 hours after IR. This time frame
(24–48 hours) coincides with the second phase of BBB opening and
the peak of proinflammatory and proapoptotic factors [18]. The
goal of this study was to limit early stage BBB injuries, an outcome
that would protect against the second phase of stroke damage.
To investigate the impact of iPSC-NSCs between 24 to 48 hours
after stroke, we used a well-defined rodent experimental stroke
model: filamentous middle cerebral artery occlusion (60-minute
MCAO) with subsequent reperfusion (MCAO/R) [19]. We used
well-characterized iPSC-derived NSCs; iPSCs are reprogramed
from foreskin fibroblasts by ectopic expression of chromatin-
remodeling transcription factors (OCT4, KLF4, SOX2, and c-MYC)
and differentiated into NSCs. We transplanted hiPSC-NSCs into
the ipsilateral hippocampus, where neurogenesis is ongoing and
signals for migration, proliferation, differentiation, and integration
are widespread.
This study is the first to show that, following stroke, behavioral
functions recover quickly (within 24 hours after transplantation),
and pathophysiology is attenuated when iPSC-NSCs are transplanted
intracranially (hippocampus) during the acute phase after stroke.
MATERIALS AND METHODS
Middle Cerebral Artery Occlusion
All experiments were approved by the institutional animal care
and use committee of Tulane University and conducted according
to the guidelines of the American Veterinary Medical Association
and the National Institutes of Health Guide for the Care and Use of
Laboratory Animals. Animals and procedures include 24- to 26-g
male C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME, http://
www.jax.org) that were treated with either a sham operation or
MCAO/R with or without transplantation.
We induced focal ischemia via intraluminal MCA occlusion, as
described previously [20]. Briefly, mice were anesthetized with
1% isoflurane in 30% oxygen, with body temperature continu-
ously maintained at 37 6 0.5°C. An incision was made to the mid-
line of the neck to expose both the left common and external
carotid arteries. To the left internal carotid artery, we introduced
a 6-0 nylon monofilament coated with silicon rubber (Doccol
©AlphaMed Press 2015
iPSC-NSCs Quickly Decrease Symptoms in Stroke
Corporation, Sharon, MA, http://www.doccol.com) through the
external carotid stump to block the origin of the middle cerebral
artery (MCA). After 60-minute occlusion, reperfusion was intro-
duced by withdrawing the filament. Sham-operated control mice
underwent a similar surgery, but filament removal was carried
out immediately after insertion (no occlusion, no reperfusion).
We assessed regional cerebral blood flow (rCBF) through the
MCA using a transcranial laser Doppler (Perimed, Stockholm,
Sweden, http://www.perimed-instruments.com); severely re-
duced rCBF (.80%) was established after successful MCAO,
and reperfusion resulted in recovery of blood flow (.90%). Brains
were collected 48 hours after MCAO/R (24 hours after transplan-
tation) for quantitative analyses.
Human iPSC-NSC Culture
We used two different types of iPSC-NSCs. One type was iPSC-
NSCs derived by EMD Millipore (Billerica, MA, http://www.
emdmillipore.com). We purchased and used viral transgene-
free human iPSC-derived neural progenitor cells (SCC035). Specif-
ically, iPSCs were generated by EMD Millipore using the kit con-
taining STEMCCA Cre-excisable constitutive polycistronic (OCT4,
KLF4, SOX2, and c-MYC) lentivirus. These hiPSC-NPCs proliferate
as an adherent cell monolayer, and more than 80% of the cells ex-
press neural stem cell markers (e.g., Nestin and Sox-2). We pur-
chased passage 3 cells and expanded the cells using our well-
defined NSC media. The other type was human iPSCs (IMR90-1,
IMR90 clone 1) purchased from WiCell Research Institute (Mad-
ison, WI, http://www.wicell.org). The iPSC line was generated by
WiCell from IMR90 fetal lung fibroblasts through viral transduc-
tion by combining the OCT4, SOX2, NANOG, and LIN28 genes. Us-
ing a previously described protocol [21] and a well-established
protocol in our laboratory, we derived NSCs from these iPSCs in
two steps, first, producing neuroepithelial cells (NEPs) from these
iPSCs and, second, producing NSCs from NEPs. Specifically, we
generated NEPs in N2B27 medium containing Dulbecco’s modi-
fied Eagle’s medium nutrient mixture F12, two supplements
(N2 and B27), and 0.55 mM b-mercaptoethanol (all from Invitrogen;
Thermo Fisher Scientific, Waltham, MA, http://www.thermofisher.
com); human recombinant noggin (300 ng/ml; PeproTech, Rocky
Hill, NJ, https://www.peprotech.com) and SB431542 (10 mM;
Tocris Bioscience, Bristol, U.K., http://www.tocris.com) were also
added. This medium was changed daily. After 8–10 days of differ-
entiation, neural rosettes containing NEPs appeared. At this
point, the medium was changed to N2B27 medium, which was
supplemented with human epidermal growth factor (20 ng/ml;
R&D Systems, Minneapolis, MN, http://www.rndsystems.com)
and human basic fibroblast growth factor (20 ng/ml; PeproTech).
These NSCs were cultured for 3–5 days until confluent, then pas-
saged (dilution factor: approximately 1:3 to 1:5) using Accutase.
Prior to confluence, the hiPSC-NSCs were transferred enzymati-
cally during refeeding once per week or within 48–72 hours prior
to transplantation. The cells proliferated as an adherent cell
monolayer, and more than 80% of the cells expressed appropriate
NSC markers (e.g., Nestin and Sox-2), as confirmed by reverse
transcriptase polymerase chain reaction (RT-PCR) and immunos-
taining during the expansion phase.
It has been previously documented that hNSCs can be success-
fully xenografted and integrated into the mouse brain [6, 11]. Pre-
vious studies by others showed naı̈ve iPSCs (not derived to NSCs) to
be tumorigenic when intracranially transplanted into a rodent
S TEM C ELLS T RANSLATIONAL M EDICINE
Eckert, Huang, Gonzalez et al.
model of MCAO [22]. We found no non-neural cell types or tumors
after the hiPSC-NSCs were engrafted.
hiPSC-NSC Transplantation
At 24 hours after MCAO/R, we transplanted hiPSC-NSCs into
the hippocampus. Adult C57BL/6J mice were anesthetized with
1% isoflurane in 30% oxygen using a face mask and placed in a
stereotaxic frame (Stoelting Co., Wood Dale, IL, http://www.
stoeltingco.com). To prevent the eyes from drying, ocular oint-
ment was applied, animal heads were wiped with 70% ethanol,
and the skin on the head was cut open at the midline. A small burr
hole (0.5-mm diameter, F.S.T.) was drilled in the skull (2-mm pos-
terior to the bregma, 1.5-mm lateral to the sagittal suture; 2 ml of
phosphate-buffered saline (PBS) containing ∼100,000 viable cells
were injected over a 3-minute period into the ipsilesional hemi-
sphere (depth of 2- to 2.5-mm dorsal). Sham controls were
injected with 2 ml PBS alone. The wound was sealed with cyano-
acrylate glue (Vector Laboratories, Burlingame, CA, https://www.
vectorlabs.com), and the brains were collected for analysis
48 hours after MCAO/R (24 hours after transplantation).
Behavioral Tests
Behavioral tests for each mouse included adhesive removal,
beam walk, and rotarod, which were carried out 3 days consecu-
tively before MCAO/R (as training) and from 2 to 30 days after
MCAO/R. For the adhesive removal test, we attached a small piece
(3 3 4 mm) of adhesive tape onto the contralateral forepaw of each
mouse using equal pressure [23]. The mouse was then transferred
to a transparent Perspex box (VetEquip, Inc., Pleasanton, CA,
http://www.vetequip.com), and we recorded the time it took for
the mouse to remove the tape (to 120 seconds maximum).
The beam walk test required that the mice be trained to tra-
verse an elevated beam (700 3 5 mm) to reach an enclosed box.
For the study, we chose only the mice that reached the box in
20 seconds without stopping. After MCAO/R, latency to traverse
the beam (to 60 seconds maximum) was recorded.
The rotarod test [24] was performed with minor modifica-
tions by using four velocities (10, 15, 20, and 25 rotations per min-
ute [rpm]) with three trials per velocity. The amount of time that
a mouse remained on the rod was recorded, but the trial ended if
the mouse fell. Speed averages shown are as a percentage relative
to the sham mice. Data shown are the total amount of time on the
rod as a percentage relative to the sham mice.
Tissue Processing
At 48 hours after MCAO/R (24 hours after transplantation), mice
were deeply anesthetized and transcardially perfused with nor-
mal saline and then with phosphate-buffered 4% paraformaldehyde
(PFA). Brains were postfixed in PFA for 24 hours and cryoprotected
in 30% sucrose; free-floating coronal sections (30 mm) were pre-
pared using a vibratome (VT1000 S; Leica Biosystems, Wetzlar,
Germany, http://www.leicabiosystems.com); frozen sections
were embedded in Tissue-Tek O.C.T. compound (catalog no.
4583; Andwin Scientific, Schaumburg, IL, https://www.andwinsci.
com), and then cut coronally at a 20-mm thickness.
Quantification of Infarct Volume
Triphenyl tetrazolium chloride (TTC) staining was used to mea-
sure infarct (lesion) volume. Fresh mouse brains were cut into
1-mm coronal sections at 48 hours after MCAO/R, then incubated
www.StemCellsTM.com
843
in 2% TTC solution (Sigma-Aldrich, St. Louis, MO, https://www.
sigmaaldrich.com). When brain tissue is viable, mitochondria con-
vert the TTC to a red substance, and the ischemic area remains
colorless. ImageJ analysis (NIH, Bethesda, MD, http://imagej.
nih.gov/ij/) was used to measure infarct size. Lesion volume
was calculated as a percentage volume of the contralateral hemi-
sphere to compensate for edema, using the following formula:
(volume of contralateral hemisphere 2 [volume of total ipsile-
sional hemisphere 2 volume of infarct area]) / volume of contra-
lateral hemisphere.
Immunohistochemistry
To block nonspecific binding to the 30-mm coronal free-floating
sections, brain sections were first incubated in 10% goat serum
in PBS, pH 7.4 (containing 0.1% Tween 20, 0.3% Triton X-100),
and then incubated with primary antibodies at 4°C overnight. Pri-
mary antibodies (and dilutions) were used as follows: rabbit anti-
Iba-1 (1:300; Wako Pure Chemical Industries, Wako, Japan,
http://www.wako-chem.co.jp/english/); mouse anti-human cy-
toplasm STEM121 (1:500; StemCells, Inc., Palo Alto, CA, http://
www.stemcellsinc.com); mouse anti-human mitochondria
(1:300; EMD Millipore); mouse anti-human Nestin (1:300; Abcam,
Cambridge, MA, http://www.abcam.com) for neural stem cells;
mouse anti-bIII-tubulin (TuJ-1, 1:300; Abcam) for neuron; rabbit
anti-S100-b for astroglia; rat anti-mouse CD31 (1:400; Abcam);
and rabbit anti-brain-derived neurotrophic factor (anti-BDNF;
1:400; Abcam). For immunofluorescent staining, we used the fol-
lowing secondary antibodies: goat anti-rabbit IgG (Alexa Fluor
594, 1:500; Invitrogen; Thermo Fisher Scientific), goat anti-
mouse IgG (Alexa Fluor 488, 1:500; Invitrogen; Thermo Fisher Sci-
entific), and goat anti-rat IgG (Alexa Fluor 594, 1:500; Invitrogen;
Thermo Fisher Scientific).
Reverse Transcriptase Polymerase Chain Reaction
To extract total RNA from ipsilesional MCAO/R mouse brains, tis-
sue (∼20–30 mg preserved in RNAlater (Ambion; Thermo Fisher
Scientific) was immersed in Trizol reagent (Invitrogen; Thermo
Fisher Scientific) and homogenized using MagNA Lyser (Roche
Diagnostics, Basel, Switzerland, http://www.roche.com). RNA
fractions were extracted with chloroform, mixed with an appro-
priate volume of 70% ethanol, and loaded onto an RNeasy column
(Qiagen, Venlo, The Netherlands, https://www.qiagen.com). DN-
ase I (Qiagen) was used to digest the retained material. To prepare
first-strand cDNA, 2 mg of total RNA was reverse transcribed with
Applied Biosystems high-capacity cDNA reverse transcriptase and
random primers (catalog no. 4368814; Invitrogen; Thermo Fisher
Scientific). The PCR amplification reaction mixture included 2 ml
of reverse transcriptase in 10 ml of SsoFast Probes Supermix with
Rox (Bio-Rad, Hercules, CA, http://www.bio-rad.com), which was
placed in a CFX96 Real-Time Thermal Cycler (Bio-Rad) under re-
action conditions including a 30-second hold at 95°C, 35 cycles
of activation for 5 seconds at 95°C, and annealing/extending
for 10 seconds at 60°C. The following TaqMan gene expression assays
were used (tumor necrosis factor a [TNF-a]: Mm00443258_m1;
interleukin 6 [IL-6]: Mm00446190_m1; IL-1b: Mm00434228_m1;
vascular cell adhesion molecule 1 [VCAM-1]: Mm01320970_m1; in-
tercellular adhesion molecule 1 [ICAM-1]: Mm00516023_m1; Ccl2
[monocyte chemotactic protein 1 (MCP-1)]: Mm00441242_
m1; Ccl3 [macrophage inflammatory protein 1a (MIP-1a)]:
Mm00441259_g1; glyceraldehyde-3-phosphate dehydrogenase
©AlphaMed Press 2015
844
[GAPDH]: Mm99999915_g1). Ct values were normalized relative to
GAPDH. The Livak (22DDCt) method was used to calculate changes
in target gene expression relative to the control mice [25].
Blood-Brain Barrier Permeability Assay
BBB integrity was assessed by Texas Red-dextran perfusion [26]
and IgG Western blots with minor modifications [27]. For Texas
Red-dextran perfusion, mice were deeply anesthetized with iso-
flurane, and Texas Red-dextran 70 kDa (D1864, Invitrogen;
Thermo Fisher Scientific) in PBS (50 mg/ml) was injected into the
inferior vena cava for 2 minutes of circulation. Mice were immedi-
ately killed by decapitation. Brains were quickly extracted and post-
fixed in 4% PFA for 24 hours and then cryoprotected with 30%
sucrose for 48 hours. After embedding in optimum cutting temper-
ature compound, brains were sliced coronally in 20-mm sections.
Slices were directly coverslipped using mounting media with DAPI
(49,6-diamidino-2-phenylindole, H-1200; Vector Laboratories).
Western Blot Analysis
After mice had been deeply anesthetized (48 hours after MCAO/
R), we collected the ipsilesional cortex and hippocampal regions
(bregma: 21.5 to 3.5 mm) and homogenized them in cold NP-40
lysis buffer (Boston BioProducts Inc., Ashland, MA, http://www.
bostonbioproducts.com) to isolate the total protein. The slurries
were centrifuged at 12,000 rpm for 15 minutes, and supernatants
were collected. We loaded 30-mg protein samples onto 4%–12%
bis-Tris NuPAGE Novex gels (Invitrogen; Thermo Fisher Scientific),
then transferred them onto nitrocellulose membranes (Invitro-
gen; Thermo Fisher Scientific). Primary antibodies included matrix
metalloproteinase 9 (MMP-9; 1:1,000; Abcam), Iba-1 (1:500;
Abcam), CD11b (1:1,000; Abcam), zonula occludens-1 (ZO-1;
1:500, Invitrogen; Thermo Fisher Scientific), BDNF (1:1,000;
Abcam), and b-actin (1:2,500; Thermo Fisher Scientific). Horse-
radish peroxidase-conjugated goat anti-mouse IgG (1:600, Invi-
trogen; Thermo Fisher Scientific) was used for detecting mouse
IgG in brain tissue. Blots were probed with appropriate horserad-
ish peroxidase-conjugated secondary antibodies (1:3,000, Invi-
trogen; Thermo Fisher Scientific) and detected using ECL
Western Blotting Substrate (Thermo Fisher Scientific). We used
ImageJ analysis to quantify protein intensities.
Gelatin Zymography
SDS-polyacrylamide gel electrophoresis zymography was used to
detect functional MMP-9 enzyme, as described previously [28].
Brain samples were prepared for Western blotting but without
denaturing the supernatants before loading the gels. Protein
(30 mg per well) was loaded onto a Novex 10% gelatin zymogram
gel (Invitrogen; Thermo Fisher Scientific). Following electropho-
resis, renaturing buffer (Invitrogen; Thermo Fisher Scientific)
was added to the gel, which was then incubated in developing
buffer (Invitrogen; Thermo Fisher Scientific) at 37°C overnight.
The gel was stained with 0.5% SimplyBlue SafeStain (Invitrogen;
Thermo Fisher Scientific) for 60 minutes, and then destained.
MMP-9 protease activity was visualized indirectly as clear bands
against a dark background. Images were analyzed using ImageJ.
Confocal Microscopy
Images were generated with a Bio-Rad MRC-1024 confocal laser
scanning microscope furnished with a single-photon Kr/Ar laser
and a Bio-Rad Radiance 2100 MP confocal microscope furnished
©AlphaMed Press 2015
iPSC-NSCs Quickly Decrease Symptoms in Stroke
with a multiphoton laser. To negate channel cross-talk, images were
acquired sequentially. Pixel resolution was 1,024 3 1,024, with neg-
ative and positive control images taken at the same settings.
Statistical Analysis
GraphPad Prism 6 (GraphPad Software, La Jolla, CA, http://www.
graphpad.com) and SPSS 19 (IBM Corp, Armonk, NY, http://www.
ibm.com) were used to carry out statistical analysis. To determine
dissimilarities between multiple groups, we performed a one-way
analysis of variance with Fisher’s least significant difference post
hoc test. Tests were considered statistically significant at p values
,.05. Data are presented as mean 6 SEM.
RESULTS
Transplanted hiPSC-NSCs Ameliorate
Neurological Dysfunction
We transplanted hiPSC-NSCs into the ipsilesional hippocampus
24 hours after stroke, and evaluated the behavioral deficits
24 hours after transplantation. The three tests [29, 30] were cho-
sen to determine sensorimotor, balance, and motor function be-
ginning 1 day after hiPSC-NSC transplantation (Fig. 1A–1C). Each
mouse was subjected to the three tests: before surgery 3 days con-
secutively (pretraining) and from 2 to 30 days after surgery. Im-
proved neurological dysfunction was clearly evident 2–8 days
after injury; the efficacy of the transplanted hiPSC-NSCs for behav-
ioral dysfunction was maintained over 1 month. Consequently, we
assessed the impact of hiPSC-NSCs at early time points and found
that the hiPSC-NSCs attenuated the impaired behavior.
For the adhesive removal test, adhesive tape was applied to the
contralateral forepaw of mice for the sham-operated and for MCAO/
R mice with or without transplanted hiPSC-NSCs. The MCAO/R mice
took much longer to remove the tape than the sham-operated mice
(p , .0001 at days 2, 4, and 6; p , .001 at day 8) (Fig. 1A). However,
mean removal time was significantly shorter for the hiPSC-NSC-
transplanted mice compared with the nontransplanted control
MCAO/R mice (p , .001 at day 2; p , .001 at day 4) (Fig. 1A), dem-
onstrating improvement in sensorimotor deficits.
To determine behavioral balance, mice were subjected to the
beam walk test [31], in which we measured the time it took for the
animal to walk across an elevated narrow beam to a platform.
MCAO/R mice were significantly slower in mean walk time com-
pared with sham controls (p , .0001). However, the mean walk
time of transplanted mice was significantly faster for hiPSC-NSC-
transplanted mice compared with the nontransplanted control
MCAO/R mice, demonstrating improved balance (p , .0001)
(Fig. 1B).
For motor coordination, we used the rotarod test to evaluate
how long an animal remained on a rotating rod. MCAO/R control
mice demonstrated extreme impairment in the ability to stay
on the rod compared with sham-operated mice, starting at 2 days
after MCAO/R and remaining low for the entire test period
(p , .0001) (Fig. 1C). However, hiPSC-NSC-transplanted mice
remained on the rod significantly longer, and this improved neuro-
logical outcome persisted for the whole test period. Sham-operated
mice showed no behavioral dysfunction. This suggests that hiPSC-
NSC transplantation into the hippocampus at 24 hours after
MCAO/R can provide long-term benefits.
As a negative control, we transplanted heat-killed hNSCs.
These engrafted cells failed to migrate, and behavioral dysfunction
S TEM C ELLS T RANSLATIONAL M EDICINE
Eckert, Huang, Gonzalez et al.
Figure 1. Behavioral deficits in MCAO/R mice ameliorated by
hiPSC-NSC transplantation. (A): Forepaw adhesive removal tests show
improvement with hiPSC-NSCs: mean adhesive removal times are signif-
icantly shorter than those for nontransplanted MCAO/R mice (n = 10).
(B): Beam walk tests show improvement with hiPSC-NSCs: beam
walking times are significantly shorter compared with those of non-
transplanted MCAO/R mice (n = 9). (C): Rotarod tests show improve-
ment with hiPSC-NSCs: rod-remaining times for transplanted mice
were significantly longer (n = 10). Four velocities (10, 15, 20, and
25 rotations per minute) were used, with three trials performed
at each velocity. Durations at all speeds are summed; data are pre-
sented as a percentage relative to sham or MCAO/R mice. p, p , .05,
ppp, p , .001, pppp, p , .0001 vs. sham group; ##, p , .01, ###,
p , .001, ####, p , .0001 vs. MCAO/R group. Data are expressed as
mean 6 SEM. Abbreviations: hiPSC-NSC, human induced pluripotent
stem cell-derived neural stem cell; MCAO/R, middle cerebral artery oc-
clusion with subsequent reperfusion; Tx, transplantation.
was not distinguishable from or worse than that of nontrans-
planted MCAO/R mice. These results validated that the effects
of the engrafted hiPSC-NSCs were not due to any transplant-
induced inflammatory response (data not shown). To validate
that hiPSC-NSC xenotransplantation was optimal in the mouse
www.StemCellsTM.com
845
brain, we compared the effect of hiPSC-NSCs with that of C17.2
[4, 32], well-characterized mouse NSCs (mNSCs). We found that
the mNSC-engrafted stroked mouse performance was similar
to hiPSC-NSC engrafted mice, demonstrating that hiPSC-NSC xen-
otransplantation is not suboptimal and allowing us to confidently
use hiPSC-NSCs for this study. To validate the behavioral findings
that we obtained from human NSCs derived from IMR-90 iPSCs,
we transplanted another source of well-characterized hiPSC-NSCs
(SCC035; EMD Millipore). We obtained similar results: the stroked
mice engrafted with these hiPSC-NSCs performed significantly
better than nontransplanted MCAO/R control mice, thus showing
a positive outcome (data not shown). Based on these results, we
continued to carry out subsequent experiments with IMR-90
hiPSC-NSCs in this study.
hiPSC-NSCs Rapidly Migrated Into the Site of
Stroke Injury
To explore the potential mechanisms involved in improved be-
havioral outcomes, we investigated the effects of hiPSC-NSCs
by measuring lesion volumes at 24 hours after transplantation
(48 hours after MCAO/R). In vitro, hiPSC-NSCs uniformly ex-
pressed the NSC-associated markers Sox2 and Nestin (Fig. 2B),
which verifies their characteristics. In viable brain tissue, TTC is
converted by mitochondria to a red substance but remains color-
less in the ischemic area (TTC staining: white, infarct) (Fig. 2C).
Previous reports show similar results to ours: 60-minute MCAO
and 48-hour reperfusion-induced ischemic damage in both the
cortex and striatum but not the hippocampus [33]. Although over-
all infarct volume measured by TTC staining was not significantly
reduced in hiPSC-NSC-transplanted mice (Fig. 2C, 2D), cresyl vio-
let staining patterns showed reduced infarct volume in the cortex,
suggesting a positive impact (data not shown).
We next examined the distribution of engrafted (donor) hu-
man iPSC-NSCs by immunostaining with the human-specific anti-
body (STEM121) against a human cytoplasmic protein (Fig. 2E–2H;
STEM121, green, arrowhead). Engrafted hiPSC-NSCs are positive
for the human-specific neural stem cell marker hNestin (Fig. 2I;
red, arrow). By 24 hours after transplant, the hiPSC-NSCs had mi-
grated to the lesion, including the cortex, but hiPSC-NSCs migrated
less extensively to the caudoputamen lesion.
It is predicted that the sooner the intervention to prevent BBB
damage, the better the clinical outcome after stroke; however,
we found that hiPSC-NSCs injected at other than 24 hours after
MCAO/R (i.e., 6 or 12 hours after MCAO/R) migrated less exten-
sively to the injured site, and this was associated with a lack of
rapid neurological improvement (data not shown). In addition,
compared with the hippocampal injection site, hiPSC-NSCs trans-
planted into the ipsilateral striatum or lateral ventricle 24 hours
after MCAO/R showed less extensive migration to the injured
sites (data not shown). Based on these findings, we focused on
the impact of hiPSC-NSCs transplanted into the hippocampus
24 hours after MCAO/R.
Transplanted hiPSC-NSCs Reduce Proinflammatory
Cytokine Levels
Ischemia and reperfusion prompt upregulation of proinflamma-
tory cytokines and chemokines [12–17]. We assessed the effect
of transplanted hiPSC-NSCs on proinflammatory cytokines using
RT-PCR. We found that at 48 hours after MCAO/R in the ipsile-
sional hemisphere, proinflammatory cytokines TNF-a, IL-6, and
©AlphaMed Press 2015
846 iPSC-NSCs Quickly Decrease Symptoms in Stroke

Figure 2. hiPSC-NSCs rapidly migrated into the site of stroke injury. (A–D): Experimental timeline. (A): MCAO/R was performed in C57BL/6J
mice at time 0, and hiPSC-NSCs were transplanted into the ipsilesional hippocampus 24 hours after MCAO/R. Outcomes were assessed at in-
dicated intervals. (B): Representative images of hiPSC-NSCs immunostained with anti-Sox2 (red) and anti-Nestin (green) in culture. Scale bar = 30
mm. (C): Triphenyl tetrazolium chloride staining of sham, MCAO/R, and hiPSC-NSC-engrafted mouse brains (red, viable tissue; white, infarct).
Sections from the same animal are shown from anterior to posterior (top to bottom). Scale bar = 2 mm. (D): Quantification of infarct volume.
Data are expressed as mean 6 SEM (n = 3; ppp, p , .001 vs. sham group. (E–I): Brain diagram. (E): Injection sites (arrowheads); hiPSC-NSC-
disseminated areas (red dots). (F–I): Human iPSC-NSCs were identified with human-specific antibody (STEM121) against human cytoplasmic
protein (green, arrowheads) and human neural stem cell marker hNestin (red, arrows) on brain slices 24 hours after transplantation. Three
sampling sites are shown in Fig. 2E (insets, higher magnification). Scale bar = 30 mm (10 mm, inset). Abbreviations: Contra, contralesional; DAPI,
49,6-diamidino-2-phenylindole; hiPSC-NSC, human induced pluripotent stem cell-derived neural stem cell; Ipsi, ipsilesional; MCAO/R, middle
cerebral artery occlusion with subsequent reperfusion; RT-PCR, reverse transcriptase-polymerase chain reaction; Tx, transplantation.

IL-1b were increased by 30.6 6 7.8-fold, 12.7 6 3.9-fold, and
5.2 6 0.8-fold, respectively (Fig. 3A). In contrast, hiPSC-NSC en-
graftment reduced TNF-a, IL-6, and IL-1b expression (Fig. 3A).
Compared with the sham group, the MCAO/R group showed
upregulation of cell adhesion molecules ICAM-1 and VCAM-1
by 10.8 6 3.5-fold and 2.7 6 0.5-fold, respectively; however,
transplanted hiPSC-NSCs reduced ICAM-1 and VCAM-1 gene ex-
pression compared with the MCAO/R group (Fig. 3A). Levels of
mRNAs encoding chemokines MCP-1 and MIP-1a (indicators of
microglial/macrophage activation) were also significantly increased
in the MCAO/R mice by 249.5 6 71.7-fold and 49.1 6 4.7-fold,
respectively, compared with sham mice (Fig. 3B). Engrafted
hiPSC-NSCs significantly downregulated expression of both che-
mokines (Fig. 3B). Consequently, injected hiPSC-NSCs decreased
gene expression associated with inflammation and microglial/
macrophage activation.
hiPSC-NSC Engraftment Ameliorates BBB Damage
Upregulation of proinflammatory cytokines and chemokines after
stroke are associated with increased BBB permeability [34, 35].

©AlphaMed Press 2015 S TEM C ELLS T RANSLATIONAL M EDICINE

Eckert, Huang, Gonzalez et al.
Figure 3. Proinflammatory gene expression reduced by hiPSC-NSC
transplantation. (A): Reduced inflammatory marker expression in
the ipsilesional hemisphere of hiPSC-NSC-transplanted MCAO/R
brains. Reverse transcriptase-polymerase chain reaction was used
to measure inflammatory gene expression, then normalized to that
of GAPDH. Transcript levels of proinflammatory cytokines TNF-a,
IL-6, IL-1b, and ICAM-1 and VCAM-1 are elevated in the MCAO/R
brains and downregulated in transplanted brains compared with
MCAO/R control brains. p, p , .05, pp, p , .01, ppp, p , .001 vs. sham
mice; #, p , .05, ##, p , .01, ###, p , .001 vs. MCAO/R mice. (B):
Chemokines MCP-1 and MIP-1a are dramatically elevated in
MCAO/R brains and significantly downregulated in hiPSC-NSC-trans-
planted brains. p, p , .05, pp, p , .01, ppp, p , .001, pppp, p , .0001
vs. sham; #, p , .05, ##, p , .01, ###, p , .001, ####, p , .0001 vs.
MCAO/R. Data are expressed as mean 6 SEM (n = 4). Abbreviations:
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; hiPSC-NSC,
human induced pluripotent stem cell-derived neural stem cell;
ICAM-1, intercellular adhesion molecule 1; IL, interleukin; MCAO/R,
middle cerebral artery occlusion with subsequent reperfusion;
MCP-1, monocyte chemotactic protein 1; MIP-1a, macrophage in-
flammatory protein 1a; TNF-a, tumor necrosis factor a; Tx, transplan-
tation; VCAM-1, vascular cell adhesion molecule 1.
Loss of BBB integrity then promotes extravasation of fluids and
intravascular proteins into the brain parenchyma. To determine
the effect of hiPSC-NSCs on BBB permeability, we assessed the
level of the blood-derived substance IgG into the brain paren-
chyma. Western blots showed significantly increased IgG levels
in the ipsilesional cortex 48 hours after injury, indicating BBB leak-
age in the MCAO/R brain compared with sham controls (Fig. 4A,
4B); however, IgG levels were significantly reduced in the hiPSC-
NSC-transplanted group compared with the MCAO/R group (Fig.
4A, 4B). We used fluorescent Texas Red-dextran (70 kDa) to trace
BBB function (Fig. 4C, 4D). We found that the tracer remained in
the lumen of the capillaries and showed a clear vascular edge, in-
dicating an intact BBB in the contralateral region. In contrast, in
the stroked ipsilateral cortex, massive tracers diffused out of
the leaky capillaries, as shown by blurred and widespread punc-
tuated regions alongside the vascular segments after MCAO/R
without treatment, indicating a compromised BBB. However,
transplanted hiPSC-NSCs noticeably reduced leakage of the
tracer, as indicated by improved BBB integrity (Fig. 4D), and
engrafted hiPSC-NSCs (STEM121-positive, green) were distrib-
uted around the blood vessels (CD31-positive, red) (Fig. 4E, 4F).
We next determined the effect of hiPSC-NSCs on matrix met-
alloproteinases MMP-2 and MMP-9. Ischemia and reperfusion leads
www.StemCellsTM.com
847
to upregulation of MMPs in the stroke lesions [36]. Specifically,
MMP-2 is upregulated a few hours after stroke [37], and MMP-9
is upregulated in the later phase. Western blot analysis showed that
MMP-9 levels in the MCAO/R group increased compared with the
sham group 48 hours after injury (Fig. 4F, 4G); however, transplanted
hiPSC-NSCs significantly attenuated MMP-9 levels (Fig. 4F, 4G). In
addition, MMP-9 enzyme activity (gel zymography using gelatin as
a substrate) was upregulated 48 hours after MCAO/R (Fig. 4H, 4I),
with transplanted hiPSC-NSCs significantly downregulating MMP-9
activity (Fig. 4H, 4I). Consistent with previous studies by others,
MMP-2 activity was low at 48 hours after MCAO/R during the second
BBB opening period (Fig. 4H).
Upregulated MMPs are linked to dysfunction of tight junc-
tions between endothelial cells (tight junction proteins are key
to maintaining BBB integrity) [38]. It has been reported that, in
vivo, BBB leakage occurs despite a structurally intact occludin
and claudin-5 network, transmembrane proteins integral to tight
junction integrity 24 hours after MCAO [26, 39]. Expression of
MMP-9 [40] during this time reportedly degrades the tight junc-
tion protein ZO-1 (links actin cytoskeleton and transmembrane
protein occludin) [41]. Western blots showed decreased ZO-1
levels in MCAO/R mice (Fig. 4J, 4K), but engrafted hiPSC-NSCs
showed that ZO-1 degradation was significantly prevented (Fig.
4J, 4K). Consistent with previous findings by others [26, 39], no
significant differences in claudin-5 and occludin levels were found
among all three groups (data not shown). These findings indicate
a role of hiPSC-NSCs in maintaining BBB integrity.
We then studied whether engrafted hiPSC-NSCs after MCAO/
R would reduce the number of activated inflammatory cells. Rest-
ing microglia (via a ramified shape) were identified in sham brains
by immunostaining for the microglial/macrophage marker Iba-1
(Fig. 5A). The number of activated inflammatory cells (Iba-1 pos-
itive and amoeboid shape) after MCAO/R were dramatically in-
creased throughout the lesion compared with the sham group
(Fig. 5B, 5D) but reduced in hiPSC-NSC-engrafted brains (Fig.
5C, 5D). BDNF is central for functional recovery and neuroprotec-
tion after stroke [42]. At 2 days after surgery, we found reduced
BDNF levels in nontransplanted MCAO/R brains (Fig. 5E), whereas
mice transplanted with hiPSC-NSCs showed increased BDNF pro-
tein levels in the ipsilesional cortex (Fig. 5F).
Although it is beyond the scope of this study, as we mainly an-
alyzed the effect of hiPSC-NSCs at 24 hours after transplantation,
we examined whether engrafted hiPSC-NSCs could differentiate
into multiple neural cell types at 30 days after MCAO/R and how
they affected proinflammatory cytokines. Proinflammatory cyto-
kines are highly expressed in the acute and subacute stages of MCAO
but return to normal baseline levels 12 days after MCAO [43]. We
analyzed the expression of inflammatory markers in the ipsilesional
hemisphere of hiPSC-NSC-transplanted brains 30 days after MCAO/R.
As expected, in the nontransplanted brains 30 days after MCAO/R
(unlike at 48 hours after MCAO/R), there was no significant in-
crease of these inflammatory cytokines and adhesion molecules,
which suggests no deleterious role of these factors 30 days after
MCAO/R (supplemental online Fig. 1). We found no significant dif-
ferences in MMP-2 and MMP-9 levels among groups (data not
shown). The numbers of activated microglia were highly in-
creased 48 hours after MCAO and still exist 30days after MCAO
[44]. We also found increases in MCP-1 and MIP-1a in the non-
transplanted MCAO/R brains. Our results showed no reduction
of these genes in the transplanted MCAO/R group. The mecha-
nism of increased MIP-1a expression in hiPSC-NSC transplanted
©AlphaMed Press 2015
848 iPSC-NSCs Quickly Decrease Symptoms in Stroke

Figure 4. Blood-brain barrier leakage lessened by hiPSC-NSC transplantation. (A): Western blot shows mouse IgG levels in ipsilesional cortex in
sham, MCAO/R, and transplanted mice. (B): Quantification of panel A (n = 3; p, p , .05, ppp, p , .001 vs. sham group; #, p , .05 vs. MCAO/R
group). (C): Brain diagram. Two sampling sites are shown (rectangles). (D): Spatial distribution of intravenously administered Texas Red-dextran
70 kDa (red) in stroke’s contralateral (Di, Diii) and ipsilateral (Dii, Div) cortical regions with or without hiPSC-NSC transplantation. Reduced
extravasation was observed in hiPSC-NSC-transplanted brains. Scale bar = 10 mm (10 mm, inset). (E): Distribution of engrafted hiPSC-NSCs
(STEM121-positive, green) around the blood vessels (CD31-positive endothelial cells in red). Scale bar = 10 mm. (F): Western blot analysis
of MMP-9 protein levels in ipsilesional cortex. (G): Quantification of panel F (n = 3; ppp, p , .001 vs. sham group; ##, p , .01 vs. MCAO/R group).
(H): Zymography assay shows MMP-9 activity in the ipsilesional cortex. (I): Quantification of panel H (n = 3; p, p , .05, pppp, p , .0001 vs. sham
group; ###, p , .001 vs. MCAO/R group). (J): Western blot analysis of ZO-1. (K): Quantification of panel J (n = 3; p, p , .05 vs. sham group;
#, p , .05 vs. MCAO/R group). Data are expressed as mean 6 SEM. Abbreviations: DAPI, 49,6-diamidino-2-phenylindole; hiPSC-NSC, human
induced pluripotent stem cell-derived neural stem cell; MCAO/R, middle cerebral artery occlusion with subsequent reperfusion; MMP, matrix
metalloproteinase; Tx, transplantation; ZO-1, zonula occludens-1.

©AlphaMed Press 2015 S TEM C ELLS T RANSLATIONAL M EDICINE

Eckert, Huang, Gonzalez et al. 849

Figure 5. Microglial activation reduced by hiPSC-NSC transplantation. Immunofluorescence staining for microglial marker Iba-1 (red)
and STEM121, an hNSC marker (green). (A): Sham mice, (B): MCAO/R mice, (C): Transplanted mice. Microglial activation increased 48 hours
after MCAO/R, whereas transplanted hiPSC-NSCs suppressed activation 24 hours after transplantation (arrowhead, transplantation site).
Sampling regions (rectangles). Infarct (gray region). (D): Quantification of Iba-1-positive active microglia per area (0.15 mm2) in mouse groups
(n = 4; ppp, p , .001 vs. sham; ###, p , .001 vs. MCAO/R). (E): Representative Western blot shows BDNF expression in ipsilesional hemisphere.
(F): Quantification of panel E (n = 3; ppp, p , .001 vs. sham; ###, p , .001 vs. MCAO/R). Data expressed as mean 6 SEM. Abbreviations: BDNF,
brain-derived neurotrophic factor; DAPI, 49,6-diamidino-2-phenylindole; hiPSC-NSC, human induced pluripotent stem cell-derived neural stem
cell; MCAO/R, middle cerebral artery occlusion with subsequent reperfusion; Tx, transplantation.

brains remains to be elucidated. Next, we analyzed engrafted do-
nor cells and found that hiPSC-NSCs are stably engrafted
(supplemental online Fig. 2) in brains 30days after MCAO/R.
The vast majority of donor cells (90% 6 3%) remained Nestin-
positive neural stem cells. Only a small percentage of donor-
derived cells coexpressed the neuronal marker (TuJ-1)
(supplemental online Fig. 2B; hMito, green; TuJ-1, red; merged,
yellow) and glial marker (S-100b) (supplemental online Fig. 2D;
hMito, green; S-100b, red; merged, yellow). These findings sup-
port the paracrine effect of hiPSC-NSCs in our experimental
conditions (transplanting cells 24 hours after MCAO/R).
DISCUSSION
We found that when hiPSC-NSCs were transplanted after stroke
into mice, they quickly migrated to the site of injury and rapidly
improved neurological and pathophysiological functions. Be-
cause proinflammatory signals recruit transplanted stem cells
to the injured site [45], we timed our transplantations 24 hours
after MCAO/R, when cell migration signals are abundant. In addi-
tion, we assessed the pathophysiological effects of hiPSC-NSCs
that lead to swift behavioral improvement 24 hours after trans-
plantation (48 hours after MCAO/R).
Specifically, we transplanted hiPSC-NSCs into the ipsilateral
hippocampus, the brain region in which neurogenesis is ongoing,
even under normal circumstances and that is rife with signals for
migration, proliferation, differentiation, and integration. Others
have previously shown only limited migration when stem cells
are injected into other areas of the brain [44, 46]. We also did
not find extensive migration of hiPSC-NSCs elsewhere, although
the subventricular zone (SVZ) is the other area of ongoing adult
neurogenesis. It is suggested that rodent SVZ harbors neural pre-
cursors that migrate mainly to the olfactory bulb, but it is contro-
versial whether the human SVZ can give rise to neurons; however,
it is well established that neurons arise from the human hippo-
campus, thus we transplanted hiPSC-NSCs into the hippocampus
24 hours after MCAO/R.
We previously reported the anti-inflammatory action of NSCs
in central nervous system disorders [4, 5]. Early phase inflamma-
tory responses in cerebral ischemia/reperfusion accompany
upregulation of proinflammatory cytokines, which alter endothe-
lial cell matrix interactions [12, 13]. Overproduced proinflammatory
mediators after MCAO/R are associated with (a) upregulation of
vascular endothelial adhesion molecules (mediates leukocyte ad-
hesion) [14], (b) factors that mediate transendothelial migration
of leukocytes [14, 15], and (c) increase in vascular permeability

www.StemCellsTM.com ©AlphaMed Press 2015

850 iPSC-NSCs Quickly Decrease Symptoms in Stroke

[47]. In this study, we showed that engrafted hiPSC-NSCs downre-
gulate expression of proinflammatory cytokines (TNF-a, IL-6, and
IL-1b), adhesion molecules (ICAM-1 and VCAM-1), and MCP-1
(CCL2) and MIP-1a [15], which mediate the infiltration process
and contribute to ischemic stroke injury. Downregulating these
factors after stroke would likely reduce neutrophil and mono-
cyte passage into the brain. In fact, we found decreased numbers
of Iba-1-positive, activated myeloid cells in hiPSC-NSC-engrafted
brains compared with nontransplanted MCAO/R brains.
MMPs, a family of more than 20 zinc-binding proteases, are
crucial to the process that results in damage to the BBB during
ischemic injury, thus the role of MMPs has been investigated ex-
tensively. Elevated MMP-9 activity is associated with degradation
of ZO-1 [48] (tight junction proteins in vascular endothelial cells)
and BBB breakdown with subsequent hemorrhagic transforma-
tion [16, 40]. In this study, we showed that engrafted hiPSC-
NSCs significantly reduced MMP-9 levels and increased ZO-1
expression, demonstrating that hiPSC-NSCs ameliorate BBB dis-
ruption by reducing MMP-9 activity and protecting ZO-1 from
degradation. BBB leakage can be demonstrated with extravasa-
tion of injected fluorescence Texas Red-dextran as early as 4 hours
after stroke in mice [26]. We confirmed recovery of BBB integrity
by reduced extravasation of both IgG and fluorescent dextran
(injected) in hiPSC-NSC-engrafted brains. Because MMP-9 upre-
gulation is linked with tPA-induced hemorrhage in stroke patients
[49] and animal models [50, 51], MMP-9 reduction by hiPSC-NSCs
suggests that a combination treatment of tPA and hiPSC-NSCs
could help reduce hemorrhagic events caused by tPA in the early
stage of stroke.
Our results showed that engrafted hiPSC-NSCs can ameliorate
secondary inflammatory brain damage by reducing cytokine pro-
duction and ameliorating later phase BBB leakage. Furthermore,
because neurotrophins promote functional recovery after stroke
[42], Western blot analysis showed significantly increased BDNF
expression, suggesting that BDNF upregulation may add a protec-
tive mechanism to this system. Neurological function rapidly im-
proved in mice 24 hours after transplantation; recovery of
behavioral function persisted throughout the month-long moni-
toring period. Consequently, we assessed the multiple actions
of hiPSC-NSCs that underlie their ability to promote beneficial
function 24 hours after transplantation (48 hours after stroke).
Although it is beyond the scope of this study, it would be
interesting to determine whether early intervention (bystander
effect) by engrafted cells could enhance neurogenesis, thus achiev-
ing even more enduring positive outcomes in stroke patients. Our
immunostaining results revealed that engrafted neural stem cells
survived 30 days after transplantation. Furthermore, most
engrafted cells remained neural stem cells, although a small
percentage of cells expressed neural and astroglial markers.
Moreover, our findings support the potential role of hiPSC-
NSC paracrine function against stroke.
CONCLUSION
Clinically, cerebral vessel occlusion is rarely permanent because
of spontaneous or thrombolytic therapy-mediated reperfusion.
Our results have clinical implications indicating a much extended
therapeutic window for hiPSC-NSC transplantation (24 hours af-
ter stroke as opposed to the 5-hour window with tPA). Moreover,
there is potential for a synergistic effect by combining trans-
planted hiPSC-NSCs with tPA to attenuate the adverse effects
of stroke.
ACKNOWLEDGMENTS
We thank Dr. Edward Neuwelt, Oregon Health and Science Uni-
versity, for helpful scientific discussion. We thank Gustavo Vallim
and Ryan Park, Tulane University, for technical assistance. This
work was supported by startup funds from the Tulane Center
for Stem Cell Research and Regenerative Medicine.
AUTHOR CONTRIBUTIONS
A.E.: collection and assembly of data, data interpretation and
manuscript writing, revision of the manuscript; L.H. and R.G.: col-
lection and/or assembly of data; H.-S.K.: data analysis and inter-
pretation; M.H.H.: data analysis and interpretation, manuscript
writing; J.-P.L.: conception and design, data analysis and interpre-
tation, manuscript writing, financial support, final manuscript
approval.
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicated no potential conflicts of interest.

REFERENCES
1 Kuroiwa T, Ting P, Martinez H et al. The bi-
phasic opening of the blood-brain barrier to
proteins following temporary middle cerebral
artery occlusion. Acta Neuropathol 1985;68:
122–129.
2 Tissue plasminogen activator for acute is-
chemic stroke. The National Institute of Neuro-
logical Disorders and Stroke rt-PA Stroke Study
Group. N Engl J Med 1995;333:1581–1587.
3 Takahashi K, Yamanaka S. Induction of plu-
ripotent stem cells from mouse embryonic and
adult fibroblast cultures by defined factors. Cell
2006;126:663–676.
4 Lee JP, Jeyakumar M, Gonzalez R et al.
Stem cells act through multiple mechanisms
to benefit mice with neurodegenerative meta-
bolic disease. Nat Med 2007;13:439–447.
5 Park KI, Teng YD, Snyder EY. The injured
brain interacts reciprocally with neural stem
cells supported by scaffolds to reconstitute
lost tissue. Nat Biotechnol 2002;20:1111–
1117.
6 Oki K, Tatarishvili J, Wood J et al. Human-
induced pluripotent stem cells form functional
neurons and improve recovery after grafting in
stroke-damaged brain. STEM CELLS 2012;30:1120–
1133.
7 Chen SJ, Chang CM, Tsai SK et al. Func-
tional improvement of focal cerebral ischemia
injury by subdural transplantation of induced
pluripotent stem cells with fibrin glue. Stem
Cells Dev 2010;19:1757–1767.
8 Yuan T, Liao W, Feng NH et al. Human in-
duced 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 2013;4:73.
9 Polentes J, Jendelova P, Cailleret M et al.
Human induced pluripotent stem cells improve
stroke outcome and reduce secondary degener-
ation in the recipient brain. Cell Transplant
2012;21:2587–2602.
10 Liu SP, Fu RH, Wu DC et al. Mouse-
induced pluripotent stem cells generated under
hypoxic conditions in the absence of viral infec-
tion and oncogenic factors and used for ische-
mic stroke therapy. Stem Cells Dev 2014;23:
421–433.
11 Chau MJ, Deveau TC, Song M et al. iPSC
Transplantation increases regeneration and
functional recovery after ischemic stroke in
neonatal rats. STEM CELLS 2014;32:3075–3087.
12 Defilippi P, Silengo L, Tarone G. Alpha 6.
beta 1 integrin (laminin receptor) is down-
regulated by tumor necrosis factor alpha and
interleukin-1 beta in human endothelial cells.
J Biol Chem 1992;267:18303–18307.
13 Pettigrew LC, Kindy MS, Scheff S et al. Fo-
cal cerebral ischemia in the TNFalpha-
transgenic rat. J Neuroinflammation 2008;5:47.

©AlphaMed Press 2015 S TEM C ELLS T RANSLATIONAL M EDICINE

Eckert, Huang, Gonzalez et al.
14 Rossi B, Angiari S, Zenaro E et al. Vascular
inflammation in central nervous system dis-
eases: Adhesion receptors controlling leukocyte-
endothelial interactions. J Leukoc Biol 2011;89:
539–556.
15 Lu B, Rutledge BJ, Gu L et al. Abnormali-
ties in monocyte recruitment and cytokine ex-
pression in monocyte chemoattractant protein
1-deficient mice. J Exp Med 1998;187:601–608.
16 Gasche Y, Fujimura M, Morita-Fujimura Y
et al. Early appearance of activated matrix
metalloproteinase-9 after focal cerebral ische-
mia in mice: A possible role in blood-brain
barrier dysfunction. J Cereb Blood Flow Metab
1999;19:1020–1028.
17 Wolpe SD, Davatelis G, Sherry B et al. Mac-
rophages secrete a novel heparin-binding protein
with inflammatory and neutrophil chemokinetic
properties. J Exp Med 1988;167:570–581.
18 Lu YY, Li ZZ, Jiang DS et al. TRAF1 is a critical
regulator of cerebral ischaemia-reperfusion injury
and neuronal death. Nat Commun 2013;4:2852.
19 O’Neill MJ, Clemens JA. Rodent models of
focal cerebral ischemia. Curr Protoc Neurosci
2001;chapter 9, unit 9.6.
20 Clark WM, Lessov NS, Dixon MP et al.
Monofilament intraluminal middle cerebral ar-
tery occlusion in the mouse. Neurol Res 1997;
19:641–648.
21 Chambers SM, Fasano CA, Papapetrou EP
et al. Highly efficient neural conversion of hu-
man ES and iPS cells by dual inhibition of SMAD
signaling. Nat Biotechnol 2009;27:275–280.
22 Kawai H, Yamashita T, Ohta Y et al. Trider-
mal tumorigenesis of induced pluripotent stem
cells transplanted in ischemic brain. J Cereb
Blood Flow Metab 2010;30:1487–1493.
23 Bouet V, Boulouard M, Toutain J et al.
The adhesive removal test: A sensitive method
to assess sensorimotor deficits in mice. Nat Pro-
toc 2009;4:1560–1564.
24 Hamm RJ, Pike BR, O’Dell DM et al. The
rotarod test: An evaluation of its effectiveness
in assessing motor deficits following traumatic
brain injury. J Neurotrauma 1994;11:187–196.
25 Schmittgen TD, Livak KJ. Analyzing real-
time PCR data by the comparative C(T) method.
Nat Protoc 2008;3:1101–1108.
26 Kuntz M, Mysiorek C, Pétrault O et al.
Stroke-induced brain parenchymal injury drives
blood-brain barrier early leakage kinetics: A
combined in vivo/in vitro study. J Cereb Blood
Flow Metab 2014;34:95–107.
See www.StemCellsTM.com for supporting information available online.
www.StemCellsTM.com
27 Tanno H, Nockels RP, Pitts LH et al. Break-
down of the blood-brain barrier after fluid
percussive brain injury in the rat. Part 1: Distri-
bution and time course of protein extravasa-
tion. J Neurotrauma 1992;9:21–32.
28 Hu X, Beeton C. Detection of functional
matrix metalloproteinases by zymography. J
Vis Exp 2010;e2445.
29 Hunter AJ, Hatcher J, Virley D et al. Func-
tional assessments in mice and rats after
focal stroke. Neuropharmacology 2000;39:806–
816.
30 Virley D, Beech JS, Smart SC et al. A tem-
poral MRI assessment of neuropathology after
transient middle cerebral artery occlusion in the
rat: Correlations with behavior. J Cereb Blood
Flow Metab 2000;20:563–582.
31 Quinn LP, Perren MJ, Brackenborough KT
et al. A beam-walking apparatus to assess be-
havioural impairments in MPTP-treated mice:
Pharmacological validation with R-(-)-deprenyl.
J Neurosci Methods 2007;164:43–49.
32 Cartwright P, McLean C, Sheppard A et al.
LIF/STAT3 controls ES cell self-renewal and plu-
ripotency by a Myc-dependent mechanism. De-
velopment 2005;132:885–896.
33 Nagasawa H, Kogure K. Correlation be-
tween cerebral blood flow and histologic
changes in a new rat model of middle cerebral
artery occlusion. Stroke 1989;20:1037–1043.
34 Carmichael ST. Gene expression changes
after focal stroke, traumatic brain and spinal cord
injuries. Curr Opin Neurol 2003;16:699–704.
35 Barone FC, Feuerstein GZ. Inflammatory
mediators and stroke: New opportunities for
novel therapeutics. J Cereb Blood Flow Metab
1999;19:819–834.
36 Chang DI, Hosomi N, Lucero J et al. Activa-
tion systems for latent matrix metalloproteinase-
2 are upregulated immediately after focal cere-
bral ischemia. J Cereb Blood Flow Metab 2003;
23:1408–1419.
37 Rosenberg GA, Estrada EY, Dencoff JE. Ma-
trix metalloproteinases and TIMPs are associated
with blood-brain barrier opening after reperfu-
sion in rat brain. Stroke 1998;29:2189–2195.
38 Risau W, Wolburg H. Development of the
blood-brain barrier. Trends Neurosci 1990;13:
174–178.
39 Knowland D, Arac A, Sekiguchi KJ et al.
Stepwise recruitment of transcellular and para-
cellular pathways underlies blood-brain barrier
breakdown in stroke. Neuron 2014;82:603–617.
851
40 Asahi M, Wang X, Mori T et al. Effects of
matrix metalloproteinase-9 gene knock-out on
the proteolysis of blood-brain barrier and white
matter components after cerebral ischemia. J
Neurosci 2001;21:7724–7732.
41 Fanning AS, Jameson BJ, Jesaitis LA et al.
The tight junction protein ZO-1 establishes a link
between the transmembrane protein occludin
and the actin cytoskeleton. J Biol Chem 1998;
273:29745–29753.
42 Emanueli C, Schratzberger P, Kirchmair R
et al. Paracrine control of vascularization and
neurogenesis by neurotrophins. Br J Pharmacol
2003;140:614–619.
43 Kang SS, Keasey MP, Arnold SA et al. En-
dogenous CNTF mediates stroke-induced adult
CNS neurogenesis in mice. Neurobiol Dis 2013;
49:68–78.
44 Darsalia V, Allison SJ, Cusulin C et al. Cell
number and timing of transplantation deter-
mine survival of human neural stem cell grafts
in stroke-damaged rat brain. J Cereb Blood Flow
Metab 2011;31:235–242.
45 Imitola J, Raddassi K, Park KI et al. Di-
rected migration of neural stem cells to sites
of CNS injury by the stromal cell-derived factor
1alpha/CXC chemokine receptor 4 pathway.
Proc Natl Acad Sci USA 2004;101:18117–
18122.
46 Kelly S, Bliss TM, Shah AK et al. Trans-
planted human fetal neural stem cells survive,
migrate, and differentiate in ischemic rat cere-
bral cortex. Proc Natl Acad Sci USA 2004;101:
11839–11844.
47 Jin R, Yang G, Li G. Inflammatory mecha-
nisms in ischemic stroke: Role of inflammatory
cells. J Leukoc Biol 2010;87:779–789.
48 Rosell A, Lo EH. Multiphasic roles for ma-
trix metalloproteinases after stroke. Curr Opin
Pharmacol 2008;8:82–89.
49 Castellanos M, Leira R, Serena J et al.
Plasma metalloproteinase-9 concentration pre-
dicts hemorrhagic transformation in acute is-
chemic stroke. Stroke 2003;34:40–46.
50 Wang X, Lee SR, Arai K et al. Lipoprotein
receptor-mediated induction of matrix metallo-
proteinase by tissue plasminogen activator. Nat
Med 2003;9:1313–1317.
51 Mishiro K, Ishiguro M, Suzuki Y et al. A
broad-spectrum matrix metalloproteinase in-
hibitor prevents hemorrhagic complications in-
duced by tissue plasminogen activator in mice.
Neuroscience 2012;205:39–48.
©AlphaMed Press 2015