Human Molecular Genetics, 2015, Vol. 24, No.

21 60666079

doi: 10.1093/hmg/ddv318
Advance Access Publication Date: 11 August 2015
Original Article

ORIGINAL ARTICLE

Elucidating the role of the A2A adenosine receptor

in neurodegeneration using neurons derived from
Huntingtons disease iPSCs
Feng-Lan Chiu1,, Jun-Tasi Lin2,, Ching-Yu Chuang1,3,, Ting Chien2,,
Chiung-Mei Chen4, Kai-Hsiang Chen5, Han-Yun Hsiao2, Yow-Sien Lin2,
Yijuang Chern2, * and Hung-Chih Kuo1,3, *
1
Institute of Cellular and Organismic Biology, Academia Sinica, 2Institute of Biomedical Sciences, Academia
Sinica and 3Genomics Research Center, Academia Sinica, Taipei, Taiwan, 4Department of Neurology, Chang Gung
Memorial Hospital, Linkou Medical Center and College of Medicine, Chang-Gung University, Taoyuan, Taiwan
and 5Department of Neurology, National Taiwan University Hospital (Hsinchu Branch), Hsinchu, Taiwan
*To whom correspondence should be addressed at: Institute of Cellular and Organismic Biology, Academia Sinica, No. 128, Sec. 2 Academia Rd., Nankang,
Taipei 115, Taiwan. Tel: +886 227899580; Fax: +886 227899587; Email: kuohuch@gate.sinica.edu.tw or Institute of Biomedical Sciences, Academia Sinica,
No.128, Sec. 2 Academia Rd., Nankang, Taipei 115, Taiwan. Tel: +886E 227893913; Fax: +886 227829143; Email: bmychern@ibms.sinica.edu.tw

Abstract
Huntingtons disease (HD) is an autosomal-dominant degenerative disease caused by a cytosine-adenine-guanine trinucleotide
expansion in the Huntingtin (htt) gene. The most vulnerable brain areas to mutant HTT-evoked toxicity are the striatum and
cortex. In spite of the extensive efforts that have been devoted to the characterization of HD pathogenesis, no disease-modifying
therapy for HD is currently available. The A2A adenosine receptor (A2AR) is widely distributed in the brain, with the highest level
observed in the striatum. We previously reported that stimulation of the A2AR triggers an anti-apoptotic effect in a rat neuron-
like cell line (PC12). Using a transgenic mouse model (R6/2) of HD, we demonstrated that A2AR-selective agonists effectively
ameliorate several major symptoms of HD. In the present study, we show that human iPSCs can be successfully induced to
differentiate into DARPP32-positive, GABAergic neurons which express the A2AR in a similar manner to striatal medium spiny
neurons. When compared with those derived from control subjects (CON-iPSCs), these HD-iPSC-derived neurons exhibited a
higher DNA damage response, based on the observed expression of H2AX and elevated oxidative stress. This is a critical
observation, because oxidative damage and abnormal DNA damage/repair have been reported in HD patients. Most
importantly, stimulation of the A2AR using selective agonists reduced DNA damage and oxidative stress-induced apoptosis in
HD-iPSC-derived neurons through a cAMP/PKA-dependent pathway. These ndings support our hypothesis that human
neurons derived from diseased iPSCs might serve as an important platform to investigate the benecial effects and underlying
mechanisms of A2AR drugs.

Equal contribution.
Received: May 12, 2015. Revised and Accepted: July 31, 2015
The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

6066

Introduction
Huntingtons disease (HD) is an inherited, autosomal-dominant,
progressive neurodegenerative disease, caused by abnormal ex-
pansion of cytosine-adenine-guanine (CAG) repeats in exon1 of
the huntingtin (htt) gene. CAG repeats in excess of 36 result in
the production of mutated Huntingtin (HTT) proteins with an ab-
normally long polyglutamine stretch; these proteins form aggre-
gates in the cytoplasm and nucleus, which cause degradation of
neurons, particularly GABAergic medium-sized spiny projection
neurons (MSNs) in the striatum. MSNs are involved in the regula-
tion of movement, emotion and cognitive ability (1,2), and thus
HD patients gradually suffer from increased irritability and de-
pression, involuntary movements (chorea), rigidity and eventually
dementia. Although the mutation which causes HD is known, our
understanding of HD pathogenesis remains poor. Moreover, there
is currently no effective strategy to delay the onset or slow down
the progression of HD.
Various models have been developed to facilitate investiga-
tion into disease pathogenesis and to develop therapeutics. How-
ever, most studies of HD have utilized transgenic mice or
immortalized neurons carrying the mutant htt gene (3), and
these models may not fully recapitulate human HD. Recent
advances in human-induced pluripotent stem cell (iPSC) tech-
nology have enabled the generation of patient-specic iPSC
lines for studying disease pathogenesis and for use as a drug
screening platform (4,5). Multiple research groups have isolated
several human HD-iPSCs, and induced their differentiation into
a disease-relevant cell type [DARPP-32-positive cells, which are
expressed in the majority of MSNs (6)]; however, the differenti-
ation efciency was low (510%) (79). Increasing such efciency
is important for the effective study of disease pathogenesis.
Cell-autonomous events contribute to mutant HTT-induced
toxicity in affected neurons in mammalian cell culture systems
and transgenic mouse models of HD. Several studies have sug-
gested that multiple mechanisms may underlie the neurotoxic
effect of mutant HTT, including excitotoxicity (10,11), oxidative
stress and mitochondrial abnormalities (10,12), defects in cal-
cium homeostasis (13,14), ubiquitin-proteasome system impair-
ments (15,16) and transcriptional dysregulation (12,17). However,
such evidence is insufcient to account for the selective vulner-
ability of neurons, because mutant HTT is ubiquitously ex-
pressed in the brain and body. Consequently, these studies
imply that additional components are involved in neuronal cell
death in HD.
Adenosine is a well-characterized regulator of neuronal sur-
vival, which exerts its function through binding to specic cell
surface adenosine receptors. To date, four adenosine receptor
subtypes, A1, A2A, A2B and A3, have been identied. The A2A ad-
enosine receptor (A2AR) is a Gs-coupled receptor. Upon stimula-
tion, the A2AR activates multiple classical signaling pathways,
including the production of cAMP that activates PKA (18,19).
The A2AR exhibits differential distribution within the brain,
with the highest expression in the encephalin-expressing MSN
of the striatum (20,21). Expression of this receptor is affected by
mutant HTT, and is gradually down-regulated at the transcrip-
tional level during the progression of HD (2224). Nonetheless,
the ability of the A2AR to enhance the cAMP level in the brain of
HD mice is similar to that of WT mice (23), suggesting that A2AR
signaling in HD is amplied. Thus, the A2AR remains a potential
drug target of HD (25). This possibility is of great interest because
it has been shown that tonic activation of the A2AR is required for
the function of several neurotrophic factors, including brain-
derived neurotrophic factor, broblast growth factor and glial
Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6067
cell line-derived neurotrophic factor (2628). In genetic mouse
models of HD, accumulating evidence indicates that stimulation
of the A2AR may promote neuronal survival. For example, it has
been shown that an A2AR agonist (CGS21680) can (i) signicantly
delay deterioration of motor coordination (23), (ii) increase prote-
asome activity and reduce the accumulation of mutant HTT ag-
gregates (23,29) and (iii) attenuate NMDA-induced toxicity in
corticostriatal slices of a transgenic mouse model of HD (R6/2)
(30). Accordingly, A2AR-related drugs have been implicated as
possible treatments for human HD.
In this study, we generated and characterized human HD-
iPSCs with a CAG expansion mutation in the htt gene, and
induced their differentiation into GABAergic and DARPP-32-
positive neurons. Consistent with previous reports (8,9,31,32),
we did not observe mutant HTT aggregation in neurons derived
from HD-iPSCs. Nevertheless, we observed selective vulnerability
of HD-iPSC-derived neurons to oxidative stress, which is an
important feature of HD pathology (3335). Furthermore, we de-
monstrated that activation of the A2AR by selective agonists
(CGS21680 and APEC) reduced oxidative stress-induced DNA
damage and caspase 3 activation in the HD-iPSC-derived GA-
BAergic neurons. Our data imply that the human HD-iPSC
model recapitulates, at least in part, the HD disease phenotype
in vitro, and thus may be a potential platform for drug screening.
Results
Derivation and characterization of HD iPSCs
To generate iPSCs, broblasts or peripheral blood mononuclear
cells obtained from four non-HD controls and two patients diag-
nosed with HD were transduced with ve reprograming factors,
namely OCT4, SOX2, KLF4, c-MYC and NANOG (OSKMN) (Supple-
mentary Material, Table S1). Following 23 weeks under human
embryonic stem cell (hESC) culture conditions, several compact,
refractive, ESC-like colonies began to emerge. Four to six iPSC
lines were isolated for each subject and characterized by various
assays. Specically, the integration of exogenous OSKMN trans-
genes into the genomes of the HD-iPSC clones was conrmed
by PCR analysis at passage six (Fig. 1A). RT-PCR analysis of the
HD-iPSC lines revealed reactivation and expression of the en-
dogenous pluripotency-associated markers, including OCT4,
SOX2, NANOG, KLF4 and c-MYC, whereas the exogenous OSKMN
transgenes were found to be predominantly silenced (Fig. 1B).
PCR analysis of genomic DNA was also used to show that HD-
iPSC-A1 and HD-iPSC-B4 contained mutant htt with CAG expan-
sion (both with 43 repeats), as also observed for their parental HD
broblasts (Fig. 1A and C). In contrast, non-HD control iPSCs
(CON-iPSCs) contained only normal htt alleles with shorter CAG
triple repeats. Immunocytochemical (ICC) analysis with anti-
bodies against ESC antigens was used to demonstrate that
HD-iPSCs express pluripotency-associated genes and markers
without losing their pluripotency characteristics after extended
culture in vitro (Fig. 1D). The in vivo developmental propensity of
HD-iPSCs was examined using the teratoma formation assay.
Following intramuscular injection of undifferentiated HD-iPSCs
into immunocompromised mice, teratomas consisting of cell types
representing all three embryonic germ layers were formed (Fig. 1E).
Differentiation of HD-iPSCs into disease-relevant
neuronal subtypes
We proceeded to investigate the neuronal differentiation poten-
tial of the HD-iPSCs. As shown in Figure 2A, HD-iPSCs were
 6068 | Human Molecular Genetics, 2015, Vol. 24, No. 21

Figure 1. Characterization of human HD-iPSC cell lines. (A) Genomic PCR was used to show that all HD-iPSC lines in this study possessed integrated viral OCT4, SOX2, KLF4,
c-MYC and NANOG (OSKMN) genes. (B) Semi-quantitative PCR analysis was used to show that the viral OSKMN exogenous genes (exo) were silenced and endogenous loci
(endo) of OSKMN were reactivated in these HD-iPSC lines. (C) Genomic PCR and sequencing conrmed the CAG expansion mutation. The H9 hESC line (hESC), normal
human foreskin broblasts (CON-HF) and HD dermal broblasts from HD patients (HD-A/B-HF) were used as controls. -actin and GAPDH were used as loading
controls. (D) Phase contrast and ICC analysis of pluripotency markers (OCT4, SOX2, Nanog and SSEA4) in HD-iPSC lines (A1, A7, B4 and B16). (E) Hematoxylin and
eosin staining of tissue sections of teratomas derived from two HD-iPSC lines (A1 and B4) grafted into immunodecient mice. Scale bar: 50 m.

subjected to embryoid body (EB) formation followed by step-wise
in vitro differentiation. Following replating of EBs onto a cell cul-
ture surface in N2 media containing 20 ng/ml bFGF, rosette-like
structures resembling early neuroepithelium and neural pro-
genitors began to emerge from EB outgrowths. ICC staining and
RT-PCR analysis were used to demonstrate that these cells ex-
press neural progenitor markers, such as Nestin, SOX1 and
PAX6, but not the ESC markers OCT4 or Nanog (Fig. 2B and C). Fur-
ther culture of the neural progenitors in N2/B27 media with
10 ng/ml bFGF directed their differentiation into various neural
populations, including TUJ1-expressing or MAP2-expressing
neurons, and Olig2-expressing or GFAP-expressing glia cells
(Fig. 2B and C). In addition to expression of neural markers, we
also detected the mHTT transcript in iPSC-derived neurons.
RT-PCR was used to show that HD-iPSC neurons contained
mutant htt with CAG expansion, while CON-iPSC neurons con-
tained only normal htt with shorter CAG triple repeats (Fig. 2D).
Since GABAergic projection neurons are the primary neuronal
subtype affected in HD brains, we proceeded to examine the
proportion of GABAergic neurons in our model. To enhance
neuronal differentiation, we cultured the differentiated neural
progenitors in B27 media for an additional 46 weeks after
4 weeks of culture in B27/N2 media. Next, we interrogated the
differentiated cells at various developmental stages, using
GABAergic neuron markers, including - aminobutyric acid
(GABA), Glutamate decarboxylase (GAD) 65, Calbindin and dopa-
mine- and cAMP-regulated neuronal phosphoprotein (DARPP-
32). We observed that cells at the early stages of differentiation
 Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6069

Figure 2. Human HD-iPSCs differentiate into neural lineages. (A) Schematic procedure of GABAergic neuron differentiation. (B) Immunostaining and (C) semi-quantitative
PCR of HD-iPSC neural derivatives at 48 weeks with the indicated antibodies and primers. Nuclei were stained with DAPI. Scale bar: 50 m. (D) RT-PCR was used to detect
htt transcripts with expanded CAG repeats (upper band). GAPDH was used as an internal control.

already expressed GABA, while some even expressed the striatal
neuron markers GAD65 and Calbindin; cells expressing DARPP-
32, a MSN marker, were observed rarely under this condition
(data not shown). At later stages of differentiation, the majority
of cells were process-bearing neurons that expressed TUJ1 and
MAP2; in addition, a high proportion of cells at this stage (over
70%) were positive for GABAergic projection neuron markers:
GABA, GAD65, DARPP-32, or Calbindin (Fig. 3AD, Supplementary
Material, Fig. S1B and Supplementary Material, Table S5). We as-
sessed the homogeneity of the neuronal populations derived
from CON- or HD-iPSCs. First, we examined viable cells by Trypan
blue exclusion assay at weeks 4, 8 and 12 of neural differentiation;
this revealed that the expansion rates are similar between CON-
and HD-iPSC-derived NPCs (Supplementary Material, Fig. S1A). In
addition, the neuronal populations derived from CON- and HD-
iPSCs exhibit similar homogeneity, as revealed by ICC, Western
blotting and RT-qPCR analysis (Fig. 3D and E, Supplementary Ma-
terial, Fig. S1BC and S2). Together, these results indicate that the
majority of neuronal cells derived from HD-iPSCs or CON-iPSCs
are GABAergic projection neurons, and that the neural differenti-
ation efciencies are similar between CON-iPSC- and HD-iPSC-
derived NPCs.
HD-iPSC neurons are more susceptible to H2O2-induced
DNA damage than controls
Increasing evidence indicates that oxidative stress plays an im-
portant role in HD pathogenesis (3335). To examine whether
HD-iPSC-derived neurons are vulnerable to oxidative stress-
induced DNA damage, we treated neurons derived from both
HD- and CON-iPSCs with H2O2. ICC analysis and immunoblotting
with an antibody against H2AX, as a DNA damage marker (36),
were used to show that H2O2 induced H2AX activation in a
time- and dose-dependent manner in neurons derived from
both HD- and CON-iPSCs (Fig. 4 and Supplementary Material,
S3); this suggests that oxidative insult promotes DNA double-
strand breaks (DSBs) in the iPSC-derived neurons. However, H2O2
induced a signicantly greater level of H2AX in HD-iPSC-derived
 6070 | Human Molecular Genetics, 2015, Vol. 24, No. 21

Figure 3. Human HD-iPSCs differentiate into GABAergic neurons. ICC staining of GABAergic projection neuron markers in representative HD-iPSC-A1 (A), -B4 (B) and CON-
iPSC-1 (C) neural derivatives at 14 weeks. Nuclei were stained with DAPI. Scale bar: 50 m. (D) The percentage of DARPP-32 and GABA double-positive cells among DAPI-
labeled nuclei of neurons derived from HD-iPSC-A1 and HD-iPSC-B4, and CON-iPSC-1. At least 200 cells were counted for each experiment. (E) Western blotting with two
primary DARPP-32 antibodies ( purchased from Abcam and SCBT) in neurons derived from CON- and HD-iPSCs. Human brain tissue (hBT) and HD- and CON-iPSCs were
used as positive and negative controls, respectively. An antibody against GAPDH was used for the protein loading control.

neurons (Fig. 4CE). We also compared the global gene expression
proles of HD- and CON-iPSC-derived neurons by microarray ana-
lysis (Supplementary Material, Table S6), which revealed that
genes encoding proteins involved in the repair of DNA DSBs
(RAD51) and protection against oxidative stress (GSST1, GSST2)
were down-regulated in HD-iPSC-derived neurons. Collectively,
these data indicate that HD-iPSC neurons are more susceptible
to H2O2-induced DNA damage.
Activation of the A2AR protects HD-iPSC neurons
from H2O2-induced DNA damage and apoptosis
through a PKA-dependent pathway
Activation of the A2AR has been previously demonstrated to have
several benecial effects in genetic mouse models of HD
(23,29,37). To gain a better understanding of how the A2AR exerts
its neuroprotective effect, we rst examined the expression of the
A2AR in HD-iPSC-derived neurons. ICC analysis, RT-qPCR and
immunoblotting were used to demonstrate that iPSC-derived
neurons express the A2AR (Fig. 5), and that HD-iPSC-derived neu-
rons express a lower level of the A2AR than control iPSC-derived
neurons (Fig. 5B and C). We subsequently investigated the effect
of the A2AR on H2O2-induced DNA damage by treating HD-iPSC-
derived neurons with two selective A2AR agonists (CGS-21680,
CGS; 2-[(2-aminoethylamino) carbonylethylphenylethylamino]-
5-N- ethylcarboxamidoadenosine, APEC), and then determining
the responses of iPSC-derived neurons during elevation of
reactive oxygen species (ROS). We hypothesized that activation
of the A2AR may prevent DNA damage and cell death arising
from elevated ROS in HD-iPSC-derived neurons. To test this hy-
pothesis, we treated HD- and CON-iPSC-derived neurons with
either CGS or APEC, followed by H2O2 insult. As demonstrated
by ICC analysis and immunoblotting, both CGS (Fig. 6A, C, E
and Supplementary Material, Fig. S4A) and APEC (Fig. 6B and
D, and Supplementary Material, Fig. S4B) signicantly reduced
the H2O2-induced activation of H2AX in HD-iPSC-derived neu-
rons in a dosage-dependent manner. We also conrmed that
CGS has the same protective effect in CON-iPSC-derived neu-
rons (Supplementary Material, Fig. S4C and D). To investigate
whether the HD-iPSC-derived neurons are susceptible to cell
death upon H2O2 exposure, we treated HD- and CON-iPSC-
derived neurons with H2O2, and performed TUNEL assays and
measured active caspase 3 to quantify the percentage of
cells with apoptotic traits. We observed that H 2O2 treatment
resulted in a greater enhancement of active caspase 3 in HD-
iPSC-derived neurons than in their counterparts derived
from CON-iPSCs (Fig. 7C, D and F). Consistent with the
effects on caspase-3 activation, H 2 O 2 treatment resulted in a
signicantly higher proportion of TUNEL-positive HD-iPSC-
derived neurons than TUNEL-positive CON-iPSC-derived
neurons. Addition of CGS reduced the TUNEL-positive cells to
10% (Fig. 7G).
 Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6071

Figure 4. H2O2 stimulation induced greater H2AX activation in HD-iPSCs than in controls. CON-iPSC-1- and HD-iPSC-B16-derived neurons were treated with (A and C)
100 M H2O2 for the indicated times or (B and D) H2O2 at the indicated concentrations for 60 min. (A and B) Immunouorescence analysis of neurons with H2AX and
TUJ1. Nuclei were stained with DAPI. Scale bar: 20 m. Quantitative analysis of H2AX intensity from (C) Figure 4A and Supplementary Material, Figure S3A and (D)
Figure 4B and Supplementary Material, Figure S3B. The total intensity of H2AX was quantied and normalized to cell numbers by counting DAPI-labeled nuclei. At
least 500 cells were quantied for each condition of each experiment. Data are presented as the mean SEM of three independent experiments. *, compared with
CON-iPSC; P < 0.05 (by one-way ANOVA). (E) Western blot analyses of H2AX in the indicated CON-iPSC and HD-iPSC neuron derivatives following treatment with 0, 50,
or 100 M H2O2 for 60 min. GAPDH was used as a protein loading control. Histograms correspond with the lanes at the top of the adjacent panel (from left to right).
Data are representative of three independent experiments. *, P < 0.05 (by Students t test).

To determine the basis for the protective effect of A2AR agon-
ist on H2O2-induced DNA damage and apoptosis, we used im-
munoblotting to examine induction of H2AX and cleaved
capase-3 in CON-iPSC neurons transfected with A2AR-shRNA or
control-shRNA. A2AR protein was readily detected in cells trans-
fected with a scrambled control shRNA, while the level of expres-
sion was signicantly reduced in cells transfected with the A2AR
shRNA (Supplementary Material, Fig. S5A). As shown in Supple-
mentary Material, Figure S5B and S5C, treatment with CGS de-
creased the H2O2-mediated induction of H2AX and cleaved
caspase-3 in control-shRNA transfectants. In contrast, the pro-
tective effect was blocked in A2AR-shRNA transfectants. These
ndings support the conclusion that A2AR activation plays an im-
portant role in H2O2-induced DNA damage and apoptosis.
As it has been previously shown that A 2AR activation pre-
vents toxic stimuli-induced apoptosis through PKA activation
(38,39), we hypothesized that PKA activation may be involved
in the A2AR-mediated protection of HD-iPSC-derived neurons.
Hence, we pre-treated HD-iPSC-derived neurons with an A2AR
antagonist (SCH58261) and a selective PKA inhibitor (H-89)
prior to H2O2 and CGS21680 exposure, and subsequently exam-
ined the extent of DNA damage (H2AX) and apoptosis (TUNEL
and caspase 3 activation). As demonstrated by ICC analysis and
immunoblotting, we found that the protective effect of CGS
against both H2O2-induced DNA damage (Fig. 7A, B and E) and
cell death (Fig. 7C, D, F and G) can be abolished by treatment
with SCH58261 or H-89. Together, these data indicate that acti-
vation of the A2AR exerts a benecial effect on HD-iPSC-derived
 6072 | Human Molecular Genetics, 2015, Vol. 24, No. 21

Figure 5. HD-iPSC-derived neurons express the A2AR. (A) Immunouorescence analysis of CON-iPSC and HD-iPSC neuron derivatives stained with A2AR and DARPP-32
antibodies. Nuclei were stained with DAPI. Scale bar: 20 m. (B) Relative expression values of A2AR mRNA. Columns represent the average fold-change values obtained
from CON- and HD-iPSC-derived neurons at week 10 of neuron differentiation. Black, control neuron combination (CON-1 and CON-2); Gray, HD neuron combination (HD-
A1, A7, B4 and B16). Error bars represent SEM. (C) Western blot analyses of A2AR protein in CON-iPSC and HD-iPSC neuron derivatives. GAPDH was used as a protein loading
control. Histograms correspond with the lanes in the panel on the right (from left to right). Data are representative of three independent experiments. *, P < 0.05 (by
Students t test).

neurons, protecting them from DNA damage and cell death via
activation of PKA.
Discussion
Here, we describe the generation of human HD-iPSCs with a CAG
expansion mutation in the endogenous htt gene, and their
differentiation into HD-relevant, DARPP-32-positive, GABAergic
neurons. We conrmed that these cells exhibit selective vulner-
ability to oxidative stress, which is an important feature of HD
pathology. Using these HD-iPSC-derived neurons as an in vitro
model, we have demonstrated that activation of the A2AR by
two A2AR-selective agonists (CGS and APEC) can reduce oxidative
stress-induced DNA damage and neuronal apoptosis in HD iPSC-
derived neurons, through a PKA-dependent pathway. This study
demonstrates the feasibility of using human-iPSC-derived MSN
neurons as a drug screening platform for HD. Importantly, the
CAG repeat number in the htt gene of our HD-iPSCs is 43, which
is in the range for most HD patients, and is much lower than
that of previously described HD-iPSCs (8,9,31). Thus, the results
obtained from our HD-iPSC cells are likely to reect HD pathogen-
esis observed in the majority of HD patients. Collectively, our
ndings suggest that the human HD-iPSC model recapitulates
HD-relevant pathogenesis, and may thus be suitable for further
development into a platform for screening novel therapeutic
strategies.
Human iPSC-based models for HD have been developed to
replicate specic molecular and biological phenotypes. However,
consistent with previous reports (8,9,31,32), we did not observe
mutant HTT aggregation, an overt HD phenotype, in neurons de-
rived from HD-iPSCs. Recently, human HD iPSC lines were used to
derive neuronal precursors; these precursors exhibited a mild in-
crease in caspase activity upon growth factor deprivation (9,31),
and lysomal activity was signicantly increased in HD-iPSCs
and HD-iPSC-derived neurons as compared with their CON-
iPSC counterparts (32). The HD-iPSC consortium and Carter
et al. (8,40) reported that HD-iPSC differentiated cultures were
highly susceptible to stressors, including BDNF withdrawal,
glutamate pulsing, or the addition of cellular stressors (H2O2
and 3-MA). Our results demonstrate that oxidative stress is in-
volved in the pathogenesis of HD; HD-iPSC-derived neurons
were more susceptible to H2O2-induced DNA damage and cell
death than their counterparts derived from normal iPSCs.
These results are consistent with previous reports that H2O2
treatment resulted in a strong DNA damage response in bro-
blasts derived from HD patients (41), and increased cell death in
human HD iPSCs (8,40). It has also been suggested that the asso-
ciation between HD and increased ROS and oxidative stimuli
(8,42) underlies mutant htt-dependent cell death (8,43).
Recent evidence indicates that oxidative stress-induced de-
fects in DNA damage repair can lead to neuronal cell death
(33,44,45). Our microarray analysis data (Supplementary Mater-
ial, Table S6) revealed that expression of RAD51, which is in-
volved in recombination and repair of damaged DNA, is
reduced in HD-iPSC neuron derivatives as compared with coun-
terparts from CON-iPSCs. We also observed that expression of
genes encoding cellular glutathione antioxidant enzymes [gluta-
thione S-transferase theta 1 and 2 (GSTT1, GSST2) and Glutathi-
one peroxidase 8 (GPX-8)] was slightly decreased in HD-iPSC-
derived neurons (Supplementary Material, Table S6). These
results are consistent with the reported increase in oxidative
stress biomarkers and reduction in antioxidant systems in HD
patients (42). Despite the presence of such defects, we did not ob-
serve increased mortality under basal neuronal culture condi-
tions in the long term. Nevertheless, these changes may
increase the sensitivity of HD-iPSC-derived neurons to oxidative
stress, thereby resulting in the accumulation of ROS, and conse-
quent DNA damage and cell death. Moreover, our results revealed
that one of the protective functions of the A2AR in human MSN
neurons derived from HD-iPSCs is the prevention of DNA damage
 Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6073

Figure 6. Stimulation of the A2AR decreased H2O2-induced H2AX activation in HD-iPSCs. HD-iPSC-A1 and -B16-derived neurons were treated with (A) CGS or (B) APEC at
the indicated concentrations for 1 h, and then exposed to 100 M H2O2 for 1 h. (A and B) Immunouorescence analysis of neurons stained with H2AX and TUJ1. Nuclei
were stained with DAPI (blue). Scale bar: 20 m. (C and D) Quantitative analysis of the intensity of H2AX from Figure 6A and B, respectively. The total intensity of H2AX
was quantied and normalized to total cell number by counting DAPI-positive nuclei. At least 500 cells were quantied for each condition of each experiment. Data are
presented as the mean SEM of three independent experiments. *, P < 0.05 (by one-way ANOVA). (E) Immunoblotting against H2AX in iPSC-derived neurons, following
treatment with CGS at the indicated concentrations for 1 h and subsequent exposure to 100 M H2O2 for 1 h. GAPDH was used as a protein loading control. Histograms
correspond with the lanes at the top of the adjacent panel (from left to right). Data are representative of three independent experiments. *, P < 0.05 (by Students t test).

evoked by elevated oxidative stress. This new nding provides
important insights into the function of the A2AR as a potential
therapeutic target for HD in human neurons.
Although the gene causing HD is known, the mechanism
underlying the increased vulnerability of a specic neuronal
population in HD remains elusive. Previous studies have sug-
gested a role of the A2AR in cell survival. For example, stimulation
of the A2AR has been shown to protect against cell death induced
by hypoxia (46) and serum-deprivation (39,47) in PC12 cells, and
NGF withdrawal in sympathetic neurons (48). Our current nd-
ings demonstrate that CGS and APEC prevent DNA damage and
apoptosis induced by H2O2 in HD-iPSC-derived neurons (Figs 6
and 7), suggesting that the A2AR exerts a positive effect on the
survival of human HD-iPSC-derived neurons. Additionally, we
conrmed the protective effect of an A2AR agonist on H2O2-
induced DNA damage and apoptosis by shRNA knockdown of
 6074 | Human Molecular Genetics, 2015, Vol. 24, No. 21

Figure 7. The protective effects of A2AR agonists are mediated through activation of PKA. The HD-iPSC-derived neurons were pre-treated with SCH58261 (SCH, 10 M) or
H-89 (H89, 10 M) and then treated with CGS21680 (CGS, 10 M) and 100 M H2O2 for 1 h (DNA damage) or 6 h (cell death), respectively. Immunouorescence analysis of
neurons stained with (A) H2AX and TUJ1 or (C) active caspase-3 and TUJ1. Nuclei were stained with DAPI. Scale bar: 20 m. Quantitative analysis of the intensity of (B)
H2AX and (D) active caspase-3, as normalized to total cell number by counting DAPI-labeled nuclei. At least 500 cells were quantied for each condition of each
experiment. Immunoblotting of neurons with antibodies against (E) H2AX and (F) active caspase 3. GAPDH was used as a protein loading control. Histograms
correspond with the lanes at the top of the adjacent panel (from left to right). Data are representative of three independent experiments. (G) Percentage of TUNEL-
positive cells within total cells. Data are presented as the mean SEM of three independent experiments. At least 200 cells were quantied for each condition of each
experiment. *, P < 0.05 [by one-way ANOVA in (B) and (D); by Students t test in (E), (F) and (G)].

the A2AR. Our results demonstrated that CGS treatment de- Fig. S5B and S5C). These ndings further support the conclusion
creased DNA damage and apoptosis induced by H2O2 insult in that A2AR activation plays an important role in H2O2-induced
control-shRNA transfectants, whereas the protective effect was DNA damage and apoptosis. Our ndings are consistent with pre-
blocked in A2AR-shRNA transfectants (Supplementary Material, vious reports, which showed that A2AR agonists are benecial in a

transgenic mouse model of HD (R6/2) and mouse striatal progeni-
tor cell lines (Hdh 7Q and Hdh 109Q) (23,37,4951). In line with the
aforementioned observations supportive of a protective role for
the A2AR in human and rodent models of HD, genetic removal
of the A2AR signicantly exacerbates the progression of HD in a
HD mouse model (N171-82Q) (52). Together, these ndings
strongly suggest that the A2AR is a worthwhile drug target for HD.
Stimulation of the A2AR activates at least two major cellular
signaling cascades: adenylyl cyclase/cAMP/PKA and PKC-
mediated pathways (19). Several reports have indicated that com-
ponents involved in regulating the PKA pathway are disrupted
during HD progression (22,53,54). Therefore, the role of PKA in
the protective function of the A2AR is worthy of further evalu-
ation. A2AR-mediated protection against H2O2-induced DNA
damage and cell death in HD-iPSC-derived neurons appears to
be mediated by PKA, because such protection was blocked by a
PKA inhibitor, H-89 (Fig. 7). Our results are consistent with earlier
reports, which demonstrated that A2AR activation protects cells
against hypoxia and growth factor withdrawal via PKA signaling
(39,46,55).
The ability of iPSCs to differentiate into GABAergic MSNs, the
most susceptible cell type in HD pathogenesis, will be the crux to
successfully model HD in culture. Procedures for inducing differ-
entiation of hESCs and iPSCs into GABAergic neurons have been
described previously (79,56,57). These protocols often involved
treatment with various morphogens, such as SHH and WNT
inhibitor (DKK1), to direct hESC differentiation toward the GA-
BAergic neuronal lineage (79,57). By using a step-wise in vitro dif-
ferentiation protocol combining EB formation, neural induction
by small molecules, treatment with inhibitors of the TGF path-
way (SB431542) and the BMP pathway (LDN193189) (58), and
mechanical isolation/purication of neural progenitors and
neurons, we induced 6070% of control iPSCs or HD-iPSCs to
differentiate into GABA- and DARPP-32-double positive neu-
rons (Fig. 3D). The high yield of GABA- and DARPP-32-double
positive neurons obtained may be attributed to effective neur-
al induction by the BMP and TGF inhibitors and the mechan-
ical isolation/purication of neural progenitors and early
neural cells. This is of particular interest because the default
neuronal differentiation of human expanded neuron precur-
sors typically produces GABAergic neurons, irrespective of
the region from which these cells were harvested from the
embryo (59,60).
In summary, we have elucidated the role of the A2AR in HD
using HD-iPSC-derived neurons as a disease modeling system.
Our results imply that A2AR activation may protect MSN against
ROS-induced damage, and suggest that the A2AR may be a
novel therapeutic target for HD.
Materials and Methods
Reagents
SB431542, 2-[ p-(2-carboxyethyl) phenylethylamino]-5-N-ethyl-
carboxamidoadenosine [CGS21680, (61)], and 5-amino-7-(2-phe-
nylethyl)-2-(2-furyl)- pyrazolo [4,3-epsilon]-1,2,4-triazolo[1,5-c)
pyrimidine [SCH58261, (62)] were purchased from Sigma Aldrich
(St Louis, Mom USA). LDN193189 was purchased from STEM-
GENT (San Diego, CA, USA). H-89 was purchased from Bio-
Mol (Plymouth Meeting, PA, USA). Adenosine deaminase
was purchased from Roche (Nutley, NJ, USA). 2-[(2-aminoethy-
lamino) carbonylethylphenylethylamino]-5-N- ethylcarboxami-
doadenosine [APEC, (63)] was a generous gift from Dr Kenneth
A. Jacobson.
Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6075
Isolation of broblasts from patients with HD
and generation of HD-iPSCs
Dermal broblasts from two patients (A and B) with HD were ob-
tained through skin biopsy; written informed consent was ob-
tained in accordance with the protocol approved by the Internal
Research Board of CHANG GUNG Memorial Hospital. Four human
non-HD control iPSC lines (CON-iPSC-1, 2, 3 and 5) were derived
from adult foreskin broblasts (CFB-50 iPSC) (64), newborn fore-
skin broblasts (ATCC, CRL-2429), adult dermal broblasts and
adult peripheral blood mononuclear cells (PBMCs), respectively
(Supplementary Material, Table S1). Derivation of iPSCs was per-
formed as described previously (64).
Cultivation and neural differentiation of hiPSCs
Undifferentiated human HD-iPSCs (HD-iPSC-A1, A7, B4 and B16)
and human CON-iPSCs (CON-iPSC-1, 2, 3 and 5) were routinely
maintained and subcultivated as previously described (64). For
neural differentiation, iPSCs were dissociated with 1 mg/ml dis-
pase (Sigma), and then left to form embryoid bodies (EBs) in
ultra-low attachment dishes (Corning) in hESC media [Dulbeccos
modied Eagles medium (DMEM)/F12, 20% knockout serum re-
placement, 1% non-essential amino acids (NEAA), 2 m L-glu-
tamine, 100 m 2-mercaptoethanol and 5 ng/ml basic broblast
growth factor (bFGF)] containing 10 M SB431542 and 100 nM
LDN193189 for 3 days; EBs were then transferred to N2 media
[DMEM/F12, 1X N2 supplement, 1% NEAA, 1 m sodium pyru-
vate, 2 m L-glutamine and 20 ng/ml bFGF (all purchased from
Invitrogen)] containing 10 M SB431542 and 100 nM LDN193189
for another 2 days. Next, EBs were plated on tissue culture dishes
(Nunc) in N2 media, and media were replaced every 2 days. After
34 weeks of differentiation, neural rosette structures (termed
neural stem cells; NSCs) were manually picked and transferred
to Matrigel-coated culture dishes in N2 medium. The iPSC-
derived NSCs were expanded as spherical aggregates.
For striatal differentiation, the iPSC-derived neurospheres
were picked and cut into small pieces (12 mm in diameter)
using a needle on a weekly basis, and then plated onto new Ma-
trigel-coated dishes in N2/B27 medium [(DMEM/F12: Neurobasal
medium (1 : 1), 0.5X N2, 0.5X B27, 1% NEAA, 0.5 m sodium pyru-
vate, 2 m L-glutamine and 10 ng/ml bFGF)] for 4 weeks; the
media were subsequently replaced with B27 medium [Neuroba-
sal medium, 1X B27, 1% NEAA, 2 m L-glutamine] for another
46 weeks. At weeks 4, 8 and 12 of neural differentiation, viable
cells were dissociated with TrypLE (Invitrogen, Carlsbad, CA,
USA) and counted using the Trypan blue exclusion assay in a
hemocytometer.
Cell stress, DNA damage and apoptosis assays
Neurons derived from iPSCs were dissociated into single cells
using TrypLE, seeded onto Matrigel-coated coverslips or appro-
priate culture dishes and then cultured for an additional 3 days
to enhance cell attachment. Cells were pre-incubated with ad-
enosine deaminase (ADA) (1 unit/ml) for 4 h to eliminate en-
dogenous adenosine, and then treated with an A2AR agonist
(CGS21680 or APEC) of the indicated concentration for 1 h, prior
to the addition of H2O2 (100 M). Cells were then incubated with
H2O2 for 1 and 6 h for the measurement of H2AX and caspase-
3 activation, respectively. To characterize the underlying mech-
anism, an A2AR antagonist (SCH58261) or a PKA inhibitor (H-89)
was applied after ADA treatment (as described above) for 10
or 30 min, respectively, followed by treatment with the indi-
cated A 2A R agonist. Cells were xed and permeabilized for
 6076 | Human Molecular Genetics, 2015, Vol. 24, No. 21
immunocytochemical analysis and the terminal deoxynucleo-
tidyl transferase-mediated dUTP-biotin nick end-labeling
(TUNEL) assay. The TUNEL assay was performed using the
DeadEnd Fluorometric TUNEL System (Promega, Wisconsin,
USA), according to the manufacturers instructions. The per-
centage of TUNEL-positive cells was calculated from the total
number of cells from three independent experiments. At
least 200 cells were quantied in each case.
Genomic DNA and RNA extraction and PCR analysis
of gene expression
Genomic DNA and total RNA were extracted with Trizol reagent
(Invitrogen, Carlsbad, CA, USA), according to the manufacturers
recommendations. RNA (1 g) was treated with DNase I (Qiagen,
Germantown, MD, USA) and puried using the RNeasy kit (Qia-
gen, Germantown, MD, USA), before being reverse-transcribed
using Superscript II (Invitrogen). The expression of total, en-
dogenous and viral transgenes of OCT4, SOX2, KLF4 and MYC,
as well as neural-related genes, were examined using semi-
quantitative PCR, with previously described primers (65,66)
and those in Supplementary Material, Tables S2 and S3. Quan-
titative PCR (RT-qPCR) was performed with SYBR Green Master
Mix and an Applied Biosystems 7900HT Fast Real-Time PCR Sys-
tem using gene-specic primers (Supplementary Material,
Table S3).
Immunocytochemistry (ICC), immunohistochemistry
and microscopy
Immunostaining was performed as previously described (29).
Cells were xed in 4% paraformaldehyde/DPBS for 20 min, per-
meabilized with 0.05% NP-40 in PBS for 20 min and then blocked
in 3% normal goat serum/DPBS for 1 h. Cells were incubated with
primary antibodies overnight at 4C, washed and incubated with
the corresponding secondary antibody for 60 min. Cells were
counterstained with 1 g/ml DAPI/DPBS for 3 min. The primary
and secondary antibodies are listed in Supplementary Material,
Table S4. Images were captured on an Olympus Ix71 microscope
or a laser-scanning confocal microscope LSM700-meta (Carl
Zeiss). The intensity of uorescence was quantied using
Image J software (National Institutes of Health, USA), and the
relative intensity of uorescence was normalized to the cell num-
ber. At least three randomly selected elds in each coverslip with
a total of over 500 cells were counted for each case. For histo-
chemical analysis, H&E staining was performed using an Auto-
stainer XL Leica ST5010 (Leica Microsystem).
Immunoblotting
Total proteins were extracted using RIPA lysis buffer fortied
with protease inhibitors (Complete Mini; Roche-Boehringer). For
detection of the A2AR, protein lysate was extracted using RIPA
lysis buffer containing 4 M urea. Protein concentration was deter-
mined using the BCA Protein Assay Kit (Pierce). Samples (1025 g
each) were then subjected to SDS-PAGE, before being electro-
blotted onto a PVDF membrane (Millipore). Membranes were
blocked in 1% BSA in PBS prior to sequential probing with primary
antibody and HRP-conjugated secondary antibody in blocking
solution. Target proteins were visualized by ECL (Amersham-
Pharmacia) with an LAS-4000 mini device (FUJI Photo Film).
Primary and secondary antibodies used in immunoblotting are
listed in Supplementary Material, Table S4.
Knockdown of the A2AR
The A2AR knockdown constructs were produced by subcloning
the 21-nucleotide shRNA sequences of the human A2AR gene
(Adora2a) (Adora2a shRNA-1, 5-AGACCTTCCGCAAGATCATTC-3;
Adora2a shRNA-2, 5- CAGAACGTCACCAACTACTTT-3) into the
pLKO.1-AS1010 vector. Preparation and transfection of lentiviral
vectors were performed as previously described (67). CON-iPSC-
derived neurons were transduced with lentiviral vectors producing
shRNA targeting the A2AR at a multiplicity of three in the presence
of polybrene (5 g/mL). Proteins were extracted from the trans-
duced cells at 4 days post-transduction.
Teratoma generation
For teratoma generation, 5 106 undifferentiated iPSCs were sub-
cutaneously injected into NOD-SCID mice. After 1012 weeks,
mice were sacriced and teratomas were isolated. All animal ex-
periments were approved by the Animal Care and Use Committee
of Academia Sinica, and performed in accordance with the Insti-
tutional Animal Care and Use Committee guidelines of Academia
Sinica.
Microarray analysis
Total RNA was extracted from CON-iPSC-1, CON-iPSC-5, HD-iPSC-
A1, HD-iPSC-A7 and HD-iPSC-B16-derived neurons at week 10 of
neural differentiation. All gene expression data were obtained
using the Affymetrix microarray platform by the Affymetrix
Gene Expression Service Laboratory at Academia Sinica, Taiwan.
Chips were scanned with an Affymetrix GeneChip Scanner 7G,
and data were analyzed with GeneSpring X software (Agilent,
Santa Clara, CA, USA). Raw data were normalized independently
for each cell line using Robust Multichip Average. The NCBI acces-
sion number for the microarray data reported in this article is
GSE59051. Changes in gene expression with a fold change >1
and an adjusted P < 0.05 were considered to be signicantly
different.
Statistical analyses
Unless stated otherwise, all experiments were performed in trip-
licate and the results are shown as the mean SEM. One-way
analysis of variance (ANOVA) and Students t test were used for
statistical analysis; P < 0.05 was considered signicant.
Supplementary Material
Supplementary Material is available at HMG online.
Acknowledgements
We thank Dr Kenneth A. Jacobson (National Institute of Diabetes
and Digestive and Kidney Diseases, National Institutes of Health,
USA) and the Chemical Synthesis Program of the National Insti-
tute of Mental Health (Rockville, MD, USA) for the generous gift of
APEC. We thank Lee Stone and Chun-Ying Yu for technical
support.
Conict of Interest statement. None of the co-authors have any con-
icts of interest to declare.

Funding
This work was supported by grants from the National Science
Council (102-2321-B-001 -067 -MY3 to H.-C.K.; 100-2320-B-001-
0110MY3 to Y.C.) and Academia Sinica (AS-103-TP-B10 and Sum-
mit project grant to H.-C.K.; AS-103-TP-B10 and 103-Academia
Sinica Investigation Award-06 to Y.C.).
References
1. Kelly, C.M., Dunnett, S.B. and Rosser, A.E. (2009) Medium
spiny neurons for transplantation in Huntingtons disease.
Biochem. Soc. Trans., 37, 323328.
2. Reiner, A., Albin, R.L., Anderson, K.D., DAmato, C.J., Penney,
J.B. and Young, A.B. (1988) Differential loss of striatal projec-
tion neurons in Huntington disease. Proc. Natl. Acad. Sci
USA., 85, 57335737.
3. Pouladi, M.A., Morton, A.J. and Hayden, M.R. (2013) Choosing
an animal model for the study of Huntingtons disease. Nat.
Re. Neurosci., 14, 708721.
4. Dimos, J.T., Griswold-Prenner, I., Grskovic, M., Irion, S.,
Johnson, C. and Vaisberg, E. (2011) Induced pluripotent
stem cells as human disease models. Ann. Rep. Med. Chem.,
46, 369383.
5. Soldner, F. and Jaenisch, R. (2012) Medicine. iPSC disease
modeling. Science, 338, 11551156.
6. Ivkovic, S. and Ehrlich, M.E. (1999) Expression of the striatal
DARPP-32/ARPP-21 phenotype in GABAergic neurons re-
quires neurotrophins in vivo and in vitro. J. Neurosci., 19,
54095419.
7. Aubry, L., Bugi, A., Lefort, N., Rousseau, F., Peschanski, M. and
Perrier, A.L. (2008) Striatal progenitors derived from human
ES cells mature into DARPP32 neurons in vitro and in quino-
linic acid-lesioned rats. Proc. Natl. Acad. Sci. USA, 105, 16707
16712.
8. HD iPSC Consortium (2012) Induced pluripotent stem cells
from patients with Huntingtons disease show CAG-repeat-
expansion-associated phenotypes. Cell Stem Cell, 11, 264278.
9. Zhang, N., An, M.C., Montoro, D. and Ellerby, L.M. (2010) Char-
acterization of human Huntingtons disease cell model from
induced pluripotent stem cells. PLoS Curr., 2, Rrn1193.
10. Kim, S.H., Thomas, C.A., Andre, V.M., Cummings, D.M.,
Cepeda, C., Levine, M.S. and Ehrlich, M.E. (2011) Forebrain
striatal-specic expression of mutant huntingtin protein
in vivo induces cell-autonomous age-dependent alterations
in sensitivity to excitotoxicity and mitochondrial function.
ASN Neuro., 3, e0060.
11. Graham, R.K., Pouladi, M.A., Joshi, P., Lu, G., Deng, Y., Wu, N.P.,
Figueroa, B.E., Metzler, M., Andre, V.M., Slow, E.J. et al. (2009)
Differential susceptibility to excitotoxic stress in YAC128
mouse models of Huntington disease between initiation
and progression of disease. J. Neurosci., 29, 21932204.
12. Thomas, E.A., Coppola, G., Tang, B., Kuhn, A., Kim, S., Gesch-
wind, D.H., Brown, T.B., Luthi-Carter, R. and Ehrlich, M.E.
(2011) In vivo cell-autonomous transcriptional abnormalities
revealed in mice expressing mutant huntingtin in striatal but
not cortical neurons. Hum. Mol. Genet., 20, 10491060.
13. Rosenstock, T.R., Bertoncini, C.R., Teles, A.V., Hirata, H., Fer-
nandes, M.J. and Smaili, S.S. (2010) Glutamate-induced al-
terations in Ca2+ signaling are modulated by mitochondrial
Ca2+ handling capacity in brain slices of R6/1 transgenic
mice. Eur. J. Neurosci., 32, 6070.
14. Panov, A.V., Gutekunst, C.A., Leavitt, B.R., Hayden, M.R.,
Burke, J.R., Strittmatter, W.J. and Greenamyre, J.T. (2002)
Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6077
Early mitochondrial calcium defects in Huntingtons disease
are a direct effect of polyglutamines. Nat. Neurosci., 5, 731
736.
15. Seo, H., Sonntag, K.C. and Isacson, O. (2004) Generalized brain
and skin proteasome inhibition in Huntingtons disease. Ann.
Neurol., 56, 319328.
16. Wang, J., Wang, C.E., Orr, A., Tydlacka, S., Li, S.H. and Li, X.J.
(2008) Impaired ubiquitin-proteasome system activity in the
synapses of Huntingtons disease mice. J. Cell. Biol., 180, 1177
1189.
17. Brown, T.B., Bogush, A.I. and Ehrlich, M.E. (2008) Neocortical
expression of mutant huntingtin is not required for altera-
tions in striatal gene expression or motor dysfunction in a
transgenic mouse. Hum. Mol. Genet., 17, 30953104.
18. Chern, Y., King, K., Lai, H.L. and Lai, H.T. (1992) Molecular
cloning of a novel adenosine receptor gene from rat brain. Bio-
chem. Biophys. Res. Commun., 185, 304309.
19. Chern, Y., Lai, H.L., Fong, J.C. and Liang, Y. (1993) Multiple me-
chanisms for desensitization of A2a adenosine receptor-
mediated cAMP elevation in rat pheochromocytoma PC12
cells. Mol. Pharmacol., 44, 950958.
20. Rosin, D.L., Robeva, A., Woodard, R.L., Guyenet, P.G. and
Linden, J. (1998) Immunohistochemical localization of ad-
enosine A2A receptors in the rat central nervous system.
J. Comp. Neurol., 401, 163186.
21. Shearman, L.P. and Weaver, D.R. (1997) [125I]4-aminobenzyl-
5-N-methylcarboxamidoadenosine (125I)AB-MECA) labels
multiple adenosine receptor subtypes in rat brain. Brain
Res., 745, 1020.
22. Chiang, M.C., Lee, Y.C., Huang, C.L. and Chern, Y. (2005) cAMP-
response element-binding protein contributes to sup-
pression of the A2A adenosine receptor promoter by mutant
Huntingtin with expanded polyglutamine residues. J. Biol.
Chem., 280, 1433114340.
23. Chou, S.Y., Lee, Y.C., Chen, H.M., Chiang, M.C., Lai, H.L.,
Chang, H.H., Wu, Y.C., Sun, C.N., Chien, C.L., Lin, Y.S. et al.
(2005) CGS21680 attenuates symptoms of Huntingtons dis-
ease in a transgenic mouse model. J. Neurochem., 93, 310320.
24. Martinez-Mir, M.I., Probst, A. and Palacios, J.M. (1991) Adeno-
sine A2 receptors: selective localization in the human basal
ganglia and alterations with disease. Neuroscience, 42,
697706.
25. Popoli, P., Blum, D., Domenici, M.R., Burnouf, S. and Chern, Y.
(2008) A critical evaluation of adenosine A2A receptors as po-
tentially druggable targets in Huntingtons disease. Curr.
Pharm. Des., 14, 15001511.
26. Canals, M., Marcellino, D., Fanelli, F., Ciruela, F., de Benedetti,
P., Goldberg, S.R., Neve, K., Fuxe, K., Agnati, L.F., Woods, A.S.
et al. (2003) Adenosine A2A-dopamine D2 receptor-receptor
heteromerization: qualitative and quantitative assessment
by uorescence and bioluminescence energy transfer. J. Biol.
Chem., 278, 4674146749.
27. Flajolet, M., Wang, Z., Futter, M., Shen, W., Nuangchamnong,
N., Bendor, J., Wallach, I., Nairn, A.C., Surmeier, D.J. and
Greengard, P. (2008) FGF acts as a co-transmitter through ad-
enosine A(2A) receptor to regulate synaptic plasticity. Nat.
Neurosci., 11, 14021409.
28. Gomes, C.A., Vaz, S.H., Ribeiro, J.A. and Sebastiao, A.M. (2006)
Glial cell line-derived neurotrophic factor (GDNF) enhances
dopamine release from striatal nerve endings in an adeno-
sine A2A receptor-dependent manner. Brain Res., 1113, 129
136.
29. Lin, J.T., Chang, W.C., Chen, H.M., Lai, H.L., Chen, C.Y., Tao, M.
H. and Chern, Y. (2013) Regulation of feedback between
 6078 | Human Molecular Genetics, 2015, Vol. 24, No. 21
protein kinase A and the proteasome system worsens Hun-
tingtons disease. Mol. Cell Biol., 33, 10731084.
30. Ferrante, A., Martire, A., Armida, M., Chiodi, V., Pezzola, A.,
Potenza, R.L., Domenici, M.R. and Popoli, P. (2010) Inuence
of CGS 21680, a selective adenosine A(2A) receptor agonist,
on NMDA receptor function and expression in the brain of
Huntingtons disease mice. Brain Res., 1323, 184191.
31. An, M.C., Zhang, N., Scott, G., Montoro, D., Wittkop, T., Moon-
ey, S., Melov, S. and Ellerby, L.M. (2012) Genetic correction of
Huntingtons disease phenotypes in induced pluripotent
stem cells. Cell Stem Cell, 11, 253263.
32. Camnasio, S., Delli Carri, A., Lombardo, A., Grad, I., Mariotti,
C., Castucci, A., Rozell, B., Lo Riso, P., Castiglioni, V., Zuccato,
C. et al. (2012) The rst reported generation of several induced
pluripotent stem cell lines from homozygous and heterozy-
gous Huntingtons disease patients demonstrates mutation
related enhanced lysosomal activity. Neurobiol. Dis., 46, 4151.
33. Browne, S.E., Bowling, A.C., MacGarvey, U., Baik, M.J., Berger,
S.C., Muqit, M.M., Bird, E.D. and Beal, M.F. (1997) Oxidative
damage and metabolic dysfunction in Huntingtons disease:
selective vulnerability of the basal ganglia. Ann. Neurol., 41,
646653.
34. Browne, S.E., Ferrante, R.J. and Beal, M.F. (1999) Oxidative
stress in Huntingtons disease. Brain Pathol., 9, 147163.
35. Chen, C.M. (2011) Mitochondrial dysfunction, metabolic de-
cits, and increased oxidative stress in Huntingtons disease.
Chang Gung Med. J., 34, 135152.
36. Rogakou, E.P., Pilch, D.R., Orr, A.H., Ivanova, V.S. and Bonner,
W.M. (1998) DNA double-stranded breaks induce histone
H2AX phosphorylation on serine 139. J. Biol. Chem., 273,
58585868.
37. Ju, T.C., Chen, H.M., Lin, J.T., Chang, C.P., Chang, W.C., Kang, J.
J., Sun, C.P., Tao, M.H., Tu, P.H., Chang, C. et al. (2011) Nuclear
translocation of AMPK-alpha1 potentiates striatal neurode-
generation in Huntingtons disease. J. Cell. Biol., 194, 209227.
38. Bshesh, K., Zhao, B., Spight, D., Biaggioni, I., Feokistov, I., De-
nenberg, A., Wong, H.R. and Shanley, T.P. (2002) The A2A re-
ceptor mediates an endogenous regulatory pathway of
cytokine expression in THP-1 cells. J. Leukoc. Biol., 72, 1027
1036.
39. Huang, N.K., Lin, Y.W., Huang, C.L., Messing, R.O. and Chern, Y.
(2001) Activation of protein kinase A and atypical protein kinase
C by A(2A) adenosine receptors antagonizes apoptosis due to
serum deprivation in PC12 cells. J. Biol. Chem., 276, 1383813846.
40. Carter, R.L., Chen, Y., Kunkanjanawan, T., Xu, Y., Moran, S.P.,
Putkhao, K., Yang, J., Huang, A.H., Parnpai, R. and Chan, A.W.
(2014) Reversal of cellular phenotypes in neural cells derived
from Huntingtons disease monkey-induced pluripotent
stem cells. Stem Cell Reports, 3, 585593.
41. Giuliano, P., De Cristofaro, T., Affaitati, A., Pizzulo, G.M., Feli-
ciello, A., Criscuolo, C., De Michele, G., Filla, A., Avvedimento,
E.V. and Varrone, S. (2003) DNA damage induced by polyglu-
tamine-expanded proteins. Hum. Mol. Genet., 12, 23012309.
42. Tunez, I., Sanchez-Lopez, F., Aguera, E., Fernandez-Bolanos,
R., Sanchez, F.M. and Tasset-Cuevas, I. (2011) Important role
of oxidative stress biomarkers in Huntingtons disease. J. Med.
Chem., 54, 56025606.
43. Goswami, A., Dikshit, P., Mishra, A., Mulherkar, S., Nukina, N.
and Jana, N.R. (2006) Oxidative stress promotes mutant hun-
tingtin aggregation and mutant huntingtin-dependent cell
death by mimicking proteasomal malfunction. Biochem. Bio-
phys. Res. Commun., 342, 184190.
44. Bogdanov, M.B., Andreassen, O.A., Dedeoglu, A., Ferrante, R.J.
and Beal, M.F. (2001) Increased oxidative damage to DNA in a
transgenic mouse model of Huntingtons disease. J. Neuro-
chem., 79, 12461249.
45. Schilling, G., Cooneld, M.L., Ross, C.A. and Borchelt, D.R.
(2001) Coenzyme Q10 and remacemide hydrochloride ameli-
orate motor decits in a Huntingtons disease transgenic
mouse model. Neurosci. Lett., 315, 149153.
46. Kobayashi, S. and Millhorn, D.E. (1999) Stimulation of expres-
sion for the adenosine A2A receptor gene by hypoxia in PC12
cells. A potential role in cell protection. J. Biol. Chem., 274,
2035820365.
47. Lee, Y.C., Yang, Y.C., Huang, C.L., Kuo, T.Y., Lin, J.H., Yang, D.
M. and Huang, N.K. (2012) When cytokinin, a plant hormone,
meets the adenosine A2A receptor: a novel neuroprotectant
and lead for treating neurodegenerative disorders? PLoS
One, 7, e38865.
48. Ramirez, S.H., Fan, S., Maguire, C.A., Perry, S., Hardiek, K.,
Ramkumar, V., Gelbard, H.A., Dewhurst, S. and Maggirwar,
S.B. (2004) Activation of adenosine A2A receptor protects
sympathetic neurons against nerve growth factor withdraw-
al. J. Neurosci. Res., 77, 258269.
49. Chiang, M.C., Chen, H.M., Lai, H.L., Chen, H.W., Chou, S.Y.,
Chen, C.M., Tsai, F.J. and Chern, Y. (2009) The A2A adenosine
receptor rescues the urea cycle deciency of Huntingtons
disease by enhancing the activity of the ubiquitin-prote-
asome system. Hum. Mol. Genet., 18, 29292942.
50. Huang, N.K., Lin, J.H., Lin, J.T., Lin, C.I., Liu, E.M., Lin, C.J.,
Chen, W.P., Shen, Y.C., Chen, H.M., Chen, J.B. et al. (2011) A
new drug design targeting the adenosinergic system for Hun-
tingtons disease. PLoS One, 6, e20934.
51. Martire, A., Pepponi, R., Domenici, M.R., Ferrante, A., Chiodi,
V. and Popoli, P. (2013) BDNF prevents NMDA-induced toxicity
in models of Huntingtons disease: the effects are genotype
specic and adenosine A(2A) receptor is involved.
J. Neurochem., 125, 225235.
52. Mievis, S., Blum, D. and Ledent, C. (2011) A2A receptor knock-
out worsens survival and motor behaviour in a transgenic
mouse model of Huntingtons disease. Neurobiol. Dis., 41,
570576.
53. Giampa, C., Laurenti, D., Anzilotti, S., Bernardi, G., Menniti, F.
S. and Fusco, F.R. (2010) Inhibition of the striatal specic
phosphodiesterase PDE10A ameliorates striatal and cortical
pathology in R6/2 mouse model of Huntingtons disease.
PLoS One, 5, e13417.
54. Hebb, A.L., Robertson, H.A. and Denovan-Wright, E.M. (2004)
Striatal phosphodiesterase mRNA and protein levels are re-
duced in Huntingtons disease transgenic mice prior to the
onset of motor symptoms. Neuroscience, 123, 967981.
55. Kobayashi, S., Conforti, L., Pun, R.Y. and Millhorn, D.E.
(1998) Adenosine modulates hypoxia-induced responses
in rat PC12 cells via the A2A receptor. J. Physiol., 508 (Pt 1),
95107.
56. Jeon, I., Lee, N., Li, J.Y., Park, I.H., Park, K.S., Moon, J., Shim, S.
H., Choi, C., Chang, D.J., Kwon, J. et al. (2012) Neuronal proper-
ties, in vivo effects, and pathology of a Huntingtons disease
patient-derived induced pluripotent stem cells. Stem Cells, 30,
20542062.
57. Delli Carri, A., Onorati, M., Lelos, M.J., Castiglioni, V., Faedo, A.,
Menon, R., Camnasio, S., Vuono, R., Spaiardi, P., Talpo, F. et al.
(2013) Developmentally coordinated extrinsic signals drive
human pluripotent stem cell differentiation toward authen-
tic DARPP-32+ medium-sized spiny neurons. Development,
140, 301312.
58. Chambers, S.M., Fasano, C.A., Papapetrou, E.P., Tomishima,
M., Sadelain, M. and Studer, L. (2009) Highly efcient neural

conversion of human ES and iPS cells by dual inhibition of
SMAD signaling. Nat. Biotechnol., 27, 275280.
59. Carpenter, M.K., Cui, X., Hu, Z.Y., Jackson, J., Sherman, S., Sei-
ger, A. and Wahlberg, L.U. (1999) In vitro expansion of a multi-
potent population of human neural progenitor cells. Exp.
Neurol., 158, 265278.
60. Jain, M., Armstrong, R.J., Tyers, P., Barker, R.A. and Rosser, A.E.
(2003) GABAergic immunoreactivity is predominant in neu-
rons derived from expanded human neural precursor cells
in vitro. Exp. Neurol., 182, 113123.
61. Zhang, G., Franklin, P.H. and Murray, T.F. (1994) Activation of
adenosine A1 receptors underlies anticonvulsant effect of
CGS21680. Eur. J. Pharmacol., 255, 239243.
62. Belardinelli, L., Shryock, J.C., Ruble, J., Monopoli, A., Dionisot-
ti, S., Ongini, E., Dennis, D.M. and Baker, S.P. (1996) Binding of
the novel nonxanthine A2A adenosine receptor antagonist
[3H]SCH58261 to coronary artery membranes. Circ. Res., 79,
11531160.
63. Barrington, W.W., Jacobson, K.A., Hutchison, A.J., Williams,
M. and Stiles, G.L. (1989) Identication of the A2 adenosine
Human Molecular Genetics, 2015, Vol. 24, No. 21 | 6079
receptor binding subunit by photoafnity crosslinking. Proc.
Natl. Acad. Sci. USA, 86, 65726576.
64. Huang, H.P., Yu, C.Y., Chen, H.F., Chen, P.H., Chuang, C.Y., Lin,
S.J., Huang, S.T., Chan, W.H., Ueng, T.H., Ho, H.N. et al. (2010)
Factors from human embryonic stem cell-derived broblast-
like cells promote topology-dependent hepatic differenti-
ation in primate embryonic and induced pluripotent stem
cells. J. Biol. Chem., 285, 3351033519.
65. Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T.,
Tomoda, K. and Yamanaka, S. (2007) Induction of pluripotent
stem cells from adult human broblasts by dened factors.
Cell, 131, 861872.
66. Takahashi, K. and Yamanaka, S. (2006) Induction of pluripo-
tent stem cells from mouse embryonic and adult broblast
cultures by dened factors. Cell, 126, 663676.
67. Lin, I.Y., Chiu, F.L., Yeang, C.H., Chen, H.F., Chuang, C.Y.,
Yang, S.Y., Hou, P.S., Sintupisut, N., Ho, H.N., Kuo, H.C. et al.
(2014) Suppression of the SOX2 neural effector gene by
PRDM1 promotes human germ cell fate in embryonic stem
cells. Stem Cell Reports, 2, 189204.