Short Article
Direct Conversion of Normal and Alzheimer’s
Disease Human Fibroblasts into Neuronal Cells by
Small Molecules
Graphical Abstract
Authors
Wenxiang Hu, Binlong Qiu, Wuqiang
Guan, ..., Yongchun Yu, Jian Zhao,
Gang Pei
Correspondence
[email protected] (J.Z.),
[email protected] (G.P.)
In Brief
In this study, Pei and colleagues
demonstrate that a cocktail of small
molecules alone can reprogram human
fibroblasts from control and Alzheimer’s
disease patients into functional neuronal
cells. These human chemical-induced
neurons resemble iPSC-derived and
transcription factor-induced human
neurons.
Highlights
d
Human fibroblasts can be directly converted into neurons
with a chemical cocktail
d
Electrophysiological properties of hciNs are similar to iPSCderived neurons and iNs
d
hciNs show high neuronal but low fibroblastic gene
expression profiles
d
hciNs derived from FAD patient fibroblasts exhibit abnormal
Ab production
Hu et al., 2015, Cell Stem Cell 17, 204–212
August 6, 2015 ª2015 Elsevier Inc.
http://dx.doi.org/10.1016/j.stem.2015.07.006
Accession Numbers
GSE69480
Cell Stem Cell
Short Article
Direct Conversion of Normal
and Alzheimer’s Disease Human Fibroblasts
into Neuronal Cells by Small Molecules
Wenxiang Hu,1,8 Binlong Qiu,1,8 Wuqiang Guan,2,8 Qinying Wang,1 Min Wang,2 Wei Li,1 Longfei Gao,1 Lu Shen,5
Yin Huang,6 Gangcai Xie,7 Hanzhi Zhao,6 Ying Jin,6 Beisha Tang,5 Yongchun Yu,2,9 Jian Zhao,1,4,9,* and Gang Pei1,3,9,*
1State Key Laboratory of Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, and Graduate
School, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai 200031, China
2Institute of Neurobiology, Institutes of Brain Science, State Key Laboratory of Medical Neurobiology and Collaborative Innovation Center for
Brain Science, Fudan University, Shanghai 200032, China
3Collaborative Innovation Center for Brain Science, School of Life Science and Technology, Tongji University, Shanghai 200092, China
4Translational Medical Center for Stem Cell Therapy, Shanghai East Hospital, School of Medicine, Tongji University, Shanghai 200120, China
5Department of Neurology, Xiangya Hospital, Central South University, Changsha, Hunan Province, 410008, China
6Key Laboratory of Stem Cell Biology, Institute of Health Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of
Sciences, Shanghai 200032, China
7Key Laboratory of Computational Biology, CAS-MPG Partner Institute for Computational Biology, 320 Yueyang Road,
Shanghai 200031, China
8Co-first author
9Co-senior author
*Correspondence: [email protected] (J.Z.), [email protected] (G.P.)
http://dx.doi.org/10.1016/j.stem.2015.07.006
SUMMARY
Neuronal conversion from human fibroblasts can be
induced by lineage-specific transcription factors;
however, the introduction of ectopic genes limits the
therapeutic applications of such induced neurons
(iNs). Here, we report that human fibroblasts can be
directly converted into neuronal cells by a chemical
cocktail of seven small molecules, bypassing a neural
progenitor stage. These human chemical-induced
neuronal cells (hciNs) resembled hiPSC-derived neurons and human iNs (hiNs) with respect to morphology, gene expression profiles, and electrophysiological properties. This approach was further applied
to generate hciNs from familial Alzheimer’s disease
patients. Taken together, our transgene-free and
chemical-only approach for direct reprogramming of
human fibroblasts into neurons provides an alternative strategy for modeling neurological diseases and
for regenerative medicine.
for disease modeling and regenerative medicine. Interestingly,
these conversions can also be induced in vivo by the introduction
of specific TFs (Guo et al., 2014; Niu et al., 2013), further implying
their therapeutic potential. However, the introduction of ectopic
transgenes limits their current therapeutic application. Thus,
tightly controlling the expression of ectopic genes and reducing
the number of TFs have been tried (Dell’Anno et al., 2014; Ladewig et al., 2012). Cell-permeable small molecules have been
shown to promote cell reprogramming (Huangfu et al., 2008; Xu
et al., 2008). A recent report has even shown that small molecules
alone can induce mouse fibroblasts into a pluripotent state (Hou
et al., 2013). Our previous report also showed that neural progenitor cells (NPCs) can be induced from mouse fibroblasts or human
urinary cells with the proper chemical cocktails (Cheng et al.,
2014). Moreover, chemical-promoted transdifferentiation from
mouse or human fibroblasts to different cell lineages has been reported (Pennarossa et al., 2013; Sayed et al., 2015).
Considering that in vitro differentiation of either ESCs or iPSCs
to generate neurons is still time-consuming and complicated
(Zhang and Zhang, 2010), here we try to establish a direct conversion of human fibroblasts into neuronal cells by small molecules, bypassing the neural progenitor stage.
INTRODUCTION
RESULTS
Terminally differentiated somatic cells could be reprogrammed
into pluripotent stem cells by forced expression of a specific
set of transcription factors (TFs), indicating that cell fate determination is reversible (Ladewig et al., 2013; Takahashi and Yamanaka, 2006). Further, different combinations of lineage-specific
TFs could directly convert cardiomyocytes (Ieda et al., 2010),
hepatocytes (Huang et al., 2014), or neurons (Pang et al., 2011;
Vierbuchen et al., 2010) from mouse or human somatic cells, bypassing the pluripotent state and providing alternative avenues
Induction of Neuronal Cells from Human Adult
Fibroblasts by Small Molecules
Our previous study showed that a small chemical cocktail, VCR
(V, valproic acid [VPA]; C, CHIR99021; and R, Repsox), could
induce NPCs from mouse and human somatic cells (Cheng
et al., 2014). We then hypothesized that VCR with the combination of chemicals known to promote neural differentiation of
NPCs might facilitate conversion of human fibroblasts into
neuronal cells. The initial human fibroblasts (HAFs, FS090609,
204 Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc.
Figure 1. Induction of Human Neuronal
Cells by Small Molecules
(A) Scheme of induction procedure. Initial fibroblasts were plated in DMEM and this day is termed
as ‘‘Day – 1.’’ 1 day later, cells were transferred
into induction medium with chemical compounds
and cultured for 8 days, and then cells were
further cultured in maturation medium with supplementary chemicals and neurotrophic factors for
2–3 weeks. V, VPA; C, CHIR99021; R, Repsox; F,
Forskolin; S, SP600625; G, GO6983; Y, Y-27632;
D, Dorsomorphin.
(B) Phase contrast images of control HAFs (left) or
hciNs at day 7 (middle) and day 14 (right). Scale
bars, 100 mm.
(C) hciNs display bipolar neuronal morphologies
and express Dcx (green), Tuj1 (red), and Map2
(green) at day 7. Scale bars, 20 mm.
(D–G) hciNs stained for NeuN (D), Tau (E), SYN (F),
and vGLUT1 (G). Scale bars, 20 mm.
(H) hciN induction efficiencies were presented
as the percentage of induced Tuj1+ Dcx+, Tuj1+
Map2+, or Tuj1+ NeuN+ neuronal cells versus
initial seeding cells at the time of quantification
(means ± SEM, n = 20 random selected 320 fields
from triplicate samples).
(I) The percentage of induced Tuj1+ Dcx+, Tuj1+
Map2+, or Tuj1+ NeuN+ neuronal cells in total
cells at the time of quantification (means ± SEM,
n = 20 random selected 320 fields from triplicate
samples).
See also Figure S1, Table S2, and Table S3.
derived from the foreskin of a 28-year-old male) showed no
contamination of neuronal cells (Figure S1A). We treated the
fibroblasts with VCR in a neuronal induction medium, consistent
with our previous results, and we did not detect any neuron-like
cells (Figure S1B). We then treated HAFs with VCR plus chemicals shown to promote neural differentiation of NPCs (Table
S1). We found that VCRF (VCR+Forskolin) treatment induced a
bipolar neuron-like cell morphology. Some of these bipolar
neuron-like cells even expressed the neuronal maker Tuj1 at
day 7 following VCRF treatment (Figure S1B). However, the
round and prominent cell bodies are unlike a typical neuron
morphology, suggesting a partial and inefficient conversion
(Liu et al., 2013; Pang et al., 2011). Therefore we added other
reported neural-promoting chemicals
to improve the neuronal conversion.
We found that the small molecules
SP600125 (JNK inhibitor, S), GO6983
(PKC inhibitor, G) and Y-27632 (ROCK inhibitor, Y) could facilitate the neuronal
conversion by VCRF and resulted in the
generation of Tuj1-positive cells with
neuron-like morphology (Figure S1B).
This VCRFSGY treatment showed the
highest potency in the induction of
neuronal cells from HAFs. 7 days after
VCRFSGY treatment, we found that a
significant fraction of the living cells
exhibited typical neuronal morphology
and expressed neuronal makers including Dcx, Tuj1, and
Map2 (Figures 1A–1C).
Most of the induced neuronal (iN) cells survived until day 10–12
and stopped developing into more mature neurons. We then
screened for additional chemicals to promote neuron survival
and maturation. Recent reports showed that CHIR99021 (C),
Forskolin (F), and Dorsomorphin (D) are beneficial for neuron survival and maturation (Ladewig et al., 2012; Liu et al., 2013).
Indeed, when we replaced the induction medium with the neuron
maturation medium supplemented with CFD and extra neurotropic factors (BDNF, GDNF, and NT3), neuronal cell survival
and maturation were significantly improved. In 2 or 3 weeks,
the iN cells were stained positive for mature neuronal markers
Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc. 205
Figure 2. hciNs Have Electrophysiological Properties of Functional Neurons
(A) Current-clamp recordings of hciNs generated from HAFs showing a representative train of action potentials (top panel). Step currents were injected
from 10 pA to 80 pA in 6 pA increments (bottom panel).
(B) Representative traces of tetrodotoxin-sensitive fast inward currents recorded in voltage-clamp mode from hciNs at day 14. Tetrodotoxin (1 mM) treatment
inhibited voltage-dependent sodium currents (right panel). Cells were depolarized from 20 mV to 70 mV in 10 mV increments.
(C) Focal application of 10 mM L-glutamic acid (left panel) or 10 mM GABA (right panel) induces inward membrane currents.
(D) Representative trace of spontaneous postsynaptic currents in hciNs generated from GFP-labeled HAFs. GFP-labeled HAFs were treated with indicated
chemicals for 1 week and then replated in high density on pure astrocytes. Two to three weeks after replating, GFP-labeled hciNs with complex neuron
(legend continued on next page)
206 Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc.
Tau, NeuN, and synapsin (SYN) (Figures 1D–1I). We referred
to them as hciN cells (hciNs). We optimized the dosages of
each small molecule (Table S3) and a cocktail with 0.5 mM
VPA, 3 mM CHIR99021, 1 mM Repsox, 10 mM Forskolin, 10 mM
SP600125, 5 mM GO6983, and 5 mM Y-27632 was able to induce
5% Tuj1-positive hciNs (Table S3). Further, the time course of
chemical treatment has been optimized. VCRFSGY treatment for
8 days resulted in the highest induction efficiency (Figure S1C).
More than 80% of Tuj1-positve cells expressed vGLUT1,
whereas GABAergic, cholinergic, or dopaminergic neurons
were merely detected (Figures 1G and S1D). We applied the
VCRFSGY induction protocol to fibroblasts obtained from eight
human individuals with different genders, ages, and passages
(Table S2). The conversion efficiency of each individual fibroblast
line was comparable as shown in Figure S1E and Table S2.
These results indicate that our chemical induction protocol could
reproducibly induce neuronal cells from human fibroblasts.
We next tested whether induction of hciNs from HAFs using
our chemical induction protocol bypassed the neural progenitor
stage. RNA expression profiling by qRT-PCR and immunostaining analysis showed that the neural progenitor genes Sox2,
Pax6, FoxG1, or Nestin were never expressed during the hciNs
induction procedure (Figures S1F and S1G). Interestingly, we
found that the proliferation marker Ki67 was not detectable after
day 3 (Figure S1H), suggesting that our chemical induction protocol might trigger a fast cell cycle exit of human fibroblasts and
that the conversion of human fibroblasts into hciNs using this
chemical strategy is probably direct.
Physiological Properties of hciNs
To test whether the hciNs from HAFs possess basic electrophysiological properties of neurons, such as the ability to fire action
potentials (APs) and the induction of membrane current, we performed whole-cell patch-clamp recording of hciNs with complex
neuron morphology at day 14 after chemical cocktail treatment.
88% of recorded hciNs (n = 89) generated repetitive trains of
APs when we depolarizing the membrane in current-clamp
mode (Figure 2A and Figure S2C). Depolarizing voltage steps
in voltage-clamp mode elicited fast inward sodium currents (Figure 2B; 93%, n = 15). Moreover, the fast inward currents were
completely blocked by the sodium channel blocker tetrodotoxin
(TTX). To test whether the hciNs have functional glutamate and
GABA receptors, we puffed exogenous L-glutamic acid (Figure 2C, left panel; 50%, n = 10) and GABA (Figure 2C, right panel;
67%, n = 9) onto the hciNs and found that inward currents were
induced, indicating that functional glutamate and GABA receptors were present in hciNs. Co-culturing neuronal cells with
astrocytes was reported to facilitate neuron maturation and synapse formation by providing neurotropic factors (Chanda et al.,
2014). Thus we infected HAFs with lentivirus encoding GFP
and seeded GFP-positive hciNs (6 days after VCRFSGY treatment) to a monolayer astrocyte culture. Two weeks after plating,
we carried out whole-cell patch-clamp recording on GFP-positive hciNs with mature neuron morphology (Figure 2D, top panel).
Spontaneous postsynaptic currents (sPSCs) was recorded
(Figure 2D; 31%, n = 26), indicating that hciNs could form functional synapses. Moreover, analysis of the kinetics of sPSCs
demonstrated that hciNs possessed fast-decay spontaneous
excitatory postsynaptic currents (sEPSCs) and slow-decay
spontaneous inhibitory postsynaptic currents (sIPSCs) (Figures
2D and S2A). Application of bicuculline selectively blocked the
slow-decay sIPSCs (Figure S2B), whereas NBQX eliminated
the fast-decay sEPSCs (data not shown), indicating the presence of GABAergic and glutamatergic synapses in the hciNs.
Together, these results show that hciNs induced by small molecules are functional neurons.
To further test the reliability of our chemical induction protocol,
we recorded more hciN cell lines and compared the electrophysiological properties of hciNs with those of human induced
pluripotent stem cell (hiPSC)-derived neurons and TF-iNs (Figures 2E–2H). The physiological properties of AP amplitude, AP
threshold, and resting membrane potential (RMP) were not
significantly different. Although there were some variations between cell lines, the overall electrophysiological properties of
hciNs resembled those of hiPSC-derived neurons and TF-iNs.
Patch-clamp analysis only qualified a small fraction of cells.
To monitor the activity of large numbers of cells at once, we
loaded hciNs (co-cultured with monolayer astrocytes) with
Cal520, a sensitive Ca2+ indicator (Figure 2I). The hciNs exhibited robust spontaneous calcium transients (Figures 2J
and 2K), suggesting that the hciN neural network was rather
active. The percentage of hciNs firing spontaneous calcium
transients was not significantly different from that of iPSCderived neurons (Figure 2L).
hciNs Possess a Neuronal Gene Expression Signature
To further characterize the hciN cell phenotype, we quantitatively analyzed the expression of 25 neuronal genes and 7
morphology were recorded. Recording configurations were shown (top panel). The insert shows two distinct types of sPSCs: fast-decay sEPSCs (indicated with
red vertical bars) and slow-decay sIPSCs (indicated with green vertical bars). The enlarged sEPSC (red line) and sIPSC (green line) were normalized. Scale bars,
20 mm.
(E) Representative APs from six different cell lines (FS090609-C1 and FS090609-C2 for hiPSCs; FS090609, SF002, and IMR90 for hciNs; FS090609 for TF-iNs).
(F–H) Statistics of intrinsic electrophysiological properties of hiPSCs (two cell lines: FS090609-C1 and FS090609-C2), hciNs (three cell lines: FS090609, SF002,
and IMR90) and TF-iNs (FS090609). F, AP amplitude; G, AP threshold; H, resting membrane potential (RMP).
(I) Fluorescence and DIC images of hciNs stained with calcium indicator Cal520. Neurons, red crosses; the number of cells is indicated (1–27). Astrocytes, green
crosses; the number of cells is 28–30.
(J) Representative traces of spontaneous calcium transients in Figure 2I. Black vertical bars indicate the initiate time of calcium transients. The rise phase of
astrocytes calcium transients (green lines) is much slower than that of neurons (red lines).
(K) Raster plots of spontaneous calcium transients in Figure 2I.
(L) Quantification of the percentage of neurons exhibiting spontaneous calcium transients. In total, 22 movies were acquired and 1,249 putative neurons were
analyzed (FS090609-C1 hiPSCs, 138 cells from 3 movies; FS090609-C2 hiPSCs, 650 cells from 3 movies; FS090609 hciNs, 108 cells from 5 movies; SF002 hciNs,
170 cells from 5 movies; IMR90 hciNs, 183 cells from 6 movies). See also Figure S2.
Data of at least three independent experiments are shown as means ± SEM.
Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc. 207
Figure 3. hciN Cells’ Expression Profiling during Chemical Reprogramming
(A) Gene expression of single cells as ascertained by the Fluidigm Biomark platform. hciNs were collected at day 21 after chemical treatment and purified using
FACS. Expression levels (expressed as Ct values) are color-coded as shown at the bottom (color scales). H1-20 indicates hciN samples, F1-20 indicates
fibroblast samples, and T1-20 represents TF-iNs samples.
(B) Heatmap and hierarchical clustering of genes with significance from microarray data. Samples of HAFs, hciNs (day 3, day 7, day 14-1, and day 14-2) and
control neurons (derived from an hES cell line) were compared. Shown are probes selected on fold change R10 in hciNs (day 14) versus HAFs. In the heatmap,
green indicates decreased expression whereas red means increased expression as compared to that in HAFs.
(legend continued on next page)
208 Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc.
non-neuronal genes at a single-cell level using the Fluidigm
Biomark platform. TF-iNs were prepared as reported previously
(Ladewig et al., 2012) and were used as a positive control (Figure S3A). hciN and TF-iN cell cultures at day 21 were purified
using fluorescence-activated cell sorting (FACS, Figures S3B
and S3C). The heatmap analysis showed that hciNs consistently
expressed major neuronal genes, channel genes, and synaptic
genes, but not glial-specific genes or fibroblast-specific genes
(Figure 3A). The expression pattern of hciNs was similar to that
of TF-iNs but distinguished from that of initial fibroblasts.
We further compared the global gene expression patterns of
human embryonic stem cell (hESC)-derived neurons (control
neurons), HAFs, pre-hciNs (VCRFSGY treated HAFs at days 3
and 7), and hciNs (induced with VCRFSGY for 14 days) by microarray analysis. The global genome heatmap with hierarchical
cluster analysis showed a significant difference between hciNs
and their parental HAFs and a high similarity between hciNs
and control neurons (Figure 3B). It is of note that many representative neuronal-specific genes including Ncam1, Dcx, Map2,
Ascl1, and Nefl were strongly upregulated (Figure 3C, left panel,
and Figure S3D), whereas glial genes were almost unaltered in
hciNs (Figure 3C, middle panel). Interestingly, the upregulation
of some neuronal genes was also observed in VCRFSGY-treated
HAFs as early as on days 3 and 7 (Figure S3E). Moreover, fibroblast-specific genes including Col12A1, Tgfb1i1, Thy1, and Ctgf
were dramatically downregulated during neuronal conversion
(Figure 3C, right panel, and Figure S3F) (Kim et al., 2010). Genes
that show a more than 2-fold alteration were subjected to further
gene ontology (GO) function enrichment analysis. We found that
the shared expression core sets of hciNs from different batches
and control neurons were mainly related to neuron differentiation, synapse, and synaptic transmission (Figure 3D). On the
other hand, the expression of genes related to the extracellular
region, the extracellular matrix, and the regulation of cell motion
were significantly downregulated GO items in hciNs versus those
in HAFs (Figure S3G).
We further detected the effect of each molecule on downstream-molecule events in our induction protocol, as shown in Figure S4A. We found that some small molecules alone could elevate
neuronal gene expression and downregulate fibroblast-specific
expression of some genes such as Col1a1, Dkk3, Thy1, and
Ctgf. Further, we removed each small molecule from our chemical
cocktails, and we found that removing CHIR99021 or SP600125
from the chemical cocktails led to no induction of neuronal gene
expression, but the downregulation of fibroblast gene expression
can still be observed, indicating that CHIR99021 and SP600125
are critical for the upregulation of neuronal gene expression (Figure S4B). These results further support our hypothesis that each
molecule of the cocktail should operate coordinately for a successful neuronal conversion (Figure S3H).
Generation of hciNs from FAD Patient Fibroblasts
To test whether our chemical induction protocol can be applied
to generate hciNs for modeling of neurological diseases, we
induced neuronal conversion of human skin fibroblasts derived
from patients with familial Alzheimer’s disease (FAD) carrying
mutations in APP (V717I) or presenilin 1 (I167del or A434T or
S169del) (Guo et al., 2010; Jiao et al., 2014). The hciNs derived
from FAD fibroblasts displayed similar neuronal characteristics
to those of hciNs derived from normal fibroblasts (Figure 4A).
The induction efficiency of hciNs from these fibroblasts was
comparable to that of normal fibroblasts (Figure S1E).
Muratore et al. (2014) show that iPSC-derived neurons
harboring the APPV717I mutation have elevated Ab42 and
Ab38 levels and Ab42/40 ratio, but the Ab40 levels are not
significantly increased as compared with those of normal control cells. Higher total Tau levels and p-Tau levels are detected,
but only p-S262 (12E8) versus total Tau levels at a very late
stage of differentiation (day 100) are upregulated. Terwel
et al. (2008) demonstrated that GSK-3 activities in APPV717I
mouse brains were higher than those of age-matched wildtype control brains, but GSK-3 activities in human neurons
with the mutation were still difficult to predict without further
exploration. We measured the levels of extracellular Ab40
and the Ab42 level of the hciNs derived from these FAD fibroblasts and compared those with those of TF-iNs. Similar to
previous reports, the extracellular Ab42 level and the Ab42/
Ab40 ratio were increased in hciNs derived from FAD109
(APPV717I) fibroblasts compared to those of hciNs derived
from healthy persons (Figures 4B–4D). In FAD hciNs from
another patient, in which newly reported mutations in presenilin 1 (I167del or S169del) were carried (Jiao et al., 2014), the
extracellular Ab42 levels were also elevated (Figure 4C). However, we did not detect obvious alternation of Ab levels in
FAD132 (PS1 A434T) fibroblasts. The Ab level in fibroblasts
was much lower than that in iN cells. We found that the levels
of phosphorylated Tau and total Tau in original fibroblasts were
very low (Figure S1I). However, the levels of phosphorylated
Tau and total Tau of FAD109 hciNs were higher than that of
controls, whereas the other three FAD hciNs (FAD131,
FAD132, and FADA16) carrying the presenilin 1 mutation did
not show obvious change in p-Tau (Thr 231) and total Tau
levels (Figure 4E). Surprisingly, the amount of active GSK-3b
in all four FAD hciNs and TF-iNs was similar to that of normal
hciNs and TF-iNs (Figure 4E). These results indicate that the
chemical induction protocol might be a feasible way to
generate patient-specific neuronal cells that might be useful
for neurological disease modeling, related mechanism studying, and drug screening.
DISCUSSION
This is a proof-of-principle study demonstrating that the direct
lineage-specific conversion of neuronal cells could be achieved
by small molecules without the introduction of ectopic genes.
Our previous study showed that a chemical cocktail, VCR, could
induce NPCs from mouse embryonic fibroblasts and human
urinary cells under a hypoxic condition. But we found the
(C) Pair-wise gene expression comparisons show that most neuronal markers (left panel) are increased, whereas glial markers (middle panel) are not activated
and some fibroblast-specific genes (right panel) are silenced. Black lines indicate 2-fold changes between the sample pairs.
(D) Gene ontology (GO) analysis of the overlapping genes whose expression changes are R2-fold identified in hciNs (day 14-1 and day 14-2) and control neurons
compared to that of HAFs. See also Figures S3 and S4.
Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc. 209
Figure 4. Generation of hciNs from FAD Fibroblasts
(A) Characterization of hciNs generated from FAD fibroblasts (FAD109 and FAD131).
(B–D) Ab levels in FAD hciN and TF-iN cell cultures compared to those of WT hciN cultures.
(E) The level of phosphorylated Tau and total Tau of FAD109 hciNs and TF-iN cells.
Data from three independent experiments are shown as means ± SEM.
Scale bars, 20 mm. *p < 0.05; **p < 0.01; ***p < 0.001; versus control hciNs.
VCRFSGY-treated fibroblasts quickly exited the cell cycle as
early as day 3 (Figure S1H), suggesting that the chemical induction of neuronal cells from human fibroblasts might bypass a proliferative intermediate.
The conversion process induced by our chemical strategy is
accompanied by the downregulation of fibroblast-specific
genes and the increased expression of endogenous neuronal
transcriptional factors. It is consistent with the previous studies
that the forced expression of certain neuronal TFs could convert
mouse or human somatic cells into neuronal cells (Caiazzo
et al., 2011; Pang et al., 2011). Consistently, our microarray
analysis showed that the gene expression patterns of related
signaling pathways in hciNs were more similar to those of control neurons than those of HAFs. Thus, by coordinating multiple
signaling pathways, these seven small molecules modulate
neuronal TF gene expression and thereby promote the neuronal
cell transition.
A proper fine-tuning of multiple signaling pathways is critical
for a successful neuronal conversion from human fibroblasts
by the chemical induction protocol. VPA induces cells into a
more amenable state for cell-fate transition through epigenetic
modification (Huangfu et al., 2008). Inhibition of TGF-b and
GSK-3 improves the neuronal conversion of stably transduced
human fibroblasts by Ascl1 and Ngn2 (Ladewig et al., 2012). Forskolin enables Ngn2 to convert human fibroblasts into cholin210 Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc.
ergic neurons (Liu et al., 2013). SP600125 could facilitate the
neural reprogramming of AHDF transduced with OCT4 alone
(Zhu et al., 2014). PKC inhibition improves naive human pluripotent stem cells and induces neuritogenesis of Neuro-2a cells
(Gafni et al., 2013; Miñana et al., 1990). Y-27632 assists in the
maintenance of pluripotent stem cells and neuron survival
(Lamas et al., 2014; Xi et al., 2013). Therefore, the chemical cocktail VCRFSGY erases fibroblast-specific gene expression of the
initial cells, specifically upregulating neuronal gene expression
and facilitating the neuronal conversion of HAFs (Figure S4).
However, the precise regulatory mechanisms of this chemical
cocktail remain to be investigated.
Most of the hciNs are glutamatergic neurons in our induction
system. Previous reports showed that different combinations
of small molecules and TFs may induce certain subtypes of neurons (Caiazzo et al., 2011; Liu et al., 2013). Thus, it should be
possible to generate different neuronal subtypes with similar
chemical approaches but with slightly modified chemical cocktails. Moreover, it also needs to be explored in the future whether
functional neurons could be induced by chemical cocktails
in vivo against neurological diseases or injury. Together, our
study presents another strategy for the generation of patientspecific neuronal cells, thus providing desirable neuron resourses for personalized disease modeling or even potential
cell replacement therapy of neurological disorders.
EXPERIMENTAL PROCEDURES
Generation of hciNs
Initial fibroblasts were seeded onto gelatin-coated culture plates (10,00015,000 cells/well in 24-well plates) and cultured in fibroblast medium for
1 day. Then, the cells were transferred into neuronal induction medium
(DMEM/F12: Neurobasal [1:1] with 0.5% N-2, 1% B-27, 100 mM cAMP, and
20 ng/ml bFGF) with indicated chemicals. Medium containing indicated
chemicals was changed every 4 days. After 8 days, cells were switched to
neuronal maturation medium (DMEM/F12: Neurobasal [1:1] with 0.5% N-2,
1% B-27, 100 mM cAMP, 20 ng/ml bFGF, 20 ng/ml BDNF, 20 ng/ml GDNF,
and 20 ng/ml NT3) with indicated chemicals. After 2 weeks, iN cells were
cultured in neuronal maturation medium without bFGF and chemicals.
Neuronal maturation medium was half-changed every other day. The concentrations of chemicals used were as follows: VPA, 0.5 mM; CHIR99021,
3 mM; Repsox, 1 mM; Forskolin, 10 mM; SP600125, 10 mM; GO6983, 5 mM;
Y-27632, 5 mM; Dorsomorphin, 1 mM. Stepwise induction protocols as well
as detailed information about the medium used are provided in Supplemental
Experimental Procedures. Information about all chemicals applied in this study
are in Table S1. The detailed information of human fibroblasts used is listed in
Table S2. Human adult fibroblasts were derived from human foreskin or skin
biopsies of healthy and patient individuals with approval for collection and
use of human samples by institutional ethical committees.
Statistical Analysis
All quantified data were statistically analyzed and presented as means ± SEM.
One-way ANOVA and Bonferroni’s multiple comparison test were used to
calculate statistical significance with p values, unless otherwise stated.
ACCESSION NUMBERS
Shanghai Zhangjiang Stem Cell Research Project (ZJ2014-ZD-002), the National Natural Science Foundation of China (31371419), and the National Science and Technology Support Program (2012BAI10B03).
Received: February 16, 2015
Revised: June 2, 2015
Accepted: July 13, 2015
Published: August 6, 2015
REFERENCES
Caiazzo, M., Dell’Anno, M.T., Dvoretskova, E., Lazarevic, D., Taverna, S., Leo,
D., Sotnikova, T.D., Menegon, A., Roncaglia, P., Colciago, G., et al. (2011).
Direct generation of functional dopaminergic neurons from mouse and human
fibroblasts. Nature 476, 224–227.
Chanda, S., Ang, C.E., Davila, J., Pak, C., Mall, M., Lee, Q.Y., Ahlenius, H.,
Jung, S.W., Südhof, T.C., and Wernig, M. (2014). Generation of induced
neuronal cells by the single reprogramming factor ASCL1. Stem Cell
Reports 3, 282–296.
Cheng, L., Hu, W., Qiu, B., Zhao, J., Yu, Y., Guan, W., Wang, M., Yang, W., and
Pei, G. (2014). Generation of neural progenitor cells by chemical cocktails and
hypoxia. Cell Res. 24, 665–679.
Dell’Anno, M.T., Caiazzo, M., Leo, D., Dvoretskova, E., Medrihan, L.,
Colasante, G., Giannelli, S., Theka, I., Russo, G., Mus, L., et al. (2014).
Remote control of induced dopaminergic neurons in parkinsonian rats.
J. Clin. Invest. 124, 3215–3229.
Gafni, O., Weinberger, L., Mansour, A.A., Manor, Y.S., Chomsky, E., BenYosef, D., Kalma, Y., Viukov, S., Maza, I., Zviran, A., et al. (2013). Derivation
of novel human ground state naive pluripotent stem cells. Nature 504,
282–286.
The accession number for the microarray data reported in this paper is GEO:
GSE69480.
Guo, J., Wei, J., Liao, S., Wang, L., Jiang, H., and Tang, B. (2010). A novel presenilin 1 mutation (Ser169del) in a Chinese family with early-onset Alzheimer’s
disease. Neurosci. Lett. 468, 34–37.
SUPPLEMENTAL INFORMATION
Guo, Z., Zhang, L., Wu, Z., Chen, Y., Wang, F., and Chen, G. (2014). In vivo
direct reprogramming of reactive glial cells into functional neurons after brain
injury and in an Alzheimer’s disease model. Cell Stem Cell 14, 188–202.
Supplemental Information for this article includes four figures, four tables, and
Supplemental Experimental Procedures and can be found with this article online at http://dx.doi.org/10.1016/j.stem.2015.07.006.
AUTHOR CONTRIBUTIONS
G.P. supervised and controlled research conception and design and interpreted data. J.Z. controlled research conception, designed the experiments,
interpreted data, and drafted and revised the manuscript critically for intellectual content. Y.Y. designed the electrophysiological analysis experiments, interpreted the correlated data, and revised the manuscript. W.H., B.Q., and
W.G. conducted the overall experiments and analyzed the data. W.H. and
B.Q. collected and analyzed data, organized figures, and drafted the manuscript. Q.W., L.G., and W.L. contributed to the hciN induction and iPSC differentiation. Y.H. and G.X. carried out single-cell PCR and heatmap analysis.
M.W. contributed to transplantation experiments. H.Z., Y.J., B.T., and L.S.
provided human fibroblast lines and commented on the manuscript. All authors contributed to the figures and the manuscript preparation and approved
the version that was submitted.
ACKNOWLEDGMENTS
We thank Dr. Duanqing Pei (Guangzhou Institutes of Biomedicine and Health)
for providing the human fibroblasts (SF001 and SF002). We thank Dr. Benedikt
Berninger (University Mainz, Germany) for providing the CAG-Ngn2-IRESDsRed plasmid. We thank Dr. Xiaoqing Zhang (Tongji University, China)
for helping with neural differentiation assay. We thank all members of the
labs for sharing reagents and advice. We also thank Yang Yu, Yan Wang,
and National Center for Protein Science Shanghai for technical support.
This work was supported by grants from Chinese Academy of Sciences
(XDA01010302), the Ministry of Science and Technology (2015CB964502),
Hou, P., Li, Y., Zhang, X., Liu, C., Guan, J., Li, H., Zhao, T., Ye, J., Yang, W., Liu,
K., et al. (2013). Pluripotent stem cells induced from mouse somatic cells by
small-molecule compounds. Science 341, 651–654.
Huang, P., Zhang, L., Gao, Y., He, Z., Yao, D., Wu, Z., Cen, J., Chen, X., Liu, C.,
Hu, Y., et al. (2014). Direct reprogramming of human fibroblasts to functional
and expandable hepatocytes. Cell Stem Cell 14, 370–384.
Huangfu, D., Maehr, R., Guo, W., Eijkelenboom, A., Snitow, M., Chen, A.E.,
and Melton, D.A. (2008). Induction of pluripotent stem cells by defined factors
is greatly improved by small-molecule compounds. Nat. Biotechnol. 26,
795–797.
Ieda, M., Fu, J.D., Delgado-Olguin, P., Vedantham, V., Hayashi, Y., Bruneau,
B.G., and Srivastava, D. (2010). Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell 142, 375–386.
Jiao, B., Tang, B., Liu, X., Xu, J., Wang, Y., Zhou, L., Zhang, F., Yan, X., Zhou,
Y., and Shen, L. (2014). Mutational analysis in early-onset familial Alzheimer’s
disease in Mainland China. Neurobiol. Aging 35, 1957.e1-e6.
Kim, K., Doi, A., Wen, B., Ng, K., Zhao, R., Cahan, P., Kim, J., Aryee, M.J., Ji,
H., Ehrlich, L.I., et al. (2010). Epigenetic memory in induced pluripotent stem
cells. Nature 467, 285–290.
Ladewig, J., Mertens, J., Kesavan, J., Doerr, J., Poppe, D., Glaue, F., Herms,
S., Wernet, P., Kögler, G., Müller, F.J., et al. (2012). Small molecules enable
highly efficient neuronal conversion of human fibroblasts. Nat. Methods 9,
575–578.
Ladewig, J., Koch, P., and Brustle, O. (2013). Leveling Waddington: the emergence of direct programming and the loss of cell fate hierarchies. Nat. Rev.
Mol. Cell Biol. 14, 225–236.
Lamas, N.J., Johnson-Kerner, B., Roybon, L., Kim, Y.A., Garcia-Diaz, A.,
Wichterle, H., and Henderson, C.E. (2014). Neurotrophic requirements of
Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc. 211
human motor neurons defined using amplified and purified stem cell-derived
cultures. PLoS ONE 9, e110324.
Liu, M.L., Zang, T., Zou, Y., Chang, J.C., Gibson, J.R., Huber, K.M., and Zhang,
C.L. (2013). Small molecules enable neurogenin 2 to efficiently convert human
fibroblasts into cholinergic neurons. Nat. Commun. 4, 2183.
Miñana, M.D., Felipo, V., and Grisolı́a, S. (1990). Inhibition of protein kinase C
induces differentiation in Neuro-2a cells. Proc. Natl. Acad. Sci. USA 87, 4335–
4339.
Muratore, C.R., Rice, H.C., Srikanth, P., Callahan, D.G., Shin, T., Benjamin,
L.N., Walsh, D.M., Selkoe, D.J., and Young-Pearse, T.L. (2014). The familial
Alzheimer’s disease APPV717I mutation alters APP processing and Tau
expression in iPSC-derived neurons. Hum. Mol. Genet. 23, 3523–3536.
Niu, W., Zang, T., Zou, Y., Fang, S., Smith, D.K., Bachoo, R., and Zhang, C.L.
(2013). In vivo reprogramming of astrocytes to neuroblasts in the adult brain.
Nat. Cell Biol. 15, 1164–1175.
Sayed, N., Wong, W.T., Ospino, F., Meng, S., Lee, J., Jha, A., Dexheimer, P.,
Aronow, B.J., and Cooke, J.P. (2015). Transdifferentiation of human fibroblasts
to endothelial cells: role of innate immunity. Circulation 131, 300–309.
Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells
from mouse embryonic and adult fibroblast cultures by defined factors. Cell
126, 663–676.
Terwel, D., Muyllaert, D., Dewachter, I., Borghgraef, P., Croes, S., Devijver, H.,
and Van Leuven, F. (2008). Amyloid activates GSK-3beta to aggravate
neuronal tauopathy in bigenic mice. Am. J. Pathol. 172, 786–798.
Vierbuchen, T., Ostermeier, A., Pang, Z.P., Kokubu, Y., Südhof, T.C., and
Wernig, M. (2010). Direct conversion of fibroblasts to functional neurons by
defined factors. Nature 463, 1035–1041.
Xi, G., Hu, P., Qu, C., Qiu, S., Tong, C., and Ying, Q.L. (2013). Induced neural
stem cells generated from rat fibroblasts. Genomics Proteomics
Bioinformatics 11, 312–319.
Xu, Y., Shi, Y., and Ding, S. (2008). A chemical approach to stem-cell biology
and regenerative medicine. Nature 453, 338–344.
Pang, Z.P., Yang, N., Vierbuchen, T., Ostermeier, A., Fuentes, D.R., Yang,
T.Q., Citri, A., Sebastiano, V., Marro, S., Südhof, T.C., and Wernig, M.
(2011). Induction of human neuronal cells by defined transcription factors.
Nature 476, 220–223.
Zhang, X.Q., and Zhang, S.C. (2010). Differentiation of neural precursors and
dopaminergic neurons from human embryonic stem cells. Methods Mol.
Biol. 584, 355–366.
Pennarossa, G., Maffei, S., Campagnol, M., Tarantini, L., Gandolfi, F., and
Brevini, T.A. (2013). Brief demethylation step allows the conversion of adult human skin fibroblasts into insulin-secreting cells. Proc. Natl. Acad. Sci. USA
110, 8948–8953.
Zhu, S., Ambasudhan, R., Sun, W., Kim, H.J., Talantova, M., Wang, X., Zhang,
M., Zhang, Y., Laurent, T., Parker, J., et al. (2014). Small molecules enable
OCT4-mediated direct reprogramming into expandable human neural stem
cells. Cell Res. 24, 126–129.
212 Cell Stem Cell 17, 204–212, August 6, 2015 ª2015 Elsevier Inc.
Cell Stem Cell
Supplemental Information
Direct Conversion of Normal
and Alzheimer’s Disease Human Fibroblasts
into Neuronal Cells by Small Molecules
Wenxiang Hu, Binlong Qiu, Wuqiang Guan, Qinying Wang, Min Wang, Wei Li, Longfei
Gao, Lu Shen, Yin Huang, Gangcai Xie, Hanzhi Zhao, Ying Jin, Beisha Tang, Yongchun
Yu, Jian Zhao, and Gang Pei
Supplemental Figure Legends
Figure S1, related to Figure 1. Screening chemicals for neuronal cells induction
and characterization of hciN cells during induction process.
(A) Initial human fibroblasts are negative for neuronal markers. The hES derived
neuron express neuronal markers Dcx, Tuj1 and Map2, but these neuronal markers are
absent in initial human fibroblasts. Scale bars: 100 μm.
(B) Quantification of induced neuronal cells in different chemicals-treated HAFs at
day 7. The percentage of Tuj1-positve neuronal cells represent the induction
efficiencies (means ±SEM, n=10 random selected ×20 fields from triplicate samples).
V, VPA; C, CHIR99021; R, Repsox; F, Forskolin; S1, SB202190; S2, SP600625; S3,
SB431542; P, PD013252; G, GO 6983; Y, Y-27632.
(C) Optimization of chemicals treatment periods for hciN cells induction. Chemicals
treatment period was from 4 days to 3 weeks. The induction efficiencies were
presented as the percentage of induced Map2+ neuronal cells in initial seeding cells at
the time of quantification (means ± SEM, n=10 random selected ×20 fields from
triplicate samples).
(D) Quantification of the percentage of different subtype neurons in Tuj1- or
Map2-positive hciN cells (right panel, means ± SEM, n=10 random selected ×20
fields from triplicate samples). Representative micrographs of hciN cells stained with
GABAnergic neuron marker GAD67, Cholinergic neuron marker Chat and
Dopaminergic neuron marker TH were shown (left panel). Scale bars: 20 μm.
(E) Induction of hciN cells from multiple human fibroblasts by small molecules.
The induction efficiencies were presented as the percentage of induced Map2+
neuronal cells in initial seeding cells at the time (Day 14) of quantification (means ±
SEM, n=10 random selected ×20 fields from triplicate samples).
(F and G) Induction of hciN cells probably bypasses neural progenitor cell stage. No
expression of neural progenitor markers Sox2, Pax6 or FoxG1 during neuronal
conversion (F). All sample data are normalized to that of d0, which is considered as 1.
Data of three independent
experiments
are shown
as
means
± SEM.
Immunofluorescent staining also showed that no detectable expression of neural
progenitor cells specific markers during neuronal conversion (G), samples from
hES-derived neural progenitor cells were used as positive control. Scale bars: 100 μm.
(H) HAFs exit cell cycle after treating with VCRFSGY for 3 days. HAFs were treated
with VCRFSGY for 3 days and stained with the proliferation marker Ki67. Scale bars:
50 μm.
(I) The level of phosphorylated Tau and total Tau were not detected in 7 individual
fibroblasts (top panel). And the level of phosphorylated GSK-3β and total GSK3-3β
of FAD109 fibroblasts were slightly higher than other individual fibroblasts (bottom
panel). Lysates from 7 individual fibroblast lines were collected and measured on
western blots. OE-tau, a sample of 293T with tau overexpressed.
Figure S2, related to Figure 2. The kinetic and pharmalogical properties of
sEPSCs and sIPSCs.
(A) sEPSCs (red circles) and sIPSCs (green circles) were sorted by decay time.
(B) Bicuculline (BIC, 50 µM) selectively block the slow-decay sIPSCs.
(C) Quantification of the percentage of putative neurons exhibiting induced action
potentials.
Figure S3, related to Figure 3. Cell preparation for Single-cell gene expression
analysis and Whole-genome profile of HAFs, hciN cells at day 3, 7 and 14 and
control neurons analyzed by cDNA microarray.
(A) Characterization of TF-iNs. HAFs were stained positive for pan-neuronal markers
Tuj1 and Map2 at day 21 after infection with Ascl1 and NeuroG2. Scale bars: 20 μm.
(B) Morphology and immunostaining of purified hciNs. Scale bars: left panel 100 μm,
right panel 20 μm.
(C) The FACS process data analysis. Cells were stained by anti-CD24-PE-Cy7, and
then CD24+ cells were sorted out.
(D) Heatmap for subsets of selected neuronal, glial and fibroblast-specific genes.
(E and F) Pairwise gene expression comparisons show that expression levels of neural
markers are increased (E), whereas expression levels of fibroblast specific genes are
silenced (F) in hciN cells at day 3 and 7. Black lines indicate 2-fold changes between
the sample pairs.
(G) Gene ontology (GO) analysis of genes with ≥ 2-fold alteration in hciN cells
(day 14-1 and day 14-2) and that in control neurons compared to that in HAFs.
(H) Schematic diagram illustrating the possible mechanism of neuronal conversion
from HAFs.
Figure S4, related to Figure 3. Gene expression profiles of HAFs induced by each
small molecule or by removing each small molecule from chemical cocktails.
(A) Upregulation of neuronal transcriptional factors and downregulation of fibroblast
specific genes at day 7 during chemical induction. HAFs were treated with indicated
chemical for 7 days. All sample data are normalized to that of d0, which is considered
as 1. Primary human neurons (Science Cell, #1525) were used as control. Data of
three independent experiments were shown as means ±SEM.
(B) Upregulation of neuronal transcriptional factors and downregulation of fibroblast
specific genes at day 7 during chemical induction. HAFs were treated with indicated
chemical for 7 days. All sample data are normalized to that of d0, which is considered
as 1. Primary human neurons (Science Cell, #1525) were used as control. Data of two
independent experiments are shown as means ±SEM.
Table S1, Related to Experimental Procedures. Chemical compounds reported to induce
neural cells differentiation and reprogramming.
Name
Forskolin
SB431542
LDN193189
CHIR99021
Mechanism
cAMP activator
ALK5 inhibitor
BMP inhibitor
GSK3 inhibitor
Structure
Effects on neural cells
Reference
induction
Improve neuronal
conversion of human
fibroblasts
Induce pluripotent stem
cells toward neural
lineages, improve
neuronal conversion of
human fibroblasts
Improve neuronal
conversion of human
fibroblasts
Improve neuronal
conversion of human
fibroblasts, induce
dopamine neurons form
human ES cells
Trigger robust neuronal
differentiation in adult
neural stem cells
Induce neural stem cells
differentiation
Isoxazole
COX-2 inhibitor
Retinoid acid
Retinoic acid receptor
agonist
db-cAMP
2'-O-Dibutyryl-cAMP
Promote neuronal surviral
and maturation
DAPT
notch inhibitor
Promote neuronal surviral
and maturation
BMP inhibitor
Promote neuronal surviral
and maturation
SB202190
p38 MAPK inhibitor
Inhibit hippocampal
apoptosis
DMH1
BMP inhibitor
Promotes neurogenesis of
hiPSCs
Dorsomorphin
(Liu et al.,
2013)
(Ladewig
et al.,
2012)
(Ladewig
et al.,
2012)
(Li et al.,
2011)
(Schneider
et al.,
2008)
(Li et al.,
2008)
(Ladewig
et al.,
2012)
(Li et al.,
2008)
(Liu et al.,
2013)
(Yang et
al., 2013a)
(Neely et
al., 2012)
Inhibitor of
PI3K/AKT
Induction of neural-like
cells from bone
marrow-mesenchymal
stem cells
(Wang et
al., 2012)
Repsox
TGFβ inhibitor
Improve survival of
neurons over long-term
culture
(Wainger
et al.,
2015)
PD0325901
Erk inhibitor
Rolipram
PDE4 inhibitor
Accelerate neural fate
conversion of EpiSCs
Facilitate the neural
reprogramming of AHDF
transduced with OCT4
alone
Facilitate the neural
reprogramming of AHDF
transduced with OCT4
alone
Facilitate the neural
reprogramming of AHDF
transduced with OCT4
alone
(Jaeger et
al., 2011)
(Zhu et
al., 2014)
LY294002
SP600125
JNK inhibitor
NaB
HDAC inhibitor
Deferoxamine
stabilize HIF by
chelating Fe2+
Regulate human neural
progenitor fate
Y-27632
ROCK inhibitor
Induce neural stem cells
from rat fibroblasts,
support human motor
neuron survival
kenpaullone
dual inhibition of
GSK3
and
HGK
kinases
Improve the survival of
human MNs
KHS101
specific binding to the
TACC3 protein
Induce a neuronal
differentiation phenotype
H7
PKC inhibitor
Induce neuritogenesis of
Neuro-2a cells
Go6983
PKC inhibitor
Improve naive human
pluripotent stem cells
(Zhu et
al., 2014)
(Zhu et
al., 2014)
(Xie et al.,
2014)
(Lamas et
al., 2014;
Xi et al.,
2013)
(Yang et
al., 2013b)
(Wurdak
et
al.,
2010)
(Minana
et
al.,
1990)
(Gafni et
al., 2013)
Table S2, Related to Figure 1 and Experimental Procedures. Detailed
information of human fibroblasts. Conversion efficiency was calculated from 10
random selected ×20 fields from triplicate samples and shown as means ± SEM.
Line
Passage
FS090609 P10-15
WT
SF001
P10-15
SF002
P10-15
IMR90
P30-35
FAD109
P5-10
FAD131
P5-10
FAD132
P5-10
FADA16
P10-15
AD
Cell type
Objects
foreskin
fibroblasts
skin
fibroblasts
skin
fibroblasts
28,
male
29,
male
23,
male
16
weeks,
female
45,
female
48,
female
38,
female
45,
female
lung
fibroblasts
skin
fibroblasts
skin
fibroblasts
skin
fibroblasts
skin
fibroblasts
Mutation
Conversion
Efficiency
(% of initial
cells)
None
11.3±1.8
None
3.9±1.2
None
4.5±1.2
None
10.5±1.1
APP: p.V717I
12.5±0.7
PSEN1: p.I167del
12.6±1.1
PSEN1:p.A434T
9.6±1.5
PSEN1:p.S169del
10.6±1.9
Table S3, Related to Figure 1. Optimal concentration of each chemical in the
cocktail for hciN conversion. IMR90 fibroblasts were treated by chemicals with
different concentrations as indicated for 4 days and then maturation medium with
CFD was added for another 3 days. Induction efficiency was calculated from 10
random selected ×20 fields from triplicate samples and shown as means ± SEM.
VPA
(mM)
CHIR99021
(μM)
Repsox
(μM)
Forskolin
(μM)
SP600125
(μM)
GO6983
(μM)
Y-27632
(μM)
0
0.2
0.5
1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
3
3
3
3
0
0.5
3
5
10
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
0
1
5
10
20
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0
1
5
10
20
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0
5
10
20
50
10
10
10
10
10
10
10
10
10
10
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
0
1
5
10
20
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
0
1
5
10
20
Induction
efficiency
(% of initial cells)
1.8±0.5
2.9±0.6
5.4±0.9
7.8±0.9
10.5±1.9
8.1±1.1
2.1±0.6
1.2±0.4
0.7±0.3
5.2±1.0
5.0±1.0
0.9±0.3
1.9±0.6
1.8±0.6
5.9±2.6
8.5±2.2
7.4±1.2
3.6±1.1
1.6±0.7
13.6±1.7
9.2±2.0
5.5±1.0
0.4±0.2
0
8.2±2.1
10.3±2.6
5.3±0.7
2.7±0.8
1.1±0.6
2.4±0.7
5.1±1.3
7.9±2.1
13.5±2.9
9.6±2.1
Table S4, Related to Experimental Procedures. Primers for qRT-PCR.
Primers
hAscl1
hBrn2
hMyt1l
hNeuroD1
hNgn2
hFoxa2
hNurr1
hCol1a1
hDkk3
hThy1
hCtgf
hChd2
hSox2
hPax6
hFoxg1
hKcnj6
hScn3b
hCacna1h
hHPRT
Forward
5’-CAAGAGAGCGCAGCCTTAG-3’
5’-AATAAGGCAAAAGGAAAGCAACT-3’
5’-CAATGGAAAGGGATTTTAAGCA-3’
5’-GTTATTGTGTTGCCTTAGCACTTC-3’
5’-TCAGACATGGACTATTGGCAG-3’
5’-AGCAGAGCCCCAACAAGATG-3’
5’-GTTCAGGCGCAGTATGGGTC-3’
5’-GAGGGCCAAGACGAAGACATC-3’
5’-CTGGGAGCTAGAGCCTGATG-3’
5’-ATCGCTCTCCTGCTAACAGTC-3’
5’-CATCTCCACCCGGGTTACCAA-3’
5’-AGTCAGTCGGAAAGTGAGCAG-3’
5’-CAAGATGCACAACTCGGAGA-3’
5’-GGCAACCTACGCAAGATGGC-3’
5’-AGAAGAACGGCAAGTACGAGA-3’
5’-GAACTGGAGACTGAAGAGGAA G-3’
5’-GAAAGGTCTCAAAAGCCGAAG-3’
5’-TGACCTTCGGCAACTATGTG-3’
5’-CCTGGCGTCGTGATTAGTGAT-3’
Reverse
5’-GCAAAAGTCAGTGCTGAACG-3’
5’-CAAAACACATCATTACACCTGCT-3’
5’-TTTGAGATTATGTACCAACGTTAGATG-3’
5’-AGTGAAATGAATTGCTCAAATTGT-3’
5’-GGGACAGGAAAGGGAACC-3’
5’-TCTGCCGGTAGAAGGGGAAGA-3’
5’-CTCCCGAAGAGTGGTAACTGT-3’
5’-CAGATCACGTCATCGCACAAC-3’
5’-TCATACTCATCGGGGACCTC-3’
5’-CTCGTACTGGATGGGTGAACT-3’
5’-AGTACGGATGCACTTTTTGC-3’
5’-ACATCAGCTATCCGTTCCTTCT-3’
5’-CGGGGCCGGTATTTATAATC-3’
5’-TGAGGGCTGTGTCTGTTCGG-3’
5’-TGTTGAGGGACAGATTGTGGC-3’
5’-GACAAGGAAAGATTGTGTTGGG-3’
5’-CACTGCTCCTGTTCTATTCCTC-3’
5’-GGAGTTCTCTGAGCTTGTGG-3’
5’-AGACGTTCAGTCCTGTCCATAA-3’
Supplemental Experimental Procedures
Cell culture
Primary human adult fibroblasts (FS090609; provided by Dr. Ying Jin) were
established from dissociated foreskin tissue of a healthy donor collected by Shanghai
Renji Hospital (Ma et al., 2012). Primary human adult fibroblasts (SF001 and SF002;
provided by Dr. Duanqing Pei) were obtained from skin biopsy of healthy donors
collected by Guangzhou Institutes of Biomedicine and Health (Huang et al., 2014; Liu
et al., 2013). IMR90 was purchased from ATCC (CCL-186). FAD patient fibroblasts
(FAD109, FAD131, FAD132 and FADA16) were derived from skin biopsies of
familial Alzheimer disease patients collected by Hunan Xiangya Hospital (Caiazzo et
al., 2011; Jiao et al., 2014). The Ethical Committees of the Shanghai Renji Hospital,
Guangzhou Institutes of Biomedicine and Health and Hunan Xiangya Hospital
approved the collection and use of human samples. Informed consents were obtained
from all subjects. Primary fibroblast cultures were established from biopsies using
standard methods (Li et al., 2009). The detail information of these human fibroblasts
was listed in Supplement information, Table S2. Human fibroblasts were maintained
in DMEM (Life Technologies, C11965) supplemented with 10% fetal bovine serum
(Ausbian, VS500T), 1mM GlutaMAX (Life Technologies, 35050-061), 0.1 mM
non-essential amino acid (NEAA, Millipore, TMS-001-C) and penicillin/streptomycin
at 37oC with 5% CO2. H9 human ES cells (WiCell, transferred from Dr. Xiaoqing
Zhang) were expanded in mTeSR1 (Stem Cell Technologies) and passaged as clumps
with dispase or EDTA. Neural stem cells were induced from hES cells by dual
inhibition of SMAD signaling as previously described (Chambers et al., 2009).
hES-derived NPCs were cultured in neuronal differentiation medium with BDNF,
GDNF, IGF (all at 10 ng/ml, PeproTech), cAMP (1 μM, Sigma-Aldrich) and ascorbic
acid (200 ng/ml, Sigma-Aldrich) to induce neuronal differentiation (Zhang and Zhang,
2010).
Generation of hciN cells
Reagents:
VPA, 0.5 mM (Calbiochem, 676380, 1 M stock in H2O);
CHIR99021, 3 μM (Axon, 1386, 10 mM stock in DMSO);
Repsox, 1 μM (BioVision, 1894, 10 mM stock in DMSO);
Forskolin, 10 μM (Cayman, 66575-29-9, 50 mM stock in DMSO);
SP600125, 10 μM (Sigma-Aldrich, S5567, 25 mM stock in DMSO);
GO6983, 5 μM (Sigma, G1918, 10 mM stock in DMSO);
Y-27632, 5 μM (Sigma-Aldrich, Y0503, 20 mM stock in DMSO);
Dorsomorphin, 1 μM (Cayman, 11967, 10 mM stock in DMSO);
cAMP, 100 μM (Sigma-Aldrich, D0260, 100 mM stock in H2O);
bFGF, 20 ng/ml (Invitrogen, PHG0024, 100 μg/ml stock in H2O);
BDNF, 20 ng/ml (PeproTech, 450-02, 100 μg/ml stock in H2O);
GDNF, 20 ng/ml (PeproTech, 450-10, 100 μg/ml stock in H2O);
NT3, 20 ng/ml (PeproTech, 450-03, 100 μg/ml stock in H2O).
Medium:
Fibroblast medium: DMEM (Life Technologies, C11965) supplemented with 10%
fetal bovine serum (Ausbian, VS500T), 1mM GlutaMAX (Life Technologies,
35050-061), 0.1 mM non-essential amino acid (NEAA, Millipore, TMS-001-C) and
penicillin/streptomycin.
Neuronal induction medium: (DMEM/F12, Life Technologies, 11330-032):
(Neurobasal, Life Technologies, 21103-049) (1:1), 0.5% N-2 (Invitrogen, 17502048),
1% B-27 (Invitrogen, 17504044), 100 μM cAMP and 20 ng/ml bFGF and
penicillin/streptomycin.
Neuronal maturation medium: DMEM/F12: Neurobasal (1:1), 0.5% N-2, 1% B-27,
100 μM cAMP, 20 ng/ml bFGF, 20 ng/ml BDNF, 20 ng/ml GDNF, 20 ng/ml NT3 and
penicillin/streptomycin.
Neuron medium: DMEM/F12: Neurobasal (1:1), 0.5% N-2, 1% B-27, 20 ng/ml
BDNF, 20 ng/ml GDNF, 20 ng/ml NT3 and penicillin/streptomycin.
Procedure:
1. 24-wells or 6-wells plates with or without coverslips were coated with 0.1% gelatin
for more than half hour.
2. Initial fibroblasts were seeded onto gelatin-coated culture plates (10,000-20,000
cells/well in 24-well plates) and cultured in fibroblast medium for 24 hours.
3. The cell confluence was about 40-60%, and the cells were transferred into neuronal
induction medium with chemical cocktails VCRFSGY. The concentration of
chemicals used is: VPA, 0.5 mM; CHIR99021, 3 μM; Repsox, 1 μM; Forskolin, 10
μM; SP600125, 10 μM; GO6983, 5 μM; Y-27632, 5 μM .
4. Medium containing chemical compounds was changed every four days.
5. After eight days (4 days for IMR90 cells), cells were switched to neuronal
maturation medium with CFD (CHIR99021, 3 μM; Forskolin, 10 μM; and
Dorsomorphin, 1 μM).
6. Maturation medium was half-changed every other day.
7. After two weeks, induced neuronal cells were cultured in neuron medium to
promote neuron survival and maturation.
8. From now on, half-change the neuron medium every other day until the cells are
subjected to further analysis.
TF-based neuron induction
TF (Ascl1+NeuroG2)-induced neurons were prepared as described previously with
little
modification
(Ladewig
et
al.,
2012).
Tet-O-FUW-Ascl1
and
CAG-Ngn2-IRES-DsRed were used to express Ascl1 and NeuronG2. For production
of lentiviral particles, HEK-293T cells were co-transfected with the packaging
plasmid PAX2, the envelope plasmid VSVG and Tet-O-FUW-Ascl1 lentiviral vector
plasmid. For production of retroviral particles, HEK-293T cells were co-transfected
with the packaging plasmid Gag-pol, the envelope plasmid VSVG and
CAG-Ngn2-IRES-DsRed retroviral vector plasmid. All viral particles were enriched
by centrifugation. Fibroblasts were transduced with these viral particles in the
presence of 6 ug/ml polybrene. After overnight incubation, fibroblasts were grown in
fibroblast medium with doxycycline (1 μg/ml, Selleck). One day later, these cells
were grown in neuronal medium containing DMEM/F12 (1% (vol/vol) N2
supplement, Invitrogen) and Neurobasal (2% (vol/vol) B27 supplement, Invitrogen)
mixed at a 1:1 ratio, cAMP (100 ng/ml, Sigma-Aldrich) and doxycycline (1 μg/ml,
Selleck) with SB-431542 (10 μM, Sigma-Aldrich), noggin (100 ng/ml Peprotech),
LDN-193189 (0.5 μM, Selleck) and CHIR99021 (2 μM, Axon). Medium were
changed every three days. Two weeks later, neurons was promoted by adding 10
ng/ml BDNF, 2 ng/ml GDNF, 10 ng/ml NT3 (Peprotech) and 0.5 mM dcAMP
(Sigma-Aldrich) to the neuronal medium.
Conversion efficiency
Conversion efficiency was calculated as previously described (Vierbuchen et al.,
2010). Briefly, we randomly selected 10–20 view fields for each sample on an
Olympus IX-51 microscope at indicated time points, and counted the total cell
number of Tuj1+/Dcx+, Tuj1+/Map2+ or Tuj1+/NeuN+ cells with neuron morphology.
The conversion efficiency was calculated as the ratio of Tuj1+/Dcx+, Tuj1+/Map2+or
Tuj1+/NeuN+ hciN cells to the initial seeding cells in each visual field. The purity
represent the percentage of induced Tuj1+/Dcx+, Tuj1+/Map2+ or Tuj1+/NeuN+
neuronal cells in total cells stained by DAPI. Quantitative data are presented as means
±SEM of at least three independent experiments.
Immunofluoresent microscopy
Immunostaining of cells were done as previously described (Hu et al., 2013). In brief,
cells plated on glass coverslips were fixed by 4% PFA solution for 10 min at room
temperature. Then cells were incubated in blocking buffer (1% bovine serum albumin
in PBS) with or without 0.5% Triton X-100 for 30 min at room temperature (RT).
Afterwards, samples were incubated with primary antibodies at 4 °C overnight and
then with appropriate fluorescent probe-conjugated secondary antibodies for 1 h at RT.
Nuclei were counterstained with DAPI. Images were captured with fluorescence
microscope (Olympus IX-51) or Leica SP-8 confocal microscope. The primary
antibodies used : Dcx (1:200, Abcam), Tuj1 (1:500, Covance), Map2 (1:500,
Millipore), NeuN (1:500, Millipore), SYN1 (1:500, Millipore), vGLUT1 (1:500,
Synaptic system), GAD67 (1:500, Millipore), Chat (1:200, Millipore), TH (1:1000,
Sigma-Aldrich), Nestin (1:1000, Millipore), Sox2 (1:50, R&D).
Electrophysiological analysis
Whole-cell patch clamp recordings were carried out just as our previous report
(Cheng et al., 2014). Recordings were made using Multiclamp 700B amplifier
(Molecular Devices). The bath was constantly perfused with fresh artificial
cerebrospinal fluid (ACSF) at 32-34℃. The ACSF contained (in mM) 126 NaCl, 3
KCl, 1.25 KH2PO4, 1.3 MgSO4, 2.4 CaCl2, 26 NaHCO3, and 10 glucose, bubbled with
95% O2/5% CO2. Signals were sampled at 10 kHz with a 2 kHz low-pass filter. The
whole-cell capacitance was fully compensated. Recordings with Ra > 50M or
fluctuation > 20 % were excluded. The intracellular solution contained (in mM): 93
K-gluconate, 16 KCl, 2 MgCl2, 10 HEPES, 4 ATP-Mg, 0.3 GTP-Na2, 10
Na-phosphocreatine, 0.5% Alexa Fluor 568 hydrazide (pH 7.25, 290/300 mOsm), and
0.4% neurobiotin. Membrane potentials were hold around -83 mV. Action potentials
were recorded in current clamp and sodium currents were recorded in voltage clamp.
Step voltages with an increment of 10 mV were injected to elicit inward sodium
current. To block sodium current, 1 μM TTX were added into the chamber, step
voltages were injected again to observe the TTX-sensitive currents 10 minutes after
TTX applying. GABA or L-glutamic acid (10 μM each, Sigma-Aldrich) were
dissolved in ACSF, loaded into a pipette , and puffed onto hciN cells using a
picosprizer III (Parker Instrument) with 10 psi and 1s duration. The puffing pipette
was ~10 MΩ and placed ~10 μm away from the soma of the recorded neuron. To
record spontaneous synaptic currents, we labeled HAFs with lentivirus encoding GFP
and cultured GFP-pre-hciN cells with monolayer astrocytes culture. Two weeks after
replating, we carried out whole-cell patch-clamp recording on GFP-labeled hciN cells
with complex neuron morphology. EPSCs were blocked by NBQX (10 µM) and
IPSPs were inhibited by bicuculline (10-50 µM). Data were analyzed using pClamp10
(Clampfit) and Mini analysis (Synaptosoft).
Calcium imaging
Culture medium was removed and cultures were incubated with 22.5 µM Cal520
(AAT Bioquest), 4.8% DMSO, 0.1% Pluronic F-127 (Life technologies) and 0.05%
Cremophor EL (Sigma-Aldrich) in normal ACSF solution (126 mM NaCl, 3 mM KCl,
1.25 mM KH2PO4, 1.3 mM MgSO4, 2.4 mM CaCl2, 26 mM NaHCO3, and 10 mM
glucose) for 30 min at room temperature. Cultures were gently washed with ACSF
solution and incubated in ACSF for additional 2 min. Calcium imaging (×20
magnification) were acquired at 15.6 Hz using a Hamamatsu ORCA-ER digital
camera with a 488 nm (FITC) filter on an Olympus BX51WI microscope. The bath
was constantly perfused with normal ACSF solution at 32℃. In each region of interest,
time-lapse image sequences of 4000 frames were acquired with Winfluor. In total, 22
movies of spontaneous calcium transients were made and 1249 putative neurons were
analyzed.
FACS Sorting
hiN cells and TF-iNs were typsinized and harvested. After gentle trituration, cell were
then pelleted by centrifugation at 1000 rpm and resuspended in 200 ul staining buffer
(PBS, 2% fetal bovine serum [FBS]). Cells were stained with PE-cy7-conjugated
anti-CD24 antibodies according to the manufacturer’s protocol. Sorting threshold was
setup based on control groups.
Single-Cell Gene Expression Analysis
3 weeks after chemicals treatment, hciN cells were harvested and cells were collected
by FACS sorting. Single cells were captured on a microfluidic C1™ integrated fluidic
circuits (IFCs) using Fluidigm C1™ Single-Cell Auto Prep system. Amplicons were
made with Fluidigm C1™ modules and Single Cell-to-CT™ Kit (Life Technologies).
Products were subjected to target-specific reverse transcription and PCR
pre-amplification with a mixture of primers specific to the target genes (STA).
Primers used were as reported previously (Victor et al., 2014; Zhang et al., 2013).
Sequences of fibroblast specific gene primers, ctgf，col1a1，dkk3 and thy1 were the
same as that shown in Table S4.Single-cell gene expression profiling was then
performed using Fluidigm Biomark™ HD system. STA products were processed for
real-time PCR analysis on Biomark Dynamic Array IFCs (Fluidigm). Clustering
analysis was performed with Ct values of corresponding genes using the statistical
software R. The Ct values for each gene of interest were normalized to the median Ct
for each cell; unexpressed genes with no detectable Ct value were given a value of
max detectable Ct value plus 4 for computational purposes.
Microarray analysis
RNA was extracted from 0.5-1×106 cells. The total RNAs were reverse-transcribed
and hybridized to PrimeView™ Human Gene Expression Array (Affymetrix)
following the manufacturer’s instruction. Microarray hybridization and whole genome
expression analysis were performed by Shanghai OE Biotech Co., Ltd. Affymetrix
GeneChip Command Console (version 4.0, Affymetrix) was used to analyze array
images to get raw data. Next, Genespring software (version 12.5, Agilent
Technologies) was employed to finish the basic analysis with the raw data.
Hierarchical clustering analysis was applied using a Euclidean distance matrix and the
complete-linkage clustering method with selected genes that show 10-fold change in
hciN cells (day 14) versus that in HAFs. Afterwards, differentially expressed genes
with expression level changes ≥ 2-fold were applied for GO analysis by DAVID
Bioinformatics Resources.
Quantitative real-time PCR
Total RNA samples were collected with Trizol (Sigma-Aldrich) according to
manufacturer’s instructions. 1.0 μg RNA for each reaction was reverse-transcribed to
cDNA with random hexamers and M-MLV Reverse Transcriptase (Promega).
Quantitative real-time PCR was subsequently conducted with specific primers and 2 ×
HotStart SYBR Green qPCR Master Mix (Excell, MB000-3013) in an MX3000P
Stratagene PCR machine. The relative expression levels were normalized against the
internal control (HPRT). Primers used were listed in Supplement information, Table
S4.
ELISA analysis
The accumulation of human Aβ40 and Aβ42 in the cell culture medium was
quantified using ELISA kits (Invitrogen) according to manufacturer’s instructions.
Medium was harvested at day 14. Samples collected from three independent
experiments were measured, and the data were shown means ±SEM.
Western blot
HciN and TF-iN cell were washed with cold PBS and lysed with Laemmli's sample
buffer, and then cell lysates were separated on SDS-PAGE, and transferred onto
nitrocellulose
membrane.
Primary
antibodies
were
as
follows:
rabbit
anti-p-GSK-3(1:1000Cell signaling Technology), mouse anti-t-GSK-3 (1:1000,
Cell signaling Technology), mouse anti-p-Tau (1:1000, AT-180, Thermo scientific),
mouse anti-t-Tau (1:1000, Thermo scientific) and rabbit anti-actin (1:1000, Sigma).
The membrane was detected by secondary antibody conjugated to a CW800
fluorescent probe (Rockland) using an infrared imaging system (Odyssey, LI-COR).
Statistical Analysis
All quantified data were statistically analyzed and presented as means ± SEM.
One-way ANOVA and Bonferroni’s Multiple comparison Test were used to calculate
statistical significance with P values, unless otherwise stated.
Supplemental Reference
Caiazzo, M., Dell'Anno, M.T., Dvoretskova, E., Lazarevic, D., Taverna, S., Leo, D., Sotnikova, T.D.,
Menegon, A., Roncaglia, P., Colciago, G., et al. (2011). Direct generation of functional dopaminergic
neurons from mouse and human fibroblasts. Nature 476, 224-227.
Chambers, S.M., Fasano, C.A., Papapetrou, E.P., Tomishima, M., Sadelain, M., and Studer, L. (2009).
Highly efficient neural conversion of human ES and iPS cells by dual inhibition of SMAD signaling.
Nat Biotechnol 27, 275-280.
Cheng, L., Hu, W., Qiu, B., Zhao, J., Yu, Y., Guan, W., Wang, M., Yang, W., and Pei, G. (2014).
Generation of neural progenitor cells by chemical cocktails and hypoxia. Cell Res 24, 665-679.
Hu, W., Zhao, J., and Pei, G. (2013). Activation of aryl hydrocarbon receptor (ahr) by tranilast, an
anti-allergy drug, promotes miR-302 expression and cell reprogramming. J Biol Chem 288,
22972-22984.
Huang, K., Liu, P., Li, X., Chen, S., Wang, L., Qin, L., Su, Z., Huang, W., Liu, J., Jia, B., et al. (2014).
Neural progenitor cells from human induced pluripotent stem cells generated less autogenous immune
response. Science China Life sciences 57, 162-170.
Jiao, B., Tang, B.S., Liu, X.Y., Xu, J., Wang, Y.J., Zhou, L., Zhang, F.F., Yan, X.X., Zhou, Y.F., and
Shen, L. (2014). Mutational analysis in early-onset familial Alzheimer's disease in Mainland China.
Neurobiol Aging 35.
Ladewig, J., Mertens, J., Kesavan, J., Doerr, J., Poppe, D., Glaue, F., Herms, S., Wernet, P., Kogler, G.,
Muller, F.J., et al. (2012). Small molecules enable highly efficient neuronal conversion of human
fibroblasts. Nature methods 9, 575-578.
Li, C., Yu, H., Ma, Y., Shi, G., Jiang, J., Gu, J., Yang, Y., Jin, S., Wei, Z., Jiang, H., et al. (2009).
Germline-competent mouse-induced pluripotent stem cell lines generated on human fibroblasts without
exogenous leukemia inhibitory factor. PLoS One 4, e6724.
Liu, P., Chen, S., Li, X., Qin, L., Huang, K., Wang, L., Huang, W., Li, S., Jia, B., Zhong, M., et al.
(2013). Low immunogenicity of neural progenitor cells differentiated from induced pluripotent stem
cells derived from less immunogenic somatic cells. PloS one 8, e69617.
Ma, Y., Li, C.L., Gu, J.J., Tang, F., Li, C., Li, P., Ping, P., Yang, S., Li, Z., and Jin, Y. (2012). Aberrant
Gene Expression Profiles in Pluripotent Stem Cells Induced from Fibroblasts of a Klinefelter
Syndrome Patient. J Biol Chem 287, 38970-38979.
Victor, M.B., Richner, M., Hermanstyne, T.O., Ransdell, J.L., Sobieski, C., Deng, P.Y., Klyachko, V.A.,
Nerbonne, J.M., and Yoo, A.S. (2014). Generation of human striatal neurons by microRNA-dependent
direct conversion of fibroblasts. Neuron 84, 311-323.
Vierbuchen, T., Ostermeier, A., Pang, Z.P., Kokubu, Y., Sudhof, T.C., and Wernig, M. (2010). Direct
conversion of fibroblasts to functional neurons by defined factors. Nature 463, 1035-1041.
Zhang, X.Q., and Zhang, S.C. (2010). Differentiation of neural precursors and dopaminergic neurons
from human embryonic stem cells. Methods Mol Biol 584, 355-366.
Zhang, Y.S., Pak, C., Han, Y., Ahlenius, H., Zhang, Z.J., Chanda, S., Marro, S., Patzke, C., Acuna, C.,
Covy, J., et al. (2013). Rapid Single-Step Induction of Functional Neurons from Human Pluripotent
Stem Cells. Neuron 78, 785-798.