Research ArticleStem Cells

Drug Screening for ALS Using Patient-Specific Induced Pluripotent Stem Cells

See allHide authors and affiliations

Science Translational Medicine  01 Aug 2012:
Vol. 4, Issue 145, pp. 145ra104
DOI: 10.1126/scitranslmed.3004052

Abstract

Amyotrophic lateral sclerosis (ALS) is a late-onset, fatal disorder in which the motor neurons degenerate. The discovery of new drugs for treating ALS has been hampered by a lack of access to motor neurons from ALS patients and appropriate disease models. We generate motor neurons from induced pluripotent stem cells (iPSCs) from familial ALS patients, who carry mutations in Tar DNA binding protein-43 (TDP-43). ALS patient–specific iPSC–derived motor neurons formed cytosolic aggregates similar to those seen in postmortem tissue from ALS patients and exhibited shorter neurites as seen in a zebrafish model of ALS. The ALS motor neurons were characterized by increased mutant TDP-43 protein in a detergent-insoluble form bound to a spliceosomal factor SNRPB2. Expression array analyses detected small increases in the expression of genes involved in RNA metabolism and decreases in the expression of genes encoding cytoskeletal proteins. We examined four chemical compounds and found that a histone acetyltransferase inhibitor called anacardic acid rescued the abnormal ALS motor neuron phenotype. These findings suggest that motor neurons generated from ALS patient–derived iPSCs may provide a useful tool for elucidating ALS disease pathogenesis and for screening drug candidates.

Introduction

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder characterized by a loss of upper and lower motor neurons that typically develops in the fifth or sixth decade of life, with a survival of less than 5 years and a prevalence of 2 in 100,000 (1, 2). The histopathological hallmarks of this fatal disease include cytosolic aggregates in the motor neurons of most ALS patients with the sporadic form of the disease. These aggregates are composed of Tar DNA binding protein-43 (TDP-43) (35), a 414–amino acid nuclear mRNA binding protein containing two RNA recognition motifs. Genetic analysis has identified more than 30 mutations in the TDP-43 gene in both familial and sporadic ALS cases (6). ALS-associated abnormalities have been reported in patient samples and cellular and animal models (610), and several compounds have been identified as abrogating the disease phenotype in an ALS mouse model. However, when these compounds were tested in ALS patients, no clinical improvements were observed (11).

Induced pluripotent stem cells (iPSCs) have been generated from ALS patients and differentiated into motor neurons (12, 13), but it is not yet clear whether the abnormal cellular and molecular phenotypes of ALS can be recapitulated in vitro. A lack of access to human motor neurons and appropriate disease models has hampered efforts to test new drug candidates for ALS. Here, we generated human motor neurons from iPSCs derived from familial ALS patients carrying TDP-43 mutations and used them to identify a compound that rescued the ALS-associated phenotype.

Results

Spinal motor neurons were generated from iPSCs derived from dermal fibroblasts from patients with familial ALS or from control individuals by means of retroviral or episomal vectors. Seven control human iPSC lines were derived from five unrelated individuals without mutations in the TDP-43 gene, and nine ALS iPSC lines were generated from three ALS patients with mutations in TDP-43 (14, 15) (Fig. 1A, fig. S1A, and table S1). ALS patients A21, A34, and ND32947 were heterozygous for the Q343R, M337V, and G298S mutations in TDP-43, respectively. Neural populations including motor neurons derived from the ALS patient iPSCs retained these TDP-43 mutations (Fig. 1B). The iPSCs generated from ALS and control dermal fibroblasts expressed human embryonic stem cell (ESC) markers (Fig. 1A and fig. S1A) and were fully characterized (figs. S1 to S5 and table S2). The ALS and control iPSC lines were then differentiated into motor neurons (16) (Fig. 1C and fig. S6A). Differentiated motor neurons were identified by expression of motor neuron markers including Islet-1, HB9, SMI-32, and ChAT (choline acetyltransferase) (Fig. 1C and fig. S6B), and their function was verified by coculture with myotubes (fig. S6, D to F). To visualize live motor neurons, we transduced them with lentivirus expressing green fluorescent protein (GFP) under the control of the HB9 promoter (HB9::GFP) (17) (Fig. 1C). HB9::GFP-positive neurons colocalized with ChAT (fig. S6C) and showed spontaneous action potentials and synaptic potentials (fig. S6, G to K). Immunocytochemical analysis with MAP2, SMI-32, and GFAP (glial fibrillary acidic protein) did not reveal any differences in differentiation propensity between ALS and control iPSCs (Fig. 1, D to G, and fig. S7).

Fig. 1

Generation of ALS patient–specific iPSCs and iPSC-derived motor neurons. (A) ALS and control iPSC lines were generated from dermal fibroblasts from ALS patients carrying mutations in TDP-43 or from control individuals who did not carry TDP-43 mutations. iPSCs were morphologically identical to human ESCs and expressed the human ESC markers Nanog and SSEA-4. Scale bars, 200 μm. (B) Human dermal fibroblasts (HDFs) from ALS patients ALS21, ALS34, and ND32947 carried TDP-43 mutations Q343R, M337V, and G298S, respectively, and the mutations were maintained in motor neuron–containing neural populations (neural cells) for more than 22 days after differentiation from iPSCs. (C) After differentiation of iPSCs into motor neurons, they were stained with motor neuron markers Islet-1 and HB9. Punctate staining of synapsin was detected as a marker of presynaptic structure in SMI-32–positive ALS iPSC–derived motor neurons. HB9::GFP-positive motor neurons colocalized with ChAT. Scale bars, 50 μm. (D to G) Motor neuron differentiation from ALS and control iPSCs. After 2 months of differentiation from iPSCs in culture, motor neurons were stained with the MAP2 neuronal marker and the motor neuron–specific marker SMI-32. Astrocytes were revealed by staining with GFAP. Scale bar, 50 μm. Proportions of ALS and control iPSC–derived motor neurons staining positive for MAP2 (E), GFAP (F), and SMI-32 (G) are shown. Error bars are SD.

Next, we examined whether ALS iPSC–derived motor neurons have shorter neurites as reported in an ALS zebrafish model (6) and decreased neurofilament mRNA as reported in postmortem tissues from ALS patients (7, 18). ALS iPSC–derived motor neurons expressing HB9::GFP were purified by fluorescence-activated cell sorting (FACS) (17, 19) (Fig. 2A and fig. S8, A to C). ALS iPSC–derived motor neurons showed shorter neurites [33.5 ± 9.9 μm (mean ± SD) compared to 63.8 ± 13.1 μm for control; P = 2.0 × 10−4 by t test] (Fig. 2, A and B, and fig. S9A). Gene expression profiling of the purified ALS iPSC–derived motor neurons showed that there was a decrease in expression of genes encoding components of cytoskeletal intermediate filaments (control > ALS, fold change >1.2, P < 0.01) (Fig. 2D, fig. S8, F and G, and table S4). The expression of medium polypeptide neurofilament (NEFM) and light polypeptide neurofilament (NEFL) was significantly decreased in ALS compared to control iPSC–derived motor neurons [P = 7.9 × 10−3 (NEFL); P = 5.3 × 10−3 (NEFM); t test] (Fig. 2C and table S6).

Fig. 2

Phenotypes of ALS iPSC–derived motor neurons. (A) ALS iPSC–derived HB9::GFP-positive motor neurons are shown before and after sorting by FACS. (Lower panels) Representative images of purified ALS and control iPSC–derived motor neurons. Scale bars, 10 μm. (B) The length of neurites was measured in purified ALS and control iPSC–derived motor neurons. P = 2.0 × 10−4 by t test. Error bars are SEM. (C) The expression of NEFL and NEFM was decreased in ALS versus control iPSC–derived motor neurons. P = 7.9 × 10−3, NEFL; P = 5.3 × 10−3, NEFM by t test. (D) Major GO terms showed both increases and decreases in gene expression in ALS versus control iPSC–derived motor neurons. (E) qPCR showed increased expression of TDP-43 mRNA relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA in ALS versus control iPSC–derived motor neurons. P = 0.040 by t test. (F) Soluble (upper panel) and insoluble (lower panel) TDP-43 fractions from ALS and control iPSC–derived motor neuron–containing neural populations. (G) ALS and control iPSC–derived motor neurons were immunostained for TDP-43 and for the SMI-32 motor neuron marker; DAPI nuclear stain, blue. Scale bars, 10 μm. (H) Number of TDP-43–positive cytosolic aggregates measured by high-content analysis. P = 0.0458 by t test. Error bars are SEM. (I) Immunoblots of different proteins including TDP-43 with antibodies against indicated proteins. Insoluble TDP-43 was immunoprecipitated (IP) from the insoluble fraction and immunoblotted (IB) with each antibody. (J) Quantification of protein band densities of TDP-43 and SNRPB2 in insoluble fractions from total lysates and in TDP-43–immunoprecipitated insoluble fractions relative to β-actin. P = 0.034 (TDP-43 in the insoluble fraction from total lysates of ALS compared to control motor neuron–containing neural populations) by t test; P = 1.0 × 10−4 (SNRPB2 coimmunoprecipitated with TDP-43 in ALS compared to control motor neuron–containing neural populations) by t test. (K) HB9::GFP-labeled ALS and control iPSC–derived motor neurons were immunostained for SNRPB2 and TDP-43 and counterstained with DAPI nuclear stain (blue). Scale bar, 10 μm. (L) Representative images showing immunostaining for SNRPB2 in postmortem spinal cord tissue from ALS patients or control individuals. Arrowheads, ALS spinal cord tissue immunoreactive for both TDP-43 and SNRPB2 in serial sections. Scale bars, 10 μm. *P < 0.05, **P < 0.01.

TDP-43 is involved in multiple steps of RNA metabolism, including transcription, splicing, and transport of mRNA (2022). TDP-43 protein autoregulates its synthesis by binding to the 3′ untranslated region of its own mRNA in a negative feedback loop (8). Gene expression profiles of purified ALS iPSC–derived motor neurons demonstrated that gene ontology (GO) terms related to RNA binding, splicing, processing, and transcriptional initiation were enriched compared to the profiles of control iPSC–derived motor neurons (Fig. 2D). The transcripts of nuclear transport/RNA granule and spliceosomal complex–related genes were also up-regulated in ALS iPSC–derived motor neurons (Fig. 2D, fig. S8, D, E, and G, and tables S3 and S5), indicating that RNA metabolism may be perturbed in ALS compared to control iPSC–derived motor neurons (8, 20). The expression of TDP-43 mRNA was significantly increased in ALS iPSC–derived motor neurons compared to control (P = 0.040 by t test; Fig. 2E and fig. S8, D and G) (8, 23).

TDP-43 has been reported in the detergent-insoluble fraction of postmortem tissues from ALS patients (3, 4, 24) and has been shown to be mislocalized, forming cytoplasmic preinclusions (20). To analyze the biochemical properties of TDP-43 in ALS iPSC–derived neural populations containing motor neurons, we performed Western blot analysis. The amount of detergent-insoluble TDP-43 including full-length and smaller fragments increased markedly in ALS iPSC–derived motor neuron–containing neural populations (Fig. 2F and fig. S9B). Immunocytochemical analysis revealed that TDP-43 in control iPSC–derived motor neurons was mainly localized in the nucleus, whereas TDP-43 in ALS iPSC–derived motor neurons was distributed in both the nucleus and the cytoplasm (Fig. 2G), with TDP-43 in the cytoplasm forming preinclusion-like aggregates. High-content analysis revealed that the number of such TDP-43 aggregates in ALS motor neurons was increased compared to that in control motor neurons (P = 0.0458 by t test; Fig. 2H).

Next, we analyzed SNRPB2, a highly expressed spliceosomal factor, in ALS iPSC–derived motor neurons (fig. S8E). The amount of TDP-43 protein was increased (P = 0.034 by t test; Fig. 2, I and J), and the amount of SNRPB2 bound to TDP-43 was increased (P = 1.0 × 10−4 by t test; Fig. 2, I and J) in the insoluble fraction of ALS iPSC–derived motor neuron–containing neural populations compared to control. TDP-43 colocalized with SNRPB2 to form aggregates in the nucleus of ALS iPSC–derived motor neurons (Fig. 2K) and motor neurons from postmortem tissue from ALS patients (Fig. 2L).

Next, we established an assay to measure death of ALS iPSC–derived motor neurons in response to arsenite, which induces oxidative stress and an increase in the amount of insoluble TDP-43 in these motor neurons (25). Arsenite increased the amount of TDP-43 in the insoluble fraction of ALS iPSC–derived motor neuron–containing neural populations [P = 7.8 × 10−3 by two-way analysis of variance (ANOVA); Fig. 3, A and B]. The number of surviving ALS iPSC–derived motor neurons was lower compared to control iPSC–derived motor neurons (18% reduction in ALS motor neurons relative to control; P = 5.0 × 10−4 by two-way ANOVA) (Fig. 3, C and D). Staining with ethidium homodimer-1, which selectively permeates the broken membranes of dying cells, revealed that the proportion of dying ALS iPSC–derived motor neurons was greater than that of control iPSC–derived motor neurons [33.5 ± 1.9% (mean ± SD) for ALS compared to 16.1 ± 3.0% for control; P = 2.9 × 10−3 by two-way ANOVA] (Fig. 3, E and F).

Fig. 3

Arsenite-induced death of ALS and control iPSC–derived motor neurons. (A and B) Decreased solubility of TDP-43 after treatment with arsenite. (A) After treatment with vehicle or arsenite (0.5 mM, 1 hour), cell lysates were separated into soluble and insoluble fractions and immunoblotted with TDP-43 antibody. (B) Arsenite increased the amount of TDP-43 in the insoluble fraction in ALS iPSC–derived motor neurons (n = 4) compared to control (n = 3). P = 7.8 × 10−3 by two-way ANOVA. (C) Images of control versus ALS iPSC–derived motor neurons after treatment with vehicle or arsenite. Arrowheads, dying motor neurons. Scale bar, 100 μm. (D) Proportion of HB9::GFP-positive control and ALS iPSC–derived motor neurons after treatment with vehicle or arsenite. P = 5.0 × 10−4 by two-way ANOVA. Error bars are SEM. (E) Death of motor neurons was assessed by ethidium homodimer-1 (EthD-1) uptake after incubation with vehicle or arsenite. P = 2.9 × 10−3 by two-way ANOVA. Error bars are SEM. (F) Images of ALS versus control iPSC–derived motor neurons after arsenite treatment. Scale bar, 100 μm. (G) Treatment with 5 μM anacardic acid (n = 6) for 16 hours blocked arsenite-induced death of ALS iPSC–derived motor neurons. Trichostatin A (3 μM) (n = 3), spliceostatin A (100 ng/ml) (n = 3), and garcinol (5 μM) (n = 6) failed to block arsenite-induced cell death. P = 0.048 by one-way ANOVA. Error bars are SEM. (H) Images of anacardic acid–treated motor neurons followed by arsenite treatment. Scale bar, 100 μm. *P < 0.05, **P < 0.01.

Given that gene expression analysis suggested that transcription and RNA splicing were perturbed in ALS iPSC–derived motor neurons, we tested four drugs that had been reported to modulate transcription through histone modification or RNA splicing in the arsenite-induced motor neuron death assay. The drugs were trichostatin A (a histone deacetyltransferase inhibitor), spliceostatin A (a spliceosomal factor inhibitor) (26), and two histone acetyltransferase inhibitors, anacardic acid (27) and garcinol (27). We identified anacardic acid as protecting against arsenite-induced death of ALS iPSC–derived motor neurons [survival rate, 81.7 ± 4.5% (mean ± SEM) for arsenite-treated ALS motor neurons compared to 97.1 ± 4.8% for arsenite-treated ALS motor neurons pretreated with anacardic acid; P = 0.048 by one-way ANOVA] (Fig. 3, G and H).

Anacardic acid decreased TDP-43 mRNA expression in ALS iPSC–derived motor neurons by 147-fold compared to untreated motor neurons (P = 0.047 by two-way ANOVA) (Fig. 4A). To investigate the effect of anacardic acid on the production of TDP-43 protein, we treated ALS iPSC–derived motor neuron–containing neural populations with anacardic acid for 48 hours. Anacardic acid reduced the amount of TDP-43 in the insoluble fraction but not in the soluble fraction of ALS iPSC–derived motor neuron–containing neural populations (P = 8.2 × 10−3 by t test) (Fig. 4, B and C). Anacardic acid also increased the length of neurites of purified ALS iPSC–derived motor neurons compared to untreated ALS iPSC–derived motor neurons [average neurite length, 75.4 ± 7.5 μm (mean ± SEM) for anacardic acid–treated ALS motor neurons compared to 36.2 ± 2.5 μm for vehicle-treated ALS motor neurons; P = 0.014 by two-way ANOVA] (Fig. 4D and fig. S9C). The drug also increased expression of NEFM mRNA (Fig. 4E), down-regulated expression of RNA metabolism–related genes (Fig. 4F), and reversed changes in the tumor necrosis factor–α (TNFα)/nuclear factor κB (NF-κB) signaling pathway (Fig. 4G and fig. S10, A to D).

Fig. 4

Anacardic acid–induced phenotypic changes in ALS and control motor neurons. (A) qPCR confirmed that 5 μM anacardic acid treatment for 16 hours down-regulated TDP-43 mRNA expression in purified ALS iPSC–derived motor neurons. P = 0.047 by two-way ANOVA. Error bars are SEM. (B) After treatment with vehicle or 5 μM anacardic acid for 48 hours, cells were lysed, separated into soluble and insoluble fractions, and immunoblotted with TDP-43. (C) Quantification of protein band densities after immunoblotting. The dotted line indicates the baseline (vehicle only) of relative protein levels (anacardic acid/vehicle). P = 8.2 × 10−3 by t test. Error bars are SEM. (D) Neurite length of purified motor neurons was measured 16 hours after treatment with vehicle or 5 μM anacardic acid. P = 0.014 by two-way ANOVA. Error bars are SEM. (E) qPCR revealed that anacardic acid treatment up-regulated expression of NEFM mRNA in ALS iPSC–derived motor neurons. P = 0.032 by t test. Error bars are SEM. (F and G) After purified motor neurons were treated for 16 hours with vehicle or 5 μM anacardic acid, cells were lysed and analyzed by expression array profiling. (Yellow bars) P value for significant changes in expression of genes and signaling pathways in ALS versus control iPSC–derived motor neurons was expressed on a logarithmic scale. (Green bars) P value for GO terms and signaling pathways of anacardic acid–treated ALS motor neurons compared to vehicle-treated ALS motor neurons was expressed on a logarithmic scale. *P < 0.05, **P < 0.01. TGFβ, transforming growth factor–β.

Discussion

We found cellular and molecular phenotypes associated with ALS in motor neurons derived from ALS patient–specific iPSCs harboring mutant TDP-43. Consistent with a previous study describing the generation of iPSCs from ALS patients carrying the TDP-43 M337V mutation (13), in our study, the TDP-43 mutation did not affect the differentiation of ALS iPSCs into motor neurons. We report an increase in insoluble TDP-43 in motor neuron–containing neural populations differentiated from ALS patient–derived iPSCs. We also observed a punctate cytoplasmic distribution of TDP-43 in ALS iPSC–derived motor neurons and a cellular vulnerability to arsenite treatment, in agreement with the previous report, which showed increased vulnerability of ALS motor neurons to other cellular stressors (13). There were, however, two differences between these studies. In the previous study (13), TDP-43 mRNA expression was similar between control iPSC–derived motor neuron culture and iPSC-derived motor neuron culture from ALS patients with the TDP-43 (M337V) mutation. In contrast, we found that TDP-43 mRNA was up-regulated in ALS iPSC–derived motor neurons with the same TDP-43 M337V mutation or with two other mutations (Q343R and G298S) compared to control iPSC–derived motor neurons. We speculate that there may be several reasons for this difference between the two studies. First, we analyzed multiple iPSC lines to account for clonal variation. We found that in one ALS iPSC line with the TDP-43 M337V mutation, TDP-43 mRNA expression was close to control, whereas the other ALS iPSC lines containing the same mutation showed up-regulation of TDP-43 mRNA. Second, we analyzed the expression of TDP-43 mRNA in purified ALS iPSC–derived motor neurons to avoid contamination with other cell types. Another difference between the two studies was the result of the lactate dehydrogenase cytotoxicity assay (fig. S11). Unlike the previous study, we found no difference in survival between control and ALS motor neurons using this assay. We speculate that this discrepancy stemmed from the fact that we used multiple iPSC lines from three different ALS patients and that this reduced the impact of clonal variation. Further progress in culture methods for motor neurons might make it possible to recapitulate ALS motor neuron death under basal conditions.

In our study, we used multiple iPSC lines from three different ALS patients for our phenotyping assays to address the challenge of clonal variation because we discovered variations even between iPSC lines derived from the same ALS patient. Although we did not use all of the iPSC clones from all ALS patients in all of the assays, we used at least two clones from each ALS patient in each phenotyping assay including immunoblot analysis of insoluble TDP-43, neurite length quantification, and quantitative polymerase chain reaction (qPCR) analysis of TDP-43 mRNA. In the drug screening assay, we used selected ALS iPSC lines and tested a positive hit on phenotyped clones that were not included in the primary screening assay. We demonstrated that the results of drug screening were not specific to the subset of screened clones. Clonal variation of iPSC lines continues to be an obstacle for cellular modeling of disease.

We found that mutant TDP-43 was more insoluble and more prone to forming aggregates than wild-type TDP-43. These properties may contribute to loss of the negative feedback loop that autoregulates synthesis of TDP-43 (8), which could in turn lead to increased expression of TDP-43 mRNA and protein. Increased amounts of TDP-43 protein bound to other RNA binding proteins such as the spliceosomal factor SNRPB2 could lead to perturbation of RNA metabolism. We generated iPSCs from three ALS patients carrying distinct TDP-43 mutations: Q343R, M337V, and G298S. These mutations and other mutations are located in the glycine-rich domain of TDP-43 (28). We speculate that mutations in the glycine-rich domain of TDP-43 may disrupt the interaction of mutant TDP-43 with other proteins or RNAs, leading to up-regulation of TDP-43 mRNA, an increase in mutant TDP-43 protein, and the formation of more aggregates. In our study, mutant TDP-43 was more sensitive to oxidative stress, which increased the amount of insoluble TDP-43 and hence aggregate formation. Anacardic acid may reverse ALS-associated phenotypes potentially by down-regulation of TDP-43 mRNA expression. However, anacardic acid might partially exert its effect through the suppression of other genes that are regulated by NF-κB (29) or through control of redox signaling (30) because pathogenic TDP-43 has been implicated in both pathways (23, 25).

In summary, we identified cellular and molecular phenotypes associated with ALS using ALS patient–specific iPSC–derived motor neurons and set up a screening assay that identified anacardic acid as a drug that could reverse some of the ALS phenotypes. Our data suggest that ALS iPSC–derived motor neurons may be useful for elucidating disease pathogenesis and for identifying new candidate drugs.

Materials and Methods

Derivation of patient-specific fibroblasts

Control and ALS-derived human dermal fibroblasts (HDFs) other than ND32947 were generated from explants of 3-mm dermal biopsies. ND32947 was obtained from Coriell Institute (Camden, NJ). After 1 to 2 weeks, fibroblast outgrowths from the explants were passaged.

iPSC generation

Human complementary DNAs for reprogramming factors were transduced in HDF with retrovirus (Sox2, Klf4, Oct3/4, and/or c-Myc) or episomal vectors (Sox2, Klf4, Oct3/4, L-Myc, Lin28, short hairpin RNA for p53). Several days after transduction, fibroblasts were harvested and replated on an SNL feeder layer. On the next day, the medium was changed to primate embryonic stem cell medium (ReproCELL) supplemented with basic fibroblast growth factor (4 ng/ml) (Wako Chemicals). The medium was changed every other day. Thirty days after transduction, iPSC colonies were picked up.

Immunocytochemistry

Cells were fixed in 4% paraformaldehyde (pH 7.4) for 30 min at room temperature and rinsed with phosphate-buffered saline (PBS). The cells were permeabilized in PBS containing 0.2% Triton X-100 for 10 min at room temperature, followed by rinsing with PBS. Nonspecific binding was blocked with PBS containing 10% donkey serum for 60 min at room temperature. Cells were incubated with primary antibodies overnight at 4°C and then labeled with appropriate fluorescently tagged secondary antibodies. DAPI (4′,6-diamidino-2-phenylindole) (Life Technologies) was used to label nuclei. Fluorescence images were acquired on DeltaVision (Applied Precision). The following primary antibodies were used in this assay: Nanog (R&D Systems, 1:10), SSEA-4 (Millipore, 1:100), SOX-17 (R&D Systems, 1:50), αSMA (DAKO A/S, 1:500), Tuj1 (Covance, 1:2000), Islet-1 (Developmental Studies Hybridoma Bank, 1:50), HB9 (Epitomics, 1:2000), ChAT (Millipore, 1:100), SMI-32 (Covance, 1:500), synapsin (Millipore, 1:500), MAP2 (Millipore, 1:200), GFAP (DAKO, 1:1000), TDP-43 (Proteintech, 1:500), and SNRPB2 (Proteintech, 1:300).

Quantitating the number of TDP-43 aggregates in cytoplasm

Differentiated motor neurons were visualized with anti–SMI-32 antibody. Nuclei were then stained with DAPI. The number of TDP-43 dots in cytoplasm was quantified by IN Cell Analyzer 6000 (GE Healthcare). The TDP-43 dot was defined by a size of more than 0.324 μm.

Immunoprecipitation

After being washed with PBS, cells cultured in 24-well plates were solubilized for 30 min on ice in immunoprecipitation assay buffer (50 mM tris-HCl, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% SDS, and 0.1% sodium deoxycholate) with a protease inhibitor cocktail and a phosphatase inhibitor with Bioruptor (high mode, ON: 30 s, OFF: 60 s, 20 times). Insoluble fraction was collected by centrifugation (15,000 rpm, 3 min), in which the supernatant was defined as soluble fraction, and lysed in SDS lysis solution (50 mM tris-HCl, 10 mM EDTA, 1% SDS) with Bioruptor. Soluble lysates were incubated overnight with protein G–Sepharose bound to anti–TDP-43 antibody (Abnova). Lysates from insoluble fraction were diluted to 10% by dilution buffer (50 mM tris-HCl, 167 mM NaCl, 1.1% Triton X-100, and 0.11% sodium deoxycholate) including phosphatase inhibitor and then incubated overnight by protein G–Sepharose bound to anti–TDP-43 antibody (Abnova). Sepharose beads were collected by brief centrifugation and washed with TBS buffer (50 mM tris-HCl, 150 mM NaCl, pH 7.4) three times. Immunoprecipitated material was eluted by boiling for 3 min in 2× sample buffer [125 mM tris-HCl (pH 6.8), 4% SDS, 20% glycerol, 0.004% bromophenol blue].

Induction of motor neurons by quick embryoid body–like aggregate method (SFEBq)

Human iPSCs were dissociated to single cells and quickly reaggregated in low cell adhesion U-shaped 96-well plates (Lipidure-Coat Plate A-U96, NOF Corporation). Aggregations were cultured in 5% DFK medium [5% KSR Medium (DFK5%), Dulbecco’s modified Eagle’s medium/Ham’s F12 (Sigma-Aldrich), 5% KSR (Invitrogen), minimum essential medium–nonessential amino acids (Invitrogen), l-glutamine (Sigma-Aldrich), 2-mercaptoethanol (Wako)] with 2 μM dorsomorphin and SB431542 in a neural inductive stage (P1) for 12 days. After patterning with neurobasal medium supplemented with B27, 1 μM retinoic acid, Sonic Hedgehog (100 to 500 ng/ml), and fibroblast growth factor 2 (12.5 ng/ml), the aggregates were adhered to Matrigel (BD Biosciences)–coated dishes on day 22. Adhesive embryoid bodies were cultured in neurobasal medium with brain-derived neurotrophic factor (10 ng/ml), glial cell line–derived neurotrophic factor (10 ng/ml), and NT-3 (10 ng/ml) in P2 culture. They were separated from the dish by Accutase, dissociated into a small clump or single cells, and cultured at 500,000 cells per well on Matrigel-coated 24-well dishes on day 35 as P3 maturation stage.

Supplementary Materials

www.sciencetranslationalmedicine.org/cgi/content/full/4/145/145ra104/DC1

Materials and Methods

Fig. S1. ESC marker, genotype, and karyotype analysis.

Fig. S2. Transgene expression.

Fig. S3. In vitro three-germ layer and teratoma assay.

Fig. S4. Bisulfite genomic sequencing of the promoter regions of Nanog and Oct4.

Fig. S5. Comparison of global gene expression profiles of human iPSCs.

Fig. S6. Motor neuron differentiation and functional assay of motor neurons.

Fig. S7. Estimation of neuronal differentiation from control and ALS iPSCs.

Fig. S8. Motor neuron purification for gene expression analysis.

Fig. S9. Decreased neurite length of ALS motor neurons.

Fig. S10. Signaling pathways significantly activated in ALS motor neurons.

Fig. S11. Cell viability and cytotoxicity.

Table S1. List of clones for human iPSC lines.

Table S2. Correlation coefficients of global gene expression in human iPSC lines.

Table S3. GO analysis of microarray data (fold change >1.2) ALS UP.

Table S4. GO analysis of microarray data (fold change >1.2) ALS DOWN.

Table S5. Genes identified as significantly increased (P < 0.05) with >1.2-fold change in motor neurons derived from ALS human iPSCs.

Table S6. Genes identified as significantly decreased (P < 0.05) with >1.2-fold change in motor neurons derived from ALS human iPSCs.

Table S7. Primers list.

References

References and Notes

  1. Acknowledgments: We thank our co-workers and collaborators; S. L. Pfaff for providing plasmids; M. Kawada, Y. Karatsu, and T. Enami for technical assistance; and K. Murai for editing the manuscript. Funding: This research was funded in part by a grant from the Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program) of the Japan Society for the Promotion of Science (S. Yamanaka), a grant from the JST Yamanaka iPS cell special project (S. Yamanaka and H. Inoue), Core Research for Evolutional Science and Technology (H. Inoue), Grant-in-Aid from the Ministry of Health and Labour (R.T. and H. Inoue), a Grant-in-Aid for Scientific Research on Innovative Area “Foundation of Synapse and Neurocircuit Pathology” (22110007) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (H. Inoue), and a research grant from the Novartis Foundation for Gerontological Research (H. Inoue). Author contributions: H. Inoue planned the project; H. Inoue, N.E., and S.K. designed the experiments; S. Yamanaka, H. Inoue, N.E., and S.K. wrote the manuscript; K.Y., S. Yamawaki, M.N., S.S., K.M., K. Okamoto, H.T., A.T., K.H., and R.T. recruited patients; N.E., S.K., K. Takahashi, K.M., F.A., K.Y., T. Kondo, K. Tsukita, K. Okita, I.A., H. Inoue, and S. Yamanaka generated and characterized iPSCs; T. Aoi conducted karyotyping; A.W. performed methylation analysis of iPSCs; Y.Y. performed immunohistochemical analysis of teratomas; N.E., S.K., K. Takahashi, K.M., and F.A. performed generation and characterization of human motor neurons; A.M. and J.T. provided technical support for neuronal differentiation; D.W. performed whole-cell patch clamp; H.H. and T. Kaneko generated lentivirus vectors with high titers; T.Y. supported the microarray analysis; H. Ito and T. Ayaki performed immunohistochemistry of human spinal cords; M.Y. provided spliceostatin A; A.K. provided human materials; T. Nakahata provided NOG mice; T. Nonaka and M.H. provided antibodies, plasmids, and scientific discussions; R.T. provided scientific discussions; M.C.N.M. and F.H.G. provided plasmids and scientific discussions. Competing interests: A.T. is a paid consultant to Chugai Pharmaceutical Co. Ltd. S. Yamanaka is a member of the scientific advisory boards of iPierian, iPS Academia Japan, and Megakaryon Corporation. H. Inoue, N.E., S.K., and K. Tsukita have filed a patent (Prophylactic and therapeutic drug for amyotrophic lateral sclerosis and method of screening, 61/587,323) related to this work. The other authors declare that they have no competing interests.
View Abstract

Navigate This Article