Stem Cell-Derived Neurons as Cellular Models of Sporadic Alzheimer’s Disease

Abstract

Alzheimer’s disease (AD) occurs as either an autosomal dominant inherited disease or sporadically. While familial mutant genes can be expressed in cells or in animal models to assess dysregulated functions, sporadic AD cannot be replicated in models given our lack of understanding of causality. Furthermore, the study of sporadic forms of AD is difficult given the inaccessibility of brain tissues in living individuals and the manifestation of symptoms years after the onset of disease. Here, the objective was to assess if induced pluripotent stem cell-derived neurons from well-ascertained sporadic AD individuals could represent potential cellular models to determine the underlying molecular mechanisms of disease. We used cryopreserved peripheral blood mononuclear cells from three well-ascertained sporadic AD and three non-cognitively impaired (NCI) individuals of the CIMA-Q cohort to obtain iPSC-derived neurons. Microtubule associated protein 2 was decreased in AD neurons, whereas expression of AD-associated amyloid precursor protein, tau, and amyloid-β peptide was similar in AD and NCI individuals. RNA sequencing identified several upregulated and downregulated mRNAs in AD relative to NCI neurons. Of these, complement Factor H (CFH), signal regulatory protein beta1 (SIRPB1), and insulin like growth factor binding protein 5 (IGFBP5) were previously associated with AD. In addition, several transcription factors not previously associated with AD, but involved in neuronal proliferation and differentiation were differentially expressed. The results identify novel avenues for the study of the underlying causes of sporadic AD and support the establishment of additional lines to identify mechanisms of disease in sporadic AD individuals.

Keywords

Alzheimer’s disease cholinergic neurons cryopreserved cell gene expression induced pluripotent stem cells

INTRODUCTION

Induced pluripotent stem cell (iPSC)-derived neurons offer much hope for the identification of biomarkers and the elucidation of underlying molecular mechanisms of neurodegenerative diseases such as Alzheimer’s disease (AD). Pluripotent stem cells are capable of long-term proliferation in culture and can be differentiated into all cell types of the body. Yamanaka and Thomson first reprogrammed human skin fibroblasts into iPSC, by retroviral-directed overexpression of embryonic-related transcription factors Oct4 and Sox2, together with c-Myc and Klf4, or with Lin28 and Nanog [1, 2]. Currently, non-integrating Sendai viral delivery of the reprogramming genes are used, thus avoiding disruption of the genome [3 –7].

Many iPSC lines, generated from familial AD fibroblasts carrying presenilin 1 (PSEN1) gene mutations (L166P, A246E, N141L, H163R, M146L), were successfully differentiated into neurons [8 –13]. These neurons generally have increased Aβ₄₂/Aβ₄₀ amyloid ratios or increased Aβ₄₀ levels and thus mimic one of the expected pathological features of AD. Duplicate amyloid precursor protein (APP) gene (APP^Dupl)-, or mutant APP^V717I-, and APP^V717L-carrying iPSC-derived neurons showed increased Aβ levels, phosphorylated Tau²³¹ and increased glycogen synthase kinase 3β (GSK3β) [8 , 15]. In contrast, iPSC-derived cortical neurons carrying the AD protective APP E693A mutation showed decreased production of Aβ₄₀ and Aβ₄₂ [16]. Production of isogenic iPSC lines carrying AD-related mutations offer an excellent system to directly pinpoint the effect of specific gene mutations [17]. Recently, three dimensional cultures established from mutant PSEN1- [18] or APP-iPSC-derived neurons or lentiviral-transfected normal human neural progenitor cells differentiated into neurons [19] showed increased secreted Aβ, tau protein, and deposition of extracellular Aβ.

Familial AD can offer considerable insight for the underlying molecular mechanism of neurodegeneration, but it is unclear if the sporadic forms of AD, which represent more than 90% of all AD cases, arise from similar etiological problems. Only a few cell lines have been developed from sporadic AD [8 , 20–26]. Of these, 5 lines have been assessed for Aβ production; 3 lines did not display changes in Aβ ratios [8, 16], while one line had increased Aβ₄₀ [8] and one line had increased Aβ₄₂/Aβ₄₀ [26]. One line showed increased phosphorylated tau/total tau [8] and one line showed increased sensitivity to ionomycin [26]. Clearly, cellular models from sporadic forms of AD require more validation and characterization. If informative, sporadic AD iPSC-derived neurons could help determine if there are unique degenerative pathways specific to each sporadic AD individual that leads to common pathologies and symptoms or whether there are shared degenerative features in all individuals with sporadic AD. If unique paths exist, the iPSC-derived neurons from sporadic AD patients could serve to personalize the treatment of sporadic AD, thereby increasing chances of successfully preventing or delaying the disease. If common degenerative pathways are discovered, these could become therapeutic targets to identify novel therapies against AD.

The establishment of sporadic AD iPSC should be done from individuals that have been ascertained as normal or AD by clinical, neuropsychological, neuroimaging, and biochemical assessments. CIMA-Q, the “Consortium pour l’identification précoce de la maladie Alzheimer au Québec (Quebec Consortium for the early identification of Alzheimer Disease)” is currently studying a cohort of approximately 300 individuals older than 65 years of age with elaborate clinical and cognitive/behavioral evaluations, and some neuroimaging evaluations. Recently, re-programming of peripheral blood mononuclear cells (PBMNC), which can be more easily obtained than fibroblasts and limit risk of infection, facilitate the process, especially in elderly individuals [7 , 27–31]. Each of the CIMA-Q participants except one agreed to provide PBMNC for cryopreservation as a source for iPSC lines. Here, PBMNC from three confirmed sporadic AD and from three age-matched non-cognitively impaired (NCI) individuals were reprogrammed into iPSC lines and differentiated into neurons. Neurons were assessed for preservation of genomic integrity and for differential gene expression.

It is the hope that the iPSC lines developed from these individuals will allow identification of underlying molecular mechanisms of disease, possibly revealing unique pathological pathways in different subjects and thus allow the development of personalized treatments.

MATERIAL AND METHODS

CIMA-Q cohort and ethical approval

Human age- and sex-matched non-cognitively impaired and confirmed sporadic AD PBMNC were obtained from the Quebec Consortium for the early Identification of Alzheimer Disease (CIMA-Q) cohort individuals. CIMA-Q is studying a cohort of 290 individuals mainly between 65–93 years of age with elaborate clinical and cognitive evaluations with some neuroimaging. This study has been approved by the Institut universitaire de gériatrie de Montréal Institutional Review Board of the committee.

Isolation and cryopreservation of PBMNC

Eight ml of blood have been collected into a CPT Cell Preparation Vacutainer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). These tubes contain a sodium heparin anticoagulant, a density fluid (FICOLL Hypaque) and a polyester gel barrier to permit the mononuclear cells separation in the tubes during a single centrifugation step [41]. The number of viable PBMNC isolated using CPT tubes varied between 1×10⁶ and 4×10⁶ cells /ml of whole blood (average of 2.6×10⁶ cells/ml of blood) and viability was between 85% and 99% (average of 97%). Thawed cryopreserved (in liquid nitrogen and –135°C) PBMNC have between 20% to 95% (average of 50%) viability.

Reprogramming PBMNC using the Sendai virus into iPSC

PBMNC-derived iPSCs were generated by Dr. Jack Puymirat’s (Université Laval, Québec) stem cell platform (iPSC Québec platform). The cryopreserved PBMNC provided the 0.25–0.5×10⁶ PBMNC required to obtain iPSCs using the CytoTune – iPS 2.0 Sendai Reprogramming Kit (Life Technologies, Burlington, ON, Canada) containing Sendai virus vectors encoding the four reprogramming factors published by Takahashi et al. (Oct-4, Sox2, Klf4, and cMyc) [1 , 43]. The Sendai virus in this kit has temperature sensitivity mutations, which inhibits the virus proliferation in cell culture conditions and the virus is cleared after 10–13 passages [42]. The iPSC colonies were cultured separately and analyzed for 1) loss of Sendai virus by PCR; 2) expression of pluripotency and stemness markers, by immunocytochemistry or PCR; 3) pluripotency by formation of embryoid bodies in vitro followed by RT-PCR analysis; 4) karyotype stability by cytogenetic analysis, and genotyping by Short Tandem Repeat (STR) analysis [44]. iPSCs were expanded on matrigel (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) in serum-free defined feeder-free culture media (mTeSR1 medium; StemCell Technologies, Vancouver, BC, Canada) [45]. The 0.5 mM EDTA (Life Technologies, Burlington, ON, Canada) or 1 mg/ml dispase (Life Technologies, Burlington, ON, Canada) splitting method were used to passage the cells [46] At each passage, a fraction of iPSC from each clone will be cryopreserved in liquid nitrogen, using CryoStor CS10 medium (StemCell Technologies, Vancouver, BC, Canada), an animal component-free, defined cryopreservation medium, recommended for extremely sensitive cell types. iPSC cell lines are made available upon request and approval to the CIMA-Q user access committee (http://www.cima-q.ca/).

Establishing NSC lines and differentiation into cholinergic neurons

Neural stem cells (NSC) and differentiated neurons were prepared from iPSC by Dr. Thomas Durcan’s and his stem cell platform (Montreal Neurological Institute and McGill University, Montreal, Quebec). Methods for neural differentiation of human embryonic stem cells are well established and show to be appropriate for iPSCs [47]. Synergistic action of two inhibitors of SMAD signaling— Noggin and SB431542— is sufficient to induce rapid and complete conversion of >80% of iPSCs into NSCs [48]. iPSC-derived NSCs form rosette-like structures, which can be mechanically isolated to obtain pure cultures of NSCs [49]. NSCs were expanded in response to fibroblast growth factor (FGF)-2 and epithelial growth factor (EGF), while maintaining the potential to differentiate into functional neurons and glia [50, 51]. iPSCs were cultured in suspension, in STEMdiff Neural Induction Medium (Stemcell Technologies, Vancouver, BC, Canada). Floating embryoid bodies were plated on matrigel, neural rosettes were isolated and dissociated with STEMdiff Neural Rosette Selection Reagent (Stemcell Technologies, Vancouver, BC, Canada), and NSC expanded in StemPro NSC Serum Free Medium (Life Technologies, Burlington, ON, Canada), passaged using StemPro Accutase (Life Technologies, Burlington, ON, Canada), and cryopreserved in CryoStor CS10 medium (StemCell Technologies, Vancouver, BC, Canada), without losing their proliferation and differentiation capacities. NSC formed aggregates in StemPro NSC Serum Free Medium after 5–7 days. These aggregates were plated on Poly-L-ornithine/Laminin coated dishes in Neural medium (Neurobasal medium, 2% serum free B-27 supplement, and 2 mM L-glutamine; from Life Technologies, Burlington, ON, Canada). Cholinergic neurons were prepared according to Bissonette [52]. From day 0 to day 2, cells were cultured in neural medium with 200 ng/ml SHH (Sonic Hedgehog recombinant human; R&D Systems, Minneapolis, MN, USA) and 100 ng/ml FGF-8 (Fibroblast Growth Factor 8b recombinant human/mouse, R&D Systems, Minneapolis, MN, USA). At day 2, cells are treated 1 h with 2μg/ml Mitomycin C and 10 ng/ml BMP-9 (Bone morphogenic protein-9 recombinant human; R&D Systems, Minneapolis, MN, USA) was added to the medium. At day 4, the media was changed to contain BMP-9 and 100 ng/ml NGF (Nerve Growth Factor recombinant human; R&D Systems, Minneapolis, MN, USA). From day 6 to day 9, the NGF-containing neural medium was changed.

RNA extraction

Total RNA was extracted using Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. Human AD and normal cortex tissues were provided by Dr. Bradley Hyman. After DNAse treatment with the RNase-free DNase set (Qiagen, Valencia, CA, USA), RNA was purified using RNeasy MinElute Cleanup Kit (Qiagen, Valencia, CA, USA). The concentrations of RNA samples were measured using the DS-11 FX+Spectrophotometer/Fluorometer (DeNovix, Wilmington, DE, USA).

PCR and qPCR

cDNA was prepared using avian myeloblastosis reverse transcriptase (AMV-RT) (Roche, Mannheim, Germany). PCR was performed with GoTaq DNA polymerase (Promega, Madison, WI, USA) or Taq DNA polymerase (New England Biolabs, Whitby, ON, Canada) following the manufacturer’s protocol. The products were visualized on Redsafe (FroggaBio, Toronto, ON, Canada) stained 2% or 1.5% agarose gels. The PCR bands on gels were analyzed using Image J software (NIH, Bethesda, MD, USA). Real-time PCR experiments were performed using SYBR Green Taq Mastermix (Quanta BioSciences, Gaithersburg, MD, USA) on an Applied Biosystems 7500 Fast Real-Time PCR system machine (Applied Biosystems, Foster City, CA, USA). Results are presented using 2-ΔΔCt with GAPDH as internal control [53]. Primers used for the gene amplification are GCAGTTCTCAAAGGCTAGAC and TTGATCGTGGAACTCCAT CT for MAP2, ATGGAGCCTCTCCGGCTGCT and TCACGTGTCTCTCAGCCCTG for TREM2, ACTGGGTGTCTGAGTACGG and TTGGAAGCCATTTTGACTAT for CHAT, GTGGGCA GGTGGGAGCTTGATTCT and CTGGGGCGGCCTGGTATGACA for GFAP, ACCACAGTCCATGCC and TCCACCACCCTGTTG for GAPDH, GGTGGTTCGAGTTCCT ACAA and CCTCTCTTTGGCTTTCTGGA for APP₆₉₅, CAGCGCCATTCCTACAACAG and CCTCTCTTTGGCTTTCTGGA for APP₇₅₁, CCCGAGATCCTGTTAAACTTC and CCTCT CTTTGGCTTTCTGG for APP₇₇₀, AACCAGTGACCATCCAGAAC and ACTTGTCAGGAAC GAGAAGG for total APP, CTCCAAAATCAGGGGATCGC and CCTTGCTCAGGTCAACTG GT for Tau.

RNA-sequencing

Total RNA extracted from neuronal cultures (CQ1-3, CQ2-4, CQ3-11, CQ4-11, CQ5-1, and CQ6-4 done in duplicates) was used for RNA sequencing analysis. Each sample had an A260/A280 ratio≥1.95 and an RNA Integrity Number≥8.5. An mRNA stranded library preparation (NEB) was used to obtain cDNA libraries from total RNA. Libraries were sequenced on an Illumina HiSeq 2500 PE 125bp (4 samples per lane). This work was done by Genome Quebec Innovation Center and analysis performed by Canadian Centre for Computational Genomics (C3G) - Montréal Node.

Total RNA was quantified using a NanoDrop Spectrophotometer ND-1000 (NanoDrop Technologies, Inc.) and its integrity was assessed on a 2100 Bioanalyzer (Agilent Technologies). Libraries were generated from 250 ng of total RNA as following: mRNA enrichment was performed using the NEBNext Poly(A) Magnetic Isolation Module (New England BioLabs). cDNA synthesis was achieved with the NEBNext RNA First Strand Synthesis and NEBNext Ultra Directional RNA Second Strand Synthesis Modules (New England BioLabs). The remaining steps of library preparation were done using and the NEBNext Ultra II DNA Library Prep Kit for Illumina (New England BioLabs). Adapters and PCR primers were purchased from New England BioLabs. Libraries were quantified using the Kapa Illumina GA with Revised Primers-SYBR Fast Universal kit (Kapa Biosystems). Average size fragment was determined using a LabChip GX (PerkinElmer) instrument.

The libraries were normalized at 2nM, denatured in 0.05 N NaOH and then were diluted to 8pM using HT1 buffer. The clustering was done on an Illumina cBot and the flowcell was run on a HiSeq 2500 for 2×125 cycles following the manufacturer’s instructions. A phiX library was used as a control and mixed with libraries at 1% level. The Illumina control software was HCS 2.2.58, the real-time analysis program was RTA v. 1.18.63. Program bcl2fastq v1.8.4 was then used to demultiplex samples and generate fastq reads.

Raw reads derived from the sequencing instrument are clipped for adapter sequence, trimmed for minimum quality (Q30) in 3’ and filtered for minimum length of 32 bp using Trimmomatic [54]. Surviving read pairs were aligned to the Homo sapiens assembly GRCh37 by the ultrafast universal RNA-seq aligner STAR [55] using the recommended two passes approach. Aligned RNA-Seq reads were assembled into transcripts and their relative abundance was estimated using Cufflinks [56] and Cuffdiff [57].

Exploratory analysis was conducted using various functions and packages from R and the Bioconductor project [58]. Differential expression was conducted using both EdgeR [59] and DEseq [60]. Terms from the Gene Ontology were tested for enrichment with the GOseq [61] R package.

Whole exome sequencing

Genomic DNA from parental PBMNC and derived-neuronal cultures was prepared using the DNeasy Blood and Tissue kit (Qiagen, Valencia, CA, USA) following the manufacturer’s protocol. DNA concentration was measured using the DS-11 FX+Spectrophotometer / Fluorometer (DeNovix) and DNA integrity was assessed on 1% agarose gel. Libraries were prepared using Roche Nimblegen SeqCap EZ Human Exome capture and used for sequencing on an Illumina HiSeq4000 PE100 sequencer (6 samples per lane). This work was done by Genome Quebec Innovation Center and analysis performed by Canadian Centre for Computational Genomics (C3G) - Montréal Node. Genomic DNA was quantified using the Quant-iT™ PicoGreen^® dsDNA Assay Kit (Life Technologies). Libraries were generated on a BioMek robot (Beckman) using the KAPA HTP Library Preparation Kit (Kapa Biosystems) as per the manufacturer’s recommendations. TruSeq adapters and PCR primers were purchased from IDT. Libraries were quantified using the Kapa Illumina GA with Revised Primers-SYBR Fast Universal kit (Kapa Biosystems). Average size fragment was determined using a LabChip GX (PerkinElmer) instrument. 166 ng of 6 libraries were pooled together (total of 1000 ng per capture) prior to proceeding with the enrichment of the targeted regions using the SeqCap EZ Exome v3 baits. Captures were performed robotically according the manufacturer’s recommendations. Final libraries were quantified using the Kapa Illumina GA with Revised Primers-SYBR Fast Universal kit (Kapa Biosystems). Average size fragment was determined using a LabChip GX (PerkinElmer) instrument.

The libraries were normalized, denatured in 0.05N NaOH and then were diluted to 200 pM and neutralized using HT1 buffer. ExAMP was added to the mix and the clustering was done on an Illumina cBot and the flowcell was run on a HiSeq 4000 for 2×100 cycles (paired-end mode) following the manufacturer’s instructions. A phiX library was used as a control and mixed with libraries at 1% level. The Illumina control software was HCS HD 3.4.0.38, the real-time analysis program was RTA v. 2.7.7. Program bcl2fastq v2.18 was then used to demultiplex samples and generate fastq reads. Reads were trimmed from the 3’ end to have a phred score of at least 30. Illumina TruSeq adapters were removed from the reads. Resulting reads shorter than 50 bp and orphaned reads were discarded. Trimming and clipping were performed using Trimmomatic v0.35. The trimmed reads were aligned to the human genome build GRCh37 using a fast, memory-efficient Burrows-Wheeler transform (BWT) aligner BWA-mem. Mapped reads were further refined using GATK and Picard program suites (Picard program suite: http://broadinstitute.github.io/picard). Specifically, GATK indel realigner was used to improve mapping near insertions and deletions, Picard mark duplicates removed duplicate reads with same paired start site and GATK base recalibration improved quality scores. Variants were called using GATK haplotype caller in gvcf mode to allow efficient downstream merging of multiple samples into one variant file to streamline downstream variant processing procedures which included functional annotation with SNPeff [62] and variant annotations with Gemini.

Detection of amyloid (Aβ) peptide levels by ELISA

Secreted Aβ_38,40,42 peptides were measured from 72 h conditioned media with the Multi-Spot^® Aβ peptide panel 1 (6E10) kit (Meso Scale Discovery, Gaithersburg, MD). Aβ_42,40,38 will be expressed relative to total Aβ. Each sample was analyzed in duplicates and then averaged for a mean value. Cellular protein was extracted and measured by BCA protein assay to control for input cell number.

Characterization of neuronal culture by immunofluorescence

After washing cells plated on coverslips with PBS, cells were fixed in 4% paraformaldehyde (SigmaAldrich, Oakville, ON, Canada) – 4% sucrose (BioRad, Mississauga, ON, Canada) for 20 min at room temperature (RT). After three washes in PBS, cells were blocked with 10% normal goat serum (Sigma, Oakville, ON, Canada) and 0.25% TritonX-100 (BioShop, Burlington, ON, Canada) in PBS (blocking solution) for 1 h at RT. Incubation of mouse anti-β111-tubulin TUJI, and rabbit anti-MAP2 (all from Abcam) diluted in blocking solution was done at 4°C overnight in a humid chamber. Following three PBS washes, Alexa conjugated secondary antibodies (Abcam, Toronto, ON, Canada) in blocking solution were added for 2 h at RT in the dark. After PBS washes, nuclei were counter-stained with DAPI (ref) for 10 min at RT in the dark. After PBS wash, coverslips were mounted with Fluorescent Mounting Medium (Dako, Burlington, ON, Canada). Pictures were taken using the Nikon TE2000U fluorescent microscope and the OpenLab software (PerkinElmer, Woodbridge, ON, Canada) and quantified using ImageJ software (NIH, Bethesda, MD, USA).

Neuroimaging

For morphometric measures, preprocessing and segmentation were performed by FreeSurfer ([http://freesurfer.net)/] http://freesurfer.net) image analysis suite (5.3), using recon -all with default parameters. The technical details of these procedures are described in prior publications [32, 33] We used Desikan-Killiany-Tourville atlas (aparc.DKTatlas40 files) for cortical measures and default atlas for subcortical measures (aseg.stats file). All morphometric measures were adjusted for age, sex, estimated total intracranial volume (eTIV), scanner manufacturer, and magnetic field strength through normative Z scores [34, 35]. SNIPE score were provided by a patented True Positive Medical Device (TPMD) technique in which images preprocessing included denoising based on an optimized nonlocal means filter [36]; correction of inhomogenities using N3 [37]; registration to stereotaxic space using ICBM152 template linear transform (1 × 1 × 1 mm³ voxel size) using a template derived from the Alzheimer Disease Neuroimaging Initiative (ADNI) study database [38]; linear intensity normalization of each subject on template intensity; and brain extraction using BEaST [39]. Hippocampal (HC) and entorhinal (EC) SNIPE scores [39, 40] are grading by estimating the nonlocal similarity of the subject’s image to training samples of healthy aging subjects and participants with AD. For each voxel in a new subject to be analyzed, the method defines a 7 × 7 × 7 voxel patch centered on the voxel. The procedure then searches the template library for similar patches. Template region labels are weighted by the patch similarity, and the region label with the maximum weight is then associated with the voxel. At the same time, the template group (1 for healthy controls and 2 for AD subjects) is also weighted by the patch similarity. The resulting average weight is used as a grading value to indicate how similar this voxel is to one or the other group. The SNIPE score is the average of the voxels within a region (e.g., HC or EC). Normative SNIPE Z scores adjusted for age and sex were computed from the healthy participants of the AD Neuroimaging Initiative 1 (ADNI-1).

RESULTS

Establishment and characterization of iPSC derived from non-AD or sporadic AD PBMNC

PBMNC were cryopreserved from 290 individuals in the Consortium for the early identification of Alzheimer Disease in Quebec (CIMA-Q) cohort. Three non-cognitively impaired individuals (NCI) and three sex- and age-matched clinically confirmed AD (Table 1) were selected to establish iPSC cell lines from the PBMNC. Only one of the three NCI subjects displayed some medical problems, whereas AD subjects all showed hypertension, and two AD subjects had cancer. NCI subjects scored 28 or 29, 0, and 25 or 26, whereas AD subjects scored below 18, 0.5, and below 23, on the Montreal Cognitive Assessment (MoCA), clinical dementia rating-global test (CDR), or telephone Mini-Mental State Examination (MMSE), respectively. Magnetic resonance imaging (MRI) indicated that the entorhinal cortex and hippocampi regions of interest had morphometric (Fig. 1A) and SNIPE (Fig. 1B) normative Z scores close to 0 in healthy controls, as expected for their respective age, sex, and estimated total intracranial volume. In contrast, participants with a clinical AD diagnosis showed Z scores markedly below what was expected for their age, sex, and estimated total intracranial volume. In aggregate, these clinical, cognitive and neuroimaging markers solidified our probability estimate that these individuals had AD.

Table 1

Demographics of the 3 NCI and 3 AD subjects used to establish the iPSC lines

					Cognitive assessment			Clinical assessment		Neuroimaging		Biochemistry
Code	Diagnosis	Sex	Age PBMNC	Age onset	MoCA	CDR	T-MMSE	Other CNS disease	Medical History	MRI	PET	LP
CQ1	NCI	F	74	–	28	0	26	NO	No significant health problems	YES	NO	YES
CQ4	NCI	M	71	–	28	0	26	NO	TIA; hypertension; sleep apnea; dyslipidemia; MGUS; asthma; vitamin B12 deficit; lumbar discal hernia	YES	NO	YES
CQ5	NCI	M	89	–	29	0	25	NO	No significant health problems	YES	NO	YES
Mean	–	1F/2M	78.0	–	28.3	0.0	25.7	–	–	–	–	–
SD	–	–	9.6	–	0.6	0.0	0.6	–	–	–	–
CQ2	AD	F	78	72	18	0.5	23	NO	Breast cancer; hypertension	YES	NO	NO
CQ3	AD	M	74	73	7	0.5	18	NO	Hypertension	YES	NO	NO
CQ6	AD	M	78	74	16	0.5	19	NO	Prostate cancer; hypertension; dyslipidemia	YES	YES	YES
Mean	–	1F/2M	76.7	73.0	13.7	0.5	20.0	–	–	–	–	–
SD	–	–	2.3	1.0	5.8	0.0	2.6	–	–	–	–	–

NCI, non-cognitively impaired; AD, Alzheimer disease; F, female; M, male; MoCA, Montreal cognitive assessment; CDR-GS, clinical dementia rating-global score; T-MMSE, telephone Mini–Mental State Examination; TIA, transient ischemia attack; MGUS, monoclonal gammopathy of undetermined significance; MRI, magnetic resonance imaging; PET, positron emission tomography; LP, lumbar puncture; SD, standard deviation.

Fig.1

Neuroimaging of NCI and AD individuals. Morphometric (A) and SNIPE (B) scores for entorhinal cortex and hippocampi from NCI and AD individuals.

The PBMNC were reprogrammed using Sendai viral expression of Oct4, Sox2, Klf4, and c-Myc genes (Fig. 2A). Many iPSC clones were obtained but only two clones were characterized per subject. The reprogramming factors, c-myc, klf4, Sox2, Oct3/4, and Sox2 (KOS), and SeV viral vector (Fig. 2A) were detected by RT-PCR in SeV-transfected PBMNC (CQ-PBMNC) and significantly reduced or lost after the establishment of iPSC CQ1-3,4 and CQ2-2,4 cell lines (Fig. 2B). The loss of SeV was also obtained in CQ5-01, CQ5-02, CQ6-04, and CQ6-06 cell lines (Supplementary Figure 1). The weak expression of SeV virus in CQ-1-4 and CQ2-2 was lost after more cell passages. Pluripotency of the iPSC clones was confirmed by RT-PCR detection of Oct4, Nanog, hTERT and Rex1 mRNAs (Fig. 2C) and the ability of iPSC to form embryoid bodies (EBs) in vitro. All iPSC lines differentiated into the three germ layers as shown by gene expression of Pax6, β-tubulin or NCAM ectoderm markers, MSx1, Flk1 or PECAM mesoderm markers, and Sox17, AFP or Gata4 endoderm markers (Supplementary Table 1). All iPSC clones showed a normal karyotype by G-banding analyses (Fig. 2D), except the CQ5-2 clone. This clone presented some cells with non-clonal chromosome aberrations and was replaced with the clone CQ5-3. Analysis of 15 Short Tandem Repeats (STR) and Amelogenin for sex determination indicated that iPSC clones were identical to their parental PBMNC (Supplementary Table 2). Together, these results show that the PBMNC iPSC lines from AD subjects behave as those derived from NCI elderly subjects. The results also indicate that each iPSC line established as models for disease should be verified for chromosomal abnormalities that have occurred either artificially from the reprogramming of PBMNC or from a rare abnormal PBMNC.

Fig.2

Establishment and characterization of iPSC derived from NCI or AD individuals PBMNC. A) Schematic representation of reprogramming of PBMNC into iPSC. B) RT-PCR of reprogramming factors (cMyc, Klf4, and KOS (hKlf4, hOct3/4, hSox2)) and SeV viral vector expression. Myoblasts were used as negative control and CQ1 and CQ2 PBMNC as positive control. C) RT-PCR of pluripotency markers hTERT, Nanog, Oct4, and Rex1. Myoblasts were used as negative control and CQ1-Sev Trans/CQ2 SeV Trans, non-selected pooled iPSC, as positive control. D) Representative picture of G-banding analysis showing a normal karyotype (46, XY) for iPSC CQ3-10 (passage 8).

Generation and characterization of iPSC-derived neuronal cultures

Neural stem cells (NSC) were established from each iPSC clone and differentiated into neuronal cultures for 9 days (Fig. 3A). iPSC-derived neuronal cells were variably immunopositive for microtubule associated protein 2 (MAP2) and for βIII-tubulin (Fig. 3B-D). The number of MAP2 immunopositive neurons was significantly lower in iPSC-derived neurons from the AD subjects compared to NCI but no difference in βIII-tubulin was observed (Fig. 3C). To further characterize neuronal cultures, RT-PCR to detect MAP2 and choline acetyltransferase (ChAT) mRNAs was performed (Fig. 3E-G). MAP2 mRNAs were present in all iPSC-derived neuronal cultures (Fig. 3E). ChAT mRNA levels were more variable and not very reproducible even between duplicates of the same line, suggesting some differential regulation between iPSC clones. Nevertheless, no statistically significant difference was observed in MAP2 and ChAT mRNA levels between NCI and AD neurons (Fig. 3F, G). This was further supported by RNA Seq data which does not show a significant difference in MAP2 (57124 NCI; 58653 AD) or ChAT (14.1 NCI; 14.5 AD) mRNA levels. Astrocyte glial fibrillary acidic protein (GFAP) and microglia triggering receptor expressed on myeloid cells 2 (TREM2) mRNAs were rare or absent. These results indicate that 9-day old iPSC-derived neuronal cultures are composed mainly of neurons with a few astrocytes and no microglia.

Fig.3

Generation and characterization of iPSC-derived neuronal cultures. A) Schematic representation of neuronal differentiation from iPSC. NSC: neural stem cell. B) Representative pictures of neuronal cultures immunostained for neuronal markers MAP2 and TUJ1 and DAPI nuclear DNA stain. Percent of MAP2 (C) and TUJ1 (D) positive cells in neuronal cultures quantified using ImageJ software. E) RT-PCR of neuronal marker MAP2, cholinergic neurons marker ChAT, astrocyte marker GFAP and microglia marker TREM2 in iPSC-derived neurons. GAPDH was used as a reference gene. Human primary neuron, astrocyte and microglia cultures were used as controls. F, G) Amplicons were quantified by ImageJ software.

Genome stability of iPSC-derived neurons relative to their parental PBMNC

Whole exome sequencing (WES) revealed high number of reads (more than 100 million reads for each sample) with over 99% average alignment rate to the GRCh37 reference genome in NCI samples and AD cells) (Table 2). The average coverage depth is 73-fold and 68-fold for NCI and AD, respectively. In addition, 90% of each genome was sequenced with at least a 50-fold degree of redundancy.

Table 2

Genome stability of iPSC-derived neurons relative to their parental PBMNC

Alignment reads
Sample			Surviving Reads #	Mapped Reads	Mapped %	Mean coverage
NCI	CQ1-4	PBMNC	133066350	132360338	99.4694	80.03
		Neurons	130429706	129748097	99.4774	79.30
	CQ4-11	PBMNC	122140772	121606263	99.5624	74.60
		Neurons	102520012	101956396	99.4502	63.10
	CQ5-1	PBMNC	123376626	122779862	99.5163	73.17
		Neurons	107172488	106568441	99.4364	67.05
	Mean				99.4853	72.88
	StDev				0.0465	6.71
AD	CQ2-2	PBMNC	104475570	104088639	99.6296	65.63
		Neurons	103038538	102470766	99.449	63.31
	CQ3-11	PBMNC	104267108	103751027	99.505	63.44
		Neurons	119882312	119056020	99.3107	71.80
	CQ6-6	PBMNC	107850112	107262866	99.4555	67.30
		Neurons	122643346	122084904	99.5447	74.70
	Mean				99.4824	67.70
	StDev				0.1071	4.64
Shared and unique SNPs
Sample			Shared SNPs	Unique SNPs	% unique SNPs
NCI	CQ1-4	PBMNC	11736	258	2.15
		Neurons	11737	315	2.61
	CQ4-11	PBMNC	11569	296	2.49
		Neurons	11518	283	2.40
	CQ5-1	PBMNC	11682	268	2.24
		Neurons	11649	315	2.63
AD	CQ2-2	PBMNC	11764	246	2.05
		Neurons	11771	245	2.04
	CQ3-11	PBMNC	11708	222	1.86
		Neurons	11829	369	3.03
	CQ6-6	PBMNC	11557	228	1.93
		Neurons	11572	335	2.81

When compared to GRCh37, approximately 12,000 single-nucleotide polymorphisms (SNPs) were identified in each sample (Table 2). Among SNPs, 97.41% were shared between parental PBMNC and iPSC-derived neurons (97.45% for NCI and 97.37% for AD). Approximately 2% of SNPs were unique for either PBMNC or for neurons in NCI and AD.

To investigate the functional relevance of SNPs, we screened neurons for known AD-associated variants [63, 64]. No mutations in familial AD-associated APP, PSEN1, and PSEN2 genes were found in AD samples. The apolipoprotein E (APOE) genotype was E3/E3 for CQ1 to CQ5 and E4/ E4 for CQ6 (rs429358, T for CQ1 to 5 and C for CQ6), similar to exome sequencing results. Together these results indicate that AD neuronal cell lines do not present more instability than NCI neuronal cell lines and that the sporadic AD lines lack the genomic mutations associated with familial AD.

APP and tau mRNA levels in iPSC-derived neurons from normal and sporadic AD subjects

No statistically significant difference was observed in total APP or alternatively spliced averaged APP₆₉₅, APP₇₅₁, and APP₇₇₀ mRNA levels between NCI and AD iPSC-derived neuronal cells by Real Time PCR, although one of the AD lines had higher levels of total APP, APP₆₉₅, and APP₇₇₀ mRNA levels compared to all NCI and other AD lines (Fig. 4A). Tau 3R, but not Tau 4R, mRNA was present in the neuronal cultures after 9 days of differentiation (Fig. 4B) and did not differ between NCI and AD neurons (Fig. 4C). To address the potential of AD iPSCs-derived neuronal cells in reflecting neuropathological features found in sporadic AD patients, we analyzed neuronal Aβ₃₈, Aβ₄₀, and Aβ₄₂ secretion by ELISA in 72 h conditioned media from 9-day-old differentiated neural cell lines. The Aβ₄₂/Aβ_total ratio in sporadic AD-iPSC-derived neurons was similar to that in NCI neurons (Fig. 4D). Secreted Aβ₄₀ and Aβ₄₂ levels did not differ significantly between NCI and AD but tended to be lower in AD neuronal cells (Fig. 4E, F). The levels of intracellular Aβ were below the limit of detection.

Fig.4

APP and Tau metabolism in iPSC-derived neurons from NCI and sporadic AD subjects. A) Relative mRNA levels of total APP, APP₆₉₅, APP₇₅₁, and APP₇₇₀ isoforms measured by Real-Time PCR. Data represent mean and SEM from 3 NCI and 3 AD iPSC-derived neuronal cultures. Statistical difference was assessed by an unpaired t-test. B) RedSafe agarose-stained gel of Tau and GAPDH RT-PCR amplicons from iPSC-derived neuronal cells, human AD and normal cortex tissues as a positive control for 3R and 4R tau mRNA isoforms. C) Tau bands intensities were quantified by ImageJ software. D) Ratios of secreted Aβ₄₂ to total Aβ in conditioned medium from NCI and AD neuronal cell lines. Levels of secreted Aβ₄₀ (E) and Aβ₄₂ (F) in conditioned medium from NCI and AD neuronal cell lines.

Gene expression profiling identified differentially expressed genes in AD iPSC-derived neurons

Gene expression profiling by RNA sequencing of NCI and AD iPSC-derived neurons generated more than 110,000,000 reads in each sample and the mapping rate on the reference genome Homo sapiens assembly GRCh37 was over 95% (Supplementary Table 3). We analyzed the expression profile of different cellular markers to define the cellular composition of the cultures (Supplementary Table 4). Each cell line expressed neuron-specific mRNAs such as microtubule associated protein tau, microtubule associated protein 2, RNA binding Fox-1 homologue 3 (NeuN), synaptophysin, and PSD95. Neuronal subtype-specific mRNA expression for glutamatergic (vesicular glutamate transporter 1 and 2, glutamate ionotropic receptor NMDA type subunit 1 and 2B), dopaminergic (LMX1B, NR4A2 and tyrosine hydroxylase), and cholinergic neurons (acetylcholine esterase, SLC18A3 and CHAT) were found in the cell lines. The number of mRNA copies for those genes with over 100 copies varied between 10–15% between the two clones from one iPSc cell line but was much more variable between different iPSC lines. However, no significant difference was observed between iPSc-derived neurons from NCI and AD subjects.

No significant difference in glial mRNAs were detected between NCI and AD neuronal cultures. Astrocyte-specific glial high affinity glutamate transporter member 2 and 3 mRNAs encoded by the SLC1A2 and SLC1A3 genes, respectively, were present in relatively high copy numbers in the cultures, but astrocyte aldehyde dehydrogenase 1 family member L1 and glial fibrillary acidic protein mRNAs encoded by the ALDH1L1 and GFAP genes, respectively, were expressed at less than 60 copies. Only a few copies of microglial-specific allograft inflammatory factor 1 and scavenger receptor class D, member 1 mRNAs encoded by the AIF1 and CD68 genes, respectively, were detected.

DeSeq and EdgeR differential mRNA expression analyses between the 3 AD and the 3 NCI neuronal cell lines revealed 7 (2 upregulated and 5 downregulated in AD samples) and 258 (95 upregulated and 163 downregulated) differentially expressed genes, respectively (Supplementary Table 5). The top 10 up- and downregulated most significant differentially expressed genes, from the less stringent EdgeR method, are listed in Table 3 and the majority are protein coding. The signal regulatory protein beta1 encoded by SIRPB1 and the cytochrome P450 family 4 subfamily F member 29, pseudogene encoded by CYP4F29P, previously detected to be expressed in brain and other tissues by RNA Sequencing, were the top upregulated mRNAs in AD compared to NCI samples. Both these mRNAs were also significantly different between NCI and AD using the more stringent DeSeq method. The next top 8 genes were only significantly different with the EdgeR method. Four mRNAs were significantly downregulated in AD samples by both the DeSeq and EdgeR method: complement factor H encoded by CFH, insulin like growth factor binding protein 5 encoded by IGFBP5, RP11-726G1,1, and collagen Type IV alpha5 chain encoded by COL4A5 genes. Among the 171 AD-related genes described in Alzheimer’s disease KEGG pathway (http://www.genome.jp/dbget-bin/www_bget?pathway+hsa05010) only 2 mRNAs were significantly downregulated in AD neurons (Supplementary Table 5): interleukin -1β (logFC = –4.95; p = 0.01) and tumor necrosis factor receptor superfamily, member 6 also known as FAS (logFC = –1.94; p = 0.04).

Table 3

Differentially expressed genes between AD and NCI neuronal cell lines

Gene symbol	Fold Change (log FC)	p (DeSeq)	p (EdgeR)	Locus	Biotype	Evidence in AD
Based on p value
Top 10 upregulated genes (out of 95 genes) EdgeR or DeSeq p < 0.05
SIRPB1	3.541	0.024	1.00E-05	20p13	Protein coding
CYP4F29P	8.879	0.004	8.00E-05	21q11.2	Pseudogene
FGF3	5.908	0.366	9.50E-05	11q13.3	Protein coding
RP11-700N1,1	6.292	0.112	0.00036	4q27	lincRNA
MTRNR2L8	2.279	0.342	0.00048	11p15.4	Protein coding
GAD2	3.850	1.000	0.00048	10p12.1	Protein coding
LHX8	8.043	1.000	0.00049	1p31.1	Protein coding
SIM2	2.866	0.072	0.001	21q22.13	Protein coding
SIX3	3.902	0.846	0.001	2p21	Protein coding
LGR5	4.026	0.667	0.001	12q21.1	Protein coding
Top 10 downregulated genes (out of 163 genes) EdgeR or DeSeqp < 0.05
CFH	–3.263	0.004	4.40E-05	1q31.3	Protein coding
IGFBP5	–2.165	0.024	9.50E-05	2q35	Protein coding
COLEC12	–2.521	0.147	9.50E-05	18p11.32	Protein coding
RP11-726G1,1	–4.010	0.034	0.00027	12p13.31	Protein coding
ASPA	–3.746	0.361	0.00027	17p13.2	Protein coding
RP11-720L2,3	–3.156	0.801	0.00027	18p11.32	lincRNA
COL4A5	–1.833	0.004	0.00045	Xq22.3	Protein coding
MAF	–1.621	0.085	0.00072	16q23.2	Protein coding
SEPP1	–1.966	0.104	0.00083	5p12	Protein coding
NPY	–3.406	0.784	0.001	7p15.3	Protein coding

Biotype: lincRNA (long intergenic non-coding RNA); proc. (processed) (processed transcripts also known as lncRNA – long non-coding RNA).

DISCUSSION

In this paper, we provide a proof of concept for the utility of PBMNC reprogrammed iPSC-derived neurons as cellular models for the investigation of underlying molecular mechanism of sporadic AD. The results are tantalizing and indicate common deregulation of genes that could be implicated in AD as well as confirm deregulation of several genes already associated with sporadic AD. The cryopreserved PBMNC from 290 cases in the CIMA-Q cohort well-ascertained clinically, neuropsychologically, and by neuroimaging offers the basic resources to endeavor the study of a large and costly establishment of more iPSC lines.

We characterized in detail two clones each from six iPSC lines reprogrammed from the PBMNC of three AD and three non-AD individuals well-ascertained clinically, neuropsychologically, and by neuroimaging. These iPSC or iPSC-derived neuronal lines did not show any significant genetic variability between clones from one subject or between NCI and AD iPSC lines. Of the 12 iPSC clones assessed, one showed chromosomal abnormalities and it was replaced with another clone with normal karyotype. Genetic variations have previously been observed in iPSC lines before and arise in about 12.5% of cultures, consistent with our results [65]. These chromosomal abnormalities may arise from genetically heterogeneous PBMNC from one subject, from the reprograming step, during the numerous passages of the iPSC cultures, or represent a positive selection step [66]. Whole exome sequencing of the iPSC-derived neurons demonstrated that over 97% of 12,000 SNPs were identical in PBMNC and iPSC-derived neurons and this value was essentially the same for NCI- and AD-derived cells.

The expression of Tau and APP mRNAs and the production of Aβ was not different between sporadic AD and NCI lines. Consistently, three sporadic AD lines were reported to have no changes in Aβ ratios [8, 16]. Nevertheless, two iPSC lines had increased Aβ₄₀ [8] or Aβ₄₂/Aβ₄₀ ratios [26]. The production of Aβ may vary depending on the time and conditions of the culture and therefore will need to be systematically analysed from a large number of iPSC-derived neurons to conclusively determine if these sporadic AD iPSC-derived neurons have altered levels of tau and Aβ. Despite a lack of change in Aβ and tau, the iPSC-derived neurons seem to replicate other features of AD. MAP2 immunopositive neurons were significantly decreased in the AD lines, consistent with previous observations where mature MAP2 was decreased in the dentate gyrus of AD brains [67]. Stringent analysis of iPSC-derived neuronal gene expression by RNA Seq (DeSeq method) revealed the upregulation of SIRBP1, CYP4F29P mRNAs and the downregulation of the CFH, IGFBP5, RP11-726G1,1, and COL4A5 mRNAs in AD compared to the NCI cases. Of these, CFH mRNA was observed to be downregulated in AD and primary brain cells due to increased levels of the miRNA146a [68 –70]. In contrast, CFH mRNA levels were not changed in other studies [71, 72]. Nevertheless, plasma CFH protein was convincingly shown to be increased and to correlate negatively with the mini mental status examination score [73] and CSF CFH correlated with the severity of AD [74]. SIRBP1 protein was increased in AD superior temporal cortex and in the APP J20 mouse model microglia and proposed to play a role by increasing phagocytosis of neural debris and amyloid [75]. IGFBP5 mRNA was previously shown to be upregulated by sublethal concentrations of Aβ₄₂ in cells, the protein increased in the TgCRND8 AD mouse model, and in the CSF of AD individuals [76]. The other genes identified through stringent RNA Seq analyses were not previously associated with AD. However, some of the differentially expressed genes are clearly related to brain development and neurogenesis. Of these, four were transcriptional factors. LHX8 is a transcription factor involved in the differentiation and development of basal forebrain cholinergic neurons [77, 78]. Six3 is a transcription factor regulating cellular proliferation and differentiation during neurogenesis [79]. SIM2, present in the Down syndrome duplicated region of chromosome 21, is a master transcriptional regulator of CNS development [80]. The MAF family of transcriptional factors play a role in peripheral and central nervous system development [81 –85]. Dysregulation of these transcription factors could lead to predisposition to AD in the aged individual, therefore, it might be important to investigate the functional significance of these dysregulated transcription factors in the brain and whether there is any association with sporadic forms of AD.

Assessing the RNA Seq less stringent Edger results by STRING analysis did not reveal any highly significant functional protein association between the top ten upregulated and the top ten downregulated genes, except for ASPA and GAD2 (high confidence 0.700), which are involved in alanine, aspartate and glutamate metabolism (KEGG pathway).

The identification of genes already identified as potential AD biomarkers in iPSC-derived neurons provides positive support for the use of iPSC-derived neurons as cellular models to identify underlying molecular mechanisms of sporadic AD. Nevertheless, there are several limitations to this type of study that need to be considered. First, it is unclear if each sporadic form of AD derives from a common aberrant pathway. It is possible that individual sporadic AD subjects differ from each other significantly and thus it may be necessary to examine the results in a personalized manner rather than investigate mechanisms common to all sporadic AD individuals. For a personalized approach, it would be necessary to derive data from hundreds of normal individuals to establish a database to compare each sporadic AD with the norm. Second, we question whether iPSC-derived neurons are sufficient to identify abnormal mechanisms. The brain is composed of many cell types such as astrocytes and microglial cells, as well as vascular systems, that coordinately work together for normal function. There is a risk that dissecting out mechanisms occurring in one cell type may not be representative of the disease state. Third, the NCI normal subjects could be pre-AD themselves. Therefore, it is important to use PBMNC that came from longitudinally studied individuals to partially eliminate the risk of adding pre-AD subjects in the normal controls. Fourth, there are many types of neurons in brain and restricting analysis to one specific type of differentiated neurons could restrict identification of major disease pathways. Fifth, extended time in culture may be necessary to accurately replicate the differentiated state of mature neurons from aged individuals. Sixth, while studying mRNA gene expression is relatively easy, adding proteomic analyses would be essential to assess the functional consequence of differential gene expression between AD and NCI. Seventh, the age-dependent risk factors that lead to AD may need to be considered especially when using these cellular models to identify underlying molecular mechanisms of disease. Nevertheless, despite these limitations, it is hoped that the study of iPSC-derived neurons from this well ascertained CIMA-Q cohort will lead to the identification of molecular mechanisms of disease and maybe in the future, as cellular models for the development of drugs against AD.

Footnotes

ACKNOWLEDGMENTS

We acknowledge Dr. Jack Puymirat and the iPSC Québec platform (Laval University, Québec) for generating iPSC from PBMNC and Dr. Tom Durcan, and the Montreal Neurological Institute (McGill University, Montreal) iPSC/CRISPR platform for generating iPSC-derived neurons. We acknowledge McGill University and the Genome Quebec Innovation Center for RNA sequencing and Whole Exome Sequencing as well as the Canadian Centre for Computational Genomics (C3G) - Montréal Node team for the analysis of exome and RNA sequencing data. The study was supported by Fonds de Recherche Québec-Santé-Pfizer and Réseau Québecois de la Recherche sur le Vieillissement grants to CIMA-Q to SB, ALB, SD and an Alzheimer Society of Canada grant to ALB.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Takahashi

, Tanabe

, Ohnuki

, Narita

, Ichisaka

, Tomoda

, Yamanaka

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

, Vodyanik

, Smuga-Otto

, Antosiewicz-Bourget

, Frane

, Tian

, Nie

, Jonsdottir

, Ruotti

, Stewart

, Slukvin

, Thomson

(2007) Induced pluripotent stem cell lines derived from human somatic cells. Science 318, 1917–1920.

, Zhu

, Asakawa

, Kuma

, Hirata

, Ueda

, Lee

, Fukumura

, Iida

, Kato

, Nagai

, Hasegawa

(2000) A cytoplasmic RNA vector derived from nontransmissible Sendai virus with efficient gene transfer and expression. J Virol 74, 6564–6569.

Fusaki

, Ban

, Nishiyama

, Saeki

, Hasegawa

(2009) Efficient induction of transgene-free human pluripotent stem cells using a vector based on Sendai virus, an RNA virus that does not integrate into the host genome. Proc Jpn Acad Ser B Phys Biol Sci 85, 348–362.

Seki

, Yuasa

, Oda

, Egashira

, Yae

, Kusumoto

, Nakata

, Tohyama

, Hashimoto

, Kodaira

, Okada

, Seimiya

, Fusaki

, Hasegawa

, Fukuda

(2010) Generation of induced pluripotent stem cells from human terminally differentiated circulating T cells. Cell Stem Cell 7, 11–14.

Ban

, Nishishita

, Fusaki

, Tabata

, Saeki

, Shikamura

, Takada

, Inoue

, Hasegawa

, Kawamata

, Nishikawa

(2011) Efficient generation of transgene-free human induced pluripotent stem cells (iPSCs) by temperature-sensitive Sendai virus vectors. Proc Natl Acad Sci U S A 108, 14234–14239.

, Muench

, Fusaki

, Beyer

, Wang

, Qi

, Yu

, Kan

(2013) Blood cell-derived induced pluripotent stem cells free of reprogramming factors generated by Sendai viral vectors. Stem Cells Transl Med 2, 558–566.

Israel

, Yuan

, Bardy

, Reyna

, Mu

, Herrera

, Hefferan

, Van Gorp

, Nazor

, Boscolo

, Carson

, Laurent

, Marsala

, Gage

, Remes

, Koo

, Goldstein

(2012) Probing sporadic and familial Alzheimer’s disease using induced pluripotent stem cells. Nature 482, 216–220.

Yagi

, Ito

, Okada

, Akamatsu

, Nihei

, Yoshizaki

, Yamanaka

, Okano

, Suzuki

(2011) Modeling familial Alzheimer’s disease with induced pluripotent stem cells. Hum Mol Genet 20, 4530–4539.

10.

Koch

, Tamboli

, Mertens

, Wunderlich

, Ladewig

, Stuber

, Esselmann

, Wiltfang

, Brustle

, Walter

(2012) Presenilin-1 L166P mutant human pluripotent stem cell-derived neurons exhibit partial loss of gamma-secretase activity in endogenous amyloid-beta generation. Am J Pathol 180, 2404–2416.

11.

Liu

, Waltz

, Woodruff

, Ouyang

, Israel

, Herrera

, Sarsoza

, Tanzi

, Koo

, Ringman

, Goldstein

, Wagner

, Yuan

(2014) Effect of potent gamma-secretase modulator in human neurons derived from multiple presenilin 1-induced pluripotent stem cell mutant carriers. JAMA Neurol 71, 1481–1489.

12.

Sproul

, Jacob

, Pre

, Kim

, Nestor

, Navarro-Sobrino

, Santa-Maria

, Zimmer

, Aubry

, Steele

, Kahler

, Dranovsky

, Arancio

, Crary

, Gandy

, Noggle

(2014) Characterization and molecular profiling of PSEN1 familial Alzheimer’s disease iPSC-derived neural progenitors. PLoS One 9, e84547.

13.

Mahairaki

, Ryu

, Peters

, Chang

, Li

, Park

, Burridge

, Talbot

Jr , Asnaghi

, Martin

, Zambidis

, Koliatsos

(2014) Induced pluripotent stem cells from familial Alzheimer’s disease patients differentiate into mature neurons with amyloidogenic properties. Stem Cells Dev 23, 2996–3010.

14.

Muratore

, Rice

, Srikanth

, Callahan

, Shin

, Benjamin

, Walsh

, Selkoe

, Young-Pearse

(2014) The familial Alzheimer’s disease APPV717I mutation alters APP processing and Tau expression in iPSC-derived neurons. Hum Mol Genet 23, 3523–3536.

15.

Muratore

, Zhou

, Liao

, Fernandez

, Taylor

, Lagomarsino

, Pearse

2nd , Rice

, Negri

, He

, Srikanth

, Callahan

, Shin

, Zhou

, Bennett

, Noggle

, Love

, Selkoe

, Young-Pearse

(2017) Cell-type Dependent Alzheimer’s disease phenotypes: Probing the biology of selective neuronal vulnerability. Stem Cell Reports 9, 1868–1884.

16.

Kondo

, Asai

, Tsukita

, Kutoku

, Ohsawa

, Sunada

, Imamura

, Egawa

, Yahata

, Okita

, Takahashi

, Asaka

, Aoi

, Watanabe

, Kadoya

, Nakano

, Watanabe

, Maruyama

, Hori

, Hibino

, Choshi

, Nakahata

, Hioki

, Kaneko

, Naitoh

, Yoshikawa

, Yamawaki

, Suzuki

, Hata

, Ueno

, Seki

, Kobayashi

, Toda

, Murakami

, Irie

, Klein

, Mori

, Asada

, Takahashi

, Iwata

, Yamanaka

, Inoue

(2013) Modeling Alzheimer’s disease with iPSCs reveals stress phenotypes associated with intracellular Abeta and differential drug responsiveness. Cell Stem Cell 12, 487–496.

17.

Woodruff

, Young

, Martinez

, Buen

, Gore

, Kinaga

, Li

, Yuan

, Zhang

, Goldstein

(2013) The presenilin-1 DeltaE9 mutation results in reduced gamma-secretase activity, but not total loss of PS1 function, in isogenic human stem cells. Cell Rep 5, 974–985.

18.

Raja

, Mungenast

, Lin

, Ko

, Abdurrob

, Seo

, Tsai

(2016) Self-organizing 3D human neural tissue derived from induced pluripotent stem cells recapitulate Alzheimer’s disease phenotypes. PLoS One 11, e0161969.

19.

Choi

, Kim

, Hebisch

, Sliwinski

, Lee

, D’Avanzo

, Chen

, Hooli

, Asselin

, Muffat

, Klee

, Zhang

, Wainger

, Peitz

, Kovacs

, Woolf

, Wagner

, Tanzi

, Kim

(2014) A three-dimensional human neural cell culture model of Alzheimer’s disease. Nature 515, 274–278.

20.

Hossini

, Megges

, Prigione

, Lichtner

, Toliat

, Wruck

, Schroter

, Nuernberg

, Kroll

, Makrantonaki

, Zouboulis

, Adjaye

(2015) Induced pluripotent stem cell-derived neuronal cells from a sporadic Alzheimer’s disease donor as a model for investigating AD-associated gene regulatory networks. BMC Genomics 16, 84.

21.

Chandrasekaran

, Varga

, Nemes

, Tancos

, Kobolak

, Dinnyes

(2016) Establishment of induced pluripotent stem cell (iPSC) line from a 63-year old patient with late onset Alzheimer’s disease (LOAD). Stem Cell Res 17, 78–80.

22.

Ochalek

, Nemes

, Varga

, Tancos

, Kobolak

, Dinnyes

(2016) Establishment of induced pluripotent stem cell (iPSC) line from a 57-year old patient with sporadic Alzheimer’s disease. Stem Cell Res 17, 72–74.

23.

Tancos

, Varga

, Kovacs

, Dinnyes

, Kobolak

(2016) Establishment of induced pluripotent stem cell (iPSC) line from a 75-year old patient with late onset Alzheimer’s disease (LOAD). Stem Cell Res 17, 81–83.

24.

Tancos

, Varga

, Kovacs

, Dinnyes

, Kobolak

(2016) Establishment of induced pluripotent stem cell (iPSC) line from an 84-year old patient with late onset Alzheimer’s disease (LOAD). Stem Cell Res 17, 75–77.

25.

Zhang

, Lv

, Zhang

, Liu

, Li

, Gong

, Sha

, Wu

(2017) Characterization of human induced pluripotent stem cell (iPSC) line from a 72year old male patient with later onset Alzheimer’s disease. Stem Cell Res 19, 34–36.

26.

Duan

, Bhattacharyya

, Belmadani

, Pan

, Miller

, Kessler

(2014) Stem cell derived basal forebrain cholinergic neurons from Alzheimer’s disease patients are more susceptible to cell death. Mol Neurodegener 9, 3.

27.

Loh

, Hartung

, Li

, Guo

, Sahalie

, Manos

, Urbach

, Heffner

, Grskovic

, Vigneault

, Lensch

, Park

, Agarwal

, Church

, Collins

, Irion

, Daley

(2010) Reprogramming of T cells from human peripheral blood. Cell Stem Cell 7, 15–19.

28.

Staerk

, Dawlaty

, Gao

, Maetzel

, Hanna

, Sommer

, Mostoslavsky

, Jaenisch

(2010) Reprogramming of human peripheral blood cells to induced pluripotent stem cells. Cell Stem Cell 7, 20–24.

29.

Chou

, Mali

, Huang

, Ye

, Dowey

, Resar

, Zou

, Zhang

, Tong

, Cheng

(2011) Efficient human iPS cell derivation by a non-integrating plasmid from blood cells with unique epigenetic and gene expression signatures. Cell Res 21, 518–529.

30.

Daheron

, D’Souza

(2008) Blood - SeV derived fibroblast generated iPSCs. In StemBook, Cambridge, MA.

31.

Churko

, Burridge

, Wu

(2013) Generation of human iPSCs from human peripheral blood mononuclear cells using non-integrative Sendai virus in chemically defined conditions. Methods Mol Biol 1036, 81–88.

32.

Dale

, Fischl

, Sereno

(1999) Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9, 179–194.

33.

Fischl

, Sereno

, Dale

(1999) Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. Neuroimage 9, 195–207.

34.

Potvin

, Dieumegarde

, Duchesne

, Alzheimer’s Disease Neuroimaging Initiative (2017) Normative morphometric data for cerebral cortical areas over the lifetime of the adult human brain. Neuroimage 156, 315–339.

35.

Potvin

, Mouiha

, Dieumegarde

, Duchesne

, Alzheimer’s Disease Neuroimaging Initiative (2016) Normative data for subcortical regional volumes over the lifetime of the adult human brain. Neuroimage 137, 9–20.

36.

Coupe

, Yger

, Prima

, Hellier

, Kervrann

, Barillot

(2008) An optimized blockwise nonlocal means denoising filter for 3-D magnetic resonance images. IEEE Trans Med Imaging 27, 425–441.

37.

Sled

, Zijdenbos

, Evans

(1998) A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging 17, 87–97.

38.

Fonov

, Evans

, Botteron

, Almli

, McKinstry

, Collins

, Brain Development Cooperative Group (2011) Unbiased average age-appropriate atlases for pediatric studies. Neuroimage 54, 313–327.

39.

Eskildsen

, Coupe

, Fonov

, Manjon

, Leung

, Guizard

, Wassef

, Ostergaard

, Collins

, Alzheimer’s Disease Neuroimaging Initiative (2012) BEaST: Brain extraction based on nonlocal segmentation technique. Neuroimage 59, 2362–2373.

40.

Coupe

, Eskildsen

, Manjon

, Fonov

, Collins

, Alzheimer’s Disease Neuroimaging Initiative (2012) Simultaneous segmentation and grading of anatomical structures for patient’s classification: Application to Alzheimer’s disease. Neuroimage 59, 3736–3747.

41.

Ruitenberg

, Mulder

, Maino

, Landay

, Ghanekar

(2006) VACUTAINER CPT and Ficoll density gradient separation perform equivalently in maintaining the quality and function of PBMC from HIV seropositive blood samples. BMC Immunol 7, 11.

42.

Chen

, Fukuda

, Fusaki

, Iida

, Hasegawa

, Lichtler

, Reichenberger

(2013) Induced pluripotent stem cell reprogramming by integration-free Sendai virus vectors from peripheral blood of patients with craniometaphyseal dysplasia. Cell Reprogram 15, 503–513.

43.

Lieu

, Fontes

, Vemuri

, Macarthur

(2013) Generation of induced pluripotent stem cells with CytoTune, a non-integrating Sendai virus. Methods Mol Biol 997, 45–56.

44.

Bayart

, Cohen-Haguenauer

(2013) Technological overview of iPS induction from human adult somatic cells. Curr Gene Ther 13, 73–92.

45.

Ludwig

, Levenstein

, Jones

, Berggren

, Mitchen

, Frane

, Crandall

, Daigh

, Conard

, Piekarczyk

, Llanas

, Thomson

(2006) Derivation of human embryonic stem cells in defined conditions. Nat Biotechnol 24, 185–187.

46.

(2008) Feeder-independent culture protocol - EDTA splitting. In StemBook, Cambridge, MA.

47.

Liu

, Zhang

(2011) Specification of neuronal and glial subtypes from human pluripotent stem cells. Cell Mol Life Sci 68, 3995–4008.

48.

Chambers

, Fasano

, Papapetrou

, Tomishima

, Sadelain

, Studer

(2009) Highly efficient neural conversion of human ES and iPS cells by dual inhibition of SMAD signaling. Nat Biotechnol 27, 275–280.

49.

Elkabetz

, Panagiotakos

, Al Shamy

, Socci

, Tabar

, Studer

(2008) Human ES cell-derived neural rosettes reveal a functionally distinct early neural stem cell stage. Genes Dev 22, 152–165.

50.

Falk

, Koch

, Kesavan

, Takashima

, Ladewig

, Alexander

, Wiskow

, Tailor

, Trotter

, Pollard

, Smith

, Brustle

(2012) Capture of neuroepithelial-like stem cells from pluripotent stem cells provides a versatile system for in vitro production of human neurons. PLoS One 7, e29597.

51.

Koch

, Opitz

, Steinbeck

, Ladewig

, Brustle

(2009) A rosette-type, self-renewing human ES cell-derived neural stem cell with potential for in vitro instruction and synaptic integration. Proc Natl Acad Sci U S A 106, 3225–3230.

52.

Bissonnette

, Lyass

, Bhattacharyya

, Belmadani

, Miller

, Kessler

(2011) The controlled generation of functional basal forebrain cholinergic neurons from human embryonic stem cells. Stem Cells 29, 802–811.

53.

Livak

, Schmittgen

(2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408.

54.

Bolger

, Lohse

, Usadel

(2014) Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120.

55.

Dobin

, Davis

, Schlesinger

, Drenkow

, Zaleski

, Jha

, Batut

, Chaisson

, Gingeras

(2013) STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21.

56.

Trapnell

, Williams

, Pertea

, Mortazavi

, Kwan

, van Baren

, Salzberg

, Wold

, Pachter

(2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28, 511–515.

57.

Trapnell

, Hendrickson

, Sauvageau

, Goff

, Rinn

, Pachter

(2013) Differential analysis of gene regulation at transcript resolution with RNA-seq. Nat Biotechnol 31, 46–53.

58.

Huber

, Carey

, Gentleman

, Anders

, Carlson

, Carvalho

, Bravo

, Davis

, Gatto

, Girke

, Gottardo

, Hahne

, Hansen

, Irizarry

, Lawrence

, Love

, MacDonald

, Obenchain

, Oles

, Pages

, Reyes

, Shannon

, Smyth

, Tenenbaum

, Waldron

, Morgan

(2015) Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods 12, 115–121.

59.

Robinson

, McCarthy

, Smyth

(2010) edgeR: A Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140.

60.

Anders

, Huber

(2010) Differential expression analysis for sequence count data. Genome Biol 11, R106.

61.

Young

, Wakefield

, Smyth

, Oshlack

(2010) Gene ontology analysis for RNA-seq: Accounting for selection bias. Genome Biol 11, R14.

62.

Cingolani

, Platts

, Wang le

, Coon

, Nguyen

, Wang

, Land

, Lu

, Ruden

(2012) A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin) 6, 80–92.

63.

Lambert

, Ibrahim-Verbaas

, Harold

, Naj

, Sims

, Bellenguez

, DeStafano

, Bis

, Beecham

, Grenier-Boley

, Russo

, Thorton-Wells

, Jones

, Smith

, Chouraki

, Thomas

, Ikram

, Zelenika

, Vardarajan

, Kamatani

, Lin

, Gerrish

, Schmidt

, Kunkle

, Dunstan

, Ruiz

, Bihoreau

, Choi

, Reitz

, Pasquier

, Cruchaga

, Craig

, Amin

, Berr

, Lopez

, De Jager

, Deramecourt

, Johnston

, Evans

, Lovestone

, Letenneur

, Morón

, Rubinsztein

, Eiriksdottir

, Sleegers

, Goate

, Fiévet

, Huentelman

, Gill

, Brown

, Kamboh

, Keller

, Barberger-Gateau

, McGuiness

, Larson

, Green

, Myers

, Dufouil

, Todd

, Wallon

, Love

, Rogaeva

, Gallacher

, St George-Hyslop

, Clarimon

, Lleo

, Bayer

, Tsuang

, Yu

, Tsolaki

, Bossú

, Spalletta

, Proitsi

, Collinge

, Sorbi

, Sanchez-Garcia

, Fox

, Hardy

, Deniz Naranjo

, Bosco

, Clarke

, Brayne

, Galimberti

, Mancuso

, Matthews

; European Alzheimer’s Disease Initiative (EADI); Genetic and Environmental Risk in Alzheimer’s Disease; Alzheimer’s Disease Genetic Consortium; Cohorts for Heart and Aging Research in Genomic Epidemiology, Moebus

, Mecocci

, Del Zompo

, Maier

, Hampel

, Pilotto

, Bullido

, Panza

, Caffarra

, Nacmias

, Gilbert

, Mayhaus

, Lannefelt

, Hakonarson

, Pichler

, Carrasquillo

, Ingelsson

, Beekly

, Alvarez

, Zou

, Valladares

, Younkin

, Coto

, Hamilton-Nelson

, Gu

, Razquin

, Pastor

, Mateo

, Owen

, Faber

, Jonsson

, Combarros

, O’Donovan

, Cantwell

, Soininen

, Blacker

, Mead

, Mosley

Jr , Bennett

, Harris

, Fratiglioni

, Holmes

, de Bruijn

, Passmore

, Montine

, Bettens

, Rotter

, Brice

, Morgan

, Foroud

, Kukull

, Hannequin

, Powell

, Nalls

, Ritchie

, Lunetta

, Kauwe

, Boerwinkle

, Riemenschneider

, Boada

, Hiltuenen

, Martin

, Schmidt

, Rujescu

, Wang

, Dartigues

, Mayeux

, Tzourio

, Hofman

, Nöthen

, Graff

, Psaty

, Jones

, Haines

, Holmans

, Lathrop

, Pericak-Vance

, Launer

, Farrer

, van Duijn

, Van Broeckhoven

, Moskvina

, Seshadri

, Williams

, Schellenberg

, Amouyel

(2013) Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat Genet 45, 1452–1458.

64.

Zhang

, Jiao

, Zhou

, Shen

(2016) Modeling Alzheimer’s disease with induced pluripotent stem cells: Current challenges and future concerns. Stem Cells Int 2016, 7828049.

65.

Taapken

, Nisler

, Newton

, Sampsell-Barron

, Leonhard

, McIntire

, Montgomery

(2011) Karotypic abnormalities in human induced pluripotent stem cells and embryonic stem cells. Nat Biotechnol 29, 313–314.

66.

Liang

, Zhang

(2013) Genetic and epigenetic variations in iPSCs: Potential causes and implications for application. Cell Stem Cell 13, 149–159.

67.

, Yamamori

, Tatebayashi

, Shafit-Zagardo

, Tanimukai

, Chen

, Iqbal

, Grundke-Iqbal

(2008) Failure of neuronal maturation in Alzheimer disease dentate gyrus. J Neuropathol Exp Neurol 67, 78–84.

68.

Lukiw

, Alexandrov

(2012) Regulation of complement factor H (CFH) by multiple miRNAs in Alzheimer’s disease (AD) brain. Mol Neurobiol 46, 11–19.

69.

Lukiw

, Zhao

, Cui

(2008) An NF-kappaB-sensitive micro RNA-146a-mediated inflammatory circuit in Alzheimer disease and in stressed human brain cells. J Biol Chem 283, 31315–31322.

70.

Pogue

, Li

, Cui

, Zhao

, Kruck

, Percy

, Tarr

, Lukiw

(2009) Characterization of an NF-kappaB-regulated, miRNA-146a-mediated down-regulation of complement factor H (CFH) in metal-sulfate-stressed human brain cells. J Inorg Biochem 103, 1591–1595.

71.

Finehout

, Franck

, Lee

(2005) Complement protein isoforms in CSF as possible biomarkers for neurodegenerative disease. Dis Markers 21, 93–101.

72.

Strohmeyer

, Shen

, Rogers

(2000) Detection of complement alternative pathway mRNA and proteins in the Alzheimer’s disease brain. Brain Res Mol Brain Res 81, 7–18.

73.

Hye

, Lynham

, Thambisetty

, Causevic

, Campbell

, Byers

, Hooper

, Rijsdijk

, Tabrizi

, Banner

, Shaw

, Foy

, Poppe

, Archer

, Hamilton

, Powell

, Brown

, Sham

, Ward

, Lovestone

(2006) Proteome-based plasma biomarkers for Alzheimer’s disease. Brain 129, 3042–3050.

74.

Wang

, Hancock

, Bradner

, Chung

, Quinn

, Peskind

, Galasko

, Jankovic

, Zabetian

, Kim

, Leverenz

, Montine

, Ginghina

, Edwards

, Snapinn

, Goldstein

, Shi

, Zhang

(2011) Complement 3 and factor h in human cerebrospinal fluid in Parkinson’s disease, Alzheimer’s disease, and multiple-system atrophy. Am J Pathol 178, 1509–1516.

75.

Gaikwad

, Larionov

, Wang

, Dannenberg

, Matozaki

, Monsonego

, Thal

, Neumann

(2009) Signal regulatory protein-beta1: A microglial modulator of phagocytosis in Alzheimer’s disease. Am J Pathol 175, 2528–2539.

76.

Barucker

, Sommer

, Beckmann

, Eravci

, Harmeier

, Schipke

, Brockschnieder

, Dyrks

, Althoff

, Fraser

, Hazrati

, George-Hyslop

, Breitner

, Peters

, Multhaup

(2015) Alzheimer amyloid peptide abeta42 regulates gene expression of transcription and growth factors. J Alzheimers Dis 44, 613–624.

77.

Mori

, Yuxing

, Takaki

, Takeuchi

, Iseki

, Hagino

, Kitanaka

, Takemura

, Misawa

, Ikawa

, Okabe

, Wanaka

(2004) The LIM homeobox gene, L3/Lhx8, is necessary for proper development of basal forebrain cholinergic neurons. Eur J Neurosci 19, 3129–3141.

78.

Zhao

, Marin

, Hermesz

, Powell

, Flames

, Palkovits

, Rubenstein

, Westphal

(2003) The LIM-homeobox gene Lhx8 is required for the development of many cholinergic neurons in the mouse forebrain. Proc Natl Acad Sci U S A 100, 9005–9010.

79.

Appolloni

, Calzolari

, Corte

, Perris

, Malatesta

(2008) Six3 controls the neural progenitor status in the murine CNS. Cereb Cortex 18, 553–562.

80.

Dahmane

, Charron

, Lopes

, Yaspo

, Maunoury

, Decorte

, Sinet

, Bloch

, Delabar

(1995) Down syndrome-critical region contains a gene homologous to Drosophila sim expressed during rat and human central nervous system development. Proc Natl Acad Sci U S A 92, 9191–9195.

81.

Benkhelifa

, Provot

, Lecoq

, Pouponnot

, Calothy

, Felder-Schmittbuhl

(1998) mafA, a novel member of the maf proto-oncogene family, displays developmental regulation and mitogenic capacity in avian neuroretina cells. Oncogene 17, 247–254.

82.

, Huang

, Li

, Guo

, Cheng

(2012) c-Maf is required for the development of dorsal horn laminae III/IV neurons and mechanoreceptive DRG axon projections. J Neurosci 32, 5362–5373.

83.

Katsuoka

, Motohashi

, Tamagawa

, Kure

, Igarashi

, Engel

, Yamamoto

(2003) Small Maf compound mutants display central nervous system neuronal degeneration, aberrant transcription, and Bach protein mislocalization coincident with myoclonus and abnormal startle response. Mol Cell Biol 23, 1163–1174.

84.

Kurschner

, Morgan

(1995) The maf proto-oncogene stimulates transcription from multiple sites in a promoter that directs Purkinje neuron-specific gene expression. Mol Cell Biol 15, 246–254.

85.

Wende

, Lechner

, Birchmeier

(2012) The transcription factor c-Maf in sensory neuron development. Transcription 3, 285–289.