| Series | GSE86977 |
| Title | REGION-SPECIFIC NEURAL STEM CELL LINEAGES REVEALED BY SINGLE-CELL RNA-SEQ FROM HUMAN EMBRYONIC STEM CELLS [Cel-seq] |
|---|
| Year | 2016 |
|---|
| Country | USA |
|---|
| Article | Not set |
|---|
| PMID | NA |
|---|
| Bio Project | BioProject: https://www.ncbi.nlm.nih.gov/bioproject/PRJNA343284 |
|---|
| Sra | SRA: https://www.ncbi.nlm.nih.gov/sra?term=SRP090072 |
|---|
| Overall Desgin | The transcriptomes of 2684 single cells were profiled by CelSeq at different timepoints throughout a 54-day differentiation protocol that converted H1 human embryonic stem cells to a variety of brain cell types. |
|---|
| Summary | During development of the human brain, multiple cell types with diverse regional identities are generated. Here we report a system to generate early human brain forebrain and mid/hindbrain cell types from human embryonic stem cells (hESCs), and infer and experimentally confirm a lineage tree for the generation of these types based on single-cell RNA-Seq analysis. We engineered SOX2Cit/+ and DCXCit/Y hESC lines to target progenitors and neurons throughout neural differentiation for single-cell transcriptomic profiling, then identified discrete cell types consisting of both rostral (cortical) and caudal (mid/hindbrain) identities. Direct comparison of the cell types were made to primary tissues using gene expression atlases and fetal human brain single-cell gene expression data, and this established that the cell types resembled early human brain cell types, including preplate cells. From the single-cell transcriptomic data a Bayesian algorithm generated a unified lineage tree, and predicted novel regulatory transcription factors. The lineage tree highlighted a prominent bifurcation between cortical and mid/hindbrain cell types, confirmed by clonal analysis experiments. We demonstrated that cell types from either branch could preferentially generated by manipulation of the canonical Wnt/beta-catenin pathway. In summary, we present an experimentally validated lineage tree that encompasses multiple brain regions, and our work sheds light on the molecular regulation of region-specific neural lineages during human brain development. |
|---|
| Experimental Protocol | To generate single cell suspensions, hESC-derived cultures were dissociated from plates using Accutase (ThermoFisher) at 37°C. Light trituration using a P1000 pipette was done every 5 min until nearly all clumps had been dissociated (up to 1 h). Cell suspension was washed and filtered through a 40 μm cell strainer. Cells were washed in PBS with 1% FBS and stained with 0.5-1 μg/mL DAPI. Single-cell suspensions were loaded onto a FACSAria II SORP (Becton Dickinson) and sorted directly into PCR strip tubes or plates held in chilled aluminum blocks. Doublets and dead cells were excluded based on forward scatter, side scatter and DAPI fluorescence. Sorting was done using the 130 μm nozzle with the sort mode set to single cell. Accuracy of single-cell sorts was confirmed by sorting DAPI-stained fixed cells onto a dry well of a 96-well plate and analyzing by fluorescence microscopy.; CelSeq protocol with a few modifications. Single cells were sorted with a FACSAria (BD) into 96-well plates containing 1.2 μL 2× CellsDirect Buffer (Thermo Fisher) with 0.1 μL of External RNA Controls Consortium (ERCC) control RNAs diluted to 1 × 10-6 molecules (Thermo Fisher). After sorting, plates were then frozen and stored at -80C. For library preparation, plates were thawed on ice. mRNA was reverse transcribed using 1.25 pmol or 0.15625 pmol of oligoT primer carrying a cell-specific 8 NT barcode and a 5 NT unique molecular identifier (UMI) (Islam et al., 2014) (see Table S4). Barcode design ensured at least three nucleotide differences from any other barcode. Samples were incubated in a PCR machine (Tetrad, BioRad) at 70 °C with a 70 °C heated lid for 3 min, spun, and heated again for two more minutes. Samples were reverse transcribed using Superscript III (Thermo Fisher) for two hours at 50 °C with a 52 °C lid and subsequently treated with 1 μL of ExoSAP-IT (Affymetrix). Samples were cooled on ice for second strand synthesis, where Second Strand Synthesis Buffer, dNTPs, DNA Polymerase, and RNAse H (NEB) were added to the samples for a 10 μL total volume and incubated at 16 °C for 2 h. Single cells were pooled by 24 wells per library, with each library containing a water-only well and an ERCC-only well. Single cell pools or population RNA libraries were purified with an equal volume of RNA Clean Beads (Beckman Coulter), linearly amplified at 37 C for 15 h using the HiScribe T7 High Yield RNA Synthesis kit (NEB), and treated with DNAse I (Thermo Fisher). RNA was fragmented using the NEBNext RNA Fragmentation Module (NEB), purified using an equal volume of RNA Clean Beads, and visualized (RNA Pico Kit, Bioanalyzer 2100, Agilent). The RNA fragments were repaired by treating with Antarctic Phosphatase and Polynucleotide Kinase (NEB) and purified with an equal volume of RNA Clean Beads. cDNA libraries were made using the NEBNext Small Library Prep Kit according to the manufacturer’s protocol, except Superscript III was used for the RT step. Index primers were used in PCR amplification. Libraries were purified using an equal volume of RNA Clean Beads and were quantified on the Bioanalyzer using the DNA High Sensitivity Kit (Agilent). Approximately 160-200 nmol of a pool of libraries were size selected to exclude species <180 bp on a 2% Dye-Free cassette on the Pippin Prep (Sage) and Speed Vac concentrated to approximately 14 μL. Libraries were then quantified by qRT-PCR using p5 (5’-AATGATACGGCGACCACCGAGA-3’) and p7 (5’-CAAGCAGAAGACGGCATACGAGAT-3’) primers and visualized (DNA High Sensitivity Kit, Bioanalyzer 2100). Library pools were then sequenced on an Illumina HiSeq using a custom read1 primer (5’-TCTACACGTTCAGAGTTCTACAGTCCGACGATC-3’) and the Illumina primer HP10. Standard Illumina primers HP12 and HP11 were used for the index read and the transcript read, respectively. PE50 kits (Illumina) were used for sequencing with read lengths of 25 nt, 6 nt, and 47 nt for read1, index, and read2, respectively.
|
|---|
| Data processing | Reads were de-multiplexed by CelSeq index, allowing for one sequence mismatch.; The transcript reads for each cell were aligned to the RefSeq transcriptome (downloaded March 2013) using Tophat with default parameters; Unaligned reads were then aligned to the genome using Bowtie (Langmead et al., 2009), followed by alignment to the ERCC spike-in controls.; Remaining unaligned reads were mapped to the genome again using GSNAP; Reads mapping to exons of the same gene were collapsed by their Unique Molecular Identifier (UMI); Reads mapped within 1kb of the 3′ end of a gene in the proper orientation are ascribed to that gene; Cells with fewer than 20,000 total cellular UMIs were discarded, and data for all remaining cells was randomly subsampled to 20,000 total cellular UMIs.; Genome_build: hg19; Supplementary_files_format_and_content: UMI_20K.2684.csv: UMI gene counts (subsampled to 20K); Supplementary_files_format_and_content: UMI.2684.csv: UMI gene counts
|
|---|
| Platform | GPL16791 |
|---|
| Public On | Public on Sep 30 2016 |
|---|
Differential expression gene List between two groups.
