results EN ISO ) Module Vof 14 Technical solutions Example: reduction of the dissipation [ ] BGI Evaluation of the. A median number of 7, to 8, expressed genes were detected per cell ( Additional file 4: Supplementary Fig. S4d), including TFs that were. ; 7(10): – .. We wish to acknowledge the help of the BGI- Shenzhen for sequencing and Biochain-Beijing for array CGH.
|Published (Last):||17 September 2016|
|PDF File Size:||20.9 Mb|
|ePub File Size:||8.44 Mb|
|Price:||Free* [*Free Regsitration Required]|
Investigating cell fate decision and subpopulation specification in the context of the neural lineage is fundamental to understanding neurogenesis and neurodegenerative diseases. The differentiation process of neural-tube-like rosettes in vitro is representative of neural tube structures, which are composed of radially organized, columnar epithelial cells and give rise to functional neural cells.
However, the underlying regulatory network of cell fate commitment during early neural differentiation remains elusive. In this study, we investigated the genome-wide transcriptome profile of single cells from six consecutive reprogramming and neural differentiation time points and identified cellular subpopulations present at each differentiation stage.
Based on the inferred reconstructed trajectory and the characteristics of subpopulations contributing the most toward commitment to the central nervous system lineage at each stage during differentiation, we identified putative novel transcription factors in regulating neural differentiation. In addition, we dissected the dynamics of chromatin accessibility at the neural differentiation stages and revealed active cis -regulatory elements for transcription factors known to have a key role in neural differentiation as well as for those that we suggest are also involved.
Further, communication network analysis demonstrated that cellular interactions most frequently occurred in the embryoid body stage and that each cell subpopulation possessed a distinctive spectrum of ligands and receptors associated with neural differentiation that could reflect the identity of each subpopulation. Our study provides a comprehensive and integrative study of the transcriptomics and epigenetics of human early neural differentiation, which paves the way for a deeper understanding of the regulatory mechanisms driving the differentiation of the neural lineage.
The nervous system contains complex molecular circuitry in developmental processes. In humans, there is a paucity of data describing early neural development and the corresponding cellular heterogeneity at various stages. To our knowledge, neural tube formation and closure are crucial for embryonic central nervous system CNS development and the process of neurulation.
Previous studies have reported that neural tube closure is strongly controlled by both genetic and epigenetic factors and is sensitive to environmental influences [ 1—3 ]. Perturbations in this delicately balanced and orchestrated process can result in neural tube defects NTDswhich give rise to birth defects such as spina bifida, anencephaly, and encephaloceles.
However, the formation and closure of the neural tube in vivo during weeks 3 and 4 of human gestation are transient events and therefore difficult to capture.
Moreover, the limited accessibility of human abortive fetuses at such an early stage precludes a thorough investigation of human early neural development. Human pluripotent stem cells hPSCsincluding embryonic stem cells ESCs and induced pluripotent stem cells iPSCscan be differentiated into all cell types, including neural cells, offering a promising in vitro model for tracing early cell lineages and studying the cell fate specification of human neural differentiation [ 45 ].
Previous studies have indicated that inhibition of bone morphogenetic protein BMP signaling or activation of fibroblast growth factor FGF signaling is needed for induction of the neuroectoderm from ESCs [ 67 ]. A striking feature of differentiating stem cells in vitro is that they form neural tube-like rosettes that are composed of radially organized columnar epithelial cells that resemble the process of neurulation.
The progenitor cells in rosettes gradually give rise to functional cells e. These cellular processes suggest that distinct cell fate decisions and lineage commitments occur during rosette formation. However, the corresponding underlying mechanisms of the regulation of cell fate commitment during early neural differentiation remain largely unknown.
Genetic effects of a 13q microdeletion detected by noninvasive prenatal testing (NIPT)
The advance of single-cell trans-omics technology has offered incisive tools for revealing heterogeneous cellular contexts and developmental processes [ 9—11 ]. Single-cell RNA sequencing scRNA-seq has been applied to the study of cellular heterogeneity as well as to the identification of novel subtypes or intermediate cell groups in multiple contexts [ 12—15 70003 and may help delineate unexpected features of neural developmental biology and facilitate the study of cellular states and neurogenesis processes.
In the present study, we used scRNA-seq and assay for transposase-accessible chromatin using sequencing ATAC-seq to investigate human early neural differentiation.
Bgo analysis reveals the landscape of the transcriptome and cis- regulatory elements during this process and creates an unbiased classification of cell subpopulations during differentiation, providing a comprehensive description of transcriptomic and epigenetic patterns in cell fate decisions.
The differentiation system of human induced pluripotent bggi cells hiPSCs provides access to the very early stage of neural development and may serve as a source of specialized cells for regenerative medicine as well as support for further investigations of neural 0703 defects.
Here, we applied a well-adopted neural induction protocol and generated neural progenitor cells NPCs by forming neural rosettes in vitro [ 816 ]. We analyzed several differentiation stages of cells, including hiPSCs, embryoid body EBearly rosettes hereafter termed Ros-E, post-3 days of rosette formationlate rosettes hereafter termed Ros-L, post-5 days of rosette formationNPCs, and the original somatic fibroblasts Fib.
We also captured bulk transcriptome profiles of the corresponding neural differentiation stages derived from iPSCs and ESCs for validation. The quality of sequencing data was evaluated and filtered by a quality control QC pipeline developed in-house see Methods section for details. Transcriptome and regulome dynamics during human early neural differentiation.
Since the development of human ESCs 703 iPSCs, the ability to investigate human neurogenesis and neurological diseases via an in vitro differentiation model has vastly improved [ 417 ]. Subsequently, artificial neural cells have been successfully generated using a byi of protocols by several laboratories [ 7030 ]. First, pluripotency-associated transcription factors TFs e.
Cell stages are usually determined by a complement of TFs or master regulators, which regulate hundreds of genes associated with various cellular functions. To study the genomic features associated with open chromatin regions, we classified ATAC peaks based on the location of the peak center.
More than 16, peaks were identified for each cell stage Additional file 1: It is widely reported that chromatin gbi undergo widespread reprogramming during cell status transition, with some genomic regions becoming compacted or opened, leading to the switching on bggi off of a repertoire of genes responsible for cell fate decision [ 24—29 ]. We studied the dynamic chromatin landscape by tracing the temporal vgi of ATAC peaks at each stage with peaks nonoverlapping with existing ones that were annotated as novel peaks.
We assumed that those peaks, conserved among differentiation stages, are associated with housekeeping genes, while stage-dynamic peaks are likely to represent cis -regulatory elements important for cell status transition. Notably, more novel peaks appeared at the NPCs stage than at any other stage Fig. To reveal bgl detail of chromatin accessibility dynamics during neural differentiation, we also analyzed the gained or lost peaks at each stage compared with the previously neighboring one.
We observed that the number bi gained peaks was with the largest increase at the NPCs stage, while the number of lost peaks was relatively high at Ros-E stage Additional file 2: Next, we studied the genomic distribution of these dynamic peaks and found that both the gained and lost peaks were located mostly in distal intergenic regions and promoter regions Additional file 2: This observation indicates that distal and promoter regions are more dynamic compared to other genomic regions during the neural differentiation process.
To gain insight into bgo potential function of closing lost peaks dynamics, we carried out GO enrichment analysis on the genes associated with lost peaks at each stage. Further GO term and KEGG enrichment analysis showed very similar results with annotation analysis of novel peaks in corresponding cell stages Additional file 3: These findings strongly suggest that the novel, gained and lost, as well as stage-specific peaks, represent cell status and cell fate transitions that progress neural differentiation and that the landscape of cis -regulatory element accessibility throughout the differentiation process is highly dynamic.
To more thoroughly investigate the molecular mechanisms 703 neural differentiation, we profiled the transcriptomes of single cells. For subsequent analysis, we focused on cells that passed the QC Methods section, Additional file 4: A median number of 7, to 8, expressed genes were detected per cell Additional file 4: Because the neural rosette recapitulates neural bgo development in vitrowe paid particular attention to the Ros-E and Ros-L stages.
Of particular interest is the gene GRHL3. Expression of this gene is associated with bgl tube closure in mice [ 3132 ], and we observed this gene to be highly expressed at Ros-E in human cells, suggesting that its role in neural tube closure may be conserved across mammals or possibly chordates. TFAP2A transcription factor AP-2 alpha and TFAP2B transcription factor AP-2 beta have been proposed 70033 master regulators of the neural crest cell, and loss of function of transcription factor AP-2 in mice is strongly associated with a cranial neural tube defect phenotype [ 33 ].
We also observed expression of ANLN anillin actin binding protein at the Ros-L stage, suggesting that neuronal migration and neurite growth might occur by 700 linking of RhoG to the actin cytoskeleton in neural rosettes [ 34 ].
An unexpected finding was that some of the most important neural TFs exhibited heterogeneous expression within the same cell stage e. This inspired us to dissect the subpopulations of cells within each cell stage to better understand the significance of this result. To evaluate the overall distribution of ggi at each of the six stages during reprogramming and neural differentiation, we first performed an unsupervised analysis using all expressed genes QC, see Methods section as input to t-distributed bhi neighbor embedding t-SNE for visualization.
This analysis showed distinct clusters for each differentiation stage, supporting our observation of heterogeneous gene expression during these stages Fig. Because previous studies have shown that TFs and cis -regulatory elements are highly informative in reflecting cell identity [ 36 ], we used a machine classifier to determine the subsets of TFs that best clustered cells ngi putative cell populations.
As we found no remarkable differential expression of pluripotency-associated genes e. S4gwe did not include iPSCs in the following analyses. Cell heterogeneity and identification of subsets within Ros-E stage. Number of successfully profiled single cells per cell stage: Each dot represents an individual cell.
Color scheme is based on z -score distribution from —1 70003 to 2 yellow.
Fibs are a very well-adopted original somatic cell resource for iPSCs reprogramming; many direct conversions bbi fibroblast to functional neurons have been reported [ 3738 ]. Furthermore, we observed that Fibs were distributed into two distinct groups called Fib-Group1 and Fib-Group2 based on their location in Fig. Moreover, cells from the Fib2 subset clustered together with EB cells Additional file 5: Together with the molecular features of Fib2 subset Additional file 5: S5bwe proposed the Fib2 subset might possess high potential for iPSCs reprogramming and neural conversion.
S6awhich suggests that the biological processes 70033 brain development and neural differentiation initiation are occurring during the iPSCs-to-EB stage transition and that these processes are bgu by each EB subpopulation. Moreover, most neural TFs and cell-specific markers were expressed commonly among EB subpopulations e. S6band some of these TFs play a crucial role in neural tube formation.
This suggests that different subpopulations contain specific molecular signatures and different differentiation states or potentials. However, genes involved in 703 crest specifiers, such as TWIST1 [ 42 ] and SOX9which contribute to the induction and maintenance of neural stem cells and are enriched in neural crest cells [ 43—45 bgu, and ETS1which regulates neural crest development through mediating BMP signaling [ 46 ], were preferentially expressed in the Ros-E1 subpopulation Fig.
The ectoderm marker, OTX1and genes involved in the ventral hindbrain marker e. We further performed single-cell differential expression SCDE on both Ros-E subpopulations and identified additional differentially expressed genes between the two groups. A high proportion of cardiac development terms was enriched in Ros-L1, whereas DNA replication- and chromatin remodeling-related terms and pathways were significantly associated with Ros-L2.
In addition, cell-substrate adhesion-related terms and cell cycle-related pathways were enriched in Ros-L3 Additional file 7: These observations suggest that significant TF expression patterns describe discrepant cell differentiation states or differentiation commitments inside the neural conversion process.
Taken together, our results suggest that the subpopulation analyses accurately describe specific gene expression dynamics at each cell stage, which are likely masked in bulk sequencing analyses. Additionally, extrapolating from these observations, we can reason that reconstructing a differentiation trajectory based on ggi gene expression dynamics of individual subpopulations would allow us to dissect neural differentiation processes that we would otherwise be unable to observe.
Based on the subpopulations identified before, we wanted to track the gene expression dynamics 7003 individual subpopulations to parse the neural differentiation bgu and dissect the subpopulation with the highest contribution 700 commitment to the CNS lineage.
Reference SNP (refSNP) Cluster Report: rs
First, we reconstructed the differentiation trajectory using 8, genes with variable expression. This showed that cells in stages from iPSCs to NPCs followed a sequential differentiation process where each stage exhibited a relatively discriminative region with some of the subpopulations overlapping Fig. Subsequently, based on the pairwise comparisons of TF expression levels, we inferred the connection of the subpopulations from the iPSCs stage to the NPCs stage across the five-stage differentiation process Fig.
TF expression levels were considered as strong indicators of cell state and identity [ 36 ]. Here, we used the Pearson correlation coefficient to identify more biologically and molecularly similar cell subpopulations and considered them as cells within the same developmental lineage [ 49 ].
These findings were consistent with the specific gene expression pattern in individual subpopulations.
Genetic effects of a 13q31.1 microdeletion detected by noninvasive prenatal testing (NIPT)
Additionally, neural crest regulators e. S7bsuggesting that cell fate specification and differential cell status might exist even within subsets. Strikingly, Ros-E2 and Ros-L3 that were identified in the dominant path to CNS lineage by correlation analysis were shown as a process of sequential conversion in our reconstructed trajectory Fig.
The molecular signature described by these subpopulations was consistent with the analysis that identified the key contributing subpopulations and encouraged us to perform additional cell fate decision analyses. Cell fate specification revealed by reconstructed trajectory. The Pearson correlation coefficient of the two comparisons is indicated on the arrow line, respectively. Selected discriminative TFs specific to the respective branch are indicated.
The columns represent the components of branch 1, branch 2, and branch 3, respectively.