scispace - formally typeset
Search or ask a question
Posted ContentDOI

Single-Cell Multiomics Reveals Distinct Cell States at the Top of the Human Hematopoietic Hierarchy

TL;DR: In this article, a pseudotime ordering of both mRNA and chromatin data revealed a bifurcation of megakaryocyte/erythroid and lympho/myeloid trajectories immediately downstream a subpopulation with an HSC-specific enhancer signature.
Abstract: The advent of single cell (Sc) genomics has challenged the dogma of haematopoiesis as a tree-like structure of stepwise lineage commitment through distinct and increasingly restricted progenitor populations Instead, analysis of ScRNA-seq has proposed that the earliest events in human hematopoietic stem cell (HSC) differentiation are characterized by only subtle molecular changes, with hematopoietic stem and progenitor cells (HSPCs) existing as a continuum of low-primed cell-states that gradually transition into a specific lineage (CLOUD-HSPCs) Here, we combine ScRNA-seq, ScATAC-seq and cell surface proteomics to dissect the heterogeneity of CLOUD-HSPCs at different stages of human life Within CLOUD-HSPCs, pseudotime ordering of both mRNA and chromatin data revealed a bifurcation of megakaryocyte/erythroid and lympho/myeloid trajectories immediately downstream a subpopulation with an HSC-specific enhancer signature Importantly, both HSCs and lineage-restricted progenitor populations could be prospectively isolated based on correlation of their molecular signatures with CD35 and CD11A expression, respectively Moreover, we describe the changes that occur in this heterogeneity as hematopoiesis develops from neonatal to aged bone marrow, including an increase of HSCs and depletion of lympho-myeloid biased MPPs Thus, this study dissects the heterogeneity of human CLOUD-HSPCs revealing distinct HSPC-states of relevance in homeostatic settings such as ageing

Summary (4 min read)

Introduction

  • HSCs are essential for maintaining hematopoietic homeostasis and have substantial clinical relevance as the critical cellular component of stem cell transplantations.
  • Simultaneously, hematopoietic differentiation has been viewed as a marked hierarchical process where lineagerestriction is gained in a stepwise manner from HSCs through distinct progenitor populations.
  • A recent study using inheritable transcribed barcodes and ScRNA-seq showed that methods such as these readily captures the lineage relationships of cells.

Results

  • Characterizing human HSPC heterogeneity using CITE-seq analysis.
  • In contrast, the MPP-III cluster was distant from the HSC populations in the UMAP and lacked expression of HSC-associated genes, but a gain of gene expression related to myeloid and lymphoid clusters.
  • The copyright holder for this preprint (whichthis version posted April 2, 2021.
  • As with the TFs, mRNA expression levels for this HSC enhancer target signature were subsequently measured within the ScRNA-seq , HSCs together with CD35 and CD11A, which again showed enrichment of HSC-I .

Discussion

  • Traditionally, the heterogeneity of hematopoietic progenitor cells has been approached by retrospective definition of cell-type composition following functional read-outs of prospectively isolated cell fractions.
  • Interestingly, gene expression levels for HSC-specific enhancer-enriched TFs were promiscuous and enriched within HSCs, MPP as well as MLP (LMPP) populations.
  • The copyright holder for this preprint (whichthis version posted April 2, 2021.
  • Surprisingly, when comparing the gene signatures between the corresponding clusters at different stages of life, the authors observed that while most progenitor populations, including the HSCs, only experienced minor age-related changes to their gene expression program, substantial transcriptional changes were observed at the lympho-myeloid biased MPP-I and MPP-III stages.
  • Trajectory analysis allowed for prospective isolation of lineage-biased cell states within the CLOUD-HSPCs as well as the identification of an HSC-specific enhancer signature tightly linked to stemness.

Sample preparation

  • Human cord blood and bone marrow was obtained from either consenting mothers or donors and processed according to guidelines approved by Lund University.
  • The copyright holder for this preprint (whichthis version posted April 2, 2021.
  • BioRxiv preprint 17 density centrifugation was performed for 20 min at 800xg.
  • The mononuclear layer was collected, washed 1:1 with Iscove’s modified Dulbecco’s medium (IMDM, Thermofisher) and cells was either frozen down as mononuclear cells (MNCs) or processed further for CD34 enrichment.

CITE-seq sample preparation

  • Cellular indexing of transcriptomes and epitopes was performed according to previous publication [22] with minor alterations.
  • After FC block cells were stained for 30 min with antibodies for Lineage (CD14, CD16, CD19, CD2, CD3 and CD235a), CD34-FITC and the first set of CITE-seq antibodies (TableS1), cells were then washed before a second stain for 30 min with CITE-seq antibodies, sample specific Hashing antibodies (TableS1) and either a CD38PECy7 FACS antibody or a CD38-CITE-seq antibody.
  • After all libraries are ready their quality and size are measured using the bioanalyzer before being sequenced on a NEXTseq .
  • After digestion nuclei were loaded on to the 10x genomics platform and sc-ATAC-seq were performed.
  • After staining the screens were run on a FACS canto (BD).

Transplantations and in-vivo

  • All animals were processed according to Lund university ethics committee, NSG (Jackson laboratories) or NRG (Jackson laboratories) mice with ages 8-12 weeks was irradiated with 250 cGY and transplanted with the cells, positive controls of 10,000 CD34+.
  • CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • BioRxiv preprint 19 analysed for human engraftment on an LSR Fortessa (BD).
  • FACS data was then analysed in FlowJo(BD) to determine reconstitution and lineage output, the data was then visualized and analysed in graph pad prism where students t-test was used to determine significance.

CITE-seq analysis

  • Post sequencing libraries were demultiplexed using Cellranger version 3.0.2, and loaded into Seurat[45] (ver. 3.0.1).
  • Cells that did not have more than 50% of all HTO UMIs from only one HTO were discarded as doublets/multiplets.
  • Filtering criteria’s for the rest of the samples can be found in the sup.
  • CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • This subset of cells was sub-clustered (resolution=0.3) using Seurat using top 1000 HVGs and four PCA dimensions.

Pseudotime analysis CITE-seq

  • Trajectory analysis was performed using Slingshot (Bioconductor version 3.9), in short, the umap and cluster identities defined using Seurat were loaded into Slingshot.
  • By using the HSC-I cluster as the starting point and the default parameters, Slingshot defined six trajectories traversing the intermediate cell states.

Projection CB, yBM and aBM

  • The size of the yBM cells was set proportional to the mapping scores in the UMAP to indicate the degree of projection (weighted number of nearest neighbours) onto each cell.
  • CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • After cells were classified into clusters using Scarf, each cluster and sample was pseudo-bulked into three replicates.
  • ScATAC-seq analysis Barcode filtering and MACS2 peak calling.

Barcode filtering

  • Post-sequencing nuclei were processed using the cellranger-atac pipeline (version 1.2.0), after alignment to the human genome (GRCh38) and barcode identification.
  • Each barcode identified by Cellranger in the aligned reads are categorized into either cell associated barcodes or background barcodes.
  • Thereafter the minima in the gradient of this list were calculated using Numpy’s ‘gradient’ function.
  • The minima in the first 100 and beyond 10000th elements are ignored because the expected number of cells were between that range.
  • Cells were further removed if they had had either too high (log2 % > 4) or too low (log2 % < -3) percentage of reads from mitochondrial genes.

MACS2 peak calling

  • Only the fragments were recovered from the BAM by considering only those reads that aligned as proper pairs and with mapping quality (MAPQ score) higher than 20.
  • The terminal ends of the fragments were saved thus recording all the cut sites in the BAM files.
  • The sorted, replicate merged, chromosome-wise BED files were then used to identify regions of high accessibility (‘peaks’) using MACS2 software.
  • CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • The cut-sites from yBM CD38- cells were taken and recounted in the yBM CD34+ to create a cell-peak matrix of yBM CD38- in CD34+ defined feature set .

Peak filtering

  • The cell-peak matrices for each population were then processed using Scarf to perform another round of cell and feature filtering.
  • Peaks that existed in the ENCODE-defined blacklisted regions (version 2) and those from X and Y chromosomes were removed.
  • Gene Score and enhancer matrix calculation and cell filtering for yBM CD34+ cells.

UMAP and clustering:

  • The cell-peak (all peaks, except those filtered as mentioned in the ‘Peak filtering’ section) matrices were processed using Scarf (version 0.4.4).
  • Top 51 principal components were used to create a KNN graph of cells (15 neighbours).
  • The yBM cells were filtered based on nCounts (between 20000 and 175000) and nFeatures (between 10000 and 50000), additionally ENCODE blacklisted regions and autosomes were excluded as done previously for yBM CD34+ cells.
  • Top 50,000 most variable peaks which were accessible in at least 50 cells were used.

Identification of enhancer clusters

  • To create the enhancer clusters for each lineage trajectory, the cell-enhancer peak was subsetted for the cells that had a Slingshot predicted weight value of 0.75 for that trajectory.
  • The cells in this subsetted matrix were ordered based on the pseudotime values of the cells and rolling mean transformation was applied to each enhancer with a window size of 200.
  • Thereafter, standard scaling transformation was applied to the enhancer.
  • The resulting smoothened and binned matrix was subjected to hierarchical clustering using correlation metric as distance function and ward method for calculating linkage.
  • The dendrogram obtained was subjected to a straight cut with the aim of obtaining 20 clusters.

Enrichment of motifs in enhancers

  • Test of enrichment of TFBS motifs in enhancer clusters was performed using Fisher’s exact test (scipy.stats.fisher_exact).
  • The test was performed individually on each of the six lineage trajectories.
  • .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • The copyright holder for this preprint (whichthis version posted April 2, 2021.
  • BioRxiv preprint 25 Enhancers clusters from each trajectory were grouped into categories: HSC enhancers and terminal enhancers (the grouping scheme is shown in the TableS 5).

Establishing enhancer-linked genes

  • Target genes for enhancer were obtained from GeneHancer database.
  • Only those targets that had an interaction score higher than 5 were considered.
  • The expression of enhancer target genes was queried in the scRNA-Seq data.
  • The gene expression values were smoothened and imputed using MAGIC algorithm[48].
  • MAGIC (magic-impute Python package) was used with the default parameters.

Projection of CD34, CD38, CD11A and CD35 populations

  • Projection of CD34+/CD38- cell populations of yBM and aBM onto yBM CD34+ or CD38- cells was done using Scarf.
  • Same methodology (for projection and predicting cell clusters) as in the case of scRNA-Seq data was used.
  • Only the top 5 neighbours of each target cell during mapping were recorded.

HSPCs.

  • A. Schematic of experimental approach combining immunophenotypic screens and CITE-seq to link surface immunophenotype and transcriptome.
  • C. Heterogeneity of CD34+CD38- population in CB, yBM and aBM projected on CD34+ yBM UMAP.
  • .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • E. Accessibility of strict HSC enhancers within CD34+CD38clusters.
  • H. Proportion of cells within ScRNA-seq defined clusters with enhancer target gene mRNA values in the top quantile.

SP2

  • The copyright holder for this preprint (whichthis version posted April 2, 2021.
  • CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
  • The copyright holder for this preprint (whichthis version posted April 2, 2021.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

Citations
More filters
Posted ContentDOI
27 Jan 2022-bioRxiv
TL;DR: This work systematically characterized the BM stromal compartment which led to the identification of six transcriptionally and functionally distinctStromal cell populations, providing the basis for a comprehensive understanding of the cellular complexity of the human BM microenvironment and the intricate stroma- hematopoiesis crosstalk mechanisms.
Abstract: Hematopoiesis is regulated by the bone marrow (BM) stroma. However, cellular identities and functions of the different BM stromal elements in humans remain poorly defined. Based on single-cell RNA sequencing, we systematically characterized the BM stromal compartment which led to the identification of six transcriptionally and functionally distinct stromal cell populations. Stromal cell differentiation hierarchy was recapitulated based on RNA velocity analysis, in vitro proliferation capacities and differentiation potentials. Potential key factors that govern the transition from stem and progenitor cells to fate- committed cells were identified. In silico cell-cell communication prediction and in situ localization analyses demonstrated distinct hematopoietic stromal cell niches in specific BM locations, which used either the CXCL12 or SPP1 axis as the major hematopoiesis-regulating mechanism. These findings provide the basis for a comprehensive understanding of the cellular complexity of the human BM microenvironment and the intricate stroma- hematopoiesis crosstalk mechanisms, thus refining our current view on hematopoietic niche organization.

2 citations

Journal ArticleDOI
06 Mar 2023-eLife
TL;DR: Based on single-cell RNA sequencing (scRNAseq), the authors systematically characterized the human non-hematopoietic BM stroma compartment and investigated stromal cell regulation principles based on RNA velocity analysis using scVelo.
Abstract: Hematopoiesis is regulated by the bone marrow (BM) stroma. However, cellular identities and functions of the different BM stromal elements in humans remain poorly defined. Based on single-cell RNA sequencing (scRNAseq), we systematically characterized the human non-hematopoietic BM stromal compartment and we investigated stromal cell regulation principles based on the RNA velocity analysis using scVelo and studied the interactions between the human BM stromal cells and hematopoietic cells based on ligand-receptor (LR) expression using CellPhoneDB. scRNAseq led to the identification of six transcriptionally and functionally distinct stromal cell populations. Stromal cell differentiation hierarchy was recapitulated based on RNA velocity analysis and in vitro proliferation capacities and differentiation potentials. Potential key factors that might govern the transition from stem and progenitor cells to fate-committed cells were identified. In situ localization analysis demonstrated that different stromal cells were localized in different niches in the bone marrow. In silico cell-cell communication analysis further predicted that different stromal cell types might regulate hematopoiesis through distinct mechanisms. These findings provide the basis for a comprehensive understanding of the cellular complexity of the human BM microenvironment and the intricate stroma-hematopoiesis crosstalk mechanisms, thus refining our current view on human hematopoietic niche organization.

2 citations

Journal ArticleDOI
TL;DR: Motixafortide + granulocyte colony-stimulating factor (G-CSF) was shown to increase the number of CD34+ hematopoietic stem and progenitor cells (HSPC) mobilization in the GENESIS trial as discussed by the authors .
Abstract: Autologous hematopoietic stem cell transplantation (ASCT) improves survival in multiple myeloma (MM). However, many individuals are unable to collect optimal CD34+ hematopoietic stem and progenitor cell (HSPC) numbers with granulocyte colony-stimulating factor (G-CSF) mobilization. Motixafortide is a novel cyclic-peptide CXCR4 inhibitor with extended in vivo activity. The GENESIS trial was a prospective, phase 3, double-blind, placebo-controlled, multicenter study with the objective of assessing the superiority of motixafortide + G-CSF over placebo + G-CSF to mobilize HSPCs for ASCT in MM. The primary endpoint was the proportion of patients collecting ≥6 × 106 CD34+ cells kg-1 within two apheresis procedures; the secondary endpoint was to achieve this goal in one apheresis. A total of 122 adult patients with MM undergoing ASCT were enrolled at 18 sites across five countries and randomized (2:1) to motixafortide + G-CSF or placebo + G-CSF for HSPC mobilization. Motixafortide + G-CSF enabled 92.5% to successfully meet the primary endpoint versus 26.2% with placebo + G-CSF (odds ratio (OR) 53.3, 95% confidence interval (CI) 14.12-201.33, P < 0.0001). Motixafortide + G-CSF also enabled 88.8% to meet the secondary endpoint versus 9.5% with placebo + G-CSF (OR 118.0, 95% CI 25.36-549.35, P < 0.0001). Motixafortide + G-CSF was safe and well tolerated, with the most common treatment-emergent adverse events observed being transient, grade 1/2 injection site reactions (pain, 50%; erythema, 27.5%; pruritis, 21.3%). In conclusion, motixafortide + G-CSF mobilized significantly greater CD34+ HSPC numbers within two apheresis procedures versus placebo + G-CSF while preferentially mobilizing increased numbers of immunophenotypically and transcriptionally primitive HSPCs. Trial Registration: ClinicalTrials.gov , NCT03246529.

1 citations

Journal ArticleDOI
06 Mar 2023-eLife
TL;DR: Based on single-cell RNA sequencing (scRNAseq), the authors systematically characterized the human non-hematopoietic BM stroma compartment and investigated stromal cell regulation principles based on RNA velocity analysis using scVelo.
Abstract: Hematopoiesis is regulated by the bone marrow (BM) stroma. However, cellular identities and functions of the different BM stromal elements in humans remain poorly defined. Based on single-cell RNA sequencing (scRNAseq), we systematically characterized the human non-hematopoietic BM stromal compartment and we investigated stromal cell regulation principles based on the RNA velocity analysis using scVelo and studied the interactions between the human BM stromal cells and hematopoietic cells based on ligand-receptor (LR) expression using CellPhoneDB. scRNAseq led to the identification of six transcriptionally and functionally distinct stromal cell populations. Stromal cell differentiation hierarchy was recapitulated based on RNA velocity analysis and in vitro proliferation capacities and differentiation potentials. Potential key factors that might govern the transition from stem and progenitor cells to fate-committed cells were identified. In situ localization analysis demonstrated that different stromal cells were localized in different niches in the bone marrow. In silico cell-cell communication analysis further predicted that different stromal cell types might regulate hematopoiesis through distinct mechanisms. These findings provide the basis for a comprehensive understanding of the cellular complexity of the human BM microenvironment and the intricate stroma-hematopoiesis crosstalk mechanisms, thus refining our current view on human hematopoietic niche organization.

1 citations

Posted ContentDOI
07 Jan 2023-bioRxiv
TL;DR: In this article , the authors defined the sequence of transcriptional and functional events occurring during the first ex vivo division of human LT-HSCs and demonstrated that loss of long-term repopulation capacity during culture is independent of cell cycle progression.
Abstract: Loss of long-term haematopoietic stem cell function (LT-HSC) hampers the success of ex vivo HSC gene therapy and expansion procedures, but the kinetics and the mechanisms by which this occurs remain incompletely characterized. Here through time-resolved scRNA-Seq, matched in vivo functional analysis and the use of a reversible in vitro system of early G1 arrest, we define the sequence of transcriptional and functional events occurring during the first ex vivo division of human LT-HSCs. We demonstrate that contrary to current assumptions, loss of long-term repopulation capacity during culture is independent of cell cycle progression. Instead it is a rapid event that follows an early period of adaptation to culture, characterised by transient gene expression dynamics and constrained global variability in gene expression. Cell cycle progression however contributes to the establishment of differentiation programmes in culture. Our data have important implications for improving HSC gene therapy and expansion protocols.
References
More filters
Journal ArticleDOI
TL;DR: This work presents DESeq2, a method for differential analysis of count data, using shrinkage estimation for dispersions and fold changes to improve stability and interpretability of estimates, which enables a more quantitative analysis focused on the strength rather than the mere presence of differential expression.
Abstract: In comparative high-throughput sequencing assays, a fundamental task is the analysis of count data, such as read counts per gene in RNA-seq, for evidence of systematic changes across experimental conditions. Small replicate numbers, discreteness, large dynamic range and the presence of outliers require a suitable statistical approach. We present DESeq2, a method for differential analysis of count data, using shrinkage estimation for dispersions and fold changes to improve stability and interpretability of estimates. This enables a more quantitative analysis focused on the strength rather than the mere presence of differential expression. The DESeq2 package is available at http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html .

47,038 citations

Journal ArticleDOI
13 Jun 2019-Cell
TL;DR: A strategy to "anchor" diverse datasets together, enabling us to integrate single-cell measurements not only across scRNA-seq technologies, but also across different modalities.

7,892 citations

Journal ArticleDOI
TL;DR: Monocle is described, an unsupervised algorithm that increases the temporal resolution of transcriptome dynamics using single-cell RNA-Seq data collected at multiple time points that revealed switch-like changes in expression of key regulatory factors, sequential waves of gene regulation, and expression of regulators that were not known to act in differentiation.
Abstract: Defining the transcriptional dynamics of a temporal process such as cell differentiation is challenging owing to the high variability in gene expression between individual cells. Time-series gene expression analyses of bulk cells have difficulty distinguishing early and late phases of a transcriptional cascade or identifying rare subpopulations of cells, and single-cell proteomic methods rely on a priori knowledge of key distinguishing markers. Here we describe Monocle, an unsupervised algorithm that increases the temporal resolution of transcriptome dynamics using single-cell RNA-Seq data collected at multiple time points. Applied to the differentiation of primary human myoblasts, Monocle revealed switch-like changes in expression of key regulatory factors, sequential waves of gene regulation, and expression of regulators that were not known to act in differentiation. We validated some of these predicted regulators in a loss-of function screen. Monocle can in principle be used to recover single-cell gene expression kinetics from a wide array of cellular processes, including differentiation, proliferation and oncogenic transformation.

4,119 citations

Journal ArticleDOI
TL;DR: A single-cell digital gene expression profiling assay with only a single mouse blastomere is described, which detected the expression of 75% more genes than microarray techniques and identified 1,753 previously unknown splice junctions called by at least 5 reads.
Abstract: Next-generation sequencing technology is a powerful tool for transcriptome analysis. However, under certain conditions, only a small amount of material is available, which requires more sensitive techniques that can preferably be used at the single-cell level. Here we describe a single-cell digital gene expression profiling assay. Using our mRNA-Seq assay with only a single mouse blastomere, we detected the expression of 75% (5,270) more genes than microarray techniques and identified 1,753 previously unknown splice junctions called by at least 5 reads. Moreover, 8-19% of the genes with multiple known transcript isoforms expressed at least two isoforms in the same blastomere or oocyte, which unambiguously demonstrated the complexity of the transcript variants at whole-genome scale in individual cells. Finally, for Dicer1(-/-) and Ago2(-/-) (Eif2c2(-/-)) oocytes, we found that 1,696 and 1,553 genes, respectively, were abnormally upregulated compared to wild-type controls, with 619 genes in common.

2,659 citations

Journal ArticleDOI
TL;DR: In this article, a method called cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) is proposed, in which oligonucleotide-labeled antibodies are used to integrate cellular protein and transcriptome measurements into an efficient, single-cell readout.
Abstract: High-throughput single-cell RNA sequencing has transformed our understanding of complex cell populations, but it does not provide phenotypic information such as cell-surface protein levels. Here, we describe cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq), a method in which oligonucleotide-labeled antibodies are used to integrate cellular protein and transcriptome measurements into an efficient, single-cell readout. CITE-seq is compatible with existing single-cell sequencing approaches and scales readily with throughput increases.

1,904 citations

Related Papers (5)
Frequently Asked Questions (2)
Q1. What contributions have the authors mentioned in the paper "Single-cell multiomics reveals distinct cell states at the top of the human hematopoietic hierarchy" ?

In this paper, the authors proposed a method to quantify the lineage relations of cells by placing cells in a trajectory and order them along a pseudotime to investigate dynamic changes in gene expression causing lineage determination. 

Future studies will evaluate the importance of stem cell-related enhancer programs also in disease settings i. e., leukaemia, and their therapeutic potential.