Alternative platelet differentiation pathways initiated by nonhierarchically related hematopoietic stem cells.

Rare multipotent stem cells replenish millions of blood cells per second through a time-consuming process, passing through multiple stages of increasingly lineage-restricted progenitors. Although insults to the blood-forming system highlight the need for more rapid blood replenishment from stem cells, established models of hematopoiesis implicate only one mandatory differentiation pathway for each blood cell lineage. Here, we establish a nonhierarchical relationship between distinct stem cells that replenish all blood cell lineages and stem cells that replenish almost exclusively platelets, a lineage essential for hemostasis and with important roles in both the innate and adaptive immune systems. These distinct stem cells use cellularly, molecularly and functionally separate pathways for the replenishment of molecularly distinct megakaryocyte-restricted progenitors: a slower steady-state multipotent pathway and a fast-track emergency-activated platelet-restricted pathway. These findings provide a framework for enhancing platelet replenishment in settings in which slow recovery of platelets remains a major clinical challenge.

Clonally heritable gene expression imparts a layer of diversity within cell types.

Cell types can be classified according to shared patterns of transcription. Non-genetic variability among individual cells of the same type has been ascribed to stochastic transcriptional bursting and transient cell states. Using high-coverage single-cell RNA profiling, we asked whether long-term, heritable differences in gene expression can impart diversity within cells of the same type. Studying clonal human lymphocytes and mouse brain cells, we uncovered a vast diversity of heritable gene expression patterns among different clones of cells of the same type in vivo. We combined chromatin accessibility and RNA profiling on different lymphocyte clones to reveal thousands of regulatory regions exhibiting interclonal variation, which could be directly linked to interclonal variation in gene expression. Our findings identify a source of cellular diversity, which may have important implications for how cellular populations are shaped by selective processes in development, aging, and disease. A record of this paper's transparent peer review process is included in the supplemental information.

Tissue-specific RNA Polymerase II promoter-proximal pause release and burst kinetics in a Drosophila embryonic patterning network.

BACKGROUND: Formation of tissue-specific transcriptional programs underlies multicellular development, including dorsoventral (DV) patterning of the Drosophila embryo. This involves interactions between transcriptional enhancers and promoters in a chromatin context, but how the chromatin landscape influences transcription is not fully understood. RESULTS: Here we comprehensively resolve differential transcriptional and chromatin states during Drosophila DV patterning. We find that RNA Polymerase II pausing is established at DV promoters prior to zygotic genome activation (ZGA), that pausing persists irrespective of cell fate, but that release into productive elongation is tightly regulated and accompanied by tissue-specific P-TEFb recruitment. DV enhancers acquire distinct tissue-specific chromatin states through CBP-mediated histone acetylation that predict the transcriptional output of target genes, whereas promoter states are more tissue-invariant. Transcriptome-wide inference of burst kinetics in different cell types revealed that while DV genes are generally characterized by a high burst size, either burst size or frequency can differ between tissues. CONCLUSIONS: The data suggest that pausing is established by pioneer transcription factors prior to ZGA and that release from pausing is imparted by enhancer chromatin state to regulate bursting in a tissue-specific manner in the early embryo. Our results uncover how developmental patterning is orchestrated by tissue-specific bursts of transcription from Pol II primed promoters in response to enhancer regulatory cues.

Erythroid Differentiation Enhances RNA Mis-Splicing in SF3B1-Mutant Myelodysplastic Syndromes with Ring Sideroblasts.

Myelodysplastic syndromes with ring sideroblasts (MDS-RS) commonly develop from hematopoietic stem cells (HSC) bearing mutations in the splicing factor SF3B1 (SF3B1mt). Direct studies into MDS-RS pathobiology have been limited by a lack of model systems that fully recapitulate erythroid biology and RS development and the inability to isolate viable human RS. Here, we combined successful direct RS isolation from patient samples, high-throughput multiomics analysis of cells encompassing the SF3B1mt stem-erythroid continuum, and functional assays to investigate the impact of SF3B1mt on erythropoiesis and RS accumulation. The isolated RS differentiated, egressed into the blood, escaped traditional nonsense-mediated decay (NMD) mechanisms, and leveraged stress-survival pathways that hinder wild-type hematopoiesis through pathogenic GDF15 overexpression. Importantly, RS constituted a contaminant of magnetically enriched CD34+ cells, skewing bulk transcriptomic data. Mis-splicing in SF3B1mt cells was intensified by erythroid differentiation through accelerated RNA splicing and decreased NMD activity, and SF3B1mt led to truncations in several MDS-implicated genes. Finally, RNA mis-splicing induced an uncoupling of RNA and protein expression, leading to critical abnormalities in proapoptotic p53 pathway genes. Overall, this characterization of erythropoiesis in SF3B1mt RS provides a resource for studying MDS-RS and uncovers insights into the unexpectedly active biology of the "dead-end" RS. SIGNIFICANCE: Ring sideroblast isolation combined with state-of-the-art multiomics identifies survival mechanisms underlying SF3B1-mutant erythropoiesis and establishes an active role for erythroid differentiation and ring sideroblasts themselves in SF3B1-mutant myelodysplastic syndrome pathogenesis.

Spatial and single-cell profiling of the metabolome, transcriptome and epigenome of the aging mouse liver.

Tissues within an organism and even cell types within a tissue can age with different velocities. However, it is unclear whether cells of one type experience different aging trajectories within a tissue depending on their spatial location. Here, we used spatial transcriptomics in combination with single-cell ATAC-seq and RNA-seq, lipidomics and functional assays to address how cells in the male murine liver are affected by age-related changes in the microenvironment. Integration of the datasets revealed zonation-specific and age-related changes in metabolic states, the epigenome and transcriptome. The epigenome changed in a zonation-dependent manner and functionally, periportal hepatocytes were characterized by decreased mitochondrial fitness, whereas pericentral hepatocytes accumulated large lipid droplets. Together, we provide evidence that changing microenvironments within a tissue exert strong influences on their resident cells that can shape epigenetic, metabolic and phenotypic outputs.

Compound models and Pearson residuals for normalization of single-cell RNA-seq data without UMIs.

Before downstream analysis can reveal biological signals in single-cell RNA sequencing data, normalization and variance stabilization are required to remove technical noise. Recently, Pearson residuals based on negative binomial models have been suggested as an efficient normalization approach. These methods were developed for UMI-based sequencing protocols, where unique molecular identifiers (UMIs) help to remove PCR amplification noise by keeping track of the original molecules. In contrast, full-length protocols such as Smart-seq2 lack UMIs and retain amplification noise, making negative binomial models inapplicable. Here, we extend Pearson residuals to such read count data by modeling them as a compound process: we assume that the captured RNA molecules follow the negative binomial distribution, but are replicated according to an amplification distribution. Based on this model, we introduce compound Pearson residuals and show that they can be analytically obtained without explicit knowledge of the amplification distribution. Further, we demonstrate that compound Pearson residuals lead to a biologically meaningful gene selection and low-dimensional embeddings of complex Smart-seq2 datasets. Finally, we empirically study amplification distributions across several sequencing protocols, and suggest that they can be described by a broken power law. We show that the resulting compound distribution captures overdispersion and zero-inflation patterns characteristic of read count data. In summary, compound Pearson residuals provide an efficient and effective way to normalize read count data based on simple mechanistic assumptions.

PARP14 is a novel target in STAT6 mutant follicular lymphoma.

The variable clinical course of follicular lymphoma (FL) is determined by the molecular heterogeneity of tumor cells and complex interactions within the tumor microenvironment (TME). IL-4 producing follicular helper T cells (TFH) are critical components of the FL TME. Binding of IL-4 to IL-4R on FL cells activates JAK/STAT signaling. We identified STAT6 mutations (STAT6MUT) in 13% of FL (N = 33/258), all clustered within the DNA binding domain. Gene expression data and immunohistochemistry showed upregulation of IL-4/STAT6 target genes in STAT6MUT FL, including CCL17, CCL22, and FCER2 (CD23). Functionally, STAT6MUT was gain-of-function by serial replating phenotype in pre-B CFU assays. Expression of STAT6MUT enhanced IL-4 induced FCER2/CD23, CCL17 and CCL22 expression and was associated with nuclear accumulation of pSTAT6. RNA sequencing identified PARP14 -a transcriptional switch and co-activator of STAT6- among the top differentially upregulated genes in IL-4 stimulated STAT6MUT lymphoma cells and in STAT6MUT primary FL cells. Quantitative chromatin immunoprecipitation (qChIP) demonstrated binding of STAT6MUT but not STAT6WT to the PARP14 promotor. Reporter assays showed increased IL-4 induced transactivation activity of STAT6MUT at the PARP14 promotor, suggesting a self-reinforcing regulatory circuit. Knock-down of PARP14 or PARP-inhibition abrogated the STAT6MUT gain-of-function phenotype. Thus, our results identify PARP14 as a novel therapeutic target in STAT6MUT FL.

Ly6D+Siglec-H+ precursors contribute to conventional dendritic cells via a Zbtb46+Ly6D+ intermediary stage.

Plasmacytoid and conventional dendritic cells (pDC and cDC) are generated from progenitor cells in the bone marrow and commitment to pDCs or cDC subtypes may occur in earlier and later progenitor stages. Cells within the CD11c+MHCII-/loSiglec-H+CCR9lo DC precursor fraction of the mouse bone marrow generate both pDCs and cDCs. Here we investigate the heterogeneity and commitment of subsets in this compartment by single-cell transcriptomics and high-dimensional flow cytometry combined with cell fate analysis: Within the CD11c+MHCII-/loSiglec-H+CCR9lo DC precursor pool cells expressing high levels of Ly6D and lacking expression of transcription factor Zbtb46 contain CCR9loB220hi immediate pDC precursors and CCR9loB220lo (lo-lo) cells which still generate pDCs and cDCs in vitro and in vivo under steady state conditions. cDC-primed cells within the Ly6DhiZbtb46- lo-lo precursors rapidly upregulate Zbtb46 and pass through a Zbtb46+Ly6D+ intermediate stage before acquiring cDC phenotype after cell division. Type I IFN stimulation limits cDC and promotes pDC output from this precursor fraction by arresting cDC-primed cells in the Zbtb46+Ly6D+ stage preventing their expansion and differentiation into cDCs. Modulation of pDC versus cDC output from precursors by external factors may allow for adaptation of DC subset composition at later differentiation stages.

Scalable single-cell RNA sequencing from full transcripts with Smart-seq3xpress.

Current single-cell RNA sequencing (scRNA-seq) methods with high cellular throughputs sacrifice full-transcript coverage and often sensitivity. Here we describe Smart-seq3xpress, which miniaturizes and streamlines the Smart-seq3 protocol to substantially reduce reagent use and increase cellular throughput. Smart-seq3xpress analysis of peripheral blood mononuclear cells resulted in a granular atlas complete with common and rare cell types. Compared with droplet-based single-cell RNA sequencing that sequences RNA ends, the additional full-transcript coverage revealed cell-type-associated isoform variation.

Molecular spikes: a gold standard for single-cell RNA counting.

Single-cell sequencing methods rely on molecule-counting strategies to account for amplification biases, yet no experimental strategy to evaluate counting performance exists. Here, we introduce molecular spikes-RNA spike-ins containing built-in unique molecular identifiers (UMIs) that we use to identify critical experimental and computational conditions for accurate RNA counting in single-cell RNA-sequencing (scRNA-seq). Using molecular spikes, we uncovered impaired RNA counting in methods that were not informative for cellular RNA abundances due to inflated UMI counts. We further leverage molecular spikes to improve estimates of total endogenous RNA amounts in cells, and introduce a strategy to correct experiments with impaired RNA counting. The molecular spikes and the accompanying R package UMIcountR ( https://github.com/cziegenhain/UMIcountR ) will improve the validation of new methods, better estimate and adjust for cellular mRNA amounts and enable more indepth characterization of RNA counting in scRNA-seq.

Prime-seq, efficient and powerful bulk RNA sequencing.

Cost-efficient library generation by early barcoding has been central in propelling single-cell RNA sequencing. Here, we optimize and validate prime-seq, an early barcoding bulk RNA-seq method. We show that it performs equivalently to TruSeq, a standard bulk RNA-seq method, but is fourfold more cost-efficient due to almost 50-fold cheaper library costs. We also validate a direct RNA isolation step, show that intronic reads are derived from RNA, and compare cost-efficiencies of available protocols. We conclude that prime-seq is currently one of the best options to set up an early barcoding bulk RNA-seq protocol from which many labs would profit.

Transcriptional kinetics and molecular functions of long noncoding RNAs.

An increasing number of long noncoding RNAs (lncRNAs) have experimentally confirmed functions, yet little is known about their transcriptional dynamics and it is challenging to determine their regulatory effects. Here, we used allele-sensitive single-cell RNA sequencing to demonstrate that, compared to messenger RNAs, lncRNAs have twice as long duration between two transcriptional bursts. Additionally, we observed increased cell-to-cell variability in lncRNA expression due to lower frequency bursting producing larger numbers of RNA molecules. Exploiting heterogeneity in asynchronously growing cells, we identified and experimentally validated lncRNAs with cell state-specific functions involved in cell cycle progression and apoptosis. Finally, we identified cis-functioning lncRNAs and showed that knockdown of these lncRNAs modulated the nearby protein-coding gene's transcriptional burst frequency or size. In summary, we identified distinct transcriptional regulation of lncRNAs and demonstrated a role for lncRNAs in the regulation of mRNA transcriptional bursting.

BAMboozle removes genetic variation from human sequence data for open data sharing.

The risks associated with re-identification of human genetic data are severely limiting open data sharing in life sciences, even in studies where donor-related genetic variant information is not of primary interest. Here, we developed BAMboozle, a versatile tool to eliminate critical types of sensitive genetic information in human sequence data by reverting aligned reads to the genome reference sequence. Applying BAMboozle to functional genomics data, such as single-cell RNA-seq (scRNA-seq) and scATAC-seq datasets, confirmed the removal of donor-related single nucleotide polymorphisms (SNPs) and indels in a manner that did not disclose the altered positions. Importantly, BAMboozle only removes the genetic sequence variants of the sample (i.e., donor) while preserving other important aspects of the raw sequence data. For example, BAMboozled scRNA-seq data contained accurate cell-type associated gene expression signatures, splice kinetic information, and can be used for methods benchmarking. Altogether, BAMboozle efficiently removes genetic variation in aligned sequence data, which represents a step forward towards open data sharing in many areas of genomics where the genetic variant information is not of primary interest.

Transcriptional bursts explain autosomal random monoallelic expression and affect allelic imbalance.

Transcriptional bursts render substantial biological noise in cellular transcriptomes. Here, we investigated the theoretical extent of allelic expression resulting from transcriptional bursting and how it compared to the amount biallelic, monoallelic and allele-biased expression observed in single-cell RNA-sequencing (scRNA-seq) data. We found that transcriptional bursting can explain the allelic expression patterns observed in single cells, including the frequent observations of autosomal monoallelic gene expression. Importantly, we identified that the burst frequency largely determined the fraction of cells with monoallelic expression, whereas the burst size had little effect on monoallelic observations. The high consistency between the bursting model predictions and scRNA-seq observations made it possible to assess the heterogeneity of a group of cells as their deviation in allelic observations from the expected. Finally, both burst frequency and size contributed to allelic imbalance observations and reinforced that studies of allelic imbalance can be confounded from the inherent noise in transcriptional bursting. Altogether, we demonstrate that allele-level transcriptional bursting renders widespread, although predictable, amounts of monoallelic and biallelic expression in single cells and cell populations.

Non-canonical Wnt/PCP signalling regulates intestinal stem cell lineage priming towards enteroendocrine and Paneth cell fates.

A detailed understanding of intestinal stem cell (ISC) self-renewal and differentiation is required to treat chronic intestinal diseases. However, the different models of ISC lineage hierarchy1-6 and segregation7-12 are subject to debate. Here, we have discovered non-canonical Wnt/planar cell polarity (PCP)-activated ISCs that are primed towards the enteroendocrine or Paneth cell lineage. Strikingly, integration of time-resolved lineage labelling with single-cell gene expression analysis revealed that both lineages are directly recruited from ISCs via unipotent transition states, challenging the existence of formerly predicted bi- or multipotent secretory progenitors7-12. Transitory cells that mature into Paneth cells are quiescent and express both stem cell and secretory lineage genes, indicating that these cells are the previously described Lgr5+ label-retaining cells7. Finally, Wnt/PCP-activated Lgr5+ ISCs are molecularly indistinguishable from Wnt/ß-catenin-activated Lgr5+ ISCs, suggesting that lineage priming and cell-cycle exit is triggered at the post-transcriptional level by polarity cues and a switch from canonical to non-canonical Wnt/PCP signalling. Taken together, we redefine the mechanisms underlying ISC lineage hierarchy and identify the Wnt/PCP pathway as a new niche signal preceding lateral inhibition in ISC lineage priming and segregation.

Recent evolution of a TET-controlled and DPPA3/STELLA-driven pathway of passive DNA demethylation in mammals.

Genome-wide DNA demethylation is a unique feature of mammalian development and naïve pluripotent stem cells. Here, we describe a recently evolved pathway in which global hypomethylation is achieved by the coupling of active and passive demethylation. TET activity is required, albeit indirectly, for global demethylation, which mostly occurs at sites devoid of TET binding. Instead, TET-mediated active demethylation is locus-specific and necessary for activating a subset of genes, including the naïve pluripotency and germline marker Dppa3 (Stella, Pgc7). DPPA3 in turn drives large-scale passive demethylation by directly binding and displacing UHRF1 from chromatin, thereby inhibiting maintenance DNA methylation. Although unique to mammals, we show that DPPA3 alone is capable of inducing global DNA demethylation in non-mammalian species (Xenopus and medaka) despite their evolutionary divergence from mammals more than 300 million years ago. Our findings suggest that the evolution of Dppa3 facilitated the emergence of global DNA demethylation in mammals.

Single-cell RNA counting at allele and isoform resolution using Smart-seq3.

Large-scale sequencing of RNA from individual cells can reveal patterns of gene, isoform and allelic expression across cell types and states1. However, current short-read single-cell RNA-sequencing methods have limited ability to count RNAs at allele and isoform resolution, and long-read sequencing techniques lack the depth required for large-scale applications across cells2,3. Here we introduce Smart-seq3, which combines full-length transcriptome coverage with a 5' unique molecular identifier RNA counting strategy that enables in silico reconstruction of thousands of RNA molecules per cell. Of the counted and reconstructed molecules, 60% could be directly assigned to allelic origin and 30-50% to specific isoforms, and we identified substantial differences in isoform usage in different mouse strains and human cell types. Smart-seq3 greatly increased sensitivity compared to Smart-seq2, typically detecting thousands more transcripts per cell. We expect that Smart-seq3 will enable large-scale characterization of cell types and states across tissues and organisms.