Skip to main content
NCBI home page
Blood Science logoLink to Blood Science
. 2025 Sep 17;7(4):e00240. doi: 10.1097/BS9.0000000000000240

Single-cell profiling of long non-coding RNAs in the hematopoietic system

Yu Zhang a,b, Qiuju Huang c, Yawen Zhang d, Shihong Lu a,b, Lianshuo Li a,b, Shangda Yang a,b, Tao Cheng a,b,*, Hui Cheng a,b,*, Xiaowei Xie a,b,*
PMCID: PMC12445416  PMID: 40979441

Abstract

Long non-coding RNAs (lncRNAs) play a crucial role in normal and dysregulated hematopoiesis. However, the functional repertoire of lncRNAs across various hematopoietic cell types remains elusive. In this study, we constructed a comprehensive single-cell lncRNA atlas containing 207,113 cells, spanning hematopoietic stem and progenitor cells (HSPCs) to differentiated blood cells, by integrating nine single-cell RNA sequencing (scRNA-seq) datasets derived from 30 healthy donors. The hematopoietic hierarchy based on lncRNA expression was highly consistent with that based on protein-coding genes. We identified 3,463 lineage-specific lncRNAs in HSPCs, neutrophils, monocytes, B cells, and T/natural killer cells; 23 of 30 selected lncRNAs were experimentally validated. Importantly, upregulated lncRNAs in pediatric patients with B cell acute lymphoblastic leukemia, T cell acute lymphoblastic leukemia, and acute myeloid leukemia were primarily associated with oxygen response and immune regulation, indicating the potential contribution of lncRNAs to leukemogenesis. In conclusion, our results portray the landscape of lncRNAs in the hematopoietic system, revealing the functional significance of lncRNAs in both normal and abnormal hematopoiesis and providing potential therapeutic targets for the clinical treatment of leukemia.

Keywords: Hematopoietic system, Leukemia, Long noncoding RNA

1. INTRODUCTION

Long non-coding RNAs (lncRNAs) are RNA molecules that exceed 200 nucleotides in length and lack protein-coding potential.1 Accumulating evidence underscores their pivotal roles in regulating blood cell development and contributing to hematologic malignancies.2,3 For example, GATA2AS, transcribed in antisense orientation relative to GATA2, its knockout accelerates erythroid differentiation and dysregulates erythroblast gene expression.4 Similarly, overexpression of H19 promotes MCL-1 translation by targeting miR-29b-3p, which inhibits cell apoptosis and increases the resistance of multiple myeloma cells to bortezomib.5 In acute myeloid leukemia (AML), SNHG29 overexpression inhibits proliferation and reduces the drug sensitivity of FLT3-ITD AML cells.6 Notably, a signature comprising of 37 lncRNAs has shown prognostic value in pediatric AML, even though the functional roles of most of these lncRNAs remain unknown.7 Overall, these findings highlight the importance of lncRNAs in both normal and malignant hematopoiesis, necessitating further investigation into their functions and regulatory mechanisms.

Despite these advances, most human lncRNAs remain functionally uncharacterized, primarily because of their low sequence conservation, cell- and tissue-specific expression, and complex regulatory mechanisms.1,8 Single-cell RNA sequencing (scRNA-seq) allows the delineation of cellular heterogeneity at an unprecedented resolution. Studies employing scRNA-seq have demonstrated that lncRNAs can distinguish cell types and identify novel subpopulations, such as normal breast cells.9 A study using the Fluidigm C1 platform identified 3173 lncRNAs in 979 human hematopoietic stem and progenitor cells (HSPCs), revealing that lncRNA expression is co-regulated with lineage-specific protein-coding genes during early hematopoiesis.10 A study using STRT-seq showed that the hematopoietic differentiation trajectory reconstructed by lncRNAs was highly consistent with that revealed by protein-coding genes.11 However, most of these insights were gained using low-throughput methods, limiting the number of cells analyzed and hindering broad applicability.

To overcome these limitations, we used nine large-scale scRNA-seq datasets generated using the 10X Genomics platform, which is currently the most widely used technology owing to its high-throughput and cost-effectiveness. Our analysis integrated data from both public sources and our laboratory-generated datasets, capturing transcriptomes from 207,113 bone marrow and peripheral blood cells. This enabled the construction of a comprehensive lncRNA expression atlas across the human hematopoietic hierarchy. We identified 5 major cell type–specific lncRNA signatures and validated 23 representative lncRNAs in corresponding hematopoietic cell types using reverse transcription quantitative polymerase chain reaction (RT-qPCR). Additionally, we explored shared lncRNA-associated regulatory pathways in pediatric leukemia to shed light on their potential contributions to the disease. Overall, our study presents a large-scale, single-cell lncRNA resource for the human blood system and lays the groundwork for future functional studies on both normal hematopoiesis and leukemia.

2. METHODS

2.1. Data collection and raw data processing

Nine normal and 3 leukemia 10X Genomics scRNA-seq datasets were selected from public databases and our lab’s previous publications.12–21 Details of the datasets and samples are provided in Supplementary Tables S1 and S2, https://links.lww.com/BS/A121. Public raw data in SRA or BAM format were downloaded from the Gene Expression Omnibus22 and China National Center for Bioinformation.23 A custom lncRNA reference for Cell Ranger was constructed using human lncRNA annotations from NONCODE v5.24 Raw sequencing reads were aligned to both the custom lncRNA reference genome and the human GENCODE reference genome (GRCh38). The feature-barcode matrices were quantified using Cell Ranger (v8.0.1) in 10X Genomics.

2.2. Single-cell RNA-seq analysis

Computational analyses were performed using Seurat (v4.3.0)25 and Scanpy (v1.10.2).26 Doublets were removed using Scanpy. Quality control was based on the proportion of mitochondrial genes per cell, number of detected genes, number of unique molecular identifiers, and expression proportion of the top 20 genes. Automatic thresholding via median absolute deviation was employed. Cell cycle phase scores were calculated based on canonical markers for each sample. Each dataset was normalized and 2000 highly variable genes were selected for analysis. Batch effects were corrected and the datasets were integrated using BBKNN in Scanpy with 30 principal components. Unsupervised clustering was performed using the Leiden algorithm, and marker genes for each cell type were identified using scanpy.tl.rank_genes_groups. To identify signature lncRNAs for each major cell type, differential expression analysis was performed using the rank_genes_groups function in Scanpy (v1.10.2) based on the Wilcoxon rank-sum test. Significantly upregulated lncRNAs were selected for each cell type with an adjusted p value (Benjamini–Hochberg correction) <0.050 and log2 fold change >0.5.

2.3. Flow cytometry

Flow cytometry was performed according to the manufacturer’s instructions. The following Biolegend antibodies were used: hCD34 (PE/Cy7) for HSPCs, hCD3 (FITC) for T cells, hCD19 (PE) for B cells, hCD15 (APC) for granulocytes, and hCD14 (APC/Cy7) for monocytes. Cells were harvested, washed with phosphate-buffered saline (PBS) and 2% fetal bovine serum (FBS), and resuspended in the staining buffer. Antibodies were added to the cell aliquots, and the mixtures were incubated in the dark at 4°C for 30 minutes. After incubation, cells were washed and resuspended for further analysis. Samples were analyzed using a BD FACSCalibur flow cytometer (BD Biosciences, San Jose, California) and data were processed using FlowJo. The gates were set using forward scatter (FSC), side scatter (SSC) and 4′,6-diamidino-2-phenylindole (DAPI) to exclude debris and dead cells, and specific cell populations were identified using fluorescence signals for quantification.

2.4. RNA isolation, reverse transcription, and RT-qPCR

Total RNA was extracted from frozen cord blood cells and reverse transcribed into cDNA using the TransScript® All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (one-Step gDNA removal) (catalog number: AT341-01). RT-qPCR was performed using the Hieff UNICON® Universal Blue qPCR SYBR Green Master Mix. Glyceraldehyde 3-phosphate dehydrogenase served as an internal control. The delta-delta cycle threshold method was used for quantification. Primer sequences used for RT-qPCR are listed in Supplementary Table S3, https://links.lww.com/BS/A121. All human samples were collected in accordance with protocols approved by the Institutional Ethics Review Board of the Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College (approval number: KT-2019011-EC-1). Written informed consent for laboratory research was obtained from all donors or legal guardians of minors, in compliance with the Declaration of Helsinki.

2.5. Malignant cell analysis

To distinguish leukemic cells from non-leukemic cells, CopyKat27 (v1.1.0) was used to calculate the genomic copy number profiles in the pediatric B cell acute lymphoblastic leukemia (B-ALL), T cell ALL (T-ALL), and AML scRNA-seq datasets. A reference comprising 2 healthy samples was used to improve the performance of the leukemia datasets. Differentially expressed genes between malignant and normal cells were identified using the rank_genes_groups function in Scanpy (v1.10.2) with Wilcoxon rank-sum test. The comparison was performed by setting group by = “compare,” groups = [“malignant_cells”], and reference = “normal_cells.” Malignant cells were leukemic cells in leukemia patients at diagnosis, whereas normal cells included all cells from healthy control samples. Genes with an adjusted p value (Benjamini–Hochberg correction) <0.050 and an absolute log2 fold change >1 were considered significantly differentially expressed. Genes with log fold change >1 were defined as upregulated, whereas those with log fold change <−1 were defined as downregulated.

2.6. Statistics and reproducibility

Statistical analyses were performed using R (v4.4.1) and Python (v3.11.9). Specific statistical methods for scRNA-seq data are detailed in the Methods section.

3. RESULTS

3.1. Identification of cell type–specific lncRNA signatures of the human hematopoietic system

We first constructed a comprehensive human single-cell transcriptomic atlas comprising 207,113 high-quality cells by integrating 9 published scRNA-seq datasets from 30 human bone marrow and peripheral blood samples (Fig. 1A). Unsupervised clustering was performed on the datasets annotated with GENCODE-annotated genes (primarily mRNAs) and NONCODE-annotated lncRNAs. Five major cell types were identified based on canonical protein-coding gene markers: HSPCs, neutrophils, monocytes, B cells, and T/natural killer (NK) cells (Supplementary Fig. S1A and B, https://links.lww.com/BS/A121). The cell type annotations were subsequently transferred to the lncRNA atlas (Fig. 1B). Notably, hematopoietic differentiation trajectories could be reconstructed using lncRNAs alone. Among the identified cell types, HSPCs exhibited the highest lncRNA expression (Fig. 1C), which is consistent with our previous findings.11

Figure 1.

Figure 1.

Open in a new tab

Integration of single-cell lncRNA transcriptome data in hematological system. (A) Workflow of the study. (B) UMAP plot displaying an integrated cell map of 5 major cell types by NONCODE-annotated lncRNA genes. (C) Bar plot showing the average detected lncRNA number for 5 major cell types. (D) Representative lncRNAs that are highly expressed in each major cell type. Red font highlights lncRNAs that have adjacent canonical marker protein-coding genes within 5 kb distance. HSPC = hematopoietic stem and progenitor cell, lncRNA = long non-coding RNAs, NK = natural killer, UMAP = uniform manifold approximation and projection.

To explore the potential roles of lncRNAs in blood cell differentiation, we identified 3463 cell type–specific lncRNAs by comparing their expression across different cell types (Fig. 1D). Of these, 331 were identified in HSPCs, 1078 in T/NK cells, 402 in B cells, 833 in monocytes, and 819 in neutrophils. Specifically, NONHSAG031143.2 (adjacent to AVP) showed high expression in HSPCs, NONHSAG073805.1 (adjacent to CD79B) and NONHSAG016106.2 (adjacent to IGHA1) in B cells, NONHSAG009755.2 (adjacent to CD3D) in T/NK cells, and NONHSAG056897.1 (adjacent to MNDA) in neutrophils.

3.2. Functional profiling of the cell type–specific lncRNA signatures

To infer the potential functions of these cell type–specific lncRNAs, we first calculated their PhastCons conservation scores (UCSC 100-way) and compared them with the scores of other lncRNAs in the atlas. The signature lncRNAs had significantly higher conservation scores than other detected lncRNAs (Fig. 2A). Next, we performed gene ontology (GO) enrichment analysis on neighboring protein-coding genes located within 5 kb of each signature lncRNA. The enriched GO terms corresponded well with the functional identities of the associated cell types (Fig. 2B, Supplementary Fig. S2A–D, https://links.lww.com/BS/A121). Specifically, HSPC-specific lncRNAs were enriched in somatic stem cell population maintenance, negative regulation of cell differentiation, and negative regulation of cell proliferation. B cell–specific lncRNAs were associated with terms including adaptive immune response and B cell receptor signaling pathway. T/NK cell-specific lncRNAs were enriched in alpha-beta T cell activation, T cell selection, and cell killing. Monocyte-specific lncRNAs were linked to cellular responses to cytokine stimuli and inflammatory responses, whereas neutrophil-specific lncRNAs were enriched for the positive regulation of responses to external stimuli.

Figure 2.

Figure 2.

Open in a new tab

Identification of cell type–specific lncRNAs in the hematological system. (A) Signature lncRNAs show higher PhastCons conservation scores (p < 2.22e−16, Mann–Whitney U test). (B) GO enrichment analysis of protein-coding genes adjacent to HSPC-specific lncRNAs. (C) The gating strategy of HSPCs, neutrophils, monocytes, B cells, and T/NK cells for RT-qPCR. (D) RT-qPCR result verifying lncRNA expression levels in 5 major cell types. GO = gene ontology, HSPC = hematopoietic stem and progenitor cell, lncRNA = long non-coding RNA, NK = natural killer, RT-qPCR = reverse transcription quantitative polymerase chain reaction.

To further validate these findings, we selected 30 highly expressed cell type–specific lncRNAs from the 5 major cell types and assessed their expression in primary cord blood cells using flow cytometry and RT-qPCR (Fig. 2C). Among them, 23 lncRNAs were successfully validated (Fig. 2D). Collectively, these findings highlight a set of conserved, lineage-enriched lncRNAs with potential regulatory roles in hematopoietic differentiation.

3.3. Refinement of cellular subpopulations using lncRNA expression

To assess the ability of lncRNAs to resolve finer cellular heterogeneity, we independently performed sub-clustering of the 5 major cell types using lncRNAs and protein-coding genes. Protein-coding gene expression resolved HSPCs into 13 clusters, B cells into 6, T/NK cells into 10, and both monocytes and neutrophils into 5 (Supplementary Fig. S3A–J, https://links.lww.com/BS/A121). These sub-cluster annotations were then projected onto the lncRNA expression atlas, enabling the identification of sub-cluster–specific lncRNAs (Fig. 3A–J). Interestingly, clustering based solely on lncRNA expression exhibited deviations from that defined by protein-coding genes. This observation aligns with previous studies using 10X Genomics datasets from peripheral blood mononuclear cell (PBMC) and liver tissues,28 where CD4+ naïve and memory T cells were less distinguishable from CD8+ naïve and effector T cells when using lncRNA expression compared to GENCODE-annotated genes.

Figure 3.

Figure 3.

Open in a new tab

Subpopulations of hematological system based on lncRNAs. (A, C, E, G, and I) UMAP plots displaying subclustered cell map of 5 major cell types based on NONCODE-annotated lncRNA genes. (B, D, F, H, and J) Representative lncRNAs that are highly expressed in each major cell type. ATM = ataxia telangiectasia mutated, COCH = cochlin, EBM = eosinophil–basophil–mast cell, EryP = erythroid progenitor, GMP = granulocyte–monocyte progenitor, HLA = human leukocyte antigen, HSC = hematopoietic stem cell, HSPC = hematopoietic stem and progenitor cell, lncRNA = long non-coding RNA, LMPP = lymphoid-primed multipotent progenitor, MAIT = mucosal-associated invariant T cell, MD = monocyte–dendritic cell progenitor, MEP = megakaryocyte–erythroid progenitor, MLP = multi-lymphoid progenitor, MPP = multipotent progenitor, NK = natural killer, PMN = polymorphonuclear leukocyte, UMAP = uniform manifold approximation and projection.

Several factors may have contributed to this discrepancy. First, lncRNAs are generally expressed at lower levels than protein-coding genes, resulting in higher dropout rates and increased noise in clustering.29 Second, differences in biological signal strength may play a role, while protein-coding genes often serve as key markers of cell identity, and lncRNAs tend to be more cell state-specific.30 Third, the mapping strategy of Cell Ranger may also influence quantification, as it excludes reads that map to overlapping exons, potentially affecting lncRNA quantification.31 Finally, the unique expression patterns and regulatory functions of lncRNAs may further contribute to the variation in clustering outcomes.32 Together, these factors limit the resolution of closely related subpopulations when relying exclusively on lncRNA data.

3.4. Common regulatory mechanisms of lncRNAs in pediatric malignant hematopoiesis

To investigate the shared regulatory mechanisms of lncRNAs in pediatric malignant hematopoiesis, we analyzed three scRNA-seq datasets from pediatric patients diagnosed with B-ALL, T-ALL, and AML. Data from each leukemia subtype, along with 2 healthy control samples from the normal human lncRNA atlas (sample id: num13_s1, num13_s2), were integrated and leukemic cells were identified using CopyKat. As expected, normal cells were largely absent in leukemia patient samples (Fig. 4A and Supplementary Fig. S4A, D, G, https://links.lww.com/BS/A121).

Figure 4.

Figure 4.

Open in a new tab

Upregulated lncRNAs in pediatric B-ALL, T-ALL, and AML leukemic cells. (A) Identification of leukemic cells in 3 pediatric leukemia subtypes. (B) GO enrichment analysis of protein-coding genes adjacent to upregulated lncRNAs in leukemic cells. (C) Shared upregulated lncRNAs in leukemic cells of 3 pediatric leukemia subtypes compared to non-leukemic cells. (D) Expression of lncRNA NONHSAG054156.2 and NONHSAG043494.2 in 3 pediatric leukemia subtypes. AML = acute myeloid leukemia, B-ALL = B cell acute lymphoblastic leukemia, GO = gene ontology, lncRNA = long non-coding RNA, T-ALL = T cell acute lymphoblastic leukemia.

To further explore the common lncRNA-associated regulatory programs in leukemic cells, we conducted differential expression analysis between leukemic cells from patients at diagnosis and non-leukemic cells from healthy controls. GO enrichment analysis was performed on protein-coding genes located within 5 kb of the differentially expressed lncRNAs (Fig. 4B). Across all 3 leukemia subtypes, the upregulated GO biological processes based on lncRNAs included response to oxygen levels and immune response, with both B-ALL and T-ALL showing enrichment in oxidative phosphorylation pathways, consistent with previous research based on protein-coding genes.19 Furthermore, we identified 19 upregulated and 20 downregulated lncRNAs that were shared among all 3 leukemia subtypes (Fig. 4C, D), suggesting the existence of common lncRNA-mediated regulatory networks in pediatric leukemia.

Next, we examined the expression of several lncRNAs, whose functions have been reported in previous studies, in leukemic cells from all patients. Notably, LINC01013 (NONHSAG044851.3) showed high expression in the complete remission group within the scRNA-seq B-ALL dataset (Supplementary Fig. S4A–C, https://links.lww.com/BS/A121), consistent with prior findings that low LINC01013 expression is associated with early relapse and mortality in pediatric B-ALL.33 Additionally, RNA sequencing analysis of 25 pediatric T-ALL patients revealed that LINC01221 was significantly upregulated in T-ALL patients compared to healthy controls,34 and a similar pattern was observed in our scRNA-seq T-ALL dataset (Supplementary Fig. S4D–F, https://links.lww.com/BS/A121).

Our findings provide novel insights into the shared and distinct roles of lncRNAs in pediatric leukemia. By identifying commonly dysregulated lncRNAs across B-ALL, T-ALL, and AML, we identified key candidates that may contribute to leukemogenesis and serve as potential biomarkers or therapeutic targets. The enrichment of these lncRNAs in pathways such as oxidative phosphorylation and oxygen response suggests their involvement in metabolic reprogramming and hypoxic adaptation in leukemia cells. Moreover, validation of previously reported lncRNAs strengthens the reliability of our approach, and the discovery of novel differentially expressed lncRNAs provides a foundation for future functional studies. Collectively, these findings enhance our understanding of lncRNA-mediated regulation of pediatric malignant hematopoiesis and offer new directions for biomarker discovery and targeted therapies.

4. DISCUSSION

Owing to their high specificity, different lncRNAs may play distinct roles across different hematopoietic lineages, making them suitable for single-cell resolution studies. Recent advances in sequencing technologies35,36 and alignment algorithms37 have significantly improved our ability to detect and quantify lncRNA expression. Although previous studies using STRT-seq and the Fluidigm C1 system have provided valuable insights, their limited throughput and high cost restrict their broader application.

In contrast, our use of the 10X Genomics platform enabled high-throughput profiling of lncRNA expression across a large number of hematopoietic cells. This approach allowed us to identify lineage-specific lncRNA signatures and explore their potential regulatory roles in hematopoiesis. Notably, similar to the findings of STRT-seq-based studies,11 we observed that lncRNA expression patterns could be used to reconstruct hematopoietic differentiation trajectories, echoing the behavior of protein-coding genes. Among all cell populations, HSPCs exhibited the highest lncRNA diversity.

Despite these advances, lncRNAs showed somewhat different performance compared to protein-coding genes in resolving subtle cellular subpopulations likely because of their generally lower expression levels, increased technical noise, and intrinsic cell state specificity. This suggests that although lncRNAs contribute to broad lineage specification, their role in defining finer subpopulations requires further investigation.

Given the high cost and complexity of experimental validation, it is essential to prioritize candidate lncRNAs for functional studies. In this study, we employed a “guilt-by-association” approach to infer potential functions by examining nearby protein-coding genes. Functional enrichment analysis revealed that these neighboring genes were often involved in lineage-relevant biological processes, supporting the idea that adjacent lncRNAs may participate in related regulatory pathways.

Beyond their roles in normal hematopoiesis, we identified a set of commonly dysregulated lncRNAs in pediatric leukemia subtypes, including B-ALL, T-ALL, and AML. These lncRNAs were associated with key biological processes such as oxygen response and immune regulation. Notably, the differential expression of LINC01013 and LINC01221 in leukemia subtypes aligned with previous findings, suggesting their potential as disease biomarkers or therapeutic targets. These findings underscore the broad relevance of lncRNAs in leukemogenesis.

However, several challenges remain in the study of lncRNAs at the single-cell level. First, lncRNAs typically have lower expression levels than protein-coding genes, making their detection and quantification more challenging. While the 10X Genomics platform provides high cell throughput, its relatively low sequencing depth may limit the capture of low-abundance lncRNAs and may not effectively detect the expression of non-polyadenylated lncRNAs. Future studies using random-primed single-cell or single-nucleus total RNA sequencing technologies (eg, VASA-seq, SUPeR-seq) or full-length poly(A)-based methods, such as Smart-seq3, as well as long-read sequencing platforms (Oxford Nanopore, Oxford, UK; PacBio, Menlo Park, California ) could provide higher sensitivity and broader transcript coverage for lncRNA detection, including non-polyadenylated transcripts.

Additionally, our functional predictions relied on a guilt-by-association approach, which inferred lncRNA functions based on co-expression with neighboring protein-coding genes. Although this method provides valuable insights, it only suggests potential functional relationships and requires further experimental validation. Future studies using clustered regularly interspaced short palindromic repeats (CRISPR)-based perturbation screens or RNA immunoprecipitation may provide more direct evidence of the function of lncRNAs.

In summary, our study provides the most extensive lncRNA atlas of the human hematopoietic system to date, highlighting its role in normal differentiation and leukemogenesis. Although our findings advance our understanding of lncRNA functions at the single-cell level, further experimental and multi-omics studies are required to fully elucidate their regulatory mechanisms. This resource serves as a valuable foundation for future research on the role of lncRNAs in hematopoiesis and related diseases.

ACKNOWLEDGMENTS

This work was supported by the National Key Research and Development Program of China (2021YFA1100900 and 2023YFF1204700), the National Natural Science Foundation of China (92368202), the Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (2021-I2M-1-040), the CAMS Fundamental Research Funds for Central Research Institutes (3332021093), the Haihe Laboratory of Cell Ecosystem Innovation Fund (HH23KYZX0004).

ETHICAL APPROVAL

All human studies were approved by the Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College. Bone marrow and blood sample collection and processing and laboratory operations were approved by the Ethics Committee of the Institute of Hematology and Blood Disease Hospital.

AUTHOR CONTRIBUTIONS

T.C., H.C., and X.X. supervised the project. Y.Z. performed the bioinformatics analysis. Q.H. and Y.Z. carried out the experiments. S.L., LL, S.Y., X.X., and Y.Z. wrote the manuscript. All authors discussed the results and contributed to manuscript revisions. The order of the co-first authors’ names reflects their respective contributions to the study.

Supplementary Material

bs9-7-e00240-s001.pdf (1.3MB, pdf)
Open in a new tab

Footnotes

Conflict of interest: The authors declare that they have no conflict of interest.

The single-cell data in this study have been deposited in the National Genomics Data Center with accession code HRA011078. Supporting data relevant to the main manuscript and supplementary materials are available in the Supporting Data Values file.

Y.Z. and Q.H. are co-first authors and contributed equally to this work.

This work was supported by the National Key Research and Development Program of China (2021YFA1100900 and 2023YFF1204700), the National Natural Science Foundation of China (92368202), the Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (2021-I2M-1-040), the CAMS Fundamental Research Funds for Central Research Institutes (3332021093), the Haihe Laboratory of Cell Ecosystem Innovation Fund (HH23KYZX0004).

REFERENCES

  • [1].Mattick JS, Amaral PP, Carninci P, et al. Long non-coding RNAs: definitions, functions, challenges and recommendations. Nat Rev Mol Cell Biol 2023;24(6):430–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Alvarez-Dominguez JR, Lodish HF. Emerging mechanisms of long noncoding RNA function during normal and malignant hematopoiesis. Blood 2017;130(18):1965–1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Qiu Y, Xu M, Huang S. Long noncoding RNAs: emerging regulators of normal and malignant hematopoiesis. Blood 2021;138(23):2327–2336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Liu G, Kim J, Nguyen N, Zhou L, Dean A. Long noncoding RNA GATA2AS influences human erythropoiesis by transcription factor and chromatin landscape modulation. Blood 2024;143(22):2300–2313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Pan Y, Zhang Y, Liu W, et al. LncRNA H19 overexpression induces bortezomib resistance in multiple myeloma by targeting MCL-1 via miR-29b-3p. Cell Death Dis 2019;10(2):106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Liu S, Zhou J, Ye X, et al. A novel lncRNA SNHG29 regulates EP300-related histone acetylation modification and inhibits FLT3-ITD AML development. Leukemia 2023;37(7):1421–1434. [DOI] [PubMed] [Google Scholar]
  • [7].Farrar JE, Smith JL, Othus M, et al. Long noncoding RNA expression independently predicts outcome in pediatric acute myeloid leukemia. J Clin Oncol 2023;41(16):2949–2962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Uszczynska-Ratajczak B, Lagarde J, Frankish A, Guigó R, Johnson R. Towards a complete map of the human long non-coding RNA transcriptome. Nat Rev Genet 2018;19(9):535–548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Bitar M, Rivera IS, Almeida I, et al. Redefining normal breast cell populations using long noncoding RNAs. Nucleic Acids Res 2023;51(12):6389–6410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Wu Z, Gao S, Zhao X, et al. Long noncoding RNAs of single hematopoietic stem and progenitor cells in healthy and dysplastic human bone marrow. Haematologica 2019;104(5):894–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Xie X, Liu M, Zhang Y, et al. Single-cell transcriptomic landscape of human blood cells. Natl Sci Rev 2021;8(3):nwaa180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Wang H, Liu C, Xie X, et al. Multi-omics blood atlas reveals unique features of immune and platelet responses to SARS-CoV-2 Omicron breakthrough infection. Immunity 2023;56(6):1410–1428.e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Gao S, Wu Z, Kannan J, et al. Comparative transcriptomic analysis of the hematopoietic system between human and mouse by single cell RNA sequencing. Cells 2021;10(5):973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Hua P, Roy N, De La Fuente J, et al. Single-cell analysis of bone marrow–derived CD34+ cells from children with sickle cell disease and thalassemia. Blood 2019;134(23):2111–2115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Rodríguez A, Zhang K, Färkkilä A, et al. MYC promotes bone marrow stem cell dysfunction in Fanconi Anemia. Cell Stem Cell 2021;28(1):33–47.e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Wang S, Zhang Q, Hui H, Agrawal K, Karris MAY, Rana TM. An atlas of immune cell exhaustion in HIV-infected individuals revealed by single-cell transcriptomics. Emerg Microbes Infect 2020;9(1):2333–2347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Roy A, Wang G, Iskander D, et al. Transitions in lineage specification and gene regulatory networks in hematopoietic stem/progenitor cells over human development. Cell Rep 2021;36(11):109698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Zhao J, Zhang S, Liu Y, et al. Single-cell RNA sequencing reveals the heterogeneity of liver-resident immune cells in human. Cell Discov 2020;6(1):22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Zhang Y, Wang S, Zhang J, et al. Elucidating minimal residual disease of paediatric B-cell acute lymphoblastic leukaemia by single-cell analysis. Nat Cell Biol 2022;24(2):242–252. [DOI] [PubMed] [Google Scholar]
  • [20].Zhang J, Duan Y, Wu P, et al. Clonal evolution dissection reveals high MSI2 level promotes chemo-resistance in T-cell acute lymphoblastic leukemia. Blood 2023;142(Supplement 1):846–846. [DOI] [PubMed] [Google Scholar]
  • [21].Bailur JK, McCachren SS, Pendleton K, et al. Risk-associated alterations in marrow T cells in pediatric leukemia. JCI Insight 2020;5(16):e140179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional genomics data sets—update. Nucleic Acids Res 2012;41(D1):D991–D995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Bai X, Bao Y, et al. Database Resources of the National Genomics Data Center, China National Center for Bioinformation in 2024. Nucleic Acids Res 2024;52(D1):D18–D32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Fang S, Zhang L, Guo J, et al. NONCODEV5: a comprehensive annotation database for long non-coding RNAs. Nucleic Acids Res 2018;46(D1):D308–D314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Hao Y, Hao S, Andersen-Nissen E, et al. Integrated analysis of multimodal single-cell data. Cell 2021;184(13):3573–3587.e29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Wolf FA, Angerer P, Theis FJ. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol 2018;19(1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Gao R, Bai S, Henderson YC, et al. Delineating copy number and clonal substructure in human tumors from single-cell transcriptomes. Nat Biotechnol 2021;39(5):599–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Rahman RU, Ahmad I, Sparks R, Saad AB, Mullen A. Singletrome: a method to analyze and enhance the transcriptome with long noncoding RNAs for single cell analysis. bioRxiv 2022:2022.10.31.514182. [Google Scholar]
  • [29].Cabili MN, Trapnell C, Goff L, et al. Integrative annotation of human large intergenic noncoding RNAs reveals global properties and specific subclasses. Genes Dev 2011;25(18):1915–1927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Lorenzi L, Chiu HS, Avila Cobos F, et al. The RNA Atlas expands the catalog of human non-coding RNAs. Nat Biotechnol 2021;39(11):1453–1465. [DOI] [PubMed] [Google Scholar]
  • [31].Zheng GXY, Terry JM, Belgrader P, et al. Massively parallel digital transcriptional profiling of single cells. Nat Commun 2017;8(1):14049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Statello L, Guo CJ, Chen LL, Huarte M. Gene regulation by long non-coding RNAs and its biological functions. Nat Rev Mol Cell Biol 2021;22(2):96–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Bárcenas-López DA, Núñez-Enríquez JC, Hidalgo-Miranda A, et al. Transcriptome analysis identifies LINC00152 as a biomarker of early relapse and mortality in acute lymphoblastic leukemia. Genes 2020;11(3):302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Sharma P, Kaur P, Bhatia P, Trehan A, Sreedharanunni S, Singh M. Novel lncRNAs LINC01221, RP11-472G21.2 and CRNDE are markers of differential expression in pediatric patients with T cell acute lymphoblastic leukemia. Cancer Cell Int 2024;24(1):65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Salmen F, De Jonghe J, Kaminski TS, et al. High-throughput total RNA sequencing in single cells using VASA-seq. Nat Biotechnol 2022;40(12):1780–1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Sheng K, Cao W, Niu Y, Deng Q, Zong C. Effective detection of variation in single-cell transcriptomes using MATQ-seq. Nat Methods 2017;14(3):267–270. [DOI] [PubMed] [Google Scholar]
  • [37].Goñi E, Mas AM, Gonzalez J, et al. Uncovering functional lncRNAs by scRNA-seq with ELATUS. Nat Commun 2024;15(1):9709. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Blood Science are provided here courtesy of Wolters Kluwer Health

RESOURCES