The mechanisms underlying haematopoietic lineage decisions remain disputed. Lineage-affiliated transcription factors1,2 with the capacity for lineage reprogramming3, positive auto-regulation4,5 and mutual inhibition6,7 have been described as being expressed in uncommitted cell populations8. This led to the assumption that lineage choice is cell-intrinsically initiated and determined by stochastic switches of randomly fluctuating cross-antagonistic transcription factors3. However, this hypothesis was developed on the basis of RNA expression data from snapshot and/or population-averaged analyses9,10,11,12. Alternative models of lineage choice therefore cannot be excluded. Here we use novel reporter mouse lines and live imaging for continuous single-cell long-term quantification of the transcription factors GATA1 and PU.1 (also known as SPI1). We analyse individual haematopoietic stem cells throughout differentiation into megakaryocytic–erythroid and granulocytic–monocytic lineages. The observed expression dynamics are incompatible with the assumption that stochastic switching between PU.1 and GATA1 precedes and initiates megakaryocytic–erythroid versus granulocytic–monocytic lineage decision-making. Rather, our findings suggest that these transcription factors are only executing and reinforcing lineage choice once made. These results challenge the current prevailing model of early myeloid lineage choice.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Pevny, L. et al. Erythroid differentiation in chimaeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1. Nature 349, 257–260 (1991)
Scott, E. W., Simon, M. C., Anastasi, J. & Singh, H. Requirement of transcription factor PU.1 in the development of multiple hematopoietic lineages. Science 265, 1573–1577 (1994)
Graf, T. & Enver, T. Forcing cells to change lineages. Nature 462, 587–594 (2009)
Tsai, S. F., Strauss, E. & Orkin, S. H. Functional analysis and in vivo footprinting implicate the erythroid transcription factor GATA-1 as a positive regulator of its own promoter. Genes Dev. 5, 919–931 (1991)
Chen, H. et al. PU.1 (Spi-1) autoregulates its expression in myeloid cells. Oncogene 11, 1549–1560 (1995)
Zhang, P. et al. PU.1 inhibits GATA-1 function and erythroid differentiation by blocking GATA-1 DNA binding. Blood 96, 2641–2648 (2000)
Nerlov, C., Querfurth, E., Kulessa, H. & Graf, T. GATA-1 interacts with the myeloid PU.1 transcription factor and represses PU.1-dependent transcription. Blood 95, 2543–2551 (2000)
Miyamoto, T. et al. Myeloid or lymphoid promiscuity as a critical step in hematopoietic lineage commitment. Dev. Cell 3, 137–147 (2002)
Huang, S., Guo, Y.-P., May, G. & Enver, T. Bifurcation dynamics in lineage-commitment in bipotent progenitor cells. Dev. Biol. 305, 695–713 (2007)
Akashi, K. et al. Transcriptional accessibility for genes of multiple tissues and hematopoietic lineages is hierarchically controlled during early hematopoiesis. Blood 101, 383–389 (2003)
Månsson, R. et al. Molecular evidence for hierarchical transcriptional lineage priming in fetal and adult stem cells and multipotent progenitors. Immunity 26, 407–419 (2007)
Hoppe, P. S., Coutu, D. L. & Schroeder, T. Single-cell technologies sharpen up mammalian stem cell research. Nat. Cell Biol. 16, 919–927 (2014)
Akashi, K., Traver, D., Miyamoto, T. & Weissman, I. L. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404, 193–197 (2000)
Orkin, S. H. & Zon, L. I. Hematopoiesis: an evolving paradigm for stem cell biology. Cell 132, 631–644 (2008)
Etzrodt, M., Endele, M. & Schroeder, T. Quantitative single-cell approaches to stem cell research. Cell Stem Cell 15, 546–558 (2014)
Kueh, H. Y., Champhekar, A., Nutt, S. L., Elowitz, M. B. & Rothenberg, E. V. Positive feedback between PU.1 and the cell cycle controls myeloid differentiation. Science 341, 670–673 (2013)
Kirstetter, P., Anderson, K., Porse, B. T., Jacobsen, S. E. W. & Nerlov, C. Activation of the canonical Wnt pathway leads to loss of hematopoietic stem cell repopulation and multilineage differentiation block. Nat. Immunol. 7, 1048–1056 (2006)
Rosenbauer, F. et al. Acute myeloid leukemia induced by graded reduction of a lineage-specific transcription factor, PU.1. Nat. Genet. 36, 624–630 (2004)
Moreau-Gachelin, F. et al. Spi-1/PU.1 transgenic mice develop multistep erythroleukemias. Mol. Cell. Biol. 16, 2453–2463 (1996)
Heyworth, C., Pearson, S., May, G. & Enver, T. Transcription factor-mediated lineage switching reveals plasticity in primary committed progenitor cells. EMBO J. 21, 3770–3781 (2002)
Filipczyk, A. et al. Network plasticity of pluripotency transcription factors in embryonic stem cells. Nat. Cell Biol. 17, 1235–1246 (2015)
Pronk, C. J. H. et al. Elucidation of the phenotypic, functional, and molecular topography of a myeloerythroid progenitor cell hierarchy. Cell Stem Cell 1, 428–442 (2007)
Arinobu, Y. et al. Reciprocal activation of GATA-1 and PU.1 marks initial specification of hematopoietic stem cells into myeloerythroid and myelolymphoid lineages. Cell Stem Cell 1, 416–427 (2007)
Iwasaki, H. et al. Identification of eosinophil lineage-committed progenitors in the murine bone marrow. J. Exp. Med. 201, 1891–1897 (2005)
Miyawaki, K. et al. CD41 marks the initial myelo-erythroid lineage specification in adult mouse hematopoiesis: redefinition of murine common myeloid progenitor. Stem Cells 33, 976–987 (2015)
Pietras, E. M. et al. Functionally distinct subsets of lineage-biased multipotent progenitors control blood production in normal and regenerative conditions. Cell Stem Cell 17, 35–46 (2015)
Rieger, M. A., Hoppe, P. S., Smejkal, B. M., Eitelhuber, A. C. & Schroeder, T. Hematopoietic cytokines can instruct lineage choice. Science 325, 217–218 (2009)
Eilken, H. M., Nishikawa, S. & Schroeder, T. Continuous single-cell imaging of blood generation from haemogenic endothelium. Nature 457, 896–900 (2009)
Hilsenbeck, O. et al. Software tools for single-cell tracking and quantification of cellular and molecular properties. Nat. Biotechnol. http://dx.doi.org/10.1038/nbt.3626 (2016)
Mancini, E. et al. FOG-1 and GATA-1 act sequentially to specify definitive megakaryocytic and erythroid progenitors. EMBO J. 31, 351–365 (2012)
Hu, M. et al. Multilineage gene expression precedes commitment in the hemopoietic system. Genes Dev. 11, 774–785 (1997)
Liu, P., Jenkins, N. A. & Copeland, N. G. A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res. 13, 476–484 (2003)
Shaner, N. C. et al. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22, 1567–1572 (2004)
Dymecki, S. M. Flp recombinase promotes site-specific DNA recombination in embryonic stem cells and transgenic mice. Proc. Natl Acad. Sci. USA 93, 6191–6196 (1996)
Kiel, M. J. et al. SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 121, 1109–1121 (2005)
Wilson, A. et al. Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair. Cell 135, 1118–1129 (2008)
Schwarzfischer, M. et al. Efficient fluorescence image normalization for time lapse movies. Proc. Microsc. Image Anal. with Appl. Biol. (2011)
Kreutz, C. et al. An error model for protein quantification. Bioinformatics 23, 2747–2753 (2007)
We are grateful to S. Ammersdoerfer, H. Oller, C. Raithel, B. Vogel and A. Ziegler for technical support. This work was supported by an EMBO long-term fellowship to M.Et., the ‘EUCOMM: Tools for Functional Annotation of the Mouse Genome’ (EUCOMMTOOLS) project (FP7-HEALTH-F4-2010-261492) to A.B., the ERC (starting grant Latent Causes), the German Federal Ministry of Education and Research (BMBF), the German Research Foundation (DFG) within the SPPs 1395 (InKoMBio) and 1356 to F.J.T., and DFG SFB 684 to T.S. and the SNF to T.S. T.S. and O.H. acknowledge financial support from SystemsX.ch.
The authors declare no competing financial interests.
Extended data figures and tables
Extended Data Figure 1 Knock-in of fluorescent proteins does not alter the composition of fetal livers and adult peripheral blood.
a, Gata1-mCherry knock-in strategy. BamHI and XbaI were chosen to generate restriction fragment length polymorphisms (from 11.1 kbp to 5.7 kbp in the case of XbaI and from 9.9 kbp to 11.1 kbp in the case of BamHI) to screen for successful knock-ins. Genomic sequences for Southern probes were identified at indicated positions. The final knock-in construct contained a 5.0 kilo base pairs (kbp) long 5′ homology arm until the last codon of Gata1, a short linker sequence (5'-AGAGCATCAGGTACCAGTGGAGCT-3'), the open reading frame (ORF) of mCherry, a FRT-flanked Neomycin-resistance cassette (including a eukaryotic and a prokaryotic promoter and a polyadenylation signal) and a 4.6 kbp long 3′ homology arm. b, Fetal livers of E14.5 embryos were collected, subjected to Ficoll-density centrifugation, pooled (C57Bl/6, 7 fetal livers; GATA1–mCherry, 6 fetal livers) and analysed by flow cytometry. Shown are percentages of the parental gate. c, Peripheral blood counts of adult mice (C57Bl/6 n = 6 biological replicates; PU.1eYFPGATA1mCherry n = 9; error bars, mean + s.d.). wbc, white blood cells (200 cells per mm3); rbc, red blood cells (4 × 105 cells per mm3); plt, platelets (20 cells per mm3); hgb, haemoglobin (0.4 g dl−1); hct, haematocrit (%); mcv, mean corpuscular volume (μm3); mch, mean corpuscular haemoglobin (0.4 pg); mchc, mean corpuscular haemoglobin concentration (g dl−1); rdw, red cell distribution width (0.4%); mpv, mean platelet volume (0.2 μm3); lypro, % lymphocytes of wbc (2%); mopro, percentage monocytes of wbc (0.1%); grpro, percentage of granulocytes of wbc (%); eopro, percentage eosinophils of wbc (0.2%). No significant difference, one-way MANOVA (P > 0.09).
Extended Data Figure 2 MegE versus GM-lineage differentiation of GATA1–mCherry/PU.1–eYFP cells in vivo is unaltered upon competitive transplantation.
a, 106 bone marrow cells each from CD45.1 C57Bl/6 and CD45.2 C57Bl/6 or from CD45.1 C57Bl/6 and CD45.2 PU.1eYFPGATA1mCherry mice were transplanted into lethally irradiated recipient mice and bone marrow progenitor cell composition was analysed after 6–7 weeks. X and Y (X.Y, e.g. 1.1) denote donor pair (X) and recipient mouse (Y). MkP, megakaryocyte progenitor; proEry, proerythroblast; preCFU-E, pre-colony-forming unit erythrocyte; preMegE, premegakaryocyte-erythrocyte progenitor; preGM, pregranulocyte–macrophage progenitor. Asterisk (*) indicates data excluded owing to low donor contribution. b, Summarized bone marrow lineage contribution per donor pair (from a).
Extended Data Figure 3 Normal reprogramming capacity of PU.1–eYFP and GATA1–mCherry in both wild-type and PU.1eYFPGATA1mCherry cells.
a, b, LSK CD34+Flt3+ were sorted, transduced with lentivirus expressing the indicated proteins and plated in methylcellulose under permissive conditions. Cells from C57Bl/6 mice (a) (3 independent experiments) and cells from PU.1–eYFP knock-in mice (n = 2; except tdTomato–GATA1 n = 1) (b) were used. c, d, PreMegE cells (c) or preGM cells (d) from both C57Bl/6 wild-type and PU.1eYFPGATA1mCherry (PU/GATA) knock-in mice were sorted and transduced with mock, PU.1 or PU.1–eYFP expressing lentivirus (c) or with mock, GATA1 or GATA1–mCherry expressing lentivirus (d), respectively. After 24 h, cells were seeded in methylcellulose under permissive conditions. Colonies were scored after 8–10 days of culture (3 independent experiments; except for C57Bl/6, PU/GATA, C57Bl/6 + PU.1 and PU/GATA + PU.1 n = 4). There was no significant difference between PU.1 and PU.1–eYFP overexpression in either C57Bl/6 or PU/GATA cells (P > 0.23; Kruskal–Wallis test). In contrast, C57Bl/6 and PU/GATA colonies without PU.1–eYFP overexpression were significantly different (P < 0.05 for M, Meg and E) (c). There was no significant difference between GATA1 and GATA1–mCherry overexpression in either C57Bl/6 or PU/GATA cells (P > 0.77; Kruskal–Wallis test). In contrast, C57Bl/6 and PU/GATA colonies without GATA1–mCherry overexpression were significantly different (P < 0.007, MegE) (d). Data are mean values. GEMMeg, granulocytic, erythroid, monocytic, megakaryocytic; MegE, megakaryocytic–erythroid; Meg, megakaryocytic; E, erythroid; GM, granulocytic–monocytic; M, monocytic; G, granulocytic.
Extended Data Figure 4 PU.1 is expressed in nuclei of all HSCs, MPPs, GMPs and MEPs of wild-type and PU.1eYFPGATA1mCherry mice, and overlaps with eYFP expression.
a, b, Indicated cell populations from wild-type C57Bl/6 (a) and PU.1eYFPGATA1mCherry (b) mice were sorted, fixed and probed with PU.1 antibody followed by staining with secondary antibody. Representative examples from controls (‘secondary’, without primary antibody) and staining of GMPs, MPPs, HSCs, and MEPs (CD150+ progenitors) are shown. DAPI stains nuclei. Scale bar, 10 μm. c, d, Quantifications of relative PU.1 expression levels determined by pixel intensities (rel. intens.). Data includes samples from three independent experiments (biological replicates), each of which was normalized to the mean expression levels of the respective MEP population. Individual data points for c are 884 MEP secondary only (475, 148 and 261 data points from the individual experiments), 1,218 MPP secondary only (553, 260 and 405), 755 GMP secondary only (599, 122 and 34), 1,213 MEP (659, 371 and 183), 603 HSCs (360, 194 and 49), 1,458 MPP (749, 283 and 426) and 819 GMP (571, 183 and 65). Individual data for d points are 1,530 MEP secondary only (739, 449 and 342 data points from the individual experiments), 1,194 MPP secondary only (547, 394 and 253), 1,866 GMP secondary only (1,616, 126 and 124), 1,521 MEP (518, 581 and 422), 273 HSCs (116, 79 and 78), 1,531 MPP (616, 463 and 452) and 2,339 GMP (1,351, 673 and 315). e, Correlation plot of PU.1 and eYFP staining in the indicated cell populations from one experiment. Pixel intensities were normalized to the mean expression in MEPs. Pearson correlation coefficients are displayed. Mean Pearson correlation values (± s.d.) of all three independent experiments were 0.82 ± 0.04 (MEP), 0.69 ± 0.10 (HSC), 0.71 ± 0.06 (MPP) and 0.63 ± 0.08 (GMP).
Extended Data Figure 5 GATA1 is expressed in nuclei of MegE committed cells of wild-type and PU.1eYFPGATA1mCherry mice, and overlaps with mCherry expression.
a, b, Indicated cell populations from wild-type C57Bl/6 (a) and PU.1eYFPGATA1mCherry (b) mice were sorted, fixed and probed with GATA1 (a, b) and mCherry (b) antibody followed by staining with secondary antibodies. Representative examples from controls (‘secondary’, without primary antibody) and staining of GMP, LSK and MEP (CD150+ progenitors) are shown. DAPI stains nuclei. Scale bar, 10 μm. c, d, Quantifications of relative GATA1 expression levels determined by pixel intensities. Data includes samples from three independent experiments (biological replicates), each of which was normalized to the mean expression levels of the respective MEP population. Individual data points for c are 292 GMP secondary only (56, 188 and 48 data points from the individual experiments), 698 LSK secondary only (155, 287 and 256), 563 MEP secondary only (308, 216 and 39), 344 GMP (64, 158 and 122), 1167 LSK (394, 294 and 479) and 590 MEP (252, 155 and 183). Individual data points for d are 485 GMP secondary only (171, 173 and 141 data points from the individual experiments), 1,295 LSK secondary only (360, 552 and 383), 886 MEP secondary only (561, 203 and 122), 462 GMP (73, 114 and 275), 1,184 LSK (252, 441 and 491) and 865 MEP (632, 115 and 118). e, Correlation plot of GATA1 and mCherry staining in the indicated cell populations from one experiment. Pixel intensities were normalized to the mean expression in MEPs. The mean Pearson correlation value of all three independent experiments was 0.87 ± 0.06 (s.d.). f, Representative example of GATA1–mCherry expression in MPP1–4 from three independent experiments (biological replicates). GATA1–mCherry is also expressed in some MPP2 cells36. These are already committed to the MegE lineage (data not shown). Upper panels PU.1eYFP mouse (negative control), lower panels PU.1eYFPGATA1mCherry mouse.
Extended Data Figure 6 CD16/32 and GATA1–mCherry expression can be used as a lineage-marker for the GM and MegE lineage, respectively.
a, Overlapping PU.1–eYFP expression histograms of MEPs, LSKs and GMPs show that PU.1–eYFP levels alone are not sufficient to attribute individual cells to a specific HSPC population. Mean ± s.d.; n = 4 (biological replicates); one representative example shown. b, Flow cytometry of day 4 HSC culture. Representative example from three independent experiments (biological replicates). GATA1–mCherry and CD16/32 are mutually exclusive. c, Single HSCs were sorted into single wells of a 384-well plate. Colonies were observed for 10 days by one brightfield image per day. Expression of PU.1–eYFP, GATA1–mCherry, and CD16/32 was qualitatively assessed on each day. Colonies were scored into exclusive GATA1–mCherry+ (white bars), exclusive CD16/32+, or GATA1–mCherry+ and CD16/32 double positive (DP) colonies (57%). DP colonies were further subdivided into colonies that started to express GATA1–mCherry and CD16/32 on the same day (blue bars), expressed GATA1–mCherry at least one day before CD16/32 (red bars), or expressed CD16/32 before GATA1–mCherry (green bars). Missing percentages to 100% mean that colonies have either not expressed any marker yet or that individual colonies have died during the course of 10 days. All surviving colonies have turned on at least one marker (GATA1–mCherry or CD16/32) by day 9. Mean (+s.d.) of three biological replicates (n = 141, 185, 129 colonies). d, Colony potential of sorted cells from b, mean ± s.d. (n = 3). e, Cytospin of cells from day 4 cultures, representative example from three independent experiments. Scale bars, 10 μm.
Example trees with mixed GM/MegE (1st and 2nd tree), only GM (3rd tree) and only MegE differentiation (4th tree). Compare Fig. 4.
This file contains the uncropped western blots. (PDF 202 kb)
GM differentiating HSC, displaying one daughter cell of each generation. Brightfield images (‚BF‘), PU.1eYFP signal (‚PU.1‘), GATA1mCHERRY signal (‚Gata1‘) and CD16/32 expression (‚CD16/32‘, determined by live antibody staining). Time format (relative to movie start): d (day) – hh (hours):mm (minutes):ss (seconds). (MP4 1446 kb)
Meg differentiating HSC, displaying one consecutive daughter cell each in brightfield images (‚BF‘) with PU.1eYFP signal (‚PU.1‘), GATA1mCHERRY signal (‚Gata1‘) and CD16/32 expression (‚CD16/32‘, determined by live antibody staining). Time format (relative to movie start): d (day) – hh (hours):mm (minutes):ss (seconds). (MP4 1618 kb)
Density scatter plot of PU.1eYFP and GATA1mCHERRY expression levels during differentiation of all tracked HSCs with CD16/32 or GATA1mCHERRY expression. Normalized to the onset (t=0h) of CD16/32 and GATA1mCHERRY, respectively. Light blue circle highlights initial expression profile of HSCs. (MP4 3397 kb)
About this article
Cite this article
Hoppe, P., Schwarzfischer, M., Loeffler, D. et al. Early myeloid lineage choice is not initiated by random PU.1 to GATA1 protein ratios. Nature 535, 299–302 (2016). https://doi.org/10.1038/nature18320
Trends in Genetics (2021)
E-cadherin is regulated by GATA-2 and marks the early commitment of mouse hematopoietic progenitors to the basophil and mast cell fates
Science Immunology (2021)
EMBO Molecular Medicine (2021)
How haematopoiesis research became a fertile ground for regulatory network biology as pioneered by Eric Davidson
Current Opinion in Hematology (2021)