Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jan 2;215(1):233-248.
doi: 10.1084/jem.20170807. Epub 2017 Dec 7.

In vivo single cell analysis reveals Gata2 dynamics in cells transitioning to hematopoietic fate

Christina Eich  1 , Jochen Arlt  2 , Chris S Vink  3 , Parham Solaimani Kartalaei  1 , Polynikis Kaimakis  1 , Samanta A Mariani  3 , Reinier van der Linden  1 , Wiggert A van Cappellen  4 , Elaine Dzierzak  5   3
Affiliations

In vivo single cell analysis reveals Gata2 dynamics in cells transitioning to hematopoietic fate

Christina Eich et al. J Exp Med. .

Abstract

Cell fate is established through coordinated gene expression programs in individual cells. Regulatory networks that include the Gata2 transcription factor play central roles in hematopoietic fate establishment. Although Gata2 is essential to the embryonic development and function of hematopoietic stem cells that form the adult hierarchy, little is known about the in vivo expression dynamics of Gata2 in single cells. Here, we examine Gata2 expression in single aortic cells as they establish hematopoietic fate in Gata2Venus mouse embryos. Time-lapse imaging reveals rapid pulsatile level changes in Gata2 reporter expression in cells undergoing endothelial-to-hematopoietic transition. Moreover, Gata2 reporter pulsatile expression is dramatically altered in Gata2+/- aortic cells, which undergo fewer transitions and are reduced in hematopoietic potential. Our novel finding of dynamic pulsatile expression of Gata2 suggests a highly unstable genetic state in single cells concomitant with their transition to hematopoietic fate. This reinforces the notion that threshold levels of Gata2 influence fate establishment and has implications for transcription factor-related hematologic dysfunctions.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
G2V reporter reveals expression of Venus in different EHT cell subsets. (A) Diagram of a whole mount of a 35-SP embryo with the dorsal aorta outlined in red and confocal images of the ventral aspect of the G2V dorsal aorta. HECs and BCs (left) and IAHCs (right) are indicated by asterisks (CD31, red; ckit, blue; and Venus, green). Bars, 20 µm. (B) Diagram of an embryo (33–34 SPs) transversal slice prepared for vital confocal time-lapse imaging. Confocal images of representative immunostained G2V embryo traversal sections (CD31, red; ckit, blue; and Venus, green). Bars, 40 µm. Ventral side downward. DIC, differential interference contrast. (C) Schematic representation of Gata2 (green) expression in the different EHT cell subsets in the E10.5 mouse dorsal aorta (ventral aspect; ECs, red, red outline; HECs, light green, red outline; BCs, medium green, blue outline; and IAHCs, dark green, blue outline). Cells with blue outline express ckit in addition to CD31. (D) Flow cytometric contour plot of CD31 gated cells. Percentages of CD31+ckit+Venusmed and CD31+ckit+Venushigh expressing cells from E10.5 G2V AGMs. (E) Bar graph of Gata2 transcription in E10.5 AGM G2V sorted CD31+ckit+Venusmed (light green) and CD31+ckit+Venushigh (dark green) cells. y axis shows FPKM values obtained from RNA-sequencing analysis. The data represent the mean ± SEM of three independent experiments and were compared using a Student’s t test (*, P = 0.0431).
Figure 2.
Figure 2.
Dynamic expression of the Gata2 reporter during EHT. (A–C) Confocal time-lapse imaging of E10 (33–34 SPs) G2V embryos (Venus, green) immunostained with anti–CD31 (red) and anti–ckit (blue) antibodies (A) or anti–CD31 (red) antibody only (B and C). Arrows indicate cells undergoing EHT events. (D) Confocal time-lapse imaging of IAHC at E10 (33–34 SPs; Venus, green), immunostained with anti–CD31 (red) antibody. Arrows indicate three IAHC cells. (Ai–Di) Quantification of Venus MFI over time (hours) corresponding to the highlighted cells during the process of EHT (Ai–Ci) and in IAHC (Di). Transverse aortic sections were imaged for 10 h at 15-min intervals. A–C (middle) show 30-min snapshots, and D shows snapshots at 0, 2.5, and 7 h. Bars, 50 µm.
Figure 3.
Figure 3.
Time-lapse imaging reveals differences in Gata2 dynamics between ECs, BCs, and IAHCs. (A) Mean number of Venus+ EHT subset cells per G2V embryo according to their ventral and dorsal aortic location, as determined by microscopy. Venus+ HECs, BCs, or IAHCs were counted in the first frame of time-lapse imaging experiments (n = 15, 42 sections) of G2V embryo slices of 150 µm thickness and calculated per embryo (E10, 32–37 SPs). The data represent the mean ± SEM of 15 independent experiments, and dorsal and ventral location were compared using two-way ANOVA with Bonferroni post test (*, P < 0.05). (B) Left: Example of a BC showing decreasing and increasing levels of Venus expression during a 12-h imaging period, imaged at a time interval of 15 min. Middle: Three-dimensional projections (x-y axes and x-z axes) of the same BC as shown in the top panel with time (hours) indicated. Right: Corresponding Venus (green) MFI profile in time. Bar, 25 µm. Sections were stained with anti–CD31 (red) antibody. (C) Venus MFI (averaged over frames 3–12) in single EHT subset cells (HECs, BCs, and IAHCs; n = 15, 718 cells). The data represent the mean ± range. Statistical significance was calculated on the pooled data of 15 independent experiments using Mann–Whitney U test (***, P < 0.0001). (D) Venus MFI in single EHT subset cells plotted according to their dorsal (d) and ventral (v) location (n = 15, 718 cells). The data represent the mean ± range. Statistical significance was calculated on the pooled data of 15 independent experiments using Mann-Whitney U test (*, P < 0.0288; **, P = 0.0020; ***, P < 0.0001). (E) Top: Temporal variation of Venus MFI for individual Venus+ cells (colored lines) plotted according to their affiliation to one of the EHT subsets (EC, BC, or IAHC). Bottom: Gray bands represent the standard deviation of the mean MFI of all tracks (black line) over time.
Figure 4.
Figure 4.
Pulse frequency and amplitude of Venus expression distinguishes EHT subsets. (A) Schematic representation of the automatic peak detection code. A local MFI maximum is considered a peak if it has at least a 15% higher intensity than its neighboring minima (see Materials and methods). The pulse period is the time between two adjacent peaks and the trough-to-peak amplitude the change between peak (highest value) and the preceding trough (lowest value). (B) Examples of normalized MFI profiles with no peak, one peak, and two pulse peaks showing increasing trough-to-peak amplitudes. (C) Distribution of the occurrence (percentage) of normalized pulse peak numbers in ECs, BCs, and IAHCs tracked over at least 10 consecutive frames (718 cells). To normalize for differences in track length, the data are presented as peaks per 10 frames (2.5 h) and represent the mean ± SEM (n = 15). Statistical significance was calculated using two-way ANOVA with Bonferroni post test (**, P < 0.01; ***, P < 0.001). (D) Distribution of the pulse periodicities of Venus+ cells showing at least two pulse peaks (n = 15, 221 cells). (E) Distribution of the pulse periodicities in EHT subset cells showing at least two pulse peaks (n = 15, 86 HECs, 80 BCs, and 55 IAHCs). The data represent the mean ± SD. Statistical significance was calculated on the pooled data (n = 15) using two-way ANOVA with Bonferroni post test (**, P ≤ 0.01; ***, P < 0.001). (F and G) Peak intensity (F) and trough-to-peak amplitude (G) in the EHT cell subsets, plotted according to their ventral (v) or dorsal (d) location in the aorta (n = 13, cells showing at least one peak: 170 HECs, 151 BCs, and 65 IAHCs). The data represent the mean ± range. Statistical significance was calculated on the pooled data (n = 15) using Mann–Whitney U test (**, P = 0.0054; ***, P < 0.0008).
Figure 5.
Figure 5.
Hematopoietic potential correlates with Venus expression levels. (A) Hematopoietic progenitor numbers in E10.5 (30–36 SPs) CD31+ckit+Vmed and CD31+ckit+Vhigh AGM sorted cells. Colony-forming unit-culture per embryo equivalent (CFU-C/ee) is shown, with colony types designated by colored bars. BFU-E, burst-forming unit erythroid; CFU-G/M/GM, CFU granulocyte, CFU macrophage, and CFU granulocyte-macrophage; CFU-GEMM, CFU granulocyte, erythroid, macrophage, megakaryocyte. The data represent the mean ± SEM of four independent experiments. (B) Percentage donor cell chimerism in recipient mice injected with CD31+ckit−V+, CD31+ckit+V−, CD31+ckit+Vmed, or CD31+ckitVhigh sorted E11 (41–49 SPs) AGM cells. Engraftment at 4 mo after transplantation was determined by flow cytometric analysis of Ly5.1/Ly5.2 marker expression of peripheral blood cells. Each dot represents one recipient receiving 1.3 to 4.1 embryo equivalent (ee) of sorted AGM cells. The data represent the mean ± SD. *, P ≤ 0.024; **, P = 0.0085; ***, P = 0.0003). (C) Overrepresentation of up-regulated differentially expressed genes (DEGs) in E10.5 CD31+ckit+Vmed and CD31+ckitVhigh sorted cells in IPA canonical pathways. (D and E) Mean FPKM values for genes in the Notch pathway (D) and heptad factor genes in E10.5 CD31+ckit+Vmed and CD31+ckitVhigh sorted E10.5 AGM cells (E). The data were compared using Student’s t test (*, P = 0.0404; **, P = 0.0096). The data represent the mean ± SEM of three independent experiments.
Figure 6.
Figure 6.
Gata2 expression parameters, hematopoietic fate, and EHT are interrelated. (A) Maximum projections of confocal time-lapse images of E10.5 Gata2V/+ and Gata2V/− aortas immunostained with anti-CD31 (red; G2V, green). Bars, 40 µm. Gata2V/+ and Gata2V/− embryos were harvested from the same litter. Ventral side downward. (B) Venus MFI (averaged over frames 3–12) in single Gata2V/- EHT subset cells (n = 6; 75 ECs, 37 BCs, and 12 IAHCs). The data were compared with Gata2V/+ EHT subset cells using one-way ANOVA with Bonferroni post test (mean ± SD; *, P ≤ 0.024; **, P = 0.0085; ***, P = 0.0003). (C and D) Distribution of the pulse periodicities in Venus+ EHT subset cells: HECs (C) and BCs (D) from E10.5 Gata2V/− aortas showing at least two pulsatile peaks (n = 6; 18 HECs and 15 BCs). The data represent the mean ± SD. The data were compared with Gata2V/+ HECs and BCs using two-way ANOVA with Bonferroni post test (**, P < 0.01). (E and F) Peak intensity (E) and trough-to-peak amplitude (F) in the Gata2V/− EHT subsets (n = 6, cells showing at least one peak: 36 ECs, 20 BCs, and 6 IAHCs). The data represent the mean ± range. The data were compared with Gata2V/+ EHT subset cells using Mann–Whitney U test (*, P = 0.0321; **, P = 0.0056). (G) Model of Gata2 expression dynamics and pulsatile characteristic during EHT. EHT cell types (top) are shown with accompanying Gata2 dynamic expression changes EHT directly below. G2V MFI (bright green) and pulse parameters (dark green sinusoids) are shown for Gata2V/+ (middle) and Gata2V/− (bottom) EHT subset cells.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Albeck J.G., Mills G.B., and Brugge J.S.. 2013. Frequency-modulated pulses of ERK activity transmit quantitative proliferation signals. Mol. Cell. 49:249–261. 10.1016/j.molcel.2012.11.002 - DOI - PMC - PubMed
    1. Anders S., and Huber W.. 2010. Differential expression analysis for sequence count data. Genome Biol. 11:R106 10.1186/gb-2010-11-10-r106 - DOI - PMC - PubMed
    1. Batchelor E., Loewer A., Mock C., and Lahav G.. 2011. Stimulus-dependent dynamics of p53 in single cells. Mol. Syst. Biol. 7:488 10.1038/msb.2011.20 - DOI - PMC - PubMed
    1. Boisset J.C., van Cappellen W., Andrieu-Soler C., Galjart N., Dzierzak E., and Robin C.. 2010. In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium. Nature. 464:116–120. 10.1038/nature08764 - DOI - PubMed
    1. Boisset J.C., Andrieu-Soler C., van Cappellen W.A., Clapes T., and Robin C.. 2011. Ex vivo time-lapse confocal imaging of the mouse embryo aorta. Nat. Protoc. 6:1792–1805. 10.1038/nprot.2011.401 - DOI - PubMed

Publication types

Substances