C3G, through its GEF activity, induces megakaryocytic differentiation and proplatelet formation

Abstract

Background: Megakaryopoiesis allows platelet formation, which is necessary for coagulation, also playing an important role in different pathologies. However, this process remains to be fully characterized. C3G, an activator of Rap1 GTPases, is involved in platelet activation and regulates several differentiation processes.

Methods: We evaluated C3G function in megakaryopoiesis using transgenic mouse models where C3G and C3GΔCat (mutant lacking the GEF domain) transgenes are expressed exclusively in megakaryocytes and platelets. In addition, we used different clones of K562, HEL and DAMI cell lines with overexpression or silencing of C3G or GATA-1.

Results: We found that C3G participates in the differentiation of immature hematopoietic cells to megakaryocytes. Accordingly, bone marrow cells from transgenic C3G, but not those from transgenic C3GΔCat mice, showed increased expression of the differentiation markers CD41 and CD61, upon thrombopoietin treatment. Furthermore, C3G overexpression increased the number of CD41⁺ megakaryocytes with high DNA content. These results are supported by data obtained in the different models of megakaryocytic cell lines. In addition, it was uncovered GATA-1 as a positive regulator of C3G expression. Moreover, C3G transgenic megakaryocytes from fresh bone marrow explants showed increased migration from the osteoblastic to the vascular niche and an enhanced ability to form proplatelets. Although the transgenic expression of C3G in platelets did not alter basal platelet counts, it did increase slightly those induced by TPO injection in vivo. Moreover, platelet C3G induced adipogenesis in the bone marrow under pathological conditions.

Conclusions: All these data indicate that C3G plays a significant role in different steps of megakaryopoiesis, acting through a mechanism dependent on its GEF activity.

Keywords: C3G; Differentiation; Megakaryocyte; Megakaryopoiesis; Platelet.

Fig. 1

Overexpression of C3G in K562 cells increases megakaryocytic markers. a May-Grünwald-Giemsa staining of the indicated K562 clones untreated or treated with 20 nM PMA for 72 h. Scale bars: 20 μm. b Representative images showing the morphology of HEL cells after 24, 48 and 72 h of PMA treatment. Images were obtained using a Zeiss Axiovert 135 inverted microscope. Scale bars: 50 μm, c Analysis of C3G expression in K562 and HEL cells stimulated with 20 nM PMA at the indicated times. Tubulin was used as loading control. C3G expression values are relative to non-treated cells and normalized against tubulin (C3G/Tub). d Expression of CD41, CD61 and GPA markers was analyzed by flow cytometry using specific fluorochrome-conjugated antibodies (CD41-APC, CD61-PE and GPA-FITC). Representative flow cytometry plots of untreated cells are shown. Histograms represent the mean ± SEM of the fluorescence intensity (relative units) of CD41, CD61 and GPA from at least 4 independent experiments of each clone, treated as indicated. 2-way ANOVA and Holm-Sidak analysis were done. **p < 0.01, ***p < 0.001. e Representative Western blots showing C3G expression in the pLTR2 and pLTR2/GFP clones. The numbers represent C3G expression values relative to empty vectors and normalized against tubulin (C3G/Tub), which was used as loading control. f Analysis of CD61 and GPA expression in untreated K562 clones, pLTR2-CT and pLTR2-C3G, by RT-PCR. The numbers represent expression values relative to pLTR2-CT expressing cells and normalized against GAPDH, which was used as housekeeping gene. g Western blot analysis of α-globin and C3G in the indicated clones of K562, treated or not with 2 μM STI-571, an inducer of erythroid differentiation in these cells, [46]. The expression of total ERK was used as loading control. α-globin/ERK ratios are shown. All values are relative to control, non-treated cells. STI: STI-571 (Imatinib mesylate); CT: control

Fig. 2

Overexpression of C3G in K562 increases polyploidization and p21 expression. The polyploidization state of K562-pLTR2/GFP and -CRISPR clones was identified by propidium iodide staining after 10 days of differentiation with 20 nM PMA. Representative flow cytometry plots of ploidy distribution, stacked bars and histograms corresponding to CRISPR clones (a) and pLTR2/GFP clones (b). We identified up to 4 different populations: diploid (2n), tetraploid (4n) and polyploidy cells (8n and ≥ 16n), indicating the percentage of cells corresponding to each population. Stacked bar histograms represent the mean of the percentage of cells of each genotype belonging to the different ploidy populations. Histograms represent the mean ± SEM of the quantification of the percentage of individual ploidy population. c Time course Western blot analysis of the expression of C3G (indicated by an arrow) and p21 in K562 cells transfected with pLTR2/GFP plasmids (upper panels) or CRISPR plasmids (lower panels) treated with PMA (20 nM) for the indicated times. Left panel: representative images of Western blot. Tubulin was used as loading control. The asterisk indicates a non-specific band. Right panel: Line/scatter plots of p21 expression. Values are normalized with tubulin and relativized to control, non-treated cells (n = 2). The ANOVA analysis indicates that there are statistically significant differences in p21 expression between pLTR2-CT and pLTR2-C3G at the different time points (p = 0.019)

Fig. 3

PMA induces C3G phosphorylation in K562 cells, which correlates with a sustained Rap1 activation. a Left panel: time course Western blot analysis of phospho-Y504-C3G (p-C3G) expression in non-transfected K562 cells treated with 20 nM PMA for 2, 5, 10, 30 and 60 min. The expression of total C3G and tubulin were used as loading controls. The asterisk indicates a non-specific band. Right panel: Histogram showing pY504-C3G/C3G ratios, relativized to control, non-treated cells. b Representative immunofluorescence confocal microscopy images of the indicated K562 clones treated with 20 nM PMA for 2, 5 and 10 min, fixed, permeabilized and incubated with: anti-phospho-C3G/anti-rabbit Cy3 antibodies (green), purified GST-RalGDS-RBD and anti-GST/anti-mouse Cy5 antibodies (red) to detect active Rap1-GTP, and DAPI (blue). The overlay images are made without DAPI channel. Images of GFP expression are shown. Histograms represent the mean ± SEM of the integrated density (I.D.) of p-C3G (left panel) and Rap1-GTP staining (right panel). The ANOVA analysis indicates that there is a significant variability between pLTR2-C3G and pLTR2-CT in the fluorescence intensities of p-C3G and Rap1-GTP, independently of the treatment used (p < 0.05). In addition, Holm-Sidak method was done. *p < 0.05

Fig. 4

GATA-1 regulates C3G expression in Dami cells. a Dami cells with a permanent GATA-1 knockdown (shGATA-1) or overexpression of the fused protein V5-GATA-1 (a gift from Dr. Laura Gutiérrez, University of Oviedo), as well as their control cells (expressing empty vectors) were used. Representative flow cytometry plots of the expression of CD41 marker in Dami cells treated with 100 ng/ml PMA for 96 h. Histograms represent the mean ± SEM of the fluorescence intensity (relative units) of CD41 from 3 to 5 independent experiments of each clone. b Histograms represent the mean ± SEM of the percentage of polyploidy (>4n) Dami cells stimulated with PMA for 96 h (n = 5). c Ectopic expression of GATA-1 in Dami cells induces C3G expression. Left panel: Western blot showing the expression of V5-GATA-1 protein or endogenous GATA-1 in Dami cells. V5-GATA-1/β-actin or endogenous (End.) GATA-1/β-actin ratios are shown. Values were normalized against endogenous GATA-1 levels in cells transfected with the empty vector. Right panel: C3G expression in Dami cells transfected with V5-GATA-1 or its control, stimulated with 100 ng/ml PMA for 96 h. Normalized C3G/β-actin ratios are shown. d Silencing of GATA-1 abolished PMA-induced C3G expression in Dami cells. Left panel: Western blot showing GATA-1 levels in Dami cells. Values were normalized using β-actin. Righ panel: C3G expression in Dami cells with silenced GATA-1 expression and treated with 100 ng/ml PMA for 96 h. Normalized C3G/β-actin ratios are shown

Fig. 5

Tg-C3G expression promotes CFU-MKs and increases the percentage of mature megakaryocytes in BM in vitro. a Freshly isolated BM cells were cultured with TPO for 6 days. The percentage of CD41+, CD61+ and double CD41+/CD61+ cells was analyzed by flow cytometry. Box plots represent the median ± SEM of the percentage of positive cells of 6 different measures from three independent cultures of each genotype. Mann-Whitney U test was done. *p < 0.05, **p < 0.01 and ***p < 0.001. b Table indicating the mean ± SEM of the percentage of positive cells of each genotype. c Representative images of a CFU-MK from WT-C3G and Tg-C3G (2C1 lineage) BMs. Histograms represent the mean ± SEM of the total number of CFU-MKs in BM cultures from 3 mouse of each genotype and 2 cultures per mouse. d Representative flow cytometry plots of ploidy distribution of FSChigh/CD41+ MKs (WT and Tg). We identified up to 5 different populations; 2n, 4n, 8n, 16n and ≥ 32n. The percentage of FSChigh/CD41+ MKs corresponding to each population is indicated. e Stacked bar histograms of the percentage of MKs of each genotype belonging to the different ploidy populations (2n, 4n, 8n, 16n and ≥ 32n) of BM cells. Vertical bars indicate the percentage of cells with a DNA content ≥8n. f Histograms represent the mean ± SEM of the percentage of individual ploidy populations of BM cells. Data correspond to 3 different experiments from 3 mouse of each genotype. Data were analyzed using the t-test. *p < 0.05 and **p < 0.01

Fig. 6

C3G regulates megakaryocyte motility and promotes the formation of proplatelets. a Transverse sections of bone marrows from the different genotypes were plated in an incubation chamber and maintained at 37 °C for 6 h. MKs at the periphery of the explants were tracked under the microscope and images were acquired at 10 min intervals. Upper box plots represent the median ± SEM velocity (μm/second) of individual megakaryocytes 6 h after their release from bone marrow explants. Lower box plots represent the median distance (μm) covered by megakaryocytes from bone marrow explants. Three mice of each genotype were analyzed. Mann Whitney U test were done. **p < 0.01, ***p < 0.001. b Percentage of MKs most strongly associated to the osteoblastic niche. After extraction of the bone marrow, small pieces of femur were treated with collagenase and dispase for 2 h and the percentage of CD61+ cells was analyzed by flow cytometry. Mann-Whitney U test was done. c Representative images of the different stages of MK maturation: spherical megakaryocytes, megakaryocytes with extending protrusions and megakaryocytes with proplatelets. Fluorescence microscope images (Left panels) and brightfield inverted microscope images (right panels). The histograms represent the mean ± SEM of the percentage of cells of each phenotype measured in two different experiments, with at least 4 explants from each genotype. d C3G did not modify platelet count. Counts were performed in peripheral blood collected from 6-month mice of the different genotypes using Hemavet Counter HV950FS. The histograms represent the mean ± SEM of the number of platelets. Mann-Whitney U test was done, but no significant differences were observed between Tg-C3G or Tg-C3GΔCat vs WT.

Fig. 7

C3G increases platelet production in response to TPO. Mice were injected intravenously with 5 μg TPO/mouse and blood samples were collected at the indicated time points. a Time course of platelet counts in Tg-C3G (left panel), Tg-C3GΔCat (right panel) and their corresponding wild-types. The graphs represent the median of the number of platelets (1000/μl) of each phenotype. b The histograms represent the mean ± SEM of the maximum fold change in the number of platelets in blood after TPO treatment (platelet count at day 5–6 /platelet count at day 0)

Fig. 8

C3G induces adipocyte density in bone marrow after tumor implantation. B16-F10 metastatic melanoma cells (1 × 10⁶) were injected subcutaneously in Tg-C3G and WT-C3G mice. After 15 days of tumor growth, the bone marrow was harvested, fixed with 4% formaldehyde, embedded in paraffin and stained with H&E. a Representative images of bone marrow from femurs of transgenic and wild type mice. b Histogram represents the median ± SEM of the total adipocyte area relative to the bone marrow area. Mann-Whitney U test was done, but no significant differences were observed between Tg-C3G vs WT. c Box plot represent the median of the adipocyte size (μm2) of individual adipocytes. At least 4 different mice of each genotype were analyzed. Mann Whitney U test was done. ***p < 0.001