Tescalcin is an essential factor in megakaryocytic differentiation associated with Ets family gene expression (original) (raw)
Upregulation of tescalcin correlates with megakaryocytic differentiation. We previously showed that tescalcin is expressed in some hematopoietic cell lines, including chronic myelogenous leukemia K562 cells (33). K562 cells are regarded as pluripotent hematopoietic progenitors that express specific markers of granulocytic, monocytic, erythroid, and megakaryocytic lineages. In addition, K562 cells can be induced to differentiate in vitro following stimulation by a variety of specific agents (36). To determine whether tescalcin is regulated in hematopoietic differentiation, K562 cells were stimulated with 1.5% DMSO, 30 μM hemin, or 10 nM PMA to promote granulocytic, erythroid, or megakaryocytic differentiation, respectively. Treatment with PMA, but not with DMSO or hemin, led to a dramatic increase in the level of tescalcin protein (Figure 1A). A slight increase in tescalcin expression was also observed in PMA-treated HEL cells (data not shown), another pluripotent hematopoietic cell line expressing markers of megakaryocytic differentiation (37, 38). Real-time quantitative PCR analysis (RQ-PCR) on total RNAs isolated from K562 and HEL cells demonstrated that PMA-induced upregulation of tescalcin was at least partially due to an increase in the accumulation of its mRNA (Figure 1B). We also noticed that the level of tescalcin mRNA in unstimulated HEL cells was at least 4-fold higher than in K562 (Figure 1B). Accordingly, HEL cells expressed higher levels of megakaryocytic markers (37). To find out whether tescalcin is expressed in primary cells, we used an established method to obtain mature megakaryocytes (MKs) from mouse fetal liver cells stimulated by TPO (39). Western blot analysis showed that while tescalcin was not detected in fetal liver cells, it was highly expressed in terminally differentiated MKs (Figure 1C).
The expression of tescalcin is increased during megakaryocytic differentiation. (A) K562 cells were cultured in the presence of DMSO, hemin, or PMA for 72 hours, as described in Methods. Cells lysates were subjected to Western blot analysis with antibody against tescalcin. (B) Stimulation with PMA upregulates tescalcin at the mRNA level. K562 and HEL cells were cultured in the absence or presence of PMA for 72 hours. Total RNA subjected to quantitative RQ-PCR with tescalcin-specific primers (TaqMan). Obtained values (n = 3; mean ± SD) were normalized to 18S ribosomal RNA, and expressed relative to unstimulated K562. (C) Tescalcin expression in primary MKs. Mature mouse MKs were obtained from fetal livers as described in Methods. Lysates of fetal liver cells (FLC) and MKs were analyzed for tescalcin expression by Western blot. K562 lysate was used as positive control. (D) K562 cells were cultured in the presence of PMA (10 nM) for the indicated times, and the accumulation of tescalcin was determined by Western blot. (E) Bryostatin blocks PMA-induced upregulation of tescalcin. K562 cells were stimulated by 10 nM PMA in the absence or presence of 100 nM bryostatin (Bryo). The expression of tescalcin was determined by Western blot. (F) Sustained ERK activity is required for tescalcin upregulation. K562 cells were stimulated by PMA (10 nM) in the absence or presence of MEK1/2 inhibitor (20 μM; U0126). Cell lysates were probed with antibodies to tescalcin, total p44/42 MAPKs (ERK1/2), and phospho-p44/42 MAP kinases (ERK1*/2*).
Upregulation of tescalcin requires sustained ERK activity. Stimulation with PMA causes the activation of PKC, which in K562 cells can lead to a number of differentiation events including the acquisition of megakaryocytic markers, growth arrest, and development of polyploid nuclei (4, 40). It has been previously shown that irreversible commitment of K562 cells to the megakaryocytic lineage depends on sustained (>24 hours) ERK activity and can be reached by treatment with PMA (6). We found that there was a delay of at least 48 hours between initiation of stimulus and the rise in tescalcin accumulation (Figure 1D). This expression pattern parallels the PMA-induced expression of other factors involved in megakaryocytopoiesis (41) and indicates that tescalcin may be a part of the differentiation program. To test this idea, we treated cells with bryostatin, a structurally distinct PKC activator. Unlike PMA, bryostatin is only capable of transient activation of ERK, which is insufficient to induce differentiation. Moreover, bryostatin effectively blocks PMA-induced differentiation in K562 cells (40). In our experiments, bryostatin inhibited upregulation of tescalcin induced by PMA (Figure 1E). To determine whether ERK activity is necessary for PMA-induced upregulation of tescalcin, we used U0126, a highly selective inhibitor of both MEK1 and -2, known direct upstream regulators of ERK. Pretreatment of cells with 20 μM U0126 effectively blocked both the activation of ERK and upregulation of tescalcin in response to PMA (Figure 1F).
Altogether, our results suggest that upregulation of tescalcin is primarily dependent on sustained ERK activity and correlates with the induction of megakaryocytic differentiation.
K562 cells overexpressing tescalcin recapitulate the events of early megakaryocytic differentiation. To establish the cause and effect relationship between upregulation of tescalcin and megakaryocytic differentiation, we generated several stable clones of K562 cells with constitutive overexpression of tescalcin. Cells transfected with empty pcDNA3 vector were used as negative controls. For the biochemical and functional studies and to avoid clonal variability, we pooled 3 independent clones of tescalcin overexpressors and corresponding controls. These pools of transfectants were designated K562-Tsc(+) and K562-Ctrl(+), respectively.
We noticed that upon overexpression of tescalcin the fraction of enlarged K562 cells had considerably increased. This observation led us to the idea that high levels of tescalcin in K562 may promote spontaneous cell differentiation along megakaryocytic lineage. To test this hypothesis, we examined typical characteristics of megakaryocytic differentiation, such as the expression of GPIIb, polyploidy, and the rate of cell proliferation (Figure 2, A–E). The expression of GPIIb, which is a part of GPIIb/IIIa (also known as αIIbβIII, CD41/CD61) integrin complex, on the cell surface is one of the key events of early megakaryocytopoiesis. Flow cytometry analysis confirmed that K562-Tsc(+) cells show increased level of GPIIb and that such an increase is mostly found on a subpopulation of larger cells (Figure 2A and Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI27465DS1).
Overexpression of tescalcin in K562 cells induces events of early megakaryocytic differentiation. (A) Overexpression of tescalcin leads to an increased level of surface GPIIb. Control [K562-Ctrl(+)] and tescalcin overexpressing [K562-Tsc(+)] cells were stained with FITC-conjugated CD41-specific antibody. Samples were analyzed using BD FACScan, with a minimum of 10,000 events acquired per sample. Expression profile of surface CD41 in K562-Ctrl(+) and K562-Tsc(+) cells is shown as filled histogram. FITC-conjugated isotype mouse IgG was used as negative control (open histogram). (B) Overexpression of tescalcin promotes polyploidy in K562 cells. K562-Ctrl(+) and K562-Tsc(+) cells were fixed and stained with propidium iodide and their DNA content was analyzed by FACS. PI log, propidium iodide, logarithmic scale. (C) The onset of PMA-induced polyploidy occurs faster in tescalcin-overexpressing cells. K562-Ctrl(+) and K562-Tsc(+) cells were cultured in the presence of PMA for indicated times and analyzed as described in B. (D) Cyclin D3 accumulation in tescalcin-overexpressing cells is increased. Wild-type [K562-WT], K562-Ctrl(+), and K562-Tsc(+) cell lysates were probed with cyclin D3– and tescalcin-specific antibodies. β-Actin was used as a loading control. (E) Reduction of proliferation rate in tescalcin-overexpressing cells. Absorbance at 490 nm of K562-Ctrl(+) and K562-Tsc(+) cell lines were compared in the MTS-based cell proliferation assay, as described in Methods. Data represent 3 independent experiments (mean ± SD).
Another indicator of megakaryocytic differentiation is polyploidization, which is essential for the efficient production and release of platelets. It is known that in response to PMA stimulation, K562 nuclear DNA ploidy rises to 4N–16N concurrently with an increase in cell volume (36). Using FACS analysis of propidium iodide–stained cells, we compared the DNA content in K562-Ctrl(+) and K562-Tsc(+) cells. In unstimulated K562-Tsc(+) cells the number of cells with DNA content of 8N and more was increased to 8%–10% compared with less than 1% in controls (Figure 2B). Further, we found that high levels of tescalcin caused accelerated onset of polyploidy in response to PMA. More than 20% of 2N and 4N cells overexpressing tescalcin shifted to 8N and higher ploidy after only 24 hours of stimulation. After 72 hours of PMA stimulation, a significant number of K562-Tsc(+) cells were already shifted to 16N and higher ploidy, while the majority of control K562-Ctrl(+) cells remained at 4N and 8N (Figure 2C). It was reported earlier that the G1-phase cyclin, cyclin D3, is upregulated in polyploidizing megakaryocytic cells and that overexpression of cyclin D3 in transgenic mice leads to increased MK number and ploidy (42, 43). In agreement with these earlier observations, we found that K562-Tsc(+) cells had an increased accumulation of cyclin D3 (Figure 2D). Next we determined the effect of tescalcin overexpression on the growth rate of K562 cells. For this we cultured cells in low-serum growth medium (1% FBS) in order to decrease the proliferative drive while preserving viability. As expected of differentiating cells undergoing polyploidy and growth inhibition, proliferation of K562-Tsc(+) cells was significantly reduced (Figure 2E). Consistent with this observation, our attempt to generate stable constitutive overexpression of tescalcin in HEL cells resulted in a rapid cell enlargement and death within several days (Supplemental Figure 2).
Thus our results strongly indicate that overexpression of tescalcin is sufficient to induce the early events of megakaryocytic differentiation in K562 cells, including the expression of the MK-specific marker GPIIb, increased cell size, polyploidization, and inhibition of growth rate.
Effect of tescalcin knockdown on cell proliferation and PMA-induced polyploidy. To determine whether tescalcin is necessary for megakaryocytic differentiation, we inhibited its expression in K562 and HEL cells by vector-based short hairpin RNA (shRNA). In a number of generated stable clones the expression of tescalcin was dramatically reduced. Pools of 3 independent clones of each type, termed “K562-Tsc(–)” and “HEL-Tsc(–),” were used in all further experiments. PMA stimulation could cause only a minor increase of tescalcin expression in these cells (Figure 3A). For a negative control, we established several cell lines stably transfected with the shRNA construct encoding a scrambled target sequence and termed them “K562-Ctrl(–)” and “HEL-Ctrl(–).” The level of tescalcin expression in control cells was similar to wild-type K562 and HEL.
Tescalcin knockdown inhibits PMA-induced polyploidy and increases cell proliferation. (A) Western blot analysis of wild-type cells and cells expressing scrambled Ctrl(–) and tescalcin-specific Tsc(–) target sequence shRNAs. Cells were cultured in the absence and presence of PMA and then analyzed by Western blot with antibody to tescalcin. β-Actin was used as a loading control. (B) Comparison of the DNA content in control HEL-Ctrl(–) and tescalcin knockdown HEL-Tsc(–) cells prior to or after stimulation by PMA for 72 hours. Cells were fixed, stained with propidium iodide, and analyzed by FACS. (C) Increase in cell proliferation after tescalcin knockdown. HEL-Ctrl(–) and HEL-Tsc(–) cells were compared in the MTS-based cell proliferation assay, as described in Methods. Data represent 3 independent experiments (mean ± SD).
The phenotype caused by tescalcin knockdown was exactly opposite that observed with tescalcin overexpression. As expected, control HEL-Ctrl(–) cells responded to PMA treatment, with the DNA content rising up to 16N ploidy level within 72 hours, whereas the DNA content in tescalcin knockdown HEL-Tsc(–) cells did not increase beyond 4N (Figure 3B). The DNA histogram for the uninduced HEL-Tsc(–) shows the accumulation of cells in S phase of cell cycle, indicative of their higher proliferative activity. Indeed, the proliferation rate of HEL-Tsc(–) was significantly higher than that of HEL-Ctrl(–) cells (Figure 3C). Thus knockdown of tescalcin prevents HEL cells from undergoing PMA-induced polyploidy and increases the rate of cell proliferation.
Effects of tescalcin knockdown on adhesion of HEL cells. Adhesion and spreading on extracellular matrix is an important step in megakaryocytic differentiation and maturation in vivo (44). We examined the effect of tescalcin knockdown on fibronectin-specific adhesion. While HEL cells normally grow in suspension, they also strongly adhere to a plastic surface after treatment with PMA (Figure 4, A and B). Knockdown of tescalcin significantly reduced this ability. Instead we observed a homotypic aggregation of HEL-Tsc(–) cells, which started approximately 36 hours after induction with PMA (Figure 4C). Moreover, fibronectin-specific adhesion was strongly inhibited as well. Under normal growing conditions 75%–80% of either HEL-WT or HEL-Ctrl(–) cells attached to fibronectin-coated plates within 30 minutes. The adherence increased to almost 100% when cells were stimulated with 10 nM PMA. In contrast, only about 3% of unstimulated HEL-Tsc(–) cells adhered to fibronectin, and the number of adhering cells did not increase above 12% even in the presence of PMA (Figure 4D).
Tescalcin knockdown inhibits cell adherence to plastic and extracellular matrix proteins. Plated in plastic culture dishes, HEL-WT (A), HEL-Ctrl(–) (B), and HEL-Tsc(–) (C) cells were stimulated by 10 nM PMA. Cells were photographed 36 hours later at ×200 magnification. Scale bar: 25 μm. (D) HEL-WT, HEL-Ctrl(–), HEL-Tsc(–) cells were allowed to adhere for 30 minutes in a 96-well culture plate precoated with fibronectin in the absence or presence of 10 nM PMA. Adhesion was measured as described in Methods. Bar graphs represent results of 3 independent experiments performed in triplicate (mean ± SD).
Tescalcin is necessary for the expression of integrins. The major cell surface receptors for extracellular matrix molecules are integrins, a family of αβ heterodimeric transmembrane glycoproteins. Integrins of β1 and β3 subfamilies, primarily α4β1, α5β1, and αIIbβ3, have been identified as being responsible for the binding of hematopoietic progenitor cells to fibronectin (45, 46). We used FACS analysis to assess the surface expression of GPIIb, α4, and α5 integrins which are known to be present in HEL cells. Consistent with previous reports, PMA treatment of HEL-Ctrl(–) resulted in a dramatic increase of GPIIb surface expression, whereas the expression of α4 and α5 integrins did not change significantly (Figure 5A). In tescalcin knockdown HEL-Tsc(–) cells, the surface expression of GPIIb and α4 molecules was below detectable levels either before or after PMA stimulation. This finding evidently explains the low adhesion phenotype displayed by HEL-Tsc(–) cells (Figure 4). In contrast, the expression level of α5 integrin did not differ from that of control cells, indicating that the effect of tescalcin knockdown in HEL cells is specific to GPIIb and α4 integrin subunits. A similar effect of tescalcin knockdown on GPIIb expression was observed in K562-Tsc(–) cells (Figure 5B). Western blot analysis of total cell lysates demonstrated that not only surface expression but also the total level of GPIIb before and after PMA stimulation were drastically affected by the knockdown of tescalcin in both HEL and K562 cells (Figure 5, C and D).
Tescalcin knockdown inhibits the expression of MK-specific marker GPIIb. (A) HEL-Ctrl(–) and HEL-Tsc(–) cells were cultured in the absence or presence of 10 nM PMA for 72 hours and then analyzed for the cell surface expression of GPIIb (CD41), α4 (CD49d), and α5 (CD49e) integrins by FACS. (B) K562-Ctrl(–) and K562-Tsc(–) cells were cultured in the absence or presence of PMA for 72 hours and analyzed for the cell surface expression of GPIIb by FACS. (C) HEL-Ctrl(–) and HEL-Tsc(–) cells were cultured in the absence or presence of PMA for 72 hours and then subjected to Western blot analysis of the total expression of GPIIb. β-Actin was used as a loading control. (D) K562 cells were analyzed as described in C.
Downregulation of tescalcin in human bone marrow CD34+ cells inhibits megakaryocytic differentiation. To examine the role of tescalcin in primary cells, we knocked down tescalcin in human bone marrow CD34+ progenitors (Stem Cell Technologies Inc.) and tested their ability to differentiate using a CFU assay. Cells were transduced with lentiviral vectors encoding scrambled and tescalcin-specific shRNA. According to GFP coexpression, approximately 50% of cells were infected at MOI = 5. The increase in MOI resulted in adverse effects on cell viability compared with noninfected CD34+ cells. To assess the extent of tescalcin knockdown, infected cells were cultured in liquid MK differentiation medium for 7 days and analyzed by Western blot. Tescalcin level was reduced approximately 2-fold in the entire pool of cells, indicating that knockdown was quite efficient in infected cell population (Figure 6A). The collagen-based CFU-MK analysis of the transduced CD34+ progenitors revealed the following effects: the strongest effect of tescalcin knockdown was the 5-fold increase in frequency of small (3–20 cells/colony) CFU-MK–derived colonies (Figure 6, B and C). Importantly, the majority of CFU-MK colonies in the control consisted of 7–20 GPIIb-positive cells, whereas CFU-MK colonies after tescalcin knockdown contained only 3–5 cells (Figure 6, D and E). Another distinct effect of tescalcin knockdown was a 2-fold (P = 0.044) decrease in a number of large colonies resembling the primitive burst-forming unit–MK (BFU-MK). The number of well defined colonies of other hematopoietic cells (negative for GPIIb) increased more than 3-fold (P = 0.017) upon tescalcin knockdown (Figure 6, B and C). These results show that tescalcin plays an important role in megakaryocytic differentiation of primary bone marrow CD34+ progenitors.
Silencing of tescalcin in CD34+ progenitors inhibits megakaryopoiesis. (A) 1 × 104 human bone marrow CD34+ cells were transduced with control or tescalcin-specific shRNA lentiviral vectors at MOI = 5, cultured in liquid differentiation media (MegaCult; StemCell Technologies Inc.) with cytokines (50 ng/ml TPO, 10 ng/ml IL-6, 10 ng/ml IL-3) for 7 days, and analyzed for tescalcin expression by Western blot. (B and C) CD34+ cells were transduced by spinoculation with scrambled control (B) or tescalcin-specific shRNA (C) lentiviral vectors at MOI = 5 in serum-free medium, as described in Methods. After recovering for 16 hours, cells were suspended in collagen-based MK differentiation media and plated in chamber slides. Twelve days later, cells were fixed, stained with anti-GPIIb antibody, and scored for 5 types of colonies. Large BFU-MK colony (LG) was defined as a cluster of 50–200 and more MKs with 2 or more different foci of development. Medium BFU-MK colony (MD) was defined as a cluster of 21–50 MKs. Small CFU-MK colony (SM) contained 3–20 MKs. Mixed (MX) colony was defined as a cluster of 2 or more MKs mixed with other hematopoietic cells. Non-MK (NON) colony was defined as cell cluster negative for GPIIb staining. Results of 6 experiments are presented as box-and-whisker plots, with the line within the box indicating the median, the box representing 25% and 75% percentiles, and whiskers indicating maximum and minimum values (*P < 0.05; **P < 0.005, compared with scrambled control). (D and E) Typical small CFU-MK colonies formed after transduction with control (D) and tescalcin-specific (E) shRNA lentiviral vectors.
Downregulation of tescalcin inhibits GPIIb gene transcription. To determine whether tescalcin inhibits the expression of GPIIb at the mRNA or protein level, we analyzed the relative expression of GPIIb mRNA in HEL and K562 cells. RQ-PCR with TaqMan set of probe and primers (Applied Biosystems) showed that tescalcin knockdown in HEL-Tsc(–) results in almost complete inhibition of GPIIb expression (Figure 7A). Such a reduction of GPIIb mRNA level in cells lacking tescalcin can be explained by either lower transcriptional activity of GPIIb promoter or by mRNA instability. To assess the transcriptional activity of the full-length human GPIIb promoter we used a luciferase reporter assay. The GPIIb promoter was highly active in the untreated HEL-Ctrl(–) and less so in K562-Ctrl(–) cells, consistent with the higher basal expression of tescalcin in HEL cells. However, in both HEL-Tsc(–) and K562-Tsc(–) cells, the activity of the promoter was blocked (Supplemental Figure 3). Therefore, our results show that the downregulation of tescalcin strongly inhibits GPIIb gene transcription.
Tescalcin controls expression of Ets family transcription factors at the mRNA level. (A) Total RNA was isolated from HEL-Ctrl(–) (white bars) and HEL-Tsc(–) cells (gray bars). Gene expression was analyzed by RQ-PCR (TaqMan) with tescalcin-, GPIIb-, Ets-1–, Ets-2–, Fli-1–, PU.1-, GATA-1–, and MafB-specific primers as described in Methods. Data are expressed relative to wild-type HEL cells (n = 3; mean ± SD). (B) Control HEL-Ctrl(–) and tescalcin knockdown HEL-Tsc(–) cells were cultured in absence or presence of PMA for 72 hours. Expression of Ets-1 and Fli-1 transcription factors was assessed by Western blot with specific antibodies. (C) Total RNA was isolated from K562-Ctrl(+) and tescalcin overexpressing K562-Tsc(+) cells cultured in absence or presence of PMA for 18 hours. RQ-PCR using GPIIb-, Ets-1–, Ets-2–, and GATA-1–specific primers was performed as described in A. Data are expressed relative to wild-type unstimulated K562 (n = 3; mean ± SD). (D) Cytospin preparations of K562 cells on glass slides were stained with antibody to tescalcin and then with FITC-labeled donkey anti-rabbit antibody. Slides were mounted with DAPI containing mounting media (Prolong; Molecular Probes) to identify cell nuclei. All images were acquired at ×1,000 magnification. Scale bar: 10 μm. (E) Outline of tescalcin-mediated pathway. TPO binds to its receptor, c-Mpl, on the cell surface and activates the ERK signaling cascade. Treatment of cells with PMA leads to a similar activation of ERK. The sustained ERK activity causes upregulation of tescalcin, which in turn promotes expression of the Ets family genes, orchestrating terminal differentiation along megakaryocytic lineage.
Tescalcin is necessary and sufficient for the expression of Ets family genes. The expression of GPIIb mRNA in megakaryocytopoiesis is known to be controlled by a number of transcriptional factors, such as GATA-1, Ets-1, Ets-2, Fli-1, PU.1, MafB, and others (41, 47–51). We used RQ-PCR to quantitatively examine the expression of these factors in cells with both knocked down and overexpressed tescalcin.
Our results demonstrate that the levels of Ets-1 and Fli-1 mRNA in HEL-Tsc(–) cells were almost undetectable. Ets-2 and PU.1 mRNAs were downregulated 10- and 4-fold, respectively, by tescalcin knockdown. The levels of GATA-1 and MafB mRNAs were not affected (Figure 7A), indicating the specificity of the tescalcin knockdown effect. Knockdown of tescalcin in K562 had a similar effect on expression of Ets family genes (data not shown). Consistent with the RQ-PCR data, Western blot analysis confirmed the decrease of Ets-1 and Fli-1 protein levels in cells lacking tescalcin. PMA stimulation caused only a minor increase in their expression (Figure 7B). Importantly, in overexpressing tescalcin K562-Tsc(+) cells the basal levels of Ets-1 and Ets-2 mRNA along with their target gene, GPIIb, was significantly higher. The level of GATA-1 mRNA in these cells was not affected (Figure 7C). PMA stimulation for only 18 hours did not have a significant effect on GPIIb or Ets-2 mRNA levels in control cells, although their expression was induced almost 4-fold in cells overexpressing tescalcin (Figure 7C). Downregulation of GATA-1 upon PMA stimulation is typical for K562 (52).
Taken together, our results demonstrate that expression levels of tescalcin in K562 and HEL cells regulate their megakaryocytic differentiation by controlling the genes encoding transcription factors of the Ets family.