This Article Abstract Full Text (PDF) Data Supplement Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by den Dekker, E. Articles by Akkerman, J.-W. N. Search for Related Content PubMed PubMed Citation Articles by den Dekker, E. Articles by Akkerman, J.-W. N. Related Collections Cell signalling/signal transduction Platelet function inhibitors Platelets

(Arteriosclerosis, Thrombosis, and Vascular Biology. 2002;22:179.)
© 2002 American Heart Association, Inc.

Thrombosis

Cyclic AMP Raises Intracellular Ca²⁺ in Human Megakaryocytes Independent of Protein Kinase A

Els den Dekker; Johan W.M. Heemskerk; Gertie Gorter; Hans van der Vuurst; José Donath; Christine Kroner; Katsuhiko Mikoshiba; Jan-Willem N. Akkerman

From the Thrombosis and Haemostasis Laboratory (E.d.D., G.G., H.v.d.V., J.D., C.K., J.-W.N.A.), Department of Haematology, University Medical Center Utrecht and Institute for Biomembranes, Utrecht University, Utrecht, and the Department of Biochemistry and Human Biology (J.W.M.H.), Maastricht University, Maastricht, The Netherlands, and the Department of Molecular Neurobiology (K.M.), Institute of Medical Science, Tokyo University, Tokyo, Japan.

Correspondence to Els den Dekker, Thrombosis and Haemostasis Laboratory, Department of Haematology, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands. E-mail e.dendekker{at}lab.azu.nl

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
The immature megakaryoblastic cell line MEG-01 responds to iloprost with an increase in cytosolic Ca²⁺ and cAMP. The Ca²⁺ response is almost absent in CHRF-288-11 cells, but cAMP formation is preserved in this more mature megakaryoblastic cell line. Also, in human hematopoietic stem cells, iloprost induces a Ca²⁺ response and cAMP formation. The Ca²⁺ response is downregulated during megakaryocytopoiesis, but cAMP formation remains unchanged. The Ca²⁺ increase may be caused by cAMP-mediated inhibition of Ca²⁺ sequestration, because it is (1) independent of Ca²⁺ entry; (2) mimicked by forskolin, an activator of adenylyl cyclase, and isobutylmethylxanthine, an inhibitor of phosphodiesterases; and (3) preserved in the presence of inhibitors of protein kinase A and inositol-1,4,5-triphosphate receptors. The small GTPase Rap1 has been implicated in the control of Ca²⁺ sequestration. Indeed, Rap1 activation parallels the iloprost- and forskolin-induced Ca²⁺ increase and is unaffected by the calcium chelator 1,2-bis(2-aminophenoxy)ethane-N,N,N',N',-tetraacetic acid-AM. These findings reveal a novel mechanism for elevating cytosolic Ca²⁺ by cAMP, possibly via GTP-Rap1.

Key Words: calcium • cAMP • stem cells • megakaryocytes • Rap1

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Megakaryocytopoiesis is accompanied by downregulation of stem cell properties and upregulation of properties that later determine platelet functions. One of the first characteristics of megakaryocyte differentiation is the appearance of the fibrinogen receptor, integrin _IIbß₃ (glycoprotein IIb/IIIa, or CD41/CD61), together with the disappearance of the stem cell marker CD34.¹ A second early event is the synthesis of von Willebrand factor (vWF), which starts in immature, both CD61⁺ and CD34⁺ megakaryocytes. At a later stage, the vWF receptor, glycoprotein Ib (CD42b), is expressed, which marks the beginning of polyploidization.¹ The transition from proliferating to differentiating megakaryocytes is accompanied by loss of nuclear-associated acetylcholinesterase activity.²

The megakaryoblastic cell lines MEG-01, DAMI, and CHRF-288-11 have properties in common with normal megakaryocytes at different stages of maturation.³ The immature MEG-01 cells already show an increase in cytosolic Ca²⁺ concentration, [Ca²⁺]_i, on stimulation by thrombin and platelet-activating factor. The more mature DAMI and CHRF-288-11 cells upregulate this property and, in addition, become sensitive to thromboxane A₂. These cells respond to the prostacyclin analogue iloprost with an increase in cAMP, a response that is also upregulated in the more mature cell lines. Hence, the immature megakaryoblast already has the capacity to regulate Ca²⁺ and cAMP via mechanisms also seen in platelets. An interesting exception is that in MEG-01 cells, iloprost-induced cAMP formation is accompanied by a rise in [Ca²⁺]_i. This is in sharp contrast with platelets, which do not raise [Ca²⁺]_i when treated with prostacyclin and show Ca²⁺ responses by thrombin and other platelet-activating agents that are completely blocked by an increase in cAMP.⁴

In the present study, we aimed to clarify how prostacyclin raises [Ca²⁺]_i in immature megakaryoblasts. Studies were focused on iloprost-induced signaling pathways to [Ca²⁺]_i and on the small GTPase Rap1, which has been implicated in Ca²⁺ regulation in platelets in the same way that phospholamban regulates Ca²⁺ signaling in cardiac and muscle cells. Earlier studies have shown that megakaryoblastic cell lines express the prostacyclin receptor⁵ Rap1⁶ and the 97-kDa sarco/endoplasmic reticulum Ca²⁺-ATPase SERCA3b, which functions in Ca²⁺ sequestration in platelets.⁶ The data show that in addition to MEG-01 cells, human stem cells and immature megakaryocytes respond to iloprost with an increase in cAMP and a rise in [Ca²⁺]_i and that cAMP controls cytosolic Ca²⁺, possibly via Rap1-GTP.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
For the Methods section, please see http://www.atvb.ahajournals.org.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Iloprost Induces a Ca²⁺ Increase in Megakaryoblastic Cell Lines
Stimulation of MEG-01 cells with iloprost induced both an increase in [Ca²⁺]_i and a rise in cAMP, confirming earlier observations.³ In CHRF-288-11 cells, the rise in [Ca²⁺]_i had almost disappeared, but the increase in cAMP was 2-fold higher (Figures 1A and 1B). These data are consistent with the concept that cell maturation is accompanied by the downregulation of iloprost-induced Ca²⁺ increases and the upregulation of iloprost-induced cAMP formation. Stimulation of MEG-01 cells with thrombin induced a Ca²⁺ response similar to that of iloprost. Thrombin induced a 3-fold higher Ca²⁺ response in CHRF-288-11 than in MEG-01 cells, illustrating that the failure of iloprost to raise [Ca²⁺]_i in CHRF-288-11 cells was not caused by abnormalities in Ca²⁺ storage or influx. To assess the contributions of Ca²⁺ mobilization and influx in the total [Ca²⁺]_i increase, experiments were repeated in Ca²⁺-free medium. Ca²⁺ mobilization by iloprost and thrombin was 42±6% and 31±12%, respectively, of the peak Ca²⁺ level in MEG-01 cells; for CHRF-288-11 cells, these values were 18±3% and 22±2%, respectively. As expected, stimulation with thrombin failed to raise cAMP levels in both cell lines. Interestingly, in MEG-01 cells, iloprost was able to induce a further Ca²⁺ rise after thrombin (Figure 1C), suggesting that these 2 agonists raise [Ca²⁺]_i by a different mechanism.

View larger version (30K): [in this window] [in a new window]   Figure 1. Calcium and cAMP increases in megakaryoblastic cell lines. Megakaryoblastic cell lines were loaded with fura-2/AM, and Ca2+ increases, in the presence of 1 mmol/L extracellular Ca2+ (A, C), and cAMP formation (B), induced by iloprost (1 µmol/L) or thrombin (5 U/mL), were measured in MEG-01 cells (open bars) and CHRF-288-11 cells (hatched bars). Basal levels of [Ca2+]i were 131±15 and 112±9 nmol/L for MEG-01 and CHRF-288-11 cells, respectively. Basal levels of cAMP were 12±3 and 8.0±3 pmol/106 cells for MEG-01 and CHRF-288-11 cells, respectively. Data are expressed as mean±SD (n=3).

Iloprost Induces a Ca²⁺ Increase in Megakaryocytes
To investigate whether iloprost induced similar increases in Ca²⁺ and cAMP in megakaryocytes, CD34⁺ stem cells were immunomagnetically purified from umbilical cord blood and cultured in vitro with recombinant human thrombopoietin and recombinant human stem cell factor. After 7 and 14 days, immature and mature megakaryocytes were isolated by a second immunomagnetic sorting based on expression of CD61 and CD42b, respectively. The 3 isolation procedures resulted in almost pure (>95%) suspensions of CD34-, CD61-, and CD42b-expressing cells. The CD34⁺ cells were predominantly diploid and contained few megakaryocytes. The immature megakaryocytes, represented by the CD61-expressing cells, were diploid (77%) and tetraploid (23%) and contained properties of stem cells (44%) and mature megakaryocytes (25%), illustrating an intermediate stage of megakaryocyte maturation. The mature megakaryocyte suspension consisted for >95% of CD61⁺ and CD42b⁺ cells, without a further increase in ploidy. For a detailed characterization of the cell suspensions, please see Table I at http://www.atvb.ahajournals.org. As shown in Figure 2A (upper panel), iloprost induced a Ca²⁺ response in stem cells as well as in immature and mature megakaryocytes. This response was the sum of Ca²⁺ mobilization (60%) and influx (40%), and the peak value of [Ca²⁺]_i was higher in stem cells (36±13 nmol/L) and immature megakaryocytes (33±3 nmol/L) than in mature megakaryocytes (12±6 nmol/L), illustrating a similar downregulation as observed in mature megakaryoblastic cell lines. Similar results were found with prostacyclin (data not shown). To investigate whether the Ca²⁺ measurements in cell suspensions reflected the responses of the individual cells, studies were repeated in single cells. These measurements showed that 30% (11 of 37) of the stem cells responded to iloprost with a Ca²⁺ response (Figure 2A, lower panel). This percentage was the same for immature megakaryocytes (7 of 23) but decreased to 15% for mature megakaryocytes (8 of 54). Both the number of iloprost-responsive cells and the magnitude of the Ca²⁺ increase declined during cell maturation, which is in line with the findings in cell suspensions. No such downregulation was observed for the thrombin-induced Ca²⁺ increases (Figure 2B). Stem cells responded to thrombin with a rise in [Ca²⁺]_i of 85±15 nmol/L, and in immature and mature megakaryocytes, the increases were 285±34 and 440±35 nmol/L, respectively (Figure 2B, upper panel). Analysis of single-cell responses showed that 95% of the stem cells (35 of 37), 91% of the immature megakaryocytes (31 of 34), and 90% (36 of 40) of the mature megakaryocytes responded to thrombin and that the Ca²⁺ responses increased with increasing cell maturity (Figure 2B, lower panel). The same studies in MEG-01 cells showed that iloprost and thrombin induced a Ca²⁺ increase in 96±4% and 95±2%, respectively, of the cells (not shown).

View larger version (28K): [in this window] [in a new window]   Figure 2. Ca2+ responses in stem cells and cultured megakaryocytes. A (upper panel) shows Ca2+ responses in the presence of 1 mmol/L extracellular Ca2+ induced by iloprost (1 µmol/L) in suspensions of stem cells (1), immature (2), and mature megakaryocytes (3). A representative tracing for 3 observations with similar results is shown. A (lower panel) shows Ca2+ responses of single immobilized cells: (1) stem cells (day 1 cells adhering to anti-CD34); (2) immature megakaryocytes (CD42blow day 7 cells adhering to anti-CD61); and (3) mature megakaryocytes (CD42bhigh day 14 cells adhering to anti-CD61). A representative tracing for at least 7 observations with similar results is shown. B shows similar experiments for cells stimulated by 1 U/mL thrombin.

Iloprost-Induced Ca²⁺ Signaling Is Mimicked by cAMP-Elevating Agents
Iloprost is known to bind with high affinity to receptors of the IP class (which bind prostaglandins of the I type) and with a low affinity to receptors of the EP1 class (which bind prostaglandins of the E type).⁷ To clarify which receptor takes part in iloprost signaling to Ca²⁺, experiments were repeated with carbaprostacylin, which is a more specific agonist of IP receptors than is iloprost.⁸ Figure 3A shows that carbaprostacylin induced an increase in [Ca²⁺]_i in immature megakaryocytes, consisting of several Ca²⁺ spikes. The response was observed in 38% (11 of 29) of the cells. The increase was also seen in nominally Ca²⁺-free buffer (Figure 3A, insert), illustrating that carbaprostacylin triggered both mobilization and influx. Stimulation of IP receptors is known to activate G_s, the trimeric G protein that activates adenylyl cyclase. To investigate whether this enzyme takes part in iloprost-induced Ca²⁺ signaling, immature megakaryocytes were stimulated with the adenylyl cyclase activator forskolin. In 23% (11 of 48) of the cells, this treatment induced a Ca²⁺ response (Figure 3B) and Ca²⁺ mobilization (Figure 3B, insert) similar to that of iloprost, indicating that iloprost presumably signals to Ca²⁺ by way of adenylyl cyclase.

View larger version (22K): [in this window] [in a new window]   Figure 3. cAMP-elevating agents induce release of Ca2+ from stores and influx. A and B show Ca2+ responses of immobilized immature megakaryocytes in the presence of 1 mmol/L extracellular Ca2+ after stimulation with (A) carbaprostacylin (1 µmol/L) or (B) forskolin (100 µmol/L). The inserts show Ca2+ responses in nominally Ca2+-free buffer. A representative tracing for at least 10 observations with similar results is shown. C shows Ca2+ responses of suspensions of MEG-01 cells in the presence of 1 mmol/L extracellular Ca2+ after stimulation with forskolin (40 µmol/L, left panel), IBMX (500 µmol/L, middle panel), and iloprost (1 µmol/L) after preincubation with IBMX (500 µmol/L, 5 minutes, 37°C; right panel).

In MEG-01 cells, forskolin (40 to 100 µmol/L) induced a Ca²⁺ increase of 105±20 nmol/L (Figure 3C, left panel). The structurally dissimilar compound isobutylmethylxanthine (IBMX, 500 µmol/L), which inhibits phosphodiesterases and thereby prevents cAMP breakdown, also induced an increase in [Ca²⁺]_i, although this response was much lower than the forskolin-induced Ca²⁺ increase (Figure 3C, middle panel). A similar, albeit weaker, Ca²⁺ increase was also induced by dibutyryl cAMP (data not shown). To further investigate the role of cAMP in Ca²⁺ increases, the effect of an inhibitor of cAMP metabolism was investigated. MEG-01 cells were preincubated with IBMX (500 µmol/L, 5 minutes, 37°C) and subsequently treated with iloprost (1 µmol/L). As shown in Figure 3C (right panel), IBMX treatment potentiated the iloprost-induced rise in Ca²⁺ by 15%, strongly supporting the concept that cAMP plays a role in the iloprost-induced Ca²⁺ increase.

As expected, stimulation of IP receptors and direct activation of adenylyl cyclase induced formation of cAMP in cultured megakaryocytes (please see Table II at http://www.atvb.ahajournals.org). Thus, the increase in [Ca²⁺]_i by these treatments was accompanied by an increase in cAMP.

The Forskolin-Induced Ca²⁺ Increase Does Not Involve IP₃ Receptors or Protein Kinase A
To investigate whether increases in intracellular inositol triphosphate (IP₃) levels mediated the forskolin-induced Ca²⁺ response, MEG-01 cells in nominally Ca²⁺-free buffer were treated with the IP₃ receptor antagonist 2-aminoethoxydiphenylborate (2-APB). In platelets, 100 µmol/L 2-ABP completely abolished thrombin- and thromboxane A₂-induced Ca²⁺ responses (Figure 4A, left panel).⁹ At concentrations of 50 µmol/L (not shown) and 100 µmol/L (Figure 4A), 2-APB partially inhibited the forskolin-induced Ca²⁺ increases (middle panel) and totally inhibited the thrombin-induced Ca²⁺ increases (right panel) in MEG-01 cells. Similar results were obtained with immature megakaryocytes (data not shown). Thus, in contrast to thrombin, which fully depended on IP₃ receptors for raising [Ca²⁺]_i, forskolin-induced Ca²⁺ increases were partly independent of IP₃ receptors.

View larger version (31K): [in this window] [in a new window]   Figure 4. Forskolin-induced Ca2+ increases do not involve IP3 receptors or PKA. A, Platelets and MEG-01 cells were preincubated with the IP3 receptor inhibitor 2-APB (100 µmol/L, 2 minutes, 37°C), and Ca2+ increases were measured in nominally Ca2+-free buffer on addition of thrombin (1 U/mL; left and right panels) or forskolin (100 µmol/L; middle panel). B, MEG-01 cells were preincubated with the PKA inhibitor H89 (15 minutes, 37°C, 10 µmol/L), and phosphorylation of VASP induced by forskolin (100 µmol/L) was determined (left panel). Ca2+ increases were measured in H89-treated cells in nominally Ca2+-free buffer on addition of forskolin (100 µmol/L, middle panel) or thrombin (1 U/mL, right panel). C, Phosphorylation of VASP (left panel) and Ca2+ increases in nominally Ca2+-free buffer (right panel) were measured after stimulation of PKA with the activator Sp-5,6-DCl-cBIMPS (800 µmol/L) and subsequently forskolin (100 µmol/L, right panel). Tracings are representative for 3 observations with similar results.

To investigate whether the forskolin-induced Ca²⁺ increase also involved protein kinase A (PKA), MEG-01 cells were treated with the PKA inhibitor H89. As shown in Figure 4B (left panel), the forskolin-induced phosphorylation of vasodilator-stimulated phosphoprotein (VASP) was completely abolished by pretreatment with H89, indicating that H89 is a good inhibitor of PKA in MEG-01 cells. Treatment with H89, however, did not abolish the forskolin-induced Ca²⁺ increase. Instead, a slight but consistent increase in the release of Ca²⁺ from intracellular stores was observed (Figure 4B, middle panel). PKA inhibition thus "unmasks" a slight suppression of Ca²⁺ increases by PKA. The same effect was seen when MEG-01 cells were stimulated with thrombin (Figure 4B, right panel). Immature megakaryocytes reacted the same way as MEG-01 cells did (data not shown). The specific PKA activator Sp-5,6-DCl-cBIMPS induced the same extent of VASP phosphorylation as forskolin (Figure 4C, left panel) but did not raise [Ca²⁺]_i in MEG-01 cells (Figure 4C, right panel). Even after stimulation of PKA with Sp-5,6-DCl-cBIMPS, forskolin induced a Ca²⁺ increase (Figure 4C, right panel). These findings argue in favor of a direct effect of cAMP on Ca²⁺ homeostasis, without involvement of PKA.

Iloprost and Thrombin Induce Rap1 Activation
The small GTPase Rap1 has been implicated in the regulation of Ca²⁺ sequestration, because (1) it coimmunoprecipitates with the sarco/endoplasmic reticulum Ca²⁺-ATPase isoform 3b (SERCA3b) and (2) binding of GTPS to Rap1 has been suggested to be related to SERCA3b activity.^10,11 Because the guanine nucleotide exchange factor (GEF) for Rap1, Epac1/2, is a target for PKA-independent signaling by cAMP, Rap1 could be the link between cAMP and Ca²⁺.^12,13 Using a pulldown technique based on the specific binding of active, GTP-bound Rap1 to the Rap-binding domain of RalGDS, we investigated the effect of iloprost and forskolin on Rap1 activation. Because the quantities of cultured megakaryocytes were below the detection limit of this technique, the studies were performed with MEG-01 and CHRF-288-11 cells only. Results from different experiments were related to internal standards as defined in the legend to Figure 5. In MEG-01 cells, addition of iloprost induced a 10-fold, sustained increase in GTP-Rap1 (Figure 5A). A much weaker 5-fold activation of Rap1 was observed on stimulation with thrombin, which was transient and disappeared after 1 minute. Incubation of blots with antibodies that specifically recognize Rap1a or Rap1b showed that iloprost activated both Rap1a and Rap1b to the same extent and with the same kinetics (Figure 5B). In CHRF-288-11 cells, the responses were completely different. In these cells, iloprost induced only little Rap1 activation, but thrombin induced a 15-fold increase in GTP-Rap1, which remained high throughout the incubation period (Figure 5C). Thus, the downregulation of iloprost-induced Ca²⁺ responses and upregulation of thrombin-induced Ca²⁺ responses, observed when immature and mature megakaryoblastic cell lines were compared, were accompanied by similar changes in the formation of GTP-Rap1, suggesting that these responses go hand in hand.

View larger version (23K): [in this window] [in a new window]   Figure 5. Rap1 activation in MEG-01 and CHRF-288-11 cells. MEG-01 cells (A, B) and CHRF-288-11 cells (C) were treated with iloprost (1 µmol/L, ) or thrombin (1 U/mL, ). After lysis, GTP-Rap1 was precipitated with glutathione S-transferase-coupled Rap binding domain of guanine nucleotide dissociation stimulator for Ral-coupled glutathione-agarose beads and identified by Western blotting with a monoclonal antibody against Rap1 (A, C) or polyclonal antibodies specifically recognizing Rap1a or Rap1b (B). To quantify the Rap1 activation induced by iloprost and thrombin, films were scanned and bands were quantified with ImageQuant software. The pixel density for each lane within the same experiment and analyzed on the same film was related to the reference sample present on the same blot. Data represent relative amounts of GTP-Rap1 (mean±SD, n=3).

cAMP Induces Rap1 Activation and Ca²⁺ Signaling in MEG-01 but Not in CHRF-288-11 Cells
To investigate whether direct activation of cAMP formation induced similar changes in [Ca²⁺]_i and GTP-Rap1 as seen with iloprost, MEG-01 and CHRF-288-11 cells were treated with forskolin. In agreement with the effect of iloprost, forskolin induced an increase in [Ca²⁺]_i and formation of GTP-Rap1 in MEG-01 but not in CHRF-288-11 cells (Figures 6A, 6B, left panel, and 6C). The onset of forskolin-induced Ca²⁺ increases in MEG-01 cells was 10 seconds later than the beginning of Rap1 activation, suggesting that Rap1 activation precedes the Ca²⁺ increase. Rap1 was also activated by the phosphodiesterase inhibitor IBMX (Figure 6B, right panel), although activation was much weaker and slower than that induced by forskolin. This difference corresponds with the difference in Ca²⁺ responses induced by these compounds (Figure 3C, left and middle panels).

View larger version (35K): [in this window] [in a new window]   Figure 6. Forskolin releases Ca2+ from stores and activates Rap1 independent of PKA. MEG-01 and CHRF-288-11 cells were treated with forskolin (100 µmol/L), and Ca2+ increases in nominally Ca2+-free buffer (A) and Rap1 activation (B, left panel) were measured. B, right panel, shows Rap1 activation in MEG-01 cells induced by IBMX (500 µmol/L). Representative examples of 3 experiments with similar results are shown. GTP-Rap1 bands on Western blots were scanned and quantified as described in the Methods section and the legend to Figure 5. Mean±SD (n=3) for activated Rap1 in MEG-01 and CHRF-288-11 cells is shown in C. D, MEG-01 cells were preincubated with H89 (15 minutes, 37°C, 10 µmol/L) and BAPTA-AM (30 minutes, 37°C, 30 µmol/L) or vehicle, and Rap1 activation induced by forskolin (100 µmol/L) was determined after 15 seconds. Data show mean±SD, n=3. GTP-Rap1 was significantly increased after forskolin treatment (P<0.05, n=3, compared with control). Pretreatment with H89 and BAPTA-AM did not significantly change the forskolin-induced Rap1 activation (P>0.05). Rap1 phosphorylation was determined by a gel-shift assay for samples derived from MEG-01 cells treated with forskolin (100 µmol/L; E, upper panel) and platelets treated with iloprost (5 µmol/L; E, lower panel).

To clarify the role of PKA in the forskolin-induced Rap1 activation, studies were repeated in the presence of the PKA inhibitor H89. As shown in Figure 6D, the forskolin-induced Rap1 activation was not affected by this inhibitor. To determine whether Rap1 activation is an upstream step of Ca²⁺ increases or a result of changes in [Ca²⁺]_i, Ca²⁺ increases were prevented by treating the cells with 1,2-bis(2-aminophenoxy)ethane-N,N,N',N',-tetraacetic acid (BAPTA)-AM. The Ca²⁺-chelating capacity of BAPTA-AM was sufficient to abolish the forskolin-induced Ca²⁺ increase completely (data not shown). As also shown in Figure 6D, this treatment failed to interfere with the activation of Rap1. Together, these data favor the concept that Rap1 is activated by a rise in cAMP and that this activation is independent of changes in [Ca²⁺]_i.

cAMP-Induced Rap1 Activation Is Not Accompanied by Phosphorylation
To determine whether Rap1 phosphorylation by PKA occurs concurrently with Rap1 activation, MEG-01 cells were treated with forskolin (100 µmol/L), and Rap1 phosphorylation was determined by a gel-shift assay (Figure 6E, upper panel). Because Rap1 has been described to be phosphorylated in iloprost-treated platelets after 15 minutes, control samples of platelets treated with iloprost (5 µmol/L) were analyzed for Rap1 phosphorylation (Figure 6E, lower panel). At the times when forskolin induced Rap1 activation in MEG-01 cells (between 5 and 25 seconds), Rap1 was not phosphorylated. However, at later times (15 and 30 minutes), weak phosphorylation was visible (not shown). In platelets, a clear iloprost-induced phosphorylation of Rap1 was observed after 15 and 30 minutes, which corresponds with results from previous studies.^14,15

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The present report shows that iloprost induces a Ca²⁺ response in human hematopoietic stem cells and immature megakaryocytes. Analysis of the signaling pathway by which iloprost raises [Ca²⁺]_i reveals a central role of cAMP. First, iloprost-induced [Ca²⁺]_i increases are accompanied by cAMP production. Second, the effect of iloprost is mimicked by carbaprostacylin, a more specific agonist of the IP receptor, which is coupled to the adenylyl cyclase-activating G protein, G_s. Third, direct activation of adenylyl cyclase by forskolin raises cAMP and [Ca²⁺]_i. Fourth, the phosphodiesterase inhibitor IBMX, which prevents cAMP breakdown, induces a Ca²⁺ response and potentiates the iloprost-induced rise in [Ca²⁺]_i. Together, these data illustrate that in immature megakaryocytes, rises in cAMP and [Ca²⁺]_i go hand in hand.

The observations that dibutyryl cAMP, IBMX, and forskolin raise [Ca²⁺]_i indicate that cAMP is an upstream regulator of Ca²⁺. Elevated cAMP levels have been shown to induce Ca²⁺ mobilization in HEK-293 cells, hepatocytes, neuronal cells, and articular chondrocytes.^16–19 In these cells, the increase in [Ca²⁺]_i was caused by PKA-mediated phosphorylation of IP₃ receptors. This resulted in sensitization for IP₃, leading to a 4-fold leftward shift of the dose-response curve of IP₃-mediated Ca²⁺ release.¹⁷

The present findings in MEG-01 cells and immature megakaryocytes differ from those observations, because neither PKA nor IP₃ receptors seem to be involved in the cAMP-induced [Ca²⁺]_i increase. The PKA inhibitor H89, which inhibits PKA-mediated suppression of IP₃-induced Ca²⁺ responses in rat megakaryocytes²⁰ and completely suppresses forskolin-induced VASP-Ser157 phosphorylation in MEG-01 cells, did not suppress the forskolin-induced Ca²⁺ response. Moreover, direct activation of PKA by the specific activator Sp-5,6-DCl-cBIMPS induced the same extent of PKA activation as forskolin, but this did not result in an increase in [Ca²⁺]_i. This indicates that PKA activation does not lead to a Ca²⁺ increase. The IP₃ receptor inhibitor 2-APB, which abolishes thrombin- and thromboxane A₂–mediated Ca²⁺ responses in platelets,⁹ also blocked the thrombin-induced Ca²⁺ response in megakaryocytes. However, this inhibitor only partly blocked the forskolin-induced Ca²⁺ response, illustrating IP₃-dependent as well as IP₃-independent routes. Cerebellar microsomes,⁹ which express predominantly the type 1 IP₃ receptor, and submandibular gland cells, which possess mainly type 2 and type 3 IP₃ receptors,²¹ show the same sensitivity to 2-APB for IP₃-mediated Ca²⁺ responses (authors’ unpublished results, 2001). Thus, it is unlikely that forskolin raises Ca²⁺ via an IP₃ receptor subtype that is resistant to 2-APB inhibition. An explanation for the IP₃ receptor-mediated Ca²⁺ response might be sought in the Ca²⁺-induced Ca²⁺ release triggered by the cAMP-mediated release of Ca²⁺ from intracellular stores. Although these responses occur independently of PKA, it is clear that phosphorylation of IP₃ receptors by this kinase plays a role in MEG-01 cells like it does in platelets.²² PKA inhibition leads to increased Ca²⁺ rises, illustrating suppression of Ca²⁺ mobilization by PKA-mediated phosphorylation of IP₃ receptors.

The findings in this report reveal a novel mechanism by which cAMP directly regulates [Ca²⁺]_i, independent of PKA and IP₃ receptors. Interestingly, iloprost further increases [Ca²⁺]_i after thrombin in MEG-01 cells, indicating that these 2 compounds activate different Ca²⁺-mobilizing mechanisms. Few PKA-independent effects of cAMP have been reported so far. Recently, De Rooij et al¹² and Kawasaki et al¹³ described that the small GTPase Rap1 is activated by cAMP via cAMP-sensitive guanine nucleotide exchange factors (cAMP GEFs). Two types of cAMP GEF for Rap1 have been reported: cAMP-GEF-I, also called exchange protein, directly activated by cAMP (Epac), and cAMP-GEF-II. Interestingly, earlier work by Corvazier et al¹¹ and Lacabaratz-Porret et al¹⁰ indicated that Rap1 might be involved in Ca²⁺ regulation in platelets. First, proteins in crude platelet plasma membrane vesicles, ranging in molecular mass from 22 to 29 kDa, bound GTPS. One of these proteins was identified as Rap1, and GTPS binding was accompanied by inhibition of Ca²⁺-ATPases in these vesicles. Second, Rap1 coimmunoprecipitated with the sarco/endoplasmic reticulum Ca²⁺-ATPase isoform 3b (SERCA3b), suggesting a physical interaction of these 2 proteins. The association of Rap1 with SERCA3b was lost on PKA-mediated phosphorylation of Rap1 on Ser179. The present results show that Rap1 activation is correlated with the release of Ca²⁺ from stores in stem cells and immature megakaryocytes. The cAMP-elevating agents iloprost, forskolin, and IBMX induce an increase in the active, GTP-bound Rap1 together with Ca²⁺ release from stores in MEG-01 cells. Both Rap1a and Rap1b were activated by iloprost, indicating that cAMP-mediated Rap1 activation does not involve a specific subtype. The Rap1 activation and Ca²⁺ increase have similar kinetics and are independent of PKA, as shown by the insensitivity to H89. Rap1 was not phosphorylated by forskolin in MEG-01 cells within the time range (5 to 25 seconds) when this compound induced a Ca²⁺ increase and Rap1 activation. This result corresponds with that obtained in platelets, in which iloprost and prostacyclin induced Rap1 phosphorylation after a lag time of 15 minutes or later.¹⁴ Furthermore, Rap1 activation was not affected by PKA-mediated phosphorylation.¹⁵ Because the Ca²⁺ chelator BAPTA-AM did not suppress the cAMP-induced Rap1 activation, it is unlikely that Rap1 is downstream of Ca²⁺ increases. These findings are in correspondence with the hypothesis that Rap1 is a molecular "switch" that regulates SERCA activity: the cAMP-induced Rap1 activation is correlated with the inhibition of SERCA and a rise in [Ca²⁺]_i (present data), and the PKA-mediated phosphorylation of Rap1 relieves the inhibition, resulting in activation of Ca²⁺-ATPases and a decrease in [Ca²⁺]_i.^10,11 The latter property is also observed in Ca²⁺ regulation by phospholamban in cardiac and muscle cells.^23,24 Phosphorylation of this 24-kDa protein by PKA triggers stimulation of cardiac sarcoplasmic reticulum Ca²⁺-ATPases, leading to a fall in [Ca²⁺]_i.

In CHRF-288-11 cells, the Rap1 activation and release of Ca²⁺ from stores by forskolin had disappeared, suggesting that one of the steps in cAMP-induced Rap1 activation is downregulated during maturation and is probably absent in platelets. Indeed, previous studies indicated that cAMP does not induce Rap1 activation in platelets.¹⁵ Although the expression of Rap1 is upregulated during megakaryocytopoiesis,^6,10,25 the mechanism of cAMP-mediated Rap1 activation is downregulated. A likely candidate to be downregulated is a cAMP-dependent GEF for Rap1. However, we were unable to demonstrate expression of cAMP-GEF-I and cAMP-GEF-II mRNA in MEG-01 cells, CHRF-288-11 cells, or platelets by reverse transcription-polymerase chain reaction (data not shown). This suggests that either another subtype of cAMP-dependent GEF may exist or that downregulation of a yet-unknown cAMP-inhibitable GTPase-activating protein for Rap1 causes the loss of cAMP-induced Ca²⁺ rises during megakaryocyte maturation.

In MEG-01 and CHRF-288-11 cells, thrombin induced Rap1 activation, a process that is upregulated during maturation. This points to a major difference between Rap1 activation by forskolin and that by thrombin. In platelets, the first phase of thrombin-induced Rap1 activation critically depends on a rise in Ca²⁺ and is followed by an activation phase, in which PKC, phosphatidylinositol-3-kinase, and the integrin _IIbß₃ play a role.^15,26 The first phase of Rap1 activation appears to be present already in MEG-01 cells, possibly reflecting the activity of Ca²⁺-dependent GEFs, such as Ca²⁺- and diacylglycerol-dependent GEF. The higher thrombin-induced activation of Rap1 in CHRF-288-11 cells is probably due to the higher Ca²⁺ response compared with that in MEG-01 cells. The sustained Rap1 activation in CHRF-288-11 cells may arise from a more sustained PKC or phosphatidylinositol-3-kinase activity in these more mature cells.

An interesting question is whether prostacyclin-induced Rap1 activation also occurs in vivo and what its role might be in megakaryocytopoiesis. Functions of active Rap1 include induction of cell adhesion,²⁷ inhibition of gene expression,²⁸ and inhibition of cell proliferation by suppression of Ras-mediated activation of Raf and mitogen-activated protein kinase.^29,30 It is possible that immature megakaryocytes become exposed to prostacyclin in vivo, because bone marrow stromal cells, which are in close proximity to maturing megakaryocytes, may produce prostacyclin on stimulation by interleukin-1ß.^31,32 Thus, stromal cells may regulate megakaryocyte gene expression and proliferation via prostacyclin-induced Rap1 activation.

The present report shows that in immature megakaryocytes, cAMP, independent of PKA, induces release of Ca²⁺ from intracellular stores, a property that is downregulated during megakaryocyte maturation. Rap1 activation might play a role in the cAMP-induced [Ca²⁺]_i increase, but further studies are needed to provide definite proof. Together these data reveal a novel mechanism for the regulation of cytosolic Ca²⁺ concentration, in stem cells and immature megakaryocytes, in which cAMP plays a central role.

	Acknowledgments

Top Abstract Introduction Methods Results Discussion References

 
This work was supported by the Netherlands Heart Foundation (grant No. 97.142) and the Netherlands Organization for Scientific Research (grant No. 902.68.241). J.-W.N.A. is supported by the Netherlands Thrombosis Foundation. The authors gratefully acknowledge the cooperation of the donors and the assistance of the Department of Obstetrics, University Medical Center Utrecht. We thank Dr J.L. Bos (University Medical Center Utrecht, Laboratory for Physiological Chemistry) for helpful discussions and for providing reagents for the Rap1 assay and N.M. Boontje for technical support.

Received February 8, 2001; accepted October 4, 2001.

	References

Top Abstract Introduction Methods Results Discussion References

 
1. Debili N, Issaad C, Masse JM, Guichard J, Katz A, Breton-Gorius J, Vainchenker W. Expression of CD34 and platelet glycoproteins during human megakaryocytic differentiation. Blood. 1992; 80: 3022–3035.[Abstract/Free Full Text]

2. Lev-Lehman E, Deutsch V, Eldor A, Soreq H. Immature human megakaryocytes produce nuclear-associated acetylcholinesterase. Blood. 1997; 89: 3644–3653.[Abstract/Free Full Text]

3. van der Vuurst H, van Willigen G, van Spronsen A, Hendriks M, Donath J, Akkerman JWN. Signal transduction through trimeric G proteins in megakaryoblastic cell lines. Arterioscler Thromb Vasc Biol. 1997; 17: 1830–1836.[Abstract/Free Full Text]

4. Rink TJ, Sage SO. Calcium signaling in human platelets. Annu Rev Physiol. 1990; 52: 431–449.[CrossRef][Medline] [Order article via Infotrieve]

5. Sasaki Y, Takahashi T, Tanaka I, Nakamura K, Okuno Y, Nakagawa O, Narumiya S, Nakao K. Expression of prostacyclin receptor in human megakaryocytes. Blood. 1997; 90: 1039–1046.[Abstract/Free Full Text]

6. Magnier C, Bredoux R, Kovacs T, Quarck R, Papp B, Corvazier E, de Gunzburg J, Enouf J. Correlated expression of the 97 kDa sarcoendoplasmic reticulum Ca²⁺-ATPase and Rap1B in platelets and various cell lines. Biochem J. 1994; 297: 343–350.[Medline] [Order article via Infotrieve]

7. Sheldrick RLG, Coleman RA, Lumley P. Iloprost, a potent EP1- and IP-receptor agonist. Br J Pharmacol. 1988; 94: 334.Abstract.

8. Hourani SMO, Cusack NJ. Pharmacological receptors on blood platelets. Pharmacol Rev. 1991; 43: 243–298.[Medline] [Order article via Infotrieve]

9. Maruyama T, Kanaji T, Nakade S, Kanno T, Mikoshiba K. 2APB, 2-aminoethoxydiphenyl borate, a membrane-penetrable modulator of Ins(1,4,5)P₃-induced Ca²⁺ release. J Biochem. 1997; 122: 498–505.[Abstract/Free Full Text]

10. Lacabaratz-Porret C, Corvazier E, Kovacs T, Bobe R, Bredoux R, Launay S, Papp B, Enouf J. Platelet sarco/endoplasmic reticulum Ca²⁺ATPase isoform 3b and Rap 1b: interrelation and regulation in physiopathology. Biochem J. 1998; 332: 173–181.[Medline] [Order article via Infotrieve]

11. Corvazier E, Enouf J, Papp B, de Gunzburg J, Tavitian A, Levy-Toledano S. Evidence for a role of rap1 protein in the regulation of human platelet Ca²⁺ fluxes. Biochem J. 1992; 281: 325–331.[Medline] [Order article via Infotrieve]

12. de Rooij J, Zwartkruis FJT, Verheijen MHG, Cool RH, Nijman SMB, Wittinghofer A, Bos JL. Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature. 1998; 396: 474–477.[CrossRef][Medline] [Order article via Infotrieve]

13. Kawasaki H, Springett GM, Mochizuki N, Toki S, Nakaya M, Matsuda M, Housman DE, Graybiel AM. A family of cAMP-binding proteins that directly activate Rap1. Science. 1998; 282: 2275–2279.[Abstract/Free Full Text]

14. Siess W, Winegar DA, Lapetina EG. Rap1-B is phosphorylated by protein kinase A in intact human platelets. Biochem Biophys Res Commun. 1990; 170: 944–950.[CrossRef][Medline] [Order article via Infotrieve]

15. Franke B, Akkerman JWN, Bos JL. Rapid Ca²⁺-mediated activation of Rap1 in human platelets. EMBO J. 1997; 16: 252–259.[CrossRef][Medline] [Order article via Infotrieve]

16. Wayman GA, Hinds TR, Storm DR. Hormone stimulation of type III adenylyl cyclase induces Ca²⁺ oscillations in HEK-293 cells. J Biol Chem. 1995; 270: 24108–24115.[Abstract/Free Full Text]

17. Burgess GM, Bird GS, Obie JF, Putney JW Jr. The mechanism for synergism between phospholipase C- and adenylylcyclase-linked hormones in liver: cyclic AMP-dependent kinase augments inositol trisphosphate-mediated Ca²⁺ mobilization without increasing the cellular levels of inositol polyphosphates. J Biol Chem. 1991; 266: 4772–4781.[Abstract/Free Full Text]

18. Nakade S, Rhee SK, Hamanaka H, Mikoshiba K. Cyclic AMP-dependent phosphorylation of an immunoaffinity-purified homotetrameric inositol 1,4,5-trisphosphate receptor (type I) increases Ca²⁺ flux in reconstituted lipid vesicles. J Biol Chem. 1994; 269: 6735–6742.[Abstract/Free Full Text]

19. D’Andrea P, Paschini V, Vittur F. Dual mechanism for cAMP-dependent modulation of Ca²⁺ signalling in articular chondrocytes. Biochem J. 1996; 318: 569–573.[Medline] [Order article via Infotrieve]

20. Tertyshnikova S, Fein A. Inhibition of inositol 1,4,5-trisphosphate-induced Ca²⁺ release by cAMP-dependent protein kinase in a living cell. Proc Natl Acad Sci U S A. 1998; 95: 1613–1617.[Abstract/Free Full Text]

21. Lee MG, Xu X, Zeng W, Diaz J, Wojcikiewicz RJ, Kuo TH, Wuytack F, Racymaekers L, Muallem S. Polarized expression of Ca²⁺ channels in pancreatic and salivary gland cells: correlation with initiation and propagation of [Ca²⁺]_i waves. J Biol Chem. 1997; 272: 15765–15770.[Abstract/Free Full Text]

22. Keularts IMLW, van Gorp RMA, Feijge MAH, Vuist WMJ, Heemskerk JWM. _2A-Adrenergic receptor stimulation potentiates calcium release in platelets by modulating cAMP levels. J Biol Chem. 2000; 275: 1763–1772.[Abstract/Free Full Text]

23. Toyofuku T, Kurzydlowski K, Tada M, MacLennan DH. Identification of regions in the Ca²⁺-ATPase of sarcoplasmic reticulum that affect functional association with phospholamban. J Biol Chem. 1993; 268: 2809–2815.[Abstract/Free Full Text]

24. Fischer TH, White GC. Partial purification and characterization of thrombolamban, a 22,000 dalton cAMP-dependent protein kinase substrate in platelets. Biochem Biophys Res Commun. 1987; 149: 700–706.[CrossRef][Medline] [Order article via Infotrieve]

25. Nagata K, Okano Y, Nozawa Y. Differential expression of low Mr GTP-binding proteins in human megakaryoblastic leukemia cell line, MEG-01, and their possible involvement in the differentiation process. Thromb Haemost. 1997; 77: 368–375.[Medline] [Order article via Infotrieve]

26. Franke B, van Triest M, de Bruijn KM, van Willigen G, Nieuwenhuis HK, Negrier C, Akkerman JWN, Bos JL. Sequential regulation of the small GTPase Rap1 in human platelets. Mol Cell Biol. 2000; 20: 779–785.[Abstract/Free Full Text]

27. Reedquist KA, Ross E, Koop EA, Wolthuis RMF, Zwartkruis FJT, van Kooyk Y, Salmon M, Buckley CD, Bos JL. The small GTPase, Rap1, mediates CD31-induced integrin adhesion. J Cell Biol. 2000; 148: 1151–1158.[Abstract/Free Full Text]

28. Boussiotis VA, Freeman GJ, Berezovskaya A, Barber DL, Nadler LM. Maintenance of human T cell anergy: blocking of IL-2 gene transcription by activated Rap1. Science. 1997; 278: 124–128.[Abstract/Free Full Text]

29. Cook SJ, McCormick F. Inhibition by cAMP of Ras-dependent activation of Raf. Science. 1993; 262: 1069–1072.[Abstract/Free Full Text]

30. Cook SJ, Rubinfeld B, Albert I, McCormick F. RapV12 antagonizes Ras-dependent activation of ERK1 and ERK2 by LPA and EGF in Rat-1 fibroblasts. EMBO J. 1993; 12: 3475–3485.[Medline] [Order article via Infotrieve]

31. Denizot Y, Dulery C, Trimoreau F, Desplat V, Praloran V. Arachidonic acid and human bone marrow stromal cells. Biochim Biophys Acta. 1998; 1402: 209–215.[Medline] [Order article via Infotrieve]

32. Chaudhary LR, Cheng SL, Avioli LV. Induction of early growth response-1 gene by interleukin-1ß and tumor necrosis factor- in normal human bone marrow stromal and osteoblastic cells: regulation by a protein kinase C inhibitor. Mol Cell Biochem. 1996; 156: 69–4177.[CrossRef][Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				I. A. Ferreira, A. I.M. Mocking, M. A.H. Feijge, G. Gorter, T. W. van Haeften, J. W.M. Heemskerk, and J.-W. N. Akkerman Platelet Inhibition by Insulin Is Absent in Type 2 Diabetes Mellitus Arterioscler. Thromb. Vasc. Biol., February 1, 2006; 26(2): 417 - 422. [Abstract] [Full Text] [PDF]

				E. Morel, A. Marcantoni, M. Gastineau, R. Birkedal, F. Rochais, A. Garnier, A.-M. Lompre, G. Vandecasteele, and F. Lezoualc'h cAMP-Binding Protein Epac Induces Cardiomyocyte Hypertrophy Circ. Res., December 9, 2005; 97(12): 1296 - 1304. [Abstract] [Full Text] [PDF]

				L. E. Wagner II, W.-H. Li, and D. I. Yule Phosphorylation of Type-1 Inositol 1,4,5-Trisphosphate Receptors by Cyclic Nucleotide-dependent Protein Kinases: A MUTATIONAL ANALYSIS OF THE FUNCTIONALLY IMPORTANT SITES IN THE S2+ AND S2- SPLICE VARIANTS J. Biol. Chem., November 14, 2003; 278(46): 45811 - 45817. [Abstract] [Full Text] [PDF]

				E. den Dekker, G. Gorter, J. W. M. Heemskerk, and J.-W. N. Akkerman Development of Platelet Inhibition by cAMP during Megakaryocytopoiesis J. Biol. Chem., August 2, 2002; 277(32): 29321 - 29329. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by den Dekker, E. Articles by Akkerman, J.-W. N. Search for Related Content PubMed PubMed Citation Articles by den Dekker, E. Articles by Akkerman, J.-W. N. Related Collections Cell signalling/signal transduction Platelet function inhibitors Platelets