Signal Transduction Through Trimeric G Proteins in Megakaryoblastic Cell Lines
Abstract The biogenesis of trimeric G proteins was investigated by measurement of the expression of α-subunits in the megakaryoblastic cell lines MEG-01, DAMI, and CHRF-288-11, representing stages of increasing maturation, and compared with platelets. Megakaryoblasts and platelets contained approximately equal amounts of Giα-1/2, Giα-3, Gqα, and G12α protein. Maturation was accompanied by (1) downregulation of mRNA for Gsα and disappearance of iloprost-induced Ca2+ mobilization, (2) upregulation of the long form of Gsα protein (Gsα-L) and an increase in iloprost-induced cAMP formation, and (3) upregulation of G16α mRNA and G16α protein and appearance of thromboxane A2-induced signaling (Ca2+ mobilization and stimulation of prostaglandin I2–induced cAMP formation). Gzα protein was absent in the megakaryoblasts despite weak expression of Gzα mRNA in DAMI and relatively high levels of Gzα mRNA and Gzα protein in platelets. These findings reveal major changes in G protein–mediated signal transduction during megakaryocytopoiesis and indicate that G16α couples the thromboxane receptor to phospholipase Cβ.
- Received March 22, 1996.
- Accepted December 12, 1996.
Platelet stimulation by PAF, TxA2, and α-thrombin is mediated through transmembrane receptors coupled to trimeric G proteins. These molecular switches transduce signals to effector enzymes such as PLCβ,1 PLA2, and AC and to ion channels for Ca2+, K+, and Na+/H+ exchange.2 3 The identification of G proteins that take part in signal transduction by different platelet-activating agents has been difficult. The IP, EP2, and DP receptors (receptors for I-, E2- and D-type PGs, eg, PGI2, PGE2, and PGD2, respectively4 ) and β-adrenergic receptors couple to the stimulatory G protein of AC, Gs. The EP1 and EP3 receptors, the α2A-adrenergic receptor, and the seven-transmembrane thrombin receptor couple to the inhibitory G protein of AC (Gi), whereas coupling to the PAF receptor is controversial.5 The thrombin receptor is also coupled to Gq, initiating phosphoinositide hydrolysis via PLCβ6 ; the TxA2 receptor also might signal through this G protein.7 8 Another G protein identified in platelets is Gz,9 which is phosphorylated by PKC,10 but its function remains unresolved.11 Additional regulatory properties are mediated via the βγ-subunits, which may inhibit AC12 and activate PLCβ.13
Because human platelets lack DNA and have no detectable protein synthesis, their signal-transducing properties evolve in the maturing megakaryocyte. Little is known about the biogenesis of signaling elements in the megakaryocyte, but megakaryoblastic cell lines have provided insight into the development of the activation apparatus. MEG-01 cells, representing an early maturation stage, show thrombin-induced Ca2+ mobilization and possess receptors for TxA2 and PGI2.14 15 MEG-01 and also the more mature DAMI cell line respond to PGE1 with increases in both cAMP and cytosolic Ca2+,14 a property not seen in platelets. CHRF-288-11 is the most mature megakaryoblastic cell line,16 containing secretion granules and signal transduction for α-thrombin, PAF, and TxA2 receptors to PLC.17
An illustration of differential expression of G proteins during development is found in cell lines used as a model for the human B-cell differentiation, where G16α is downregulated during transition from the pre-B to the B stage.18 Upregulation of Giα and Gsα is observed during maturation of human thymocytes,19 whereas differentiation of the murine erythroleukemia cell line RED-1 leads to loss of Giα-3 and a fall in PTX-sensitive Ca2+ mobilization.20
Assuming a similar asynchronous expression of trimeric G proteins in megakaryocytes, we set out to explore the transcription of Gα subunits during megakaryocytopoiesis in an attempt to clarify the role of individual G proteins. To compare G-protein expression with signal processing, the studies were carried out with megakaryoblastic cell lines that approximate different maturation stages of normal megakaryocytopoiesis.
FCS, horse serum, and Superscript II reverse transcriptase were from Gibco BRL. SuperTaq DNA polymerase was from HT Biotechnology, and Pfu polymerase was from Stratagene. Human α-thrombin, the cAMP derivative dbcAMP, PTX, and CTX were from Sigma. PAF, fura 2-AM, DNAse-free RNAseA, and Pwo DNA polymerase were from Boehringer Mannheim. The prostacyclin analogue iloprost was a gift from Schering, Berlin, Germany. The mastoparan analogue Mas-7, the PLC inhibitor U73122, and its inactive analog U73343 were purchased from Biomol. The thromboxane A2 analogue U46619, the IP receptor agonist carbaprostacyclin, and the EP1 receptor agonist 17-phenyl trinor PGE2 were from Cayman Chemical Co. SQ29548 was from Bristol-Meyers Squibb. Nitrocellulose filters were from Schleicher & Schuel, and Immobilon-P PVDF membranes (PVDF filters) were from Millipore. The cAMP-[125I] kit and [α-32P]dCTP were purchased from Amersham Life Science, and [3H]SQ29548 was from New England Nuclear. All other chemicals were of analytical grade.
The antibodies against Gαcommon (GA/1), Giα-1/2 (AS/7), Giα-3 (EC/2), Gsα (RM/1), and Gq/11α (QL) were rabbit polyclonal antibodies to C-terminal peptides of the different α-subunits and were purchased from NEN DuPont. Anti-G12α (N-terminal 2 to 21) and anti-Gzα (N-terminal 3 to 18) were rabbit polyclonal antibodies from Santa Cruz Biotechnology and Calbiochem, respectively. The rabbit polyclonal anti-G16α (AS 339, directed against amino acids 361 to 374 of the C-terminus21 ) was a kind gift of Dr K. Spicher, Free University of Berlin, Germany. Mouse monoclonal antibody 6F6 (anti-CD42b) recognizes free and complexed GPIb and was provided by Dr H.K. Nieuwenhuis, Department of Hematology, University Hospital, Utrecht, Netherlands. A FITC-conjugated antibody to GPIIIa (F803, anti-CD61, Dako) was used with a negative control of the same isotype (IgG1-FITC, X949, Dako). FITC-conjugated goat anti-mouse (gαm-FITC) was from the Central Laboratory for Blood Transfusion, Amsterdam, Netherlands. SWARPOs were from Dako.
MEG-01 cells,22 kindly provided by Dr H. Saito, Nagoya University School of Medicine, Nagoya, Japan, were grown in plastic culture flasks in RPMI 1640 medium supplemented with 20% FCS. DAMI cells,23 a gift of Dr R.I. Handin, Brigham and Women’s Hospital, Boston, Mass, were grown in IMDM supplemented with 10% horse serum. CHRF-288-11 cells16 were a gift of Dr M. Lieberman, Department of Molecular Genetics, Biochemistry, and Microbiology, University of Cincinnati (Ohio) and were grown in Fischer’s medium supplemented with 20% horse serum. All cells were cultured in the presence of 2 mmol/L l-glutamine, 100 U/mL penicillin, and 100 U/mL streptomycin at 37°C in a humidified atmosphere with 5% CO2 and were subcultured twice a week to maintain a concentration of ≈1×106 cells/mL. For some experiments, cells were incubated with PTX (200 ng/mL) for 14 hours or CTX (0.5 μg/mL) for 5 hours at 37°C in serum-free RPMI supplemented with 0.2% BSA.
Flow Cytometric Analysis
Expression of GPIb and GPIIIa was evaluated by FACS analysis. GPIb content was measured by labeling with mouse monoclonal antibody 6F6 followed by gαm-FITC. As a negative control, the same gαm-FITC antibody was used without primary antibody. GPIIIa was measured by incubation with FITC-labeled antibody F803; as a negative control, a nonspecific antibody of the same isotype, X949, was used. Briefly, cells were incubated with 6F6 or F803 in HEPES–Tyrode’s solution (pH 7.2) supplemented with 1.0% (wt/vol) BSA, for 20 minutes at 4°C. The cells incubated with the 6F6 antibody were washed and incubated with gαm-FITC for 30 minutes at 4°C. Finally, cells were washed in PBS (pH 7.4) with 0.2% (wt/vol) BSA and analyzed on a FACScan (Becton Dickinson).
Platelet concentrates (Red Cross blood bank, Utrecht, Netherlands) were made leukocyte-poor by filtration through a PALL50 leukocyte removal filter (PALL Biomedical Ltd), reducing the leukocyte/platelet ratio from 1:3000 to <1:106. The platelets were washed twice by centrifugation (20 minutes, 700g at 20°C) in PBS supplemented with 1/10 vol ACD (2.5 g trisodium citrate, 1.5 g citric acid, 2.0 g d-glucose in 100 mL distilled water [pH 6.5]).
Cell lines and platelets were collected by centrifugation (10 minutes, 100g at 20°C) and lysed in Laemmli’s electrophoresis sample buffer (0.001% bromphenol blue, 2% wt/vol SDS, 5% β-mercaptoethanol, and 10% glycerol in 62.5 mmol/L Tris [pH 6.8]). The samples were boiled for 5 minutes and stored at −20°C; 40 μg protein was subjected to SDS-PAGE (10% gels) and electroblotted onto PVDF filters. The blots were blocked in PBS (pH 7.4) containing 5% (wt/vol) fat-free dry milk (Protifar, Nutricia) and 0.05% Tween 20 for 1 hour at room temperature.
The primary antibody incubation was performed overnight at a 1:1000 dilution in PBS supplemented with 0.5% (wt/vol) Protifar and 0.05% Tween 20. The membranes were then washed three times and incubated with SWARPOs in the same buffer. Bands were visualized by chemiluminescence with the Renaissance Western blot chemiluminescence reagent of NEN-DuPont and DuPont NEF-496 Reflection autoradiography films.
Total RNA was extracted as described by Davis et al.24 In short, megakaryoblastic cells and platelets were pelleted, lysed in guanidine isothiocyanate buffer (4 mol/L guanidine isothiocyanate, 25 mmol/L sodium citrate [pH 7.0], 0.5% sarkosyl, and 0.1 mol/L β-mercaptoethanol), and layered on a CsCl cushion (5.7 mol/L CsCl, 25 mmol/L sodium citrate). After centrifugation (Beckmann SW41 Ti rotor, 32 000 rpm, 20 hours at 20°C), the pellet was resuspended in 300 μL 0.3 mol/L sodium acetate, extracted with phenol/chloroform (1/1, vol/vol), precipitated with 2.5 vol 96% ethanol, and resuspended in H2O.
Northern blot analysis was carried out as described.24 25 In brief, 30 μg of total RNA was migrated on a 1% agarose/formaldehyde gel in MOPS buffer (20 mmol/L 4-morpholinopropanesulfonic acid, 5 mmol/L sodium acetate, and 1 mmol/L EDTA [pH 7]), blotted onto nitrocellulose filter, and hybridized overnight at 42°C in hybridization buffer N,24 with 2×106 cpm of denatured probe/mL. The blots were washed for 10 minutes in 2×SSC buffer at 65°C, followed by a wash in 0.5×SSC with 0.1% SDS for 30 minutes at 65°C.
Full-length DNA probes were made by random-primer labeling essentially as described by Feinberg and Vogelstein.26 27 A Promega Prime-a-gene kit was used with 32P-labeled dCTP. The 32P signal was measured on a Phosphor Imager (Molecular Dynamics) and quantified by reprobing of the blots with a probe for rat GAPDH.
Amplification of Gα mRNA by RT-PCR
First-strand cDNA was synthesized from total RNA with oligo-dT primers using Superscript II reverse transcriptase and was used for amplification of Gsα, G16α, and Gzα. The oligonucleotides were designed to be specific for the different α-subunits and were located in the 5′ coding region and the 3′ noncoding region. The sequences of the primers were as follows. Gsα: 5′ primer, 5′-GGAATTCCATATGGGATGTCTCGGGAA-3′; 3′ primer, 5′-CCGCTCGAGGCCCTATGGTGGGTGATTATTA-3′; G16α:5′ primer, 5′-GGAATTCCATATGAGCGCTTGGCGTCACCCGCAGTTCGGTGGTATGGCCCGCTCGCTGACCTGGCGCTGCT-3′; 3′ primer, 5′-GAAGATCTGGCGTTCCTTCTCCTGTCCACTAGAGTGCG-3′; Gzα: 5′ primer, 5′-GGAATTCCATATGGGATGTCGGCAAAGCTCAGAGG-3′; 3′ primer, 5′-GGATGATCAAAAGTGAAGGGGCAGGTTGGG-3′. The PCR was performed for 30 cycles using Pwo polymerase for Gsα, SuperTaq for G16α, and Pfu polymerase for Gzα. The conditions for denaturation between cycles, annealing, and extension were, respectively, 1 minute at 95°C, 1 minute at 52°C, 2 minutes 72°C for Gsα; 1 minute at 95°C, 1 minute at 60°C, 2 minutes at 72°C for G16α; and 1 minute at 95°C, 1 minute at 53°C, 2 minutes at 72°C for Gzα. All PCRs were performed with a hot start at 95°C; SuperTaq and Pfu were used with 5% DMSO.
MEG-01, DAMI, and CHRF-288-11 cells were cultured overnight in serum-free RPMI, pelleted (5 minutes, 200g at 20°C), and resuspended in Ca2+-free HEPES–Tyrode’s buffer (in mmol/L: NaCl 145, KCl 5, Na2HPO4 0.5, MgSO4 1, HEPES 10, and glucose 5, and 2 g/L BSA [pH 6.5]; buffer A). Cells were incubated at 37°C for 1 hour with 3 μmol/L fura 2-AM, pelleted, resuspended in albumin-free buffer A, and stored at room temperature. Five minutes before the start of each measurement, the suspensions were diluted to 3×105 cells/mL in albumin-free buffer A (pH 7.2) and prewarmed at 37°C. Cells were incubated with 1 mmol/L CaCl2 for 1 minute to load intracellular Ca2+ stores. Subsequently, 1 mmol/L EGTA (final concentration) was added to trap extracellular Ca2+ immediately before the addition of agonist, making the measurements specific for Ca2+ mobilization. Measurements were performed at 37°C under mild stirring (50 rpm) in a Hitachi F-4500 fluorescence spectrophotometer with a multiwavelength timescan program. Fura 2 fluorescence was measured at 340 nm (F1) and 380 nm (F2) excitation, chosen on both sides of the isosbestic point, and 510 nm emission. [Ca2+]i was calculated from the ratio (R=F1/F2) of fluorescence intensities obtained from the formula of Grynkiewicz et al28 as modified by Gillis and Gailly29 : [Ca2+]i=Kd×(R−Rmin)/(Rmax−R)×F2min/F2max.
Fmax was measured at saturated Ca2+ concentration, and Fmin in the absence of calcium ions. Rmax was determined by addition of Triton X-100 (final concentration, 0.1%) in the presence of 1 mmol/L Ca2+, and Rmin by the subsequent addition of 10 mmol/L EDTA and 20 mmol/L Tris base. Calculations were based on a dissociation constant (Kd) of the fura 2–Ca2+ complex of 224 nmol/L.
Determination of cAMP
Samples (300 μL) of control and agonist-treated suspensions (1×105 cells/mL in HEPES-Tyrode’s buffer [pH 7.2]) were added to 600 μL of ice-cold 96% ethanol and transferred to liquid nitrogen. Defrosted samples were centrifuged (15 minutes, 14 000g at 4°C), and the pellets were washed with ice-cold 65% ethanol. The combined supernatants were evaporated under a stream of nitrogen at 60°C. cAMP was determined according to the manufacturer’s instructions with a cAMP-[125I] radioimmunoassay.
Analysis of TxA2 Receptors
MEG-01, DAMI, and CHRF cells were suspended in HEPES-Tyrode’s buffer (1.5×106 cells/mL in buffer A, without BSA [pH 7.2]). Aliquots of 0.2 mL were incubated with increasing concentrations (1 to 50 nmol/L) of [3H]SQ29548 at 25°C for 60 minutes. The incubations were stopped with 1.5 mL of ice-cold buffer, and the samples were rapidly filtered through a Whatman GF/C filter. The filters were rinsed three times with 3 mL of ice-cold buffer. Scintillation fluid was added, and radioactivity was counted according to standard procedures. Nonspecific binding was assessed in the presence of 10 μmol/L excess of unlabeled ligand. Data were analyzed by nonlinear regression (GraphPad Prism) and were displayed as Scatchard plots.
Statistical significances were calculated by a two-tailed paired or unpaired (where appropriate) Student’s t test. Data are expressed as mean±SD (for n data).
Characterization of Megakaryoblastic Cell Lines
Flow cytometric analysis showed that almost all cells expressed GPIb (Fig 1⇓); small differences in relative fluorescence intensity indicated that the number of expressed GPIb molecules per cell was somewhat higher on MEG-01 and DAMI than on CHRF. Analysis of GPIIIa expression showed that only 36±2% of MEG-01 cells expressed this protein, in contrast to 91±2% and 99±1% for DAMI and CHRF, respectively. The relative fluorescence intensities differed considerably, ranging from 6±1% in MEG-01 and 15±1% in DAMI to 75±2% (n=4) in CHRF. This indicates that the number of GPIIIa molecules per cell on MEG-01 and DAMI is much lower than on CHRF. On the basis of these data and morphological criteria (not shown), these data confirm that MEG-01, DAMI, and CHRF represent stages of increasing maturation.
Identification of Gα Subunits in Megakaryoblastic Cell Lines and Platelets
Western blots revealed that MEG-01, DAMI, CHRF, and platelets contain Giα-1/2, Giα-3, Gqα, G12α, and Gsα (Fig 2⇓). Gsα is known to consist of a short (45-kD) and a long (52-kD) isoform, designated Gsα-S and Gsα-L, respectively.30 MEG-01 and DAMI contained mainly Gsα-S; CHRF and platelets contained both types in approximately equal amounts, although the total amount of Gsα appeared lower. MEG-01 did not contain G16α protein, but DAMI, CHRF, and platelets showed increasing amounts of this protein. The three cell lines appeared devoid of Gzα protein, although this α-subunit was easily detectable in platelets.
To study the expression of G16α and Gzα in more detail, mRNAs for these subunits were analyzed and compared with mRNA for Gsα as an example of an early expressed α-subunit. Northern blot analysis (Fig 3⇓) showed Gsα mRNA in MEG-01, DAMI, and CHRF cells, in agreement with the Western blots. G16α mRNA was absent in MEG-01, but DAMI and especially CHRF showed expression for this α-subunit. In contrast, Northern blots failed to show mRNA for Gzα in the three cell lines (not shown). Reprobing of the blots with a probe for GAPDH and quantification of the amount of mRNA on the Phosphor Imager revealed that MEG-01 contained the highest amount (100%) of mRNA for Gsα, whereas DAMI and CHRF contained 45% to 54% and 61% to 55%, respectively, in two separate experiments. These data are in agreement with the weaker expression of Gsα on the Western blot. In contrast, G16α mRNA signal was absent in MEG-01, 20% to 21% in DAMI, and 100% in CHRF.
The amplification of megakaryoblastic and platelet mRNAs by RT-PCR with oligonucleotides specific for G16α was in line with these observations. Although Gsα mRNA was present in all cell types, a concurrently run sample was negative for G16α mRNA in MEG-01. A weak band of ≈1.6 kb was detected in DAMI and CHRF, and a strong band was found in platelets (Fig 4⇓). A similar amplification of Gzα mRNA was negative for MEG-01 and CHRF, which again accords with the Northern blots. However, a weak band of ≈1.4 kb was found in DAMI, despite the negative Northern blot for this cell line. Again, amplification of platelet RNA led to a clear band of the same size.
To analyze whether incomplete Gα expression led to abnormal signal transduction, MEG-01, DAMI, and CHRF cells were stimulated with agonists known to affect Ca2+ homeostasis in platelets (Table 1⇓). α-Thrombin and PAF induced Ca2+ mobilization in the three cell lines, albeit to different extents. These responses could be completely blocked by a preincubation with the PLC inhibitor U73122, whereas the inactive analogue U73343 had no effect. Preincubation with PTX abolished the α-thrombin–induced Ca2+ mobilization by ≈40%, indicating that the thrombin receptor was coupled to PLC via a G protein of the Gi class and via a different G protein. In contrast, the PAF-induced Ca2+ mobilization was insensitive to PTX. Interestingly, the TxA2 analogue U46619 failed to mobilize Ca2+ in MEG-01, whereas DAMI and CHRF responded with increases of ≈60% and 70%, respectively, compared with unstimulated cells. As in platelets, these responses were insensitive to PTX.5
The PGI2 analogue iloprost also triggered Ca2+ mobilization. The response was particularly evident in MEG-01, lower in DAMI, and virtually absent in CHRF. The rise in cytosolic Ca2+ was abolished by the PLC inhibitor U73122 and could be partially blocked by PTX. Although iloprost shows the highest affinity for IP receptors, binding to EP1 receptors has been reported.31 To better discriminate between these receptors, studies were repeated with carbaprostacyclin, an agonist with optimal specificity for IP receptors, and with 17-phenyl trinor PGE2, an agonist for EP1 receptors. After stimulation of DAMI cells with 17-phenyl trinor PGE2 (1 μmol/L), the rise in cytosolic Ca2+ in response to carbaprostacyclin (1 μmol/L) was 110±16% (n=3) of normal carbaprostacyclin signaling. Also, a primary incubation with carbaprostacyclin failed to change subsequent responses by 17-phenyl trinor PGE2. Thus, both ligands were specific for their respective receptors. The carbaprostacyclin data were similar to those obtained with iloprost, confirming that the IP receptor signals to Ca2+ via PLC, a property that disappears at later maturation stages. In contrast, the Ca2+ responses triggered by 17-phenyl trinor PGE2 did not disappear during maturation.
In platelets, a slight increase in cAMP is sufficient to abolish agonist-induced Ca2+ mobilization.32 As expected, iloprost raised cAMP in the three megakaryoblasts (see below), but this treatment did not change the Ca2+ responses by platelet ligands. Also, the cell-permeable analogue dbcAMP (250 μmol/L, 5 minutes) failed to change these Ca2+ responses, revealing a marked discrepancy with those patterns in platelets (data not shown).
The cAMP accumulation in the megakaryoblasts is listed in Table 2⇓. Unstimulated cells contained ≈10 pmol cAMP/106 cells. Incubation with CTX increased cAMP 2-fold in all cell lines, reflecting the presence of a functional Gs protein. Iloprost raised the level of cAMP from 4- to 6-fold in MEG-01 and DAMI to 12-fold in CHRF. Preincubation with the mastoparan analogue Mas-7 completely inhibited the iloprost-induced cAMP accumulation, reflecting the presence of a functional Gi protein. As in platelets, α-thrombin, PAF, and the thromboxane analogue U46619 failed to increase the cAMP level. Surprisingly, α-thrombin amplified the cAMP accumulation by iloprost, a property not seen in platelets. There was a 2-fold to 3-fold enhancement in MEG-01 and DAMI and a 6-fold increase in CHRF. PAF failed to affect the iloprost-induced cAMP formation in any of the cells. Potentiation of cAMP formation was also absent when MEG-01 was treated with TxA2 analogue, but both DAMI and CHRF showed a stimulation by TxA2 similar to that seen with α-thrombin.
The inability of TxA2 analogue to mobilize Ca2+ or to stimulate iloprost-induced cAMP formation in MEG-01 but not in DAMI and CHRF pointed to a defect in TxA2-induced signal transduction. Because all cell lines showed PLC-mediated Ca2+ responses (Table 1⇑) and AC-mediated cAMP formation (Table 2⇑), the impairment was either at the receptor level or caused by the absence of G16α (Figs 2 through 4⇑⇑⇑). Analysis of [3H]SQ29548 binding to intact cells revealed that MEG-01, DAMI, and CHRF all contained TxA2 receptors (Fig 5⇓). MEG-01 showed a binding affinity of 6.9 nmol/L, with a Bmax of 14 800 binding sites per cell, which is in good agreement with the Kd of 8.2 nmol/L and 12 800 binding sites per cell reported earlier.15 DAMI cells had a similar Kd (7.5 nmol/L) but contained 44 000 binding sites per cell. CHRF had a Kd of 5.4 nmol/L and 21 600 binding sites per cell, which is again close to reported values (Kd=2.1 nmol/L; Bmax=35 00033 ). These data indicated that neither receptor affinity nor the number of receptors varied appreciably between the cell lines.
With MEG-01, DAMI, and CHRF cell lines used as models for different stages of megakaryocyte maturation and the platelet taken as the most matured cell, the following stages in the biogenesis of Gα subunits can be identified.
There is an early stage represented by MEG-01 that already contains the proteins for Giα-1/2, Giα-3, Gqα, G12α, and Gsα. These cells respond to α-thrombin and PAF with Ca2+ mobilization, indicating that the respective receptors are present and that steps downstream of the α-subunits are intact. Indeed, the fact that the Ca2+ response is abolished by U73122 is evidence for the presence of PLC. In platelets, α-thrombin mobilizes Ca2+ via Gi-β/γ and a poorly characterized PTX-insensitive G protein, possibly Gq.34 MEG-01 cells also accumulate cAMP in response to iloprost, in line with the presence of Gsα and the CTX-induced cAMP formation in these cells. The Gsα-type is mainly the short isotype, Gsα-S, which in reconstitution experiments had a much lower activity than the long isotype, Gsα-L.30
The maturation stage represented by DAMI contains the same α-subunits as present in MEG-01 and, in addition, G16α. G16 is a member of the Gq family and is expressed primarily in hematopoietic cells.35 A recent publication reported that platelet RNA contains the message for this G protein.36 The appearance of G16α is accompanied by Ca2+ mobilization by TxA2. Because MEG-01, DAMI, and CHRF all contain TxA2 receptors and the sequences that mediate Ca2+ mobilization by α-thrombin, these findings suggest that G16α couples the TxA2 receptor to PLC. A second property that accompanies the appearance of G16α is stimulation of iloprost-induced cAMP formation by TxA2, again pointing to the formation of a receptor-G16α complex that is activated by this agonist.
The late stage of megakaryocyte maturation, represented by CHRF, shows downregulation of Gsα-S and the appearance of Gsα-L, which is accompanied by an increase in iloprost-induced cAMP formation. At the same time, there is a downregulation of total Gsα, as evident from the weaker bands in the Western blot and the lower expression of Gsα mRNA on Northern blot. Platelets also contain approximately equal amounts of Gsα-S and Gsα-L. Thus, the extreme responsiveness of platelets to PGI237 and its analogues might result from the prevalence of the long isotype, which has a 3- to 10-fold higher activity than the short isotype.30 A similar shift is seen in rat platelets during aging.38
Apart from the signaling properties that develop during maturation, a property that disappears is the iloprost-induced Ca2+ mobilization. Whereas MEG-01 showed a 95% increase in cytosolic Ca2+ content, there was only a 30% increase in DAMI. In CHRF, iloprost failed to mobilize Ca2+, a property these cells share with platelets. A similar change is seen in cultures of human CD34+ cells, in which young megakaryoblasts, cultured in the presence of thrombopoietin, show iloprost-induced Ca2+ mobilization and matured megakaryocytes lose this property (van der Vuurst et al, unpublished results, 1997). The Ca2+ mobilization by iloprost in MEG-01 was for the major part inhibited by PTX, suggesting the involvement of a G protein of the Gi family. The PTX-insensitive part might reflect a role for a Gs protein. HEL cells also show PTX-insensitive Ca2+ mobilization in response to iloprost.39 The PLC inhibitor U73122 completely abolished the iloprost-induced Ca2+ mobilization. Similar findings were obtained with carbaprostacyclin, which is a more specific agonist for IP receptors, indicating that these receptors couple to PLC via a member of the Gs class. Ca2+ mobilization through IP receptors disappears during maturation, in contrast to responses evoked by EP1 receptors.
In all cell lines, stimulation with iloprost raises cAMP. This leads to the peculiar observation that PGI2 stimulates both inhibitory and stimulatory G proteins for AC regulation. The fact that iloprost raises cAMP indicates that the stimulating signals prevail and that additional mechanisms separate the two pathways. Equally surprising is the fact that Ca2+ mobilization by α-thrombin, PAF, or TxA2 analogue in megakaryoblasts is insensitive to an increase in cAMP, induced either via the PGI2 receptor or by direct addition of dbcAMP. In platelets, cAMP inhibits multiple steps in agonist-induced Ca2+ mobilization, such as the binding of α-thrombin to its receptor and activation of PLC.32 In many cell types, cAMP is not an inhibitor or even signals as an activating factor. Activation of adenosine A1 receptors in smooth muscle cells led to a Gi-mediated Ca2+ mobilization through PLC, which was insensitive to cAMP.40 A dopamine receptor expressed in human embryonic kidney 293 cells mediated Ca2+ mobilization, which depended on cAMP.41
The insensitivity to PTX and the fact that the protein is phosphorylated during platelet activation makes Gz a topic of specific interest. None of the megakaryoblasts contained detectable Gzα protein, and there was only a weak 1.4-kb band in DAMI after RT-PCR. Nevertheless, these cells showed a normal Ca2+ mobilization on stimulation with α-thrombin and PAF and undisturbed cAMP production by iloprost, indicating that these pathways can function without Gz. In platelets, activation of PKC results in phosphorylation of Gzα10 and suppression of PGI2-induced cAMP formation,42 but a causal relationship has never been established.
A property not seen in platelets is the amplification of iloprost-induced cAMP formation by α-thrombin and TxA2 analogue. For TxA2 analogue, this property required the presence of G16α, suggesting that a TxA2-G16 complex was involved. α-Thrombin enhanced cAMP formation even in MEG-01, which lacks G16α, and might therefore involve a complex between the thrombin receptor and a Gi subtype or Gq. Preliminary studies show that the enhancement by both agonists is mediated via protein kinase C and a second step upstream of PLC (van der Vuurst et al, unpublished results, 1997). The stimulation in megakaryoblasts is in sharp contrast with platelets, whereas PKC inhibits this pathway.35
Most classifications of megakaryocyte maturation presume their development from an incomplete precursor to a cell that contains most of the proteins present in platelets, such as membrane receptors, glycoproteins, and α-granule proteins. Williams and Levine43 separated megakaryocyte development into four stages, ranging from the early megakaryoblast to the mature polyploid megakaryocyte capable of platelet formation. The protein composition in developing megakaryocytes is unknown, and it is equally uncertain how well megakaryoblastic cell lines resemble the different maturation stages of megakaryocytopoiesis. The present results illustrate that maturation might imply that expression increases for one protein and decreases for another. Whereas the more mature DAMI and CHRF cells show increased levels of G16α mRNA compared with the immature MEG-01 cells, the levels of Gsα mRNA are lower. Furthermore, the shift to the more active Gsα-L subtype illustrates that maturation also includes changes between closely related proteins.
Selected Abbreviations and Acronyms
|dbcAMP||=||N6,2′-O-dibutyryladenosine 3′:5′-cyclic monophosphate|
|PKC||=||protein kinase C|
|SWARPO||=||peroxidase-conjugated swine antibodies against rabbit IgG|
This study was supported by the Netherlands Organization for Scientific Research/the Netherlands Heart Foundation (grant 940.50.102/900.526.094).
Siffert W, Siffert G, Scheid P, Akkerman J-WN. Na+/H+ exchange modulates Ca2+ mobilization in human platelets stimulated by ADP and the thromboxane mimetic U 46619. J Biol Chem. 1990;264:719-725.
Brass LF, Woolkalis MJ, Manning DR. Interactions in platelets between G proteins and the agonists that stimulate phospholipase C and inhibit adenylyl cyclase. J Biol Chem. 1988;263:5348-5355.
Hung DT, Wong YH, Vu T-KH, Coughlin SR. The cloned platelet thrombin receptor couples to at least two distinct effectors to stimulate phosphoinositide hydrolysis and inhibit adenylyl cyclase. J Biol Chem. 1992;267:20831-20834.
Offermanns S, Laugwitz K-L, Spicher K, Schultz G. G proteins of the G12 family are activated via thromboxane A2 and thrombin receptors in human platelets. Proc Natl Acad Sci U S A. 1994;91:504-508.
Knezevic I, Borg C, Le Breton GC. Identification of Gq as one of the G-proteins which copurify with human platelet thromboxane A2/prostaglandin H2 receptors. J Biol Chem. 1993;268:26011-26017.
Williams Gagnon A, Manning DR, Catani L, Gewirtz A, Poncz M, Brass LF. Identification of Gzα as a pertussis toxin-insensitive G protein in human platelets and megakaryocytes. Blood. 1991;78:1247-1253.
Lounsbury KM, Casey PJ, Brass LF, Manning DR. Phosphorylation of Gz in human platelets. J Biol Chem. 1991;266:22051-22056.
Wong YH, Conklin BR, Bourne HR. Gz-mediated hormonal inhibition of cyclic AMP accumulation. Science. 1992;255:339-342.
Taussig R, Gilman AG. Mammalian membrane-bound adenylyl cyclases. J Biol Chem. 1995;270:1-4.
Watanabe T, Yatomi Y, Sunaga S, Miki I, Ishii A, Nakao A, Higashihara M, Seyama Y, Ogura M, Saito H, Kurokawa K, Shimizu T. Characterization of prostaglandin and thromboxane receptors expressed on a megakaryoblastic leukemia cell line, MEG-01s. Blood. 1991;78:2328-2336.
Fugman DA, Witte DP, Jones CLA, Aronow BJ, Lieberman MA. In vitro establishment and characterization of a human megakaryoblastic cell line. Blood. 1990;75:1252-1261.
Dorn GW II, Davis MG. Differential megakaryocytic desensitization to platelet agonists. Am J Physiol. 1992;263:C864–C872.
Mapara MY, Bommert K, Bargou RC, Leng C, Beck C, Ludwig WD, Gierschik P, Dorken B. G protein subunit Gα16 expression is restricted to progenitor B cells during human B-cell differentiation. Blood. 1995;85:1836-1842.
Pessa-Morikawa T, Mustelin T, Andersson LC. Functional maturation of human T lymphocytes is accompanied by changes in the G-protein pattern. J Immunol. 1990;144:2690-2695.
Kesselring F, Spicher K, Porzig H. Changes in G protein pattern and in G protein-dependent signaling during erythropoietin- and dimethylsulfoxide-induced differentiation of murine erythroleukemia cells. Blood. 1994;84:4088-4098.
Ogura M, Morishima Y, Ohno R, Kato Y, Hirabayashi N, Nagura H, Saito H. Establishment of a novel human megakaryoblastic leukemia cell line, MEG-01, with positive Philadelphia chromosome. Blood. 1985;66:1384-1392.
Greenberg AM, Rosenthal DS, Greeley TA, Tantravahi R, Handin RL. Characterization of a new megakaryocytic cell line: the Dami cell. Blood. 1988;72:1968-1977.
Davis LG, Dibner MD, Battey JF. Basic Methods in Molecular Biology. New York, NY: Elsevier Science Publishing Co Inc; 1986.
Thomas PS. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Natl Acad Sci U S A. 1980;77:5201-5205.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescent properties. J Biol Chem. 1985;260:3440-3450.
Nanoff C, Boehm S, Hohenegger M, Schutz W, Freissmuth M. 2′,3′-Dialdehyde GTP as an irreversible G protein antagonist. J Biol Chem. 1994;269:3199-3207.
Sheldrick RLG, Coleman RA, Lumley P. Iloprost: a potent EP1- and IP-receptor agonist. Br J Pharmacol. 1988;94:334P.
Kroll MH, Schafer AI. Biochemical mechanisms of platelet activation. Blood. 1989;74:1181-1195.
Dorn GW II. Regulation of response to thromboxane A2 in CHRF-288 megakaryocytic cells. Am J Physiol. 1992;262:C991–C999.
Benka ML, Lee M, Wang G-R, Buckman S, Burlacu A, Cole L, DePina A, Dias P, Granger A, Grant B, Hayward-Lester A, Karki S, Mann S, Marcu O, Nussenzweig A, Piepenhagen P, Raje M, Roegiers F, Rybak S, Salic A, Smith-Hall J, Waters J, Yamamoto N, Yanowitz J. The thrombin receptor in human platelets is coupled to a GTP binding protein of the Gαq family. FEBS Lett. 1995;363:49-52.
Amatruda TT III, Steele DA, Slepak VZ, Simon MI. Gα16, a G protein specially expressed in hematopoietic cells. Proc Natl Acad Sci U S A. 1991;88:5587-5591.
Giesberts AN, Van Willigen G, Lapetina EG, Akkerman J-WN. Regulation of platelet glycoprotein IIb/IIIa (integrin αIIBβ3) function via the thrombin receptor. Biochem J. 1995;309:613-620.
Brass LF, Woolkalis MJ. Dual regulation of cyclic AMP formation by thrombin in HEL cells, a leukaemic cell line with megakaryocytic properties. Biochem J. 1992;281:73-80.
Gerwins P, Fredholm BB. Stimulation of adenosine A1 receptors and bradykinin receptors, which act via different G proteins, synergistically raises inositol 1,4,5-triphosphate and intracellular free calcium in DDT1MF-2 smooth muscle cells. Proc Natl Acad Sci U S A. 1992;89:7330-7334.
Lin CW, Miller TR, Witte DG, Bianchi BR, Stashko M, Manelli AM, Frail DE. Characterization of cloned human dopamine D1 receptor-mediated calcium release in 293 cells. Mol Pharmacol. 1995;47:131-139.
Tsu RC, Lai HWL, Allen RA, Wong YH. Differential coupling of the formyl peptide receptor to adenylate cyclase and phospholipase C by the pertussis toxin-insensitive Gz protein. Biochem J. 1995;309:331-339.