Satya P. Kunapuli, Department of Physiology, and James L. Daniel, Department of Pharmacology, Temple University Medical School, 3420 N. Broad Street Philadelphia, PA 19140, USA. Tel.: +1 215 707 4615; +215 707 4457; fax: +1 215 707 4003. E-mail: email@example.com; firstname.lastname@example.org
Summary. Background: Myosin IIA is an essential platelet contractile protein that is regulated by phosphorylation of its regulatory light chain (MLC) on residues (Thr)18 and (Ser)19 via the myosin light chain kinase (MLCK).
Objective: The present study was carried out to elucidate the mechanisms regulating MLC (Ser)19 and (Thr)18 phosphorylation and the functional consequence of each phosphorylation event in platelets.
Results: Induction of 2MeSADP-induced shape change occurs within 5 s along with robust phosphorylation of MLC (Ser)19 with minimal phosphorylation of MLC (Thr)18. Selective activation of G12/13 produces both slow shape change and comparably slow MLC (Thr)18 and (Ser)19 phosphorylation. Stimulation with agonists that trigger ATP secretion caused rapid MLC (Ser)19 phosphorylation while MLC (Thr)18 phosphorylation was coincident with secretion. Platelets treated with p160ROCK inhibitor Y-27632 exhibited a partial inhibition in secretion and had a substantial inhibition in MLC (Thr)18 phosphorylation without effecting MLC (Ser)19 phosphorylation. These data suggest that phosphorylation of MLC (Ser)19 is downstream of Gq/Ca2+-dependent mechanisms and sufficient for shape change, whereas MLC (Thr)18 phosphorylation is substantially downstream of G12/13-regulated Rho kinase pathways and necessary, probably in concert with MLC (Ser)19 phosphorylation, for full contractile activity leading to dense granule secretion. Overall, we suggest that the amplitude of the platelet contractile response is differentially regulated by a least two different signaling pathways, which lead to different phosphorylation patterns of the myosin light chain, and this mechanism results in a graded response rather than a simple on/off switch.
Platelets undergo a series of coordinated responses after activation, which play an important role in thrombosis and are essential for the maintenance of hemostasis [1,2]. Physiological agonists such as collagen, thrombin and ADP are capable of activating platelets, resulting in events including shape change, aggregation, generation of thromboxane and secretion of granule contents [3–8]. Platelet shape change is the earliest functional response after activation with physiological agonists and is accompanied by rearrangement of the cytoskeleton . Properties that may lead to cytoskeletal rearrangements such as filament assembly, surface membrane folding and centralization of secretory granules are thought to be mediated by the phosphorylation of Myosin IIA [9–11]. The phosphorylation of the myosin light chain (MLC) results in an increased development of actin-activated ATPase activity and reflects the contractile activity of actomyosin [12,13]. In intact cells, MLC is found to be phosphorylated on residues threonine 18 (Thr)18 or serine 19 (Ser)19 [14,15] and is regulated through an increase in calcium/calmodulin-mediated myosin light chain kinase (MLCK) activity and/or through the activation of Rho kinase, which can either directly phosphorylate MLC or phosphorylate myosin phosphatase, thereby inhibiting its activity [14,16]. Diphosphorylation was first described by Ikebe and Hartshorne . They showed that (Thr)18 phosphorylation occurs in smooth muscle more slowly than at MLC (Ser)19. They concluded that phosphorylation at MLC (Thr)18 markedly increases the actin-activated ATPase activity of myosin over singly phosphorylated myosin. Reports from Kiss et al.  suggest other platelet cytoskeletal kinases such as zipper-interacting protein kinase (ZIPK) and integrin-linked kinase (ILK) may contribute to the regulation and balance of kinase and phosphatase activity leading to MLC phosphorylation.
It has been previously reported that the phosphorylation of myosin light chains are important for triggering platelet shape change . Studies from our laboratory have demonstrated that platelet shape change is mediated by MLC phosphorylation in a Ca2+-dependent and Ca2+-independent manner through Gq and RhoA pathways, respectively . The contribution of these signaling pathways in the regulation of MLC phosphorylation has been investigated with the use of pharmacological inhibitors 5,5′-dimethyl-BAPTA (a calcium chelator), YM-254890 (a Gq inhibitor [20–22], which prevents signaling leading to Ca2+mobilization), or Y-27632 (a selective inhibitor of P160ROCK ).
Rho A signaling has been shown to cause MLC phosphorylation and this event is important for platelet internal contraction . Suzuki et al.  demonstrated that a reduction in ATP secretion as well as MLC phosphorylation occurred in platelets pre-treated with Y-27632, and activated with the thromboxane analog, STA2 (1 μmol L−1) or thrombin (0.05 U mL−1), implicating a contribution of Rho kinase in ATP release. We have also shown that protease-activated receptor (PAR)-mediated dense granule release is inhibited upon blockade of Rho A signaling  suggesting Rho A or downstream effectors may contribute to dense granule release.
In the present study, we examined the mechanism and the role of phosphorylation of MLC residues (Thr)18 and (Ser)19 in platelet function. The results demonstrate that the phosphorylation of MLC (Ser)19 occurs by Ca2+-dependent and Ca2+-independent signaling and correlates with platelet shape change. Phosphorylation of MLC on (Thr)18 occurs in a Ca2+-independent manner through both a G12/13 and Rho kinase mechanism. We provide evidence that the phosphorylation of MLC (Thr)18 may be important for the full release of ATP after PAR-1 activation.
Materials and methods
Apyrase (Type VII), 2MeSADP, α-thrombin, acetylsalicylic acid (ASA) and indomethacin were obtained from Sigma (St. Louis, MO, USA). Hexapeptide SFLLRN was obtained from New England Peptide (Gardner, MA, USA). AYPGKF was from GenScript Corp. (Piscataway, NJ, USA). Serotonin was from ACROS Organics (Morris Plains, NJ, USA). Convulxin was purified according to the method of Polgar et al. . DuPONT Instruments luminescence biometer reagent kit (Wilmington, DE, USA) was used for detection of secreted ATP. YM-254890 was a gift from Yamanouchi Pharmaceutical Co., Ltd. (Ibaraki, Japan). The calcium chelator dimethyl-BAPTA [bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetra-acetic acid] was purchased from Biomol (Plymouth Meeting, PA, USA). AR-C69931MX was a gift from Astra-Zeneca (Wilmington, DE, USA). ROCK inhibitor Y-27632 was from Calbiochem (San Diego, CA, USA). Fura-2 acetoxylmethyl ester (Fura-2 AM) was from Molecular Probes (Eugene, OR, USA). Phospho-myosin light chain 2 (Ser)19 Mouse mAb and p44/42 MAPK (Erk1/2) (3A7) mouse mAb were obtained from Cell Signaling (Beverly, MA, USA) and phospho-specific MLC (Thr)18, horseradish peroxidase (HRP)-conjugated goat anti-mouse or anti-rabbit immunoglobulin G (IgG) were from Santa Cruz (Santa Cruz, CA, USA). Millipore Immobilon Western Chemiluminescent HRP substrate and polyvinylidene fluoride (PVDF) membrane was used for all immunoblotting (Billerica, MA, USA).
Isolation of human platelets
Blood was collected from informed healthy volunteers into a one-sixth volume of acid/citrate/dextrose (2.5 g sodium citrate, 2 g glucose, 1.5 g citric acid in 100 mL deionized water). Platelet-rich plasma (PRP) was obtained by centrifugation at 250 × g for 15 min at ambient temperature and incubated with 1 mmol L−1 aspirin for 30 min at 37 °C. Platelets were isolated from plasma by centrifugation at 800 × g for 10 min at ambient temperature and resuspended in Tyrode buffer (138 mmol L−1 NaCl, 2.7 mmol L−1 KCl, 2 mmol L−1 MgCl2, 0.42 mmol L−1 NaH2PO4, 5 mmol L−1 glucose, 10 mmol L−1 HEPES [N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid, pH 7.4] and 0.2 units per mL apyrase). The platelet count was adjusted to 2 × 108 per mL. Approval was obtained from the institutional review board of Temple University for these studies. Informed consent was provided prior to blood donation.
Measurement of platelet secretion
Platelet secretion was determined by measuring the release of ATP as previously described  using the Dupont Instruments luminescence biometer reagent. In experiments where inhibitors were used, the platelet sample was incubated with the inhibitors for 5 min at 37 °C prior to the addition of agonists. The secretion was subsequently measured as described above.
Aggregation of 0.5 mL of washed platelets was analyzed using a lumi-aggregometer (Chrono-log Corp., Havertown, PA, USA). Aggregation was measured using light transmission under stirring conditions (900 rpm) at 37 °C. Each sample was allowed to aggregate for the indicated time. The chart recorder (Kipp and Zonen, Bohemia, NY, USA) was set for 0.2 mm s−1.
Measurement of intracellular Ca2+mobilization
PRP was incubated with the fluorescent calcium indicator Fura-2 AM (5 μm) and aspirin (1 mm), fluorescence was measured and the Ca2+concentration was calculated as previously described using Triton X-100 (0.1%) instead of digitonin for standardization .
Western blot analysis
Platelets were stimulated with agonists in the presence of inhibitors or vehicle for the appropriate time under stirring conditions at 37 °C and the reaction was stopped by the addition of 0.6 N HClO4. The resulting acid precipitate was collected and chilled on ice. The pellets were centrifuged at 13 000 × g for 10 min followed first by rinsing and then resuspension in 0.5 mL of deionized water. The protein was again pelleted by centrifugation at 13 000 × g for 10 min. The protein pellets were solubilized in sample buffer containing 0.1 mol L−1 Tris, 2% sodium dodecylsulfate (SDS), 1% (v/v) glycerol, 0.1% bromophenol blue and 100 mmol L−1 DTT then boiled for 10 min. Proteins were separated by 15% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred onto PVDF membrane. Blots were blocked with 5% non-fat milk in TBS-T (TBS+0.05% Tween) for 1 h at room temperature and probed overnight at 4 °C with appropriate antibodies. Blots were washed four times with TBS-T for 10 min. HRP-conjugated secondary antibody was incubated for 60 min at room temperature. Blots were washed an additional four times with TBS-T and antigen-antibody complexes were detected using chemiluminescent HRP substrate. Bands were visualized on a Fuji imaging system and densities calculated with Image Gauge software (Fujifilm Medical Systems, Stamford, CT, USA).
Measurement of percent myosin light chain phosphorylation
For absolute mass method measurement of myosin light chain phosphorylation, urea gel electrophoresis was used as previously described . In brief, HClO4 acid precipitates were centrifuged at 13 000 × g for 10 min, washed and reconstituted in 50 μL of sample buffer containing 8 mol L−1 urea, 20 mmol L−1 Tris–HCl (pH 8.6), 122 mmol L−1 glycerin, 5 mmol L−1 dithiothreitol, with 0.1% bromphenol blue dye. Platelet samples where sonicated in a Branson (Shelton, CT, USA) sonication bath. Electrophoresis was performed using 10% polyacrylamide gels containing 40% (v/v) glycerol with 3.6% polyacrylamide stacking gel containing 8 mol L−1 urea. The running buffer used in the top chamber was 20 mmol L−1 Tris, 122 mmol L−1 glycine at pH 8.6 containing 4 mmol L−1 urea. The samples were electrophoresed at 8–9 mA per gel and electrophoreses was terminated 1 h after the bromphenol blue marker had come off the bottom of the gel. Gels were either stained with GelCode blue staining reagent or transferred and probed with appropriate antibody.
All densitometric analysis was performed using Fujifilm Science Lab 2003, Image Gauge, Version 4.22 software (Edison, NJ, USA). Percent MLC phosphorylation was derived first by calculation of the ratio of MLC phosphorylation to total ERK. Second, the zero time point ratio was subtracted from each stimulated sample in all experiments and the highest determination was set to 100%. KaleidaGraph 2003, Version 3.62 by Synergy software and Prism 2007, Version 5 by Graphpad software were used for statistical analysis and graphical representation.
Agonist-induced phosphorylation of specific residues on MLC in human platelets
Actin–myosin interactions play a crucial role in the activation of platelets . These interactions are initiated upon phosphorylation of MLC on (Ser)19 and (Thr)18 residues. Although it has been known that these two MLC residues are phosphorylated in platelets upon agonist stimulation , it is not known whether they are phosphorylated by the same signaling pathways. We investigated the regulation of myosin light chain phosphorylation, using the PAR-1 agonist SFLLRN in aspirin-treated platelets. The platelet lysates were then probed for phosphorylation of either MLC (Thr)18 or (Ser)19 using phospho-specific antibodies. After the stimulation of PAR-1 receptors, platelets undergo shape change, aggregation and secretion (Fig. 1A), while myosin light chain is phosphorylated on both (Thr)18 and (Ser)19 residues. As shown in Fig. 1B,C, phosphorylation of both residues occur in a time-dependent manner and peak within 15 s of the addition of agonist. We also observed a slight difference in the kinetics of phosphorylation of MLC (Thr)18 compared with (Ser)19. Most notably at 5 s (Thr)18 was proportionally less phosphorylated than (Ser)19. These data suggest possible differences in the regulation of these two residues of MLC.
Gq-dependent shape change and MLC phosphorylation
Prior studies from our group correlated myosin phosphorylation with the onset of ADP-induced shape change . However, it was not determined whether both residues of MLC were phosphorylated during this response as diphosphorylation of myosin had yet to be described. To evaluate Gq-mediated MLC phosphorylation, we selectively activated the Gq-coupled P2Y1 receptors with 2MeSADP in the presence of a P2Y12 receptor antagonist AR-C69931 MX. As shown in Fig. 2A, platelets activated with 2MeSADP in the presence of AR-C69931 MX undergo rapid shape change, which is completed by 10 s. As shown in Fig. 2B, both MLC residues are phosphorylated downstream of selective Gq activation. In each blot, a platelet sample stimulated with SFLLRN was used both as a positive control and as a measure of a full phosphorylation of each residue. Fig. 2C represents the relative percent of MLC phosphorylation of both residues after selective Gq stimulation in relationship to the maximal phosphorylation levels when normalized to the SFLLRN-stimulated sample. The onset of MLC phosphorylation strongly correlates with the timing of platelet shape change. 2MeSADP-induced MLC (Ser)19 phosphorylation peaks at about 60% of that induced by SFLLRN at the peak time of approximately 5 seconds. The 2MeSADP-induced MLC (Ser)19 phosphorylation rapidly declines consistent with our pervious studies on total myosin phosphorylation [9,19]. In contrast, (Thr)18 phosphorylation is only 20% of the SFLLRN level and is sustained for the full time period studied.
ADP-mediated calcium-dependent and calcium-independent MLC phosphorylation
Our previous studies have shown that ADP-induced platelet shape change is mediated by both Ca2+-dependent and Ca2+-independent signaling pathways . To examine whether the ADP-mediated phosphorylation of MLC (Ser)19 occurs by a Ca2+-dependent pathway, we used 5,5′-dimethyl-BAPTA, a high-affinity Ca2+chelator . As shown in Fig. 3A, 2MeSADP-induced MLC (Ser)19 phosphorylation is almost totally inhibited in BAPTA-treated platelets, while the phosphorylation status of MLC (Thr)18 was minimally affected (Fig 3B). These results suggest that the Ca2+-dependent pathway plays an important role in the phosphorylation of MLC (Ser)19 but does not regulate the phosphorylation of MLC (Thr)18. These data suggest that this partial MLC (Thr)18 phosphorylation may be regulated by Gq-dependent RhoA activation [25,29,30]. Rho-kinase is capable of phosphorylating MLC directly . We have shown that activation of RhoA/P160ROCK pathways downstream of Gq are important for ADP-induced Ca2+-independent shape change . In order to determine whether RhoA-dependent signaling regulates (Thr)18 phosphorylation, we used the Rho kinase inhibitor Y-27632 . As shown in Fig. 3A, in platelets treated with Y-27632, there is small inhibition of MLC (Ser)19 phosphorylation at 5 s, while by 20 s a significant inhibition is observed. MLC (Thr)18 phosphorylation was completely inhibited by Y-27632 (Fig. 3C). As Y-27632-treated platelets activated with 2MeSADP undergo rapid shape change (Fig. 3D), these results suggest MLC (Thr)18 phosphorylation plays little role in this response. Thus, the slight increase in MLC (Thr)18 phosphorylation after stimulation with ADP is probably mediated through a P160ROCK-dependent mechanism.
Effect of G12/13 signaling on MLC phosphorylation
As activation of Gq failed to cause moderate to high levels of MLC (Thr)18 phosphorylation, we hypothesized that MLC (Thr)18 phosphorylation is at a great extent downstream of G12/13 signaling. We activated platelets with the PAR-1 agonist SFLLRN in the presence of YM-254890, a selective Gq inhibitor . Such platelets undergo a delayed change in shape and fail to aggregate, secrete, or as shown in Fig. 4A to mobilize intracellular calcium. As shown in Fig. 4B–D, YM-254890 treated platelets exhibited a significant reduction in MLC (Ser)19 phosphorylation and only a slight decrease in MLC (Thr)18 phosphorylation that is not statistically significant. Similar results were obtained when thrombin-stimulated platelets were treated with BAPTA; MLC (Ser)19 phosphorylation was inhibited by about 50% whereas MLC (Thr)18 phosphorylation was not affected (data not shown). The contribution of G12/13 to the phosphorylation of of MLC Thr(18) was confirmed using a low concentration of U46619 (30 nmol L−1) which does not activate Gq pathways (4E) [24,32,33]. All these data support the hypothesis that RhoA signaling is essential for MLC (Thr)18 phosphorylation and that G12/13-dependent RhoA signaling provides the greater portion of this phosphorylation with strong agonists, whereas Gq-mediated RhoA activation only contributes weakly. These results also suggest that both Gq-pathways are important for contributing to significant levels of MLC (Ser)19 phosphorylation during the rapid calcium transient whereas RhoA pathways contribute to sustained phosphorylation (Ser)19 which can be attributed to inhibition of myosin phosphatase.
Correlation of MLC (Thr) 18 phosphorylation with dense granule secretion
MLC diphosphorylation has previously been reported to play a role in secretion [34–36]. We correlated MLC phosphorylation levels in aspirin or indomethacin-treated platelets with agonists that either cause or fail to cause secretion. Activation of the PAR or GPVI receptors with peptides SFLLRN, AYPGKF, or thrombin, and convulxin, respectively, induced platelet secretion (Fig. 5A), and a moderate increase in both MLC (Ser)19 and MLC (Thr)18 phosphorylation levels were observed (Fig. 5B). In comparison, when aspirin or indomethacin-treated platelets were activated with 2MeSADP or serotonin, agonists which failed to cause secretion, MLC (Ser)19 phosphorylation levels are inhibited compared with agonists which caused secretion, and only minimal levels of MLC (Thr)18 phosphorylation were detected (Fig. 5B). As shown in Fig. 5C, when non-aspirin-treated platelets were activated with 2MesADP, these platelets underwent secretion because of the positive feedback of thromboxane on its receptor. Significant levels of MLC (Ser)19 phosphorylation occurred and significant levels of MLC (Thr)18 phosphorylation preceded the initiation of ATP release. These data suggest a threshold level of both MLC (Ser)19 and MLC (Thr)18 may be required for ATP release.
Role of MLC (Thr)18 phosphorylation in dense granule secretion
As MLC (Thr)18 phosphorylation occurs through a G12/13/RhoA-dependent mechanism we evaluated ATP release in the presence of a RhoA pathway inhibitor, Y-27632. ATP secretion was partially inhibited in aspirin-treated platelets activated with PAR-1 peptide SFLLRN in the presence of Y-27632 (Fig. 6A). As shown in Fig. 6B, there is no significant difference in MLC (Ser)19 phosphorylation levels. However, Y-27632 treatment produces a significant inhibition in MLC (Thr)18 phosphorylation levels (Fig. 6C). The inhibition of secretion is similar to the inhibition in MLC (Thr)18 phosphorylation. Y-27632 failed to completely inhibit the MLC (Thr)18 phosphorylation that occurs after PAR-1 activation, suggesting Y-27632 insensitive kinases upstream of P160ROCK may be able to phosphorylate MLC (Thr)18.
Effect of YM-254890 or Y-27632 on PAR-1-induced myosin light chain phosphorylation
In order to quantitate the actual percentage of myosin phosphorylation that occurs on each of the two residues of MLC, we used urea gels to detect the shifts in the mobility of the phosphorylated proteins. This method was first used by Perrie and Perry , who showed that MLC mobility increased on a native acrylamide gel when it become phosphorylated. Platelets were activated with SFLLRN and the proteins resolved on urea gels and then either stained for protein (Fig. 7A), or probed with either total MLC antibodies or phospho-specific antibodies for MLC (Ser)19 or (Thr)18 (Fig. 7B,C). There was a large SFLLRN-induced increase in the percentage of MLC (Ser)19 phosphorylation and this change correlates well with the change seen both in the protein staining pattern Fig. 7A (70%) and the antibody-dependent detection of total MLC (Fig. 7B1,C1). Using an overlay (Fig. 7C3) of one blot probed for total MLC and the same blot reprobed for (Thr)18 phosphorylation shows a band that runs slightly faster than the total phosphorylated light chain. This result indicates that phosphorylation of (Thr)18 adds a slight increase to the mobility of MLC beyond that induced by MLC (Ser)19 phosphorylation alone. Unfortunately, we were not able to detect an increase in mobility of total MLC either by protein staining or by probing for total MLC. These data suggest that only a small but important fraction of MLC becomes phosphorylated on (Thr)18. The other two lanes in each blot show the effect of the Gq inhibitor, YM-254890 or the P160ROCK inhibitor, Y-27632 on these staining patterns. In the presence of YM-254890, the amount of MLC (Ser)19 phosphorylation was dramatically inhibited (Fig. 7B) consistent with Fig. 4. Two bands of MLC phosphorylated on (Thr)18 are seen in Fig. 7C (lane 3) a prominent band probably representing MLC phosphorylation on both (Ser)19 and (Thr)18 and a less prominent band phosphorylated on only (Thr)18. In the presence of Y-27632, (Thr)18 phosphorylation was dramatically reduced, whereas (Ser)19 phosphorylation was close to the stimulated control. Overall these experiments are consistent with the results of our other experiments.
In the present study, we investigated the significance of diphosphorylation of MLC on the functional responses of human platelets. We also addressed the role of Gq and G12/13-mediated signaling pathways regulating agonist-dependent phosphorylation of MLC (Thr)18 and MLC (Ser)19. Previous studies in platelets focused on the gross phosphorylation of MLC without distinguishing the relative contribution of each residue. As a result, little is known about the role each distinct MLC phosphorylation plays in human platelet functional responses. We have demonstrated that when platelets are fully activated by PARs MLC becomes phosphorylated on both (Thr)18 and (Ser)19.
PARs couple to both heterotrimeric G proteins Gq and G12/13 [38,39]. In the present study, we have shown that activation of the PAR-1 receptor under conditions where platelets change shape, aggregate and secrete dense granules [16,32] undergo a time-dependent increase in both MLC (Ser)19 and (Thr)18 phosphorylation (Fig. 1A,B). The kinetics of phosphorylation of MLC (Thr)18 were slightly slower, which is consistent with the differences in kinetics of MLC (Ser)19 and (Thr)18 observed in smooth muscle [17,40]. The difference in kinetics led us to explore the possibility that MLC (Ser)19 and (Thr)18 may be differentially regulated downstream of Gq and G12/13 pathways and play a role in different functional responses.
Stimulation of aspirinated human platelets with ADP, through P2Y1, causes activation of Gq and PLCβ2, which results in mobilization of intracellular Ca2+and PKC activation [41–45]. It has been shown that platelets activated with ADP under these conditions undergo phosphorylation of MLC, which coincides with shape change, suggesting its importance in the initiation of the shape change event . ADP-stimulated platelet myosin phosphorylation is regulated by Ca2+-dependent MLCK activation and Rho/p160ROCK-activated inhibition of the myosin phosphatase, MYPT1, apparently through Gq-mediated mechanisms as ADP does not activate G12/13 [19,25,32].
Whether diphosphorylation of MLC is required or plays a role in Ca2+-dependent and -independent ADP-mediated shape change had not been investigated. We show that after activation of Gq pathways by ADP, phosphorylation of MLC (Ser)19 is rapid and robust (Fig. 2B,C), but the level of (Thr)18 phosphorylation was relatively weak compared with the level seen after concomitant Gq and G12/13 activation (Fig. 2C). In the presence of the p160ROCK inhibitor, Y-27632, this low level of MLC (Thr)18 phosphorylation is abolished without effecting the initial rate of shape change (Fig. 3D) . These data indicate that MLC (Thr)18 phosphorylation is mediated through Rho kinase and plays little role in initiation of shape change. At 5 s the level of MLC (Ser)19 phosphorylation is minimally effected by Y-27632 indicating that shape change depends primarily on a p160ROCK-independent activation of MLC (Ser)19 phosphorylation (Fig. 3A). At later times, MLC (Ser)19 phosphorylation is inhibited by Y-27632, suggesting that maintenance of inactive MYPT1 is required for prolonged MLC (Ser)19 phosphorylation. These conclusions are completely consistent with previous observations [19,32].
YM-254890 and BAPTA have been used to inhibit Gq-dependent pathways, leaving G12/13 pathways intact. PAR-1-stimulated phosphorylation of MLC (Ser)19 is dependent on both Gq and G12/13-coupled events (Fig. 4B). As MLC (Thr)18 phosphorylation is not significantly attenuated when Gq is inhibited (Fig. 4B), the majority MLC (Thr)18 phosphorylation must be downstream of G12/13 with a small component dependent of Gq-activated RhoA and not dependent on either Ca2+or PLCβII activation. Secretion does not occur under selective G12/13 activation because other requirements, such as intracellular Ca2+mobilization and PKC activation, are not met. At the current time, there is no condition where we can activate platelets to solely cause MLC (Thr)18 phosphorylation independent of MLC (Ser)19 phosphorylation to test whether phosphorylation of MLC (Thr)18 alone could support platelet functional responses. This is probably because of the fact that (Thr)18 phosphorylation may require prior phosphorylation on MLC (Ser) 19 (Fig. 7C) .
It has been shown that the diphosphorylation of MLC at (Thr)18 and (Ser)19 further increases actin-activated myosin ATPase activity over that of monophosphorylated MLC (Ser)19 [17,40]. Furthermore, in support of a conclusion of the present study, diphosphorylation of MLC by MLCK and PKC was required for secretion in RBL-2H3 cells . To evaluate whether MLC diphosphorylation is required for secretion, we activated platelets with strong agonists. Activation of platelet receptors with strong agonists such as thrombin and collagen are known to cause platelet dense granule release (Fig. 5A). However, weaker agonists such as ADP or serotonin are unable to secrete dense granule contents in the absence of thromboxane generation [41–44,47] possibly because of the inability of these receptors to activate G12/13 pathways and thus provide sufficient MLC phosphorylation (Fig. 5A) . We show that the strong platelet agonists thrombin, SFLLRN, AYPGKF and convulxin caused moderate to high levels of both MLC (Ser)19 and (Thr)18 phosphorylation, whereas the weak agonists ADP and serotonin caused only modest levels of MLC (Ser)19 and minimal levels of (Thr)18 phosphorylation when thromboxane generation is blocked (Fig. 5B). In non-aspirinated platelets, stimulation with ADP can cause dense granule secretion because of potentiation by thromboxane A2. Activation of dense granule secretion is delayed under these conditions allowing a clear separation between shape change and onset of secretion. We find that phosphorylation of MLC (Ser)19 correlates with the initial shape change whereas MLC (Thr)18 phosphorylation is delayed and occurs prior to ATP secretion (Fig. 5C). Finally, when Y-27632 is used in platelets stimulated with a PAR-1 agonist, both ATP secretion and MLC (Thr)18 phosphorylation are attenuated. MLC (Ser)19 phosphorylation is not affected in spite of the fact that a significant portion of MLC (Ser)19 phosphorylation is downstream of G12/13 activation. This suggests that p160ROCK-independent pathways regulate a portion of MLC (Ser)19 and perhaps MLC (Thr)18 phosphorylation. We suggest that either ZIPK or ILK may contribute to the regulation of Gq-independent MLC phosphorylation . One caveat should be noted in that all these data are correlative in nature and do not provide absolute proof of this hypothesis.
From the current study and prior studies, activation of G12/13/Rho A pathways prolongs MLC phosphorylation [48,49] and this is necessary for irreversible aggregation . Rho A activation and MLC phosphorylation have also been implicated in playing a role in ‘internal contraction’ which could be important for driving dense granule release [24,50]. We demonstrated that under conditions where intracellular Ca2+is mobilized and there is PKC activation, the maintenance of moderate levels of both MLC (Ser)19 and MLC (Thr)18 phosphorylation are required for full dense granule secretion. While this is the first proposal of such a mechanism in platelets, a similar conclusion was reached in the case of RBL-2H3 cells .
Attempts to measure absolute amounts of MLC (Thr)18 phosporylation by urea gels indicated that only low levels of (Thr)18 phosphorylation occurred under conditions where secretion ensued. This is consistent with the results of Itoh et al.  who directly detected low levels of diphosphorylated MLC after thrombin stimulation. We propose this is consistent with a small regulated fraction of MLC that is directly coupled to the release of dense granules.
In conclusion, we have linked MLC (Ser)19 phosphorylation to platelet shape change and have shown that MLC (Ser)19 is phosphorylated downstream of both Ca2+-dependent Gq activation and Ca2+-independent G12/13 pathways. We have also provided evidence that MLC (Thr)18 phosphorylation occurs subsequent to MLC (Ser) 19 primarily through G12/13/Rho A activation and that diphosphorylated MLC may play a role in dense granule secretion.
This work is supported by HL81322, HL80444 and HL60683 from National Institutes of Health to S.P. Kunapuli. T. M. Getz is supported by NIH training grant in Thrombosis (HL07777).
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.