|Home | About | Journals | Submit | Contact Us | Français|
Protein Kinase C delta (PKCδ) is expressed in platelets and activated downstream of protease-activated receptors (PAR)s and glycoprotein VI (GPVI) receptors.
To investigate the role of PKCδ in platelets.
We evaluated the role of PKCδ in platelets using two approaches - pharmacological and molecular genetic approach. In human platelets pretreated with isoform selective antagonistic RACK peptide (δV1-1)TAT, and in the murine platelets lacking PKCδ, PAR4-mediated dense granule secretion was inhibited, whereas GPVI-mediated dense granule secretion was potentiated. These effects were statistically significant in the absence and presence of thromboxane A2 (TXA2). Furthermore, TXA2 generation was differentially regulated by PKCδ. However, PKCδ had a small effect on platelet P-selectin expression. Calcium- and PKC-dependent pathways independently activate fibrinogen receptor in platelets. When calcium pathways are blocked by dimethyl-BAPTA, AYPGKF-induced aggregation in PKCδ null mouse platelets and in human platelets pretreated with (δV1-1)TAT, was inhibited. In a FeCl3–induced injury in vivo thrombosis model, PKCδ −/− mice occluded similar to their wild type littermates.
Hence, we conclude that PKCδ differentially regulates platelet functional responses such as dense granule secretion and TXA2 generation downstream of PARs and GPVI receptors, but PKCδ deficiency does not affect the thrombus formation in vivo.
Protein kinase Cs (PKCs) are members of the extended AGC (Protein Kinase A, G and C) family of differentially expressed serine/threonine kinases implicated in a diverse array of cellular functions1, 2. Following activation, these kinases migrate to different subcellular locations including the plasma membrane and cytoskeletal elements where they regulate different physiological functions3. PKC isoforms are subdivided into three groups based on their lipid and cofactor requirements: the diacylglycerol (DAG) and calcium-sensitive conventional isoforms (α, βI, βII and γ), the DAG-sensitive and calcium-insensitive novel isoforms (δ, η, θ, and ε) and the phosphatidylinositide trisphosphate-sensitive atypical isoforms (ζ, ι, μ and γ)4.
Intrinsic function of PKCs is regulated by three mechanisms: a) binding of the co-factor that allosterically activates the enzyme b) phosphorylation on the activation loop residue that primes the enzyme for catalysis and c) interaction with proteins that position it close to its regulators and substrates1. RACKs (Receptors for activated C-Kinase) bind to individual PKC isozymes following their activation and translocate them to their target substrates. Each PKC has a unique RACK that regulates the positioning of the PKC in close proximity to their target substrates5. The PKCs then phosphorylate their substrates and mediate downstream signaling events6, 7. Modulation of PKC translocation and function has been disrupted using the inhibitor in various cells, organs and animals8, 9. (δV 1-1)TAT, which is an anti-PKCδ RACK peptide conjugated to a TAT carrier, binds to the RACK binding region on PKCδ. This prevents the binding of endogenous RACK to PKCδ. By using (δV 1-1)TAT in human platelets, we demonstrated the role of PKCδ in regulating platelet functional responses. In order to substantiate the findings from the pharmacological inhibition, we performed similar experiments in murine platelets isolated from PKCδ−/− and their age-matched PKCδ+/+ wild type littermates.
Platelet granule secretion is one of the early events of platelet functional responses. Platelet dense granules contain ADP, ATP, Ca2+and serotonin that amplify platelet responses by acting on their respective receptors35. α–granules on the other hand, contain homologs of plasma proteins and platelet-specific proteins10. One of the components of α–granules is P-selectin that is expressed on the surface of activated platelets and aids in platelet-leukocyte interaction11. Following platelet secretion, TXA2 is generated which diffuses across the platelet membrane and recruits more platelets to the site of injury. Activation of platelets by physiological agonists such as thrombin and collagen result in the generation of thromboxane. At low doses of agonist, secretion is dependent on TXA2 generation and thus further amplification of platelet functional responses ensue. These events result in the formation of platelet plug. Inside-out signaling results in the activation of fibrinogen receptor αIIbβ3. There are two pathways by which the fibrinogen receptor can be activated, a Ca2+-dependent pathway and a PKC-dependent pathway12. However, the specific PKC isoforms that can cause activation of αIIbβ3 in the absence of Ca2+ has not been investigated yet.
Protein Kinase Cs have been implicated in platelet secretion13. Of the novel isoforms, PKCδ is implicated in inhibition of cell growth, cell differentiation, apoptosis and tumor suppression in immune cells14. PKCδ is activated by strong agonists such as thrombin and collagen, but not by ADP 15. Activated PKCδ positively regulates PAR-mediated dense granule secretion and negatively regulates GPVI-mediated dense granule secretion 15, 16. These studies used rottlerin, a inhibitor of PKCδ at IC50 of 3-6 μM17. However, recent reports have raised concerns regarding the specificity of rottlerin18-20 21, 22.
In this study, by combining the pharmacological approaches with the use of platelets from PKCδ knockout mice, we investigated the role of PKCδ in platelet functional responses. We show that PKCδ differentially regulates PAR and GPVI-mediated dense granule secretion and TXA2 generation. These effects translate into faster time to occlusion leading to thrombus formation in vivo.
Approval for this study was obtained from the Institutional Review Board of Temple University (Philadelphia, PA) and mice were used for physiological measurements using the protocol approved by the Institutional Animal Care and Use Committee (IACUC).
Convulxin, fibrinogen (Fraction I, type I), apyrase grade VII, thrombin and acetylsalicylic acid were obtained from Sigma (St.Louis, MO). Luciferin-luciferase reagent was purchased from Chrono-Log (Havertown, PA). Anti-phospho and anti- PKCδ antibodies were obtained Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The acetoxymethyl ester of the calcium chelator 5,5′-dimethyl-bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetra-acetic acid (dimethyl BAPTA) was purchased from NEN (Boston, MA). All the other reagents were of reagent grade, and de-ionized water was used throughout.
PKCδ −/− (C57BL/6 background) mice and their age-matched wild type littermates were used. These mice were obtained from Dr. Keiko Nakayama (Tohoku Graduate School of Medicine, Japan). All the mice obtained from the source were transferred to the Central Animal Facility (CAF) under the head and director of the CAF in Temple University School of Medicine.
The platelets were isolated from humans and mice according to previously established protocol23.
ATP secretion from platelet dense granules was determined in aspirin-treated human platelets or indomethacin-treated mouse platelets by using the Luciferin-Luciferase assay. The platelets were treated with aspirin (1 mM) or indomethacin (10 μM) for 1 minute and then stimulated with agonists in a lumi-aggregometer at 37°C with stirring at 900 rpm, and the corresponding luminescence was measured.
P-selectin (CD62) expression on platelets from PKCδ −/− murine platelets and those from wild type littermates was measured according to a established protocol using FITC-labeled anti-CD62 antibody (Becton Dickinson, San Jose, CA)24
Washed murine platelets without indomethacin treatment were stimulated in a platelet aggregometer and levels of TXB2 were determined according to established protocol25.
Adult mice (6-8 weeks old: weight ~25 gms) were anesthetized by intraperitoneal injection of pentobarbital (40 mg/kg). Experimental groups consisted of PKCδ +/+ and PKCδ−/− mice (n=11). The mice were subjected to thrombosis according to a established protocol24. The operator was blinded to mouse genotype while performing all experiments.
The results were quantitated, expressed as Mean ± S.D. The data was statistically analyzed using Student's t test and P ≤ 0.05 was considered significant.
Previous studies from our lab and others have shown that PKCδ is important for PAR-mediated dense granule release, but it negatively regulates GPVI-mediated secretion15. However, as these studies used a non-selective inhibitor rottlerin, we investigated the role of PKCδ using more selective tools to block this isoform. First, we used δ (V1-1)TAT (1 μM), an anti-PKCδ RACK peptide, that specifically blocks translocation of PKCδ to its target substrate9, 26. When human platelets were pre-incubated for 15 min with δ (V1-1)TAT (1 μM), 200 μM AYPGKF-induced dense granule release in aspirin-treated human platelets was inhibited (Figure 1A). However, when the platelets were treated with a control peptide, there was no effect on AYPGKF-induced secretion (Figure 1A) and these were statistically significant (P< 0.05). Similar experiments were performed by stimulating platelets with 60 ng/ml convulxin, a GPVI agonist. Contrarily, convulxin-induced dense granule release was potentiated by the pretreatment of platelets for 15 min with PKCδ specific antagonistic RACK peptide (Figure 1B). However, the control peptide did not affect the GPVI-mediated secretion (Figure 1B). The potentiation of convulxin-induced ATP secretion by δ(V1-1)TAT was also statistically significant (Figure 1B). These results indicate that PKCδ positively regulates PAR-mediated dense granule secretion, whereas it negatively regulates GPVI-mediated dense granule secretion.
As the specificity of the pharmacological agents might be argued, we used platelets from mice lacking PKCδ isoform. PKCδ knock out mice were generated by two groups14, 27-29. Although this PKC is ubiquitously expressed, the mice that do not express PKCδ show a clear phenotype only in immune cells14. Indomethacin-treated platelets from PKCδ −/− mice and their age-matched wild type littermate controls were stimulated with sub-maximal doses of AYPGKF (200 μM), or GPVI agonists- convulxin (60 ng/ml), collagen-related peptide (CRP, 5 μg/ml), or collagen (20 μg/ml). Platelet aggregations and dense granule secretions were measured in these platelets upon agonist stimulation. As shown in Figure 2A, aggregations and dense granule secretions were inhibited upon stimulation with AYPGKF in PKCδ−/− murine platelets compared to wild type littermates. On the other hand, secretion induced by GPVI agonists were potentiated. These results are consistent with the findings using (δ V1-1)TAT in human platelets. Dense granule secretions were also measured in the absence of indomethacin and inhibition or potentiation was found to be statistically significant (Figure 2B). The actual ATP released in nmoles was quantitated and the inhibition or potentiation were statistically significant (Figure 2A and 2B). These data suggest that PKCδ differentially regulates PARs and GPVI-mediated dense granule secretions in murine platelets.
The exposure of P-selectin from α–granules aids in different cell-cell interaction between platelets, leukocytes and endothelial cells11, 30. As PKCδ differentially regulates dense granule secretion, we investigated the role of this isoform in α–granule secretion. Washed indomethacin-treated murine platelets from PKCδ −/− and wild type littermates were stimulated with sub-maximal doses of AYPGKF or convulxin and P-selectin exposure was measured by flow cytometry. Our results showed that PAR-mediated α–granule secretion, expressed as Mean Fluoroscence Intensity (MFI), was inhibited and GPVI-mediated α–granule secretion was potentiated marginally in PKCδ −/− compared to wild type littermates (Figure 3). These results indicate that PKCδ plays a differential role in regulating α–granule secretion downstream of PARs and GPVI receptors.
We have previously shown, using rottlerin, that PKCδ regulates PAR-mediated thromboxane A2 (TXA2) generation31. Therefore, we next investigated the contribution of PKCδ to TXA2 generation downstream of PARs and GPVI receptors using platelets from PKCδ null mice. PAR-mediated TXA2 generated was inhibited in PKCδ −/− murine platelets compared to the wild type littermates (Figure 4A). On the other hand, GPVI-mediated TXA2 generated was potentiated in PKCδ −/− murine platelets compared to the wild type littermates (Figure 4B). These statistically significant results (P < 0.05) suggest that PKCδ positively regulates PAR-mediated TXA2 generation, whereas it negatively regulates GPVI-mediated TXA2 generation.
Calcium and PKC pathways can independently cause PAR and GPVI-mediated fibrinogen receptor activation12. To investigate the role of PKCδ in PKC-dependent fibrinogen receptor activation, washed aspirin-treated human platelets were pretreated with dimethyl BAPTA (10 μM) for 5 minutes, which completely blocks the calcium-dependent pathways. We confirmed the effectiveness of BAPTA by measuring secretion upon stimulation with AYPGKF, which was abolished. These BAPTA-treated platelets were then pre-incubated with (δV 1-1)TAT or control peptide for 10 minutes, stimulated with sub-maximal doses of PAR4 agonist peptide, AYPGKF (200 μM) in the presence of fibrinogen (1 mg/ml). This resulted in the inhibition of AYPGKF-induced aggregation, while the control peptide had no effect (Figure 5A). Similarly, when indomethacin-treated PKCδ −/− and wild type platelets were preincubated with BAPTA for 5 minutes, AYPGKF-induced (200 μM), aggregation was inhibited (Figure 5B). These results indicate that PKCδ is one of the isoforms contributing to PKC-dependent platelet aggregation. However, as the aggregation was not abolished in the case of pan-PKC inhibitors12, suggesting that other PKC isoforms might also be contributing to this event.
We investigated whether the results using ex vivo platelets translate into an in vivo aberration in thrombus formation in mice32. Hence, we evaluated the role of PKCδ in thrombus formation in vivo using FeCl3-induced carotid artery injury model. As shown in Figure 6A & B, the time to occlusion (TTO) in a representative wild type and PKCδ−/− mice were 6.5 minutes and 4 minutes, respectively. However, the time to occlusion of wild type (12.85 ± 11.13 min) and PKCδ−/− (7.94 ± 7.56 min) shown in Figure 6C (n =11) indicate that the differences in times to occlusion are not statistically significant (p> 0.1) as judged by student t test. These results indicate that PKCδ deficiency does not affect the thrombus formation in vivo.
Previously, differential regulation of dense granule secretion by PKCδ isoform downstream of PAR and GPVI receptors was studied using rottlerin15. However, recent reports have indicated problems regarding the specificity of rottlerin as a PKCδ inhibitor 18-20. It inhibits other protein kinases such as MAPK-activated protein kinase 2 and p38 regulated/activated kinase, and other enzymes such as β–lactamase, chymotrypsin and malate dehydrogenase18-20. In order to substantiate the previous findings with rottlerin, we used two complementary approaches in the current study. The first approach is the pharmacological approach, using (δV 1-1)TAT, an anti-PKCδ RACK peptide, and the second approach, extending the pharmacological findings, is the use of mice lacking PKCδ.
For the pharmacological inhibition of PKCδ, (δV 1-1)TAT was used. The specificity of the deltaV1-1 TAT peptide has been well established by previous studies 6, 26, 33, wherein the anti-PKCδ RACK peptide was shown to inhibit the translocation of PKCδ but not PKCα, βI, ε. Furthermore in platelets, 2MeSADP does not activate PKCδ 15. 2MeSADP-induced aggregations were not affected in human platelets that were pretreated with anti-PKCδ RACK peptide suggesting that this peptide is specific for PKCδ isoform (data not shown). Our results using (δV 1-1)TAT in human platelets and platelets from mice lacking PKCδ showed inhibition of PAR-mediated and potentiation of GPVI-mediated platelet dense granule secretions and TXA2 generation. The effects of PKCδ were statistically significant only at sub-maximal doses of the agonist and not at higher doses (data not shown). Thus, PKCδ differentially regulates dense granule secretion and thromboxane generation in platelets downstream of PARs and GPVI. Consistently, we observed a more rapid thrombus formation in PKCδ null mice in comparison to its wild type littermates in vivo.
Previous findings from Pula G et al.34 reported that PKCδ does not play any significant role in GPVI-mediated dense granule secretion and fibrinogen binding. Our results are in contrast with these findings, because we observed a negative regulatory role for this PKC isoform downstream of GPVI in this study. There are plausible explanations for these differences in the results. Previous studies in the PKCδ −/− were carried out using a very low concentration of 3 μg/ml collagen to measure the release of 5-HT as a marker for dense granule secretion. At low doses of collagen, the effects are mostly dependent on feedback mediators such as ADP and TXA2. Although, ADP was scavenged using 0.2 U/ml apyrase, the effects that were reported may be mediated by TXA2, because dense granule secretions were measured in the absence of indomethacin. Therefore, under such low collagen concentrations, the role of PKCδ in dense granule secretion cannot be observed. Moreover, dense granule secretions were measured using serotonin assay (5-HT assay) where radioactive 14C is used. In this case, the platelet response cannot be monitored continuously. In our ATP secretion assay, we monitored both aggregation and secretion continuously and hence observed the differences in the responses. Moreover, we strengthened our findings using other GPVI receptor agonists like CRP and convulxin in addition to collagen. Therefore, these discrepancies seen between our data and data from Pula et al could be attributed to different experimental approaches34. Such discrepancies have been noted in the case of other studies using the same knock out mice strains. For example, results from SHIP-1−/− murine platelets by Severin et al 35. and Pasquet et al 36are contradictory. Although both the groups used mice of the same strain, there were differences in the experimental results. These may be attributed to the differences in experimental methodology.
Following platelet secretion, TXA2 is generated which amplifies the platelet functional responses. Previous reports indicated a role for PKCδ in regulating PAR-mediated TXA2 generation31, but its role downstream of GPVI receptors is not known. Our current data suggest that PKCδ positively regulates AYPGKF-induced TXA2 generation and negatively regulates GPVI-mediated TXA2 generation. The enhanced platelet functional responses, viz. secretion and thromboxane generation, downstream of GPVI stimulation in PKCδ null mice, however, did not translate into statistically significant difference in the time to occlusion of the ferric chloride-injured artery in these mice.
We have previously shown that PAR-mediated platelet aggregation occurs in a calcium- and PKC-dependent manner12. Here, we show that PKCδ contributes to the PKC-dependent platelet aggregation by AYPGKF. As pan-PKC isoforms completely abolish PKC-dependent aggregation12 but this aggregation was only partially blocked by PKCδ antagonistic RACK peptides or in PKCδ null murine platelets. These results suggest that other PKC isoforms might contribute to the PKC-dependent aggregation. Previous studies have shown that BAPTA treatment enhances the membrane translocation of the PKCε isoform37. Thus, it is possible that the PKCε isoform might be more activated and take over the function of PKCδ in BAPTA treated platelets.
Signaling through PARs involves G-protein signaling pathways unlike GPVI receptor signaling that is mediated by tyrosine kinases. Therefore, the downstream signaling molecules are different in PAR and GPVI-signaling pathways. PAR-mediated signaling involves PLCβ2, whereas the key signaling molecule in GPVI signaling is, PLCγ2. These signaling molecules upstream of PKCδ may differentially regulate PKCδ and result in the differential regulation of platelet functional responses. However, the molecular mechanism of such differential regulation is not clear. Recent reports have indicated Rab27b (a Rab27 GTPase subfamily) and Ral (a GTPase) to regulate platelet dense granule secretion38, 39. It is possible that PKCδ may regulate small G-proteins, that aid in the fusion of t-SNAREs and v-SNAREs, differentially downstream of PAR and GPVI-mediated platelet exocytosis. Future studies have to address the mechanism of such differential regulation by this PKC isoform.
In conclusion, PKCδ differentially contributes to the regulation of PAR- and GPVI-mediated secretion and TXA2 generation. However, the altered platelet functional responses in PKCδ null mouse platelets do not affect thrombus formation in the injured artery in these mice.
The authors thank Dr. Mochly-Rosen and Dr. Grant Budas (Department of Chemical and Systems Biology, Stanford University School of Medicine, Stanford, CA) for generous gift of (δV 1-1)TAT and the control peptide. They also thank Ms. Monica Dupon for her technical assistance.
Sources of Funding:
This work was supported by pre-doctoral fellowships from American Heart Association, Great Rivers Affiliate to R.C (0715319U) and S.M (0415440U) and grants from NIH, HL60683, HL80444, and HL81322.
This work was supported by Research Grants HL60683, HL81322 and HL80444 from the National Institutes of Health (S. P. K.) and Pre-doctoral fellowships from the American Heart Association, Great Rivers Affiliate (R.C and S.M).