Yun Zhang, Kunming Institute of Zoology, The Chinese Academy of Sciences, 32 East Jiao Chang Road, Kunming, Yunnan 650223, China. Tel/fax: +86 871 5198515. E-mail: email@example.com
Summary. Background: Prohibitins (PHBs), comprising the two homologous members PHB1 and PHB2, are ubiquitously expressed and highly conserved. The membrane PHBs have been reported to be involved in typhoid fever, obesity, and cancer metastasis. Proteomic studies have revealed the presence of PHBs in human platelets, but the roles of PHBs during platelet aggregation are unknown.Objectives: To investigate the role of PHBs in platelet aggregation. Methods and results: PHB1 and PHB2 were detected on the surfaces of human platelets by flow cytometry and confocal microscopy. The PHBs were distributed in lipid rafts, as determined by sucrose density centrifugation. In addition, the PHBs were associated with protease-activated receptor 1 (PAR1), as determined by Bm-TFF2 (a PAR1 agonist)-affinity chromatography, coimmunoprecipitation, and confocal microscopy. The platelet aggregation, αIIbβ3 activation, granular secretion and calcium mobilization stimulated by low concentrations of thrombin (0.05 U mL−1) or PAR1-activating peptide (PAR1-AP) (20 μm) were reduced or abolished as a result of the blockade of PHBs by anti-PHB antibodies or their Fab fragments; however, the same results were not obtained with induction by high concentrations of thrombin (0.6 U mL−1) or protease-activated receptor 4-activating peptide (300 μm). The calcium mobilization in MEG-01 megakaryocytes stimulated by PAR1-AP was significantly suppressed by PHB depletion with RNA interference against PHB1 and PHB2. Conclusions: PHBs are localized on the human platelet membrane and are involved in PAR1-mediated platelet aggregation. Until recently, PHBs were unknown as regulators of PAR1 signaling, and they may be effective targets for antiplatelet therapy.
Prohibitins (PHBs) belong to a larger family of proteins that share an evolutionarily conserved stomatin/prohibitin/flotillin/Hf1K/C (SPFH) domain, also known as the PHB domain [1–4]. Members of this protein family, which exhibit a propensity to oligomerize and an enrichment in lipid raft microdomains in diverse cellular membranes, are implicated in crucial cellular processes, such as signal transduction, membrane protein chaperoning, and vesicle and protein trafficking [3,4]. PHBs, comprising the two homologous members PHB1 (32 kDa) and PHB2 (37 kDa), are ubiquitously expressed and highly conserved proteins that have important functions [4,5]. The absence of PHBs leads to severe phenotypes and deficiencies in higher eukaryotes, and PHB2 knockout is lethal in mice [6,7]. PHBs have been found to be localized to several cellular compartments, such as mitochondria, the plasma membrane, and the nucleus, and have been implicated in the stabilization of mitochondrial proteins, transcriptional regulation, the regulation of sister chromatid cohesion, and cellular signaling [4,5]. In human platelets, proteomic studies have revealed the presence of these proteins , but their roles in platelet activation are unknown.
Platelet activation plays a central role in thrombosis and hemostasis . The initiation of a platelet thrombus is induced by a variety of stimuli, including collagen, thrombin, ADP, and thromboxane A2, which act in concert to ensure the rapid formation of a platelet plug at sites of vascular injury [9,10]. The most potent activator of platelets is thrombin [10,11]. Thrombin induces platelet activation through a unique family of G-protein-coupled receptors, namely protease-activated receptors (PARs). Two of these receptors (PAR1 and PAR4) are present in human platelets. PAR1 is a high-affinity receptor that mediates the activation of human platelets at low thrombin concentrations. In the absence of PAR1 function, PAR4 can mediate platelet activation, but only at high thrombin concentrations [12,13]. Inhibition of the thrombin-stimulated PAR1 signaling pathway in platelets has emerged as a target for clinical development, but the specificity of this type of treatment is a major problem, because PAR1 is widely expressed in various cell types [13,14]. Recently, we found that a frog trefoil factor, Bm-TFF2, from skin secretions of the frog Bombina maxima, is able to bind and activate PAR1 to induce platelet activation. PAR1 activation by the protein is independent of receptor cleavage and tethered-ligand unmasking. Bm-TFF2 is much more potent than PAR1-activating peptide (PAR1-AP) in inducing human platelet activation [15,16].
In the present study, we show that PHBs are localized to the platelet membrane, are associated with PAR1, and are involved in PAR1-mediated platelet aggregation.
Materials and methods
The thrombin used in the present study was purchased from Calbiochem (La Jolla, CA, USA). Collagen was purchased from Chrono-Log Corp. (Havertown, PA, USA). PAR1-AP (SFLLRN) and PAR4-activating peptide (PAR4-AP) (AYPGKF) were synthesized by GL Biochem (Shanghai, China). CNBr-activated Sepharose 4B was purchased from Amersham Biosciences (Uppsala, Sweden). Bm-TFF2, a frog trefoil factor that activates human platelets via PAR1, and stejnulxin, a snake venom C-type lectin-like protein that activates platelets via glycoprotein (GP)VI, were purified from B. maxima skin secretions and Trimeresurus stejnegeri venom, respectively, as previously described [15,17]. The anti-CD32 mAb IV.3 was purchased from StemCell Technologies (Vancouver, BC, Canada). The anti-CD61 mAb fluorescein 5(6)-isothiocyanate (FITC)-coupled anti-CD62P, anti-αIIbβ3 (PAC-1) antibodies and control IgGs were purchased from BD Biosciences (San Jose, CA, USA). The anti-PAR1 mAb (ATAP2), anti-flotillin-1, anti-ERK2 and the horseradish peroxidase-conjugated, FITC-conjugated and phycoerythrin (PE)-conjugated secondary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The anti-PHB1 polyclonal antibody was purchased from Neomarkers (Fremont, CA, USA) (RB-292-P0) or R&D Systems (Minneapolis, MN, USA) (AF3470). The anti-PHB2 antibody (07-234) was purchased from Millipore (Billerica, MA, USA). The specificities of the antibodies against PHBs were confirmed by full gel western blotting (Fig. S1). All other reagents were purchased from Sigma (St Louis, MO, USA). The protein concentrations were determined with a protein assay kit (Bio-Rad, Hercules, CA, USA), with bovine serum albumin (BSA) as a standard. Human platelet-rich plasma was purchased from the Yunnan Blood Center (Kunming, China). Washed human platelets were prepared as previously described . The Fab fragments of the anti-PHB1 and anti-PHB2 antibodies were prepared with a commercial Fab preparation kit (Thermo Fisher Scientific, Rockford, IL, USA), according to the manufacturer’s instructions.
The flow cytometry method that was used in the present study was similar to that described in our previous report . For the detection of the surface expression of PHBs by immunofluorescence staining, washed platelets were incubated with the appropriate primary and secondary antibodies. After being washed three times, the sample was analyzed with a flow cytometer (FACSVantage SE; Becton Dickinson, Franklin Lakes, NJ, USA ). For the examination of αIIbβ3 activation or P-selectin expression, the washed platelets were activated with an agonist in the presence of FITC–PAC-1 (1 : 100) or FITC–anti-CD62P (1 : 100) at 37 °C without stirring. FITC-labeled isotype-matched antibodies were used as controls. After incubation for 20 min, the platelets were washed and analyzed as described above.
The washed platelets were incubated with the appropriate primary and secondary antibodies (FITC-conjugated anti-rabbit and anti-goat IgGs, and PE-conjugated anti-mouse IgGs). Isotype-matched normal rabbit and mouse IgGs were used as controls, and did not show any staining on platelets. After being washed, the platelets were resuspended in phosphate-buffered saline (PBS), and the suspension was placed on a slide. The slide was observed under a confocal microscope (Zeiss LSM 510 Meta; Jena, Germany) equipped with a Plan-Neofluar 40 × 1.3 oil DIC objective.
Platelet membrane preparations
The platelet membranes were prepared according to the method of Kim et al. . Briefly, washed platelets suspended in buffer A (25 mm Tris, 5 mm MgCl2, pH 7.4) were sonicated (four 30-s bursts with 30-s intervals) and centrifuged (3000 × g, 10 min). The supernatant was centrifuged (100 000 × g, 30 min), and the resulting pellet was solubilized by homogenization with a glass homogenizer in buffer B (50 mm Tris, 5 mm MgCl2, 10 mm Chaps, pH 7.4). The solubilized mixture was centrifuged (100 000 × g, 30 min), and the supernatant was collected as the platelet membrane preparation.
Isolation of lipid rafts
Sucrose density gradient centrifugation was used to isolate the lipid rafts, according to the methods of Quinton et al. . Briefly, platelets (1 × 1011 mL−1, 500 μL) were lysed with lysis buffer and incubated for 5 min at 4 °C. The lysate was mixed with an equal volume of ice-cold 80% sucrose in Mes-buffered saline (150 mm NaCl, 25 mm Mes, 1 mm phenylmethanesulfonyl fluoride, pH 6.5), resulting in a mixture containing 40% homogenate (1 mL), and were placed at the bottom of a centrifuge tube. Successive additions of 30% (1.5 mL) and 5% (0.75 mL) sucrose in Mes-buffered saline were consecutively layered upon the 40% homogenate, and the tube was centrifuged at 100 000 × g for 18 h at 4 °C (Beckman Instruments, Fullerton, CA, USA). Gradient fractions (0.3 mL each) were sequentially removed from the top of the gradient.
Bm-TFF2–Sepharose 4B affinity chromatography
Bm-TFF2, stejnulxin or glycine was coupled to CNBr-activated Sepharose 4B beads, according to the manufacturer’s instructions. The platelet surface biotinylation was performed according to the method of Lu et al. . Briefly, washed platelets (5 × 109 mL−1) in PBS were incubated with 1 mg mL−1 EZ-link Sulfo-NHS-LC-Biotin (Pierce, Rockford, IL, USA) for 30 min at room temperature. After the free biotin had been removed, the biotinylated platelets were used to prepare platelet membranes. The affinity chromatography method that was used in the present study was described in our previous report . The column was washed extensively with lysis buffer. The binding proteins were analyzed by SDS-PAGE or western blotting.
In-gel protein tryptic digestion and mass spectrometry (MS) analysis
For the MS analysis, unlabeled washed platelets were used, and the SDS-PAGE gel was silver-stained. After destaining, washing, and dehydration, the gel bands were incubated with 10 μg mL−1 trypsin in 25 mm NH4HCO3 at 37 °C for 3 h. The tryptic peptides were cocrystallized with a saturated solution of α-cyano-4-hydroxycinnamic acid in 50% Acetonitrile/0.1% trifluoroacetic acid on a MALDI plate. The MS and MS/MS spectra were obtained with a 4700 proteomics analyzer MALDI-TOF/TOF mass spectrometer (Applied Biosystems, Foster City, CA, USA). The peaks of a calibration mixture (Applied Biosystems) in six calibration spots were used as external standards to calibrate each spectrum to a mass accuracy within 5 p.p.m. for MS or within 10 p.p.m. for MS/MS. The MS spectra were acquired in the reflector positive ion mode, and the peptide masses were acquired in the 850–4000 m/z range. A database search was performed with the mascot search tool (Version 2.0; Matrix Science, London, UK) to screen Swiss-Prot.
Immunoprecipitation and western blotting
The washed platelets were lysed with NP-40 buffer (50 mm Tris, 150 mm NaCl, 0.5% NP-40, 2% BSA, and complete protease inhibitor cocktail, pH 7.4). The lysate was clarified by centrifugation (12 000 × g, 25 min, 4 °C). After preclearance with protein A/G–agarose beads, the lysate was incubated with 2 μg of the primary antibody overnight at 4 °C. The protein A/G–agarose beads were added to the lysate and incubated for 2 h at 4 °C. The beads were pulled down and washed three times, and the immune complexes were removed from the beads by boiling for 10 min in SDS-PAGE sample buffer for electrophoresis and western blotting. The western blotting was performed as described previously . The lysate (containing 30 μg of protein) or immunoprecipitated proteins were loaded onto an SDS-PAGE gel, and, after electrophoresis, the proteins were electrotransferred onto a poly(vinylidene difluoride) membrane. The membrane was subsequently blocked with 3% BSA, and incubated with appropriate primary and secondary antibodies. The protein bands were visualized with an enhanced chemiluminescence reagent (Thermo).
Platelet aggregation and cytoplasmic calcium measurements
The platelet aggregation and cytoplasmic calcium measurements were performed with platelets from several donors, as previously described [15,16]. For the platelet inhibition assays, the platelets were pretreated with a mAb (IV.3) to block anti-FcγRIIA (CD32), to prevent the non-specific actions of the anti-PHB antibodies .
Cell culture and RNA interference (RNAi)
The human megakaryocytic MEG-01 cell line (platelet progenitor cells) was obtained from the American Type Culture Collection (Manassas, VA, USA), and cultured in RPMI-1640 supplemented with 10% (v/v) fetal bovine serum, 100 U mL−1 penicillin and 100 U mL−1 streptomycin at 37 °C in humidified air containing 5% CO2. For RNAi, the cells were cultured to 40–50% confluence before transfection. Small interfering RNAs (siRNAs) against PHB1 (siPHB1) and PHB2 (siPHB2) or the negative control siRNA were used to transfect the cells with the Lipofectamine RNAiMAX Transfection Reagent (Invitrogen, Carlsbad, CA, USA). Two days after transfection, the cells were collected, and 50% of the cells were used for Ca2+ mobilization analysis, whereas the other 50% of the cells were used for western blot analysis. siPHB1 (5′-CAGAAAUCACUGUGAAAUUTT-3′), siPHB2 (5′-CCCAGGAAUUCUCAAUAAATT-3′) and negative control siRNA (4390843) were obtained from Qiagen (Valencia, CA, USA).
The data were representative of three to four independent experiments, and were analyzed with Student’s t-test for variance. The experimental values are expressed as means ± standard deviations. The level of statistical significance was set at P <0.05.
PHBs are present in human platelet membranes
A previous report found that PHBs are expressed in human platelets by using a proteomic analysis , but the functions of the PHBs were unclear. In the present study, flow cytometry revealed that PHB1 and PHB2 were expressed on the surfaces of human platelets (Fig. 1A). The use of confocal microscopy revealed that PHB1 and PHB2 are colocalized with CD61, a platelet membrane-associated receptor (Fig. 1B). The present study also revealed that PHB1 and PHB2 interact with each other in platelet membrane preparations, as determined by coimmunoprecipitation (Fig. 1C). In addition, PHB1 and PHB2, like flotillin-1 (a lipid raft marker protein ), were distributed in platelet lipid rafts (5–20% low-density fractions from sucrose density centrifugation; Fig. 1D). The disruption of lipid raft integrity by cholesterol depletion with MβCD (5 mm) eliminated the lipid raft localization of PHB1, PHB2, and flotillin-1 (Fig. 1D). These results indicate that PHB1 and PHB2 are present in platelet membranes.
PHBs are associated with PAR1 in human platelet membranes
Bm-TFF2 binds and activates PAR1 to induce human platelet aggregation[15,16]. During the process of identifying the membrane receptor(s) of Bm-TFF2 in human platelets, western blotting revealed the enrichment of PAR1 in the Bm-TFF2–Sepharose 4B-affinity chromatography column . In addition, two protein bands with molecular masses of approximately 31 and 35 kDa specifically appeared in the eluate of the Bm-TFF2 column that was loaded with surface biotinylated platelet membrane preparations but not in the eluates from the stejnulxin-coupled or glycine-coupled columns (Fig. 2A). MS analysis (Fig. 2B) and western blotting (Fig. 2C,D) identified these two protein bands as PHB1 and PHB2, respectively. PHB1 and PHB2 did not directly bind to Bm-TFF2 (Fig. S2), and PHB coupling to the Bm-TFF2-affinity chromatography column may be mediated by PAR1. PHBs and PAR1 were detected in the lipid raft fraction (Fig. 1D). We next examined the possible interaction between PHBs and PAR1 by using platelet membrane preparations. The results revealed that PHB1 and PHB2 were associated with immunoprecipitated PAR1, and that PAR1 was also coprecipitated with PHB1 and PHB2 (Fig. 2E). The association between PHBs and PAR1 was observed in MEG-01 megakaryocytes (platelet progenitor cells) (Fig. S3). However, PHBs did not interact with PAR4 in human platelet membranes (Fig. S4). When the platelets were double-immunostained with anti-PHB1 and anti-PAR1 antibodies, colocalization of PAR1 and PHB1/PHB2 was observed (Fig. 2F). These results showed that PHBs are associated with PAR1.
PHBs are involved in PAR1-mediated platelet aggregation, αIIbβ3 activation, and granular secretion
To determine the role of PHBs in platelet aggregation, we used anti-PHB antibodies or their Fab fragments as specific inhibitors to block the PHBs. The specific action of the anti-PHB antibodies on the membrane PHBs was supported by the fact that the antibodies lost their inhibitory capacity after being blocked by recombinant PHBs (data not shown). The results revealed that the blockade of PHBs with anti-PHB antibodies or their Fab fragments significantly reduced or abolished the platelet aggregation that was stimulated by a low concentration of thrombin (0.05 U mL−1) or PAR1-AP (20 or 50 μm) (Figs 3A and S5). However, the anti-PHB antibodies or their Fab fragments did not exhibit an effect on the human platelet aggregation that was stimulated by a high concentration of thrombin (0.6 U mL−1) or PAR4-activating peptide (PAR4-AP) (100 or 300 μm) (Figs 3A and S5), collagen (5 μg mL−1) (Fig. S6), stejnulxin (5 nm; a GPVI agonist), mucetin (300 nm; a GPIb agonist ), A23187 (10 μm; a Ca2+ ionophore), or phorbol 12-myristate 13-acetate (200 nm; an activator of protein kinase C) (data not shown). In addition, the effect of the blockade of PHBs on αIIbβ3 activation and granular secretion was examined. The use of anti-PHB antibodies significantly inhibited the low concentration thrombin-stimulated or PAR1-AP-stimulated αIIbβ3 activation (68% or 75%), CD62P release (56% or 58%), and ATP release (58% or 65%) (Fig. 3B–D). In contrast, the blockade of PHBs did not affect αIIbβ3 activation or the granular secretion activated by PAR4-AP (Fig. S7A–C). These results indicate that PHBs are involved in PAR1-mediated platelet aggregation.
PHBs affect PAR1-mediated Ca2+ mobilization
Ca2+ mobilization is a critical step in PAR1-mediated platelet activation ; therefore, we investigated the effects of PHBs on Ca2+ mobilization stimulated by PAR1 activation. Human platelets and MEG-01 cells were used in these assays. The blockade of PHBs with anti-PHB antibodies significantly inhibited the Ca2+ mobilization in the platelets or MEG-01 cells that was induced by 20 μm PAR1-AP (80% or 73% inhibition) (Fig. 4A). In contrast, the Ca2+ mobilization stimulated by PAR4-AP was not different in the presence or absence of the anti-PHB antibodies (Fig. S7D). Furthermore, the role of PHBs in the PAR1-mediated Ca2+ mobilization was investigated by PHB depletion with RNAi. Treatment of the cells with siPHB1 and siPHB2 significantly decreased the expression of these proteins in MEG-01 cells, as determined by western blotting (Fig. 4B). The PAR1-AP-stimulated Ca2+ mobilization was largely inhibited in the cells transfected with selective siRNAs against PHBs (70% inhibition, P <0.01) as compared with the MEG-01 cells transfected with control siRNAs (Fig. 4B). In contrast, PHB depletion had no significant effect on the Ca2+ mobilization that was induced by U46619 (1 μm), stejnulxin, or A23187 (data not shown). These results indicate that PHBs are involved in PAR1-activated Ca2+ mobilization.
The present study is the first to show that PHBs are localized to the platelet membrane and regulate PAR1-mediated platelet aggregation. Our finding that PHB1 and PHB2 are localized to the platelet membrane is consistent with those of previous studies on the surface expression of PHBs in several cell types, such as human B lymphocytes, intestinal epithelial cells, and colorectal cancer cells [23–25].
How membrane-associated PHBs act at the molecular level has remained poorly understood [3,4]. Interestingly, in the present study, a substantial proportion of the PHBs were observed to be resident in the lipid rafts of platelets, which is in agreement with the properties and proposed functions of SPFH family proteins. It has been hypothesized that all PHB domain-containing proteins act in the generation and maintenance, or the recruitment, of proteins into lipid raft domains [3,26]. In the present study, we demonstrated that PHBs participate in PAR1 signaling. Our finding is consistent with the previous report that flotillin-2, another lipid raft marker protein, is associated with PAR1 in melanoma cells , although the molecular mechanisms are largely unknown. PHB1 and PHB2 were found to be associated with each other in platelet membranes (Fig. 1C). Furthermore, the enrichment of PHBs in platelet membrane lipid rafts, together with PAR1, was observed in platelets (Fig. 1D). It is reasonable to speculate that PHBs may form membrane-bound complexes, and participate in the formation of PAR1 signaling platforms in lipid raft microdomains. Disruption and/or blockade of the functional availability of membrane PHBs resulted in the blockade of the scaffolding and signaling platform formation function of the proteins, leading to inhibition of PAR1 signaling and PAR1-mediated platelet activation. The physiologic relevance of lipid raft proteins in platelet function and the relative contributions to various pathophysiologic conditions are worthy of further investigation.
Platelets play a key role in hemostasis, but are also responsible for the formation of pathogenic thrombi. Unappreciated local and systemic platelet activation are implicated in diverse disorder processes, such as atherosclerosis, vascular restenosis, acute lung injury, and transplant rejection [9,28]. Thrombin is a very potent platelet agonist, and thrombin-dependent platelet activation and aggregation have been shown to be increased as a result of angioplasty and stenting, which may cause clinical complications, including acute myocardial infarction and death [29,30]. Thrombin activates platelets at extremely low concentrations (lower than those required for the activation of the coagulation cascade). This activation occurs mainly through PAR1, which is a strong and high-affinity receptor [12,14]. In the present study, PHBs were identified as regulators of PAR1-stimulated platelet aggregation. Targeting of PHBs might be a useful therapeutic approach for antiplatelet therapy.
In conclusion, we have shown that PHBs are localized to the platelet membrane and are involved in PAR1-mediated platelet aggregation. These findings offer new opportunities for studies on the regulation of thrombin-induced signaling pathways in platelet activation and the functional implications in various physiologic and disease states.
Y. Zhang, Y. Wang, Y. Xiang, and W. Lee: designed the study, performed the research, and analyzed the data; Y. Zhang and Y. Zhang: designed the study, performed the research, analyzed the data, and wrote the paper.
This work was supported by grants from the National Basic Research Program of China (973 Program, 2010CB529800) and the National Natural Science Foundation of China (NSFC-Yunnan joint funding U1132601, 30630014, and 30870304) to Y. Zhang.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.