Journal of Thrombosis and Haemostasis, 12: 1921–1927
DOI: 10.1111/jth.12676
IN FOCUS
Group V secretory phospholipase A2 impairs endothelial protein C receptor-dependent protein C activation and accelerates thrombosis in vivo I . T A M A Y O , * S . E . V E L A S C O , * C . P U Y , * C . T . E S M O N , † M . G . D I C H I A R A , * R . M O N T E S * and J. HERMIDA* *Division of Cardiovascular Sciences, Laboratory of Thrombosis and Hemostasis, Center for Applied Medical Research, University of Navarra, Pamplona, Spain; and †Oklahoma Medical Research Foundation, Oklahoma City, OK, USA
To cite this article: Tamayo I, Velasco SE, Puy C, Esmon CT, Dichiara MG, Montes R, Hermida J. Group V secretory phospholipase A2 impairs endothelial protein C receptor-dependent protein C activation and accelerates thrombosis in vivo. J Thromb Haemost 2014; 12: 1921–7. See also Rao LVM. Secretory group V phospholipase A2: a new player in thrombosis? This issue, pp 1918–20.
Summary. Background: Endothelial protein C receptor (EPCR) must be bound to a molecule of phosphatidylcholine (PC) to be fully functional, i.e. to interact with protein C/activated protein C (APC) properly. PC can be replaced with other lipids, such as lysophosphatidylcholine or platelet-activating factor, by the action of group V secretory phospholipase A2 (sPLA2-V), an enzyme that is upregulated in a variety of inflammatory conditions. Studies in purified systems have demonstrated that the substitution of PC notably impairs EPCR function in a process called EPCR encryption. Objectives: To analyze whether sPLA2-V was able to regulate EPCR-dependent protein C activation in vivo, and its impact on thrombosis and the hemostatic system. Methods: Mice were transfected with sPLA2-V by hydrodynamic gene delivery. The effects on thrombosis were studied with the laser carotid artery occlusion model, and APC generation capacity was measured with ELISA. Global hemostasis was analyzed with thromboelastometry. Results: We found that sPLA2V overexpression in mice significantly decreased their ability to generate APC. Furthermore, a murine carotid artery laser thrombosis model revealed that higher sPLA2-V levels were directly associated with faster artery thrombosis. Conclusions: sPLA2-V plays a thrombogenic
Correspondence: Jose Hermida, Laboratory of Thrombosis and Hemostasis, Division of Cardiovascular Sciences, Center for Applied Medical Research (CIMA), University of Navarra, Avenida Pıo XII 55, 31008 Pamplona, Spain. Tel.: +34 948194700; fax: +34 948194716. E-mail: [email protected] Received 12 February 2014 Manuscript handled by: W. Ruf Final decision: P. H. Reitsma, 6 July 2014 © 2014 International Society on Thrombosis and Haemostasis
role by impairing the ability of EPCR to promote protein C activation. Keywords: activated protein C resistance; endothelial cell protein C receptor; group V secretory phospholipase A2; protein C; thrombosis.
Introduction The protein C anticoagulant system is one of the cornerstones of the hemostatic system. Indeed, low levels of activated protein C (APC) constitute a risk factor for venous thromboembolism [1]. The efficiency of protein C activation relies on protein C’s ability to interact with endothelial protein C receptor (EPCR). After thrombin generation, binding between protein C and EPCR facilitates the activation of the former by the thrombin–thrombomodulin complex on the endothelial surface [2]. In fact, it has been reported that ~ 90% of APC generation in vivo depends on the presence of EPCR [3]. For EPCR to function, a phosphatidylcholine (PC) molecule must be present within its hydrophobic groove [4]. We have recently shown that when the PC molecule is replaced by either lysophosphatidylcholine (lysoPC) or platelet-activating factor (PAF), EPCR is encrypted to diminish its binding affinity for protein C [5]. Accordingly, we observed that when endothelial cells were incubated with group V secretory phospholipase A2 (sPLA2-V), which is able to generate lysoPC and PAF by using PC as the substrate, the ability of thrombin to generate APC was severely affected. In the present study, we sought to determine the in vivo significance of the abovedescribed in vitro observations. For this purpose, the effects of sPLA2-V overexpression and inhibition in mice on APC generation and thrombus formation were analyzed. We provide evidence that sPLA2-V plays a proco-
1922 I. Tamayo et al
agulant role by modulating the circulating levels of APC, which results in a shortening of the time required for thrombus formation after vessel injury. Materials and methods Plasmid cloning of the transcript encoding sPLA2-V
The cDNA corresponding to the functional transcript of the human sPLA2-V gene, PLA2G5 (NM_000929.2), which shares homology with mouse sPLA2-V, was amplified from mRNA of tumor necrosis factor-a-stimulated human aortic endothelial cells with the following primers: forward, 50 -CACCATGAAAGGCCTCCTCCCACTGG-30 ; and reverse, 50 -CTAGGAGCAGAGGATGTTGGG-30 . It was then cloned into the pcDNA3.2/V5/GW/D-TOPO plasmid (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions, to be under the control of the cytomegalovirus (CMV) promoter. The integrity of the sequence was confirmed by direct sequencing. The pCMVsPLA2-V (vector containing the sPLA2-V cDNA) and pCMV (empty vector for control purposes) plasmids were used to transform One Shot Top 10 Chemically Competent Escherichia coli cells (Invitrogen). DNA from the pCMV and pCMV-sPLA2-V plasmids was purified by use of the Qiagen EndoFree Megaprep plasmid purification kit (Qiagen, Crawley, UK), and the quantity and purity were measured in a nanodrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).
Quantification of thrombomodulin and EPCR gene expression was carried out with mRNA from vessels. Briefly, mice were killed and perfused with 10 mL of phosphate-buffered saline. Samples were extracted, snap frozen, and stored at 80 °C until RNA extraction with TriReagent (Sigma Chemical Co., Poole, UK) was performed. After retrotranscription, RT-PCR was performed on an ABI PRISM 7900 detector (Applied Biosystems, Carlsbad, CA, USA) with TaqMan gene expression assayson-demand for human sPLA2-V, murine EPCR, and murine thrombomodulin (Hs00173472_m1, Mm00440992_m1, and Mm00437014_s1, respectively; Applied Biosystems). Mouse b-actin (Mm00607939_s1; Applied Biosystems) was used to normalize results. Administration to mice of vesicles of phospholipids and factor Xa
The animal experiments were performed with ICR mice weighing 20–25 g (Harlan Interfauna Iberica S.A., Barcelona, Spain). All of the experiments were performed under a protocol approved by the Institutional Animal Care and Use Committee (001/12) at the University of Navarra.
To study the effect of sPLA2-V overexpression on in vivo APC generation ability, vesicles consisting of a mixture of PC and phosphatidylserine (PS) together with FXa were used. Their administration to mice induces the generation of measurable amounts of circulating APC. PCPS (molar proportion 80 : 20) vesicles were first prepared by sonication. Briefly, 2.6 lmol of total phospholipids was placed in a glass test tube and dried under a gentle stream of nitrogen. HEPES-buffered saline solution (2.6 mL) was then added at room temperature; after 1 h, the mixture vortexed until complete resuspension of the phospholipid layer was achieved. Finally, the tube was bath-sonicated in a Microson ultrasonicator (Misonix, New York, NY, USA) for 30 min. Forty-eight hours after hydrodynamic gene delivery, PCPS vesicles and FXa were injected through the tail vein in a final volume of 150 lL at 541.5 nmol kg1 PCPS and 352 pmol kg1 FXa. After 10 min, mice were killed by CO2 inhalation, and blood samples were collected in tubes containing sodium citrate.
Hydrodynamic sPLA2-V gene delivery to the liver
Measurement of mouse circulating APC levels
Overexpression of sPLA2-V was induced in ICR mice by hydrodynamic PLA2G5 gene delivery to the liver. Briefly, mice were weighed, and 100 lg of pCMV or pCMVsPLA2-V plasmid was given through the tail vein in saline solution (Braun Medical, Barcelona, Spain), in a volume equivalent to 10% of the weight of each mouse [6].
Mouse circulating APC levels were measured as previously described [7]. Briefly, ELISA plates were coated with antimurine protein C mAb 1587. After blocking with 1% bovine serum albumin, 1 : 2 diluted murine plasma samples were added in quadruplicate. Serial dilutions of recombinant mouse APC were used to prepare the standard curve. After extensive washing, 1 mM Spectrozyme PCA (American Diagnostica, Greenwich, CT, USA) was added. Twenty-four hours later, absorbance was measured at 405 nm. The standard curve was linear (R2 = 0.99) from 0.78 to 50 ng mL1. As this is a proteolytic assay, background generated by non-specific binding to the well of other proteases has to be subtracted. For this purpose, samples were first incubated in parallel with excess mAb 1587. Consequently, all APC in those in parallel samples would be bound to this mAb before its addition
Mice
Measurement of gene expression of sPLA2-V, thrombomodulin, and EPCR
To confirm PLA2G5 overexpression, RT-PCR was used to measure the mRNA levels 48 h after the hydrodynamic gene delivery procedure. Tail vein hydrodynamic gene delivery is primarily a hepatic gene delivery procedure, so mRNA for sPLA2-V quantification was extracted from liver samples.
© 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1923
to the well and eliminated in subsequent washings. Values corresponding to those wells were considered to be the background, which was subtracted from sample wells in order to obtain the APC concentration in plasma.
signed rank test, and the Friedman test; correlations were performed with the non-parametric Spearman rank test. P-values of < 0.05 were considered to be significant. GRAPHPAD PRISM (Version 5.0c; GraphPad Software, San Diego, CA, USA) was used throughout.
Rose Bengal/laser carotid artery thrombosis model
To study the effect of sPLA2-V overexpression on thrombosis, the Rose Bengal/laser carotid artery thrombosis model was used [8]. Forty-eight hours after the hydrodynamic gene delivery, mice were anesthetized with a mixture of 100 mg kg1 ketamine and 10 mg kg1 xylazine; 100 mg kg1 Rose Bengal (Sigma-Aldrich, St Louis, MO, USA) was then administered intravenously. The left carotid artery was exposed, a pulse Doppler flow probe (diameter, 0.5 mm; Transonic, Sidney, Australia) was applied, and the baseline blood flow was measured. The left carotid artery was subsequently illuminated with a 1.5-mW green light (540 nm) laser (Melles Griot, Carlsbad, CA, USA), and the time to thrombus formation was recorded. Thrombus was considered to be stable if the flow was not recovered after 30 min. In the first set of experiments, the time to total vessel occlusion was compared between sPLA2-Voverexpressing and control mice. A second set of experiments was performed to determine the effect of constitutive sPLA2-V expression on time to vessel occlusion. For this purpose, the comparison was performed between mice intraperitoneally given the sPLA2 inhibitor manoalide and mice given the vehicle. A third set of experiments was carried out to assess the EPCR dependency of the sPLA2-V overexpression effect on thrombosis. In this case, 50 lg of the EPCR-blocking mAb RCR-16 [8] was administered intravenously 1 h before application of the laser radiation. Phospholipase inhibitor administration
The phospholipase inhibitor manoalide (Santa Cruz, Santa Cruz, CA, USA) (1.42 mg kg1 in 1% dimethylsulfoxide in saline solution) or vehicle was intraperitoneally administered five times every 12 h before PCPS/FXa administration or Rose Bengal/laser carotid artery injury induction. Influence of sPLA2-V on hemostasis
Whole blood thromboelastometry was performed to analyze the influence of sPLA2-V on hemostasis. Blood was collected by cardiac puncture from four mice for each group in citrated buffer and pooled. Three hundred microliters of whole blood was used to analyze both intrinsic and extrinsic coagulation pathways and platelets with a ROTEM thromboelastometer (Pentapharm, Munich, Germany). Statistical analysis
Experimental variables were analyzed with the non-parametric Mann–Whitney U-test, the Wilcoxon matched-pairs © 2014 International Society on Thrombosis and Haemostasis
Results Transfection with pCMV-sPLA2-V induces high hepatic expression of sPLA2-V in mice
Twenty-four hours after the intravenous administration of 100 lg of the sPLA2-V-encoding plasmid (pCMVsPLA2-V), sPLA2-V mRNA levels were analyzed in liver. First, murine sPLA2-V mRNA was not detected in mouse livers, independently of whether or not tranfection with pCMV-sPLA2-V had been performed. As expected, human sPLA2-V was expressed at a high level (77.91 60.44 relative quantitation [RQ]) in livers from mice transfected with pCMV-sPLA2-V. Hepatic expression of sPLA2-V was also confirmed by immunohistochemistry (not shown). Finally, human sPLA2-V mRNA was undetectable in vascular tissue of transfected mice. Overexpression of sPLA2-V impaires the ability of mice to activate protein C
We next assessed whether sPLA2-V overexpression influenced the ability of mice to activate protein C. For this purpose, PCPS/FXa vesicles were intravenously administered to mice in order to induce the generation of measurable amounts of APC. Mice subjected to pCMV-sPLA2V transfection showed significantly decreased circulating APC levels upon induction (Fig. 1A) (60.33 15.39 ng mL1, mean standard deviation [SD], vs. 30.57 20.70 ng mL1 in pCMV-transfected and pCMV-sPLA2V-transfected mice, respectively, P = 0.021). To further analyze the association between sPLA2-V expression and the ability to generate APC, a separate set of pCMV-sPLA2-V-transfected mice was used: an inverse correlation was found between the mRNA transcript levels and the circulating APC levels (q = 0.786, P = 0.018) (Fig. 1B). Overexpression of sPLA2-V accelerates thrombus formation in mice
In order to investigate the relevance of sPLA2-V overexpression for thrombus formation, we used a Rose Bengal/ laser-induced carotid artery injury model 48 h after transfection. sPLA2-V overexpression significantly shortened the time to total vessel occlusion (68.84 32.68 min, mean SD, vs. 45.56 27.93 min in pCMV-transfected and pCMV-sPLA2-V transfected mice, respectively, P = 0.012) (Fig. 1C). Again, in the pCMV-sPLA2-V group, an inverse correlation was found between the
P = 0.0206
80 60 40 20 0 Control
D
sPLA2-V expression (RQ)
sPLA2-V
C 150
100 Spearman Rho = – 0.786 80
P = 0.018
60 40 20 0
Occlusion time (min)
B
P = 0.0123 100
50
0 20
40 50 60 APC (ng mL–1)
30
70
Control
sPLA2-V
E 80
250 Spearman Rho = – 0.733 P = 0.025
200 150 100 50 0
Oclussion time (min)
APC (ng mL–1)
A 100
sPLA2-V expression (RQ)
1924 I. Tamayo et al
P = 0.6953 60 40 20 0
0
20 40 60 Occlusion time (min)
80
Anti-EPCR moAb Anti-EPCR mAb+sPLA2-V
Fig. 1. Group V secretory phospholipase A2 (sPLA2-V) overexpression in mice reduces their ability to generate activated protein C (APC) and increases thrombogenicity. (A) Circulating APC levels after phosphatidylcholine–phosphatidylserine/factor Xa challenge were compared between sPLA2-V-overexpressing and control mice. (B) The correlation between RT-PCR-assessed sPLA2-V expression and plasma APC levels was determined in pCMV-sPLA2-V-transfected mice. (C) After induction of carotid artery injury with the Rose Bengal/laser method, the time to total carotid artery occlusion was compared between sPLA2-V-overexpressing and control mice. (D) The correlation between RT-PCRassessed sPLA2-V expression and total carotid artery occlusion time was determined in pCMV-sPLA2-V-transfected mice. (E) Carotid artery occlusion time in control and sPLA2-V-overexpressing mice after blocking of endothelial protein C receptor (EPCR) with 50 lg of RCR16. Bars represent median values. RQ, relative quantitation.
mRNA transcript levels in the liver and the occlusion time (q = 0.733, P = 0.025) (Fig. 1D). EPCR is needed for sPLA2-V to exert its prothrombotic effect
To confirm the involvement of EPCR in the prothrombotic effect of sPLA2-V, control and sPLA2-V-overexpressing mice were given the anti-EPCR mAb RCR-16, which blocks PC/APC binding to EPCR, and the time to carotid artery occlusion subsequent to Rose Bengal/laser stimulation was compared between both groups. sPLA2-V overexpression was unable to accelerate thrombus formation when EPCR was blocked by RCR-16 (27.64 19.38 min, mean SD, vs. 25.37 13.67 min in pCMV-transfected and pCMVsPLA2-V-transfected mice, respectively, P = 0.696) (Fig. 1E). In a separate set of mice, we measured APC generation, and confirmed that, in the presence of RCR-16, sPLA2-V overexpression did not significantly influence APC generation (45.55 22.50 ng mL1, mean SD, vs. 38.27 14.20 ng mL1 in pCMV-transfected and pCMVsPLA2-V-transfected mice, respectively, P = 0.5566). sPLA2-V does not influence EPCR and thrombomodulin expression
In order to confirm that the effect of sPLA2-V on APC generation and thrombosis was not attributable to a decrease in
the expression of EPCR and/or thrombomodulin, vascular mRNA levels of both proteins were analyzed with RTPCR. pCMV-transfected and pCMV-sPLA2-V-transfected mice showed similar levels of EPCR (1.00 0.43 RQ, mean SD, vs. 1.02 0.41 RQ, P = 0.937, n = 6) and thrombomodulin (1.00 0.22 RQ, mean SD, vs. 0.87 0.14 RQ, P = 0.393, n = 6). Other hemostatic pathways are not altered upon sPLA2-V overexpression
Thromboelastometry revealed that sPLA2-V overexpression did not alter other hemostatic pathways (Table 1). Thus, these results point to the sPLA2-V-induced decrease in protein C activation as the mechanism responsible for the accelerated thrombus formation.
In vivo inhibition of sPLA2-V confirms its thrombogenic role
In agreement with our previous in vitro experiments in which the sPLA2-V inhibitor manoalide improved protein C activation [5], we confirmed that the effect of manoalide on constitutive sPLA2-V levels resulted in significantly increased APC generation after challenge with PCPS/FXa vesicles (Fig. 2A) (77.3 41.8 ng mL1, mean SD, vs. 114.5 42.8 ng mL1 in vehicle-treated and manoalide-treated mice, respectively, n = 10, P = © 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1925 Table 1 Effect of overexpressing group V secretory phospholipase A2 (sPLA2-V) or inhibiting sPLA2-V expression on the hemostatic balance in mice Intrinsic coagulation pathway
pCMV pCMV-sPLA2-V
Extrinsic coagulation pathway
CT (s)
CFT (s)
MFC (mm)
CT (s)
CFT (s)
MFC (mm)
223 198
105 86
59 63
50 47
61 39
60 68
The clotting time (CT), clot formation time (CFT) and maximum clot firmness (MCF) of the intrinsic and extrinsic coagulation pathways were compared between a pool of plasma obtained from four pCMV-transfected mice and a pool of plasma obtained from four pCMV-sPLA2-Vtransfected mice. A representative experiment of two independent experiments is shown.
A
Discussion
200
APC (ng mL–1)
P = 0.0363 150
100
50
0 Vehicle
Occlusion time (min)
B
Manoalide
250 P = 0.1025
200 150 100 50 0 Vehicle
Manoalide
Fig. 2. In vivo inhibition of constitutive levels of group V secretory phospholipase A2 (sPLA2-V) increases the ability to generate activated protein C (APC). (A) Circulating APC levels after phosphatidylcholine–phosphatidylserine/factor Xa challenge were compared between sPLA2-V inhibitor-treated and control mice. (B) After induction of carotid artery injury with the Rose Bengal/laser method, the time to total carotid artery occlusion was compared between sPLA2-V inhibitor-treated and control mice. Bars represent median values.
0.036). Furthermore, when manoalide-treated mice were subjected to carotid artery-induced thrombosis with Rose Bengal/laser treatment (Fig. 2B), a slight, albeit non-significant, antithrombotic effect was seen (70.55 39.78 min, mean SD, vs. 87.10 48.31 min in vehicletreated and manoalide-treated mice, respectively, P = 0.102). © 2014 International Society on Thrombosis and Haemostasis
We demonstrate here, for the first time, that in vivo overexpression of sPLA2-V leads to impairment of APC generation and accelerates thrombosis. We recently showed that EPCR must have PC within its hydrophobic pocket to activate protein C efficiently. Indeed, the ligand-binding ability of EPCR was reduced by the substitution of PC with lysoPC or PAF. Accordingly, sPLA2-V, which generates these lipids, reduced the ability of endothelial cells to generate APC [4,5]. Now we have demonstrated that sPLA2-V overexpression in mice results in a proportional decrease in circulating APC levels upon PCPS/FXa challenge. This modulation of APC ability is rooted in the PC–APC–EPCR anticoagulant pathway, as was shown by the fact that differences in circulating APC levels resulting from sPLA2-V overexpression diminished after EPCR blockade. Thus, sPLA2-V is also able to reduce the ability to activate protein C in vivo. We ruled out the possibility that these observations were attributable to an effect of sPLA2-V on EPCR or thrombomodulin expression, as sPLA2-V does not induce the shedding of EPCR or thrombomodulin [5], and their in vivo expression was unaltered in sPLA2-V-overexpressing mice. Downregulation of APC generation owing to liver injury induced by abnormally high sPLA2-V activity could also constitute an alternative explanation for these findings. However, we observed that inhibition of endogenous sPLA2-V in non-manipulated mice induced a significant increase in circulating APC levels, thus supporting the notion that the effect of sPLA2-V on APC generation does not rely on liver injury. Additionally, mice overexpressing sPLA2-V showed normal hemostasis, which indicates that the liver synthesis of coagulation factors is preserved. All of these results suggest that impairment of EPCR function in vivo by sPLA2-V could be responsible for the reduction in APC generation. Monitoring of thrombus formation after carotid artery injury revealed that sPLA2-V overexpression significantly shortened the time needed for total vessel occlusion. It is of note that, as with the ability to generate protein C upon challenge, the magnitude of the effect was proportional to the success of the gene delivery, i.e. the higher the sPLA2-V overexpression, the shorter the time
1926 I. Tamayo et al
required for formation of the occlusive thrombus. The lack of human sPLA2-V expression in vascular tissues after hydrodynamic delivery (not shown) points towards hepatic overexpression and release of sPLA2-V as the actions causing the prothrombotic effect. Thus, we identified a link between the capacity to activate protein C and the susceptibility to thrombosis in these mice. The connection between APC and thrombosis has been widely reported in the literature: low APC levels are associated with venous thrombosis in humans [1], and EPCR blockade is prothrombotic both in a murine thrombosis model and in patients with anti-EPCR blocking antibodies [8,9]. Notably, a lack of a prothrombotic effect of sPLA2-V overexpression in EPCR-blocked mice was noted. This observation is consistent with the notion that, in vivo, EPCR can also be encrypted by sPLA2-V, thus decreasing the efficiency of protein C activation. Finally, there do not seem to be alternative mechanisms explaining the prothrombotic effect of sPLA2-V. Thromboelastometry is useful to examine the role played by the coagulation pathways, fibrinolysis and platelets in clot formation, and in these experiments the plasma from sPLA2-V-overexpressing mice did not behave differently from the plasma from the other mice. Another interesting feature of this study is that mice treated with manoalide showed an increased capacity to generate APC. This finding suggests that a fraction of EPCR is basally encrypted. This notion is consistent with the documented existence of constitutive baseline expression of sPLA2-V [10], and is in agreement with our previous in vitro observations of a gain of function of cell surface EPCR in manoalide-treated endothelial cells [5]. Accordingly, the in vivo inhibition of sPLA2-V resulted in a modest, but statistically insignificant, decrease in thrombus formation upon challenge. Thus, the possibility of an antithrombotic strategy aimed at inhibiting sPLA2-V activity in circumstances of abnormally high sPLA2-V levels should be explored in the future. Finally, future studies should address the effect of sPLA2-V modulation on other actions that EPCR is known to be involved in, i.e. APCtriggered cellular signaling leading to antiapoptotic, antiinflammatory and vascular barrier protective actions [11], or FVII/VIIa interactions with endothelial cells [11–16]. Finally, the lack of a method with which to effectively measure the plasma levels of sPLA2-V is a limitation of the study. There is currently no suitable assay to measure this, although several groups are working on its development. Nevertheless, the correlations observed between mRNA sPLA2-V levels and APC and thrombosis are strong enough to conclude that sPLA2-V plays a role in vivo in our model. To summarize, these experiments suggest that the sPLA2-V encryption of EPCR donwregulates protein C activation in vivo and thus contributes to the promotion of thrombus formation. Controlling sPLA2-V upregulation may help to ensure an adequate anticoagulant response.
Acknowledgements We acknowledge G. Gonz alez-Aseguinolaza, J. H. Morrissey and G. L. Ferrell for their assistance with gene delivery, phospholipid vesicles, and APC capture procedures, respectively. This work was supported through the Uni on Temporal de Empresas project CIMA and by grants from Instituto de Salud Carlos III (PI08/1349, PI11/1458, PI13/ 00072, PI10/01432, and RECAVA RD06/0014/0008), the Health Department of the Gobierno de Navarra (15/09), Ministerio de Economıa y Competitividad (InncorporaPTQ) (PTQ-11-04780 and PTQ-09-02-02217 to I. Tamayo and S. E. Velasco), and MCYT ERANET-NEURON ‘PROTEA’ PRI-PIMNEU-2011-1334. Addendum I. Tamayo, S. E. Velasco, R. Montes, and J. Hermida designed research. I. Tamayo, S. E. Velasco, C. Puy, and M. G. Dichiara performed research. C. T. Esmon contributed vital new reagents. I. Tamayo, S. E. Velasco, C. T. Esmon, R. Montes, and J. Hermida analyzed and interpreted data. I. Tamayo, R. Montes, and J. Hermida wrote the manuscript. Disclosure of Conflict of Interests R. Montes reports receiving grants from Instituto de Salud Carlos III and MCYT ERANET-NEURON ‘PROTEA’, during the conduct of the study. J. Hermida reports receiving grants from Instituto de Salud Carlos III and the Health Department of the Gobierno de Navarra, during the conduct of the study. S. E. Velasco reports receiving grants from Ministerio de Economıa y Competitividad (Inncorpora-PTQ), during the conduct of the study. M. G. Dichiara reports receiving grants from the Foundation for Applied Medical Research, during the conduct of the study. I. Tamayo reports receiving grants from Ministerio de Economıa y Competitividad (Inncorpora-PTQ), during the conduct of the study. C. T. Esmon and C. Puy state that they have no conflict of interest. References 1 Espa~ na F, Vaya A, Mira Y, Medina P, Estelles A, Villa P, Falc o C, Aznar J. Low level of circulating activated protein C is a risk factor for venous thromboembolism. Thromb Haemost 2001; 86: 1368–73. 2 Stearns-Kurosawa DJ, Kurosawa S, Mollica JS, Ferrell GL, Esmon CT. The endothelial cell protein C receptor augments protein C activation by the thrombin–thrombomodulin complex. Proc Natl Acad Sci USA 1996; 93: 10212–16. 3 Taylor FB. Endothelial cell protein C receptor plays an important role in protein C activation in vivo. Blood 2001; 97: 1685–8. 4 Oganesyan V, Oganesyan N, Terzyan S, Qu D, Dauter Z, Esmon NL, Esmon CT. The crystal structure of the endothelial protein C receptor and a bound phospholipid. J Biol Chem 2002; 277: 24851–4. © 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1927 5 L opez-Sagaseta J, Puy C, Tamayo I, Allende M, Cerver o J, Velasco SE, Esmon CT, Montes R, Hermida J. sPLA2-V inhibits EPCR anticoagulant and antiapoptotic properties by accommodating lysophosphatidylcholine or PAF in the hydrophobic groove. Blood 2012; 119: 2914–21. 6 Liu F, Song Y, Liu D. Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther 1999; 6: 1258–66. 7 Li W, Zheng X, Gu J, Hunter J, Ferrell GL, Lupu F, Esmon N, Esmon CT. Overexpressing endothelial cell protein C receptor alters the hemostatic balance and protects mice from endotoxin. J Thromb Haemost 2005; 3: 1351–9. 8 Centelles MN, Puy C, L opez-Sagaseta J, Fukudome K, Montes R, Hermida J. Blocking endothelial protein C receptor (EPCR) accelerates thrombus development in vivo. Thromb Haemost 2010; 103: 1239–44. 9 Hurtado V, Montes R, Gris J-C, Bertolaccini ML, Alonso A, Martınez-Gonz alez MA, Khamashta MA, Fukudome K, Lane DA, Hermida J. Autoantibodies against EPCR are found in antiphospholipid syndrome and are a risk factor for fetal death. Blood 2004; 104: 1369–74. 10 Sawada H, Murakami M, Enomoto A, Shimbara S, Kudo I. Regulation of type V phospholipase A2 expression and func-
© 2014 International Society on Thrombosis and Haemostasis
11 12
13
14
15
16
tion by proinflammatory stimuli. Eur J Biochem 1999; 263: 826–35. Montes R, Puy C, Molina E, Hermida J. Is EPCR a multi-ligand receptor? Pros and cons. Thromb Haemost 2012; 107: 815–26. Ghosh S, Pendurthi UR, Steinoe A, Esmon CT, Rao LVM. Endothelial cell protein C receptor acts as a cellular receptor for factor VIIa on endothelium. J Biol Chem 2007; 282: 11849–57. Preston RJS, Ajzner E, Razzari C, Karageorgi S, Dua S, Dahlb€ack B, Lane DA. Multifunctional specificity of the protein C/activated protein C Gla domain. J Biol Chem 2006; 281: 28850–7. Gleeson EM, O’Donnell JS, Preston RJS. The endothelial cell protein C receptor: cell surface conductor of cytoprotective coagulation factor signaling. Cell Mol Life Sci 2012; 69: 717–26. L opez-Sagaseta J, Montes R, Puy C, Dıez N, Fukudome K, Hermida J. Binding of factor VIIa to the endothelial cell protein C receptor reduces its coagulant activity. J Thromb Haemost 2007; 5: 1817–24. Puy C, L opez-Sagaseta J, Hermida J, Montes R. The endothelial cells downregulate the generation of factor VIIa through EPCR binding. Br J Haematol 2010; 149: 111–17.
Copyright of Journal of Thrombosis & Haemostasis is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
DOI: 10.1111/jth.12676
IN FOCUS
Group V secretory phospholipase A2 impairs endothelial protein C receptor-dependent protein C activation and accelerates thrombosis in vivo I . T A M A Y O , * S . E . V E L A S C O , * C . P U Y , * C . T . E S M O N , † M . G . D I C H I A R A , * R . M O N T E S * and J. HERMIDA* *Division of Cardiovascular Sciences, Laboratory of Thrombosis and Hemostasis, Center for Applied Medical Research, University of Navarra, Pamplona, Spain; and †Oklahoma Medical Research Foundation, Oklahoma City, OK, USA
To cite this article: Tamayo I, Velasco SE, Puy C, Esmon CT, Dichiara MG, Montes R, Hermida J. Group V secretory phospholipase A2 impairs endothelial protein C receptor-dependent protein C activation and accelerates thrombosis in vivo. J Thromb Haemost 2014; 12: 1921–7. See also Rao LVM. Secretory group V phospholipase A2: a new player in thrombosis? This issue, pp 1918–20.
Summary. Background: Endothelial protein C receptor (EPCR) must be bound to a molecule of phosphatidylcholine (PC) to be fully functional, i.e. to interact with protein C/activated protein C (APC) properly. PC can be replaced with other lipids, such as lysophosphatidylcholine or platelet-activating factor, by the action of group V secretory phospholipase A2 (sPLA2-V), an enzyme that is upregulated in a variety of inflammatory conditions. Studies in purified systems have demonstrated that the substitution of PC notably impairs EPCR function in a process called EPCR encryption. Objectives: To analyze whether sPLA2-V was able to regulate EPCR-dependent protein C activation in vivo, and its impact on thrombosis and the hemostatic system. Methods: Mice were transfected with sPLA2-V by hydrodynamic gene delivery. The effects on thrombosis were studied with the laser carotid artery occlusion model, and APC generation capacity was measured with ELISA. Global hemostasis was analyzed with thromboelastometry. Results: We found that sPLA2V overexpression in mice significantly decreased their ability to generate APC. Furthermore, a murine carotid artery laser thrombosis model revealed that higher sPLA2-V levels were directly associated with faster artery thrombosis. Conclusions: sPLA2-V plays a thrombogenic
Correspondence: Jose Hermida, Laboratory of Thrombosis and Hemostasis, Division of Cardiovascular Sciences, Center for Applied Medical Research (CIMA), University of Navarra, Avenida Pıo XII 55, 31008 Pamplona, Spain. Tel.: +34 948194700; fax: +34 948194716. E-mail: [email protected] Received 12 February 2014 Manuscript handled by: W. Ruf Final decision: P. H. Reitsma, 6 July 2014 © 2014 International Society on Thrombosis and Haemostasis
role by impairing the ability of EPCR to promote protein C activation. Keywords: activated protein C resistance; endothelial cell protein C receptor; group V secretory phospholipase A2; protein C; thrombosis.
Introduction The protein C anticoagulant system is one of the cornerstones of the hemostatic system. Indeed, low levels of activated protein C (APC) constitute a risk factor for venous thromboembolism [1]. The efficiency of protein C activation relies on protein C’s ability to interact with endothelial protein C receptor (EPCR). After thrombin generation, binding between protein C and EPCR facilitates the activation of the former by the thrombin–thrombomodulin complex on the endothelial surface [2]. In fact, it has been reported that ~ 90% of APC generation in vivo depends on the presence of EPCR [3]. For EPCR to function, a phosphatidylcholine (PC) molecule must be present within its hydrophobic groove [4]. We have recently shown that when the PC molecule is replaced by either lysophosphatidylcholine (lysoPC) or platelet-activating factor (PAF), EPCR is encrypted to diminish its binding affinity for protein C [5]. Accordingly, we observed that when endothelial cells were incubated with group V secretory phospholipase A2 (sPLA2-V), which is able to generate lysoPC and PAF by using PC as the substrate, the ability of thrombin to generate APC was severely affected. In the present study, we sought to determine the in vivo significance of the abovedescribed in vitro observations. For this purpose, the effects of sPLA2-V overexpression and inhibition in mice on APC generation and thrombus formation were analyzed. We provide evidence that sPLA2-V plays a proco-
1922 I. Tamayo et al
agulant role by modulating the circulating levels of APC, which results in a shortening of the time required for thrombus formation after vessel injury. Materials and methods Plasmid cloning of the transcript encoding sPLA2-V
The cDNA corresponding to the functional transcript of the human sPLA2-V gene, PLA2G5 (NM_000929.2), which shares homology with mouse sPLA2-V, was amplified from mRNA of tumor necrosis factor-a-stimulated human aortic endothelial cells with the following primers: forward, 50 -CACCATGAAAGGCCTCCTCCCACTGG-30 ; and reverse, 50 -CTAGGAGCAGAGGATGTTGGG-30 . It was then cloned into the pcDNA3.2/V5/GW/D-TOPO plasmid (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions, to be under the control of the cytomegalovirus (CMV) promoter. The integrity of the sequence was confirmed by direct sequencing. The pCMVsPLA2-V (vector containing the sPLA2-V cDNA) and pCMV (empty vector for control purposes) plasmids were used to transform One Shot Top 10 Chemically Competent Escherichia coli cells (Invitrogen). DNA from the pCMV and pCMV-sPLA2-V plasmids was purified by use of the Qiagen EndoFree Megaprep plasmid purification kit (Qiagen, Crawley, UK), and the quantity and purity were measured in a nanodrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).
Quantification of thrombomodulin and EPCR gene expression was carried out with mRNA from vessels. Briefly, mice were killed and perfused with 10 mL of phosphate-buffered saline. Samples were extracted, snap frozen, and stored at 80 °C until RNA extraction with TriReagent (Sigma Chemical Co., Poole, UK) was performed. After retrotranscription, RT-PCR was performed on an ABI PRISM 7900 detector (Applied Biosystems, Carlsbad, CA, USA) with TaqMan gene expression assayson-demand for human sPLA2-V, murine EPCR, and murine thrombomodulin (Hs00173472_m1, Mm00440992_m1, and Mm00437014_s1, respectively; Applied Biosystems). Mouse b-actin (Mm00607939_s1; Applied Biosystems) was used to normalize results. Administration to mice of vesicles of phospholipids and factor Xa
The animal experiments were performed with ICR mice weighing 20–25 g (Harlan Interfauna Iberica S.A., Barcelona, Spain). All of the experiments were performed under a protocol approved by the Institutional Animal Care and Use Committee (001/12) at the University of Navarra.
To study the effect of sPLA2-V overexpression on in vivo APC generation ability, vesicles consisting of a mixture of PC and phosphatidylserine (PS) together with FXa were used. Their administration to mice induces the generation of measurable amounts of circulating APC. PCPS (molar proportion 80 : 20) vesicles were first prepared by sonication. Briefly, 2.6 lmol of total phospholipids was placed in a glass test tube and dried under a gentle stream of nitrogen. HEPES-buffered saline solution (2.6 mL) was then added at room temperature; after 1 h, the mixture vortexed until complete resuspension of the phospholipid layer was achieved. Finally, the tube was bath-sonicated in a Microson ultrasonicator (Misonix, New York, NY, USA) for 30 min. Forty-eight hours after hydrodynamic gene delivery, PCPS vesicles and FXa were injected through the tail vein in a final volume of 150 lL at 541.5 nmol kg1 PCPS and 352 pmol kg1 FXa. After 10 min, mice were killed by CO2 inhalation, and blood samples were collected in tubes containing sodium citrate.
Hydrodynamic sPLA2-V gene delivery to the liver
Measurement of mouse circulating APC levels
Overexpression of sPLA2-V was induced in ICR mice by hydrodynamic PLA2G5 gene delivery to the liver. Briefly, mice were weighed, and 100 lg of pCMV or pCMVsPLA2-V plasmid was given through the tail vein in saline solution (Braun Medical, Barcelona, Spain), in a volume equivalent to 10% of the weight of each mouse [6].
Mouse circulating APC levels were measured as previously described [7]. Briefly, ELISA plates were coated with antimurine protein C mAb 1587. After blocking with 1% bovine serum albumin, 1 : 2 diluted murine plasma samples were added in quadruplicate. Serial dilutions of recombinant mouse APC were used to prepare the standard curve. After extensive washing, 1 mM Spectrozyme PCA (American Diagnostica, Greenwich, CT, USA) was added. Twenty-four hours later, absorbance was measured at 405 nm. The standard curve was linear (R2 = 0.99) from 0.78 to 50 ng mL1. As this is a proteolytic assay, background generated by non-specific binding to the well of other proteases has to be subtracted. For this purpose, samples were first incubated in parallel with excess mAb 1587. Consequently, all APC in those in parallel samples would be bound to this mAb before its addition
Mice
Measurement of gene expression of sPLA2-V, thrombomodulin, and EPCR
To confirm PLA2G5 overexpression, RT-PCR was used to measure the mRNA levels 48 h after the hydrodynamic gene delivery procedure. Tail vein hydrodynamic gene delivery is primarily a hepatic gene delivery procedure, so mRNA for sPLA2-V quantification was extracted from liver samples.
© 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1923
to the well and eliminated in subsequent washings. Values corresponding to those wells were considered to be the background, which was subtracted from sample wells in order to obtain the APC concentration in plasma.
signed rank test, and the Friedman test; correlations were performed with the non-parametric Spearman rank test. P-values of < 0.05 were considered to be significant. GRAPHPAD PRISM (Version 5.0c; GraphPad Software, San Diego, CA, USA) was used throughout.
Rose Bengal/laser carotid artery thrombosis model
To study the effect of sPLA2-V overexpression on thrombosis, the Rose Bengal/laser carotid artery thrombosis model was used [8]. Forty-eight hours after the hydrodynamic gene delivery, mice were anesthetized with a mixture of 100 mg kg1 ketamine and 10 mg kg1 xylazine; 100 mg kg1 Rose Bengal (Sigma-Aldrich, St Louis, MO, USA) was then administered intravenously. The left carotid artery was exposed, a pulse Doppler flow probe (diameter, 0.5 mm; Transonic, Sidney, Australia) was applied, and the baseline blood flow was measured. The left carotid artery was subsequently illuminated with a 1.5-mW green light (540 nm) laser (Melles Griot, Carlsbad, CA, USA), and the time to thrombus formation was recorded. Thrombus was considered to be stable if the flow was not recovered after 30 min. In the first set of experiments, the time to total vessel occlusion was compared between sPLA2-Voverexpressing and control mice. A second set of experiments was performed to determine the effect of constitutive sPLA2-V expression on time to vessel occlusion. For this purpose, the comparison was performed between mice intraperitoneally given the sPLA2 inhibitor manoalide and mice given the vehicle. A third set of experiments was carried out to assess the EPCR dependency of the sPLA2-V overexpression effect on thrombosis. In this case, 50 lg of the EPCR-blocking mAb RCR-16 [8] was administered intravenously 1 h before application of the laser radiation. Phospholipase inhibitor administration
The phospholipase inhibitor manoalide (Santa Cruz, Santa Cruz, CA, USA) (1.42 mg kg1 in 1% dimethylsulfoxide in saline solution) or vehicle was intraperitoneally administered five times every 12 h before PCPS/FXa administration or Rose Bengal/laser carotid artery injury induction. Influence of sPLA2-V on hemostasis
Whole blood thromboelastometry was performed to analyze the influence of sPLA2-V on hemostasis. Blood was collected by cardiac puncture from four mice for each group in citrated buffer and pooled. Three hundred microliters of whole blood was used to analyze both intrinsic and extrinsic coagulation pathways and platelets with a ROTEM thromboelastometer (Pentapharm, Munich, Germany). Statistical analysis
Experimental variables were analyzed with the non-parametric Mann–Whitney U-test, the Wilcoxon matched-pairs © 2014 International Society on Thrombosis and Haemostasis
Results Transfection with pCMV-sPLA2-V induces high hepatic expression of sPLA2-V in mice
Twenty-four hours after the intravenous administration of 100 lg of the sPLA2-V-encoding plasmid (pCMVsPLA2-V), sPLA2-V mRNA levels were analyzed in liver. First, murine sPLA2-V mRNA was not detected in mouse livers, independently of whether or not tranfection with pCMV-sPLA2-V had been performed. As expected, human sPLA2-V was expressed at a high level (77.91 60.44 relative quantitation [RQ]) in livers from mice transfected with pCMV-sPLA2-V. Hepatic expression of sPLA2-V was also confirmed by immunohistochemistry (not shown). Finally, human sPLA2-V mRNA was undetectable in vascular tissue of transfected mice. Overexpression of sPLA2-V impaires the ability of mice to activate protein C
We next assessed whether sPLA2-V overexpression influenced the ability of mice to activate protein C. For this purpose, PCPS/FXa vesicles were intravenously administered to mice in order to induce the generation of measurable amounts of APC. Mice subjected to pCMV-sPLA2V transfection showed significantly decreased circulating APC levels upon induction (Fig. 1A) (60.33 15.39 ng mL1, mean standard deviation [SD], vs. 30.57 20.70 ng mL1 in pCMV-transfected and pCMV-sPLA2V-transfected mice, respectively, P = 0.021). To further analyze the association between sPLA2-V expression and the ability to generate APC, a separate set of pCMV-sPLA2-V-transfected mice was used: an inverse correlation was found between the mRNA transcript levels and the circulating APC levels (q = 0.786, P = 0.018) (Fig. 1B). Overexpression of sPLA2-V accelerates thrombus formation in mice
In order to investigate the relevance of sPLA2-V overexpression for thrombus formation, we used a Rose Bengal/ laser-induced carotid artery injury model 48 h after transfection. sPLA2-V overexpression significantly shortened the time to total vessel occlusion (68.84 32.68 min, mean SD, vs. 45.56 27.93 min in pCMV-transfected and pCMV-sPLA2-V transfected mice, respectively, P = 0.012) (Fig. 1C). Again, in the pCMV-sPLA2-V group, an inverse correlation was found between the
P = 0.0206
80 60 40 20 0 Control
D
sPLA2-V expression (RQ)
sPLA2-V
C 150
100 Spearman Rho = – 0.786 80
P = 0.018
60 40 20 0
Occlusion time (min)
B
P = 0.0123 100
50
0 20
40 50 60 APC (ng mL–1)
30
70
Control
sPLA2-V
E 80
250 Spearman Rho = – 0.733 P = 0.025
200 150 100 50 0
Oclussion time (min)
APC (ng mL–1)
A 100
sPLA2-V expression (RQ)
1924 I. Tamayo et al
P = 0.6953 60 40 20 0
0
20 40 60 Occlusion time (min)
80
Anti-EPCR moAb Anti-EPCR mAb+sPLA2-V
Fig. 1. Group V secretory phospholipase A2 (sPLA2-V) overexpression in mice reduces their ability to generate activated protein C (APC) and increases thrombogenicity. (A) Circulating APC levels after phosphatidylcholine–phosphatidylserine/factor Xa challenge were compared between sPLA2-V-overexpressing and control mice. (B) The correlation between RT-PCR-assessed sPLA2-V expression and plasma APC levels was determined in pCMV-sPLA2-V-transfected mice. (C) After induction of carotid artery injury with the Rose Bengal/laser method, the time to total carotid artery occlusion was compared between sPLA2-V-overexpressing and control mice. (D) The correlation between RT-PCRassessed sPLA2-V expression and total carotid artery occlusion time was determined in pCMV-sPLA2-V-transfected mice. (E) Carotid artery occlusion time in control and sPLA2-V-overexpressing mice after blocking of endothelial protein C receptor (EPCR) with 50 lg of RCR16. Bars represent median values. RQ, relative quantitation.
mRNA transcript levels in the liver and the occlusion time (q = 0.733, P = 0.025) (Fig. 1D). EPCR is needed for sPLA2-V to exert its prothrombotic effect
To confirm the involvement of EPCR in the prothrombotic effect of sPLA2-V, control and sPLA2-V-overexpressing mice were given the anti-EPCR mAb RCR-16, which blocks PC/APC binding to EPCR, and the time to carotid artery occlusion subsequent to Rose Bengal/laser stimulation was compared between both groups. sPLA2-V overexpression was unable to accelerate thrombus formation when EPCR was blocked by RCR-16 (27.64 19.38 min, mean SD, vs. 25.37 13.67 min in pCMV-transfected and pCMVsPLA2-V-transfected mice, respectively, P = 0.696) (Fig. 1E). In a separate set of mice, we measured APC generation, and confirmed that, in the presence of RCR-16, sPLA2-V overexpression did not significantly influence APC generation (45.55 22.50 ng mL1, mean SD, vs. 38.27 14.20 ng mL1 in pCMV-transfected and pCMVsPLA2-V-transfected mice, respectively, P = 0.5566). sPLA2-V does not influence EPCR and thrombomodulin expression
In order to confirm that the effect of sPLA2-V on APC generation and thrombosis was not attributable to a decrease in
the expression of EPCR and/or thrombomodulin, vascular mRNA levels of both proteins were analyzed with RTPCR. pCMV-transfected and pCMV-sPLA2-V-transfected mice showed similar levels of EPCR (1.00 0.43 RQ, mean SD, vs. 1.02 0.41 RQ, P = 0.937, n = 6) and thrombomodulin (1.00 0.22 RQ, mean SD, vs. 0.87 0.14 RQ, P = 0.393, n = 6). Other hemostatic pathways are not altered upon sPLA2-V overexpression
Thromboelastometry revealed that sPLA2-V overexpression did not alter other hemostatic pathways (Table 1). Thus, these results point to the sPLA2-V-induced decrease in protein C activation as the mechanism responsible for the accelerated thrombus formation.
In vivo inhibition of sPLA2-V confirms its thrombogenic role
In agreement with our previous in vitro experiments in which the sPLA2-V inhibitor manoalide improved protein C activation [5], we confirmed that the effect of manoalide on constitutive sPLA2-V levels resulted in significantly increased APC generation after challenge with PCPS/FXa vesicles (Fig. 2A) (77.3 41.8 ng mL1, mean SD, vs. 114.5 42.8 ng mL1 in vehicle-treated and manoalide-treated mice, respectively, n = 10, P = © 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1925 Table 1 Effect of overexpressing group V secretory phospholipase A2 (sPLA2-V) or inhibiting sPLA2-V expression on the hemostatic balance in mice Intrinsic coagulation pathway
pCMV pCMV-sPLA2-V
Extrinsic coagulation pathway
CT (s)
CFT (s)
MFC (mm)
CT (s)
CFT (s)
MFC (mm)
223 198
105 86
59 63
50 47
61 39
60 68
The clotting time (CT), clot formation time (CFT) and maximum clot firmness (MCF) of the intrinsic and extrinsic coagulation pathways were compared between a pool of plasma obtained from four pCMV-transfected mice and a pool of plasma obtained from four pCMV-sPLA2-Vtransfected mice. A representative experiment of two independent experiments is shown.
A
Discussion
200
APC (ng mL–1)
P = 0.0363 150
100
50
0 Vehicle
Occlusion time (min)
B
Manoalide
250 P = 0.1025
200 150 100 50 0 Vehicle
Manoalide
Fig. 2. In vivo inhibition of constitutive levels of group V secretory phospholipase A2 (sPLA2-V) increases the ability to generate activated protein C (APC). (A) Circulating APC levels after phosphatidylcholine–phosphatidylserine/factor Xa challenge were compared between sPLA2-V inhibitor-treated and control mice. (B) After induction of carotid artery injury with the Rose Bengal/laser method, the time to total carotid artery occlusion was compared between sPLA2-V inhibitor-treated and control mice. Bars represent median values.
0.036). Furthermore, when manoalide-treated mice were subjected to carotid artery-induced thrombosis with Rose Bengal/laser treatment (Fig. 2B), a slight, albeit non-significant, antithrombotic effect was seen (70.55 39.78 min, mean SD, vs. 87.10 48.31 min in vehicletreated and manoalide-treated mice, respectively, P = 0.102). © 2014 International Society on Thrombosis and Haemostasis
We demonstrate here, for the first time, that in vivo overexpression of sPLA2-V leads to impairment of APC generation and accelerates thrombosis. We recently showed that EPCR must have PC within its hydrophobic pocket to activate protein C efficiently. Indeed, the ligand-binding ability of EPCR was reduced by the substitution of PC with lysoPC or PAF. Accordingly, sPLA2-V, which generates these lipids, reduced the ability of endothelial cells to generate APC [4,5]. Now we have demonstrated that sPLA2-V overexpression in mice results in a proportional decrease in circulating APC levels upon PCPS/FXa challenge. This modulation of APC ability is rooted in the PC–APC–EPCR anticoagulant pathway, as was shown by the fact that differences in circulating APC levels resulting from sPLA2-V overexpression diminished after EPCR blockade. Thus, sPLA2-V is also able to reduce the ability to activate protein C in vivo. We ruled out the possibility that these observations were attributable to an effect of sPLA2-V on EPCR or thrombomodulin expression, as sPLA2-V does not induce the shedding of EPCR or thrombomodulin [5], and their in vivo expression was unaltered in sPLA2-V-overexpressing mice. Downregulation of APC generation owing to liver injury induced by abnormally high sPLA2-V activity could also constitute an alternative explanation for these findings. However, we observed that inhibition of endogenous sPLA2-V in non-manipulated mice induced a significant increase in circulating APC levels, thus supporting the notion that the effect of sPLA2-V on APC generation does not rely on liver injury. Additionally, mice overexpressing sPLA2-V showed normal hemostasis, which indicates that the liver synthesis of coagulation factors is preserved. All of these results suggest that impairment of EPCR function in vivo by sPLA2-V could be responsible for the reduction in APC generation. Monitoring of thrombus formation after carotid artery injury revealed that sPLA2-V overexpression significantly shortened the time needed for total vessel occlusion. It is of note that, as with the ability to generate protein C upon challenge, the magnitude of the effect was proportional to the success of the gene delivery, i.e. the higher the sPLA2-V overexpression, the shorter the time
1926 I. Tamayo et al
required for formation of the occlusive thrombus. The lack of human sPLA2-V expression in vascular tissues after hydrodynamic delivery (not shown) points towards hepatic overexpression and release of sPLA2-V as the actions causing the prothrombotic effect. Thus, we identified a link between the capacity to activate protein C and the susceptibility to thrombosis in these mice. The connection between APC and thrombosis has been widely reported in the literature: low APC levels are associated with venous thrombosis in humans [1], and EPCR blockade is prothrombotic both in a murine thrombosis model and in patients with anti-EPCR blocking antibodies [8,9]. Notably, a lack of a prothrombotic effect of sPLA2-V overexpression in EPCR-blocked mice was noted. This observation is consistent with the notion that, in vivo, EPCR can also be encrypted by sPLA2-V, thus decreasing the efficiency of protein C activation. Finally, there do not seem to be alternative mechanisms explaining the prothrombotic effect of sPLA2-V. Thromboelastometry is useful to examine the role played by the coagulation pathways, fibrinolysis and platelets in clot formation, and in these experiments the plasma from sPLA2-V-overexpressing mice did not behave differently from the plasma from the other mice. Another interesting feature of this study is that mice treated with manoalide showed an increased capacity to generate APC. This finding suggests that a fraction of EPCR is basally encrypted. This notion is consistent with the documented existence of constitutive baseline expression of sPLA2-V [10], and is in agreement with our previous in vitro observations of a gain of function of cell surface EPCR in manoalide-treated endothelial cells [5]. Accordingly, the in vivo inhibition of sPLA2-V resulted in a modest, but statistically insignificant, decrease in thrombus formation upon challenge. Thus, the possibility of an antithrombotic strategy aimed at inhibiting sPLA2-V activity in circumstances of abnormally high sPLA2-V levels should be explored in the future. Finally, future studies should address the effect of sPLA2-V modulation on other actions that EPCR is known to be involved in, i.e. APCtriggered cellular signaling leading to antiapoptotic, antiinflammatory and vascular barrier protective actions [11], or FVII/VIIa interactions with endothelial cells [11–16]. Finally, the lack of a method with which to effectively measure the plasma levels of sPLA2-V is a limitation of the study. There is currently no suitable assay to measure this, although several groups are working on its development. Nevertheless, the correlations observed between mRNA sPLA2-V levels and APC and thrombosis are strong enough to conclude that sPLA2-V plays a role in vivo in our model. To summarize, these experiments suggest that the sPLA2-V encryption of EPCR donwregulates protein C activation in vivo and thus contributes to the promotion of thrombus formation. Controlling sPLA2-V upregulation may help to ensure an adequate anticoagulant response.
Acknowledgements We acknowledge G. Gonz alez-Aseguinolaza, J. H. Morrissey and G. L. Ferrell for their assistance with gene delivery, phospholipid vesicles, and APC capture procedures, respectively. This work was supported through the Uni on Temporal de Empresas project CIMA and by grants from Instituto de Salud Carlos III (PI08/1349, PI11/1458, PI13/ 00072, PI10/01432, and RECAVA RD06/0014/0008), the Health Department of the Gobierno de Navarra (15/09), Ministerio de Economıa y Competitividad (InncorporaPTQ) (PTQ-11-04780 and PTQ-09-02-02217 to I. Tamayo and S. E. Velasco), and MCYT ERANET-NEURON ‘PROTEA’ PRI-PIMNEU-2011-1334. Addendum I. Tamayo, S. E. Velasco, R. Montes, and J. Hermida designed research. I. Tamayo, S. E. Velasco, C. Puy, and M. G. Dichiara performed research. C. T. Esmon contributed vital new reagents. I. Tamayo, S. E. Velasco, C. T. Esmon, R. Montes, and J. Hermida analyzed and interpreted data. I. Tamayo, R. Montes, and J. Hermida wrote the manuscript. Disclosure of Conflict of Interests R. Montes reports receiving grants from Instituto de Salud Carlos III and MCYT ERANET-NEURON ‘PROTEA’, during the conduct of the study. J. Hermida reports receiving grants from Instituto de Salud Carlos III and the Health Department of the Gobierno de Navarra, during the conduct of the study. S. E. Velasco reports receiving grants from Ministerio de Economıa y Competitividad (Inncorpora-PTQ), during the conduct of the study. M. G. Dichiara reports receiving grants from the Foundation for Applied Medical Research, during the conduct of the study. I. Tamayo reports receiving grants from Ministerio de Economıa y Competitividad (Inncorpora-PTQ), during the conduct of the study. C. T. Esmon and C. Puy state that they have no conflict of interest. References 1 Espa~ na F, Vaya A, Mira Y, Medina P, Estelles A, Villa P, Falc o C, Aznar J. Low level of circulating activated protein C is a risk factor for venous thromboembolism. Thromb Haemost 2001; 86: 1368–73. 2 Stearns-Kurosawa DJ, Kurosawa S, Mollica JS, Ferrell GL, Esmon CT. The endothelial cell protein C receptor augments protein C activation by the thrombin–thrombomodulin complex. Proc Natl Acad Sci USA 1996; 93: 10212–16. 3 Taylor FB. Endothelial cell protein C receptor plays an important role in protein C activation in vivo. Blood 2001; 97: 1685–8. 4 Oganesyan V, Oganesyan N, Terzyan S, Qu D, Dauter Z, Esmon NL, Esmon CT. The crystal structure of the endothelial protein C receptor and a bound phospholipid. J Biol Chem 2002; 277: 24851–4. © 2014 International Society on Thrombosis and Haemostasis
EPCR encryption by sPLA2-V in vivo 1927 5 L opez-Sagaseta J, Puy C, Tamayo I, Allende M, Cerver o J, Velasco SE, Esmon CT, Montes R, Hermida J. sPLA2-V inhibits EPCR anticoagulant and antiapoptotic properties by accommodating lysophosphatidylcholine or PAF in the hydrophobic groove. Blood 2012; 119: 2914–21. 6 Liu F, Song Y, Liu D. Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther 1999; 6: 1258–66. 7 Li W, Zheng X, Gu J, Hunter J, Ferrell GL, Lupu F, Esmon N, Esmon CT. Overexpressing endothelial cell protein C receptor alters the hemostatic balance and protects mice from endotoxin. J Thromb Haemost 2005; 3: 1351–9. 8 Centelles MN, Puy C, L opez-Sagaseta J, Fukudome K, Montes R, Hermida J. Blocking endothelial protein C receptor (EPCR) accelerates thrombus development in vivo. Thromb Haemost 2010; 103: 1239–44. 9 Hurtado V, Montes R, Gris J-C, Bertolaccini ML, Alonso A, Martınez-Gonz alez MA, Khamashta MA, Fukudome K, Lane DA, Hermida J. Autoantibodies against EPCR are found in antiphospholipid syndrome and are a risk factor for fetal death. Blood 2004; 104: 1369–74. 10 Sawada H, Murakami M, Enomoto A, Shimbara S, Kudo I. Regulation of type V phospholipase A2 expression and func-
© 2014 International Society on Thrombosis and Haemostasis
11 12
13
14
15
16
tion by proinflammatory stimuli. Eur J Biochem 1999; 263: 826–35. Montes R, Puy C, Molina E, Hermida J. Is EPCR a multi-ligand receptor? Pros and cons. Thromb Haemost 2012; 107: 815–26. Ghosh S, Pendurthi UR, Steinoe A, Esmon CT, Rao LVM. Endothelial cell protein C receptor acts as a cellular receptor for factor VIIa on endothelium. J Biol Chem 2007; 282: 11849–57. Preston RJS, Ajzner E, Razzari C, Karageorgi S, Dua S, Dahlb€ack B, Lane DA. Multifunctional specificity of the protein C/activated protein C Gla domain. J Biol Chem 2006; 281: 28850–7. Gleeson EM, O’Donnell JS, Preston RJS. The endothelial cell protein C receptor: cell surface conductor of cytoprotective coagulation factor signaling. Cell Mol Life Sci 2012; 69: 717–26. L opez-Sagaseta J, Montes R, Puy C, Dıez N, Fukudome K, Hermida J. Binding of factor VIIa to the endothelial cell protein C receptor reduces its coagulant activity. J Thromb Haemost 2007; 5: 1817–24. Puy C, L opez-Sagaseta J, Hermida J, Montes R. The endothelial cells downregulate the generation of factor VIIa through EPCR binding. Br J Haematol 2010; 149: 111–17.
Copyright of Journal of Thrombosis & Haemostasis is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
