TR47, A PAR1-BASED PEPTIDE, INHIBITS MELANOMA CELL MIGRATION IN VITRO AND METASTASIS IN VIVO

Abstract

Activated Protein C (APC) is a serine-protease that displays antithrombotic and anti-inflammatory properties. In addition, cleavage of protease-activated receptor 1 (PAR1) by APC exerts endothelial cytoprotective actions. The effects of APC on endothelial cells may be reproduced by TR47, a PAR1-based peptide that mimics the novel N-terminus of PAR1 generated upon cleavage at Arg-46 by APC. In this study we demonstrate that wild-type APC and its signaling-proficient mutant, APC-2Cys (which has dramatically reduced anticoagulant activity), display similar inhibitory effects towards the transendothelial migration of A375 human melanoma cells. Consistent with this observation, APC and APC-2Cys significantly reduced the in vivo metastatic potential of the B16F10 murine melanoma cells. TR47 recapitulated the in vitro and in vivo protective profiles of APC and APC-2Cys. Treatment of EA.hy926 endothelial cells with TR47 (20 μM) significantly decreased the A375 cell migration. In addition, treatment of C57/BL6 mice with a single TR47 dose (125 μg/animal) strongly reduced the metastatic burden of B16F10 cells. Together, our results suggest that protection of the endothelial barrier by APC/TR47-mediated signaling pathways might be a valuable therapeutic approach to prevent metastasis.

1. Introduction

PARs (Protease-Activated Receptors) comprise a family of four distinct G-protein coupled receptors, which share a unique mechanism of activation that relies on the proteolytic cleavage of their N-terminal ectodomains by specific proteases, including blood coagulation factors [1]. Cleavage of PARs unmasks a new N-terminus, which serves as a tethered ligand that binds to the second extracellular domain of the receptor, resulting in a variety of cellular responses that exert important functions in hemostasis, thrombosis and vascular biology [2]. In addition to their roles in physiological processes, PARs contribute to numerous pathological processes, including tumor biology. In this context, several studies have described the contribution of PARs, in particular PAR1 and PAR2, in distinct aspects of tumor progression such as primary growth, angiogenesis, evasion from apoptosis and metastasis[2, 3].

PAR1 was firstly described as the thrombin receptor on the surface of human platelets, being responsible for rapid platelet aggregation at sub-nanomolar concentrations of thrombin [4]. Subsequent studies showed that thrombin also modulates PAR1-dependent pro-inflammatory pathways in other cell types [5]. Furthermore, thrombin enhances the expression of various cytokines (IL-1, IL-6, IL-8 and TNF-α) and adhesion molecules (E-selectin, P-selectin and VCAM-1) in cultured endothelial cells [6,7]. In addition, endothelial cell activation by thrombin causes dissociation of cell-cell junctions as well as cytoskeleton contraction, leading to endothelial barrier disruption [1]. Paradoxically, another serine-protease named activated protein C (APC) mediates anti-inflammatory and barrier-protective responses in endothelial cells upon activation of PAR1 [8]. Studies suggest that cleavage of PAR1 by thrombin at R41 position, in contrast to cleavage at R46 position by APC, generates distinct tethered ligands that promote the paradoxical effects in cultured endothelial cells[9 ].

In accordance with the opposite effects of thrombin and APC upon the endothelial barrier, contrasting effects of these enzymes have been reported concerning the tumor cell migration across endothelial cells. Treatment of cultured endothelial cells with APC decreases transendothelial migration of either murine or human melanoma cells [10]. In this study, we demonstrate that wild type APC, as well as a mutant form of APC that lacks its anticoagulant properties, but retains pro-signaling effects, decreases in vitro tumor cell migration across an endothelial cell barrier as well as in vivo metastasis. Remarkably, a soluble peptide that mimics the new PAR1 N-terminus generated upon cleavage at R46 by APC, TR47, decreases A375 migration across endothelial cell and reduces B16F10 metastatic potential. Accordingly, our data demonstrate that the PAR1 agonism evoked by APC or by TR47, a peptide mimicking the cleavage at R46, might be regarded as a potential tool to impair tumor cell extravasation during hematogenous metastasis.

2. Material and Methods

2.1. Proteins and agonist peptides

Recombinant human wild-type activated protein C (APC WT) and the non-anticoagulant mutant (APC-2Cys) were expressed and purified as previously described [11, 12]. Agonist peptides corresponding to the N-terminus of PAR1 cleaved by thrombin at R41 position (TR41, SFLLRNPN-NH₂) or cleaved by APC at R46 position (TR47, NPNDKYEP-NH₂) were synthesized by Pepmic Co. LTD, China. A synthetic peptide containing a scrambled sequence was also used in some experiments.

2.2. Cell culture

The human melanoma cell line, A375, and the murine melanoma cell line, B16F10, were purchased from ATCC (Manassas, Virginia, USA). The human umbilical vein endothelial cell line, EA.hy926, was kindly provided by Dr. C. Edgell (University of North Carolina, NC, USA). Cell lines were cultured in DMEM (Dulbecco’s modified Eagle medium; Life Technologies, SP, Brazil) supplemented with 10% fetal bovine serum (FBS) and incubated in a humidified atmosphere at 37°C in 5% CO₂.

2.3. Transendothelial migration assay

Migration assays were performed in transwell plates of 6.5 mm diameter, with 8 μm pore size filters (Corning, MA, USA). EA.hy926 endothelial cells (1 × 10⁵) were grown for 24 h to obtain confluent monolayers followed by treatment with PAR1 agonist peptides or APC constructs for 4 hours. TR47, TR41 and scrambled peptide were used at final concentration of 20 μM; APC WT and APC-2Cys were used at final concentration of 20 nM. The inserts were washed twice with PBS. A375 cells (2 × 10⁵) were resuspended in serum-free media and added to the upper compartment. FBS (10%) was added as a chemoattractant in the lower chamber. After 16 h, membranes were washed with PBS. The upper side of the membrane was gently wiped with a cotton swab and fixed with methanol for 20 minutes. The membrane was then stained with 0.2% crystal violet (Sigma-Aldrich, MO, USA). Each experiment was repeated in duplicate wells and cell counting was performed in five randomly selected microscopic high-power fields.

2.4. Experimental metastasis assay

Eight-week-old C57BL/6 mice were injected via the lateral tail vein with a 100 μL-bolus of 2.5 × 10⁵ B16F10 cells (resuspended in DMEM). The anti-metastatic effects of APC WT, APC-2Cys, TR41 or TR47, were evaluated by a single intraperitoneal injection of APC constructs (5 μg/animal) or PAR1-based peptides (125 μg/animal) one hour before the inoculation of melanoma cells. The control animals were injected with PBS. After 15 days, the animals were sacrificed, the lungs were removed, and the number of pulmonary tumor nodules only on the convex face of the organ was counted. All experiments involving animals were performed according to the guidelines for Care and Use of Laboratory Animals approved by the ethical committee of the Federal University of Rio de Janeiro.

2.5. Platelet aggregation

Human washed platelets were prepared as described previously [13]. Briefly, platelet aggregation was performed in a 490-2D platelet aggregometer (Chrono-log Corporation, CA, USA). Washed platelets were added to the cuvettes at a final concentration of 2 × 10⁸ / mL, together with 11 mM CaCl₂. Platelet aggregation curves were monitored for 6 min after the addition of 20 μM TR-41 or TR-47. The experiment was performed at 37 °C under agitation.

2.6. Analysis of ERK 1/2 phosphorylation by western blot

EA.hy926 cells were incubated in serum-free medium for 16 h followed by stimulation with TR47 or TR41 (20 μM) for a period of 15 to 120 minutes. After treatment, cells were lysed and 50 μg of proteins from each sample were run on 8% SDS–PAGE and transferred to a PVDF Hybond-P membrane (GE Healthcare, Brazil). Membranes were incubated with antibodies against phospho-p44/42 MAPK (Erk1/2) (#9101, Cell Signaling Technology, MA, USA) and β-actin (#4967, Cell Signaling Technology, MA, USA). After incubation with secondary antibodies, immunoblots were detected using the ECL reagent (GE Healthcare, Brazil) and bands were quantified using ImageJ program (NIH, MD, USA).

2.7. Statistical analysis

Comparison between groups was made using the Student’s unpaired t-test or one-way analysis of variance (ANOVA). All calculations were performed using the GraphPad Prism 5.0 software(GraphPad Software Inc., CA, USA).

3. Results

3.1. APC inhibits melanoma cell migration and metastasis

To evaluate the invasive properties of A375 cell line, we employed a transmigration assay across an endothelial cell monolayer [10]. Treatment of EA.hy926 endothelial cells with wild type APC significantly reduced A375 cell migration, as shown in figure 1A. This effect was identical to that observed with APC-2Cys, a signaling proficient mutant that retains the ability to cleave PAR1 while lacks its anticoagulant properties [12]. Consistent with this observation, both wild type APC and APC-2Cys decreased the metastatic potential of B16F10 melanoma cells in vivo (Fig. 1B). Thus, we observed lower pulmonary metastatic foci in mice injected with a single dose of wild type APC or APC-2Cys, 1 hour before the intravenous inoculation of melanoma cells (Fig. 1C).

Figure 1. APC constructs impair transmigration and metastasis of melanoma cells.

(A) Migration of A375 cells through a monolayer of endothelial cells (EA.hy926) was determined by transwell assay after 16 h. EA.hy926 cells were treated with APC WT (20 nM) or APC-2Cys (20 nM) for 4 hours prior to addition of melanoma cells to the upper compartment. Data are reported as means ± SD. *** P < 0.001. (B) B16F10 cells (2.5 × 10⁵) were injected intravenously into C57BL/6 mice after previous intraperitoneal injection of APC WT or APC-2Cys (5 μg in 50 μL PBS/animal). Control animals received PBS instead of APC constructs. After 15 days, the animals were sacrificed and the metastatic nodules on harvested lungs were counted. Lines represent the median of the number of lung metastasis observed in the animals from each group. *** P < 0.001. (C) Representative pictures of lungs of each group are shown.

3.2. The PAR1-based peptides, TR41 and TR47, display distinct biological activities

Most of the PAR1-depentent effects of thrombin can be reproduced by peptides based on the sequence of the new PAR1 N-terminus revealed by cleavage at Arg41 [14], such as TR41. Similarly, APC effects on endothelial cells are reproduced by a PAR1-based peptide that mimics the new N-terminal generated upon the cleavage at Arg46 [9], such as TR47.

PAR1 is the primary mediator of thrombin-stimulated platelet activation [4]. As expected, TR41, but not TR47, induced platelet aggregation (Fig. 2A). In addition to the effect on platelets, we investigated the effect of PAR1 agonist peptides in the signaling pathways of endothelial cells. Figure 2B shows that TR41 and TR47 stimulated ERK phosphorylation in EA.hy926 endothelial cells. However, TR41 showed higher potency than TR47. Moreover, TR41 induced potent ERK-1/2 phosphorylation after 15 min of stimulation, whereas the maximum effect of TR47 was observed 60 min after endothelial cell stimulation. The differential effects of TR41 and TR47 in the signaling pathways of endothelial cells may explain the differences observed in the transmigration and experimental metastasis assays.

Figure 2. TR41 and TR47 display distinct biological activities in platelets and in endothelial cells.

(A) Treatment of platelets with TR41 induces platelet aggregation, as opposed to TR47. Washed-platelet aggregation was measured for 6 minutes after addition of TR41 or TR47 (20 μM). Conditions were performed with 11 mM CaCl₂, under sample agitation at 37°C. (B) Endothelial cells, EA.hy926, were starved in serum-free medium for 16 h followed by stimulation with TR47 or TR41 (20 μM) for a period of 15 to 120 minutes. After treatments, cells were lysed and the levels of p-ERK 1/2 were determined by western blotting. β-actin was used as a loading control.

3.3. TR47 inhibits tumor cell migration and metastasis

We further evaluated the ability of TR47 in modulating the transmigration of A375 cells through the endothelial cell monolayer. Figure 3A shows that treatment of EA.hy926 cells with TR47 reduced the ability of A375 melanoma cells to migrate toward the chemotactic stimuli. On the other hand, TR41 showed no effect and the number of migrated melanoma cells was similar to that observed upon treatment with the scrambled peptide. Next, we investigated the ability of TR47 to prevent B16F10 metastasis in vivo. As seen in Figure 3B, treatment with TR47 significantly decreased the number of pulmonary metastasis, as compared to treatment with TR41 or with the control group. Representative images of lungs from untreated, TR41-treated or TR47-treated mice are presented in figure 3C.

Figure 3. TR47, unlike TR41, impairs transmigration and metastasis of melanoma cells.

(A) Migration of A375 cells against a monolayer of endothelial cells (EA.hy926) was determined by transwell assay after 16 h. EA.hy926 cells were treated with TR47 or TR41 (20 μM) for 4 hours prior to addition of melanoma cells to the upper compartment. Data are reported as means ± SD. * P < 0.05 and ** P < 0.01. (B) B16F10 cells (2.5 × 10⁵) were injected intravenously into C57BL/6 mice after previous intraperitoneal injection of a scrambled peptide, TR41 or TR47 (125 μg in 50 μL PBS/animal). Controls received PBS instead of peptides. After 15 days, the animals were sacrificed and the metastatic nodules on harvested lungs were counted. Lines represent the median of the number of lung metastasis observed in the animals from each group. * P < 0.05 and ** P < 0.01. (C) Representative pictures of lungs of each group are shown.

4. Discussion

Metastasis is defined as the ability of cancer cells to disseminate from primary tumors to form new tumor colonies in distant tissues [15]. This multi-step process is initiated by the local invasion of primary tumor cells into the surrounding tissues followed by its intravasation into the circulatory or lymphatic system. Tumor cells may travel across the circulatory or lymphatic system, arrest and extravasate through the vascular walls into the parenchyma of distant tissues. The ability of tumor cells to migrate through the endothelial cell barrier is a crucial event in this process[16].

It is known that APC enhances vascular barrier function, a process that involves downregulation of NF-kB signaling pathway and inhibition of expression of cell adhesion molecules [6]. In accordance with these properties, treatment with recombinant APC decreases the metastatic potential of B16F10 melanoma cells in vivo[17 ]. The anti-metastatic properties of APC have been directly correlated with its signaling properties. Blocking of APC-mediated signaling but not APC-mediated anticoagulant properties with monoclonal antibodies enhance melanoma metastasis [18]. In addition, an agonist of the sphingosine-1-phosphate receptor-1, which acts downstream to APC-PAR1, recapitulated the protective effect of APC [18]. Consistent with these studies, we observed that treatment with APC or its signaling-proficient mutant, APC-2Cys, significantly reduced the in vitro migration and the in vivo metastatic potential of B16F10 melanoma cells (Fig. 1). Interestingly, although coagulation inhibition is highly effective in preventing experimental metastasis [19,20], the loss of APC anticoagulant activity did not affect its antimetastatic effect.

Bezuhly and colleagues showed that transgenic mice overexpressing the endothelial receptor of APC, EPCR, exhibited significant reduction in liver and lung metastases in the B16F10 melanoma model [17]. On the other hand, overexpression of EPCR in tumor cells has been associated with increased tumor aggressiveness [21,22]. Therefore, it is proposed that APC may elicit pro-or anti-tumor responses depending on the presence of EPCR in neoplastic cells or in endothelial cells. In addition to the potential pro-tumor effects of APC, the incidence of hemorrhage-related adverse events has been pointed as an important limitation for the clinical use of APC in cancer [23]. Thus, there is interest in developing APC variants or peptides with no anticoagulant but normal cytoprotective activity.

It has been proposed that the cytoprotective effects of APC derives from cleavage of PAR1 at Arg46 position [9]. In this context, peptides based on this new N-terminus sequence of PAR1 display endothelial cell protective properties in vitro and in vivo [9, 24]. TR47 protects endothelial cells against staurosporin toxicity and also prevents thrombin-induced permeability in vitro [25]. In addition, TR47 reduces VEGF-induced vascular leakage in vivo. In line with these reported biological effects, we demonstrate here that an 8-amino acid long PAR1-based peptide, TR47, mimics the APC antitumor effects in vitro and in vivo (Fig. 3). It is important to note that the experimental metastasis model employed in our study, i.e., the intravenous injection of tumor cells, does not recapitulate all the steps involved in the spontaneous metastatic process, although it is widely used in the literature [26].

Taken together, PAR1 biased agonism evoked by PAR1-based peptides might be regarded as a potential tool to modulate tumor cell-mediated changes in the endothelial barrier, a process that is crucial for metastasis.

Highlights.

• APCWT and APC-2Cysdecrease transendothelial migration of A375 cells.

• APC WT and APC-2Cys have protective effect against B16F10 experimental metastasis.

• The PAR1-based peptide, TR47, inhibits transendothelial migration of A375 cells.

• TR47 decreases the in vivo metastatic potential of B16F10 cells.