Targeting melanoma growth and metastasis with systemic delivery of liposome-incorporated protease-activated receptor-1 small interfering RNA

Abstract

The thrombin receptor [protease-activated receptor-1 (PAR-1)] is overexpressed in highly metastatic melanoma cell lines and in patients with metastatic lesions. Activation of PAR-1 leads to cell signaling and up-regulation of genes involved in adhesion, invasion, and angiogenesis. Herein, we stably silence PAR-1 through the use of lentiviral short hairpin RNA and found significant decreases in both tumor growth (P < 0.01) and metastasis (P < 0.001) of highly metastatic melanoma cell lines in vivo. The use of viruses for therapy is not ideal as it can induce toxic immune responses and possible gene alterations following viral integration. Therefore, we also used systemic delivery of PAR-1 small interfering RNA (siRNA) incorporated into neutral liposomes [1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC)] to decrease melanoma growth and metastasis in vivo. Significant decreases in tumor growth, weight, and metastatic lung colonies (P < 0.001 for all) were found in mice treated with PAR-1 siRNA-DOPC. The in vivo effects of PAR-1 on invasion and angiogenesis were analyzed via immunohistochemistry. Concomitant decreases in vascular endothelial growth factor, interleukin-8, and matrix metalloproteinase-2 expression levels, as well as decreased blood vessel density (CD31), were found in tumor samples from PAR-1 siRNA-treated mice, suggesting that PAR-1 is a regulator of melanoma cell growth and metastasis by affecting angiogenic and invasive factors. We propose that siRNA incorporated into DOPC nanoparticles could be delivered systemically and used as a new modality for melanoma treatment.

Figure 1. Silencing of PAR-1 by siRNA and shRNA

A) Western blot depicting silencing of PAR-1 in A375SM transfected with PAR-1 siRNA compared to parental and non-targeting control. α-actin is used as loading control. Approximately 80% PAR-1 downregulation was observed in PAR-1 siRNA transfected cells compared to NT siRNA as determined by densitometry (0.2 and 1, respectively). B) Semi-quantitative RT-PCR depicting a decrease of PAR-1 (by ∼90%) in A375SM. C) A375SM and D) C8161 cells were transduced with lentivirus containing either NT shRNA or PAR-1 shRNA. The lentiviral vectors have GFP expression and therefore transduced cells will have green fluorescence. Cells were incubated with a PAR-1 antibody and a secondary PE-tagged antibody (red fluorescence). Top panels depict cells transduced with NT shRNA showing GFP positive/PE positive (presence of PAR-1). Bottom panels depict cells transduced with PAR-1 shRNA showing GFP positive/PE negative (decreased PAR-1 expression)

Figure 2. Effects of PAR-1 shRNA on melanoma growth and metastasis

A) In vivo tumor growth was determined on the PAR-1-positive highly metastatic melanoma cell lines, A375SM and (B) C8161 after stable silencing of PAR-1 with shRNA. Mice (n=10/group) were measured with a caliper in two dimensions for 6 weeks. Silencing of PAR-1 resulted in significant inhibition of tumor growth in both A375SM (PAR-1 shRNA- 41.1 mm³ and NT shRNA 534.3 mm³) and C8161 melanoma cells (PAR-1 shRNA- 159.7 mm³ and NT shRNA 813.5 mm³). * P<0.01 for both A375SM and C8161. C) A375SM melanoma cells stably transduced with NT lentiviral shRNA or PAR-1 shRNA were injected intravenously into nude mice (n=10/group). The mice were sacrificed after 6 weeks, the lungs were removed, fixed in Bouin's solution and the number of lung colonies was counted. Each symbol represents one mouse. There is a significant decrease in the number of lung metastases in PAR-1 shRNA transduced A375SM cells as compared to NT shRNA transduced cells (median 1 and 37, respectively. D) Decreased metastases also occurred in PAR-1 shRNA transduced C8161 cells compared to NT shRNA transduced cells (median 6 and 70 respectively). ** P<0.001 for both A375SM and C8161.

Figure 3. Localization of siRNA-DOPC particles in subcutaneous tumors after single i.v. injection

A) NT siRNA tagged with Alexa-555 (red) incorporated into DOPC liposomes were administered at Day 0 to melanoma-bearing nude mice (5mm³) (n=5/day). Mice were sacrificed on Days 2, 4, and 6 after single intravenous. administration. The siRNA was found within the tumor cells at all time-points. Even at Day 6 after intravenous injection, there were siRNA (red) particles within the subcutaneous tumor (blue Hoechst nuclear staining) in a perinuclear orientation almost imperceptible to scavenging macrophages (green). B) In vivo silencing of PAR-1 after single intravenous injection of PAR-1 siRNA-DOPC as assessed by real-time PCR. Data depict optimal decrease of PAR-1 expression in subcutaneous melanoma tumors 4 days after single intravenous injection of PAR-1 siRNA-DOPC (10μg) compared to NT siRNA-DOPC treatment. All tumors (n=5/group) were run in triplicates and averages obtained after normalization with 18s rRNA. Significant downregulation of PAR-1 was observed 4 days but not 6 days after injection.

Figure 4. Effects of PAR-1 silencing after twice weekly systemic administration of PAR-1 siRNA-DOPC

A) Nude mice (n=10/group) were subcutaneously injected with 5×10⁵ A375SM cells. After tumors grew to a size of 3-5mm³, mice were then intravenously treated with 10μg of liposomal incorporated siRNA twice per week for 4 weeks. There is a significant decrease in tumor growth in mice treated with PAR-1 siRNA-DOPC compared to mice treated with NT siRNA-DOPC (749.9 mm³ compared to176.2 mm³, respectively). ** P<0.001 B) Individual subcutaneous tumor weights were measured after 4 weeks of twice weekly intravenous treatment of 10μg of liposomal incorporated siRNA. There is significant inhibition of tumor weight between NT siRNA-treated mice (median 614 mg) and PAR-1 siRNA-treated mice (median 107 mg). ** P<0.001. C) Representative slides from immunohistochemistry analysis for PAR-1 in tumors from mice treated with liposomal PAR-1 and NT siRNA. PAR-1 silencing was observed at the end of treatment by Day 27. As a negative control, the tumors were incubated without primary antibody. D) Real-time PCR for PAR-1 expression of individual tumors from PAR-1 siRNA-DOPC-treated mice compared to the average expression of NT siRNA-DOPC-treated mice (n=10). All tumors were run in triplicates and averages obtained after normalization with 18s rRNA. Striped bar depicts the average expression of PAR-1 from PAR-1 siRNA-DOPC-treated tumors and reveals a significant reduction in expression in vivo by Day 27. Four of the tumors from the PAR-1 siRNA-DOPC-treated mice were too small to obtain sufficient RNA and were not included for this analysis.

Figure 5. Inhibition of lung metastases by melanoma cells after twice weekly systemic administration of PAR-1 siRNA-DOPC

Nude mice were intravenously injected with 1×10⁶ A375SM. Mice were subsequently treated intravenously with 10μg of either liposomal incorporated PAR-1 siRNA or NT siRNA twice weekly for 5 weeks. Lungs were harvested, fixed in Bouin's solution and individual tumor nodules counted. There is a significant decrease in the number of lung colonies in PAR-1-siRNA treated mice (median 12) compared to those treated with NT siRNA (median 62). One mouse died in the NT group two weeks after treatment began. ** P<0.001.

Figure 6. Effects of PAR-1 silencing on angiogenic and invasive factors

IHC analysis was performed on tumors from systemically treated mice with PAR-1 siRNA-DOPC or NT siRNA-DOPC. Representative images demonstrate that silencing of PAR-1 affects angiogenic (IL-8, VEGF) and invasive (MMP-2) factors. There is also a reduction in the number of blood vessels (CD31; microvessel density) in tumors treated with PAR-1 siRNA. As a negative control, the tumor samples were incubated without primary antibody. All images are at 10X magnification.