uPA/uPAR downregulation inhibits radiation-induced migration, invasion and angiogenesis in IOMM-Lee meningioma cells and decreases tumor growth in vivo

Abstract

Meningioma is a well-known tumor of the central nervous system, and is treated by surgical resection and/or radiation. Recently, ionizing radiation has been shown to enhance invasiveness of surviving tumor cells, and several proteolytic enzyme molecules, including urokinase plasminogen activator (uPA), seem to be upregulated after radiation. uPA and its receptor (uPAR) have been strongly implicated in tumor invasion, angiogenesis and progression. Hence, the tumor-associated uPA-uPAR system is considered a potential target for cancer therapy. In the present study, we show that radiation increases uPA levels in the IOMM-Lee meningioma cells, and subsequently, increases tumor invasion, migration and angiogenesis in vitro. Studies with signaling molecule inhibitors AG1478, U0126 and SB203580 (specific inhibitors of EGFR, MEK1/2 and p38 respectively) showed inhibition of uPA levels in both basal and irradiated-IOMM-Lee cells. The PI3K inhibitor (LY294002) and the AKT inhibitor (AKT inhibitor IV) also partially decreased uPA expression, whereas SP600125, a JNK inhibitor, did not affect uPA levels in either radiated or non-radiated cells. Further, a bicistronic plasmid construct with small interfering RNA (siRNA) against uPA and its receptor inhibited tumor invasion, migration and angiogenesis in radiation-treated IOMM-Lee cells. In addition, siRNA against uPA and its receptor inhibited subcutaneous tumor growth in athymic nude mice in combination with radiation in a synergistic manner. Thus, the specific targeting of proteases via RNA interference could augment the therapeutic effect of radiation and prevent the adverse effects resulting from tumor cells that receive sublethal doses of radiation within the tumor mass.

Figure 1. Radiation increases uPA and other signaling pathway molecules expression in IOMM-Lee cells

(A) Fibrinogen zymography using the conditioned medium from IOMM-Lee cell cultures radiated with 5 or 10Gy and controls for the indicated time points. Equal amounts of total protein (0.5μg) were resolved for each sample. Mean densitometric values ± SEM were calculated and plotted as a histogram. *Statistically different compared to respective control and radiated groups (p<0.05). (B) RT-PCR analysis for uPA expression. Total RNA was extracted from IOMM-Lee cells incubated in serum-free medium for 6h after cells were radiated (with 10Gy) or not irradiated (mock). GAPDH was used as a loading control. Mean densitometric values ± SEM were calculated and plotted as a histogram. *Statistically different compared to control and radiated groups (p<0.05). (C) EGFR, through ERK1/2 and p38 and to a lesser extent, pI3K/AKT, mediates radiation-induced uPA in IOMM-Lee cells. Fibrinogen zymography of conditioned medium, and western blot analysis of cell lysates from IOMM-Lee cells. Cells were treated with specific inhibitors of JNK (SP600125) (15μM), ERK1/2 (U0126) (10μM), p38 (SB203580) (10μM), AKT (AKT inhibitor IV) (10μM), pI3K (LY294002) (20μM) and EGFR (AG1478) (10μM) for 30 min before they were radiated with 10Gy and subsequently incubated for 6h in serum-free medium. Conditioned medium was collected and 0.5μg of total protein was used for fibrinogen zymography to detect uPA activity. Equal amounts of total proteins from cell extracts were resolved by SDS-PAGE and probed with antibodies against phosphorylated JNK, ERK1/2, p38, AKT, EGFR and total JNK1/2, ERK1/2, p38, AKT and EGFR. GAPDH was used as a loading control. Experiments were repeated at least three times.

Figure 2. puPA and pu2 siRNA inhibit radiation-induced uPA and uPAR

(A) Cells were transfected with mock, pSV, puPA or pu2 and 18h later, cells were radiated with 10Gy and incubated in serum-free medium for 6h. Conditioned medium (0.5μg) was used for fibrinogen zymography and equal amounts of total protein from cell extracts were resolved on SDS-PAGE and probed with anti-uPAR primary antibody. GAPDH was used as a loading control. (B) MTT proliferation assay. Cells were transfected with the plasmids and irradiated as described above. Six hours later, cells were trypsinized and 2×10⁴ cells were seeded on 96-well plates (eight wells per treatment). Cell proliferation was measured at 24, 48 and 72h as described in Materials and Methods. The bars represent the means ± SE of three different experiments. *Statistically different compared to control and puPA or pu2 treated groups or IR+control and IR + puPA or IR + pu2 groups (p<0.05).

Figure 3. Inhibition of radiation-induced migration from tumor spheroids and invasion through matrigel

(A) After IOMM-Lee cells formed single spheroids, they were transfected with mock, pSV, puPA or pu2 and transferred to 8-well chamber slides and irradiated with 10Gy. Spheroids were allowed to grow in serum-free medium for 24h and were photographed under a light microscope. (B) Matrigel invasion assay of IOMM-Lee cells transfected for 24h and then radiated with 10Gy. After 6h, 2.5×10⁵ cells were plated on matrigel-coated transwell inserts. Invasion was allowed for 24h, and then inserts were stained with HEMA stain. Cells that had migrated through the matrigel were counted and photographed under a light microscope. (C) Migration from the spheroids was measured using Image-Pro Discovery software. Average values of three separate experiments are shown and bars represent the means ± SE. *Statistically different compared to control and puPA or pu2 treated groups or IR+control and IR + puPA or IR + pu2 groups (p<0.01). (D) Quantification of invading cells. The bars represent the means ± SE of three different experiments. *Statistically different compared to control and puPA or pu2 treated groups or IR+control and IR + puPA or IR + pu2 groups (p<0.01).

Figure 4. Downregulation of uPA and uPAR decreases radiation-induced tumor angiogenesis

(A) Conditioned medium from IOMM-Lee cells, which were transfected with mock, pSV, puPA or pu2 and subsequently radiated with 10Gy, was added to human microvascular endothelial cells (HMEC-1) pre-seeded on matrigel-coated 96-well plates. After overnight incubation, HMEC-1 were observed for capillary-like network formation and photographed under a light microscope. (B) Angiogenic effect was measured by counting the relative branch-points. Bars represent the means ± SE of three different experiments. *Statistically different compared to control and puPA or pu2 treated groups or IR+control and IR + puPA or IR + pu2 groups (p<0.01).

Figure 5. puPA and pu2 along with radiation suppress subcutaneous tumor growth in athymic nude mice

(A) IOMM-Lee cells (5×10⁶) were injected subcutaneously into athymic nude mice. After tumors reached 4-5 mm in diameter, mice were treated intratumorally with mock or with 4 doses (60μg/dose) of pSV, puPA or pu2. Between the injections, animals were irradiated twice with 5Gy as described in Materials and Methods. Three weeks after treatments were completed, animals were sacrificed and tumors removed, measured and photographed. (B) Tumor volume was measured as described in Materials and Methods. Tumor volumes are shown as symbols and the average values of all tumor volume is shown as horizontal solid line (5 animals per treatment group) (p<0.01).

Figure 6. puPA or pu2 downregulate radiation-induced uPA, uPAR and CD31 in tumor xenografts

Hematoxylin and eosin staining of sections from IOMM-Lee xenograft tumors (top) and immunohistochemical analyses for uPA (second from top), uPAR (second from bottom) and CD31 (bottom) using specific antibodies for uPA, uPAR and CD31.