Expression of a urokinase-type plasminogen activator during tumor growth leads to angiogenesis via galanin activation in tumor-bearing mice

Abstract

Small-cell lung carcinoma releases progalanin. The released progalanin is activated via a nonclassical processing pathway, being processed into an active form of galanin (1-20) by plasmin in extracellular components. Plasmin is produced from plasminogen activators. To clarify the regulation of progalanin via plasminogen activation by urokinase and tissue-plasminogen activator (t-PA), we investigated the regulation mechanism for urokinase and t-PA expression and their effect on galanin activation. Additionally, we studied the effect of activated galanin on angiogenesis. To determine the effect of cell density, we measured the expression levels of urokinase and t-PA using real-time PCR and plasminogen/gelatin zymography in a cell culture. The urokinase expression increased under both high cell density and presence of cell membrane fractions. However, urokinase increments induced by conditioned medium were low. These results indicate that expression of plasminogen activators is regulated by cell membrane factors. We used tumor-bearing mice to clarify the expression of plasminogen activators and galanin activation. Real-time PCR showed that urokinase was substantially higher in the central parts of tumors compared to the periphery, and this was confirmed by plasminogen/gelatin zymography. To evaluate the biological effect of plasminogen activators on tumor growth, we used tranexamic acid as a plasminogen inhibitor. Tranexamic acid decreased galanin (1-20) and the hemoglobin content of tumors and suppressed tumor growth. Additionally, galanin had no effect on the hemoglobin content of tumors derived from cells lacking GALR2. These results demonstrate the regulation of urokinase expression in tumors through progalanin activation in extracellular compartments, and confirm that galanin plays a role in angiogenesis.

Keywords: adhesion molecule; extracellular processing; galanin; neuropeptide; plasminogen activator; tumor growth.

Figure 1

Induction of plasminogen activators in SBC‐3A cells. (A) SBC‐3A cells were seeded in a culture dish at a density of 10⁴, 10⁵, or 10⁶ cells/10 cm². (B) The effects of cell seed density on plasminogen activator mRNA expression were determined by real‐time PCR analysis. Cells seeded at several densities were cultured for 24 h, and RNA was isolated. Expression of u‐PA and t‐PA mRNA was measured by real‐time PCR analysis. β‐Actin mRNA was used as a reference gene. N = 3–4, *P < 0.05 vs. 10⁴ cells. (C) Plasminogen activator activity was measured using fluorescent substrate (Pyr‐Gly‐Arg‐MCA). The substrate was digested with culture medium, after which the digest was measured by fluorescence (355 nm/460 nm). N = 3–4, *P < 0.05 vs. 10⁴ cells. (D) Plasminogen activator activity and molecular forms were determined by plasminogen/gelatin zymography. Samples of 50 μg per lane were loaded for electrophoresis, using 10% acrylamide gel containing 1% gelatin and 50 μg·mL⁻¹ plasminogen. t‐PA and u‐PA were detected at ~70 kDa and ~50 kDa, respectively. The effect of cell membrane fraction (E) and conditioned media (F) on plasminogen activators mRNA expression was determined by real‐time PCR analysis. N = 3–4, *P < 0.05.

Figure 2

Expression of plasminogen activators in tumor tissue. (A) Hemoglobin content in the central and peripheral tumor regions. Hemoglobin content was measured with the cyanmethemoglobin method and found to be related to angiogenesis in tumor regions. N = 4, *P < 0.05. (B) HIF‐1α expression was detected by western blotting analysis under hypoxic conditions. (C) Expression of t‐PA and u‐PA mRNA in the peripheral and central tumor regions was measured by real‐time PCR. N = 4, *: P < 0.05. (D) The expression of plasminogen activators was determined by plasminogen/gelatin zymography.

Figure 3

Involvement of plasminogen activators in galanin production and tumor growth. (A) Tumors generated in tumor‐bearing mice. SBC‐3A cells were injected into the subcutaneous layer. The tumor was grown to a diameter of 7–10 mm. To suppress plasmin activity, the mice were administered tranexamic acid (TA). The molecular forms of galanin‐like immunoreactivity in the tumor (B) were treated with tranexamic acid (C) using gel filtration. The gel was carried by Sephadex G‐50 fine (1.0 cm × 80 cm) and 1 m acetic acid as an eluent. The column was calibrated with bovine serum albumin (Vo), lysozyme (14 kDa), human galanin (3 kDa), and dibutyryl cAMP (Vt). (D) Tumor growth was measured by tumor weight at 10 days after implanting SBC‐3A cells into the subcutaneous layer. The mice were treated with TA (30 mg·day⁻¹, intraperitoneal administration) to suppress plasmin activity. N = 4, *P < 0.05. (E and F) Hemoglobin content was measured in tumors derived from SBC‐3A or SBC‐3A‐Y cells, using the cyanmethemoglobin method. Hemoglobin decreased with TA treatment, but recovered under galanin administration (1 μg·day⁻¹, direct injection into tumor). N = 4, *P < 0.05.