JG6, a novel marine-derived oligosaccharide, suppresses breast cancer metastasis via binding to cofilin

Abstract

Cofilin, an actin-binding protein which disassembles actin filaments, plays an important role in invasion and metastasis. Here, we discover that JG6, an oligomannurarate sulfate, binds to cofilin, suppresses the migration of human breast cancer cells and cancer metastasis in breast cancer xenograft model. Mechanistically, JG6 occupies actin-binding sites of cofilin, thereby disrupting cofilin modulated actin turnover. Our results highlight the significance of cofilin in cancer and suggest JG6, a cofilin inhibitor, to treat metastatic cancer.

Fig1. JG6 binds to cofilin and inhibits its actin-severing activity

(A) The chemical structure of JG6. (B) Cancer cells were seeded and 100 μg/ml JG6-FITC was added. After incubated with JG6-FITC for the indicated time, cells were harvested and washed three times with PBS, and analyzed by FCM. Data were analyzed with CellQuest software and shown as means±S.E. of three independent experiments. (C) Cancer cells were seeded and 100 μg/ml JG6-FITC was added. Before and after the addition of JG6-FITC, cells were analyzed and photographed under a fluorescence microscope. Five parallel samples were prepared in each group and results are representative of three separate experiments. (D) Binding of JG6 and cofilin or pSer3-cofilin were co-immunoprecipitated with JG6 antibody, followed by immunoblotting using cofilin or pSer3-cofilin antibody. IgG co-immunoprecipitation is used as a negative control. (E) The JG6 binding rate was calculated by measuring the intensity of co-IP bands normalized with the intensity of input bands, which was quantified by ImageJ software. Data are means±S.E. of three independent experiments. n.s., not significant. (F) 4 μM pyrene-actin was polymerized for 12 h at 20 °C. Depolymerization was initiated by a 12-fold dilution with 2 μM cofilin and 50 or 100 μg/ml JG6. The decline in fluorescence was monitored immediately.

Fig2. JG6 suppresses the depolymerization/severing activities of cofilin on F-actin

(A&B) MDA-MB-435 and MTLn3 were incubated with JG6 (0, 50, 100 and 200 μg/ml) for 24 hours. Cell extracts were extracted for F-actin (F)/G-actin (G) fractionation assay as described [11]. The supernatant and pellet were separated, followed by immunoblotting using anti-actin antibody. Band intensity was quantified by ImageJ software. Data are shown as means±S.E. of three independent experiments. (C) GBA-MTLn3 cells were incubated with JG6 100 μg/ml for 24 hours and live cell images were photographed under a confocal microscope. (D) MTLn3 cells were incubated with JG6 100 μg/ml for 24 hours and then stimulated with 10% FBS or not, binding assay of actin and cofilin was analyzed by immunoprecipitation using anti-cofilin antibody and then subjected to immunoblotting analysis for actin. (E) MDA-MB-435 and MTLn3 cells were pretreated with the indicated concentration of JG6 for 24 hours and lysated for immunoblotting analysis with indicated antibodies.

Fig3. The binding mode of JG6 to cofilin

(A) The binding pattern between JG6 with cofilin. Computer molecular simulation was applied with a JG6 trimer. The colored stick structure reflects the oligosaccharide and the colored line and band around it indicate the conformation and secondary structure of cofilin. The potential interaction AA sites on cofilin were marked. (B) Binding between JG6 and cofilin WT or mutants. (C) The binding curves of JG6 with cofilin WT and K44A/D79A were determined using surface plasmon resonance. (D) 4 μM pyrene-actin was polymerized for 12 h at 20 °C. Depolymerization was initiated by a 12-fold dilution with 2 μM WT or mutant cofilin. (E) MTLn3 cells transfected with cofilin siRNA followed with cofilin WT or mutation plasmids were incubated with JG6 100 μg/ml for 24 hours. F-actin (F)/G-actin (G) fractionation and immunoblotting analysis were similar to Fig2 A&B

Fig4. JG6 suppresses cofilin-mediated cell migration

(A) JG6 inhibits serum-induced migration of breast cancer MTLn3 cells and MDA-MB-435 cells. The inhibitory effect of JG6 (50, 100 and 200 μg/ml) on cell migration was assessed using a Transwell migration assay as described in Materials and Methods. Representative pictures of three independent experiments with similar results were shown (magnification, 40×). (B) Cells that migrated into the lower chamber in (A) were counted. The data shown were the mean±S.E. of three independent experiments assuming no inhibition rate in the absence of JG6.(C) MTLn3 cells were transfected with scramble or cofilin siRNA for the indicated time, then subjected to Western blot analysis. (D&F) MTLn3 cells transfected with scramble siRNA or cofilin siRNA(D) and the mutant-cofilins or empty vector (Mock) (F)for 48 h. The inhibitory effect of JG6 (100 μg/ml) on migration was analyzed using Transwell migration assay, and representative images were shown. (E&G) The inhibition rate percentages of JG6 were determined, E for D and G for F, based on the cell numbers that migrated into the lower chamber. The mean±S.E. of three independent experiments are shown. ***, p < 0.001.

Fig 5. JG6 inhibits breast cancer metastasis

(A) Effect of JG6 on lung metastasis of MDA-MB-435 breast carcinoma orthotopic xenografts in nude mice. Top, representative photograph of metastatic nodules on lungs with H&E staining (magnification, 200×). (B) The histogram shows the inhibitory action of JG6 on the number of pulmonary metastatic nodules. Columns, mean of a typical experiment; bars, SE.