Alternatively spliced RAGEv1 inhibits tumorigenesis through suppression of JNK signaling

Abstract

Receptor for advanced glycation end products (RAGE) and its ligands are overexpressed in multiple cancers. RAGE has been implicated in tumorigenesis and metastasis, but little is known of the mechanisms involved. In this study, we define a specific functional role for an alternate splice variant termed RAGE splice variant 1 (RAGEv1), which encodes a soluble endogenous form of the receptor that inhibits tumorigenesis. RAGEv1 was downregulated in lung, prostate, and brain tumors relative to control matched tissues. Overexpressing RAGEv1 in tumor cells altered RAGE ligand stimulation of several novel classes of genes that are critical in tumorigenesis and metastasis. Additionally, RAGEv1 inhibited tumor formation, cell invasion, and angiogenesis induced by RAGE ligand signaling. Analysis of signal transduction pathways underlying these effects revealed marked suppression of c-jun-NH(2)-kinase (JNK) pathway signaling, and JNK inhibition suppressed signaling through the RAGE pathway. Tumors expressing RAGEv1 were significantly smaller than wild-type tumors and displayed prominently reduced activation of JNK. Our results identify RAGEv1 as a novel suppressor, the study of which may offer new cancer therapeutic directions.

Figure 1

RAGEv1 is secreted by tumor cells and binds RAGE-ligand. (A) Western blot analysis was performed to assess the secretion of RAGEv1 from C6 glioma cells stably transfected with empty vector control (mock) or RAGEv1. Total cell lysate or conditioned media were analyzed for RAGEv1 expression with antibodies to human RAGE or the specific RAGEv1 C-terminus epitope. GAPDH was used as a loading control for total cell lysate. (B) Binding of RAGEv1 to RAGE-ligands was assessed by incubation of GST-tag or GST-tagged s100A12 with conditioned media from mock and RAGEv1 transfected cells. Pulldown assays were performed to assess RAGEv1 binding and western blot analysis performed with antibodies against RAGEv1 on the elutant from pulldown reactions.

Figure 2

RAGEv1 inhibits RAGE-ligand induced changes in tumorigenic gene expression. (A-B) Relative gene expression of PDGFB and TNF were determined by QPCR on mock and RAGEv1 expressing cells, stimulated with RAGE-ligand (s100B), gene expression normalized to GAPDH levels. (C-D) Western blot analysis of cyclin D1 and Apaf1 performed on cell lysates from mock and RAGEv1 cells, stimulated with RAGE-ligand (s100B), normalized to GAPDH levels. Data are means ± SEM from three independent experiments and significant differences (P ≤ 0.05) indicated by an asterisk.

Figure 3

RAGEv1 inhibits mechanisms of tumorigenesis in vitro. (A) RAGEv1 blocks RAGE-ligand induced angiogenesis. Matrigel tube formation assay of ECs exposed to conditioned media from mock and RAGEv1 transfected tumor cells exposed to vehicle or RAGE-ligand (s100B). Tube number was counted in three random fields from three independent experiments. (B) RAGEv1 expression increases cellular adhesion. Serum starved cells were incubated with vehicle or RAGE-ligand (s100B) for 24 h, seeded into plates, cultivated for 2 h and bound cells fixed, stained with crystal violet and quantified by absorbance at 570nm. (C) RAGEv1 expression increases cellular apoptosis. Serum starved cells were incubated with vehicle or RAGE-ligand (s100B) for 24 h and apoptosis assessed by staining for annexin V-FITC (x-axis) and /PI (y-axis) and flow cytometry. (D) RAGEv1 blocks RAGE-ligand induced tumor cell invasion. Matrigel invasion assays were performed with mock and RAGEv1 transfected cells in response to vehicle or RAGE-ligand (s100B). Data are means ± SEM from three independent experiments. Significant differences (P ≤ 0.05) between groups are indicated by an asterisk.

Figure 4

RAGEv1 inhibits MAPK signaling (A-D) Cells (mock and RAGEv1) were stimulated with RAGE-ligand for the indicated times, lysed and subjected to western blot with antibodies for phospho-status and total of SAPK/JNK, MEK 1/2, p38 and AKT for the indicated times. Activity levels are expressed as a ratio of phospho/total levels. Data are means ± s.e.m from three independent experiments. Significant differences (P ≤ 0.05) between groups are indicated by an asterisk.

Figure 5

RAGEv1 inhibits tumorigenesis through blocking a JNK-dependent signaling mechanism. (A-C) Cells were incubated with control (DMSO) or inhibitors for SAPK/JNK (SP600125), MEK1/2 (U0126), p38 (SB203580) and AKT (Triciribine) for 1h before stimulation with RAGE-ligand (s100B). RNA was then extracted and QPCR performed for APAF1, PDGFB and TNF, normalized to GAPDH levels. (D) JNK inhibition blocks RAGE-ligand induced tumor cell invasion. Matrigel invasion assays were performed with mock transfected cells, treated with control (DMSO) or SAPK/JNK (SP600125) and in response to vehicle or RAGE-ligand (s100B). Data are means ± s.e.m from three independent experiments. Significant differences (P ≤ 0.05) between groups are indicated by an asterisk.

Figure 6

RAGEv1 inhibits tumorigenesis in vitro and in vivo. (E) RAGEv1 inhibits tumorigenic growth in vitro. Soft agar assay were performed with mock and RAGEv1 transfected cells cultured for 14 days and stained with crystal violet. (A) Athymic nude mice were injected with either mock or RAGEv1 expressing tumor cells and tumor growth measured after 31 days. (B) Protein was extracted ex vivo from mock (Ma-c) and RAGEv1 (RAGEv1a-c) tumors and subjected to western blot for phospho and total levels of SAPK/JNK. Significant differences (P ≤ 0.05) between groups are indicated by an asterisk.