RAGE, carboxylated glycans and S100A8/A9 play essential roles in colitis-associated carcinogenesis

Abstract

Patients with inflammatory bowel diseases are at increased risk for colorectal cancer, but the molecular mechanisms linking inflammation and cancer are not well defined. We earlier showed that carboxylated N-glycans expressed on receptor for advanced glycation end products (RAGE) and other glycoproteins mediate colitis through activation of nuclear factor kappa B (NF-kappaB). Because NF-kappaB signaling plays a critical role in the molecular pathogenesis of colitis-associated cancer (CAC), we reasoned that carboxylated glycans, RAGE and its ligands might promote CAC. Carboxylated glycans are expressed on a subpopulation of RAGE on colon cancer cells and mediate S100A8/A9 binding to RAGE. Colon tumor cells express binding sites for S100A8/A9 and binding leads to activation of NF-kappaB and tumor cell proliferation. Binding, downstream signaling and tumor cell proliferation are blocked by mAbGB3.1, an anti-carboxylate glycan antibody, and by anti-RAGE. In human colon tumor tissues and in a mouse model of CAC, we found that myeloid progenitors expressing S100A8 and S100A9 infiltrate regions of dysplasia and adenoma. mAbGB3.1 administration markedly reduces chronic inflammation and tumorigenesis in the mouse model of CAC and RAGE-deficient mice are resistant to the onset of CAC. These findings show that RAGE, carboxylated glycans and S100A8/A9 play essential roles in tumor-stromal interactions, leading to inflammation-associated colon carcinogenesis.

Fig. 1.

Expression of mAbGB3.1 glycans, RAGE and S100A9 in human colorectal tumors. Representative immunostained tumor sections and paired normal adjacent tissue (NAT) for tumor 1 are shown. (tumor 1, stage IIIB; tumor 2, stage IIIC; arrowhead, macrophages; asterisk, endothelial cells; arrow, tumor cells). Tumor cells are characterized by hyperchromatic nuclei (intense hematoxylin staining). Bar = 100 μm for all images except for mAbGB3.1 staining of tumor 2 that is enlarged to show staining of tumor vasculature and for the inset for tumor 1 to show distinct mAbGB3.1 staining of tumor cells and macrophages (bar = 50 μm).

Fig. 2.

(A) Colon tumor cells express RAGE and mAbGB3.1 glycans. Cell membrane proteins from CT-26 cells (lanes 1, 4 and 7), HT-29 cells (lanes 2, 5 and 8) and Caco-2 cells (lanes 3, 6 and 9) were examined for RAGE expression by western blot using anti-RAGE before deglycosylation (lanes 1–3, 20 μg protein per lane) and after deglycosylation (lanes 4–6, 20 μg protein per lane) and after mAbGB3.1 immunoprecipitation (lanes 7–9, immunoprecipitated from 1 mg of membrane proteins). (B) Purification of bovine lung RAGE and mAbGB3.1 enrichment. Left panel: protein stain by Coomassie brilliant blue of a representative RAGE preparation from bovine lung shows >98% purity. Middle panel: western blot using anti-RAGE shows that purified RAGE carries EndoH-sensitive and PNGase F-sensitive N-glycan chains. Right panel: western blot using anti-RAGE shows that mAbGB3.1 immunoprecipitates a minor subpopulation of RAGE. (C) Purified mouse S100A8/A9. S100A8/A9 complex was purified as described before, and 5 μg protein analyzed on 17% gels and purity confirmed by Coomassie brilliant blue. Arrow marks the position of covalently linked 26 kDa dimer of S100A8/A9. (D) S100A8/A9 complex binds purified RAGE. To determine saturation kinetics of binding of mouse S100A8/A9 to purified RAGE, increasing amounts of S100A8/A9 were added to total RAGE, mAbGB3.1-enriched RAGE or RAGE deglycosylated using PNGase F under non-denaturing conditions that removed both N-glycans. RAGE on plate was quantified using anti-RAGE. Bound S100A8/A9 was quantified using anti-S100A8 against standard S100A8/A9. Data were fitted to non-linear regression analysis using GraphPad Prism. Each point is the mean ± SD of two determinations.

Fig. 3.

(A) Binding of ¹²⁵I S100A8/A9 to CT-26 cells. Cells were incubated with increasing concentrations of ¹²⁵I S100A8/A9 for 1 h at 4°C followed by washing and cell lysis, and cell-bound radioactivity was measured using a gamma counter. Saturation binding kinetic analysis was performed using GraphPad Prism. Values represent mean ± SD of two determinations. (B) Inhibition of binding of ¹²⁵I S100A8/A9 to CT-26 cells. Cells were incubated with ¹²⁵I S100A8/A9 (20nM) in the presence or absence of mAbGB3.1, anti-RAGE or anti-S100A8 (10-fold molar excess) or cold ligand (50-fold molar excess). Cell-bound radioactivity was determined as above. Data represent mean ± SD of two determinations (*P ≤ 0.05 and **P ≤ 0.01). (C) S100A8/A9 induces NF-κB-dependent transcription. CT-26 cells were transiently transfected with plasmids containing firefly luciferase reporter gene under a promoter containing NF-κB-binding site and Renilla luciferase construct as an internal control. Transfected cells were stimulated with S100A8/A9 in presence or absence of inhibitors. Cell lysates were assayed for luciferase activity. Values are represented as ratio of firefly luciferase activity over Renilla luciferase (fold induction relative to unstimulated cells). Each value is the mean ± SD of two determinations (*P ≤ 0.05 and **P ≤ 0.01). (D) S100A8/A9 proteins stimulate colon cancer cell proliferation. CT-26 cells were incubated with increasing concentrations of S100A8/A9 in the presence or absence of mAbGB3.1, control antibody or anti-RAGE. At low concentrations, S100A8/A9 stimulated cell proliferation that was blocked by mAbGB3.1 and anti-RAGE. S100A8/A9-induced growth was not dependent on time or concentration as seen earlier with other tumor cells.

Fig. 4.

(A) Representative hematoxylin and eosin-stained colon sections indicating progress of CAC in untreated wild-type mice subjected to the AOM–DSS protocol. (a) Normal colon Swiss-roll ×10 magnification. (b) Normal colon. (c) Colonic inflammation, 2 weeks after DSS. (d) Colonic inflammation, 6 weeks after DSS. Arrow indicates early dysplasia in a region of inflammation. (e) High-grade dysplasia (flat polypoid) 12 weeks after DSS. (f) Adenoma 12 weeks after DSS. (g) Adenoma 20 weeks after DSS. (B) (a) A representative tumor shows S100A9-positive infiltrating cells. Cells were also positive for S100A8 (data not shown). (b) A region adjacent to the tumor from the same colon is negative for S100A8/A9-positive cells. (C) Two representative sections of tumors with infiltrating cells stained for CD11b (myeloid) or Gr-1 (granulocyte). There was evidence for CD11b⁺, Gr-1⁺ single-positive and CD11b⁺/Gr-1⁺ double-positive cells within tumors representing infiltrating myeloid progenitors. Magnification scale bar is indicated for each panel.

Fig. 5.

(A) mAbGB3.1 administration reduces colonic inflammation (6 weeks) and incidence of tumors (12 weeks in preventive and therapeutic protocols) in mice treated with AOM–DSS. Mice were administered with mAbGB3.1 or a control antibody as described in Materials and Methods. Colonic inflammation, dysplasia and adenomas were evaluated using established criteria (n = 4 per group per time point, **P ≤ 0.01). (B) TNFα and IL-6 were measured in sera of mice at different time points (n = 4 per group per time point, **P ≤ 0.01). (C) Colonic inflammation in RAGE^+/+ and RAGE^−/− mice 2 weeks after AOM–DSS evaluated using established criteria. (D) Colonic tumor incidence in RAGE^+/+ and RAGE^−/− mice 20 weeks after AOM–DSS.