Receptor for advanced glycation end products is targeted by FBXO10 for ubiquitination and degradation

Abstract

The receptor for advanced glycation end products (RAGE) is a highly expressed cell membrane receptor serving to anchor lung epithelia to matrix components, and it also amplifies inflammatory signaling during acute lung injury. However, mechanisms that regulate its protein concentrations in cells remain largely unknown. Here we show that RAGE exhibits an extended life span in lung epithelia (t_½ 6 h), is monoubiquitinated at K374, and is degraded in lysosomes. The RAGE ligand ODN2006, a synthetic oligodeoxynucleotide resembling pathogenic hypomethylated CpG DNA, promotes rapid lysosomal RAGE degradation through activation of protein kinase Cζ (PKCζ), which phosphorylates RAGE. PKCζ overexpression enhances RAGE degradation, while PKCζ knockdown stabilizes RAGE protein levels and prevents ODN2006-mediated degradation. We identify that RAGE is targeted by the ubiquitin E3 ligase subunit F-box protein O10 (FBXO10), which associates with RAGE to mediate its ubiquitination and degradation. FBXO10 depletion in cells stabilizes RAGE and is required for ODN2006-mediated degradation. These data suggest that modulation of regulators involved in ubiquitin-mediated disposal of RAGE might serve as unique molecular inputs directing RAGE cellular concentrations and downstream responses, which are critical in an array of inflammatory disorders, including acute lung injury.-Evankovich, J., Lear, T., Mckelvey, A., Dunn, S., Londino, J., Liu, Y., Chen, B. B., Mallampalli, R. K. Receptor for advanced glycation end products is targeted by FBXO10 for ubiquitination and degradation.

Keywords: F-box proteins; acute lung injury; ubiquitin.

Figure 1.

RAGE is a monoubiquitinated receptor degraded by lysosomes. A) RAGE levels decrease with CHX treatment and are rescued with Leu. B2B cells were treated with CHX (100 µg/ml) or CHX + MG-132 (60 mmol) or Leu (60 mmol) for 3 or 6 h and immunoblotted for RAGE or actin. Immunoblot is representative of 3 independent experiments. B) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3) C) B2B cells were transfected with increasing amounts of HA-ubiquitin (Ub) plasmid and immunoblotted for RAGE, HA-Ub, or actin 16 h after transfection. D) B2B cells were transfected with 1 µg empty vector or 1 µg HA-Ub and treated with CHX (100 µg/ml) for 2, 4, or 6 h. RAGE levels decreased more rapidly with HA-Ub overexpression in response to CHX. E) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). F) B2B whole-cell lysates were immunoblotted for RAGE after ubiquitin pull-down with TUBES reagent. Endogenous RAGE migrated just above 50 kDa in input samples, and a slower-migrating higher MW band was observed in Ub pull-down samples, but not agarose control samples cells, representing monoubiquitinated RAGE (Ub-RAGE). MG-132 (60 mmol) pretreatment modestly increased Ub-RAGE levels, while Leu (60 mmol) greatly enhanced Ub-RAGE levels. *P < 0.05, CHX + Leu statistically significant compared to CHX at 6 h (B) or at 4 and 6 h (E), Student’s t test.

Figure 2.

Lysine 374 regulates RAGE stability. A) B2B cells were transfected with either RAGE WT or RAGE K374R plasmids and treated with CHX (100 µg/ml) for 2, 4, or 6 h. RAGE WT levels decreased with CHX treatment, while RAGE K374R levels remained constant. B) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). C) B2B cells were transfected with empty vector, RAGE WT, or RAGE K374R. After His pull-down, samples were immunoblotted for V5 or ubiquitin. *P < 0.05 at 4 and 6 h, Student’s t test.

Figure 3.

RAGE CpG DNA ligand ODN2006 decreases intracellular RAGE levels. A) B2B cells were treated with ODN2006 for 15, 30, 60, or 120 min and immunoblotted for RAGE or actin. B) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). C) B2B cells were treated with 0.01, 0.1, 1, or 5 µM ODN2006 for 60 min and immunoblotted for RAGE or actin. D) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). E) MLE12 cells were treated with 1 µM ODN2006 for 60 min and immunoblotted for RAGE. F) B2B cells were stimulated with RAGE ligand s100a12 (5 μg/ml) for 15, 30, 60, or 120 min and immunoblotted for RAGE or actin. *P < 0.05 compared to time point 0 (B) or untreated cells (D), Student’s t test.

Figure 4.

ODN2006 promotes lysosomal RAGE degradation. A) B2B cells were pretreated with GI254023X (10 μM), ADAM10 inhibitor, for 30 min before stimulation with ODN2006 for 15, 30, 60, or 120 min. B) B2B cells were treated with PMA (1 μM) for 60 min or ODN2006 for 15, 30, or 60 min and immunoblotted for RAGE, which detects both FL-RAGE and Cl-RAGE, or actin. FL-RAGE decreased, but no Cl-RAGE was detected. C) Cells were pretreated with DAPT (2 μM) for 60 min, followed by stimulation with PMA (1 μM) for 60 min or ODN2006 for 15, 30, or 60 min. FL-RAGE decreased with ODN2006 stimulation. With long exposure, Cl-RAGE was detected, and PMA increased Cl-RAGE compared to untreated cells, while Cl-RAGE decreased similarly to FL-RAGE in ODN2006-treated cells. D) Leu preserves RAGE expression in response to ODN2006. B2B cells were pretreated with MG-132 or Leu before stimulation with ODN2006 (1 μM) for 15, 30, and 60 min, followed by immunoblotting for RAGE or actin.

Figure 5.

PKCζ modulates RAGE stability. A) PKCζ is activated by ODN2006. B2B cells were treated with ODN2006 (1 μM) for 15, 30, or 60 min, and immunoblotted for activated phospho-PKCζ, total PKCζ, or actin. B) ODN2006 increases phospho-RAGE. Cells were transfected with RAGE WT plasmid and treated with ODN2006 for 15, 30, or 60 min, followed by His pull-down and immunoblotting for V5 or phosphoserine. C) PKCζ overexpression reduces RAGE levels. Cells were transfected with 1, 2, or 3 µg PKCζ or PKCα plasmid, followed by immunoblotting for RAGE, V5, or actin. Parentheses represent densitometric values normalized to control untransfected cells. D) MLE12 cells were transfected with 3 µg PKCζ plasmid and immunoblotted for RAGE. E) Cells were transfected with empty vector or PKCζ plasmid and treated with CHX (100 µg/ml) for 2, 4, or 6 h. F) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). G) PKCζ knockdown stabilizes RAGE expression. Cells were transfected with control shRNA or PKCζ shRNA. After 72 h, cells were treated with CHX (100 µg/ml) for 2, 4, or 6 h. H) Densitometry quantification of 2 independent experiments. Data represent means ± sem (n = 2). I) PKCζ knockdown stabilizes RAGE in response to ODN2006. Cells were transfected with control shRNA or PKCζ shRNA. After 72 h, cells were treated with ODN2006 for 15, 30, or 60 min and immunoblotted for PKCζ and RAGE. *P < 0.05 at 4 and 6 h (F, H), Student’s t test.

Figure 6.

S391 modulates RAGE stability. A) RAGE S391A is resistant to degradation. B2B cells were transfected with RAGE WT or RAGE S391A plasmid and treated with CHX (100 µg/ml) for 2, 4, or 6 h. B) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). C) Cells were transfected with RAGE WT or RAGE S391D plasmid and treated with CHX (100 µg/ml) for 2, 4, or 6 h. D) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). *P < 0.05 at 4 and 6 h, Student’s t test.

Figure 7.

FBXO10 promotes RAGE degradation. A) Cells were transfected with various F-box plasmids and immunoblotted for V5 and RAGE. B) RAGE and FBXO10 interact. V5-RAGE WT plasmid was transfected into B2B cells, followed by IP for FBXO10 and immunoblotting for FBXO10, or V5. C) Cells were transfected with 1, 2, or 3 µg FBXO10 plasmid and immunoblotted for V5, RAGE, or actin. D) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). E) B2B cells were transfected with 1 µg empty vector or 1 µg FBXO10 and treated with CHX (100 µg/ml) for 2, 4, or 6 h. F) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). G) FBXO10 knockdown stabilizes RAGE in response to CHX. B2B cells were transfected with scrambled siRNA or FBXO10 siRNA. After 72 h, cells were treated with CHX (100 µg/ml) for 2, 4, or 6 h and immunoblotted for RAGE or actin. H) FBXO10 co-overexpression reduces RAGE WT but not RAGE K374R levels. Cells were transfected with RAGE WT or RAGE K374R, with or without FBXO10 plasmid, and immunoblotted for V5. I) FBXO10 co-overexpression reduces RAGE WT and RAGE S391D, but not RAGE S391A, levels. Cells were transfected with RAGE WT, RAGE S391D, or RAGE S391A, with or without FBXO10 plasmid, and immunoblotted for V5. J) FBXO10 knockdown stabilizes RAGE in response to ODN2006. Cells were transfected with control siRNA or FBXO10 siRNA. After 72 h, cells were treated with ODN2006 for 60 min and immunoblotted for FBXO10, RAGE, or actin. K) Densitometry quantification of 3 independent experiments. Data represent means ± sem (n = 3). *P < 0.05, significant compared to control (D, K); at 4 and 6 h (F), Student’s t test.