N-glycosylation of R-spondin1 at Asn137 negatively regulates its secretion and Wnt/β-catenin signaling-enhancing activity

Abstract

N-glycosylation is a post-translational protein modification with a wide variety of functions. It has been predicted that R-spondin1 (RSPO1) is N-glycosylated, although this remains unknown. The present study identified that RSPO1 was N-glycosylated at Asn137, and that N-glycosylation of RSPO1 negatively influenced its secretion and enhancing effect on Wnt/β-catenin signaling. In vitro treatment with peptide-N-glycosidase F increased the electrophoretic mobility of RSPO1. Furthermore, treatment of wild-type (wt) RSPO1-overexpressing HT1080 cells with tunicamycin (TM), which inhibits N-glycosylation, resulted in a significant reduction in the molecular weight of RSPO1. However, TM treatment had no effect in the RSPO1 mutant whereby the Asn137 residue was replaced by Gln (N137Q). These results demonstrated for the first time that RSPO1 is N-glycosylated at Asn137. RSPO1 is a secreted protein that has Wnt/β-catenin signaling-enhancing activity and is expected to have therapeutic applications. The role of N-glycosylation in RSPO1 was evaluated by conducting comparative experiments with wt and N137Q RSPO1, which revealed that the N137Q mutant increased the secretion and Wnt/β-catenin signaling-enhancing effect of RSPO1, compared with wt RSPO1. These results suggest that N-glycosylation of RSPO1 has a negative influence on its secretion and Wnt/β-catenin signaling-enhancing effect.

Keywords: N-linked glycosylation; R-spondin1; Wnt signaling; glycosylation; heparin-binding protein; protein secretion.

Figure 1.

N-glycosylation of human RSPO1. (A) Schematic diagram of human RSPO1 protein, which consists of 263 amino acids. The location of a putative N-glycosylation site (Asn137) is indicated by a sugar chain. Protein domain architecture denotes signal peptide (red box), furin repeat (blue boxes) and thrombospondin type-1 (green box). (B) HT1080-neo and HT1080-RSPO1-MH cells were treated with or without 10 µg/ml TM for 24 h. Subsequently, cells were lysed, and aliquots of each cell lysate were electrophoresed and immunoblotted with anti-c-Myc or anti-α-tubulin antibodies. (C) Deglycosylation of RSPO1 by PNGase F in vitro. HT1080-neo and HT1080-RSPO1-MH cells were treated with or without 10 µg/ml TM for 24 h. Subsequently, cells were lysed and treated with or without PNGase F for 3 h. Samples were then electrophoresed and immunoblotted with anti-c-Myc antibody. RSPO1, R-spondin1; TM, tunicamycin; PNGase F, peptide-N-glycosidase F; neo, pCI-neo vector.

Figure 2.

RSPO1 is N-glycosylated at Asn137. (A and B) Establishment of wt and N-glycosylation-defective RSPO1 mutant-expressing HT1080 cells. (A) Total RNA was isolated from each cell line, and semiquantitative reverse transcription-polymerase chain reaction analysis was performed (B) Cells were lysed, and aliquots of the cell lysates were electrophoresed and immunoblotted with anti-c-Myc or anti-α-tubulin antibodies. (C) Inhibition of N-glycosylation of RSPO1. HT1080-neo, HT1080-RSPO1-MH and HT1080-RSPO1/N137Q-MH cells were treated with the indicated concentrations of tunicamycin for 24 h. Subsequently, cells were lysed, and aliquots of the cell lysates were electrophoresed and immunoblotted with anti-c-Myc or anti-α-tubulin antibodies. RSPO1, R-spondin1; TM, tunicamycin; neo, pCI-neo vector; wt, wild-type.

Figure 3.

Effect of N-glycosylation on the secretion and trafficking of RSPO1. (A) Effect of N-glycosylation on the secretion of RSPO1. HT1080-neo, HT1080-RSPO1-MH and HT1080-RSPO1/N137Q-MH cells were cultured in serum-free DMEM for 24 h in the presence of 50 µg/ml heparin, and CM and cell lysates were collected. Samples from CM were concentrated with Ni-NTA agarose, and alongside aliquots of the cell lysates, were electrophoresed and immunoblotted with anti-c-Myc or anti-α-tubulin antibodies. (B) Effect of N-glycosylation on the kinetics of RSPO1 secretion. Each cell line was cultured in serum-free DMEM with 50 µg/ml heparin for the indicated time points, and CM were collected. Following concentration with Ni-NTA agarose, the samples were electrophoresed and immunoblotted with anti-c-Myc antibody. Protein bands were quantified using ImageQuant™ TL version 8.1 software to generate the graph. Secreted levels of RSPO1 are the levels of wt RSPO1 protein quantified by the software at 24 h, which is defined as 1. (C) Effect of N-glycosylation on the intracellular trafficking of RSPO1. HT1080-neo, HT1080-RSPO1-MH and HT1080-RSPO1/N137Q-MH cells were fixed prior to be stained with Hoechst 33258 (blue), anti-c-Myc antibody (green) and anti-Golgi reassembly-stacking protein of 65 kDa antibody (red), and observed by fluorescence microscopy. Areas of overlapping stains indicating co-localization were represented in yellow when the images were superimposed. Bar, 10 µm. Magnification, ×400. RSPO1, R-spondin1; neo, pCI-neo vector; wt, wild-type; DMEM, Dulbecco's modified Eagle's medium; NTA, nitrilotriacetic acid; CM, conditioned media.

Figure 4.

Effect of N-glycosylation on the Wnt signaling-enhancing effect of RSPO1. Wt and mutant RSPO1 proteins were purified from their corresponding CM and quantified by immunoblotting (inset). HEK293T cells were transiently transfected with TOPFlash or FOPFlash (firefly luciferase) and phRL-TK (Renilla luciferase) vectors, prior to be stimulated with equal amounts of recombinant wt or N137Q mutant RSPO1 in the presence of Wnt3A-CM. Following 24 h, luciferase activities were measured and normalized with Renilla luciferase activity. Data are the mean± standard deviation. *P<0.05. RSPO1, R-spondin1; wt, wild-type; CM, conditioned medium.