FBXO22 inhibits colitis and colorectal carcinogenesis by regulating the degradation of the S2448-phosphorylated form of mTOR

Abstract

Inflammatory bowel disease (IBD) is a considerable threat to human health with a significant risk for colorectal cancer (CRC). However, currently, both the molecular pathogenesis and therapeutic treatment of IBD remain limited. In this report, using both systemic and intestinal epithelium-specific gene knockout mouse models, we demonstrate that FBXO22, a substrate receptor within the SKP1-Cullin 1-F-box family of E3 ubiquitin ligases, plays an inhibitory role in the Azoxymethane/Dextran Sodium Sulfate-induced colorectal inflammatory responses and CRC. FBXO22 targets the serine 2448-phosphorylated form of mammalian mechanistic target of rapamycin (pS2448-mTOR) for ubiquitin-dependent degradation. This proteolytic targeting effect is established based on multiple lines of evidence including the results of colon tissue immunoblots, analysis of cultured cells with altered abundance of FBXO22 by depletion or overexpression, comparison of protein decay rate, effects on mTOR substrates S6K1 and 4E-BP1, analysis of protein-protein interactions, phosphor-peptide binding and competition, as well as reconstituted and cellular ubiquitination. Finally, we have shown that mTOR inhibitor rapamycin (RAPA) was able to alleviate the effects of fbxo22 deletion on colorectal inflammatory response and CRC. These RAPA effects are correlated with the ability of RAPA to inhibit pS2448-mTOR, pS6K1, and p4E-BP1. Collectively, our data support a suppressive role for FBXO22 in colorectal inflammation signaling and CRC initiation by targeting pS2448-mTOR for degradation.

Keywords: FBXO22; colitis; colorectal carcinogenesis; pS2448-mTOR.

Fig. 1.

FBXO22 deletion exacerbates AOM/DSS-induced inflammatory response. (A) Experimental scheme for induction of colitis in mice, introducing AOM by intraperitoneal injection and DSS in drinking water. (B–D) Changes in body weight (B), diarrhea (C), and rectal bleeding (D) of the treated mice with indicated genotypes (n ≥ 5 for each group). (E–G) Analysis of the colon sections from the AOM/DSS-treated mice. (E) Images of colon. (F) H&E staining and Histopathological score of colons was graded following analysis of H&E staining. (G) Claudin-3 staining. n ≥ 5. Quantitative comparisons are provided. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2.

FBXO22 deletion promotes CRC in mice. (A) Experimental scheme for induction of CRC in mice. (B) Changes in body weight of the treated mice with indicated genotypes (n = 5 for each group). (C) Representative images of dissected colons from the treated mice. Red arrows point to the polyps. (D) Number of colonic tumor induced by the AOM/DSS treatment in mice (n ≥ 10). (E) Size distribution of colonic polyps induced by the AOM/DSS treatment (n = 8). Note that large tumors/polyps (>2 mm) appeared more frequent in the (+/+) mice relative to (+/−) colons. It is presently unclear whether this difference bears any biological significance. (F) Mortality of mice (n = 15). (G and H) Colon histological and IHC analysis. (G) H&E staining. Dashed boxes point to tumor legions. (H) Ki67 staining. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3.

FBXO22 deletion elevates the level of pS2448-mTOR. (A) Representative immunoblots of endogenous FBXO22, mTOR, and pS2448-mTOR in colon tissues from the AOM/DSS-treated fbxo22^+/+-Villin-Cre and fbxo22^fl/fl-Villin-Cre mice and quantitation of protein expression level. (B) Representative immunoblots of mTOR and pS2448-mTOR in HCT116 and MC38 cells with FBXO22 KD and quantitation of protein expression level. (C) Representative immunoblots of mTOR and pS2448-mTOR in HCT116 and MC38 cells overexpressing FBXO22 and quantitation of protein expression level. (D) HCT116 or MC38 cells were transfected with indicated plasmids and then treated with or without MG132. Total cell lysates were subjected to immunoblotting with the indicated antibodies. All immunoblots were carried out in at least three biological repeats with quantification presented. *P < 0.05, **P < 0.01, ***P < 0.001. n.s, No significance.

Fig. 4.

FBXO22 mediates ubiquitination and degradation of pS2448-mTOR in CRC cells. (A) CHX analysis. Representative immunoblots of mTOR, pS2448-mTOR and its substrates 4E-BP1 and S6K1 in HCT116 cells overexpressing FBXO22 that were exposed to CHX and chased for indicated time period. Graphs show data quantification of three independent experiments. (B) Co-IP of FBXO22 and pS2448-mTOR in ectopic forms. (C) Substitution of S2448 by alanine inhibits the binding of mTOR to FBXO22. (D) BT-PS2448 binds to FBXO22. (E) BT-PS2448 inhibits the interactions between FBXO22 and pS2448-mTOR. (F) Reconstitution of ubiquitination of BT-PS2448 in vitro.

Fig. 5.

RAPA attenuates inflammatory responses in fbxo22-deleted mice. (A) Experimental scheme for colitis recovery by RAPA in mice. Mice were treated with RAPA or equal amounts of sterile water by gavage every day from DSS drinking. (B−D) Changes in body weight, diarrhea, and rectal bleeding of mice with indicated genotypes (n ≥ 5). (E−G) Colon analysis. (E) Representative colon images. (F) H&E staining and Histopathological score of colons were graded following analysis of H&E staining. (G) Claudin-3 staining. n ≥ 5. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 6.

RAPA inhibits colorectal tumor formation in FBXO22 deleted mice. (A) Experimental scheme for RAPA-mediated CRC inhibition in mice. (B−E) Colon analysis. (B) Representative images of dissected colons from the treated mice with indicated genotypes. Red arrows point to the polyps. (C) The average tumor number. n ≥ 5. (D) H&E staining. Dashed boxes point to tumor legions. (E) Ki67 staining. (F) Immunoblot analysis of FBXO22 and components in the mTOR signaling pathway in colonic epithelial cells from the treated mice. These immunoblots were carried out in at least three biological repeats. *P < 0.05, **P < 0.01, ***P < 0.001. Note that the effects of RAPA on decreasing pS2448-mTOR levels are statistically significant albeit that there is a considerable degree of variations in response to RAPA by mice.