Galectin-9 Is a Novel Regulator of Epithelial Restitution

Abstract

Increasingly, the ß-galactoside binding lectins, termed galectins, are being recognized as critical regulators of cell function and organismal homeostasis. Within the context of the mucosal surface, galectins are established regulators of innate and adaptive immune responses, microbial populations, and several critical epithelial functions, including cell migration, proliferation, and response to injury. However, given their complex tissue distribution and expression patterns, their role within specific processes remains poorly understood. We took a genetic approach to understand the role of endogenous galectin-9 (Gal-9), a mucosal galectin that has been linked to inflammatory bowel disease, within the context of the murine intestine. Gal-9-deficient (Gal9^-/-, also known as Lgals9^-/-) animals show increased sensitivity to chemically induced colitis and impaired proliferation in the setting of acute injury. Moreover, Gal9^-/--derived enteroids showed impaired growth ex vivo. Consistent with a model in which endogenous Gal-9 controls epithelial growth and repair, Gal9^-/- animals showed increased sensitivity to intestinal challenge in multiple models of epithelial injury, including acute irradiation injury and ectopic wound biopsies. Finally, regenerating crypts from patient biopsies showed increased expression of Gal-9, indicating these processes may be conserved in humans. Taken together, these studies implicate Gal-9 in the regulation of cellular proliferation and epithelial restitution after intestinal epithelial injury.

Figure 1

Dextran sodium sulfate (DSS)-induced colitis in wild-type (WT) and galectin-9–deficient (Gal9^−/− also known as Lgals9^−/−) mice. A: Gal-9 expression in colon preparations from vehicle-treated and DSS-treated mice (treated with 2% DSS for 6 days). B and C: C57BL/6 (WT) and Gal-9^−/− mice were administered either vehicle (veh) or 3% DSS and monitored for weight loss (B) and disease activity (C). D and E: After 8 days of treatment, mice were sacrificed, colons were dissected (D), and colon length was quantified (E). n = 5 per group (A). ∗∗∗P < 0.0005. Scale bars = 1 cm (D). DAI, disease activity index; KO, knockout.

Figure 2

Histologic analysis of dextran sodium sulfate (DSS)-treated wild-type (WT) and galectin-9–deficient (Gal-9^−/− also known as Lgals9^−/−) mice. A–D: Representative photomicrographs of hematoxylin and eosin–stained sections from the colonic mucosa obtained from day 8 preparations of either vehicle-treated WT and Gal-9^−/− mice (A and B, respectively) or DSS-treated WT and Gal-9^−/− mice (C and D, respectively). Arrows indicate crypts at ulcerated edge. Higher-magnification photomicrographs (20× objective) are shown on the right (C and D). E–G: Hematoxylin and eosin–stained sections were scored for the degree of colitis as determined by the histologic activity index (E), which is a composite of the extent of inflammatory infiltrate (F) and the degree of epithelial injury (G). H: The percentage of epithelial loss as calculated based on the amount of intact epithelia. ∗∗P < 0.005. Scale bars = 500 μm (A–D). HAI, histology activity index; KO, knockout; veh, vehicle.

Figure 3

Proliferation responses in regenerating crypts of dextran sodium sulfate (DSS)-treated wild-type (WT) and galectin-9–deficient (Gal-9^−/− also known as Lgals9^−/−) mice. A and B: Representative photomicrographs of crypts adjacent to ulcerated mucosa of DSS-treated mice from WT (A) and Gal-9^−/− (B) animals. C: Quantification of crypt length from crypts adjacent to ulcerated mucosa of DSS-treated mice from WT and Gal-9^−/− animals. D and E: Ki-67, DAPI, and ß-catenin immunostaining performed on (D) WT and (E) Gal-9^−/− animals. Boxed areas indicate ulcerated edges, also known as Lgals9^−/−, which are represented in higher magnification in images to the right. F: Quantification of the Ki-67 proliferation index of crypts adjacent to the ulcerated edge of DSS-treated WT and Gal-9^−/− mice. ∗∗P < 0.005, ∗∗∗P < 0.0005. Scale bars: 100 μm (A and B); 70 μm (D and E). KO, knockout.

Figure 4

Intestinal organoids derived from wild-type (WT) and galectin-9–deficient (Gal9^−/−, also known as Lgals9^−/−) mice. A: The number of spheroids versus enteroids on day 9 preparations. B: The number of buds per enteroid for all visualized enteroids, defined as any budding three-dimensional structure. C and D: Representative photomicrographs of 9-day–old intestinal organoids isolated from WT (C) and Gal-9^−/− (D) animals. ∗P < 0.05, ∗∗P < 0.005. Scale bars = 50 μm (C and D). KO, knockout.

Figure 5

Response to irradiation-induced injury in wild-type (WT) and galectin-9–deficient (Gal9^−/−, also known as Lgals9^−/−) mice. A and B: Representative photomicrographs of hematoxylin and eosin staining from midjejunal preparations of mucosa 5 days after irradiation from WT (A) and Gal-9^−/− (B) animals. C and D: For each group, the villus length (C) and crypt density (D) is shown. Scale bars = 100 μm (A and B). ∗P < 0.05. HPF, high-power field; KO, knockout.

Figure 6

Analysis of response to endoscopic wound biopsy in wild-type (WT) and Gal9^−/− (also known as Lgals9^−/−) mice. WT and galectin-9–deficient (Gal-9^−/−) mice were anesthetized and exposed to endoscopic wounds generated from a forceps biopsy. At 24 hours after the biopsy, wounds generated in WT (top row, left) and Gal9^−/− (top row, right) mice were photographed using miniaturized endoscopy and compared with images obtained 72 hours after the biopsy (bottom row left versus right). Dotted lines indicate the wound edge from the biopsy. To the right of the photomicrographs, the percentage of wound closure calculated at 72 hours after the biopsy in WT and Gal-9^−/− mice is shown. ∗∗P < 0.005. KO, knockout.

Figure 7

Galectin 9 (Gal-9) expression in human colonic biopsies. A: Hematoxylin and eosin staining of an endoscopic mucosal resection from a patient with a previous procedure resulting in erosion and biopsy-site changes (higher magnification on right, arrow indicates proliferative crypts at eroded edge of wound). B: Serial sections were stained with DAPI and anti-human galectin-9 (Gal-9) antibody; arrow indicates increased Gal-9 expression in proliferating crypts adjacent to biopsy site. Scale bars: 500 μm (A, left panel); 250 μm (A, right panel, and B).