Human intestinal MUC17 mucin augments intestinal cell restitution and enhances healing of experimental colitis

Abstract

The membrane-bound mucins, MUC17 (human) and Muc3 (mouse), are highly expressed on the apical surface of intestinal epithelia and are thought to be cytoprotective. The extracellular regions of these mucins contain EGF-like Cys-rich segments (CRD1 and CRD2) connected by an intervening linker domain (L). The purpose of this study was to determine the functional activity of human MUC17 membrane-bound mucin.

Methods: Endogenous MUC17 was inhibited in LS174T colon cells by stable transfection of a small hairpin RNA targeting MUC17 (LSsi cells). The effect of recombinant MUC17-CRD1-L-CRD2 protein on migration, apoptosis, and experimental colitis was determined.

Results: Reduced MUC17 expression in LSsi cells was associated with visibly reduced cell aggregation, reduced cell-cell adherence, and reduced cell migration, but no change in tumorigenicity. LSsi cells also demonstrated a 3.7-fold increase in apoptosis rates compared with control cells following treatment with etoposide. Exposure of colonic cell lines to exogenous recombinant MUC17-CRD1-L-CRD2 protein significantly increased cell migration and inhibited apoptosis. As a marker of biologic activity, MUC17-CRD1-L-CRD2 proteins stimulate ERK phosphorylation in colonic cell lines; and inhibition of ERK phosphorylation reduced the anti-apoptosis and migratory effect of MUC17-CRD1-L-CRD2. Finally, mice treated with MUC17-CRD1-L-CRD2 protein given per rectum demonstrated accelerated healing in acetic acid and dextran sodium sulfate induced colitis in vivo. These data indicate that both native MUC17 and the exogenous recombinant cysteine-rich domain of MUC17 play a role in diverse cellular mechanisms related to cell restitution, and suggest a potential role for MUC17-CRD1-L-CRD2 recombinant protein in the treatment of mucosal inflammatory diseases.

Fig. 1

Schematic representation of MUC17 protein structures. The MUC17 amino acid sequence is comprised of a signal peptide, a large tandemly repeated central domain (TR), two EGF-like domains, a SEA domain, a transmembrane domain (TM), and an 80 amino acid cytoplasmic tail (Cyt). The recombinant proteins used are derived from the two Cys-rich domains (CRD1 and CRD2), separated by a linker (L) domain which includes the SEA domain. On this figure CRD1 corresponds to the first EGF-like domain, the linker region corresponds to the next domain that includes the SEA module, and CRD2 corresponds to the second EGF-like domain.

Fig. 2

Expression analysis of MUC17. (A) RT-PCR for MUC17 mRNA expression in LSscram and LSsi cells. RT-PCR performed with B-actin primers were used to indicate mRNA in all lanes. (B) Western blot of cell lysates from LSscram and LSsi using anti-MUC17 antibody. Protein lysates from were resolved on 2% agarose gel containing SDS and transferred passively onto PVDF membrane. A unique band was revealed with a molecular weight consistent with the 500 kDa expected for MUC17. Loading control indicated by immunoreactivity using anti-actin antibody. Samples were tested in duplicate on two different occasions with similar results.

Fig. 3

LS174T cell morphology, adhesion, and cell growth. (A) Cell morphology in standard cell culture environment: 1600× magnification images of LSscram (a) and LSsi (d). Cell aggregates in a droplet environment: LS scram at 40× (b) and 100× (c) magnification and LSsi at 40× (e) and 100× (f) magnification. (B) Cell Adhesion assay. LS174T cells were seeded at 2 × 10⁵ cells/well into 6-well plates, in triplicate and performed twice. After 5 h of incubation at the medium was aspirated and unattached cells were counted and expressed as a percentage of total cells seeded, *p < 0.002 vs. LSscram. (C) Cell growth assay. LSsi and LSscram cells seeded in triplicate onto 6-well plates at 5 × 10⁴ cells/well and counted at 48 and 96 h. Mean ± SEM, *p < 0.002, **p < 0.01 vs. LSscram.

Fig. 4

(A) Colony formation in soft agar with different initial seeding densities. Cells were seeded at indicated densities, allowed to incubate for 14 days at 37 °C, then the plates were imaged and number of colonies counted with the use of ImageJ. n = 4 per initial seeding density. Mean ± SEM. (B) Tumor size in nude mice over time after initial injection of 1 × 10⁶ LSsi cells per site (n = 9) and 1 × 10⁶ LSscram cells per site (n = 9). Mean ± SEM. (C) Final tumor weights after 15 days in LS scram (n = 9) and LSsi (n = 9), Mean ± SEM.

Fig. 5

Cell migration and apoptosis. (A) The number of cells migrating across a razor scrape (arrow) was measured after 48 h at 37 °C; two images of LSscram cells and two images of LSsi cells (1600×). (B) Surface area of migration over wound over 48 h measured by computer imaging. (Mean ± SEM); *p < 0.00001. (C) Percent of apoptotic cells determined by Hoescht dye staining with and without 24 h treatment with etoposide. LSsi and LSscram cells were imaged at 400× at random fields. N = 4 per sample. Mean ± SEM. *p < 0.002 vs. LSscram no treatment, p < 0.03 vs. LSsi + etoposide; **p < 0.008 vs. LSsi no treatment.

Fig. 6

(A) Apoptosis was induced in Lovo colon cells using anti-Fas. His-tagged MUC17-CRD1-L-CRD2 (HisMUC17wt) inhibits Fas-induced apoptosis in a dose-dependent fashion. N = 4 per sample, *p < 0.001 vs. Fas Only treatment. (B) His-tagged MUC17-CRD1-L-CRD2 (hisMUC17wt) as well as His-tagged MUC17-L-CRD2 without CRD1, indicated as trunMUC17, stimulated cell migration in Lovo colon cells, N = 6 per sample, *p < 0.01 vs. serum free (sf). EGF at 10 ng/ml was a positive control for anti-apoptosis and cell migration.

Fig. 7

Activation of ERK cell signaling pathways. (A) IEC6 cells were treated with control serum-free medium (SF) or E. coli MUC17-CRD1-L-CRD2His8 (10 μg/ml) for 5–30 min and lysates were blotted with anti-phosphoERK1. Cells pretreated with ERK inhibitor U0126 before 15 min exposure to the MUC17 protein resulted in no phosphoERK1. The lower molecular weight bands indicate coumassie-stained protein demonstrating equal loading of gel. (B) YAMC cells were treated with control serum-free medium (SF. Lane 1); E. coli MUC17-CRD1-L-CRD2His8 (10 μg/ml) for 5, 10, 20, 60 or 120 min (lanes 2–6); EGF (100 ng/ml) for 5 and 10 min (lanes 7 and 8); and BSA (10 μg/ml) for 5 and 10 min; and lysates were blotted with anti-phosphoERK1. (C) Cell migration of Lovo cells treated with EGF (20 ng/ml), 10% fetal calf serum (FCS), Muc3-CRD1-L-CRD2 (hM3ecoli; 10 μg/ml), or MUC17-CRD1-L-CRD2 (hM17ecoli; 10 μg/ml); with or without pretreatment with 10 μM ERK inhibitor U0126 (u). *p < 0.01 vs. SF; **p < 0.01 vs. SF and p < 0.04 vs. hM3ecoli+u; ***p < 0.005 vs. SF and p < 0.02 vs. hM17ecoli+u. ERK inhibition has no effect on EGF-induced cell migration, partially blocks migration induced by 10% fetal calf serum (FCS), and completely blocks migration induced by both Muc3 and MUC17-CRD proteins.

Fig. 8

Colitis induced by acetic acid. Mice received 5% acetic acid per rectum at time 0, then treated with PBS, GST control protein (100 μg/dose), or GST-tagged MUC17-CRD1-L-CRD2 (gM17ecoli); 100 μg/dose per rectum at time 12 and 24 h. Mice were euthanized at 30 h for histologic examination of the distal colon for (A) crypt damage score and (B) length of total (grade III) ulceration. *p < 0.056 vs. GST. (C) Survival experiment: mice were given 5% acetic acid per rectum at time 0, then treated with PBS buffer, GST control protein (100 μg/dose), or GST-tagged MUC-17 (gM17ecoli) protein at 200 μg/dose at 12 h and 24 h (note mice treated with 100 μg/dose showed similar survival in previous experiment). Mice were then euthanized at 30 h. In this experiment, 100% of mice treated with the MUC17 protein survived, compared to 78% treated with GST, and 55% treated with PBS (*p = 0.032).

Fig. 9

Colitis induced by dextran sodium sulfate. Mice received 5% DSS for 5 days and then received control buffer or hisMUC17-CRD1-L-CRD2 (hM17ecoli) (100 μg/dose) at 12, 24, and 36 h. Mice were harvested at 48 h and distal colons examined. (A) Crypt damage score. *p < 0.02 vs. control. (B) Total length of grade III ulceration. *p < 0.056 vs. control. (C) Representative histology of undamaged distal colon crypts in mouse treated with hisMUC17-CRD1-L-CRD2. (D) Representative histology of early grade III ulceration in distal colon of control mouse treated with buffer only.