Enhanced wound healing by recombinant Escherichia coli Nissle 1917 via human epidermal growth factor receptor in human intestinal epithelial cells: therapeutic implication using recombinant probiotics

Abstract

The gastrointestinal mucosa has a remarkable ability to repair damage with the support of epidermal growth factor (EGF), which stimulates epithelial migration and proliferative reepithelialization. For the treatment of mucosal injuries, it is important to develop efficient methods for the localized delivery of mucoactive biotherapeutics. The basic idea in the present study came from the assumption that an intestinal probiotic vehicle can carry and deliver key recombinant medicinal proteins to the injured epithelial target in patients with intestinal ulcerative diseases, including inflammatory bowel disease. The study was focused on the use of the safe probiotic E. coli Nissle 1917, which was constructed to secrete human EGF in conjunction with the lipase ABC transporter recognition domain (LARD). Using the in vitro physically wounded monolayer model, ABC transporter-mediated EGF secretion by probiotic E. coli Nissle 1917 was demonstrated to enhance the wound-healing migration of human enterocytes. Moreover, the epithelial wound closure was dependent on EGF receptor-linked activation, which exclusively involved the subsequent signaling pathway of the mitogen-activated protein kinase kinase (MEK) extracellular-related kinases 1 and 2 (ERK1/2). In particular, the migrating frontier of the wounded edge displayed the strongest EGF receptor-linked signaling activation in the presence of the recombinant probiotic. The present study provides a basis for the clinical application of human recombinant biotherapeutics via an efficient, safe probiotic vehicle.

Fig 1

Expression of human EGF in E. coli Nissle 1917. (A) The recombinant EGF-LARD3 gene was expressed in E. coli Nissle 1917 after 1 mM IPTG treatment in LB broth; the released product was detected using immunoprecipitation with anti-EGF or anti-LARD3 antibody. Each recombinant bacterium was designated Nissle (the wild type of E. coli Nissle 1917), Nissle-AB (E. coli Nissle 1917 with the EGF expression plasmid and pACYC-184), or Nissle-AC (E. coli Nissle 1917 with the EGF expression plasmid and PrtDEF transporter). (B) The PrtDEF transporter gene was introduced into E. coli XL1-Blue expressing LARD3-linked EGF. After 1 mM IPTG treatment, the bacterial lysate and supernatant were assessed by protein analysis. (C) The bacterial culture supernatants were administered to human intestinal epithelial cells (HCT-8) for up to 20 min, and the cellular lysate was subjected to immunoprecipitation with either anti-EGFR or anti-phospho-EGFR antibody for the monitoring of biological activity.

Fig 2

Effect of recombinant EGF-secreting bacteria on epithelial wound healing. The y axis indicates relative migratory activity as described in Materials and Methods. The HCT-8 or IEC-18 cellular monolayer was starved overnight and physically wounded. (A) Control supernatant (RPMI 1640 medium without FBS) and each bacterial (Nissle, Nissle-AB, or Nissle-AC) culture supernatant were administered to wounded epithelial cells, and the relative length of epithelial migration was measured at each time point. Values on the y axis indicate fold change above the migration length of the control after 24 h. Values are means ± standard errors of the means (SEM) (n = 10 to 14). Bars with different letters are significantly different from each other (P < 0.05). Pairwise comparison was performed using post hoc ANOVA SNK methods in each experiment with different epithelial cell lines. (B) The experimental procedure for HCT-8 cells was performed using another general nonpathogenic E. coli strain, designated XL1 (the wild type of E. coli XL1-Blue), XL1-AB (E. coli XL1-Blue with the EGF expression plasmid and pACYC-184), or XL1-AC (E. coli XL1-Blue with the EGF expression plasmid and PrtDEF transporter). Values are means ± SEM (n = 10 to 14). Bars with different letters are significantly different from each other (P < 0.05). Pairwise comparison was performed using post hoc ANOVA SNK methods.

Fig 3

Effect of purified EGF on epithelial wound healing. (A) EGF was purified using the immunoaffinity method and detected as described in Materials and Methods. (B) The bacterial culture supernatants or purified EGF were administered to wounded monolayers of human intestinal epithelial cells (HCT-8) for 20 min, and the cellular lysate was subjected to immunoprecipitation with either anti-EGFR or anti-phospho-EGFR antibody for the monitoring of biological activity. (C) The y axis indicates relative migratory activity as depicted in Materials and Methods. The HCT-8 cellular monolayer was starved overnight and physically wounded. Control (RPMI 1640 medium without FBS), recombinant EGF, and purified EGF from Nissle-AC was administered to wounded epithelial cells, and the relative length of epithelial migration was measured at each time point. Values are means ± SEM (n = 10 to 14). An asterisk represents a significantly different result (P < 0.05) compared to that of each control group at each time point. For comparative analyses of two groups of data, Student's t test was performed.

Fig 4

Involvement of EGF receptor-linked signals in wound healing. (A) Serum-starved intestinal epithelial cells were pretreated with vehicle (DMSO) or 10 μM AG1478 and then stimulated with control supernatant (Con; RPMI 1640 medium without FBS) and each bacterial (Nissle, Nissle-AB, or Nissle-AC) supernatant. The total epithelial cellular lysate was subjected to Western blot analysis. (B and C) Serum-starved wounded intestinal epithelial monolayer cells (HCT-8) (B) or IEC-18 (C) were treated with control and Nissle-AC supernatants in the presence of vehicle (DMSO) or each inhibitor (10 μM AG1478, 5 μM LY294002, and 2 μM U0126). Relative wound closure at 48 h was measured, and the representative pictures (×100 magnification) are shown on the right. Values are means ± SEM (n = 10 to 14). An asterisk indicates significant differences from each vehicle group (P < 0.05). For the comparative analysis of two groups of data, Student's t test was performed.

Fig 5

Visualization of epithelial signaling activation by recombinant bacteria. (A) Control supernatant (RPMI 1640 medium without FBS), each bacterial (Nissle, Nissle-AB, or Nissle-AC) supernatant, or commercially available purified EGF was administered to wounded epithelial cells for 40 min. Phosphorylated ERK1/2 (red) and phosphorylated EGF receptor (green) was determined by immunofluorescence confocal microscopy. DIC, differential interference contrast. (B and C) The phosphorylation signals were quantified as described in Materials and Methods. Values are means ± SEM (n = 10 to 14). An asterisk indicates significant differences between each control group at each zone (P < 0.05). For comparative analyses of two groups of data, Student's t test was performed.