Effect of local application of growth factors on gastric ulcer healing and mucosal expression of cyclooxygenase-1 and -2

Abstract

Background/aims: Ulcer healing involves expression of various growth factors such as epidermal growth factor (EGF), hepatocyte growth factor (HGF) and basic fibroblast growth factor (bFGF) at the ulcer margin, but the influence of EGF, HGF and bFGF applied locally with or without neutralizing anti-EGF, HGF and bFGF antibodies or cyclooxygenase (COX)-1 and COX-2 inhibitors on ulcer healing and the expression of COX-1 and COX-2 during ulcer healing have only been studied a little.

Methods: Rats with gastric ulcers induced by serosal application of acetic acid (ulcer area 28 mm2 received a submucosal injection of either (1) vehicle (saline), (2) EGF, (3) HGF, and (4) bFGF with or without antibodies against EGF, HGF and bFGF or indomethacin (2 mg/kg/day i.p.), a nonspecific inhibitor of COX, or NS-398 (10 mg/kg/day i.g.) and Vioxx (5 mg/kg/day i.g.), both highly specific COX-2 inhibitors. A separate group of animals with chronic gastric fistulas was also used to assess gastric secretion during ulcer healing with and without growth factors. Each growth factor and specific antibody against EGF, HGF and bFGF (100 ng/100 microl each) were injected just around the ulcer immediately after ulcer induction and this local injection was repeated on day 2 following anesthesia and laparotomy. On days 13 and 21, the ulcer area was determined by planimetry, gastric blood flow (GBF) at the ulcer margin was examined by the H2-gas clearance technique, and mucosal generation of PGE2 and the gene expression of COX-1 and COX-2 in the non-ulcerated and ulcerated gastric mucosa were assessed. Gastric ulcers healed progressively within 21 days after induction and this effect was accompanied by a significant increase in GBF at the ulcer margin and in the expression of COX-2 in the ulcer area. Local treatment with EGF, HGF and bFGF produced a significant decrease in gastric acid secretion and significantly accelerated the rate of ulcer healing and raised GBF at the ulcer margin causing further significant upregulation of COX-2 but not COX-1 expression in the ulcerated mucosa. The acceleration of ulcer healing and hyperemia at the ulcer margin exhibited by locally applied EGF, HGF and bFGF were similar to those obtained with systemic administration of these growth factors. HGF applied submucosally, upregulated COX-2 expression and this was significantly attenuated by concurrent treatment with antibody against this peptide. Anti-EGF and anti-bFGF antibodies completely abolished the acceleration of the ulcer healing and hyperemia at the ulcer margin induced by these growth factors. Indomethacin and both COX-2 inhibitors significantly prolonged ulcer healing, while suppressing the generation of PGE2 in non-ulcerated and ulcerated gastric mucosa and GBF at the ulcer margin. The acceleration of ulcer healing by EGF, HGF and bFGF and the accompanying rise in GBF at the ulcer margin were significantly attenuated by the concurrent treatment with indomethacin or NS-398 and Vioxx.

Conclusions: (1) Growth factors accelerate ulcer healing due to enhancement in the microcirculation around the ulcer and these effects are specific because they can be abolished by neutralization with antibodies; (2) COX-2-derived prostaglandins and suppression of gastric secretion may play an important role in the acceleration of ulcer healing by various growth factors, and (3) the local effects of EGF, HGF and bFGF on ulcer healing can be reproduced by their systemic application indicating the high efficacy of growth factors to accelerate this healing.