Interaction of hepatocyte growth factor and non-steroidal anti-inflammatory drugs during gastric epithelial wound healing.

BACKGROUND/AIMS: Expression of the hepatocyte growth factor (HGF) and cyclooxygenase-2 (COX-2) is upregulated at the margins of healing gastric ulcers. We investigated in vitro the interference of HGF, the selective COX-2 inhibitor NS-398 and the nonselective COX inhibitor indomethacin with gastric epithelial wound healing and actin microfilament (actin-MF) formation. METHODS: Standardized gastric epithelial wounds, created in confluent RGM1 rat cell monolayers were treated with: HGF (10 ng/ml), NS-398 (1-100 microM) or indomethacin (0.01- 0.5 mM). The areas of re-epithelialization and cell proliferation were measured 24 h after wounding. Actin-MFs were labeled with fluorescein-conjugated phalloidin and their distribution was examined using a Nikon epifluorescence microscope. RESULTS: HGF caused a significant increase in gastric monolayer wound re-epithelialization and this was not affected by mitomycin C. Both indomethacin and NS-398 inhibited HGF-stimulated re-epithelialization, but the basal wound re-epithelialization rate and cell proliferation was only significantly inhibited by indomethacin. HGF triggered actin stress fiber formation which was inhibited by both indomethacin and NS-398, but only indomethacin interfered with actin-MF formation at the baseline condition. CONCLUSIONS: HGF significantly increased gastric wound re-epithelialization by activating cell migration which may be mediated by the COX-2 pathway.

Classification of ADRs: a proposal for harmonization and differentiation based on the experience of the Comprehensive Hospital Drug Monitoring Bern/St. Gallen, 1974-1993.

The present paper deals with the classification of adverse drug reactions (ADRs) according to today's largely accepted pathomechanisms. The classification system applied, relies primarily on the proposals of Rawlins and Thomson with type A ('augmented') and B ('bizarre') reactions. In the database of the Comprehensive Hospital Drug Monitoring (CHDM) Bern/St. Gallen on 48,005 consecutively hospitalized patients, ADRs had been attributed to 10 different pathomechanisms. These permit a versatile new system, easily adaptable to expanding knowledge. If we look at the 12,785 ADRs registered in the CHDM Bern/St. Gallen from 1974 to 1993, 76% were of type A, 13% of type B, and 11% of a pathomechanism not yet defined (type X). The main subgroups were A1 'not specified' in type A, Ba allergic/immunological and Bpa pseudoallergic/anaphylactoid in type B. Dose-related (A2) and drug-related reactions (A4, intolerance in a restricted sense), drug-to-drug interactions (A5), rebound/withdrawal effects (A6) and secondary reactions (A7) represented smaller subgroups. Patient-related reactions (A3, 'idiosyncrasy' in the strict sense) were not assessed. Today's algorithms for ADRs mainly rely on pharmacological, i.e. type A reactions. For most of the type B reactions adaptations including the experience of allergists, clinical immunologists and infectious disease specialists should be respected.