Gastric Mucosal Protection­ Ulcer healing

The area of identifying mechanisms of protecting the gastric mucosa against injury and encouraging the healing of areas that are injured has been of interest in the department of surgery for the past number of years. This research project continues this study of gastric mucosal defence dealing specifically with NSAID related side effects and their minimisations.

Nonsteroidal anti-inflammatory drugs and ulcer healing

NSAIDs are aspirin like drugs that are used widely within the hospital and within the community. They delay the healing of gastric ulcers (Fig 1) and this delay is associated with complications such as bleeding and in some cases death. This project studied an important new group of NSAIDs which are potentially less toxic to the gastric mucosa. The COX-2 inhibitors are a specific group of NSAIDs which block the isoform of the enzyme cyclooxygenase which is involved in inflammation but they do not block the constitutive cyclooxygenase enzyme which mediates the process of gastric mucosal defence. Our studies have shown that L745,337 (selective COX-2 inhibitor) blocks the acute response phase to acetic acid injury but has no effect on the healing phase of repair. Indomethacin (predominant COX-1 inhibitor) however, interferes with both the acute response phase and healing phase of repair through influence on both wound contraction and mucosal regeneration. Results of this study may lead to a better understanding of the mechanisms of delayed gastric ulcer healing induced by NSAIDs.

Figure 1 - Haematoxylin and eosin stained histological section of a gastric ulcer

Potential mechanisms of NSAIDs reduced effect on mucosal regeneration
A/ Rate of apoptosis in the regenerative mucosa

NSAIDs may reduce mucosal regeneration through stimulating a process of programmed cell death (apoptosis) or by a direct toxic effect (cell necrosis) on the newly proliferating mucosal cells. In vitro sudies have shown NSAIDs to induce apoptosis in various cell lines. It is unknown however, if NSAIDs have an effect on apoptosis in the regenerative mucosa in vivo . We have been able to demonstrate using both TUNEL method and histological assessment on haematoxylin & eosin (H&E) stained sections that treatment with indomethacin and L745,337 has not led to a change in the rate of apoptosis in the regenerative mucosa. Therefore, the reduction in mucosal regeneration after indomethacin treatment is not due to a change in apoptosis.

B/ Effect of NSAIDs on cell migration of rat gastric mucosal (RGM1) cells

NSAIDs may also influence mucosal regeneration by inhibiting the migration of existing or newly proliferating cells from migrating across the granulation tissue to cover the mucosal defect. There is some evidence to suggest that NSAIDs may inhibit the migration of wounded gastric monolayers. The effect of NSAIDs on the restorative capacity of the cell themselves solely due to cell movement remains unknown. We have clearly demonstrated both indomethacin and L745,337 inhibited cell migration of RGM1 cells in a concentration dependent manner. Indomethacin and L745,337 delayed ulcer healing even when proliferation was blocked with mitomycin C. This data indicates that COX-1 and COX-2 participate in gastric epithelial cell migration which is important in epithelialisation of the gastric mucosa during ulcer healing. Furthermore the inhibitory effect of NSAIDs in this model is independent of proliferation.

Contraction and NSAIDs

NSAIDs delay ulcer healing partly by inhibiting wound contraction of the ulcer base through an unknown mechanism.Wound contraction is mediated by myofibroblasts which are found throughout the granulation tissue of the ulcer base. Myofibroblasts are modified fibroblasts which have actin filaments aligned parallel to the long axis of the cell and cause the contraction of the entire cell. They are connected to the matrix through the fibronexus which is attached intracellularly to the actin fibers and extracellularly to the matrix. The contractile ability of the microfilaments along with their adherence to one another and the surrounding matrix leads to shrinkage of the tissue. The effect of NSAIDs on wound contraction and myofibroblasts in the rat model remains has not been investigated until now. We found that myofibroblast density in both indomethacin and L745,337 treated tissue was high even though indomethacin inhibits both ulcer healing and contraction and L745,337 has no effect (Fig 2). This suggests that the myofibroblasts present in the indomethacin group are disfunctional. Hence, a major effect of indomethacin on the repair phase is blocking myofibroblast function, whereas L745,337 does not interfere with myofibroblast function.

Figure 2 - Myofibroblast staining in the ulcer base of after two weeks of dosing with NSAIDs

Ageing and ulcer healing

The aged appear to be particularly at risk of developing gastric complications as a result of NSAID therapy. We have investigated the effect of age on natural healing and delayed ulcer healing induced by NSAIDs, both COX-1 and COX-2 inhibitors in a rat .We found there was no age-related differences in the development of gastric ulcers and in natural ulcer healing in rats. However, there was an age-related difference in delayed healing with indomethacin but not with the L745,337 treated animals. The aged mucosa has been found to express significantly less COX-1 mRNA than the young, which may explain the reduction in delayed healing seen in the aged.

Research group

• Ms Debbie Mantzaris
• Prof. Paul E O'Brien