Non-steroidal anti-inflammatory drugs (NSAIDs) inhibit the proliferation of colorectal carcinoma cell lines in vitro and reduce the risk of colorectal carcinoma in vivo. The good correlation observed between the potency of NSAIDs as inhibitors of colorectal carcinoma cell proliferation and as antagonists of a 78 kDa gastrin binding protein (GBP) suggested that blockade of the GBP might contribute to the anti-proliferative effects of NSAIDs [G.S. Baldwin, V.J. Murphy, Z. Yang, T. Hashimoto, J. Pharmacol. Exp. Ther. 286 (1998) 1110-1114]. The most potent NSAID investigated was sulindac sulphide, which had an IC50 value of 40 microM. In order to investigate the structural requirements for binding to the GBP, 26 analogues of sulindac sulphide and sulindac sulphoxide were tested for their ability to inhibit the binding of iodinated gastrin to the GBP. Six of the analogues inhibited gastrin binding by more than 50% at a concentration of 1 mM. The IC50 values estimated by computer fitting of titration data were in the range of 280-940 microM. Comparison of the analogue structures suggests that a substituent with a carboxyl group is preferred in the R2 position. In addition the location of the NSAID binding site within the GBP structure was investigated. NSAIDs bound to both the N- and C-terminal halves of the GBP, and the affinities determined were similar to the values previously reported for the full-length GBP. The results reported herein represent the first step in the rational design of more potent GBP antagonists, some of which may be useful for the treatment of colorectal carcinoma.