Identification of aspirin analogues that repress NF-κB signalling and demonstrate anti-proliferative activity towards colorectal cancer in vitro and in vivo.

Substantial evidence indicates that aspirin and related non-steroidal anti-inflammatory drugs (NSAIDs) have potential as chemopreventative/therapeutic agents. However, these agents cannot be universally recommended for prevention purposes due to their potential side-effect profiles. Here, we compared the growth inhibitory and mechanistic activity of aspirin to two novel analogues, diaspirin (DiA) and fumaryl diaspirin (F-DiA). We found that the aspirin analogues inhibited cell proliferation and induced apoptosis of colorectal cancer cells at significantly lower doses than aspirin. Similar to aspirin, we found that an early response to the analogues was a reduction in levels of cyclin D1 and stimulation of the NF-κB pathway. This stimulation was associated with a significant reduction in basal levels of NF-κB transcriptional activity, in keeping with previous data for aspirin. However, in contrast to aspirin, DiA and F-DiA activity was not associated with nucleolar accumulation of RelA. For all assays, F-DiA had a more rapid and significant effect than DiA, identifying this agent as particularly active against colorectal cancer. Using a syngeneic colorectal tumour model in mice, we found that, while both agents significantly inhibited tumour growth in vivo, this effect was particularly pronounced for F-DiA. These data identify two compounds that are active against colorectal cancer in vitro and in vivo. They also identify a potential mechanism of action of these agents and shed light on the chemical structures that may be important for the antitumour effects of aspirin.

Activity of aspirin analogues and vanillin in a human colorectal cancer cell line.

Colorectal cancer is the third most common cancer and is associated with significant morbidity and mortality. Epidemiological and animal studies indicate that regular acetylsalicylic acid (aspirin) intake is associated with a reduction in the incidence of colorectal cancer. Acetylsalicylic acid (ASA) has also been shown to inhibit colorectal cancer cell proliferation in vitro. The molecular basis for this specific cytotoxicity is an area of considerable debate. To investigate the toxicity of salicylates, the sensitivity of the DNA mismatch repair proficient SW480 human colorectal cancer cell line to four categories of compounds with varying degrees of structural similarity to acetylsalicylic acid was tested. These compounds were: i) salicylic acid analogues with substituents at the 5-position; ii) ASA analogues with extended chain lengths in the acyl group; iii) vanillin (4-hydroxy-3-methoxybenzaldehyde; and iv) bis(2-carboxyphenyl) succinate (BCS) and structurally similar derivatives thereof. It was found that compounds with amino and acetamido substituents at the salicylate 5-position were less toxic than ASA itself. Modifications to the length of the hydrocarbon chain in the acyl groups of ASA analogues also marginally reduced toxicity. Vanillin exhibited relatively limited toxicity against the SW480 colorectal cancer cell line. Commercially available and in-house synthesised BCS (diaspirin) were notably more inhibitory to cell growth than ASA itself, yet retained substantial specificity against colorectal cancer cell lines vs. non-colorectal cancer cell lines. BCS and ASA were toxic to SW480 cells through initiation of necrotic and apoptotic pathways. Fumaroyldiaspirin and benzoylaspirin exhibited greater toxicity than ASA against the SW480 cell line. A novel method for synthesis of BCS, a compound that has erratic commercial availability, is described. We propose that the anti-inflammatory and anticancer capacity of BCS and the other analogues described herein is worthy of investigation.