Experimental and dietary studies have shown that omega-3 polyunsaturated fatty acids (ω-3 PUFAs) such as docosahexaenoic acid (DHA; C22:6 ù-3) and eicosapentaenoic acid (EPA; C20:5 ù-3) inhibit the development of colorectal, breast, and prostate cancers, while ù-6 PUFAs stimulate tumourigenesis, but underlying mechanisms are not clear. The cytochrome P450-derived 17,18-epoxide of EPA (ω-3- 17,18-epoxy-EPA) and its saturated monoepoxide analogues have been found to impair tumour cell growth and activate apoptosis in triple-negative MDA-MB-231 breast cancer cells. ù-3-17,18-epoxy-EPA decreased cell viability by targeting and down- regulating cyclin D1 through the activation of the p38 mitogen-activated protein (MAP) kinase. As the epoxide was prone to degradation and was not stable in vivo, bioisosteric urea analogues of greater stability were synthesised with further structural modifications. In earlier work, a stable mimic of ù-3-17,18-EPA a chloro- trifluoromethyl urea-substituted fatty acid termed 29 was prepared that rapidly killed triple-negative MDA-MB-231 cancer cells in vitro and in vivo in mouse xenograft models.
The present study evaluated the influence of structural modifications of new 29 congeners on anti-proliferative and pro-apoptotic potency. Several regions for modifications were identified in the structure of 29. Initial modification focused on isosteric replacement of the carboxylic acid functional group in 29. Replacement of the carboxylic acid moiety in 29 with phosphonic acid, oxo-thiadiazole, hydroxamic acid, sulfonic acid, amide and methyl amide produced additional analogues that were evaluated for the inhibition of proliferation of triple-negative MDA-MB-231 breast cancer cells. The oxo-thiadiazole, hydroxamic and sulfonic bioisosteres decreased ATP
production and, in addition, disrupted adherence in crystal violet assay. In contrast, the phosphonic, amide or methyl amide isosteres did not alter ATP production relative to vehicle control. Consistent with these findings, the oxo-thiadiazole, and hydroxamic acid analogues interrupted MDA-MB-231 cell cycle progression in G0/G1 phase, with cells treated with the sulfonic acid isostere arrested in G2/M phase; thus, treated cells were unable to complete the cell cycle. The results indicate that the hydroxamic acid, oxo-thiadiazole and sulfonic acid isosteres inhibit MDA-MB-231 cells proliferation by decreasing cell confluence, energy metabolism, and halting cell cycle progression. In addition to anti-proliferative activities, the phosphonic acid (40 µM), oxo-thiadiazole, hydroxamic acid, and sulfonic acid bioisosteres (at 10 µM) were pro-apoptotic, significantly increasing caspase-3 activity and Annexin V/7AAD staining. An early feature of apoptosis is a loss of membrane asymmetry and externalization of the membrane phospholipid phosphatidylserine so that it is available to bind with Annexin
V. In late apoptosis, dead cells are permeable to 7AAD. Thus, increased Annexin V/7AAD staining by carboxylate isosteres of 29 is in agreement with an increase in tumour cell killing by the dual mechanisms of apoptosis and necrosis. Taken together, the carboxylic acid functional group can be replaced with certain bioisosteric groups to retain the anti-proliferative and pro-apoptotic activity of 29 in triple-negative MDA- MB-231 breast cancer cells.
The distribution coefficient (log D) is a useful parameter for an understanding of a drug candidate behaviour. The log Ds of carboxylic acid isosteres of 29 were calculated. Even though the log Ds of hydroxamic acid and oxo-thiadiazole isosteres were higher than 29, they appear to be promising compounds. This suggests that, although the lipophilicity of drug candidate is an important property, geometry, and other physicochemical properties of the candidate may be important additional considerations in biological activity.
In the second part of this study, the nature of the aromatic system in 29 was modified to produce new analogues and identify factors that optimise anti-proliferative activity. Additional aryl ureas with bicyclic, heteroaromatic and electron- donating/withdrawing substituents were synthesised (NK1 – NK26). The new analogues were evaluated for anti-proliferative activity in triple-negative MDA-MB- 231 breast cancer cells. In particular, four di-substituted analogues that carried strongly electron-withdrawing groups: (termed NK14, NK17, NK18, and NK24) decreased adherence and impaired ATP production in MDA-MB-231 breast cancer cells. In contrast, other analogues that carried bulky aromatic, heteroaromatic and weakly electron-withdrawing groups were inactive. MDA-MB-231 cells treated with NK14, NK17 and NK24, showed altered cell cycle distribution relative to vehicle control. NK14 and NK24 significantly increased the population of cells in the sub-G1 phase (DNA fragmentation) and decreased the proportion of cells in G0/G1 phase, which is consistent with the apparent pause in cell cycle progression and the failure of treated cells to complete the cell cycle. Consistent with these findings, NK14 and NK24 isosteres significantly increased caspase-3 activity and the proportion of cells that were stained with Annexin V/7AAD; therefore the analogues induced apoptotic and necrotic cell death. To rationalise the above findings further, physicochemical parameters of NK14, NK17, NK18 and NK24 analogues including electronic (σ), hydrophobic (π) and steric (MR) parameters were calculated. It was noted that the comparable anti- proliferative and pro-apoptotic potency observed for 29 (Hammet sum σ ≥ 0.66) and NK14 (σ ≥ 0.66) could be related to similar molecular and electronic distributions. However, to establish the structural requirements for activity of 29, NK14 and NK24, future studies should systematically assess the potential importance of substitution patterns in these analogues. In conclusion, 29 and several other ω-3-17,18-epoxy-EPA mimics show promise as potential anti-tumour agents. Further studies are required to optimise the development of these analogues in cancer chemotherapy.