Binding site of amiloride to urokinase plasminogen activator depends on species

A novel drug candidate is checked on its potency on animal models before it can advance to human phase of the research. Usually negative results on animal phase disqualify it. Targeting specific enzymes by small chemicals raises the question about the appropriateness of this approach. As an example, the urokinase (uPA) is recognized as an important enzyme responsible for cancer metastasis and angiogenesis. It is therefore important to ask the question if a small chemical will inhibit uPA of different species with the same or different potency. Using DNA sequence and known structure of uPA we have modeled 3D structures of uPAs for several different species. By theoretical calculations we have determined most probable structure of amiloride/uPAs complexes. Catalytic triad (B57, B102, B195) and specificity pocket (B187-B197, B212-B229) are highly conserved in all cases, and are the regions responsible for proteolytic activity and recognition of the substrate. Significant differences were observed in a different region (loop B93-B101), that we identified as binding site of amiloride to the tissue plasminogen activator (tPA). Although tPA shares the same function of activating plasminogen and it is structurally similar to uPA. Amiloride is a specific inhibitor of uPA but does not inhibit tPA. Our study shows that predicted position of amiloride depends on species and in some cases was located, as expected, in the specificity pocket, but in the other cases close to the loop B93-B101. This location could weaken affinity of binding or prevent inhibition of uPA. Therefore, drug screening and elimination process based solely on animal study, without careful structural analysis, could lead to the elimination of potential drugs for humans.