Bortezomib warhead-switch confers dual activity against mycobacterial caseinolytic protease and proteasome and selectivity against human proteasome

Abstract

Mycobacteria harbor two main degradative proteolytic machineries, the caseinolytic protease ClpP1P2 and a proteasome. We recently showed that Bortezomib inhibits ClpP1P2 and exhibits whole cell activity against Mycobacterium tuberculosis. Bortezomib, a dipeptide with a boronic acid warhead, is a human proteasome inhibitor approved for cancer therapy. The boronic acid warhead of the compound has been shown to drive potency against both the human proteasome and ClpP1P2 protease. Selectivity for the bacterial ClpP1P2 protease over the human proteasome is lacking but needs to be achieved to move this new anti-tuberculosis lead forward. In this study we explored whether an alternative warhead could influence Bortezomib's selectivity. We synthesized an analog containing a chloromethyl ketone instead of the boronic acid warhead and determined potencies against the bacterial and human enzymes. Surprisingly, the analog retained activity against mycobacterial ClpP1P2 and was active against the mycobacterial proteasome, but was devoid of activity against the human proteasome. Interrogation of a set of chloromethyl ketone peptides identified three additional compounds similarly inhibiting both ClpP1P2 and the proteasome in the bacteria while leaving the human proteasome untouched. Finally, we showed that these compounds display bactericidal activity against M. tuberculosis with cytotoxicity ranging from acceptable to undetectable. These results suggest that selectivity over the human proteasome is achievable. Selectivity, together with dual-targeting of mycobacterial ClpP1P2 and proteasome makes this new scaffold an attractive starting point for optimization.