Pyrazinamide (PZA) is active against major Mycobacterium tuberculosis species ( M. tuberculosis , M. africanum , and M. microti ) but not against M. bovis and M. avium . The latter two are mycobacterial species involved in human and cattle tuberculosis and in HIV coinfections, respectively. PZA is a first-line agent for the treatment of human tuberculosis and requires activation by a mycobacterial pyrazinamidase to form the active metabolite pyrazinoic acid (POA). As a result of this mechanism, resistance to PZA, as is often found in tuberculosis patients, is caused by point mutations in pyrazinamidase. In previous work, we have shown that POA esters and amides synthesized in our laboratory were stable in plasma (M. F. Simões, E. Valente, M. J. Gómez, E. Anes, and L. Constantino, Eur J Pharm Sci 37:257–263, 2009, http://dx.doi.org/10.1016/j.ejps.2009.02.012 ). Although the amides did not present significant activity, the esters were active against sensitive mycobacteria at concentrations 5- to 10-fold lower than those of PZA. Here, we report that these POA derivatives possess antibacterial efficacy in vitro and ex vivo against several species and strains of Mycobacterium with natural or acquired resistance to PZA, including M. bovis and M. avium . Our results indicate that the resistance probably was overcome by cleavage of the prodrugs into POA and a long-chain alcohol. Although it is not possible to rule out that the esters have intrinsic activity per se , we bring evidence here that long-chain fatty alcohols possess a significant antimycobacterial effect against PZA-resistant species and strains and are not mere inactive promoieties. These findings may lead to candidate dual drugs having enhanced activity against both PZA-susceptible and PZA-resistant isolates and being suitable for clinical development.

One interesting approach to fight tuberculosis is the use of prodrugs that often have shown improved biological activities over drugs with poor absorption or difficulty to cross membranes. Previous studies demonstrate that weak acids such as benzoic acid, present antimycobacterial activity. Moreover, esters of those acids revealed to be a viable alternative since they may diffuse more easily through the cell membranes. Previously we showed that mycobacteria can easily activate benzoic acid esters by conversion to the corresponding acid. Since Zhang postulated that the activity of the acids can be dependent on their pKa, we set up to synthesize a library of benzoates with different electron withdrawing groups (4-chloro, 2,6-dichloro, 3,5-dichloro, 4-nitro, and 3,5 dinitro), to modulate pKa of the liberated acid and different alkoxy substituents (propyl, hexyl, and phenyl) to modulate their lipophilicity, and tested the activity of the esters and the corresponding free acids against mycobacteria. We also studied the activation of the esters by mycobacterial enzymes and the stability of the compounds in buffer and plasma. We concluded that all the benzoates in our study can be activated by mycobacterial enzymes and that the phenyl and hexyl esters presented higher activity than the corresponding free acids, with the nitrobenzoates, and especially the dinitrobenzoates, showing very interesting antitubercular activity that deserve further exploration. Our results did not show a correlation between the activity and the pKa of the acids.