Susceptibilities of Mycobacterium tuberculosis and Mycobacterium avium complex to lipophilic deazapteridine derivatives, inhibitors of dihydrofolate reductase.

Twelve lipophilic 2,4-diamino-5-methyl-5-deazapteridine derivatives and trimethoprim were evaluated for activity against Mycobacterium tuberculosis and Mycobacterium avium in vitro. Six of the compounds had MICs of < or =12.8 mg/L and < or =1.28 mg/L against M. tuberculosis and M. avium, respectively; trimethoprim MICs were >128 mg/L and >12.8 but < or =128 mg/L, respectively. Two compounds, with either a 2-methyl-5-methoxy phenyl or 2-methoxy-5-trifluoromethyl phenyl linked at the 6-position of the deazapteridine moiety by a CH2NH bridge, had MICs of < or =0.13 mg/L against M. avium; the two compounds also had apparent I50 values for dihydrofolate reductase of 2 and 8 nM, respectively, compared with an I50 of 400 nM with trimethoprim. Four of the compounds were selectively toxic to mycobacteria as compared with Vero cells. These results demonstrated that lipophilic antifolates can be synthesized which are more active against mycobacteria than trimethoprim and which possess selective toxicity.

Homologated aza analogs of arabinose as antimycobacterial agents.

A series of hydrolytically-stable aza analogs of arabinofuranose was prepared and evaluated against Mycobacterium tuberculosis and M. avium. The compounds were designed to mimic the putative arabinose donor involved in biogenesis of the essential cell wall polysaccharide, arabinogalactan. Though most compounds displayed little activity in cell culture, one compound showed significant activity in infected macrophage models.