The ability of Mycobacterium tuberculosis (MTB) to establish latent infection poses significant problems in the treatment of TB, but a team of scientists led by researchers at Weill Cornell Medical College have now discovered compounds that are able to kill the bacterium in its dormant state. Unlike most other antibacterials, which inhibit the synthesis of bacterial proteins, the newly identified compounds inhibit protein degradation. 
Oxathiazol-2-ones, such as GLR5 and HT1171, were shown to selectively and irreversibly inhibit the proteasome of MTB, whilst having little effect on the human proteasome and showing no apparent toxicity to mammalian cells. The compounds were further shown to act as suicide-substrates which cyclocarbonylate the active site threonine of the bacterial proteasome. X-ray crystallographic studies revealed major conformational changes that protect the inhibitor-enzyme intermediate from hydrolysis, allowing formation of an oxazolidin-2-one and preventing regeneration of active protease. Some of the many amino acid residues involved in the conformational changes are remote from the active site pocket and are different from those in the human proteasome, which may account for the selectivity of oxathiazol-2-ones for the bacterial system.
The study was published online on September 16th in the journal Nature.
Its nice to see this interesting info. As for as my knowledge goes Southern Research Institute, Birmingham, has done lots of testings (High Throughput Screening) way back in 1998-99 and they did collect many new compounds from different universities & Institutes. I think they have a lots of data (1,2,4-triazoles, 1,3,4-oxadiazoles, 1,3,4-thiadiazoles & many other derivatives).