The ubiquitous glyoxalase system converts toxic alpha-keto aldehydes into their corresponding nontoxic 2-hydroxycarboxylic acids, utilizing glutathione (GSH) as a cofactor. The first enzyme in this system, glyoxalase I (GlxI), catalyzes the isomerization of the hemithioacetal formed nonenzymatically between GSH and cytotoxic alpha-keto aldehydes. To study the Escherichia coli GlxI enzyme, the DNA encoding this protein, gloA, was isolated and incorporated into the plasmid pTTQ18. Nucleotide sequencing of the gloA gene predicted a polypeptide of 135 amino acids and Mr of 14 919. The gloA gene has been overexpressed in E. coli and shown to encode for GlxI. An effective two-step purification protocol was developed, yielding 150-200 mg of homogeneous protein per liter of culture. Electrospray mass spectrometry confirmed the monomeric weight of the purified protein, while gel filtration analysis indicated GlxI to be a homodimer of 30 kDa. Zinc, the natural metal ion found in the Homo sapiens and Saccharomyces cerevisiae GlxI, had no effect on the activity of E. coli GlxI. In contrast, the addition of NiCl2 to the growth medium or to purified E. coli apo-GlxI greatly enhanced the enzymatic activity. Inductively coupled plasma and atomic absorption analyses indicated binding of only one nickel ion per dimeric enzyme, suggesting only one functional active site in this homodimeric enzyme. In addition, the apoprotein regained maximal activity with one molar equivalence of nickel chloride, indicative of tight metal binding. The effects of pH on the kinetics of the nickel-activated enzyme were also studied. This is the first example of a non-zinc activated GlxI whose maximal activation is seen with Ni2+.

HIV-1 reverse transcriptase (RT) is multifunctional, with RNA-dependent DNA polymerase (RDDP), DNA-dependent DNA polymerase (DDDP), and ribonuclease H (RNase H) activities. N-(4-tert-Butylbenzoyl)-2-hydroxy-1-naphthaldehyde hydrazone (BBNH) inhibited both the polymerase and the RNase H activities of HIV-1 RT in vitro. IC50 values for inhibition of RDDP were 0.8-3.4 microM, depending on the template/primer (T/P) used in the assay. The IC50 for DDDP inhibition was about 12 microM, while that for inhibition of RNase H was 3.5 microM. EC50 for inhibition of HIV-1 replication in cord blood mononuclear cells was 1.5 microM. BBNH inhibition of RNase H in vitro was time-dependent, whereas inhibition of RT polymerase activities was immediate. BBNH was a linear mixed-type inhibitor of RT RDDP activity with respect to both T/P and to dNTP, whereas BBNH inhibition of RT RNase H activity was linear competitive. Protection experiments using an azidonevirapine photolabel showed that BBNH binds to the non-nucleoside RT inhibitor (NNRTI) binding pocket. Importantly, the compound inhibited recombinant RT containing mutations associated with high-level resistance to other NNRTI. While BBNH did not inhibit the DNA polymerase activities of other retroviral reverse transcriptases and DNA polymerases, the compound inhibited Escherichia coli RNase HI and the RNase H activity of murine leukemia virus RT. BBNH also inhibited HIV-1 RT RNase H in the presence of high concentrations of other non-nucleoside inhibitors with higher affinities for the NNRTI binding pocket, and of RT in which the NNRTI binding pocket had been irreversibly blocked by the azidonevirapine photolabel. We conclude that BBNH may therefore bind to two sites on HIV-1 RT. One site is the polymerase non-nucleoside inhibitor binding site and the second may be located in the RNase H domain. BBNH is therefore a promising lead compound for the development of multisite inhibitors of HIV-1 RT.

The thiocarboxanilide nonnucleoside inhibitor (NNI) UC781 inhibited HIV-1 reverse transcriptase (RT) DNA polymerase activity at a 1:1 molar ratio of inhibitor to enzyme. Inhibition was linear uncompetitive with respect to template/primer (T/P) and mixed noncompetitive with respect to deoxynucleoside triphosphate (dNTP), typical of NNI. When the RT-T/P binary complex was incubated with UC781 and then separated from unbound inhibitor, recovery of enzyme activity was slow, with only about 60% activity recovered after 25 min. The inactivation of the RT-T/P complex was prevented by the presence of a large excess of UC84, another carboxanilide NNI that interacts with this RT mechanistic form. UC781 protected the RT-T/P-dNTP ternary complex from irreversible inactivation by a photoactivatable azido analog of nevirapine, implying that UC781 binds to the NNI pocket of this RT mechanistic form. UC781 did not photoprotect either the free enzyme or the RT-T/P binary complex; however, protein fluorescence quenching studies indicated that UC781 interacted with all RT mechanistic forms, with the order of affinity being RT-T/P-dNTP ternary complex > RT-T/P binary complex > free RT. Reaction progress curve analysis showed that the binding of UC781 to RT is rapid (k(on) approximately 1.7 x 10(6) M(-1) s(-1)), but that dissociation is slow (k(off) approximately 1.6 x 10(-3) s(-1)). UC781 is therefore a rapid tight-binding inhibitor of HIV-1 RT, the first NNI to demonstrate this property.