This paper presents the first solution structure of the RNase H domain of HIV-1 reverse transcriptase (RT) determined by NMR methods. The solution conditions in this study were at physiological pH in the presence of Mg(2+). An investigation of the dependence of the (1)H-(15)N HSQC spectrum of the RNase H domain on [Mg(2+)] indicates that Mg(2+) produces significant, global effects on the amide chemical shifts, implying that divalent metal ion binding is important for stabilizing the structure of the isolated domain in solution. Analysis of amide shift data as a function of MgCl(2) concentration using either a single- or two-site binding model indicated that the latter provided a significantly improved fit, with the K(D) for site A = 2.7-3.2 mM and K(D) for site B approximately 35 mM, calculated on the assumption that site A is already occupied. Resonances of the [U-(13)C,(15)N]RNase H domain, measured at pH 6.8, in 80 mM MgCl(2), were assigned and NOESY data collected in order to determine the structure. Assignment of the NOESY spectra using the ARIA program resulted in a high-resolution structure for residues 6-114 which was similar to the crystal structure of the isolated domain,. The data were insufficient to define a compact structure for the C-terminal residues after 114. Residues I134-L138 located at the C-terminus are highly disordered and give rise to relatively sharp and intense amide resonances, while the amide resonances for the segment from E124 to A132 appear to be largely absent and are presumably subject to significant exchange broadening between different conformational states. Comparisons with crystal structure data for the full reverse transcriptase molecule indicate that the corresponding region is absent in nearly all of the crystal structures determined for the P2(1)2(1)2(1) space group, while these residues adopt an alpha-helix in structures determined for other symmetry groups. This structural heterogeneity indicates that significant conformational variability exists for this segment of the full reverse transcriptase enzyme as well, and the structure of the C-terminal peptide can be selected or deselected, depending on crystallization conditions. This analysis, along with the structural characterization contained herein, challenges the previous paradigm that the dynamic behavior of the isolated RNase H domain differs substantially from the behavior in the intact enzyme. The poor Mg(2+) binding and conformational flexibility of residues located near the active site indicate that substrate binding is a precondition for metal ion binding and for selecting the active site conformation of the RNase H domain.

␤-Carbolines are tricyclic nitrogen heterocycles formed in plants and animals as Maillard reaction products between amino acids and reducing sugars or aldehydes. They are being detected increasingly in human tissues, and their physiological roles need to be understood. Two ␤-carboline carboxylates have been reported to accumulate in the human eye lens. We report here on the identification of another ␤-carboline, namely 1-methyl-1-vinyl -2,3,4-trihydro-␤-carboline-3-carboxylic acid, in the lenses of some cataract patients from India. Analysis of these three lenticular ␤-carbolines using photodynamic and antioxidant assays shows all of them to be inert as sensitizers and effective as antioxidants; they quench singlet oxygen, superoxide and hydroxyl radicals and inhibit the oxidative formation of higher molecular weight aggregates of the test protein, eye lens ␥-crystallin. Such antioxidative ability of ␤-carbolines is of particular relevance to the lens, which faces continual photic and oxidative stress. The ␤-carboline diacid IV is also seen to display an unexpected ability of inhibiting the thermal coagulation of ␥-crystallin and the dithiothreitol-induced precipitation of insulin. These results offer experimental support to earlier suggestions that one of the roles that the ␤-carbolines have is to offer protection against oxidative stress to the human tissues where they accumulate.

Rhodostomin (Rho) is a snake venom protein isolated from Calloselasma rhodostoma. Rho is a disintegrin that inhibits platelet aggregation by blocking the binding of fibrinogen to the integrin alpha(IIb)beta3 of platelets. Rho produced in Escherichia coli inhibited platelet aggregation with a K(I) value of 263 nM. Although functional, Rho produced in E. coli is misfolded based on our 2D and 3D NMR studies. In order to correct the folding problem, Rho was expressed in Pichia pastoris. The recombinant Rho expressed in P. pastoris inhibited platelet aggregation with a resulting K(I) value of 70 nM. This is the same potency as that of native Rho. CD analysis showed that the secondary structures of Rho are pH-independent and contain 3.5-7.9% alpha-helix, 48.2-50.5% beta-structures, and 42.3-47% coil. The sequential assignment and structure analysis of Rho were obtained using 2D and 3D 15N-edited NMR spectra. These results provide the first direct evidence that highly disulfide-bonded disintegrin can be expressed in P. pastoris with the correct fold. This evidence may serve as the basis for exploring the structure and function relationships as well as the dynamics of disintegrin and its variants.

Giganticine (1), a novel nonprotein amino acid, has been isolated from a methanol extract of the root bark of Calotropis gigantea and its structure established by spectroscopic methods. It exhibited a significant antifeedant activity against nymphs of the desert locust Schistocerca gregaria.