Structural requirements for the action of steroids as quenchers of albumin fluorescence.

1. Androgens, corticoids, gestagens, estrogens and related steroids are effective quenchers of the intrinsic fluorescence of bovine serum albumin. The quenching effect involves the formation of a steroid albumin complex which formation constant (Kf) and free energy of formation (delta G 0) can be determined by fluorescence titration. The fluorimetrically determined delta G 0 values range from -6.5 to -7.5 kcal/mol. 2. 5 alpha-Androstane and 5 alpha-pregnane are effective quenchers of albumin fluorescence, in accord with the essentially hydrophobic nature of the steroid-albumin interaction. Introduction of hydroxy or oxo groups in 5 alpha-androstane decreases the fluorescence quenching action, but the effect of each group declines when other polar groups are present in the steroid molecule. Similar effects occur with 5 alpha-pregnane except that 20-hydroxy (or oxo) duo-polar derivatives are more effective than the parent hydrocarbon. 3. Comparison of delta G 0 values for steroids differing in a single grouping shows that the steroid-albumin interaction is increased by (a) the benzenoid A-ring; (b) sulfate or carboxylate ions in the vicinity of C-3; (c) the 3-oxo group in place of the 3 alpha-hydroxyl (with 5 beta-pregnane derivatives; not with 5 alpha-androstane derivatives); (d) 17 beta-acetyl or 17 beta-hydroxyethyl residues; (e) acetylated or propionated 17 beta-hydroxy groups; (f) acetylated or methylated hydroxy groups at the C-3 of estrogens; (g) delta 5 and delta 6 double bonds; and (h) the 19 beta-methyl group. The maximal variation of delta G 0 determined by affinity-enhancing groups is -0.8 kcal/mol. Conversely, the steroid-albumin interaction is decreased by introduction of (i) oxygen atoms at C-3, C-6, C-11, C-16, and C-17; (j) 17 alpha-ethynyl and 17 alpha-acetoxyl residues; (k) benzoylated or hexahydro-benzoylated beta-hydroxy groups at C-17; (l) acetylated and benzoylated hydroxy groups at C-3; and delta 1 (conjugated) double bond. Oxo groups at C-3, C-6, C-16 and the 16 alpha, 17 alpha-epoxy group are more effective than the corresponding alpha-hydroxyl in decreasing affinity, while at C-11 and C-17, the alpha-hydroxyl is more effective than the beta-hydroxyl and the oxo group. The effect of substituents is influenced by the whole molecular structure, particularly, by the stereostructure at the A/B juncture, and the presence of an oxo group at C-17. 4. The stereospecific effect of substituents at different positions in the steroid molecule suggests that with non-aromatic, A/B trans (planar) steroids, binding to albumin primarily involves the (alpha) rear surface of the B-, C- and D-ring, and possibly, the 17 beta-side chain. With estrogens and A/B cis (dihedral) steroids, the benzenoid A-ring and electron attracting groups at C-3, respectively, may participate in binding.

Structural requirements for steroid binding and quenching of albumin fluorescence in bovine plasma albumin.

1. Steroids interact with bovine plasma albumin at a binding region that involves tryptophanyl, tyrosyl, arginyl and lysyl residues. The function of the tryptophanyl residues is demonstrated by: (a) the decrease of albumin binding affinity after modification of one tryptophanyl with 2-nitrophenylsulfenyl chloride; (b) steroid quenching of albumin tryptophanyl fluorescence; and (c) steroid quenching of 1-anilinonaphth-alene-8-sulfonate fluorescence, when it is excited by energy transfer from excited tryptophanyls. The function of tyrosyl residues is demonstrated by the decrease of albumin binding affinity after nitration of 30% tyrosyls with tetranitromethane, or deprotonation of tyrosyls by variation of pH. The function of arginyl and lysyl residues is demonstrated by the decrease of binding affinity after modification of these residues with glyoxal, formaldehyde or acetic anhydride. The presence of both apolar (Trp, Tyr and Lys (deprotonated)) and polar (Arg and Lys(protonated)) residues at the steroid binding site fits in well with the site relative apolarity, when expressed on the Kosower scale (Kosower, E.M. (1958) J. Am. Chem. Soc. 80, 3253-3260). 2. The contribution of specific amino acid residues to steroid binding depends to some extent on the steroid structure, as exemplified by the quantitatively different role of arginyl (or lysyl) residues in albumin interaction with testosterone acetate and epitestosterone, respectively, or that of tyrosyl residues in albumin interaction with 11-deoxycorticosterone and epitestosterone, respectively. 3. The concerted action of polar and apolar amino acid residues is an essential requirement for steroid binding, since unfolding of albumin polypeptide chain by guanidine-HC1, urea, or by reduction of disulfide bridges with 2-mercaptoethanol, strongly decreases steroid binding to albumin while, conversely, reoxidation and refolding of the unfolded polypeptide chain restore albumin affinity for steroids. 4. Parallel determinations of steroid binding constants by equilibrium dialysis and fluorimetric titration, as well as the general pattern of the pH and temperature effects on steroid quenching of albumin fluorescence, confirm the validity of the fluorescence quenching titration as an effective method for measuring albumin-steroid molecular interactions.

Increased biliary secretion and loss of hepatic glutathione in rat liver after nifurtimox treatment.

Treatment of rats with nifurtimox, a nitrofuran derivative widely used for the treatment of Chagas' disease, induced a time- and dose-dependent depletion of liver glutathione, maximal effects being obtained with 200 mg nifurtimox/kg body weight. Extra release of both oxidized (GSSG) and reduced (GSH) glutathione into bile contributed to this depletion. Glutathione excretion into bile accounted for only part of liver glutathione loss, thus indicating that, in addition to the GSH-peroxidase reaction (resulting in GSSG generation), other glutathione-related processes were involved in nifurtimox detoxification. Bile flow, bile salt excretion, liver lipid conjugated diene content, liver glutathione reductase and glutathione peroxidase activities, and serum alanine aminotransferase (ALAT) activity were not affected by the nifurtimox treatment, thus ruling out widespread damage of the liver cell by nifurtimox. Nevertheless, the extra GSH release in the nifurtimox-treated rats may indicate an alteration of the hepatocyte membrane.

Deficient metabolic utilization of hydrogen peroxide in Trypanosoma cruzi.

The glutathione peroxidase-glutathione reductase system, an alternative pathway for metabolic utilization of H2O2 [Chance, Sies & Boveris (1979) Physiol. Rev. 59, 527-605], was investigated in Trypanosoma cruzi, an organism lacking catalase and deficient in peroxidase [Boveris & Stoppani (1977) Experientia 33, 1306-1308]. The presence of glutathione (4.9 +/- 0.7 nmol of reduced glutathione/10(8) cells) and NADPH-dependent glutathione reductase (5.3 +/- 0.4 munit/10(8) cells) was demonstrated in the cytosolic fraction of the parasite, but with H2O2 as substrate glutathione peroxidase activity could not be demonstrated in the same extracts. With t-butyl hydroperoxide or cumene hydroperoxide as substrate, a very low NADPH-dependent glutathione peroxidase activity was detected (equivalent to 0.3-0.5 munit of peroxidase/10(8) cells, or about 10% of glutathione reductase activity). Blank reactions of the glutathione peroxidase assay (non-enzymic oxidation of glutathione by hydroperoxides and enzymic oxidation of NADPH) hampered accurate measurement of peroxidase activity. The presence of superoxide dismutase and ascorbate peroxidase activity in, as well as the absence of catalase from, epimastigote extracts was confirmed. Ascorbate peroxidase activity was cyanide-sensitive and heat-labile, but no activity could be demonstrated with diaminobenzidine, pyrogallol or guaiacol as electron donor. The summarized results support the view that T. cruzi epimastigotes lack an adequate enzyme defence against H2O2 and H2O2-related free radicals.