Structure-activity relationships in a series of 11-deoxy prostaglandins

A series of 11-deoxy prostaglandin derivatives and some naturally occurring prostaglandins have been investigated in the anaesthetized artificially respired guinea-pig for their effect on blood pressure, bronchial resistance (overflow pressure at constant volume), tracheal segment pressure, and on intestinal and uterine smooth muscle. The compounds were administered intravenously. Prostaglandins E₁, E₂ and F_2α produced responses that were qualitatively similar to those in the literature. Prostaglandin A₂ (100 μg) was a bronchoconstrictor, although it decreased tracheal segment pressure and blood pressure. Prostaglandin B₂ (100 μg) caused double elevations in blood pressure, tracheal segment pressure and bronchial resistance. The intensity of bronchoconstriction produced by PGB₂ was of the same order as with PGF_2α. A number of structure-activity relationships were found. 11-Deoxygenation lowered the biological activity of the natural prostaglandins PGE₁ and PGF_1α. The vasodepressor and bronchodilator responses of 11-deoxy PGE₁ were converted to vasopressor and bronchoconstrictor by epimerisation at C-15. Introduction of a methyl group at C-15 of 11-deoxy PGF_1α both increased and prolonged vasopressor and bronchoconstrictor activity. At C-9 both the keto and β-hydroxy group imparted vasodepressor and bronchodilator activity, while the α-hydroxy led to vasopressor and bronchoconstrictor activity. Extension of the omega sidechain by two methylene groups radically reduced the activity of 11-deoxy PGF_1α and its derivatives.These experiments indicate that steric differences in the prostaglandin structures studied can result in diametrically opposed profiles of biological activity. Further, small variations in the prostaglandin molecule can lead to differences in potency and/or profile of activity in the guinea-pig.     

Characterization and distribution of a phospholipase A2 activity from adult rabbit lung

A phospholipase A₂ activity was characterized in adult rabbit lung. This activity was calcium- and deoxycholate-dependent and displayed an alkaline pH optimum. K_m and V_max were 0.176mM and 256.8 pmoles/min./mg protein respectively. The microsomal fraction displayed the highest enzymatic specific activity; the lowest activity was present in the cytosol. Yet this latter fraction accounted for the majority of the total activity. Although the specific activity was high within the lamellar body fraction this compartment contained only 2% of the total activity. Phospholipase A₂ activity was inhibited by bromophenacyl bromide, chlorpromazine and mepacrine in decreasing order of effectiveness. Treatment of the microsomes with increasing concentrations of NaCl indicated that the lung phospholipase A₂ activity was related loosely bound to the microsomal membranes and was maximally removed with salt at a concentration only slightly higher than physiological. Addition of calmodulin to the enzyme assay did not significantly alter hydrolysis of labelled phosphatidylcholine.     

Specific prostaglandine E1 and A1 binding sites in rat a dipocyte plasma membranes

Rat adipocyte plasma membranes sacs have been shown to be a sensitive and specific system for studying prostaglandin binding. The binding of prostaglandin E₁ and prostaglandin A₁ increases linearly with increasing protein concentration, and is a temperature-sensitive process. Prostaglandin E₁ binding is not ion dependent, but is enhanced by GTP. Prostaglandin A₁ binding is stimulated by ions, but is not affected by GTP.Discrete binding sites for prostaglandin E₁ and A₁ were found. Scatchard plot analysis showed that the binding of both prostaglandins was biphasic, indicating two types of binding sites. Prostaglandin E₁ had association constants of 4.9 · 10⁹ 1/mole and 4 · 10⁸ 1/mole, while the prostaglandin A₁ association constants and binding capacities varied according to the ionic composition of the buffer. In Tris-HCl buffer, the prostaglandin A₁ association constants were 8.3 · 10⁸ 1/mole and 5.7 · 10⁷ 1/mole, while in the Krebs—Ringer Tris buffer, the results were 1.2 · 10⁹ 1/mole and 8.6 · 10⁶ 1/mole.Some cross-reactivity between prostaglandin E₁ and A₁ was found for their respective binding sites. Using Scatchard plot analysis, it was found that a 10-fold excess of prostaglandin E₁ inhibited prostaglandin A₁ binding by 1–20% depending upon the concentration of prostaglandin A₁ used. Prostaglandin E₁ competes primarily for the A prostaglandin high-affinity binding site. Similar Scatchard analysis using a 20-fold excess of prostaglandin A₁ inhibited prostaglandin E₁ binding by 10–40%. Prostaglandin A₁ was found to compete primarily for the E prostaglandin low-affinity receptor.All of the bound [³H]prostaglandin E₁, but only 64% of the bound [³H]-prostaglandin A₁ can be recovered unmetabolized from the fat cell membrane. There is no non-specific binding of prostaglandin E₁, but 10–15% of prostaglandin A₁ binding to adipocyte membranes is non-specific. Using a parallel line assay to measure relative affinities for the E binding site, prostaglandin E₁ > prostaglandin A₂ > prostaglandin F_2α. Prostaglandin E₂ and 16,16-dimethyl prostaglandin E₂ were equipotent with prostaglandin E₁, while other prostaglandins had lower relative affinities. 7-Oxa-13-prostynoic acid does not appear to antagonize prostaglandin activity in adipocytes at the level of the receptor.     

Role of phospholipase A2 and prostaglandin E in growth and differentiation of myeloid leukemic cells

Phospholipase A₂ activity and prostaglandin E synthesis have been studied in different clones of myeloid leukemic cells, which differ in their competence to be induced to differentiate by the macrophage and granulocyte differentiation-inducing protein or the tumor promoter 12-O-tetradecanoyl phorbol-13-acetate (TPA). Clones that could be induced to differentiate by this protein showed a higher basal phospholipase A₂ activity than clones that could not be induced to differentiate by this protein inducer. Cell competence to be induced to differentiate by TPA did not show this correlation, and the clone with the least ability to respond to TPA showed the lowest number of binding sites for [20-³H]phorbol 12,13-dibutyrate. Differentiation induced by the protein was accompanied by a 7–14-fold increase in prostaglandin E synthesis, whereas differentiation induced by TPA did not show this increase. Externally added prostaglandin E₁ did not induce differentiation but inhibited cell proliferation and the degree of inhibition in the different clones was related to the basal phospholipase A₂ activity. The results indicate that increase of prostaglandin E synthesis was not an essential pre-requisite for differentiation, that prostaglandin E seems to be involved in the inhibition of cell proliferation in association with phospholipase A₂, and that the differentiation-inducing protein and TPA can induce differentiation by different pathways. The amount of basal phospholipase A₂ activity was also related to previously found differences in the ability of the clones to develop desensitization to β-adrenergic hormones or prostaglandin E₁.     

pH dependent stabilization of S2QA − and S2QB − charge pairs studied by thermoluminescence

The pH dependence of emission peak temperature and decay time of thermoluminescence arising from S₂Q_B ^– and S₂Q_A ^– recombinations demonstrates that a stabilization of S₂Q_B ^– occurs at low pH whereas stabilization of S₂Q_A ^– occurs at high pH. Based on comparative analysis of thermoluminescence parameters of the two types of recombination, we suggest that in the pH range between 5.3 and 7.5, E_m(S₂/S₁) and E_m(Q_A/Q_A ^–) are constant, but E_m(Q_B/Q_B ^–) gradually increases with decreasing pH, while in the pH range between 7.5 and 8.5, an unusual change occurs on S₂Q_A ^– charge pair, which is interpreted as either a decrease in E_m(S₂/S₁) or an increase in E_m(Q_A/Q_A ^–).     

Mechanism of increased renal prostaglandin E2 in uranyl nitrate-induced acute renal failure

We have previously demonstrated that decreased cortical prostaglandin metabolism can contribute significantly to an increase in renal tissue levels and activity of prostaglandin E₂ in bilateral ureteral obstruction, a model of acute renal failure. In the present study, we have further investigated whether alterations in prostaglandin metabolism can occur in a nephrotoxic model of acute renal failure. Prostaglandin synthesis, prostaglandin E₂ metabolism (measured as both prostaglandin E₂-9-ketoreductase and prostaglandin E₂-15-hydroxydehydrogenase activity), and tissue concentration of prostaglandin E₂ were determined in rabbit kidneys following an intravenous administration of uranyl nitrate (5 mg/kg). No changes in the rates of cortical microsomal prostaglandin E₂ and prostaglandin F_2α synthesis were noted at the end of 1 and 3 days, while medullary synthesis of prostaglandin E₂ fell by 47% after 1 day and 43% after 3 days. Cortical cytosolic prostaglandin E₂-9-ketoreductase activity was found to be decreased by 36% and 76% after 1 and 3 days respectively. No significant changes were noted in cortical cytosolic prostaglandin E₂-15-hydroxydehydrogenase activity after 3 days. Cortical tissue levels of prostaglandin E₂ increased by 500% at the end of 3 days. These data demonstrate that in nephrotoxic acute renal failure, decreased prostaglandin metabolism (i.e., prostaglandin E₂-9-ketoreductase activity) can result in increased tissue levels of prostaglandin E₂ in the absence of increased prostaglandin synthesis and suggest that alterations in prostaglandin metabolism may be an important regulator of prostaglandin activity in acute renal failure.     

Thromboxane A2 and prostaglandin H2: Potent stimulators of the swine coronary artery

Thromboxane A₂ was generated by incubation of arachidonic acid with a suspension of human platelets. The filtrate contained 266 ± 46 ng/ml (n=10) of thromboxane A₂ and 25 ng/ml or less of prostaglandin endoperoxides (prostaglandins G₂+H₂). Thromboxane A₂ was 2–10 times more potent than prostaglandin H₂ and 9–102 times and 26–308 times more potent than prostaglandins E₂ and F₂α, respectively, in causing contractions of the superfused swine coronary artery.     

Subcellular localization of prostaglandin E2 binding sites in bovine adrenal medulla

Prostaglandins E₁ and E₂ are specifically bound by particulate fractions from bovine adrenal medulla. The subcellular localization of these binding sites has been investigated by comparing their distribution in subcellular fractions obtained by differential and gradient centrifugation to those of marker enzymes for various organelles. Prostaglandin E₂ binding sites were purified about 16-fold with respect to the homogenate in a fraction which was highly enriched in plasma membranes on the basis of the activities of the marker enzymes acetylcholinesterase and calcium-dependent ATPase, which were both purified by about 12-fold in this fraction. The plasma membrane fraction contained relatively low activities of marker enzymes for mitochondria (monoamine oxidase), lysosomes (acid phosphatase), endoplasmic reticulum (glucose-6-phosphatase), Golgi (galactosyl transferase) and chromaffin granule membranes (dopamine β-hydroxylase). The only other fractions enriched in prostaglandin E₂ binding sites were those for the endoplasmic reticulum and the Golgi, in which the binding sites were purified about 4-fold and 7-fold, respectively. This is probably due mainly to contamination with plasma membranes, since calcium-dependent ATPase and acetylcholinesterase were each purified to a similar extent in these two fractions. These data suggest that the high-affinity prostaglandin E₂ binding sites of the adrenal medulla are localized primarily on the plasma membranes of the medullary cells.     

Work in progress. Occurence of phospholipase A1 and A2 in human decidua

Phospholipase A₂, an enzyme which may regulate the formation of polyunsaturated fatty acids utilized for prostaglandin synthesis, was found to have significant higher activity in decidual than in myometrial tissue. The major part of phospholipase A₂ in the decidua had an acid pH optimum, which indicates that most of the enzyme is stored in the lysosomes of this tissue. These findings, together with previous observations, lend further support to the view that lysosomal phospholipase A₂ released within decidual cells might be a trigger of abortion and parturition.     

Stimulative effect of prostaglandins on production of hexosamine-containing substances by cultured fibroblasts (2) early effect of various prostaglandins at various doses

Early effects of various prostaglandins on the production of hexosamine-containing substances by cultured fibroblasts, which were derived from a rat carrageenin granuloma, were studied. At the stationary phase, the cells were exposed for 6 h to one of the prostaglandin A₁ (PGA₁), A₂, B₁, B₂, D₂, F_1α, F_2α, E₁, E₂ or arachidonic acid in various concentrations ranging from 0.01 to 10 μg/ml for all the stimuli and from 10 pg to 10 μg/ml for PGF_2α. The activity of the cells in incorporating ³H-glucosamine into hexosamine-containing substances (acidic) glycosaminoglycans and glycoproteins) during this period was compared with that of control cells. All the stimuli tested showed more or less stimulative effect on the synthesis of hexosamine-containing substances at their specific concentrations. PGF_2α was found to be the most potent stimulant and its stimulative effect was found significant even at the low concentration of 100 pg/ml. PGD₂, F_1α and E₂ were the next potent stimuli. Their optimum dose were around 1 μg/ml but they still had significant stimulation at the concentration of 0.01 μg/ml. Effect of PGE₂ was rather mild. Stimulation by PGA₁, A₂, B₁ and B₂ or arachidonic acid was seen at high dose, and its seemed to be non-specific. The results suggested that these prostaglandins such as PGF_2α, D₂, F_1α and E₂ play some important role on regulating the production of intercellular ground substances.     

Polymorphonuclear leukocytes produce thromboxane A2-like activity during phagocytosis

Homogenates of phagocytosing polymorphonuclear leukocytes obtained from rabbit peritoneum were incubated with the prostaglandin endoperoxides PGG₂ or PGH₂. After 2 min at 0°C, incubation mixtures contained an increased rabbit aorta contracting activity. Ether extracts of incubation mixtures contained a substance which contracted the superfused strips of rabbit aorta and coeliac artery and had a half life which was similar to thromboxane A₂. The generation of thromboxane A₂-like activity from PG endoperoxides was prevented by boiling the homogenate prior to incubation, or by pretreatment with benzydamine, a drug which blocks thromboxane formation in platelets. Production of thromboxane A₂-like material by leukocyte homogenates was compared with platelet microsomal thromboxane synthetase.     

Evidence for enhanced venous smooth muscle turnover of prostaglandin-like substance in portal veins from spontaneously hypertensive rats

The sensitivity of portal veins from 14 to 18 week-old Okamoto-Aoki spontaneously hypertensive rats to prostaglandins A₂, B₂, D₂ and F_2α were enhanced whereas the sensitivity to prostaglandin E₂ was diminished when compared with responses of veins from normotensive Wistar-Kyoto rats. Inhibition of prostaglandin synthesis with both eicosotetraynoic acid (ETYA) and indomethacin (INDO) abolished the observed differences in sensitivity to prostaglandins. Synthesis of prostaglandin-like substance (with arachidonic acid as precursor) was significantly enhanced in portal veins from spontaneously hypertensive rats. Metabolism of prostaglandins E₂ and F_2α, employing the oil-immersion technique of Kalsner and Nickerson, appeared to be similar in veins from normotensive and hypertensive rats. These findings suggest that prostaglandin synthesis is enhanced in venous smooth muscle from hypertensive rats. The increased concentration of endogenous prostaglandin at the venous smooth muscle cell may modify the responses to exogenously administered prostaglandins thus accounting, in part, for the altered sensitivity to these fatty acids.     

Characterization and localization of prostaglandin E1 receptors in rat liver plasma membranes

Methodology for measurement and characterization of prostaglandin binding to membranes has been developed. The binding assay was used to study the presence of prostaglandin receptors in high purified cell fractions derived from rat liver. High affinity binding receptors which have a saturation value of 1.0 pmole/mg protein and a dissociation constant of 1.2 nM were found exclusively in the plasma membrane. High affinity receptors were not found in cell fractions containing nuclei, rough microsomes. Golgi complex or mitochondria. The binding by other prostaglandins was competitive with prostaglandin E₁. Competitive binding studies were used to obtain dissociation constants for prostaglandins F_1α, F_2α, B₁, B₂, A₁, A₂, and 15-keto prostaglandin E₂ which were 1100, 100, 300, 180, 16. 16 and 700 nM, respectively. Eicosa-5.8.11.19-tetraynoic acid, an inhibitor of prostaglandin synthesis did not bind appreciably to the prostaglandin E receptor, whereas two prostaglandin analogues, which have high physiological activity compete effectively with prostaglandin E₁ for the receptor. Thus, the binding receptor for the E-type prostaglandins is highly specific both with respect to cell localization as well as the type of substrate. Numerical routines for the fitting of the data and a procedure for the determination of the specific activity of the labelled prostaglandin are provided.     

Comparison of the effects of prostaglandin I2 and prostaglandin E2 stimulation of the rat kidney adenylate cyclase-cyclic AMP systems

Prostacyclin (Prostaglandin I₂) effects on the rat kidney adenylate cyclase-cyclic AMP system were examined. Prostaglandin I₂ and prostaglandin E₂, from 8 · 10^?4 to 8 · ^?7 M stimulated adenylate cyclase to a similar extent in cortex and outer medulla. In inner medulla, prostaglandin I₂ was more effective than prostaglandin E₂ at all concentrations tested. Both prostaglandin I₂ and prostaglandin E₂ were additive with antidiuretic hormone in outer and inner medulla. Prostaglandin I₂ and prostaglandin E₂ were not additive in any area of the kidney, indicating both were working by similar mechanisms. Prostaglandin I₂ stimulation of adenylate cyclase correlated with its ability to increase renal slice cyclic AMP content. Prostaglandin I₂ and prostaglandin E₂ (1.5 · 10^?4 M) elevated cyclic AMP content in cortex and outer medulla slices. In inner medulla, with Santoquin® (0.1 mM) present to suppress endogenous prostaglandin synthesis, prostaglandin I₂ and prostaglandin E₂ increased cyclic AMP content. 6-Ketoprostaglandin F_1α, the stable metabolite of prostaglandin I₂, did not increase adenylate cyclase activity or tissue cyclic AMP content. Thus, prostaglandin I₂ activates renal adenylate cyclase. This suggests that the physiological actions of prostaglandin I₂ may be mediated through the adenylate cyclase-cyclic AMP system.     

Light activation of fructose bisphosphatase in photosynthetically competent pea chloroplasts.

The ipsilateral kidney was removed from a rabbit 48h after unilateral partial renal-vein-constriction and was perfused with Krebs–Henseleit media at 37°C. Hourly administration of a fixed dose of bradykinin to the renal-vein-constricted kidney demonstrated a marked time-dependent increase in the release of bioassayable prostaglandin E₂ and thromboxane A₂ into the venous effluent as compared with the response of the contralateral control kidney. The renal-vein-constricted kidney produced up to 60 times more prostaglandin E₂ in response to bradykinin after 6h of perfusion as compared with the contralateral kidney; thromboxane A₂ was not demonstratable in the contralateral kidney. Inhibition of protein synthesis de novo in the perfused renal-vein-constricted kidney with cycloheximide lessened the hormone-stimulated increase in prostaglandin E₂ by 94% and in thromboxane A₂ by 90% at 6h of perfusion. Covalent acetylation of the renal cyclo-oxygenase by prior oral administration of aspirin to the rabbit inhibited initial bradykinin-stimulated prostaglandin E₂ biosynthesis 71% at 1h of perfusion. However, there was total recovery from aspirin in the renal-vein-constricted kidney by 2h of perfusion after bradykinin stimulation. Total cyclo-oxygenase activity as measured by [¹⁴C]arachidonate metabolism to labelled prostaglandins by renal cortical and renal medullary microsomal fractions prepared from 6h-perfused kidneys demonstrated that renal-vein-constricted kidney-cortical cyclo-oxygenase activity was significantly greater than the contralateral-kidney-cortical conversion, whereas medullary arachidonate metabolism was comparable in both the renal-vein-constricted kidney and contralateral kidney. These data suggest that perfusion of a renal-vein-constricted kidney initiates a time-dependent induction of synthesis of prostaglandin-producing enzymes, which appear to be primarily localized in the renal cortex. The presence of the synthetic capacity to generate very potent vasodilator and vasoconstrictor prostaglandins in the renal cortex suggests that these substances could mediate or modulate changes in renal vascular resistance in pathological states.     

NADP-linked 15-hydroxyprostaglandin dehydrogenase for prostaglandin D2 in human blood platelets

Two forms of NADP-linked 15-hydroxyprostaglandin dehydrogenase for prostaglandin D₂ were found in the cytosol fraction of human blood platelets. These enzymes were purified by ammonium sulfate fractionation, Blue Sepharose, and Sephadex G-100 column chromatography. The two enzymes differed in molecular weights (65,000 for peak I enzyme and 31,000 for peak II as estimated by gel filtration) and their substrate specificities. The relative rates for reaction with peak I enzyme were: prostaglandin D₂, 100(%); E₂, 14; F_2α, 2; I₂, 29; and B₂, 0; whereas for peak II enzyme, D₂, 100; E₂, 23; F_2α, 61; I₂, 29; and B₂, 131. Prostaglandin D₂ was converted to 15-ketoprostaglandin D₂ and then 13,14-dihydro-15-ketoprostaglandin D₂, which were identified by spectrophotometry and gas chromatography/mass spectrometry, respectively. These metabolites were three orders of magnitude less potent in inhibiting human platelet aggregation than prostaglandin D₂. The results indicated that NADP-linked dehydrogenases participated in the metabolic inactivation of prostaglandin D₂ in the platelets. Furthermore, the dehydrogenase activity for prostaglandin D₂ was high in monkey (0.128 nmol/min · mg at 24 °C) and human platelets (0.066), but was not detectable (less than 0.007) in the rabbit, rat, and chicken. Because prostaglandin D₂, which was demonstrated by several authors to be synthesized in platelet-rich plasma during platelet aggregation, exhibited significant antiaggregatory activity only in human and monkey platelets, these prostaglandin dehydrogenases appear to play a physiological role in the circulatory system.     

Properties of solubilized prostaglandin synthetase from sheep vesicular glands.

Prostaglandin synthetase activity associated with the microsomal fraction from sheep vesicular glands has been solubilized by treatment with the non-ionic detergents Tween 20, Lubrol Px and Lubrol Wx. Approx.8 fold purification from microsomes is obtained and over 90% of the activity is recovered in the detergent solubilized fraction. The solubilized synthetase activity is stable at pH 5.0 but is gradually lost at pH 8.0; it is also heat and acid labile. The relative amounts of prostaglandins E₂, D₂ and F_2α formed by the microsomal-bound synthetase and by the solubilized synthetase are similar. Also similar are the pH optima (7.9–8.5) of the two synthetase preparations. The solubilization process appears to yield a fully active enzymatic preparation which could be employed for further purification and characterization of the prostaglandin synthetase complex.     

A comparison of imidazole and 9,11-azoprosta-5,13-dienoic acid Two selective thromboxane synthetase inhibitors

Two selective thromboxane A₂ synthetase inhibitors, imidazole and 9,11-azoprosta-5,13-dienoic acid (azo analog I) were compared to determine their effects on the quantitative formation of thromboxane B₂ and prostaglandin E₂ accompanying human platelet aggregation. Azo analog I was at least 200 times more potent, on a molar basis, than imidazole in suppressing thromboxane B₂ formation in either platelet-rich plasma or washed platelet suspensions aggregated with arachidonic acid or prostaglandin H₂. The inhibitors differed in their effect on the aggregation response itself. Azo analog I selectively suppressed thromboxane A₂ formation with an accompanying, parallel, suppression of the platelet aggregation.Imidazole selectively suppressed thromboxane A₂ formation, but only suppressed the accompanying aggregation in platelet rich plasma, and not washed platelet suspensions. The results indicate that azo analog I functions by competitive inhibition of prostaglandin H₂ on the thromboxane synthetase, and that imidazole, while it suppresses thromboxane A₂ formation, may have an associated agonist activity that enhances platelet aggregation. The data presented support this hypothesis, and they emphasize the importance of thromboxane A₂ in arachidonate mediated platelet aggregation.     

The Small Subunit AroB of Arsenite Oxidase: LESSONS ON THE [2Fe-2S] RIESKE PROTEIN SUPERFAMILY

Here, we describe the characterization of the [2Fe-2S] clusters of arsenite oxidases from Rhizobium sp. NT-26 and Ralstonia sp. 22. Both reduced Rieske proteins feature EPR signals similar to their homologs from Rieske-cyt b complexes, with g values at 2.027, 1.88, and 1.77. Redox titrations in a range of pH values showed that both [2Fe-2S] centers have constant E_m values up to pH 8 at ∼+210 mV. Above this pH value, the E_m values of both centers are pH-dependent, similar to what is observed for the Rieske-cyt b complexes. The redox properties of these two proteins, together with the low E_m value (+160 mV) of the Alcaligenes faecalis arsenite oxidase Rieske (confirmed herein), are in line with the structural determinants observed in the primary sequences, which have previously been deduced from the study of Rieske-cyt b complexes. Since the published E_m value of the Chloroflexus aurantiacus Rieske (+100 mV) is in conflict with this sequence analysis, we re-analyzed membrane samples of this organism and obtain a new value (+200 mV). Arsenite oxidase activity was affected by quinols and quinol analogs, which is similar to what is found with the Rieske-cyt b complexes. Together, these results show that the Rieske protein of arsenite oxidase shares numerous properties with its counterpart in the Rieske-cyt b complex. However, two cysteine residues, strictly conserved in the Rieske-cyt b-Rieske and considered to be crucial for its function, are not conserved in the arsenite oxidase counterpart. We discuss the role of these residues.