Oxidation of prostacyclin and its analogs by three 15-hydroxyprostaglandin dehydrogenases

A study of the oxidation of prostacyclin and some of its analogs by three 15-hydroxyprostaglandin dehydrogenases was undertaken to determine the structural features of these compounds which might influence their rate of enzymatic inactivation. The effect of some structural changes seemed to be determined by the substrate specificity of individual enzymes. Other changes influenced the rate of oxidation by all three enzymes similarily. Among this latter group it was noted that a 15S hydroxyl group is necessary for oxidation to occur and that steric changes in the carboxy side chain and structural changes in the epoxy ring have a greater effect on the affinity of the substrate for the enzyme than on its maximum rate of oxidation. Certain analogs of prostacyclin are not substrates for one or more of the enzymes tested. Of these, (5S)-9-deoxy-5,9 alpha-epoxy-PGF1 and its methyl ester are potent inhibitors of only the placental enzyme---an interesting case of apparent selective metabolic regulation.     

Conversions of prostaglandin endoperoxides by prostacyclin synthase from pig aorta

Partially purified prostacyclin synthase from pig aorta converted the prostaglandin (PG) endoperoxide PGH₂ to prostacyclin (PGI₂), and PGH₁ to 12-hydroxy-8,10-heptadecadienoic acid (HHD). Both reactions were inhibited by 15-hydroperoxy-5,8,11,13-eicosatetraenoic acid (15-HP) in a dose-dependent fashion. However, the reactions PGH₂ → PGI₂ and PGH₁ → HHD appeared to differ: substrate availability was rate limiting in the latter reaction, while the enzyme became rapidly saturated with PGH₂ and a steady rate of prostacyclin formation was observed at higher substrate levels.     

Monoclonal antibody recognition of abscisic Acid analogs

Specificities of three monoclonal antibodies (15-I-C5, DBPA 1, and MAC 62) raised against the plant hormone (S)-(+)-abscisic acid (ABA) have been compared. Immunological cross-reactivities against fifteen biologically active analogs of ABA were measured. The ABA analogs were altered at one or more of four positions: the double bonds in the ring, at C-2 C-3 and at C-4 C-5, and in the oxidation level at C-1. Several analogs were optically active with chiral centers at C-1′ and C-2′. For cross-reactivity, all three monoclonal antibodies required the carboxylic acid group, and the cis configuration of the double bond at C-2 C-3 of the ABA molecule. Monoclonals 15-I-C5 and DBPA 1 required the entire ABA sidechain from the C-1 to C-1′, but these monoclonals did cross-react with analogs with the ring double bond reduced and the C-2′ methyl cis to the sidechain. Only MAC 62 recognized analogs containing an acetylene at C-4 C-5. MAC 62 had more strict requirements for the ring double bond, but gave some cross-reactivity with acetylenic analogs having a saturated ring. All three monoclonals had higher specificity for analogs having the same absolute configuration at C-1′ as (S)-(+)-ABA. This work provides new information about the spatial regions of the ABA molecule that elicit immunological recognition, and serves as a basis for future investigations of the ABA receptor using ABA analogs and anti-idiotypic antibodies.     

Comparison of substrate specificities of the human placental NAD- and NADP-linked 15-hydroxyprostaglandin dehydrogenases

A study of the relative activity of the purified placental NAD- and NADP-linked 15-hydroxyprostaglandin dehydrogenases with various prostaglandins and thromboxane B₂(TxB₂) suggests that most, if not all, oxidation in the placenta of the 15-hydroxyl group of prostaglandins of the A, E, and F series as well as PGI₂ (prostacyclin) and 6-keto PGF_1α is catalyzed by the NAD-linked enzyme. Prostaglandin B₁ is an excellent substrate for the NADP-linked enzyme. Despite the conformational similarities between PGB₁ and PGI₂, the latter molecule is a poor substrate for the NADP-linked enzyme. Thromboxane B₂ is not oxidized by the NAD-linked enzyme and is oxidized slowly by the NADP-linked enzyme.     

Chemically modified ATP derivatives for the study of aminoacyl-tRNA synthetases from baker's yeast: ATP analogs with fixed conformations or modified triphosphate chains in the aminoacylation reaction

The systematic investigation of substrate specificity of aminoacyl-tRNA synthetases from yeast is completed by tests of ATP analogs with fixed conformation about the glycosidic bond and with modifications in the triphosphate chain as substrate analogs in the aminoacylation reaction. Two analogs with fixed high anti (8,2′-O-cyclo-ATP, 8,2′-S-cyclo-ATP) and two with fixed anti (8,3′-O-cyclo-ATP, 8,3′-S-cyclo-ATP) conformation have been tested in the esterification reaction of phenylalanyl-, seryl-, lysyl-, valyl-, isoleucyl-, arginyl-, and tyrosyl-tRNA synthetases from baker's yeast. None of the compounds was a substrate, whereas 11 K_i values could be determined. 8,2′-S-cyclo-ATP, remarkably, is the only analog which inhibits all these synthetases. Each compound with a fixed anti conformation inhibits two enzymes. Among 11 analogs with modifications in the triphosphate chain, four were substrates for phenylalanyl-, three for seryl-, one for lysyl-, three for valyl-, one for isoleucyl-, and none for arginyl- and for tyrosyl-tRNA synthetases. Two compounds were inhibitors of different types for phenylalanyl-, two for seryl-, seven for lysyl-, six for valyl-, nine for isoleucyl-, seven for arginyl-, and two for tyrosyl-tRNA synthetases. Their K_m, V, and K_i values have been determined. In the general picture of substrate specificity the subunit enzymes can tolerate substitutions in position 2, 2′, at the α-phosphorus, at the β,γ-P-X-P bridge and at the γ-phosphorus atom. The single chain enzymes tolerate substitutions in position 7 and at the γ-phosphorus. All seven synthetases from yeast need an intact NH₂ group in position 6 and an oxygen atom in position 3′.     

The use of tritiated prostaglandins in metabolism studies II: Kinetic isotopic effect: An useful tool to investigate catabolizing sequence of prostaglandins

Although numerous data exist concerning tritium kinetic isotope effect in enzymic reactions, little is related to the metabolism of tritiated prostaglandins. The present study reports an evaluation of the kinetic isotope effect which occurs during the oxidation of 15-hydroxyl group of tritium-labeled prostaglandins E₂ and F_2α by the 15-hydroxyprostaglandin dehydrogenase and during the oxidation of 9-hydroxyl group of tritium-labeled prostaglandin F_2α by the 9-hydroxyprostaglandin dehydrogenase. The large kinetic isotope effect tends to limit the validity of the dehydrogenase assay using tritium-labeled prostaglandins as substrate. However these assays can be considered to be an indication of relative enzyme activity.     

Effect of growth conditions on enzyme activities,intracellular kinetics and biomass yields of a new RuMP-type methylotroph (T15)

Summary A new methylotrophic strain (T15), which employs the ribulose monophosphate (RuMP) cycle of formaldehyde assimilation, was isolated on the basis of high in vitro activities of formaldehyde and formate dehydrogenases (19 and 678 mU per mg protein, respectively). Serial subculturing of the strain in batch cultures, on 4 g/l CH₃OH for 6 months, led to loss of substantial percentages of the NAD-linked formaldehyde (25%) and formate (98%) dehydrogenases. The activities of these two enzymes were partially recovered when cells were grown continuously at very low dilution rate (0.03 h^–1). We found large variations (40 to 1000%) in the activities of other key enzymes of carbon-substrate oxidation (both linear and cyclic) and assimilation, in batch cultures with pure and mixed substrates, and in continuous cultures of different dilution rates. Key intracellular reaction rates, including those of the cyclic and linear substrate oxidation, were measured in vivo using a ¹⁴C-tracer technique in both continuous and batch cultures. The results indicate significant variations in these reaction rates, particularly those of linear and cyclic carbon oxidation. Overall, the cyclic oxidation appears to be employed to a larger (although not predominant) extent in strain T15 compared with another RuMP strain (L3) we have previously examined. T15 exhibits high biomass yields (up to 0.63 g cells per g CH₃OH) and growth rates (up to 0.46 h^–1) on CH₃OH in batch cultures. CH₃NH₂ can also be utilized as a substrate. In continuous culture, T15 could be grown at dilution rates up to 0.36 h^–1 with a corresponding biomass yield of 0.4. Examination of a large number of data on the biomass yields of strains T15 and L3 reveals that the large variations in yields derive from the variable branching of carbon flow between linear and cyclic oxidation and assimilation, rather than changes in the biosynthetic efficiency of carbon incorporation into biomass.     

Flavin-containing heme enzymes

There are many examples of oxidative enzymes containing both flavin and heme prosthetic groups that carry out the oxidation of their substrate. For the purpose of this article we have chosen five systems. Two of these, the l-lactate dehydrogenase flavocytochrome b₂ and cellobiose dehydrogenase, carry out the catalytic chemistry at the flavin group. In contrast, the remaining three require activation of dioxygen at the heme group in order to accomplish substrate oxidation, these being flavohemoglobin, a nitric oxide dioxygenase, and the mono-oxygenases nitric oxide synthase and flavocytochrome P450 BM3, which functions as a fatty acid hydroxylase. In the light of recent advances we will describe the structures of these enzymes, some of which share significant homology. We will also discuss their diverse and sometimes controversial catalytic mechanisms, and consider electron transfer processes between the redox cofactors in order to provide an overview of this fascinating set of enzymes.     

Action of PGI2 analogs on the pulmonary and systemic circulations in the conscious newborn lamb

Previous work (Lock , J. Pharm. Exp. Ther. :156, 1980) has shown that conventional screening procedures for vasoactive PGI₂ analogs have little value in predicting pulmonary vasodilator activity in the newborn lamb. To gain a better insight into the structural requirements for pulmonary vasoactivity and possibly identify useful compounds for the management of neonatal pulmonary hypertensive disorders, we have tested the following PGI₂ analogs in normoxic and hypoxic newborn lambs: 15(S)-9-deoxy-15-methyl1–9α, 6-nitrilo-PGF₁ (analog I); 9-deoxy-9α, 5-nitrilo-PGF₁ (analog II); (6S, 15S)-15-methyl-PG1₁ (analog III); and (6R, 15S)-15-methyl-PGI₁ (analog IV). A prostaglandin analog mimicking PGI₂ (compound BW245C; (±)-5-(6-carboxyhexyl)-1-(3-cyclohexyl-3-hydroxypropyl)hydantoin) was tested as well. Compounds were injected into a branch pulmonary artery and any local pulmonary effect could be assessed from the change in the ratio of blood flow to the injected lung over total flow. None of the analogs tested proved to be a selective pulmonary dilator. BW245C was a potent peripheral vasodilator (threshold around 0.5 μg/kg) and indirectly lowered pulmonary vascular resistance through its systemic effects. Analog I also dilated the systemic circulation, but only at the highest dose tested (100 μg/kg). The latter finding is surprising because it was previously shown that the parent, non-methylated compound is a fairly potent and selective pulmonary vasodilator. Analog II and IV were inactive at a dose up to, respectively, 30 and 20 μg/kg. Analog III, on the other hand, weakly constricted the systemic circulation at a dose of 10 μg/kg. These findings suggest that the neonatal pulmonary vasculature is endowed with specific receptor sites which can discriminative between closely related PGI₂ analogs.     

Studies on the Active Site of Dihydroxy-acid Dehydratase

Several classes of substrate analogs of dihydroxy-acid dehydratase have been tested as inhibitors of this enzyme in an attempt to characterize its binding site and find what modifications in substrate structure lead to an affinity higher than that of the natural substrates. The substrate analogs were tested on dihydroxy-acid dehydratase from both spinach and Escherichia coli. One modification of the substrate that led to as much as a 1000-fold increase in binding affinity was replacement of the 3-hydroxyl group with a thiol. It has been shown previously that the 3-hydroxyl group of the substrate becomes a ligand for one Fe of the Fe-S clusters of these enzymes on binding to their active sites. It seems likely then that the tighter binding of the thiol containing analogs is due to the thiol group becoming a ligand to an iron of the Fe-S clusters of these enzymes. A second modification in substrate that led to as much as 1000-fold increase in binding affinity was the addition of a large lipophilic group. This suggests there is a large hydrophobic pocket or hydrophobic surface near the active site of dihydroxy-acid dehydratase. A modification in substrate that led to as much as a 50-fold increase in binding was the replacement of the carboxyl group of the substrate with phosphonate; however, this increase was limited to substrate analogs without a polar functionality on the carbon β to the phosphonate group. Bromopyruvate was found to irreversibly inactivate dihydroxy-acid dehydratase. Each good inhibitor we found was active on spinach dihydroxy-acid dehydratase and E. coli dihydroxy-acid dehydratase to a similar extent suggesting the active sites of the enzymes from these two organisms are similar. Some of the better inhibitors described in this report have mild herbicidal activity.     

Defective alterations in the collagen network to prostacyclin in COPD lung fibroblasts

Background

Prostacyclin analogs are potent vasodilators and possess anti-inflammatory properties. However, the effect of prostacyclin on extracellular matrix (ECM) in COPD is not well known. Collagen fibrils and proteoglycans are essential ECM components in the lung and fibroblasts are key players in regulating the homeostasis of ECM proteins. The aim was to study the synthesis of prostacyclin and its effect on fibroblast activity and ECM production, and in particular collagen I and the collagen-associated proteoglycans biglycan and decorin.

Methods

Parenchymal lung fibroblasts were isolated from lungs from COPD patients (GOLD stage IV) and from lungs and transbronchial biopsies from control subjects. The prostacyclin analog iloprost was used to study the effect of prostacyclin on ECM protein synthesis, migration, proliferation and contractile capacity of fibroblasts.

Results

TGF-β₁ stimulation significantly increased prostacyclin synthesis in fibroblasts from COPD patients (p < 0.01), but showed no effect on fibroblasts from control subjects. Collagen I synthesis was decreased by iloprost in both control and COPD fibroblasts (p < 0.05). Conversely, iloprost significantly altered biglycan and decorin synthesis in control fibroblasts, but iloprost displayed no effect on these proteoglycans in COPD fibroblasts. Proliferation rate was reduced (p < 0.05) and contractile capacity was increased in COPD fibroblasts (p < 0.05) compared to control fibroblasts. Iloprost decreased proliferative rate in control fibroblasts (p < 0.05), whereas iloprost attenuated contraction capacity in both COPD (p < 0.01) and control fibroblasts (p < 0.05).

Conclusions

Iloprost reduced collagen I synthesis and fibroblast contractility but did not affect the collagen-associated proteoglycans or proliferation rate in fibroblasts from COPD patients. Enhanced prostacyclin production could lead to improper collagen network fibrillogenesis and a more emphysematous lung structure in severe COPD patients.     

Prostacyclin in sepsis: a systematic review

According to current literature, infective processes greatly modify both vascular hemodynamics and anti-oxidant properties of affected tissues, causing a change in homeostasis that regulates the correct functioning of all cells responsible for the physiological and metabolic balance of various organs. As a consequence, the response to the infection that has caused the change is also likely to be weaker and, in the case of septic shock, ineffective. In this review, we will take into consideration these mechanisms and then focus on a group of vasodilator drugs (prostacyclin and its analogs) which, though have been used for over 20 years mainly to treat obstructive vascular diseases, have such hemodynamic and anti-inflammatory properties which prevent homeostatic changes. It is obvious that prostacyclin does not definitively have anti-infective characteristics; however, in association with anti-infective drugs (antibiotics, etc.), the effectiveness of the latter appears improved, at least in some circumstances. Similarly, the fact that prostacyclin and its analogs have a cytoprotective effect on the liver and reduce the ischemia-reperfusion damage following liver transplant is not a novelty and evidence that they improve hepatic hemodynamics suggests their use in those pathologies characterized by possible reduced perfusion or ascertained ischemia of the liver.     

Effect of oxidation of methionine in a peptide substrate for human elastases: a model for inactivation of alpha 1-protease inhibitor.

Human α₁-protease inhibitor which is an important plasma protein, contains a methionine residue at its reactive site. A model synthetic peptide substrate, succinyl-L-alanyl-L-alanyl-L-prolyl-L-methionine p-nitroanilide, has been employed to study the effect of oxidation of methionine on the rate of hydrolysis of this substrate by human elastases. The methionine sulfoxide derivative obtained by mild oxidation of this substrate is hydrolyzed by pancreatic elastase 2 and leukocyte elastase at rates that are 5% and 0.3% of the rates measured for hydrolysis of the parent compound by the respective enzymes. These results suggest that oxidation of the active site methionine residue of human α₁-protease inhibitor may decrease the rate of reaction of pancreatic or leukocyte elastase with this inhibitor.     

Structural perspectives on H2S homeostasis

The enzymes involved in H₂S homeostasis regulate its production from sulfur-containing amino acids and its oxidation to thiosulfate and sulfate. Two gatekeepers in this homeostatic circuit are cystathionine beta-synthase, which commits homocysteine to cysteine, and sulfide quinone oxidoreductase, which commits H₂S to oxidation via a mitochondrial pathway. Inborn errors at either locus affect sulfur metabolism, increasing homocysteine-derived H₂S synthesis in the case of CBS deficiency and reducing complex IV activity in the case of SQOR deficiency. In this review, we focus on structural perspectives on the reaction mechanisms and regulation of these two enzymes, which are key to understanding H₂S homeostasis in health and its dysregulation and potential targeting in disease.     

Thermodynamic and extrathermodynamic requirements of enzyme catalysis

An enzyme's affinity for the altered substrate in the transition state (symbolized here as S) matches the value of k(cat)/K(m) divided by the rate constant for the uncatalyzed reaction in water. The validity of this relationship is not affected by the detailed mechanism by which any particular enzyme may act, or on whether changes in enzyme conformation occur on the path to the transition state. It subsumes potential effects of substrate desolvation, H-bonding and other polar attractions, and the juxtaposition of several substrates in a configuration appropriate for reaction. The startling rate enhancements that some enzymes produce have only recently been recognized. Direct measurements of the binding affinities of stable transition-state analog inhibitors confirm the remarkable power of binding discrimination of enzymes. Several parts of the enzyme and substrate, that contribute to S binding, exhibit extremely large connectivity effects, with effective relative concentrations in excess of 10(8) M. Exact structures of enzyme complexes with transition-state analogs also indicate a general tendency of enzyme active sites to close around S in such a way as to maximize binding contacts. The role of solvent water in these binding equilibria, for which Walter Kauzmann provided a primer, is only beginning to be appreciated.     

The trigonal-bipyramidal NO4 ligand set in biologically relevant vanadium compounds and their inorganic models

In the present focused review, vanadate-dependent haloperoxidases and vanadate-inhibited enzymes which catalyze the hydrolysis of phosphoester bonds are addressed. In these systems, vanadate [H_xVO₄]^(3−x)− is covalently coordinated to the imidazolyl moiety of an active site histidine, with a geometrical arrangement close to a trigonal bipyramid. The resulting ligand set, NO₄, and ligand arrangement provide peroxidase activity to the haloperoxidases and, to a certain extent, also to vanadate-inhibited phosphatases. The haloperoxidases are responsible for the oxidative halogenation of a variety of organic substrates. They are also active in other oxidation reactions relying on peroxide as the oxidant, such as the oxidative cyclizations of terpenes and, specifically, the oxygenation of (prochiral) sulfides to (chiral) sulfoxides. These functions can be modeled by vanadium complexes. Attracted interest is paid to {V(NO₄)} complexes that are functional and structural models of the peroxidases. In the vanadate-inhibited phosphatases – structural analogs of the transition state in phosphoester hydrolysis by the native enzymes – the position of the axial histidine can also be taken by cysteinate or serinate, a fact which has implications for the insulin-mimetic potential of vanadate.     

Redox‐regulated methionine oxidation of Arabidopsis thaliana glutathione transferase Phi9 induces H‐site flexibility

Glutathione transferase enzymes help plants to cope with biotic and abiotic stress. They mainly catalyze the conjugation of glutathione (GSH) onto xenobiotics, and some act as glutathione peroxidase. With X‐ray crystallography, kinetics, and thermodynamics, we studied the impact of oxidation on Arabidopsis thaliana glutathione transferase Phi 9 (GSTF9). GSTF9 has no cysteine in its sequence, and it adopts a universal GST structural fold characterized by a typical conserved GSH‐binding site (G‐site) and a hydrophobic co‐substrate‐binding site (H‐site). At elevated H₂O₂ concentrations, methionine sulfur oxidation decreases its transferase activity. This oxidation increases the flexibility of the H‐site loop, which is reflected in lower activities for hydrophobic substrates. Determination of the transition state thermodynamic parameters shows that upon oxidation an increased enthalpic penalty is counterbalanced by a more favorable entropic contribution. All in all, to guarantee functionality under oxidative stress conditions, GSTF9 employs a thermodynamic and structural compensatory mechanism and becomes substrate of methionine sulfoxide reductases, making it a redox‐regulated enzyme.     

Kinetic mechanism of quinol oxidation by cytochrome bd studied with ubiquinone-2 analogs

Cytochrome bd is a heterodimeric terminal ubiquinol oxidase of Escherichia coli under microaerophilic growth conditions. The oxidase activity shows sigmoidal concentration-dependence with low concentrations of ubiquinols, and a marked substrate inhibition with high concentrations of ubiquinol-2 analogs [Sakamoto, K., Miyoshi, H., Takegami, K., Mogi, T., Anraku, Y., and Iwamura H. (1996) J. Biol. Chem. 271, 29897-29902]. Kinetic analysis of the oxidation of the ubiquinol-2 analogs, where the 2- or 3-methoxy group has been substituted with an azido or ethoxy group, suggested that its peculiar enzyme kinetics can be explained by a modified ping-pong bi-bi mechanism with the formation of inactive binary complex FS in the one-electron reduced oxygenated state and inactive ternary complex (E2S)S(n) on the oxidation of the second quinol molecule. Structure-function studies on the ubiquinol-2 analogs suggested that the 6-diprenyl group and the 3-methoxy group on the quinone ring are involved in the substrate inhibition. We also found that oxidized forms of ubiquinone-2 analogs served as weak noncompetitive inhibitors. These results indicate that the mechanism for the substrate oxidation by cytochrome bd is different from that of the heme-copper terminal quinol oxidase and is tightly coupled to dioxygen reduction chemistry.     

STRUCTURAL CHANGES IN ISOLATED LIVER MITOCHONDRIA OF RATS DURING ESSENTIAL FATTY ACID DEFICIENCY

Liver mitochondria isolated in 0.44 M sucrose from rats deficient in essential fatty acids (EFA) oxidized citrate, succinate, α-ketoglutarate, glutamate, and pyruvate at a faster rate than did mitochondria isolated from normal rats; however, the oxidation of malate, caprylate, and β-hydroxybutyrate was not significantly increased. The mitochondria from deficient rats exhibited an increased ATPase activity and extensive structural damage as revealed by electron microscope examination of thin sections. An increase in citrate oxidation and ATPase activity, together with some structural damage, could be demonstrated as early as the 4^th week in rats on a fat-free diet. Saturated fat in the diet did not prevent the change in mitochondrial structure but accelerated its appearance. Both the biochemical and structural defects could be reversed within three weeks after feeding deficient rats a source of EFA. In the presence of a phosphate acceptor the effect of EFA deficiency on substrate oxidation was largely eliminated. A trend toward a reduced efficiency of oxidative phosphorylation was noted in mitochondria from EFA-deficient rats, but significant uncoupling was found only in the case of citrate, β-hydroxybutyrate, and glutamate in the presence of malonate. Together with the increased ATPase activity, the uncoupling of phosphorylation could account for the poor respiratory control found with the deficient preparation. However, EFA deficiency was without effect on the respiration of liver slices, which supports the belief that the observed changes in oxidation and phosphorylation are an artifact resulting from damage sustained by the deficient mitochondria during their isolation.     

Synthesis and chiroptical characterization of prostacyclin diastereomers

In the mercuri- and halo-cyclizations of PGF₂α methyl ester and its 11,15-bis(α-ethoxyethyl)-ether (or other protected forms) the exo-PGI₁ derivative predominates independent of reagent and degree of protective of the PGF₂α sample used. Diastereomerically pure samples of exo- and endo-PGI₁ and prostacyclin (PGI₂) were prepared. PGI₀ epimers were prepared: catalytic hydrogenation of PGI₂ Me ester provides exclusively the endo isomer. PGI₂ methyl ester was found to be stable to extensive chromatography on silica, and to storage for at least a year in anhydrous ethanol at −20°C. At pH 7.4 in 2:1 H₂O:EtOH, the ester has a half-life in excess of 5 hr at 25°C. A reproducible small scale (0.4–3 mg) synthesis of prostacyclin uses a modification of Whittaker's iodocyclization followed by DBN treatment. This procedure, developed with 15-³H-PGF₂α, proved widely applicable to PGF₂α analogs and diastereomers. The following prostacyclins (in the Me ester and Na salt forms) bearing the 5-en-6-yl ether unit were prepared in this way: ent-PGI₂, rac-PGI₂, 15-epi-PGI₂, ent-15-epi-PGI₂, 11-epi-PGI₂, 8,9,12-epi-PGI₂, -PGI₂, 13,14-dihydro-PGI₂, and 13,14-dihydro-15-epi-PGI₂. NMR comparisons for the methyl esters reveal that of the resonances (H-5,9,11,15) that appear at δ4.0±0.6 ppm, the most deshielded is H-9 so long as the 5,6-olefin is . The 8 ,9 -6,9-oxido- -5,6-ene unit is most readily characterized by its strong positive dichroic absorption at 210–230 nm. CD spectroscopy not only serves to confirm the presence of this unit in analogs, but also can be used for quantitative analysis of PGI₂ solutions and for monitoring the rate of hydrolytic cleavage of these enol ethers.