9-Hydroxyprostaglandin dehydrogenase from rat kidney. Purification to homogeneity and partial characterization

Characterization of aldo-keto reductase 1C subfamily members encoded in two rat genes (akr1c19 and RGD1564865). Relationship to 9-hydroxyprostaglandin dehydrogenase

Rat genes, akr1c19 and RGD1564865, encode members (R1C19 and 20HSDL, respectively) of the aldo-keto reductase (AKR) 1C subfamily, whose functions, however, remain unknown. Here, we show that recombinant R1C19 and 20HSDL exhibit NAD⁺-dependent dehydrogenase activity for prostaglandins (PGs) with 9α-hydroxy group (PGF_2α, its 13,14-dihydro- and 15-keto derivatives, 9α,11β-PGF₂ and PGD₂). 20HSDL oxidized the PGs with much lower K_m (0.3-14 μM) and higher k_cat/K_m values (0.064-2.6 min^-1μM^-1) than those of R1C19. They also differed in other properties: R1C19, but not 20HSDL, oxidized some 17β-hydroxysteroids (5β-androstane-3α,17β-diol and 5β-androstan-17β-ol-3-one). 20HSDL was specifically inhibited by zomepirac, but not by R1C19-selective inhibitors (hexestrol, flavonoids, ibuprofen and flufenamic acid), although the two enzymes were sensitive to indomethacin and cis-unsaturated fatty acids. The mRNA for 20HSDL was expressed abundantly in rat kidney and at low levels in the liver, testis, brain, heart and colon, in contrast to ubiquitous expression of R1C19 mRNA. The comparison of enzymic features of R1C19 and 20HSDL with rat PG dehydrogenases and other AKRs suggests not only a close relationship of 20HSDL with 9-hydroxy-PG dehydrogenase in rat kidney, but also roles of R1C19 and rat AKRs (1C16 and 1C24) in the metabolism of PGF_2α, PGD₂ and 9α,11β-PGF₂ in other tissues.

High-pressure liquid chromatography of oxo-eicosanoids derived from arachidonic acid

Eicosanoids are a large group of biologically active metabolites of arachidonic acid and related C20 fatty acids. Many of these compounds contain hydroxyl groups which can be converted to oxo groups by a variety of substrate-specific dehydrogenases. In many cases, this results in a reduction in potency, but in others, such as the oxidation of 5-hydroxyeicosatetraenoic acid to its oxo metabolite 5-oxo-eicosatetraenoic acid, there is a dramatic increase in biological activity. Thus, it is often very important to analyze the relative amounts of oxo- and hydroxy-eicosanoids formed by various cells and tissues. The present study was designed to compare the chromatographic behavior of oxo-eicosanoids and their hydroxy counterparts in commonly used mobile phases for reversed-phase and normal-phase HPLC. We examined three groups of eicosanoids: prostaglandins, leukotriene B4 and some of its metabolites, and monohydroxy-eicosanoids and their oxo metabolites. We found that in reversed-phase HPLC, the retention times of oxo-eicosanoids were longer than those of the corresponding hydroxy-eicosanoids in mobile phases containing acetonitrile as the major organic component, whereas the reverse was true for mobile phases containing methanol. Normal-phase HPLC using mobile phases containing hexane, isopropanol, and acetic acid gave excellent separation of oxo- and hydroxy-eicosanoids. Increasing the concentration of acetic acid in the mobile phase selectively reduced the retention times of oxo-eicosatetraenoic acids compared to monohydroxy-eicosatetraenoic acids, whereas the reverse was true for isopropanol. Differences in the chromatographic behavior of oxo- and hydroxy-eicosanoids can be useful clues in the structural characterization of these compounds, as illustrated by the chromatographic properties of a complex series of LTB4 metabolites formed by rat neutrophils.

Purification and characterization of an NAD(+)-dependent dehydrogenase that catalyzes the oxidation of thromboxane B2 at C-11 from porcine liver. Development and application of 11-dehydro-thromboxane B2 radioimmunoassay to enzyme assay

11-Dehydro-thromboxane B2 has been identified as a major metabolite of infused as well as endogenous thromboxane B2 in mammalian plasma and urine. This metabolite is derived from thromboxane B2 by enzymatic oxidation at C-11 catalyzed by 11-hydroxythromboxane B2 dehydrogenase. A radioimmunoassay for 11-dehydro-thromboxane B2 has been developed and used for enzyme assay, purification and characterization. Antibodies were generated against 11-dehydro-thromboxane B2 conjugated to bovine thyroglobulin. Labeled marker was prepared by radioiodinating 11-dehydro-thromboxane B2-tyrosine methyl ester conjugate. A sensitive radioimmunoassay capable of detecting 10 pg of 11-dehydro-thromboxane B2 per assay tube was developed. The antibodies showed minimal crossreaction with thromboxane B2 (0.03%), prostaglandin D2 (2.76%) and other eicosanoids (less than 0.03%). The enzyme activity was determined by assaying NAD(+)-dependent formation of immunoreactive 11-dehydro-thromboxane B2 from thromboxane B2. The enzyme was found to be enriched in liver although significant activity was also detected in gastrointestinal tract and kidney in pig. The enzyme was purified from porcine liver cytosol to apparent homogeneity using conventional and affinity chromatography. The purified enzyme exhibited coenzyme specificity for NAD+ and used thromboxane B2 as a substrate. The enzyme also catalyzes NADH-dependent reduction of 11-dehydro-thromboxane B2 to thromboxane B2 indicating the reversibility of the enzyme catalyzed reaction. The apparent Km values for thromboxane B2, 11-dehydro-thromboxane B2 and NAD+ are 8.1, 8.0 and 23 microM, respectively. Subunit Mr was shown to be 55,000, whereas the native enzyme Mr was found to be 110,000 indicating that the enzyme is a dimer. The enzyme is sensitive to sulfhydryl inhibitions suggesting cysteine residues are essential to enzyme activity. The availability of a homogeneous enzyme preparation should allow further studies on the substrate specificity and the structure and function of the enzyme.

Organ selective conversion of prostaglandin D2 to 9 alpha, 11 beta-prostaglandin F2 and its subsequent metabolism in rat, rabbit and guinea pig

Cell-free 100,000 g supernatants from liver, kidney, lung and caecum of rat, rabbit and guinea-pig were compared for their ability to transform prostaglandins F2 alpha, D2, E2 and 9 alpha, 11 beta-prostaglandin F2 (11epi-PGF 2 alpha) to metabolic products. Experiments utilized multitritiated substrate PGs, with assessment of biotransformation by TLC, HPLC and GC/MS. PGF2 alpha was converted via the sulphasalazine analogue-inhibitable NAD+-dependent 15-hydroxy-prostaglandin dehydrogenase pathway (15-PGDH), with high activity (greater than 5 pmol/min/mg protein) in all 12 systems except rat and rabbit liver (e.g. guinea-pig kidney and rat caecum both 64 pmol/min/mg; rat liver 0.3 pmol/min/mg), forming 15-keto and 13,14-dihydro-15-keto metabolites as determined by TLC, HPLC and GC/MS. Prostaglandin D2 was not transformed in similar fashion in NAD+- or NADP+-supplemented incubations in any of the 12 cytosolic systems. However, PGD2 was converted to a single product identified by TLC, HPLC and GC/MS as 9 alpha, 11 beta-PGF2 in certain of the systems when supplemented with an NADPH regenerating system, with high activity in guinea-pig kidney (55.0 pmol/min/mg), guinea-pig liver (27.5 pmol/min/mg) and rabbit liver (13.7 pmol/min/mg) and less than 5 pmol/min/mg in 8 of the remaining 9 systems. This stereospecific 11-ketoreductase of rabbit and guinea-pig liver was stable to 10 min heating at 50 degrees, dialysis, storage at -20 degrees and repeated freeze/thawing but was not inhibited by sulphasalazine analogues. The 11-ketoreductase had a markedly different tissue profile from PGE2 9-ketoreductase, which was shown to convert PGE2 stereospecifically to 9 alpha, 11 alpha-prostaglandin F2 (PGF2 alpha) and was present at highest activity in rabbit liver and kidney. Evidence was obtained that 9 alpha, 11 beta-PGF2 was actively transformed by the sulphasalazine-inhitable 15-PGDH pathway at approximately one third of the rate of PGF2 alpha with high activity in several cytosolic systems (e.g. rat caecum, guinea-pig liver and kidney), suggesting that further transformation in vivo of this biologically active product of PGD2 metabolism could be initiated by this route.

Inhibition of the human placental NAD- and NADP-linked 15-hydroxyprostaglandin dehydrogenases by nonsteroidal anti-inflammatory drugs

A number of nonsteroidal anti-inflammatory drugs are non-competitive or mixed inhibitors of human placental NAD- and NADP-linked 15-hydroxyprostaglandin dehydrogenases. Cis- and trans-sulindac sulfide and cis- and trans-sulindac inhibit the NAD-linked enzyme as well or better than they inhibit various cyclooxygenases in vitro. The remainder of the compounds tested are at least one order of magnitude less effective as inhibitors of the 15-hydroxyprostaglandin dehydrogenases than they are as inhibitors of cyclooxygenases. Cis- and trans-sulindac sulfide are sufficiently strong inhibitors of the NAD-linked enzyme (Kis of 7.8 microM and 6.8 microM respectively) to raise the possibility that they might also inhibit this enzyme in vivo.

Bovine placental prostaglandin synthesis in vitro as it relates to placental separation

Bovine placentomes were collected during late gestation, prepartum and immediately postpartum. Postpartum tissues were collected prior to fetal membrane separation. Maternal and fetal placentomal components each were examined for their ability to synthesize prostaglandins (PG's) from arachidonic acid (AA) and metabolize PGF2 alpha and PGE2 in vitro. Maternal placental PG synthesis was lower (P less than .05) than that for fetal placental tissue and was primarily PGF's. Fetal placental PG synthesis increased (P less than .05) prepartum and was primarily PGE's. Fetal placental PGE production predominated (P less than .05) postpartum if the fetal membranes were retained, while PGF production predominated (P less than .05) if the membranes were released. Maternal and fetal placental tissues were unable to convert PGE2 to PGF2 alpha (P greater than .05). Postpartum fetal placental tissue was able to convert PGF2 alpha to PGE2 (P less than .05) if the fetal membranes were retained but not if the membranes were released (P greater than .05). These results indicate that fetal placental synthesis of PGF's may be related to placental membrane separation. The shift in fetal placental PG production from PGE's to PGF's may be due to a cessation of the ability of released fetal tissue to convert PGF2 alpha to PGE2.

Prostaglandin dehydrogenase activity of purified rat liver 3 alpha-hydroxysteroid dehydrogenase

Homogeneous 3 alpha-hydroxysteroid dehydrogenase (3 alpha-HSD) from rat liver cytosol displays 9, 11, and 15-hydroxyprostaglandin dehydrogenase activity. Using [14C]-PGF2 alpha as substrate the products of this reaction were separated by TLC and identified by autoradiography as PGE2 and PGB2. The purified enzyme catalyzes this reaction at a rate 200 times faster than cytosol. This corresponds to the rate enhancement observed when the enzyme is purified from cytosol using androsterone (a 3 alpha-hydroxysteroid) as substrate and suggests that it may represent a major 9-hydroxyprostaglandin dehydrogenase in this tissue. Although the 3 alpha-HSD has many properties in common with the 9-hydroxyprostaglandin dehydrogenase of rat kidney, rat kidney contains no protein that is immunodetectable with polyclonal antibody raised against the purified 3 alpha-HSD.

Rat liver 3 alpha-hydroxysteroid dehydrogenase

3 alpha-HSD appears to be a multifunctional enzyme. In addition to its traditional role of catalyzing early steps in androgen metabolism, it will also oxidoreduce prostaglandins and detoxify trans-dihydrodiols (proximate carcinogens). Since these novel reactions have been quantified using homogeneous enzyme it is necessary to interpret the role of the enzyme in these processes in vivo with some caution. However, it is rare that such observations on a purified hydroxysteroid dehydrogenase have led to such important questions. Is the 3 alpha-HSD the only steroid dehydrogenase that transforms prostaglandins and trans-dihydrodiols? Are hydroxysteroid dehydrogenases and prostaglandin dehydrogenases the same enzymes in certain tissues? Does 3 alpha-HSD protect against chemical carcinogenesis in vivo? The inhibition of the purified dehydrogenase by therapeutically relevant concentrations of anti-inflammatory drugs also deserves comment. Is this hydroxysteroid dehydrogenase really an in vivo target for anti-inflammatory drug action? Could these drugs exert some of their pharmacological effect either by preventing glucocorticoid metabolism in some tissues or by preventing the transformation of PGF2 alpha (non-inflammatory prostanoid) to PGE2 (a pro-inflammatory prostanoid)? Could these drugs, by inhibiting trans-dihydrodiol oxidation, potentiate the initiation of chemical carcinogenesis? These and other important questions can be answered only by developing specific inhibitors for the dehydrogenase to decipher its function in vivo.

Age-dependent changes in the synthesis and catabolism of 6 oxo PGE1 and other prostanoids by the rat kidney in vitro

Synthesis and catabolism of 6 oxo PGE1 was assessed in 100,000 g cell-free supernatant fractions of kidneys obtained from rats aged 20, 34 and 70 days. In addition the release of PGI2, TxA2 (measured as 6 oxo PGF1 alpha and TxB2, respectively), PGE2 and PGF2 alpha from kidney slices prepared from these three groups of rats was determined using specific radioimmunoassays. The conversion of PGI2 to 6 oxo PGE1 (but not 13,14 dihydro 15 oxo PGF2 alpha to 13,14 dihydro 15 oxo PGE2) was detected in supernatant fractions of kidneys from 20 day rats. Slices prepared from the kidneys of these animals spontaneously released significant amounts of three prostanoids (6 oxo PGF1 alpha greater than PGE2 greater than PGF2 alpha greater than TxB2 = 0). No formation of 6 oxo PGE1 from exogenous PGI2 was demonstrated in renal 100,000 g supernates from 34 and 70 day rats even though these supernates avidly oxidised 13,14 dihydro 15 oxo PGF2 alpha to 13,14 dihydro 15 oxo PGE2. In these animals the rank order of prostanoid release from kidney slices was PGE2 greater than 6 oxo PGF1 alpha greater than PGF2 alpha greater than TxB2 = 0. The catabolism of 6 oxo PGE1 is also age-dependent. In 20 and 34 day old rats 6 oxo PGE1 and PGE1 incubated with renal 100,000 g supernates undergo loss of biological activity as determined by the ability to inhibit ADP induced human platelet aggregation. In contrast, kidney 100,000 g supernates prepared from 70 day rats convert 6 oxo PGE1 to an unidentified metabolite with more potent anti-aggregatory activity. The possibility that 6 oxo PGE1 has a biological role in the developing rat kidney is discussed.

Formation of 6-keto prostaglandin E1 in mammalian kidneys

1 The metabolism of prostacyclin (PGI2) and 6-keto prostaglandin F1 alpha (6-keto PGF1 alpha) was studied in cell-free homogenates of rat, rabbit and guinea-pig kidney. 2 Rabbit kidney converted both PGI2 and 6-keto PGF1 alpha to a stable metabolite with chromatographic and biological activity identical to that of authentic 6-keto PGE1. Activity was found in the kidney cortex but not medulla, was inhibited by NAD+ or NADP+ (5 mM) and showed an optimum temperature requirement of 37 degrees C. 3 Guinea-pig kidney converted PGI2 but not 6-keto PGF1 alpha to a labile, biologically active metabolite which was not 6-keto pge1. 4 No conversion of prostacyclin or 6-keto PGF1 alpha to biologically active metabolites occurred in cell-free homogenates of rat kidney, liver and colon or guinea-pig liver and colon. 5 6-keto PGE1 rapidly lost spasmogenic activity on the rat stomach strip following incubation with rabbit or guinea-pig kidney supernatant in the absence of added cofactors. No loss of activity occurred on incubation with rat kidney. 6 Rutin (50 microM) potently inhibited synthesis of 6-keto PGE1 from added PGI2 by rabbit kidney cortex. This reaction was potentiated by a similar concentration of sulphasalazine, carbenoxolone, imidazole, papaverine or indomethacin. 7 The relevance of these findings for the possible physiological and pathological roles of 6-keto PGE1 in the kidney is discussed.

A radioimmunoassay for 6-ketoprostaglandin E1

A radioimmunoassay for 6-ketoprostaglandin E1 has been developed. 6-keto-prostaglandin E1 antibodies were produced in rabbits by repeated immunization with 6-ketoprostaglandin E1 coupled to bovine serum albumin. [125]-labeled hapten with high specific radioactivity was prepared by radioiodination of 6-ketoprostaglandin in E1-tyrosine methyl ester conjugate followed by purification with thin layer chromatography. The antibodies showed good specificity toward 6-ketoprostaglandin E1 and crossreacted only significantly with prostaglandin E1. The sensitivity of the assay was 10 pg per assay tube. Application of the radioimmunoassay was demonstrated by the detection of immunoreactive 6-ketoprostaglandin E1 from 6-ketoprostaglandin F1 alpha catalyzed by swine renal NADP+-linked 9-hydroxyprostaglandin dehydrogenase.

Stimulation of renin release by 6-oxo-prostaglandin E1 and prostacyclin

1 Renin release induced by 6-oxo-prostaglandin E1 (6-oxo-PGE1) was compared to release in response to prostacyclin (PGI2) and 6-oxo-PGF1 alpha in slices of rabbit renal cortex. 2 Krebs-Ringer medium bathing slices of renal cortex was collected for renin assay after four successive 20 min intervals (periods I-IV). Renin release did not increase during periods I to IV in untreated slices. Agonists were added, only once, at the beginning of period III. Between periods III and IV, the incubation solution was aspirated and replaced with fresh medium. 3 PGI2 increased renin release during period III while 6-oxo-PGE1 stimulated release during periods III and IV. 6-oxo-PGE1 stimulated renin release (24%-74%) in concentrations ranging from 1-33 microM while PGI2 stimulated release at 10 microM (60%) but not at 5 microM. 6-oxo-PGF1 alpha, 10 microM, did not release renin during period III (period III, 9%), but caused a small rise in period IV (29%). 4 6-oxo-PGE1, unlike PGI2, was stable under the incubation conditions (pH 7.4, 37 degrees C) as indicated by recovery of undiminished platelet anti-aggregatory material after 20 min. 5 In the rabbit kidney, activity of 9-hydroxyprostaglandin dehydrogenase was greatest in the cortex and negligible in the papilla, corresponding to the zonal distribution of renin. 6 The prominent and sustained in vitro renin releasing effect of 6-oxo-PGE1, as well as the cortical localization of enzyme activity capable of generating this stable prostacyclin metabolite, suggest that formation of 6-oxo-PGE1 may contribute to PGI2-induced renin release and may explain the delayed stimulation caused by 6-oxo-PGF1 alpha.