Identification of 9-cis-retinoic acid, 9,13-di-cis-retinoic acid, and 14-hydroxy-4,14-retro-retinol in human plasma after liver consumption

Vitamin A is a well-established teratogen in several animal species. Case reports as well as a recent epidemiological study suggest that vitamin A intake in excess of 25,000 or 10,000 IU respectively, can result in retinoid-specific defects in the offspring. A single meal of liver contains, on the average, a 10- to 20-fold higher amount of vitamin A than what is already suspected to be teratogenic. To evaluate the risk of liver consumption during pregnancy, we have studied levels of vitamin A and a number of potentially active retinoid metabolites in plasma of ten healthy male volunteers following consumption of fried turkey liver (2 g raw weight/kg body weight). HPLC, UV spectroscopy and mass spectrometry were used for identification and quantitation of retinoids in plasma. As shown previously, vitamin A intake via liver consumption resulted in greatly increased plasma levels of 13-cis-retinoic acid (13-cis-RA) and 13-cis-4-oxo-RA, and low levels of all-trans-RA and all-trans-4-oxo-RA. In our present investigation 9-cis-RA, 9,13-di-cis-RA, and 14-hydroxy-4,14-retro-retinol (14-HRR) were identified for the first time in humans as physiological metabolites of vitamin A. 9-cis-RA is a potent teratogen as well as a high affinity ligand of retinoid receptors, and 14-HRR was previously shown to promote lymphocyte activation in vitro. The present study bears on the issue of a possible teratogenic risk of liver consumption, as active retinoids were identified in human plasma, and their levels could be related to previous human studies as well as to experimental studies in sensitive animal species.

Identification of 9,13-dicis-retinoic acid as a major plasma metabolite of 9-cis-retinoic acid and limited transfer of 9-cis-retinoic acid and 9,13-dicis-retinoic acid to the mouse and rat embryos

9-Cis-retinoic acid (9-cis-RA) has been proposed to be the endogenous ligand of retinoid X receptors. We examined the plasma pharmacokinetics of 9-cis-RA and its metabolites in nonpregnant female NMRI mice after oral dosing with 50 mg 9-cis-RA/kg body weight. Furthermore, we studied the metabolism of 9-cis-RA and its transfer to the embryo following oral administration of the precursor 9-cis-retinaldehyde (9-cis-RAL; 100 mg/kg body weight) to pregnant mice and rats on gestational days 11 and 13, respectively. Following 9-cis-RA administration, plasma levels of 9-cis-RA reached their maximum within 40-60 min and then declined in a monoexponential manner with an apparent half-life of 64 +/- 32 min. A great variety of polar metabolites of 9-cis-RA was found; among them, the beta-glucuronides of 9-cis-RA (9-cis-RAG) and of 9-cis-4-oxo-RA (9-cis-4-oxo-RAG) could be identified. A further prominent polar metabolite of 9-cis-RA in mouse plasma was shown to be an additional RA isomer (distinct from 13-cis-RA and all-trans-RA) whose concentrations weeesimilarly high as those of 9-cis-RA.(ABSTRACT TRUNCATED AT 250 WORDS)

Exposure of newborn infants to di-(2-ethylhexyl)-phthalate and 2-ethylhexanoic acid following exchange transfusion with polyvinylchloride catheters

Infants in the neonatal intensive care unit are regularly exposed to the plasticizer di-(2-ethylhexyl)-phthalate (DEHP) following exchange transfusion or extracorporeal membrane oxygenation. Whether such exposure leads to increased morbidity is not known, although elevated levels of DEHP have been associated with necrotizing enterocolitis and cholestasis. The hypothesis that infants undergoing exchange transfusion are exposed to toxic levels of DEHP and the presumed metabolite 2-ethylhexanoic acid (EHXA) was tested by measuring serum levels of DEHP in 16 newborn infants (gas-liquid-chromatography) and urine concentrations of EHXA in 6 of these infants (gas chromatography-mass spectrometry). DEHP levels were undetectable (< 1 microgram/mL) before exchange but ranged from 6.1 to 21.6 micrograms per mL of serum (average, 12.5 +/- 6.2 micrograms/mL) after a single exchange transfusion. DEHP uptake did not result in cholestasis. EHXA peak levels were 127 to 416 ng per mL of urine, with a median of 174 ng per mL. Concentrations of EHXA were lower than anticipated, which indicates that EHXA is not a major metabolite in the neonatal infant.

Effects of valproate and E-2-en-valproate on functional and morphological parameters of rat liver. II. Influence of phenobarbital comedication

The effect of phenobarbital on the potential hepatotoxicity of E-2-en-valproate (E-2-en-VPA; trans-2-en-VPA) and VPA was studied in young male Sprague-Dawley rats. E-2-en-VPA and VPA were administered daily at 750 mg/kg i.p. (divided into three doses a day) for 7 consecutive days. Phenobarbital was coadministered i.p. once daily at 100 mg/kg for 2 days, followed by daily injections of 50 mg/kg for the subsequent days of the treatment period. Additional groups of rats were treated with phenobarbital alone or received once daily administration of 4-en-VPA (100 mg/kg), a potentially hepatotoxic metabolite of VPA. Clinical chemistry data were studied before and after the period of treatment. Furthermore, drug and metabolite levels were analyzed by gas chromatography-mass spectrometry. Treatment with VPA and phenobarbital resulted in deaths and histopathological liver alterations, such as marked microvesicular steatosis and degenerative lesions, whereas no death and hepatotoxicity occurred in rats treated with E-2-en-VPA and phenobarbital. Furthermore, hyperammonemia was recorded in VPA- but not E-2-en-VPA-treated rats. In comparison to treatment with VPA or E-2-en-VPA alone, combined treatment with phenobarbital markedly reduced plasma levels of the parent drugs and metabolites originating from beta-oxidation, but, in case of VPA, increased metabolites originating from omega-oxidation. Plasma levels of 4-en-VPA were increased by phenobarbital in VPA-treated rats, but 4-en-VPA was not detectable in rats treated with E-2-en-VPA. The most severe alterations in functional and morphological liver parameters were found in rats treated with 4-en-VPA. In these animals, the extent of steatosis was significantly correlated with plasma levels of 4-en-VPA, but not its major metabolite 2,4-dien-VPA. Plasma levels of 4-en-VPA or its major metabolite 2,4-dien-VPA in rats without steatosis were markedly higher than levels of these compounds in VPA-treated rats with steatosis, suggesting that 4-en-VPA and 2,4-dien-VPA are not critically involved in the hepatotoxic effects of VPA. The data substantiate that E-2-en-VPA is less hepatotoxic than VPA and may thus offer advantages for antiepileptic therapy.

Valproate (VPA) metabolites in various clinical conditions of probable VPA-associated hepatotoxicity

Of a cohort of 470 epileptic patients in whom valproate (VPA) serum metabolites had been measured, 170 subjects without symptoms or signs of hepatic side effects were chosen as a reference group to establish the usual metabolic pattern. A wide interindividual variation of VPA metabolite concentrations was noted. Infants receiving VPA monotherapy and comedication with other antiepileptic drugs (AEDs) showed lower concentrations of the potential hepatotoxin 4-ene-VPA than did older children. In 11 patients with early symptoms and signs of possible fatal VPA-associated hepatotoxicity, the following spectrum of benign clinical conditions was observed: unusually severe side effect during initiation of VPA therapy (1 patient), high VPA dosage (2 patients), reversible impairment of coagulation with bleeding manifestations in association with a slight increase in transaminase levels (1 child), and reversible liver dysfunction associated with febrile illness (7 patients). Reversible or irreversible fulminant liver failure had occurred in 5 children. Three of the 4 children with a fatal outcome had massive lactic acidosis. In all patients with probable VPA-associated hepatotoxicity, some aspects of VPA metabolism differed distinctly from that of the reference group, but the inter-individual profile of metabolites varied considerably, even in the subgroup of 4 children who died. Impairment of VPA beta-oxidation and increase of metabolites of alternative metabolic pathways (omega- and omega 1-hydroxylation, dehydrogenation reactions) were the most frequent findings. Increased values of 2-n-propyl-4-pentenoic acid metabolite of VPA (4-ene-VPA), could be detected only in 1 of the 5 patients with fulminant liver failure and in one other child with a slight hepatic dysfunction, indicating that this VPA metabolite is not the decisive hepatotoxin or indicator of hepatotoxicity. Because we cannot distinguish between benign and life-threatening hepatic adverse reactions on the basis of VPA metabolites, all identified changes are considered secondary to an as-yet-unknown primary metabolic event. The most toxic compound could be VPA itself, which may unmask an inborn or an acquired metabolic defect in the processing of fatty acids.

Effects of valproate and E-2-en-valproate on functional and morphological parameters of rat liver. I. Biochemical, histopathological and pharmacokinetic studies

E-2-en-Valproate (E-2-en-VPA; trans-2-en-VPA) and VPA were studied for potential hepatotoxicity in young male Sprague-Dawley rats. Both drugs were administered daily at 750 mg/kg i.p. (divided into three doses a day) for 7 consecutive days. Clinical chemistry parameters were studied before and after the period of treatment. Furthermore, the drug pharmacokinetics and metabolism were analyzed at onset and end of the prolonged administration. Treatment with VPA induced hyperammonemia and other alterations in liver function tests which were not observed after treatment with E-2-en-VPA, although plasma levels of both drugs were comparable. The pharmacokinetics of VPA and E-2-en-VPA in young rats were similar, but analysis of metabolites by gas chromatography-mass spectrometry indicated marked differences in the metabolite profile, e.g., a lack of the suspected hepatotoxic metabolite 4-en-VPA in plasma of rats treated with E-2-en-VPA. Histopathological examination of liver sections showed that VPA and E-2-en-VPA did not induce degenerative liver lesions or significant alterations in hepatic content and distribution of lipids and glycogen at the doses administered. Only one of the VPA treated rats showed fatty infiltration (microvesicular steatosis). The data demonstrate that, although E-2-en-VPA is more potent than VPA as an anticonvulsant in rats, it does not exert more potent hepatotoxic effects and does not alter ammonia metabolism. Thus the data substantiate previous experimental studies that E-2-en-VPA might be a valuable substitute for VPA.

Quantitative determination of valproic acid and 14 metabolites in serum and urine by gas chromatography/mass spectrometry

A method for the determination of the antiepileptic drug valproic acid and 14 of its metabolites in serum and urine by gas chromatography/mass spectrometry with selected ion monitoring of the trimethylsilylated derivatives has been developed. Sample preparation, including hydrolysis of VPA-conjugates and removal of urea in urine is carried out at pH 5.0 and is rapid and simple. The samples are extracted with ethyl acetate and the concentrated extracts are trimethylsilylated. Analysis with adequate separation of metabolites is achieved with a DB 1701 fused silica (Megabore) capillary column. The method exhibits high recovery and reproducibility and is sufficiently sensitive and selective for analysis of small sample volumes. Application of the method for screening patient serum and urine samples for unusual metabolite patterns, with possible predictive value for early detection of liver injury, is presented.

Valproate metabolites in serum and urine during antiepileptic therapy in children with infantile spasms: abnormal metabolite pattern associated with reversible hepatotoxicity

The purpose of this study was to identify abnormal metabolite patterns of valproate (VPA) as possible early indicators of VPA-induced liver toxicity. In a prospective study, we determined serum and urine levels of VPA metabolites by gas chromatography-mass spectrometry (GC-MS) during the course of therapy in 25 children treated for infantile spasms with high VPA doses (less than or equal to 100 mg/kg body weight/day). Most patients had similar metabolite profiles: The main metabolites in serum were the beta-oxidation products (2-en-VPA and 3-keto-VPA) and the major diunsaturated metabolite 2,3'-dien-VPA. Glucuronide conjugates and the oxidation products represent the most abundant metabolites in urine. Other metabolites, including the potential hepatotoxin 4-en-VPA, were detected only in low concentrations. Two children had transiently aberrant metabolite profiles, indicating altered beta-oxidation, (levels of 2-en-VPA, 2,3'-dien-VPA, and 3-en-VPA were markedly increased) in connection with hepatomegaly and increased liver enzyme activities at a time when both had febrile infections and were receiving dexamethasone comedication. At no time were increased levels of 4-en-VPA or its derivatives detected. Establishing the VPA metabolite profile may aid in evaluation of patients who show signs and symptoms of liver dysfunction during VPA therapy. The present study shows that initial stages of hepatotoxicity reactions to VPA may be accompanied by characteristic changes in VPA metabolism; early detection of such abnormal metabolite patterns might decrease the risk of severe hepatic injury.

Transplacental distribution of weak acids in mice: accumulation in compartments of high pH

The transplacental distribution of three weak acids was measured in mice on gestational day 11. The ratio of drug concentration in maternal, embryonic, and extraembryonic compartments compared to maternal plasma followed a consistent hierarchy such that amniotic fluid greater than or equal to exocoelomic fluid greater than embryo greater than embryo plasma greater than maternal skeletal muscle. This distribution pattern correlates well with pH in these compartments and suggests that pH gradients between compartments are an influential factor in determining weak acid drug disposition during pregnancy.

Plasma pharmacokinetics and metabolism of 13-cis- and all-trans-retinoic acid in the cynomolgus monkey and the identification of 13-cis- and all-trans-retinoyl-beta-glucuronides. A comparison to one human case study with isotretinoin

In order to compare the disposition and metabolism of 13-cis-retinoic acid (13-cis-RA) and all-trans-retinoic acid (all-trans-RA) in the nonpregnant female cynomolgus monkey, the plasma concentrations of the parent compound, the oxidized metabolites 4-oxo-13-cis-retinoic acid and 4-oxo-all-trans-retinoic acid, and the conjugate metabolites 13-cis-retinoyl-beta-glucuronide (13-cis-RAG) and all trans-retinoyl-beta-glucuronide (all-trans-RAG), were determined on day 1 and day 10 after oral dosing of 2 and 10 mg 13-cis- and all-trans-RA/kg/day. Both 13-cis-RAG and all-trans-RAG have been identified as major plasma metabolites in these studies using thermospray/HPLC/mass-spectrometry of the intact conjugates. AUC comparisons from 0-24 hr after administration indicated that 13-cis-RA treatment resulted in primarily cis metabolites and all-trans-RA treatment resulted in primarily trans metabolites, although low levels of isomerization products were observed. Comparison of the two doses (2 and 10 mg/kg, po) revealed that the AUCs were proportional to the dose administered. Although qualitatively similar, elimination of 13-cis-RA in the monkey was more rapid than in the human, and approximately a 10-fold greater dose of 13-cis-RA was required in the monkey to produce the AUC values comparable to the human. The elimination of all-trans-RA in monkey was faster than that of 13-cis-RA and tended to increase with repeated dosing.(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of oxidized and glucuronidated metabolites of retinol in monkey plasma by thermospray liquid chromatography/mass spectrometry

Eight metabolites of retinol were isolated by high-performance liquid chromatography (HPLC) from the plasma of the non-human primate Macaca fascicularis after acute exposure to 150,000 IU of vitamin A per kilogram body weight. After enrichment and further chromatographic purification, the metabolites were reinjected individually into a second HPLC system which was connected on-line by a thermospray interface to a mass spectrometer operated in the positive ionization mode. Six retinoids were identified by (i) a comparison of their retention times with those of appropriate reference compounds in the two chromatographic systems and (ii) by comparison of their mass spectra with those of reference compounds. These retinoids were: 13-cis-4-oxoretinoic acid, all-trans-4-oxoretinoic acid, 13-cis-retinoic acid, all-trans-retinoic acid, all-trans-retinoyl beta-glucuronide and all-trans-retinyl beta-glucuronide. One further metabolite could be identified for the first time as all-trans-4-oxoretinoyl beta-glucuronide by its mass spectrum and, after treatment of the unknown metabolite with beta-glucuronidase, by its hydrolysis product all-trans-4-oxoretinoic acid. The molecular structure of one metabolite could not be elucidated. A major metabolic pathway of high-dose vitamin A in the non-human primate is apparently the oxidation of the primary alcohol group of retinol resulting in the formation of all-trans-retinoic acid. Subsequently, a broad spectrum of various metabolites of all-trans-retinoic acid, including beta-glucuronides and retinoids with a 13-cis configuration, appear in the plasma.

Teratologic potential of 2-methoxyethanol and transplacental distribution of its metabolite, 2-methoxyacetic acid, in non-human primates

The embryotoxic effects of 2-methoxyethanol (2-ME) were studied in non-human primates to better assess the risk for women of child-bearing age exposed to this agent. Macaca fascicularis females were treated daily throughout the organogenetic phase of pregnancy (days 20-45) by gavage and the fetuses collected at day 100 by Caesarean section. At the highest dose (0.47 mmole/kg), all eight pregnancies ended in death of the embryo. One of these dead embryos was abnormal, missing a digit on each forelimb. At the middle dose (0.32 mmole/kg), three of 10 pregnancies ended in embryonic death, presumably due to 2-ME exposure and three of 13 pregnancies met a similar fate at the low dose (0.16 mmole/kg). In each of these two groups, an additional pregnancy was lost to abortion, but both were thought to be spontaneous, which usually occurs in 10-20% of untreated macaque pregnancies. These results indicate that 2-ME is a potent toxin to the developing primate embryo and thereby furthers the concern about exposure of pregnant women to this agent, although maternal toxicity was evident in nearly all treated pregnancies and was especially severe in the high-dosage animals. Distribution of the major metabolite of 2-ME, 2-methoxyacetic acid (2-MAA), indicated a long half-life (ca. 20 h), resulting in accumulation of metabolite in maternal serum after repeated daily dosing. Transplacental studies revealed uniform distribution in the embryo and extraembryonic fluids at a concentration similar to that in maternal serum. The yolk sac, on the other hand, accumulated a very high concentration of 2-MAA, but the embryotoxic significance of this observation is unknown.

Arginine is a physiological precursor of endothelium-derived nitric oxide

ATP dose dependently stimulated the formation and release of nitric oxide (NO) from perfused rabbit aorta. L-Canavanine, an inhibitor of various L-arginine-utilizing enzymes, abolished basal and ATP-induced NO formation and release. ATP increased the accumulation of presumably NO-derived NO2- in the medium of primary cultures of bovine aortic endothelial cells. 15NO, 15NO2- and 15NO3- formation was found when L-[guanido-15N2]arginine was added to the culture medium. We conclude that the terminal guanidino nitrogens of L-arginine are the physiological precursors of endothelium-derived NO.

Ventral duplication of the autopod: chemical induction by methoxyacetic acid in rat embryos

Administration of 2-methoxyethanol to pregnant rats on day 12 of gestation induced ventral duplication of the autopod, presumably via its oxidative metabolite, methoxyacetic acid. Morphological observations indicate that the limb bud periderm is severely damaged by methoxyacetic acid so that large patches of this structure are actually missing during an extended period of limb bud development. A high concentration of methoxyacetic acid (10 mM) was found in the extraembryonic fluid and we postulate that the damage to the periderm was initiated from this extraembryonic exposure. The ventral duplication of the autopod is thought to arise through an attempt by the embryo to repair the periderm lesion.

Microscale ultrafiltration technique for determining free drug in 50-microL serum samples

This ultrafiltration technique allows determination of free drug in 50 microL of serum. We ultrafiltered sera containing the following drugs--valproic acid (and its major metabolites), phenobarbital, diazepam, indomethacin, phenytoin, furosemide, and chloramphenicol--using both the commercially available micropartition system (MPS-1, Amicon), which requires a 200-microL sample, and our modified micro system, which requires only 50 microL. The value for the free fraction of each drug obtained in the two experiments agreed well. The smaller sample requirement makes the micro method particularly suited for pediatric samples and studies on small laboratory animals.

Microscale ultrafiltration technique for determining free drug in 50-microL serum samples

This ultrafiltration technique allows determination of free drug in 50 microL of serum. We ultrafiltered sera containing the following drugs--valproic acid (and its major metabolites), phenobarbital, diazepam, indomethacin, phenytoin, furosemide, and chloramphenicol--using both the commercially available micropartition system (MPS-1, Amicon), which requires a 200-microL sample, and our modified micro system, which requires only 50 microL. The value for the free fraction of each drug obtained in the two experiments agreed well. The smaller sample requirement makes the micro method particularly suited for pediatric samples and studies on small laboratory animals.

Valproic acid and several metabolites: quantitative determination in serum, urine, breast milk and tissues by gas chromatography-mass spectrometry using selected ion monitoring

A method has been developed for the simultaneous quantitative determination of valproic acid (2-propylpentanoic acid) and its metabolites 2-propyl-2-pentenoic acid (trans), 2-propyl-3-pentenoic acid (trans), 2-propyl-4-pentenoic acid, 3-hydroxy-2-propylpentanoic acid, 4-hydroxy-2-propylpentanoic acid, 5-hydroxy-2-propylpentanoic acid, 3-oxo-2-propyl-pentanoic acid, and and 2-propylglutaric acid. All compounds were extracted at pH 5.0 with ethyl acetate. The concentrated extracts were trimethylsilylated and the resulting mixtures analyzed by a gas chromatography-mass spectrometry-computer system operated in the selected ion monitoring mode. Linear calibration curves were obtained in the concentration ranges studied (0.1-20 microgram/ml for metabolites, 0.1-150 microgram/ml for valproic acid. The recoveries of the drugs were between 92 and 97%. The relative standard deviations were between 3.9 and 8.1% (analysis of multiple 10-microliter samples of patient urine). The lower detection limits were found to be between 2.8 and 18 ng/ml using 200-microliter serum samples. The derivatized extracts were stable for at least one week. Applications of the method described include studies of placental transfer for valproic acid and metabolites in the human, the elimination of these substances by the neonate, their transfer via mother's milk, and their levels in mouse brain.

Microassay for primidone and its metabolites phenylethylmalondiamide, phenobarbital and p-hydroxyphenobarbital in human serum, saliva, breast milk and tissues by gas chromatography--mass spectrometry using selected ion monitoring

A method for the quantitative determination of primidone and its metabolites phenobarbital, phenylethylmalondiamide (PEMA) and hydroxyphenobarbital (free and conjugated) in serum, urine, saliva, breast milk and tissue has been developed. Following the addition of the methyl analogues of primidone, phenobarbital and PEMA as internal standards and of saturated ammonium sulphate, the samples (5--100 microliter) were extracted twice with ethyl acetate--benzene (20:80). The extracts were divided into two equal portions; one portion was ethylated by Greeley's method for the analysis of primidone, phenobarbital and hydroxy-phenobarbital, while the other was trimethylsilylated for the analysis of primidone and PEMA. A gas chromatographic--mass spectrometric system was used for the analysis of the derivatized extracts. Linear calibration curves were obtained in the concentration range studied (between 100 ng/ml and 30 microgram/ml). The recoveries of the drugs were between 80 and 93%. The relative standard deviations were between 3.2 and 5.9% (100-microliter serum samples containing 1 microgram/ml of the drugs). The lower detection limits were found to be between 1.4 and 3.7 ng/ml using serum samples of 100 microliter. These methods have been applied to the study of the placental transfer and neonatal disposition of primidone and its metabolites in the human.

Quantitative analysis of prazepam and its metabolites by electron capture gas chromatography and selected ion monitoring. Application to diaplacetal passage and fetal hepatic metabolism in early human pregnancy

Methods have been developed for the determination of the benzodiazepine tranquilizer prazepam and its metabolites desmethyl diazepam, 3-hydroxy-prazepam and oxazepam by electron capture gas chromatography and selected ion monitoring with diazepam as the internal standard. The benzodiazepines were isolated from blood serum or homogenized tissue samples, either by extraction with ethyl acetate or on small Extrelut columns packed with porous silica. The concentrated extracts were directly injected into the gas chromatograph equipped with an electron capture detector. Following trimethylsilylation, analysis on a gas chromatography--mass spectrometry--computer system operated in the selected ion-monitoring mode was performed. Using 50--200 mg (microliter) biological material, concentrations of prazepam and metabolites of 5 ng/g(ml) could be determined with signal-to-noise ratios of greater than 10. Using 1 g(ml) samples, the same signal-to-noise ratios were obtained with 1 ng/g(ml) concentrations. The methods developed were applied to the analysis of the diaplacental transfer of prazepam and desmethyl diazepam in early human pregnancy. Furthermore, prazepam metabolism in human fetal liver and cell cultures was studied.