Enzymic catalysis of the accumulation of acetaldehyde from ethanol in human prenatal cephalic tissues: evaluation of the relative contributions of CYP2E1, alcohol dehydrogenase, and catalase/peroxidases

BACKGROUND

The human prenatal brain is very sensitive to the toxic effects of ethanol, but very little information is available concerning the conversion of ethanol to the highly cytotoxic metabolite, acetaldehyde, in that organ. Thus, experiments were designed to investigate rates of accumulation of acetaldehyde from ethanol in the prenatal human brain.

METHODS

Prenatal human cephalic tissue homogenates were used as enzyme sources and were compared with analogous preparations of adult rat livers. Generated acetaldehyde was derivatized with cyclohexane-1,3-dione to yield fluorescent decahydroacrizine-1,8-dione, which was readily separated, detected, and quantitated with HPLC.

RESULTS

Detected rates of accumulation were unexpectedly high, even in the absence of added NADPH, NAD+, or H2O2, which are cofactors/cosubstrates for cytochrome P-450-, alcohol dehydrogenase- and catalase/peroxidase-catalyzed reactions, respectively. Without added cofactors/cosubstrates or other components and under linear reaction conditions, rates in human prenatal cephalic preparations were approximately 20% of those observed with analogous preparations of adult rat livers. Cofactor/cosubstrate-independent reactions were localized in the cytosolic (soluble) fraction and were strongly dependent on molecular oxygen (O2). They were not inhibited substantially by carbon monoxide (CO:O2 = 80:20 vs N2:O2 = 80:20) or by pyrazole in concentrations up to 10 mM and were only weakly inhibited by azide. Preincubations with excess catalase did not result in decreased activity. Reactions exhibited substrate saturation and heat inactivation indicating enzymic catalysis.

CONCLUSIONS

Experiments indicated a relatively rapid accumulation of acetaldehyde from ethanol in human prenatal brain tissues and suggested that the observed cofactor/cosubstrate-independent reactions were largely independent of P-450 cytochromes, alcohol dehydrogenases, or catalase/peroxidases. Results were consistent with catalysis by an as yet unidentified cytosolic oxidase(s).

Catalysis of the 4-hydroxylation of retinoic acids by cyp3a7 in human fetal hepatic tissues

Cytochrome P4503A7 (CYP3A7) is the primary CYP isoform expressed in human fetal hepatic microsomes, and its potential role in human embryotoxicity has attracted considerable investigative attention. In this study, we investigated the 4-hydroxylation of highly embryotoxic and teratogenic retinoic acids (RA) as catalyzed by human fetal liver microsomes (HFLM) and demonstrated that CYP3A7 is an efficient RA hydroxylase. When all-trans-retinoic acid (tRA), 9-cis-retinoic acid (9cRA), or 13-cis-retinoic acid (13cRA) were incubated with HFLM (54-109 gestational days) plus NADPH, each of these three retinoic acids was rapidly converted to its corresponding 4-hydroxy and 4-oxo metabolites. The reactions were strongly inhibited by CO (CO:O(2), 80:20) and were NADPH-dependent, indicating that the reactions were catalyzed by P450 isoenzymes. At 54 to 89 gestational days, 4-hydroxylase activities were relatively low. However, at gestational days 96 to 109, activities were much higher. Selective inhibitors were employed for elucidation of the roles of individual CYP isoenzymes in HFLM. alpha-Naphthoflavone, paclitaxel, and diethyldithiocarbamate showed little or no effects on HFLM-catalyzed reactions, indicating that CYP1A1, CYP1A2, CYP1B1, CYP2C8, and CYP2E1 did not play significant roles in the catalysis. By contrast, troleandomycin strongly inhibited the reaction (70-75% inhibition), suggesting that CYP3A7 was primarily responsible for the observed catalysis. It was also discovered that CYP3A7 SUPERSOMES efficiently catalyzed the 4-hydroxylations of tRA, 9cRA, and 13cRA. Because 4-hydroxylated metabolites of RA are much less potent embryotoxins and teratogens, the results indicated that the 4-hydroxylation of RA, catalyzed prenatally by CYP3A7, might play an important role in protecting the human fetus against RA-induced embryotoxicity.

Patterns of CYP26 expression in human prenatal cephalic and hepatic tissues indicate an important role during early brain development

CYP26 (P450RAI) catalyzes catabolic retinoic acid (RA) hydroxylation and thereby appears to play a critical role in retinoid signaling pathways during development. In this study, a quantitative competitive reverse transcriptase-polymerase chain reaction (RT-PCR) assay was developed for evaluation of CYP26 message levels in human prenatal tissues. Statistical analyses of transcription levels in 12 prenatal human brains and six prenatal human livers demonstrated good sensitivity and reproducibility. Quantitative profiles of CYP26 gene expression in early (gestational days 57-110) prenatal cephalic and hepatic tissues and comparisons with adult counterparts are reported for the first time. Prenatal cephalic tissues at days 57-67 exhibited values of 1950+/-420 (CYP26 molecules/10(6) GAPDH molecules) whereas prenatal cephalic tissues at days 105-110 exhibited values of 22300+/-4450 (CYP26 molecules/10(6) GAPDH molecules), indicating a sharp developmental increase (approximately 11-fold). Levels in human adult cephalic tissues were slightly less than the prenatal cephalic levels measured during the earliest stages of gestation and were approximately 3-fold lower than those measured in adult human hepatic tissues. Levels in human prenatal hepatic tissues at days 63-110 gestation were less than 800 (CYP26 molecules/10(6) GAPDH molecules) and did not exhibit developmental increases. Considered together, the data have strong implications for the importance of CYP26 in early development of the human brain.

Biosynthesis of all-trans-retinoic acid from all-trans-retinol: catalysis of all-trans-retinol oxidation by human P-450 cytochromes

Oxidative conversion of all-trans-retinol (t-ROH) to all-trans-retinal (t-RAL) is recognized as the rate-limiting step for biosynthesis of all-trans-retinoic acid from t-ROH in mammalian hepatic tissues. The purpose of this study was to investigate the role of human cytochrome P-450 (CYP)-dependent monooxygenation in the conversion of t-ROH to t-RAL. Adult human liver microsomes (HLMS) were incubated with t-ROH, and retinoids generated were identified and quantified by liquid chromatography-mass spectroscopy, HPLC, and other methods. HLMS-catalyzed generation of t-RAL from t-ROH was primarily NADPH-dependent and was strongly inhibited by carbon monoxide. Rates of reactions increased linearly with time and concentrations of HLMS, and exhibited classical substrate saturation. These observations strongly indicated that the reaction proceeded via CYP-catalyzed monooxygenation. On the basis of responses to selective chemical inhibitors, isoforms from CYP family 1 and the CYP3A subfamily appeared to be very active. Members of the CYP2C subfamily and CYP2D6 exhibited lesser activities and CYP2A6, CYP2B6, and CYP2E1 were virtually inactive. cDNA-expressed human CYP enzymes (CYP SUPERSOMES) also were used to assess the capacity of individual CYP enzymes to catalyze the reaction. Based on responses to selective chemical inhibitors, specific activities, and levels present in adult human hepatic tissues, CYP1A2 and CYP3A4 strongly appeared to be the major CYP enzymes catalyzing hepatic oxidative conversion of t-ROH to t-RAL in the adult human liver. CYP1A1 and CYP1B1 SUPERSOMES both exhibited exceptionally high activities, and in extrahepatic tissues, these isoforms could play important roles in biosynthesis of all-trans-retinoic acid from t-ROH.

Catalysis of drug oxidation during embryogenesis in human hepatic tissues using imipramine as a model substrate

We investigated the catalysis of drug monooxygenation by human embryonic hepatic tissues at a very early stage of gestation (days 52-59). Imipramine was used as a model substrate and the metabolites generated were identified and quantified by electrospray mass spectroscopy and HPLC. The primary metabolite generated was desipramine. It was reported previously from this and other laboratories that cytochrome P-450 monooxygenase (CYP) 1A1, 1B1, 2E1, and 3A7 are each expressed in human embryonic hepatic tissues, and selective inhibitors were therefore used to elucidate their respective roles. Furafylline did not inhibit the reaction, supporting that CYP1A2 was not expressed in human embryonic hepatic tissues. Diethyldithiocarbamate also failed to inhibit the same reaction, suggesting that CYP2E1 did not play a significant role in catalyzing the reaction. Triacetyloleandomycin inhibited the reaction by approximately 90%, suggesting that CYP3A7 was primarily responsible for catalyzing the reaction. However, alpha-naphthoflavone inhibited the same reaction by approximately 70%, suggesting that CYP1A1 and/or CYP1B1 may also catalyze the reaction substantially. To explore this issue more, a cDNA-expressed human CYP3A7 (CYP3A7 SUPERSOMES) was incubated with alpha-naphthoflavone (1 microM). Generation of desipramine was inhibited by approximately 40 to 50%. The addition of the CYP3A subfamily selective inhibitor triacetyloleandomycin (1 microM) produced no statistically significant inhibition in reactions catalyzed by CYP1A1 or 1B1 SUPERSOMES. Taken together, the results indicated that CYP3A7 was the major if not sole isoform responsible for catalysis of the N-demethylation of imipramine in human hepatic tissues during embryogenesis.

Catalytic activity and quantitation of cytochrome P-450 2E1 in prenatal human brain

Cytochrome P-450 2E1 (CYP2E1) is a readily inducible hemoprotein that catalyzes the oxidation of endogenous compounds and many low molecular weight xenobiotics. As the major component of the microsomal ethanol oxidizing system, it contributes significantly to ethanol metabolism and the formation of the highly reactive metabolite acetaldehyde. The leaky property of this enzyme results in the generation of reactive oxygen species that can induce oxidative stress and cytotoxic conditions deleterious to development. To further investigate the proposed role of CYP2E1 in the etiology of alcohol teratogenesis, the current study focused on the quantification of CYP2E1 in prenatal human brain, a tissue that is highly vulnerable to the damaging effects of ethanol throughout gestation. In microsomal samples prepared from pools of brain tissues, immunoreactive protein was detected by Western blot analysis using enhanced chemiluminescence, whereas functional protein was estimated with an enzymatic assay using p-nitrophenol and an electrochemical detection system. CYP2E1 transcript was consistently detected in RNA samples prepared from individual brain tissues using the ribonuclease protection assay. Quantitative data were collected by scanning densitometry and phosphorimaging technology. There was a dramatic increase in human brain CYP2E1 content around gestational day 50 and a fairly constant level was maintained throughout the early fetal period, until at least day 113. The relatively low levels of the P-450 isoform present in conceptal brain may be sufficient to generate reactive intermediates that elicit neuroembryotoxicity following maternal alcohol consumption.

Inhibition of human prenatal biosynthesis of all-trans-retinoic acid by ethanol, ethanol metabolites, and products of lipid peroxidation reactions: a possible role for CYP2E1

Biotransformation of all-trans-retinol (t-ROH) and all-trans-retinal (t-RAL) to all-trans-retinoic acid (t-RA) in human prenatal hepatic tissues (53-84 gestational days) was investigated with HPLC using human adult hepatic tissues as positive controls. Catalysis of the biotransformation of t-ROH by prenatal human cytosolic fractions resulted in accumulation of t-RAL with minimal t-RA. Oxidations of t-ROH catalyzed by prenatal cytosol were supported by both NAD+ and NADP+, although NAD+ was a much better cofactor. In contrast, catalysis of the oxidation of t-RAL to t-RA appeared to be solely NAD+ dependent. Substrate Km values for conversions of t-ROH to t-RAL and of t-RAL to t-RA were 82.4 and 65.8 microM, respectively. At concentrations of 10 and 90 mM, ethanol inhibited the conversion of t-ROH to t-RAL by 25 and 43%, respectively, but did not inhibit the conversion of t-RAL to t-RA significantly. In contrast, acetaldehyde reduced the conversion of t-RAL to t-RA by 25 and 87% at 0.1 and 10 mM respective concentrations. Several alcohols and aldehydes known to be generated from lipid peroxides also exhibited significant inhibition of t-RA biosynthesis in human prenatal hepatic tissues. Among the compounds tested, 4-hydroxy-2-nonenal (4-HNE) was highly effective in inhibiting the conversion of t-RAL to t-RA. A 20% inhibition was observed at a concentration of only 0.001 mM, and nearly complete inhibition was produced at 0.1 mM. Human fetal and embryonic hepatic tissues each exhibited significant CYP2E1 expression as assessed with chlorzoxazone 6-hydroxylation, a highly sensitive western blotting technique, and reverse transcriptase-polymerase chain reaction (PCR) (RT-PCR), suggesting that lipid peroxidation can be initiated via CYP2E1-catalyzed ethanol oxidation in human embryonic hepatic tissues. In summary, these studies suggest that ethanol may affect the biosynthesis of t-RA in human prenatal hepatic tissues directly and indirectly. Ethanol and its major oxidative metabolite, acetaldehyde, both inhibit the generation of t-RA. Concurrently, the CYP2E1-catalyzed oxidation of ethanol can initiate lipid peroxidation via generation of a variety of free radicals. The lipid peroxides thereby generated could then be further converted via CYP2E1-catalyzed reactions to alcohols and aldehydes, including 4-HNE, that act as potent inhibitors of t-RA synthesis.

Expression of cytochrome P450RAI (CYP26) in human fetal hepatic and cephalic tissues

PCR amplifications with two sets of degenerate primers that were targeted to CYP26-specific regions were performed with cDNAs from human fetal liver and brain as templates. PCR products were purified, cloned, sequenced and analyzed with the BLAST program. Our results revealed expression of CYP26 in both human fetal liver and brain. Furthermore, human fetal CYP26 cDNA exhibited 99.2%-100% nucleotide sequence identity to its adult counterpart. Novel isoforms, that would have indicated additional CYP26 genes, were not found. A Northern blot containing poly(A+)RNAs from 43 human adult and 7 human fetal tissues was tested for CYP26 expression. We were able to detect CYP26 message in most tissues but hybridization signals varied in intensity. Highest levels of transcription were in adult liver, heart, pituitary gland, adrenal gland, placenta and regions of the brain. CYP26 expression in fetal tissues was strongest in the brain and comparable with message levels in adult tissues.

Recombinant human glutathione S-transferases catalyse enzymic isomerization of 13-cis-retinoic acid to all-trans-retinoic acid in vitro

The steric conversion of 13-cis-retinoic acid (13-cRA) to all-trans-retinoic acid (t-RA) has been proposed as an activation mechanism for the observed therapeutic and teratogenic activities of 13-cRA. Here we have investigated the catalysis of isomerization of 13-cRA to t-RA by recombinant human glutathione S-transferases (GSTs). Substrate was incubated with GST in 0.1 M sodium phosphate buffer, pH 7.5, at 37 degrees C in total darkness. The t-RA generated was measured quantitatively by HPLC. Under the reaction conditions used, GSTP1-1 was far more effective than human GSTM1-1 or human GSTA1-1 in catalysing the isomerization reaction. The reaction catalysed by GSTP1-1 showed substrate saturation and the Km and Vmax values for the reaction were approx. 7 microM and 650 pmol/min per nmol respectively. The reaction rate increased linearly with increasing enzyme concentration. The reaction was inhibited both by heat treatment and by S-decylglutathione (a potent inhibitor of transferase activity associated with GST). Additions of polyclonal rabbit antiserum for human GSTP1-1 to the reaction resulted in a significant decrease in generation of t-RA (70-80%). In addition, ethacrynic acid, a selective substrate for Pi isoforms of GST, also inhibited the isomerization of 13-cRA to t-RA catalysed by GSTP1-1. Under the same reaction conditions, GSTP1-1 was much less effective in catalysing the steric conversion of 9-cis-retinoic acid to t-RA, indicating that the enzyme was stereospecific for the conversion of 13-cRA to t-RA. These observations suggest that enzymic catalysis was the primary mechanism for the GSTP1-1-dependent conversion of 13-cRA to t-RA. Reactions catalysed by a purified rat hepatic GST Pi isoenzyme proceeded more slowly than reactions catalysed by human GSTP1-1. Comparative studies also showed that there were marked species differences in catalytic activities between various purified mammalian hepatic GST mixtures.

Cytochrome-P450-dependent biotransformation of xenobiotics in human and rodent embryonic tissues

Profound species differences and developmental stage differences as well as a lack of solid data prevent broad, sweeping generalizations in terms of statements that can be made concerning the prenatal expression of individual P450 isoforms. It is clear, however, that several of such isoforms are expressed at levels that can be toxicologically significant. At present, the greatest interest appears to be in P450s 1A1, 1B1, 2E1, and 3A7, each of which has been reported to be expressed at toxicologically significant levels or at least at potentially toxicologically significant levels during organogenesis. Reports of the expression of other P450 isoforms at later stages of gestation also have appeared in the recent literature.

Biotransformation of 13-cis- and 9-cis-retinoic acid to all-trans-retinoic acid in rat conceptal homogenates. Evidence for catalysis by a conceptal isomerase

The purpose of this study was to investigate whether and to what extent the steric isomerization of retinoic acids in conceptal tissues can be attributed to enzymatic catalysis in addition to thiol-dependent, nonenzymatic catalysis. Conversions of 13-cis-retinoic acid and 9-cis-retinoic acid to all-trans-retinoic acid catalyzed by cell-free preparations of conceptal rat tissues (gestational day 12.5) were investigated. Substrates and rat conceptal homogenates (RCH) were incubated in sodium phosphate buffer (0.1 M, pH 7.5) at 37 degrees C in the dark. Incubation mixtures were quantitatively analyzed by HPLC. In RCH-catalyzed reactions, conversions of 13-cis-retinoic acid or 9-cis-retinoic acid to all-trans-retinoic acid were very rapid, in comparison with uncatalyzed isomerization reactions (incubations without RCH). Comparisons of the rates of reactions catalyzed by freshly prepared vs. freshly prepared/dialyzed RCH showed no significant differences, indicating that small, suflhydryl-containing molecules such as reduced glutathione did not significantly contribute to the RCH-catalyzed reactions. Furthermore, at physiological concentrations (2.5-10 mM), reduced glutathione exhibited very low specific catalytic activities, indicating that nonenzymatic, sulfhydryl-dependent catalysis was not a major mechanism in catalyzing interconversions of retinoic acids in vivo. Enzymatic catalysis by RCH of the conversion of 13-cis-retinoic acid to all-trans-retinoic acid was further characterized by showing 1) substrate saturation kinetics, 2) reaction rates that increased proportionally with protein concentrations, and (3) much greater sensitivity of the reactions to heat inactivation and denaturation by urea, compared with nonenzymatic, glutathione-catalyzed reactions. Thus, isomerization of retinoids in conceptal tissues appeared to be under enzymatic control.

Inhibition of embryonic retinoic acid synthesis by aldehydes of lipid peroxidation and prevention of inhibition by reduced glutathione and glutathione S-transferases

Inhibition of conceptal biosynthesis of all-trans-retinoic acid (t-RA) by aldehydes generated from lipid peroxidation was investigated. Oxidative conversion of all-trans-retinal (t-RAL, 18 microM) to t-RA catalyzed by rat conceptal cytosol (RCC) was sensitive to inhibition by trans-2-nonenal (tNE), nonyl aldehyde (NA), 4-hydroxy-2-nonenal (4HNE), and hexanal. With an initial molar ratio of aldehyde/t-RAL of 2:1, tNE, NA, and 4HNE caused 70, 65, and 40% reductions of t-RA synthesis, respectively. Hexanal reduced generation of t-RA by approximately 50% as the ratio of aldehyde/t-RAL was raised to 20:1. tNE significantly increased the Km of the reaction and kinetic analyses indicated a mixed competitive/noncompetitive inhibition. By contrast, analogous reactions catalyzed by adult rat hepatic cytosol (ARHC) were highly resistant to inhibition by the same aldehydes. Significant inhibition (> 40% reduction of t-RA generation) by 4HNE, NA, and tNE were achieved at high molar ratios of aldehyde/t-RAL (> 175:1). Hexanal did not inhibit the reaction significantly even at very high ratios of aldehyde/t-RAL (> 2,000:1). Interestingly, when reduced glutathione (GSH, 10 mM) alone or GSH plus glutathione S-transferase (GST) were added to RCC-catalyzed reactions, additions of tNE or 4HNE showed either no significant inhibition or a partial lack of inhibition. Results suggested that GSH-dependent conjugation with 4HNE proceeded slowly compared to conjugation with tNE. To test the hypothesis that GST-catalyzed GSH conjugation can effectively prevent inhibition of t-RA synthesis by aldehydic products of lipid peroxidation, triethyltin bromide (TEB, a potent inhibitor of GST, 20 microM) was added to ARHC-catalyzed reactions when hexanal or tNE were present in the incubations. Eighty and 60% of hexanal and tNE inhibition, respectively, were observed. This was apparently due to TEB blockage of GST-catalyzed GSH conjugation reactions and thus strongly supported the stated hypothesis.

Chemical teratogenesis in humans: biochemical and molecular mechanisms

In this review, an attempt has been made to summarize our current understanding of the mechanisms whereby certain chemicals cause birth defects. The chemicals selected for consideration were those that have been designated as established or recognized human teratogens. It is clear that our current understanding of mechanisms whereby these agents cause teratogenic effects (birth defects) can vary dramatically from one agent to the next. Extremes include the folic acid antagonists, which are now well established as agents that produce birth defects by virtue of potent inhibition of dihydrofolate reductase as a primary biochemical mechanism. An example at the other extreme is ethanol, for which very few definitive statements can be made with regard to teratogenic mechanisms, and the probability exists that a large number of interacting, contributory mechanisms can be invoked. For nearly all chemical teratogens, the critical links in the chains of events between the initial, primary biochemical and molecular mechanistic event (e.g. dihydrofolate reductase inhibition) and the manifestations of specific abnormalities (pathogenic mechanisms) remain to be delineated. This will provide an enormous challenge for investigators for years to come.

Glutathione S-transferases act as isomerases in isomerization of 13-cis-retinoic acid to all-trans-retinoic acid in vitro

A discovery that rapid enzymic isomerization of 13-cis-retinoic acid (13-cRA) to all-trans-retinoic acid (t-RA) can be catalysed by purified hepatic glutathione S-transferases (GSTs; EC 2.5.1.18) from rat is now reported. Rates of cis-trans isomerization were determined quantitatively by HPLC. GST-catalysed reactions reached equilibrium rapidly, in marked contrast with uncatalysed or GSH-catalysed isomerizations. The GST-catalysed reaction exhibited substrate saturation kinetics with a Km of approx. 8 microM. The maximal velocity of the reaction and the catalytic efficiency of GSTs were determined. The initial rate of the reaction increased linearly as a function of enzyme concentration. Catalysis by GSTs was independent of the presence of GSH, indicating that GSTs act as GSH-independent isomerases as well as transferases. Incubation with guanidine (7-8 M) or heat-inactivation of GSTs (100 degrees C for 3 min) decreased isomerase activities by approx. 50% and 75% respectively. The same heat treatment did not significantly inhibit isomerization catalysed by GSH and apoferritin, indicating that the observed decrease in isomerase activity by heat inactivation was not primarily due to oxidation of protein thiol groups in the GSTs. The specific activity of GSTs was approx. 23- and 340-fold those of GSH and apoferritin respectively when comparisons were made on the basis of free thiol concentrations, indicating that free thiol in GSTs cannot account for the majority of observed isomerase activities and suggesting that specific conformations of GSTs are important for such activities. Complete inhibition of the reaction by low concentrations of N-ethylmaleimide (10 microM) demonstrated that intact protein thiols are required for the isomerase activities of GSTs.

Expression of CYP2E1 during embryogenesis and fetogenesis in human cephalic tissues: implications for the fetal alcohol syndrome

Reverse transcription and the polymerase chain reaction (RT-PCR) with oligonucleotide primers designed to target cDNA nucleotides 1241-1357 corresponding to exons 8 (3' end) and 9 (5' end) in human genomic CYP2E1 detected consistently strong signals in 9 of 10 prenatal human brains. Cephalic tissues analyzed were between 54 and 78 days of gestation. RT-PCR signals for expression of CYP2E1 in corresponding human hepatic or adrenal tissues were weaker or, with only 2 exceptions, undetectable. Attempts to approximate levels of P4502E1 mRNA with Northern blots and RNase protection assays indicated that levels in human prenatal whole brain tissues tended to increase as a function of gestational age but, at the early stages investigated, were far lower than the constitutive levels in hepatic tissues of adult humans or male rats. Localized, P4502E1-dependent cephalic bioactivation of ethanol, with associated generation of several reactive chemical species, could contribute significantly to the etiology of neuroembryotoxic effects of prenatal ethanol exposure.

Biotransformation of all-trans-retinal, 13-cis-retinal, and 9-cis-retinal catalyzed by conceptal cytosol and microsomes

Oxidative conversions of all-trans-retinal (t-RAL), 13-cis-retinal (13-cRAL), and 9-cis-retinal (9-cRAL) to their corresponding retinoic acids (RAs) catalyzed by rat conceptal cytosol (RCC) or microsomes (RCM) were studied. The primary product of RCC-catalyzed oxidations of both t-RAL and 13-cRAL was t-RA, with only trace amounts of 13-cRA and 9-cRA. In the RCC-catalyzed oxidation of 9-cRAL, generated t-RA, 9-cRA, and 13-cRA constituted approximately 56, 34, and 10%, respectively, of the total RAs. For all RCC-catalyzed retinal oxidations, NAD was a much more effective cofactor than NADP. And t-RAL and 13-cRAL were much better substrates than 9-cRAL. Formaldehyde, acetaldehyde, citral, and disulfiram were investigated as inhibitors, but only citral and disulfiram effectively inhibited the RCC-catalyzed conversion of t-RAL or 13-cRAL to t-RA. Methanol and ethanol failed to inhibit either reaction even at very high concentrations (> or = 10 mM). RCM exhibited lower specific enzymatic activities than RCC in catalyzing oxidations of t-RAL, 13-cRAL, and 9-cRAL, indicating that the cytosolic fraction was dominant for converting retinals to RAs. The predominant RA produced from RCM-catalyzed oxidations of t-RAL, 13-cRAL, or 9-cRAL was t-RA for each substrate, and again NAD was a much more effective cofactor than NADP in all cases. For RCM-catalyzed oxidations of RALs, 13-cRAL was a much better substrate than t-RAL or 9-cRAL. Methanol and ethanol were not effective inhibitors for RCM-catalyzed oxidations of t-RAL or 13-cRAL. In RCM-catalyzed reactions, citral (10 mM) completely inhibited oxidation of t-RAL but showed only a minor effect on oxidation of 13-cRAL. 13-cRA was converted almost completely to t-RA after 2 hr of incubation with RCC.

Catalysis of the dealkylation/debenzylation of phenoxazone ethers by hemoglobin in the absence of peroxides: implications for investigations of embryonic biotransformation

Investigations of catalysis of the O-dealkylation and O-debenzylation of phenoxazone (resorufin) ethers in human and rodent embryonic tissue homogenates indicated that, with few exceptions, each conceptal tissue investigated contained enzymes capable of catalyzing each of the reactions under study. All observable reactions exhibited NADPH dependence and strong inhibition by carbon monoxide, ketoconazole, alternate electron acceptors, and by hypoxic incubation conditions; but, they were not strongly inhibited by several other classical cytochrome P450 (P450) inhibitors. Cyanide, azide, superoxide dismutase/catalase, and glutathione/glutathione peroxidase each also failed to inhibit the reactions significantly. Subcellular fractionation experiments revealed that cytosolic fractions contained a preponderance of the observable monooxygenase activities. Attempts to identify components responsible for the cytosolic catalytic activity indicated that cytosolic nitric oxide synthases did not contribute significantly. Column fractionation of the cytosol indicated that significant catalytic activity coeluted with fractions containing hemoglobin (Hgb), and experiments with purified Hgb as enzyme source showed that Hgb would catalyze all reactions under study at very slow rates in the absence of added reductases or peroxides. Additions of either reductases or peroxides, however, resulted in marked increases in rates of Hgb-catalyzed reactions. Further investigations strongly suggested that virtually all dealkylation or debenzylation of phenoxazone ethers catalyzed by embryonic cytosolic fractions could be accounted for by the presence of Hgb in those fractions. Conceptal microsomal fractions, however, exhibited definitive, P450-dependent monooxygenase activities attributable to specific individual, identifiable P450 isoforms.

Drug metabolic enzymes in developmental toxicology

Although much is known about the metabolism of environmental toxicants in adult organisms, little information exists on the role of cytochrome P450 (CYP) enzymes during development. The developing organism is remarkably dynamic, presenting a constantly changing metabolic profile as various enzyme systems are activated or repressed. This may explain the markedly different sensitivities to various toxicants that are exhibited throughout the developmental period. The application of molecular biological methods has provided important information on the roles of these enzymes in modulating the response of the developing organism to toxicological exposures. The first talk will focus on the identification and role of CYPs during early organogenesis, particularly on how these enzymes influence the response of the conceptus and early embryo to toxic chemicals. The second presentation will discuss the identification of CYPs expressed during human development, as many of the enzymes present in adults are not expressed in the fetus. The third speaker will discuss the developmental consequences of loss of expression of particular metabolic enzymes, focusing on recent studies employing knockout mice to examine the role of drug metabolic enzymes during development. The last two talks will discuss some of the short- and long-term consequences of in utero exposures to toxic chemicals and the role of CYP in modulating the toxic response of the developing organism. The first of these will focus on the role of CYP2E1 in human fetuses during late gestation and the response of this enzyme to inducing agents such as alcohol. The last talk will discuss the role of CYP1A1 in the activation of the Ki-ras oncogene following in utero exposure to carcinogens as a mechanism for lung tumor formation in a pharmacogenetic mouse model.

Effects of ethanol on biotransformation of all-trans-retinol and all-trans-retinal to all-trans-retinoic acid in rat conceptal cytosol

Enzymatic catalysis of the oxidations of ethanol, all-trans-retinol (tretinol) and all-trans-retinal (t-retinal) were demonstrated in the cytosolic fractions of rat conceptal homogenates at day 12 of gestation. Products of the retinoid oxidation reactions were identified with HPLC by comparing elution times with those of authentic standard retinoids. NAD-dependent oxidations of each of the three substrates were demonstrable with assay conditions used; t-retinol and t-retinal each were converted to readily detectable quantities of all-trans-retinoic acid (t-RA). At 1.0 mM or higher concentrations, ethanol effectively inhibited the synthesis of t-RA from both t-retinol and t-retinal when adult hepatic cytosol was used as enzyme source. Approximately 70% and 40% inhibitions, respectively, were observed at 10 mM ethanol concentrations. By contrast, for the reactions catalyzed by rat conceptal cytosol (RCC) under the same experimental conditions, ethanol falled to inhibit significantly the conversion of either t-retinol or t-retinal to t-RA at concentrations up to 1,000 mM. For the RCC-catalyzed conversion of t-retinal to t-RA, increasing concentrations of ethanol (0 to 1.0 M) resulted in linear increases rather than decreases in quantities of t-RA generated. At a 2.0 M concentration of ethanol, the quantity of t-RA increased by > 50%. Significant inhibition of t-RA generation from t-retinal occurred only at extremely high (> 4.0 M) concentrations. The results indicated that ethanol was a very ineffective inhibitor of RCC-catalyzed synthesis of t-RA from either t-retinol or t-retinal. This contrasted strongly with effective inhibitory effects with adult hepatic cytosol as enzyme source. The results supported the concept that competitive inhibition of conversion of t-retinol to t-RA in conceptal tissues is not a significant factor in ethanol-elicited embryotoxicity and dysmorphogenesis, at least in rodents. Mechanisms for the ethanol-induced increases in conversion of t-retinal to t-RA remain to be elucidated.

Interactive dysmorphogenic effects of all-trans-retinol and ethanol on cultured whole rat embryos during organogenesis

Whole rat conceptuses (10.5 gestational days) were explanted into a culture medium containing all-trans-retinol (t-retinol, vitamin A1), ethanol, or combinations of the two alcohols at various concentrations, and were cultured at 37 degrees C for 24 hr. Parameters emphasized in morphological analyses were branchial arch development, closure of neural tube, axial rotation, and development of otic vesicles and of optic cup. Additions of t-retinol alone to the culture medium resulted in significant decreases in viability at concentrations of 7.0 microM and above. A primary target site affected by t-retinol was the second branchial arch. With initial culture medium concentrations of 3.5 microM, 28% of embryos exhibited an underdeveloped second branchial arch, and the effect was concentration dependent. Incubations with t-retinol alone also caused failure of closure of neural tubes, underdevelopment/absence of otic and optic vesicles, and failure of normal axial rotation, but these effects were statistically significant only at the higher concentrations (10.5-14.0 microM). Incubations of conceptuses with ethanol alone resulted in statistically significant decreases in viability and increases of incidence of embryonic abnormalities at 50 mM but not at 10- or 20-mM concentrations. The embryotoxicity of ethanol appeared less site-specific than that of t-retinol. However, ethanol-elicited developmental abnormalities included underdevelopment of the first and second branchial arches, abnormally open neural tubes, abnormally small or absent otic and optic vesicles, and incomplete axial rotation in common with effects elicited by t-retinol. In general, embryos incubated with combinations of t-retinol and ethanol showed lower survival rates and higher incidences of developmental abnormalities when compared to the calculated values expected for simple additive effects; i.e., interactive effects were most frequently greater than additive and probably synergistic but not antagonistic. To assist in the elucidation of possible mechanism(s) for the greater than additive/synergistic dysmorphogenic effects observed, concentrations of all-trans-retinoic acid (t-RA) and all-trans-retinal(t-retinal) in cultured conceptal tissues were determined by high-performance liquid chromatography (HPLC). HPLC analysis showed increases in conceptal tissue levels of both t-RA and t-retinal after conceptuses were exposed to t-retinol (10.5 microM) plus various quantities of ethanol for 24 hr. These observations, in combination with those of previous studies, suggested that the observed greater-than-additive/synergistic dysmorphogenic effects were not due to the inhibition by ethanol of conceptal biosynthesis of t-RA. Whether the increased levels of t-RA and t-retinal caused the observed greater than additive/synergistic dysmorphogenic effects remains to be elucidated.

Apoptosis induced in cultured rat embryos by intra-amniotically microinjected sodium nitroprusside

Previously, we reported that massive cell death was induced in the mesencephalic area of cultured rat embryos after embryos of gestational day 10.5 were intra-amniotically microinjected with sodium nitroprusside (SNP, 800 microM) and cultured for 24 hr at 37 degrees C. The massive cell death apparently was the result of NO-mediated embryotoxicity. Damage was concentration dependent and tissue specific. In follow-up studies, we now report evidence that NO generated from SNP induces apoptosis in organogenesis stage cultured rat embryos. Nile blue sulfate (NBS) staining suggested that microinjections of 400 microM SNP induced apoptosis in the mesencephalic area. Since we observed no massive cell death ("white caps") at this concentration, it appeared that early stages of apoptosis preceded "white cap" formation. At 800 microM SNP, total disintegration of cell bodies was evident and may have resulted from later stages of aoptosis or necrosis, or both. The "white caps" per se, an accumulation of disintegrated cell bodies, did not stain with NBS, probably due to total loss of cell integrity and resultant coagulation. The majority of the coagulated dead cells in the "white caps" were heavily stained with 3,3'-diaminobenzidine via in situ 3' end-labeling with terminal transferase. However, it is now known that NO can damage DNA directly and that in situ 3' end-labeling by terminal transferase detects not only apoptosis but also random DNA breakage. Increased 3' end-labeling and a "DNA ladder" were detectable within 5-10 hr after exposure of day 10.5 embryos to 400 or 800 microM of microinjected SNP. Some smear background was also observed in the "ladder." Rostral aspects of embryos exhibited more prominent indices of apoptosis than caudal regions. The results suggested that microinjections of SNP into the amniotic fluid of day 10.5 cultured rat embryos induces NO-mediated cell death in the mesencephalic and rhombencephalic regions by the process of apoptosis or of both apoptosis and necrosis, depending on the timing, concentration, and stage of gestation.

Biotransformation of all-trans-retinol and all-trans-retinal to all-trans-retinoic acid in rat conceptal homogenates

Catalysis of the oxidation of all-trans-retinol (vitamin A1) or of all-trans-retinal to all-trans-retinoic acid (all-trans-RA) by rat conceptal enzymes was investigated during organogenesis. Products of the reaction were identified and quantified with HPLC by comparing their elution times with those of authentic standard retinoids. Under the incubation and assay conditions utilized, all-trans-retinol and all-trans-retinal were converted to readily detectable quantities of all-trans-RA. Rat conceptal homogenates from gestational days 10.5, 11.5 and 12.5 each exhibited enzymatic activity for oxidation of all-trans-retinol and all-trans-retinal to all-trans-RA. Enzymatic catalysis was verified by showing that: (1) both reactions were coenzyme dependent; (2) the rates of reactions increased as concentrations of conceptal protein increased; (3) both reactions were abolished by heating the tissue homogenates (100 degrees, 5 min); and (4) both reactions exhibited substrate saturation. Under the same experimental conditions, formation of all-trans-RA from all-trans-retinol was much slower than from all-trans-retinal, suggesting that oxidation of all-trans-retinol to all-trans-retinal was the rate-limiting step for biotransformation of all-trans-retinol to all-trans-RA in embryonic tissues. When NAD or NADP were replaced by NADH or NADPH, the rate of oxidation of all-trans-retinol was reduced markedly, indicating that the reaction was catalyzed primarily by an NAD/NADP-dependent dehydrogenase(s). Carbon monoxide (CO:O2 = 90:10) did not inhibit the reaction. NAD appeared to be a more effective cofactor than NADP in catalyzing oxidation of all-trans-retinal to all-trans-RA. When NAD was omitted, formation of all-trans-RA from all-trans-retinal was reduced by approximately 55%. Replacing NAD by NADH or NADPH also reduced the conversion of all-trans-retinal to all-trans-RA by about 60%. These observations suggested at least two pathways for the generation of all-trans-RA from all-trans-retinal in embryos: oxidation catalyzed by an NAD/NADP-dependent dehydrogenase(s) and oxidation catalyzed by an oxidase(s) that did not require NAD, NADH, NADP or NADPH. Conversion of all-trans-retinol to all-trans-RA was inhibited strongly by low concentrations of citral, but not by high concentrations of sodium azide, 4-methylpyrazole, or metyrapone. Similarly, oxidation of all-trans-retinal was inhibited strongly by citral but not by metyrapone.(ABSTRACT TRUNCATED AT 400 WORDS)

Xenopus laevis: a model system for the study of embryonic retinoid metabolism. III. Isomerization and metabolism of all-trans-retinoic acid and 9-cis-retinoic acid and their dysmorphogenic effects in embryos during neurulation

These investigations provide data pertaining to the metabolism and disposition of exogenous 9-cis-retinoic acid and all-trans-retinoic acid during neurulation in Xenopus embryos. Each isomer elicited malformations of the heart, eye, and brain, but approximately 2-fold higher concentrations of all-trans-retinoic acid than 9-cis-retinoic acid were required to produce qualitatively and quantitatively similar dysmorphogenic effects. The dymorphogenic effects of all-trans-retinoic acid could not be attributed to the isomerization of all-trans-retinoic acid to 9-cis-retinoic acid. Evidence is provided that all-trans-retinoic acid and 9-cis-retinoic acid are both direct-acting dysmorphogens. After Xenopus embryos were exposed to all-trans-retinoic acid, elevated levels of 4-oxo-all-trans-retinoic acid, 4-oxo-13-cis-retinoic acid, all-trans-retinoyl-beta-glucuronide, and 13-cis-retinoic acid were detected in the embryos, whereas embryonic levels of 9-cis-retinoic acid were actually slightly lower than endogenous levels during early neurulation. After embryos were exposed to 9-cis-retinoic acid during neurulation, elevated levels of 4-oxo metabolites, glucuronides and 9,13-di-cis-retinoic acid were observed in the embryos. At equivalent concentrations, 4-oxo-13-cis-retinoic acid and 13-cis-retinoic acid elicited fewer severe multiple malformations than all-trans isomers 9,13-di-cis isomers, or 9-cis isomers. The dysmorphogenic effect of 9,13-di-cis-retinoic acid may be caused by its isomerization to 9-cis-retinoic acid. All-trans retinoyl-beta-glucuronide was only marginally teratogenic at the highest concentrations tested.

Expression of functional cytochrome P4501A1 in human embryonic hepatic tissues during organogenesis

Investigations with chemical inhibitors and with inhibitory antibodies specific for cytochrome P4501A-catalyzed ethoxyresorufin (ethoxyphenoxazone) O-deethylation and 2-acetylaminofluorene (N-2-fluorenylacetamide) ring hydroxylation indicated that cytochrome(s) P450 of the 1A subfamily was functionally expressed in human embryonic hepatic tissues at very early stages (days 50-60) of gestation. Lack of detectable capacity of hepatic microsomal enzymes to catalyze either N-hydroxylation of 2-acetylaminofluorene or O-demethylation of methoxyresorufin indicated that functional cytochrome P4501A2 is expressed minimally or negligibly in human embryonic hepatic tissues. By contrast, profound inhibition of the ring hydroxylation of 2-acetylaminofluorene and of the O-deethylation of ethoxyresorufin by 7,8-benzoflavone as well as by anti-cytochrome P4501A1 antibodies indicated the presence of significant levels of functional cytochrome P4501A1 in hepatic microsomes of human embryos. Using the reverse transcriptase-linked polymerase chain reaction with specific oligonucleotide primers, we also detected significant expression of cytochrome P4501A1 mRNA in human embryonic livers. Polymerase chain reaction amplification, cloning and sequencing of the corresponding cDNA provided evidence that the cytochrome P4501A1 mRNA expressed in human embryonic tissues was identical to that expressed in adult human tissues. The results of the study have important implications in terms of the embryotoxic effects of chemicals that are known to be substrates, inhibitors or inducers of cytochrome P4501A1 and to which pregnant women are exposed.

Xenopus laevis: a model system for the study of embryonic retinoid metabolism. II. Embryonic metabolism of all-trans-3,4-didehydroretinol to all-trans-3,4-didehydroretinoic acid

This study demonstrates early embryonic metabolism of exogenous all-trans-3,4-didehydroretinol (vitamin A2) to all-trans-3,4-didehydroretinal and to all-trans-3,4-didehydroretinoic acid in Xenopus embryos during neurulation. The latter metabolite was recently shown to bind with high affinity and to activate various retinoic acid receptors. Embryos treated with all-trans-3,4-didehydroretinol during early or late gastrulation exhibited abnormalities along the anteroposterior axis. The abnormalities were primarily in the posterior regions of the embryo, with only minor defects anteriorally. Eye malformations, typical for early exposure to 9-cis- and all-trans-retinols and retinals (companion paper), were not observed. We also present evidence that all-trans-3,4-didehydroretinoic acid is present endogenously during early neurulation and is evenly distributed along the anteroposterior axis. After treatment with all-trans-3,4-didehydroretinol, embryonic levels of all-trans-3,4-didehydroretinoic acid exceeded endogenous levels of this metabolite during early and late neurulation. We hypothesize that the dysmorphogenic effects produced by treatment of Xenopus embryos with the alcohol precursor, all-trans-3,4-didehydroretinol, are the result of its embryonic conversion to its corresponding acid ligand.

Xenopus laevis: a model system for the study of embryonic retinoid metabolism. I. Embryonic metabolism of 9-cis- and all-trans-retinals and retinols to their corresponding acid forms

Recently, the temporal and spatial distribution patterns of two established, endogenous retinoid receptor ligands, 9-cis-retinoic acid and all-trans-retinoic acid and various precursor retinoids were described in Xenopus embryos during early development (Creech Kraft et al., Proc. Natl. Acad. Sci. U.S.A. 1994; Biochem. J. 1994). Each of these two receptor ligands is a metabolite of vitamin A (all-trans-retinol), and each is also a potent dysmorphogen in Xenopus embryos as well as in embryos of several other vertebrate species. This study demonstrates early embryonic metabolism of exogenous all-trans-retinol, 9-cis-retinol, all-trans-retinal, and 9-cis-retinal to 9-cis-retinoic acid, all-trans-retinoic acid, and other metabolites in Xenopus embryos during neurulation, a specific stage of development that spans a time period of approximately 8 hr. Our results demonstrate that the Xenopus embryo provides a suitable model system for studying the embryonic bioconversion of retinoids and dysmorphogenic effects within a single time window of development.

Functional cytochrome P4503A isoforms in human embryonic tissues: expression during organogenesis

Expression of functional cytochrome P450 (CYP) isoforms in human embryonic tissues was explored during organogenesis (days 50-60 of gestation) with substrate probes, inhibitor probes, and immunoprobes and by reverse transcription-polymerase chain reaction (PCR), cloning, and sequencing. Evidence was obtained for the presence of relatively high levels of one or more functional CYP3A isoforms in embryonic livers. This was manifested as relatively extensive hydroxylation of (R)-warfarin at carbon 10 and as triacetyloleandomycin-inhibited O-debenzylation of benzyloxyresorufin when human embryonic hepatic microsomal fractions were used as enzyme sources. Immunoblots with anti-CYP3A4 antibody exhibited a strong signal in embryonic hepatic tissues but, in contrast, indicated very low or negligible CYP3A levels in human embryonic lung, kidney, heart, adrenal, and brain tissues. To explore expression of individual members of the CYP3A subfamily in human embryonic hepatic tissues at this early gestational stage, CYP3A cDNA was generated by reverse transcription, amplified by PCR, cloned, and sequenced. Oligonucleotide primers used for PCR were designed to flank target sequences unique to CYP3A but also common to all human CYP3A subfamily members for which GenBank nucleotide sequence information was available (CYP3A3, CYP3A4, CYP3A5, CYP3A5P, and CYP3A7). Sequencing data indicated that plasmids in 58 of 59 recombinant positive colonies contained an insert with a sequence identical to that present in CYP3A7 cDNA and the plasmid of only one colony contained an insert with a sequence identical to that present in CYP3A5 cDNA. No evidence was found for expression of CYP3A3 or CYP3A4. Thus, during organogenesis, human embryonic hepatic tissues express primarily CYP3A7 and are capable of significant CYP3A7-catalyzed xenobiotic monooxygenation during this very early stage of gestation.

Induction in vitro and complete coding region sequence of cytochrome P4501A1 cDNA from cultured whole rat conceptuses during early organogenesis

Exposures of cultured whole rat conceptuses during organogenesis to 3-methylcholanthrene (MC; 0.025-25 microM), 5,6-benzoflavone (BNF; 5-100 microM) or benz[a]anthracene (BA; 5-100 microM) were effected by placement of each of these "MC-type" inducing agents in the culture medium at the time of explantation on day 9.5 of gestation. Conceptuses were then cultured for 48 hr and evaluated on day 11.5 for increased expression of inducible conceptal cytochrome P450 (P450). The three agents each elicited concentration-dependent increases in 7,8-benzoflavone (ANF)-inhibitable ethoxyresorufin O-deethylase (EROD) activities and increased P4501A1 mRNA as detected by primer-specific reverse transcriptase-polymerase chain reaction (RT-PCR) in cell-free preparations of the treated, cultured conceptuses. At effective inducing concentrations, dysmorphogenic or other embryotoxic effects were not detectable. At 20 microM concentrations, the three agents exhibited roughly equal induction that was approximately equivalent in magnitude (6- to 13-fold) to that achieved previously with exposures to MC in utero. Additions to the culture medium of 2.5 to 10 microM concentrations of dexamethasone (DEX) did not alter significantly the magnitude of MC-elicited induction in vitro. Repeated full-length sequencing of an RT-PCR-amplified cDNA revealed a coding region sequence identical to that reported for the P4501A1 sequence from adult rat liver. The results provide a basis for investigations, in the absence of maternal influences, of the regulation of mammalian conceptal P4501A1 in intact tissues during organogenesis, a gestational period critical in terms of the dysmorphogenic and other embryotoxic effects of foreign organic chemicals. The results are also pertinent to studies of embryotoxicity, particularly to the transplacental carcinogenicity, mutagenicity and dysmorphogenicity of P4501A1 substrates.

Benzene and benzene metabolites as embryotoxic agents: effects on cultured rat embryos

Benzene and several of its metabolites were investigated for dysmorphogenic and embryotoxic effects after direct exposures of cultured whole rat conceptuses. Benzene produced no statistically significant effects at concentrations up to 1.6 mM. Inclusion with 1.6 mM benzene of an hepatic xenobiotic-biotransforming system (S9) resulted in only minor decreases in embryonic growth parameters and no detectable dysmorphogenesis. Phenol, a major benzene metabolite, also elicited only minimal embryotoxicity at 1.6 mM concentrations. However, inclusion of an S9 system with phenol resulted in significant dysmorphogenic and embryotoxic effects at concentrations as low as 0.01 mM. For phenol bioactivation, S9 from phenobarbital-induced rats was the most effective, with induction by pregnenolone-16 alpha-carbonitrile, isopropanol, Aroclor 1254, no inducer, and 3-methylcholanthrene following in order of effectiveness. Bioactivating activity resided solely in the microsomal fraction. Metabolites coeluting on HPLC with hydroquinone and catechol were the major metabolites generated from phenol by each S9 system, but no significant correlation between specific metabolite generation and embryotoxicity was apparent. Of the benzene metabolites studied, trans, trans-muconaldehyde exhibited the highest embryotoxic potency but was not detectably generated by any of the S9 systems. Hydroquinone, catechol, and benzoquinone were approximately equipotent, each producing 100% lethality at 0.1 mM. Combined additions to the culture medium of hydroquinone together with phenol resulted in greater than additive effects, indicating a possible synergistic interaction between these metabolites and suggesting that peroxidase activity may be important to the mechanism of phenol-elicited embryotoxicity.

Temporal distribution, localization and metabolism of all-trans-retinol, didehydroretinol and all-trans-retinal during Xenopus development

Recently, the temporal and spatial distribution patterns of the retinoid receptor ligands 9-cis-retinoic acid and all-trans-retinoic acid were described in Xenopus embryos during early development [Creech Kraft, Schuh, Juchau and Kimelman (1994) Proc. Natl. Acad. Sci. U.S.A., in the press]. The present study demonstrates the presence and distribution of their likely precursors, all-trans-retinol, didehydroretinol, didehydroretinal and all-trans-retinal, as well as the occurrence of 4-oxo metabolites, in Xenopus embryos. The temporal and spatial distribution patterns of all-trans-retinol, didehydroretinol and all-trans-retinal did not coincide with that observed for 9-cis-retinoic acid but, in certain regards, were similar to the patterns delineated for all-trans-retinoic acid and all-trans-retinoyl beta-glucuronide. Evidence is presented that 9-cis-retinoic acid can be synthesized from both all-trans-retinoic acid and all-trans-retinol in Xenopus embryos, suggesting that the difference between the distributions of 9-cis-retinoic acid and the other retinoids may be caused by selective synthesis and/or protein binding of the 9-cis isomer.

Dysmorphogenic effects of nitric oxide (NO) and NO-synthase inhibition: studies with intra-amniotic injections of sodium nitroprusside and NG-monomethyl-L-arginine

Sodium nitroprusside (SNP), a chemical that is readily converted to nitric oxide (NO) in biological systems, was microinjected into the amniotic fluids of cultured whole rat conceptuses on day 10.5 of gestation and dysmorphogenic/embryotoxic effects were evaluated after a 24 hr incubation period. Injections of 217 ng/embryo (approximately 800 microM) resulted in whitened zones of dead cells in a discretely circumscribed region within the mesencephalon closely associated with the neural tube. These zones were observed with a high incidence after SNP microinjections and were referred to as "white caps" because of their microscopic appearance. At higher concentrations, the whitened zone extended into the rhombencephalon and occasionally appeared to extend the full length of the dorsal midline. The whitened zones of tissue separated readily from the apparently normal underlying tissues upon removal or disturbance of the amniotic membrane. Coinjection of ferrous hemoglobin with SNP selectively prevented the appearance of "white caps" but not other embryotoxic manifestations. Microinjections of the breakdown products of light-exposed SNP elicited generalized embryotoxicity but "white caps" were not observed. In separate experiments, we found that embryonic enzymes catalyzed significant conversion of arginine to citrulline, indicating expression of NO-synthase during organogenesis. NG-monomethyl-L-arginine (L-NMMA), a specific inhibitor of NO-synthase, was microinjected (50-150 ng/embryo; approximately 200-600 microM) on day 10.5 of gestation and produced malformations that differed markedly from those elicited by SNP. Failure of anterior and posterior neural tube closure and profound underdevelopment of the hyoid arch and optic cup were observed at concentrations that produced no apparent growth deficit. These studies with SNP and L-NMMA indicated that both an excess and a deficiency of NO can be embryotoxic/dysmorphogenic and suggest important roles for optimal levels of NO and NO synthases in normal embryonic development.

Embryogenesis in cultured whole rat embryos after combined exposures to 3,3',5-triiodo-L-thyronine (T3) plus all-trans-retinoic acid and to T3 plus 9-cis-retinoic acid

Retinoid-induced malformations of the jaw, ears, face, skull, eyes, and heart in humans and rodents are well known. Data on nuclear receptors and developmental toxicity bioassays indicate that thyroid hormones can modulate the biologic activity of retinoids. The present investigation concerned the potential for interactions of all-trans-retinoic acid (RA) with 3,3'5-triiodo-L-thyronine (T3) and of 9-cis-retinoic acid (9-cis-RA) with T3 in the morphogenesis of cultured whole rat embryos. Varying concentrations of retinoids or T3 were microinjected into the amniotic fluid or placed in the culture medium alone or in combinations of T3 with each retinoid. At 200 ng/ml, T3 increased the incidence of branchial arch defects produced by either RA or 9-cis-RA but did not elicit branchial arch defects alone except at concentrations significantly compromising survival (2,000 ng/ml; 32% mortality). Similarly high culture medium concentrations of T3 alone were associated with failure of neural tube closure in the rhombencephalon (rhombencephalic schisis). At this concentration, other dysmorphia were minimal and at 670 ng/ml T3, no dysmorphogenic or embryotoxic effects could be detected. Modulation of T3 effects by the yolk sac placenta was suggested by failure of microinjected T3 to elicit dysmorphia at very high amniotic fluid concentrations. RA (300 ng/ml) or 9-cis-RA (600 ng/ml) alone elicited no or minimal rhombencephalic schisis at the highest concentrations studied. RA plus T3 produced a much greater than additive effect on rhombencephalic schisis, whereas 9-cis-RA plus T3 produced a less than additive effect. Conversely, much greater than additive effects on anterior schisis were observed for 9-cis-RA plus T3 whereas combined effects of RA and T3 were approximately additive. For most other dysmorphia, the combined effects of each retinoid with T3 were greater than additive and were particularly striking for cephalic defects.

Cytochrome P-450-dependent biotransformation of 2-acetylaminofluorene in cell-free preparations of human embryonic hepatic, adrenal, renal, pulmonary, and cardiac tissues

Human embryonic hepatic, renal, adrenal, pulmonary, and cardiac tissues (gestational age = 50-60 days) were probed for functional P-450 isoforms with 2-acetylaminofluorene (AAF) in cell-free preparations. Each of these tissues exhibited P-450-dependent hydroxylation at several positions on the AAF molecule, although activities in renal, pulmonary, and particularly cardiac preparations were generally low. N-hydroxylation activities were marginal to undetectable in all five tissues, but 7-hydroxylation was detectable in each tissue. Highest aromatic ring-hydroxylation activities were observed in hepatic tissues, and adrenal tissues also exhibited relatively high activities for ring-hydroxylation, particularly at carbon-7. The 9-hydroxylated AAF metabolite (9-OH-AAF) was the predominant metabolite for all human embryonic tissues, but generation via catalysis by P-450 isoforms appeared to be minimal/negligible. Activity profiles for human embryonic tissues (days 50-60 of gestation) were compared with those of 12 separate, vector-expressed human P-450 isoforms, with those of human fetal tissues (days 72-140 days of gestation), with those of various rodent embryonic tissues, and with those of adult rhesus monkey and adult rat tissues preexposed to inducing agents. These analyses suggested that each of the human embryonic tissues studied expresses functional, xenobiotic-biotransforming P-450 isoforms, but contrasted with previous investigations with phenoxazone ethers as functional P-450 probes. Resolution of these apparent differences will require further research. Early prenatal expression of functional P-450 isoforms in organogenesis-stage human embryonic tissues has important implications for our understanding and predicting of teratogenic/embryotoxic and other biologic effects of exposures to drugs and other environmental chemicals during human pregnancy.

In vitro embryotoxicity of N-methyl-N-(7-propoxynaphthalene-2-ethyl)hydroxylamine (QAB): evidence for N-dehydroxylated metabolite as a proximate dysmorphogen

The rat conceptus biotransforms N-methyl-N-(7-propoxynaphthalene-2-ethyl)hydroxylamine (QAB) in vitro to 7-propoxynaphthalen-2-ylacetic acid (QAA) and six more (M1 to M6) metabolites. Thus far, M4 has been identified as N-demethyl-QAB and M6 as N-desoxy-QAB. We investigated which of these two metabolites might be involved in QAB-embryotoxicity in vitro. Conceptuses were cultured from day 9.5 to 11.5 post-coitum, and were exposed to N-demethyl-QAB or N-desoxy-QAB either in the culture medium or by microinjection directly into the amniotic cavity. When added to the culture medium, N-demethyl-QAB (No Observed Adverse Effect Level, NOAEL, for growth 122 microM and for differentiation 41 microM) was less active than QAB itself (NOAEL for growth and differentiation 12 microM). N-desoxy-QAB caused severe growth retardation and an impairment of differentiation at a concentration of 11 microM (NOAEL 3.6 microM). As regards causing anomalies, the NOAEL of N-demethyl-QAB (41 microM) was 10-fold higher than that of QAB (NOAEL 3.9 microM) and that of N-desoxy-QAB (NOAEL 3.6 microM). At an intraamniotic concentration of 0.7 mM, N-demethyl-QAB caused no effects on growth and differentiation and no increase of anomalies was observed, whereas QAB and N-desoxy-QAB each elicited an increase in dysmorphogenic embryos at equimolar concentrations without affecting growth and differentiation. It is, therefore, concluded that N-desoxy-QAB, but not N-demethyl-QAB, could be a proximate dysmorphogen responsible for the embryotoxicity/teratogenicity of QAB in vitro.

9-cis-retinoic acid: a direct-acting dysmorphogen

Experiments in vitro with cultured rat conceptuses demonstrated that 9-cis-retinoic acid (9-cis-RA) (300 ng/mL amniotic fluid) produced branchial arch and somite defects similar to those elicited by equal concentrations of all-trans-retinoic acid (all-trans-RA), but with an increase in cephalic defects that included missing optic vesicles. After conceptuses were intraamniotically microinjected with 600 ng 9-cis-RA/mL amniotic fluid on day 10 of gestation, an unusual heart defect was also observed. HPLC analyses indicated that 9-cis-RA readily underwent conversion to the less active metabolite, 13-cis-retinoic acid (13-cis-RA), in cultured conceptuses during the first 4 hr after treatment but only after 6 hr could elevated levels of the potent dysmorphogen all-trans-RA be detected. In separate experiments, conversion of 13-cis-RA or of all-trans-RA to 9-cis-RA could not be detected during a 6-hr embryo culture period. Endogenous levels of 9-cis-RA in whole rat embryos also were below limits of detection but small quantities of this isomer could be detected in neonatal rat eye and human embryonic brain. Our present study strongly suggests that 9-cis-RA is a direct-acting dysmorphogen with probable specific target sites of action.

Tissue levels of retinoids in human embryos/fetuses

In nonhuman vertebrate embryos, two endogenous retinoids with significant morphogenic activities have been identified thus far: all-trans retinoic acid and 3, 4-didehydroretinoic acid. To date, no information is available concerning endogenous retinoid levels in developing human embryos or fetuses. The purpose of the present study was to provide data relating to normal levels of retinoids in various human embryonic and fetal tissues at various stages of gestation measurable with HPLC techniques. Our investigations show that all-trans-retinoic acid, 13-cis-retinoic acid, retinol, all-trans-retinoyl-beta-glucuronide, and one unidentified metabolite were all present and quantifiable in several human embryos and fetuses investigated. Tissue levels of retinol were consistently much higher than those of the other three detected metabolites; 4-oxo metabolites were below the levels of detection in all samples studied.

Chemical teratogenesis

This review has briefly summarized what is currently known concerning the mechanisms whereby several groups of chemicals regarded as "recognized" human teratogens elicit their respective teratogenic effects. It is evident that the extent of our understanding of mechanisms for individual chemicals varies dramatically from that of a reasonably good understanding for methotrexate and other folic acid antagonists to that of virtually no understanding for the most widely recognized human teratogen, thalidomide. Even with methotrexate, however, much remains to be learned pertaining to mechanisms--i.e., the critical links in the chain of events between dihydrofolate reductase inhibition and the manifestation of specific abnormalities. Nevertheless, we can take some comfort in being able to say that we understand the primary causative mechanism. For thalidomide, as well as several others the chemical represents both a shame and a challenge--a challenge that should be addressed with our most serious efforts.

Mechanisms of glucagon-induced increases in rates of cytochrome P450-dependent pentoxyphenoxazone O-depentylation in cultured rat conceptuses

Inclusions of glucagon (1.0 or 2.0 microM, final concentrations) in the media of cultured whole rat conceptuses resulted in concentration-dependent increases in measured rates of O-depentylation of pentoxyphenoxazone in cell-free preparations of conceptal tissues. Enzymic activities were assayed 24 h after initial exposure of the conceptuses to glucagon on day 10 of gestation. Glucagon elicited increases in tissue levels of cAMP that were parallel to those produced by 3-isobutyl-1-methylxanthine over the same time period. Tissue cAMP levels were maximal after 2 h, rapidly returned to control levels and were also equal to background levels in controls after the 24 h culture period. Dibutyryl cAMP, 3-isobutyl-1-methylxanthine, theophylline, and RO201724, a cAMP-selective phosphodiesterase inhibitor, each produced 75 to 100% increases in O-dealkylase activity. Dibutyryl cGMP and two phosphodiesterase inhibitors, enoximone (cGMP-inhibited) and zaprinast (cGMP-specific), each failed to produce statistically significant increases in O-depentylase activity. The O-depentylase was tentatively identified as a conceptus-specific P450 cytochrome that is synthesized predominantly in tissues of the visceral yolk sac. The results indicated that glucagon may upregulate a unique, xenobiotic-biotransforming P450(s) via a long-term mechanism(s) specifically involving tissue cAMP.

Conceptual biotransformation of 4-oxo-all-trans-retinoic acid, 4-oxo-13-cis-retinoic acid and all-trans-retinoyl-beta-glucuronide in rat whole embryo culture

In cultured rat conceptuses, intraamniotic microinjections of 2500 ng/mL of 4-oxo-13-cis-retinoic acid, 600 ng/mL 4-oxo-all-trans-retinoic acid or 4000 ng/mL all-trans-retinoyl-beta-glucuronide, produce qualitatively and quantitatively similar patterns of dysmorphogenesis as those reported after the intraamniotic microinjection of 250 ng/mL all-trans-retinoic acid [Lee et al., Teratology 44: 313-323, 1991; Creech Kraft et al., Teratology 45: 259-270, 1992]. In the present study, we utilized HPLC techniques to analyze retinoid levels in cultured rat conceptuses, 1.5 hr after intraamniotic microinjections of 4-oxo-13-cis-retinoic acid (2500 ng/mL), 4-oxo-all-trans-retinoic acid (600 ng/mL) or all-trans-retinoyl-beta-glucuronide (4000 ng/mL). Our findings show that, after the microinjections of 4-oxo-all-trans-retinoic acid or 4-oxo-13-cis-retinoic acid (at these selected concentrations), 4-oxo-all-trans-retinoic acid was predominant in the embryos proper at concentrations of about 200 nM. This was roughly equivalent to the levels of all-trans-retinoic acid assayed after microinjections of all-trans-retinoyl-beta-glucuronide (4000 ng/mL). We conclude from these studies that both 4-oxo-all-trans-retinoic acid and all-trans-retinoic acid behave as ultimate or proximate dysmorphogens.

Correlations between conceptal concentrations of all-trans-retinoic acid and dysmorphogenesis after microinjections of all-trans-retinoic acid, 13-cis-retinoic acid, all-trans-retinoyl-beta-glucuronide, or retinol in cultured whole rat embryos

Retinol (4,000 ng/ml), all-trans-retinoyl-beta-glucuronide (4,000 ng/ml), and 13-cis-retinoic acid (1,500 ng/ml) each produced dysmorphogenic effects qualitatively similar to those elicited by 250 ng/ml of all-trans-retinoic acid after microinjections of the respective individual retinoids into the amniotic cavities of cultured whole rat embryos. Subsequent HPLC analyses of the cultured whole conceptuses, embryos proper, yolk sacs, and culture media (24 hr after microinjections) indicated that conceptal biotransformation of each of the retinoids had occurred during the culture period. All-trans-retinoic acid was present in the embryos proper at quantitatively similar concentrations (20-100 nM) after microinjections of the selected quantities of each of the microinjected retinoids: retinol, all-trans-retinoyl-beta-glucuronide, 13-cis-retinoic acid, or all-trans-retinoic acid. The results suggested that all-trans-retinoic acid acted as an ultimate dysmorphogen for the retinoids tested with respect to the anomalies monitored in the embryo culture system.

Microinjections of cultured rat conceptuses: studies with 4-oxo-all-trans-retinoic acid, 4-oxo-13-cis-retinoic acid and all-trans-retinoyl-beta-glucuronide

4-Oxo-all-trans-retinoic acid, 4-oxo-13-cis-retinoic acid and all-trans-retinoyl-beta-glucuronide were intraamniotically microinjected in rat embryos on day 10 of gestation and cultured until day 11.5. A comparison of the concentration-effect relationships showed that the dysmorphogenic effects produced by these metabolites were qualitatively similar to those of parent all-trans-retinoic acid. Compared with all-trans-retinoic acid (300 ng/ml), the dysmorphogenic effects were elicited by a 2-fold higher concentration of 4-oxo-all-trans-retinoic acid, an approximately 10-fold higher concentration of 4-oxo-13-cis-retinoic acid and a 16-fold higher concentration of all-trans-retinoyl-beta-glucuronide. A surplus of uridine 5'-diphospho-glucuronic acid, microinjected together with 300 ng/ml all-trans-retinoic acid, decreased the observed embryo-toxicity of all-trans-retinoic acid, suggesting the possibility of glucuronidation in tissues of the conceptus per se. The results of the study provide further support for the hypothesis that 4-oxo-all-trans-retinoic acid and all-trans-retinoic acid are, in contrast to the corresponding cis-isomers and glucuronides, ultimate dysmorphogenic retinoids.

Human embryonic cytochrome P450S: phenoxazone ethers as probes for expression of functional isoforms during organogenesis

Human embryonic tissues were investigated during the period of organogenesis with a combination of substrate probes, selective inhibitors and immunoprobes in terms of their capacity to express functional P450 isoforms. A series of phenoxazone ethers utilized as substrate probes revealed that human embryonic hepatic, pulmonary, renal, adrenal and cardiac tissues each contained a complement of functional P450 isoforms when analyzed between days 50 and 60 of gestation. Preparations of each of these tissues contained isoforms capable of catalyzing O-demethylation, O-deethylation, O-depentylation and O-debenzylation of the respective phenoxazone ethers. Investigations with chemical inhibitors and inhibitory antibodies as well as comparisons with vector-expressed, human P450 isoforms suggested that isoforms of P450 subfamilies 1A, 2B, 2C or 3A were not major contributors to any of the observed reactions. The P450-dependent reactions studied exhibited several unexpected and unusual characteristics including a preference for NADH over NADPH as the initial electron donor. Results were consistent with the concept that conceptal-specific P450 isoforms participate in the human embryonic O-dealkylation/debenzylation probe reactions investigated.

Microinjection of cultured rat embryos: studies with retinol, 13-cis- and all-trans-retinoic acid

Retinol, all-trans-retinoic acid or 13-cis-retinoic acid were intraamniotically microinjected in rat embryos on day 10 of gestation and cultured until day 11.5. A comparison of the concentration-effect relationships of the retinoids showed that the dysmorphogenic effects were qualitatively similar for all three, but were elicited by a low concentration of all-trans-retinoic acid (250 ng/ml), a 6- to 7-fold higher concentration of 13-cis-retinoic acid and an approximately 16-fold higher concentration of retinol. After microinjection of 2,000 ng/ml of retinol, no dysmorphogenesis was observed but instead an increase in all growth parameters as compared to the controls.

Asymmetric development of mitochondrial activity in rat embryos as a determinant of the defect patterns induced by exposure to hypoxia, hyperoxia, and redox cyclers in vitro

Previous study has shown that midorganogenesis-stage rat embryos exposed to strong redox cyclers under moderate hypoxia in vitro develop severe necrotic defects on the right side. Similar effects can be produced by exposure to severe hypoxia alone. Studies presented here indicate that exposure to severe but survivable hyperoxia induces comparable necrotic degeneration on the left sides of all embryos. We hypothesize that the basis of these axially asymmetric defects is relatively precocious mitochondrial maturity on the left side of the embryo. In order to investigate this hypothesis, we compared mitochondrial oxygen utilization (NADH oxidase activities) on either side of rat embryos between days 11 and 14 of gestation. Activities were consistently higher on the left side during this period and significantly higher on day 11. We also found that the asymmetric embryotoxicity induced by niridazole, a strong redox cycler, could be attenuated by prior culture under hyperoxic conditions. We propose that mitochondrial immaturity on the right results in inadequate energy generation under hypoxic conditions, either directly or as a result of redox cycling. On the other hand, necrosis associated with hyperoxic conditions results from "leakage" of superoxide from functionally mature mitochondria on the left side.

Organogenesis-stage cytochrome P450 isoforms: utilization of PCR for detection of CYP1A1 mRNA in rat conceptal tissues

Utilizing the reverse transcriptase-linked polymerase chain reaction, we analyzed the capacity of three groups of rat conceptal tissues to express cytochrome P4501A1 (CYP1A1) mRNA during the dysmorphogenesis-sensitive stage of organogenesis. The visceral yolk sac, ectoplacental cone and embryo proper each were investigated on day 12 of gestation with and without prior exposure in utero to 3-methylcholanthrene as inducing agent. With two sets of discriminating oligonucleotide primers, definitive, reproducible signals were detectable only in tissues from 3-methylcholanthrene preexposed conceptuses. Signals of highest intensity were observed with visceral yolk sac tissues and signals of lowest intensity were observed with tissues of the embryo per se. Specificities of the amplified cDNAs were verified using Southern blotting with hybridization to an internal oligonucleotide probe. The results indicate that organogenesis-stage conceptual tissues of the rat will express CYP1A1 mRNA in response to environmental transregulating agents.