Monitoring of cytochrome P-450 1A activity by determination of the urinary pattern of caffeine metabolites in Wistar and hyperbilirubinemic Gunn rats.

Various studies suggest that induction of cytochrome P-450 1A (CYP1A) might be a valuable therapeutic modality for reducing the hyperbilirubinemia of infants with Crigler-Najjar syndrome type I (CNS-I), a severe form of congenital jaundice. To evaluate inducers of CYP1A as possible tools in the treatment of hyperbilirubinemia, a novel assay was established, based on the analysis of the urinary pattern of caffeine metabolites in rats. Wistar rats received [1-Me-(14)C]-caffeine (10 mg/kg i.p.), before and 48h after administration of the potent CYP1A inducer 5,6-benzoflavone (BNF) (80 mg/kg, i.p.). A substantial increase in the fractions of the terminal caffeine metabolites 1-methyluric acid (1-U), 1-methylxanthine (1-X), and a concomitant decrease in the caffeine demethylation product 1,7-dimethylxanthine (1,7-X) was observed after application of BNF. The ratio of the caffeine metabolites (1-U+1-X)/1,7-X may serve as an index of CYP1A activity in rats in vivo. Hyperbilirubinemic, homozygous (jj) Gunn rats are an accepted model for human CNS-I. In male jj Gunn rats treated with BNF or with indole-3-carbinol (I3C, 80 mg/kg, oral gavage), the inducing effect of BNF and 13C on CYP1A activity was confirmed by the urinary pattern of caffeine metabolites, and was parallelled by a decrease in plasma bilirubin levels. These data demonstrate the usefulness of the established caffeine assay for the evaluation of inducers of CYP1A as tools for reducing hyperbilirubinemia and further confirm the potential value of I3C in the treatment of CNS-I.

Extrahepatic drug metabolizing enzymes.

Drug metabolizing enzymes are known to be present and active in most extrahepatic tissues. In this review, evidence is presented that expression and activity of several extrahepatic drug-metabolizing enzymes are regulated in unique ways, which may be associated with tissue functions and activities. In several instances evidence is offered that hormonal effects may be regulated through tissue specific distribution and/or responses of transcription factors.

Interactions between bilirubin and reactive oxygen species in liver microsomes and in human neutrophil granulocytes.

Bilirubin protects polyunsaturated fatty acids from lipid peroxidation, thus preventing damage by reactive oxygen species to cell membranes and proteins. On the other hand, such reactive oxygen species may contribute to the degradation and elimination of bilirubin. We therefore examined the interactions between bilirubin and reactive oxygen species. Bilirubin is decomposed in microsomes via a NADPH-independent process. This reaction appears to be mediated by H2O2 or by the hydroxyl radical since it is stimulated by exogenous H2O2 and by cytochrome P450 inducers, which increase H2O2 production in microsomes, and is inhibited by the hydroxyl radical scavenger sodium benzoate. These results suggest that cytochrome P450 may act as a peroxidase or as a Fenton catalyst in bilirubin degradation. On the other hand, bilirubin inhibits the NADPH consumption of microsomes as well as the NADPH oxidase activity of human neutrophil granulocytes and the resulting superoxide formation in these cells. This effect on superoxide concentration may be partially due to direct interaction between superoxide and bilirubin, since bilirubin reduces the superoxide concentration in a xanthine oxidase system. Bilirubin degradation is inhibited by superoxide dismutase suggesting that bilirubin may be oxidized in this system by the superoxide radical. The bilirubin-induced reduction in superoxide concentration in the supernatant of granulocytes suggests that hyperbilirubinemia may compromise immune function.

Postnatal development of somatosensory evoked potential in jaundiced Gunn rats and effects of sulfadimethoxine administration.

Bilirubin encephalopathy results from the entry of bilirubin into the brain and is expressed by motor, sensory, and/or behavioral impairment. The jaundiced (jj) Gunn rat is a valuable animal model for studying the kinetics of bilirubin-induced neurotoxicity. This is often done by recording evoked potentials, which are also used as indices of brain damage in infants who develop neonatal jaundice, as is the case with the auditory nerve and brainstem evoked response (ABR). The present study describes the postnatal development of the somatosensory evoked potential (SEP) in Gunn rats. No effects of jaundice on the SEP were found in young jj rats (16-28 d). However, adult (3-4 mo) jj rats had prolonged latencies and decreased amplitudes of the P2 component of the SEP compared with adult nonjaundiced (Jj) rats. These changes in the SEP of jaundiced rats may reflect a synaptic lesion in these animals, possibly due to cumulative and/or progressive damage induced by bilirubin during the first 3 mo of life. After sulfadimethoxine administration, marked latency prolongations (2-6%) were observed in the early components of SEP in young (3-wk-old) jj (but not Jj) rats, as early as 2 h after injection. These changes, which became more severe (4-10%) with time, seem to be mostly peripheral. The present results suggest that the SEP may be a sensitive marker for the massive entry of bilirubin into the nervous system, and could serve as part of an evoked potential battery (in addition to visual evoked potential and ABR) in assessing bilirubin-induced neurotoxicity in jaundiced newborns and infants.

Visual evoked potential abnormalities in jaundiced Gunn rats treated with sulfadimethoxine.

The manifestations of bilirubin encephalopathy include disturbances in the visual pathway (visual gaze paralysis and distorted visual perception). In the young jaundiced Gunn rat (jj) model of hyperbilirubinemia, significant differences in visual evoked potential (VEP) patterns have been recorded during development. In the present study, the effects of sulfadimethoxine (SDM) on VEP and electroretinogram (ERG) were examined in 3-wk-old jj rats. This drug displaces bilirubin from its albumin binding sites in the circulation, shifting it into tissues including the brain. Marked latency prolongations (11-20%) and reduced amplitudes (20-64%) were observed in the different wave components of the VEP. These changes were evident as early as 2 h after injection of the drug and persisted thereafter for another 4 h. On the other hand, ERG changes (significant prolongation of wave b) became apparent in these animals only 6 h after SDM injection. These results suggest that, although some changes in the retina may occur after a massive entry of bilirubin into the nervous system, the primary damage in the visual pathway after bilirubin exposure is probably beyond the retina.

Contribution of asphyxia to the induction of hearing impairment in jaundiced Gunn rats.

OBJECTIVES: This study was designed to determine whether asphyxia contributes to the induction of hearing impairment during neonatal jaundice. METHODS: Asphyxia was induced in jaundiced and nonjaundiced Gunn rats on postnatal days 1 (low bilirubin levels) and 10 (elevated bilirubin levels). Auditory nerve-brainstem evoked response thresholds were assessed in 21- and 28-day and 3-month-old rats. RESULTS: Asphyxia by itself or jaundice by itself did not lead to any type of hearing impairment. However, the combination of both high plasma bilirubin levels and asphyxia in 10-day-old rats but not in 1-day-old rats was accompanied by a progressive hearing loss in these rats. CONCLUSIONS: The contributory effect of asphyxia on neonatal jaundice may have important clinical relevance if asphyxia, for example, respiratory distress, accompanies neonatal jaundice.

Marked endogenous activation of the CYP1A1 and CYP1A2 genes in the congenitally jaundiced Gunn rat.

The homozygous recessive jaundiced Gunn rat lacks expression of bilirubin UDP-glucuronosyltransferase and serves as a model for Crigler-Najjar syndrome type I, in which high and toxic plasma levels of bilirubin result from this genetic defect in bilirubin conjugation. Both rats and humans dispose of this heme waste product by an alternate metabolic route that involves oxidation of the compound, followed by biliary excretion of the more polar metabolites. To determine the role of cytochrome P450 in this process, hepatic levels of cytochrome P450 mRNA and protein were measured in jaundiced and nonjaundiced Gunn rats as a function of age and sex. The mRNA and protein levels of cytochrome P450(CYP) 1A1 and CYP1A2 were markedly elevated in the jaundiced rats at the age of 10 days, compared with their nonjaundiced littermates. Levels of CYP2E1 mRNA and protein did not differ between these rats, indicating that the CYP1A P450 genes were specifically induced. CYP1A1 mRNA and protein levels increased further in the jaundiced animals between 10 days and 1 month of postnatal life but remained undetectable in the nonjaundiced littermates. On the other hand, CYP1A2 mRNA and protein content increased during this time period in both jaundiced and nonjaundiced rats, but at the age of 1 month there were no major differences between the two groups. CYP1A2 mRNA and protein levels were indistinguishable in 3-month-old jaundiced and nonjaundiced Gunn rats, whereas CYP1A1 could not be detected in either group. These data suggest that young jaundiced Gunn rats cope with the degradation of toxic bilirubin by increasing hepatic levels of CYP1A1 and CYP1A2. On the other hand, normal developmental activation of CYP1A2 may provide the alternative pathway for bilirubin degradation in adult animals. This is the first demonstration of the induction of cytochrome P450 gene expression to permit the elimination of an endogenously generated neurotoxic chemical in a genetic disease in which the normal excretory mechanism is impaired.

Postnatal development of flash visual evoked potentials in the jaundiced Gunn rat.

In this study, the postnatal development of flash visual evoked potential (VEP) has been monitored in the jaundiced (jj) Gunn rat model of neonatal hyperbilirubinemia to determine whether this evoked response is affected by bilirubin-induced neurotoxicity. VEP could first be recorded at 16 d of age. The jj rats exhibited prolonged wave latencies and lower wave amplitudes during the 3rd wk of postnatal life, when compared with their nonjaundiced littermates. There was no correlation at 21 d of age between VEP parameters and either bilirubin levels or body weight. About one third of the jj animals died between 21 and 28 d of age. The average VEP wave latencies at 21 d of age of the rats who were to die was prolonged compared with those of rats who survived till at least 28 d of age. Thus, the latency of VEP waves at the age of 21 d appears to be related to the further outcome of jj Gunn rats. Although wave amplitudes were lower in jj as compared with nonjaundiced 21-d-old animals, there were no amplitude differences between the jj rats who would die and those who would survive during the 4th wk of life. These findings may contribute to the understanding of the pathogenesis of bilirubin encephalopathy in the neonatal period.

Magnesium and phosphate enrichment of culture medium stimulates the proliferation of epidermal cells from newborn and adult mice.

The proliferation and differentiation of mouse epidermal cells can be sequentially analyzed by modification of extracellular calcium. Newborn cells cultured in low calcium medium (less than 0.1 mM) proliferate as a monolayer and maintain a typical basal cell phenotype in culture but have a limited proliferative capacity and short lifespan. Elevation of the magnesium content of the culture medium from 1 to 5 mM stimulated the proliferation of newborn mouse (1-3 days old) keratinocytes. Maximal DNA synthesis rates, as determined on day 5 of culture, were up to 2-3-fold higher in the magnesium-enriched cultures. Exposure to high magnesium caused 3-4-fold increases in the DNA content of newborn keratinocyte cultures, and extended the confluent phase of epidermal cell growth to over 10 days. Other divalent cations (strontium, copper, zinc, nickel, beryllium, and barium) did not improve keratinocyte growth in culture. Keratinocytes from the tail skin of adult (3 months old) mice displayed an absolute requirement for high phosphate in the culture medium. The medium containing an optimal (10 mM) phosphate concentration prevented the cell detachment caused by the standard low (1 mM) phosphate medium, and in combination with an elevated magnesium content (10-15 mM) it markedly increased both DNA synthesis rates and DNA content of the adult cell cultures. Optimally growing, newborn or adult cultures contained less cells in the G1 phase of the cell cycle and more cells in S and G2 +M. The addition of phosphate and magnesium per se did not induce keratinocyte differentiation and did not interfere with the high calcium (1 mM)-induced differentiation.

A marked increase in latency of fetal rat liver UDP-glucuronosyltransferase accompanies the fluidization of the microsomal membrane in the perinatal period.

The activity of rat liver microsomal uridine diphosphate glucuronosyltransferase towards 4-methylumbelliferone (4-MU-UDPGT) increased markedly between day 18 of fetal life and day 1 after birth. This increase in UDPGT activity resulted largely from a gradual increase in the fraction of enzyme activity which was detected after pretreatment of microsomes with digitonin ('latent' activity). The latent fraction of 4-MU-UDPGT activity approached at birth the values obtained for adult rat liver. This perinatal increase in latency of 4-MU-UDPGT activity may be associated with the concurrent fluidization of the microsomal membrane which is observed in the perinatal period.

A rapid procedure for flow cytometric DNA analysis in cultures of normal and transformed epidermal cells.

A simple, rapid, and highly reproducible procedure for flow cytometric DNA analysis has been adapted for studying cell cycle kinetics in epidermal cell cultures. The preparation of cell nuclei and their staining with the fluorescent dye propidium iodide were performed directly on the culture dish, without prior suspension and fixation of the cells. Singly dispersed nuclei were produced by mild trypsinization of cells in the presence of the nonionic detergent Nonidet P-40 and spermine. The culture dishes could be kept frozen for prolonged periods of time before trypsinization and staining, without affecting either the recovery of nuclei or the cell cycle distribution profiles. This remarkable stability of cell nuclei greatly simplified the analysis of multiple samples in cell cycle kinetic studies. This method was used to analyze the cell cycle distribution in cultures of normal and transformed mouse epidermal cells, human colon carcinoma cells, primary bovine aortic endothelial cells, and fibroblastic and myogenic cell lines. This procedure should be very useful in studying growth kinetics, differentiation, and transformation of epidermal as well as other adherent cell types.

Immunohistochemical localization of cytochrome P-450 in rat brain.

The presence of cytochrome P-450 in rat brain was studied by immunohistochemistry, using antibodies to cytochrome P-450 purified from livers of phenobarbital- or 3-methylcholanthrene-treated rats. Immunoreactive nerves were observed only in brain sections incubated with immunoglobulin-G to 3-methylcholanthrene-induced cytochrome P-450. This immunoreactivity was abolished by preabsorption of the antibody with highly purified rat liver cytochrome P-450c, the major cytochrome P-450 isozyme induced by 3-methylcholanthrene, but was not affected by other cytochrome P-450 isozymes induced by phenobarbital, isosafrole or pregnenolone-16 alpha-carbonitrile. The most abundant concentration of nerve fibers with cytochrome P-450 immunoreactivity was observed in the globus pallidus. Immunoreactive fibers were also observed in the caudate putamen, amygdala, septum, ventromedial nucleus of the hypothalamus, medial forebrain bundle, ansa lenticularis, and ventromedial portion of the internal capsule and crus cerebri. Cell bodies with cytochrome P-450 immunoreactivity were observed in the caudate putamen and in the perifornical area of the hypothalamus. The cytochrome P-450 immunoreactive fibers in the globus pallidus and caudate putamen do not appear to emanate from cell bodies in the substantia nigra, since there was no reduction in the density of these fibers after unilateral stereotaxic electrolytic destruction of the substantia nigra (zona compacta and reticulata). Our data suggest that these striatal nerve processes are derived from cell bodies within the caudate putamen itself. The present results indicate that rat brain contains a form of cytochrome P-450 with antigenic relatedness to the hepatic 3-methylcholanthrene-inducible cytochrome P-450c.(ABSTRACT TRUNCATED AT 250 WORDS)

Increase in a specific cytochrome P-450 isoenzyme in the liver of congenitally jaundiced Gunn rats.

Congenitally jaundiced (jj) Gunn rats had a greater hepatic microsomal content of a cytochrome P-450 isoenzyme, P-450c, than did the non-jaundiced (Jj) rats. No differences in content of P-450b, P-450d and pregnenolone-16 alpha-carbonitrile-induced (PCN) P-450 were found between jj and Jj rats. This is the first demonstration of a constitutive increase in a specific cytochrome P-450 isoenzyme in association with a genetic defect.

Fetal and adult human liver differ markedly in the fluidity and lipid composition of their microsomal membranes.

The fluidity and lipid composition of the human hepatic microsomal membrane were studied in 11 livers from 16- to 21-week-old fetuses and in 5 adult livers, and compared with those of fetal and adult rat liver microsomes. Membrane fluidity was analyzed by measurement of fluorescence polarization using the fluorophore 1,6-diphenyl-1,3,5-hexatriene. The lipid apparent microviscosity (eta) of human fetal liver microsomes was 2.17 +/- 0.13 poise, as compared with 1.08 +/- 0.08 poise in adult liver (p less than 0.001). Similar differences in fluidity were found between fetal and adult rat liver microsomes. The more "fluid" adult microsomes had higher phospholipid/cholesterol and phosphatidylcholine/sphingomyelin molar ratios than those of the more "rigid" fetal microsomes. The degree of unsaturation of the adult microsomal lipids was much higher than that of the fetal lipids. The ratios of unsaturated/saturated fatty acids in microsomal lipids highly correlated with the eta values obtained for the combined group of fetal and adult human livers, suggesting that the developmental increase in degree of unsaturation of the microsomal lipids is a major determinant of the increased fluidity of adult as compared with fetal liver microsomes. These differences in fluidity and lipid composition between fetal and adult human liver microsomes may be a critical factor in the regulation of hepatic microsomal drug and carcinogen metabolizing enzyme activity, and could so determine the extent of toxicity and teratogenicity of drugs and/or their metabolites in the developing human fetus.

Glucocorticoids increase the fluidity of the fetal-rat liver microsomal membrane in the perinatal period.

Dexamethasone, a synthetic glucocorticoid, was administered to pregnant rats during the last week of pregnancy in order to examine its effects on the fluidity of the developing fetal-rat liver microsomal membrane. This early prenatal exposure to dexamethasone, which preceded the natural appearance of fetal corticosteroids, markedly accelerated the normal perinatal course of fluidization of this membrane. The lipid apparent microviscosity, which was determined by measurement of fluorescence polarization, decreased in 21-days-old treated fetuses to values that were indistinguishable from those of untreated newborn rats. This dexamethasone-mediated acceleration of membrane fluidization was associated with an increase in the index of unsaturation of the fatty acyl moiety of microsomal lipids. Dexamethasone caused a significant increase in the microsomal content of polyunsaturated fatty acids (arachidonic and linoleic acid), which was accompanied by a decrease in content of monoenoic fatty acids (oleic and palmitoleic acid). This early exposure in utero to dexamethasone precociously induced the changes in fatty acid composition of fetal-rat liver microsomal lipids that normally occur between the last day of pregnancy and the first day of extra-uterine life. These results suggest that endogenous glucocorticoids play a major role in the perinatal fluidization of the rat liver microsomal membrane.

Effect of combined treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin and phototherapy on bilirubin metabolism in the jaundiced Gunn rat.

2,3,7,8-Tetrachlorodibenzo-p-dioxin, a potent inducer of microsomal cytochrome P448-dependent monoxygenases, and phototherapy both accelerate bilirubin metabolism and decrease jaundice in Gunn rats. The effects of combined treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin and light were studied in these rats by applying phototherapy for 65 hr, beginning 5 days after induction with 2,3,7,8-tetrachlorodibenzo-p-dioxin. 2,3,7,8-Tetrachlorodibenzo-p-dioxin pretreatment caused a 75% decline in plasma bilirubin in 5 days, with no change thereafter, whether or not the rats were exposed subsequently to phototherapy. In the uninduced rats, plasma bilirubin levels declined by 55% after 40 hr of phototherapy. As determined by [14C]bilirubin kinetics, both 2,3,7,8-tetrachlorodibenzo-p-dioxin and phototherapy increased fractional bilirubin turnover and decreased the total bilirubin pool. In the 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced rats, the contracted bilirubin pool shifted from skin to liver, but these tissue pools did not change further during phototherapy. By contrast, in uninduced rats, phototherapy decreased the cutaneous bilirubin pool, which is the main target of phototherapy. 2,3,7,8-Tetrachlorodibenzo-p-dioxin was more effective than phototherapy in diminishing plasma bilirubin levels and the total bilirubin pool, but the combined treatment (2,3,7,8-tetrachlorodibenzo-p-dioxin followed by phototherapy) was no more effective than 2,3,7,8-tetrachlorodibenzo-p-dioxin alone.

Effects of phototherapy on the activity of microsomal mixed-function monooxygenases and the pharmacokinetics of [14C] hexobarbital in jaundiced and heterozygous Gunn rats.

We assessed the effects of in vivo phototherapy of Gunn rats on the activity of hepatic microsomal mixed-function monoxygenases and on the in vivo pharmacokinetics of [14C]HB. In experiment 1 no serial changes were seen in activities of hexobarbital hydroxylase or benzo(a)pyrene hydroxylase in hepatic microsomes isolated after 2, 4 or 7 days from homozygous jaundiced female Gunn rats exposed to continuous phototherapy or in matched Gunn rats maintained under dim light. In experiment 2 homozygous jaundiced (jj) and heterozygous nonjaundiced (Jj) Gunn rats of both sexes each received i.v. [14C]HB on 2 successive days. In random order, each was exposed on the first or second day to phototherapy for 5.5 hr, beginning 0.5 hr before the administration of HB; otherwise, each was kept under dim light. Plasma [14C]HB in arterial blood samples was separated chromatographically from its labeled metabolites, and biexponential plasma disappearance curves for [14C]HB were analyzed by a SAAM-23 computer program. Clearances in female rats were much slower. In both sexes, the total body clearance and volume of distribution of HB were decreased by 20% during phototherapy of the jj but not the Jj rats; terminal plasma half-life was unchanged. In experiment 3 direct in vitro illumination of [14C]HB did not cause photodegradation of this compound, despite the presence of albumin with or without bilirubin.

Effects of phenobarbital on bilirubin metabolism and its response to phototherapy in the jaundiced Gunn rat.

Jaundiced Gunn rats, treated with phenobarbital (60 mg per kg i.p. for 7 to 10 days) showed 25 and 36% decreases in mean plasma bilirubin levels in two experiments (p less than 0.01). Kinetic studies with tracer 14C-bilirubin revealed that there was no change in bilirubin turnover or total pool size due to phenobarbital, but a 49% increase in the hepatic pool and a 27% decrease in the cutaneous pool of bilirubin. The increase in the hepatic pool accounted for over 90% of the bilirubin lost from the plasma. Such pretreatment with phenobarbital did not alter the decline in plasma bilirubin or total bilirubin pool due to subsequent phototherapy. Phenobarbital followed by phototherapy produced a significantly greater reduction in plasma bilirubin levels than either treatment alone. These studies demonstrate that phenobarbital does decrease plasma bilirubin in Gunn rats primarily by shifting the pigment to the liver, and suggests that combined treatment with phenobarbital and phototherapy might be of value in patients with congenital hyperbilirubinemia due to glucuronyl transferase deficiency.

Early changes in growth control of cells in culture induced by a chemical carcinogen.

The effects of the chemical carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) on 'early events' in cellular response to growth stimulation were studied in quiescent (confluent and serum-starved) baby hamster kidney C13 and secondary mouse embryo cell cultures. After a short (one hour) exposure to MNNG (0.015 micrograms/ml), we observed an increase (up to 31%) in basal uridine uptake rates and a significant decrease in the stimulatory effect of serum on the uridine uptake capacity of the quiescent cells. These effects of MNNG are reminiscent of the permanent changes in growth regulation-related properties which accompany cell transformation. The results of this study suggest that MNNG interacts with a cellular target(s) which controls the rate of uridine uptake and other growth-related responses to serum factors. The effects of MNNG on cell responsiveness to growth regulators do not require active DNA synthesis and are observed long before transformation is evident, suggesting that non-DNA targets may be involved in the process of initiation of chemical carcinogenesis.

Bilirubin binding and neonatal acidosis.

Plasma of neonates with severe metabolic acidosis secondary to fetal hypoxia bound less bilirubin than that of neonates without acidosis, as determined by Sephadex gel filtration. There was a significant correlation between the amount of bilirubin adsorbed by Sephadex and the base deficit. The method used ruled out any influence of plasma pH per se on binding. Our results suggest that organic anions that accumulate in the plasma of asphyxiated acidotic neonates may compete with bilirubin for binding sites on albumin.