The isolation and further characterization of the bilirubin tetrapyrroles in bile-containing human duodenal juice and dog gall-bladder bile

On how the conformation of biliverdins influences their reduction to bilirubins: a biological and molecular modeling study

The cyclic 2,18-bridged biliverdin (2) is excreted in rat bile without reduction to the corresponding bilirubin. Conformational analysis, employing an optimized Monte Carlo method and a mixed Monte Carlo/stochastic dynamics, reveals that biliverdin IXalpha (1) and the cyclic analogue 2 adopt 'lock washer' conformations, stabilized by the presence of intramolecular hydrogen bonds between N23...H22N and, to a lesser extent, between N23...H24N. Although 2 is very similar in overall shape to 1, the former adopts a 'locked lock washer' conformation unable to undergo fluctuations, thus possibly hampering a proper recognition by biliverdin reductase.

Exploring the conformation of bilirubins with natural and unnatural analogues: use of positional and bridged isomers of bilirubin IXalpha

Unlike bilirubin IXalpha (1), the isomers bilirubin IXdelta (2) and neobilirubin IXbeta (3) do not require conjugation with glucuronic acid in order to be excreted. A conformational analysis employing an optimized Monte Carlo method and a mixed Monte Carlo stochastic dynamics reveals that isomer 2 exhibits a structure more closed than the well known 'ridge-tile' conformation of 1. The change in the position of both propionic acid chains causes the loss of at least four hydrogen bonds. On the other hand, the change in the configuration of the distal dipyrrinone and the blockage of the lactamic nitrogen by the presence of a bridge in isomer 3 results in an open and more elongated structure, where the chance of hydrogen bond formation in this region is obliterated. The resulting molecular models for these compounds are consistent with 1H NM R, UV-vis, and TLC data.

Changes in the composition of bilirubin-IX isomers during human prenatal development

We analyzed the isomeric composition of bilirubin-IX in human fetal bile using HPLC. The approximate ratio of the bilirubin-IX isomers obtained from the fetal bile at 20 weeks of gestation was IX alpha, 6%; IX beta, 87%; IX gamma, 0.5%; and IX delta, 6%. From 15 to 22 weeks, bilirubin-IX beta was predominant and bilirubin-IX delta and bilirubin-IX alpha were also present in the bile as minor components. By 28 weeks, bilirubin-IX alpha constituted about 50% of the total bilirubin. There was a general correlation between fetal age and the proportion of bilirubin-IX alpha to bilirubin-IX beta in the bile and the small intestinal contents of fetuses. As development proceeded from mid-gestation to near term, the isomeric composition dramatically changed, with a decrease in the IX beta isomer and a subsequent increment of the IX alpha isomer. In contrast, the IX delta isomer changes little. Recently, we identified four forms of biliverdin reductase including two biliverdin-IX alpha reductases and two biliverdin-IX beta reductases in human liver cytosolic fractions [Yamaguchi, T., Komoda, Y. & Nakajima. H. (1994) J. Biol. Chem. 269, 24,343-24,348]. The proportion of the total activity of biliverdin-IX beta reductases to that of biliverdin-IX alpha reductases was considerably higher in the fetal, than in the adult liver.

Absolute concentration of urinary endogenous beta-glucuronidase determined by an ELISA method as a sensitive but non-specific indicator for active renal parenchymal damage

OBJECTIVE

To demonstrate that the absolute concentration, instead of the activity, of urinary endogenous beta-glucuronidase is a better indicator for active renal parenchymal damage.

MATERIALS AND METHODS

Urine samples were obtained from 143 adult patients comprising 60 control subjects with no evidence of urinary tract disease and 83 patients with serum creatinine > 1.5 mg/dL. The absolute concentration of urinary endogenous beta-glucuronidase was determined by an enzyme-linked immunosorbant assay (ELISA) method recently developed by us. The maximal velocity of the enzyme was determined by the enzyme kinetic method. The bacterial beta-glucuronidase and bilirubin in the urine were also detected.

RESULTS

The urinary beta-glucuronidase activity was affected by certain inhibitors (D-glucaro-1, 4-lactone), intrinsic substrates (conjugated bilirubin) and bacterial beta-glucuronidase present in the urine. Its concentration, determined by the ELISA method, was not interfered with by such factors. When those urine samples which were contaminated with bacteria and/or bilirubin were excluded, the beta-glucuronidase concentration (X ng/mumol creatinine) was significantly correlated with its maximal velocity (Y nmol/min/mumol creatinine): Y = 0.003 + 0.103X. Contrary to blood urea nitrogen, which increases with the increase in serum creatinine, the urinary beta-glucuronidase activity and its concentration reached a peak at a serum creatinine of 1.6-3.0 mg/dL and declined at higher serum creatinine levels. Episodic elevation of serum creatinine due to acute insult to the kidney or flare up of disease activity was often preceded by an increase in the level of urinary beta-glucuronidase.

CONCLUSION

The absolute concentration of the urinary beta-glucuronidase was not affected by several factors in the urine which interfere with its activity, and was a sensitive indicator for renal parenchymal damage, particularly in the early stage when the disease is active.

In vitro formation of glutathione conjugates of the dimethylester of bilirubin

Rat hepatic microsomes catalyzed the formation of two distinct glutathione conjugates of bilirubin dimethylester (DMB). The two conjugates were identical to those isolated from the bile of Gunn rats infused with DMB. The microsomal reaction was dependent on NADPH, oxygen and glutathione and was inhibited by nitrogen and the cytochrome P450 inhibitors metyrapone, 1-benzyl-imidazole, and alpha-naphthoflavone. Conjugate formation was inducible with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) but not phenobarbital pretreatment. The rate of formation of conjugates was not affected by washings of the microsomal pellet or by the presence of superoxide dismutase and/or catalase. Cation fast atom bombardment mass spectrometry (FAB/MS) of the conjugates indicated a molecular ion of 937 atomic mass units (amu). Fragmentation revealed a loss of 307 amu, consistent with glutathione, and a residual mass of 629 amu suggesting a hydroxylated derivative of DMB (612 amu). Cation FAB/MS/MS of conjugates formed in vitro under an atmosphere of oxygen-16 and oxygen-18 demonstrated the incorporation of molecular oxygen by a difference of 2 amu in the respective molecular ions. Our results suggest that DMB is oxidized by the cytochrome P450 IA gene family to an epoxide intermediate which is then subsequently conjugated with glutathione.

High-performance liquid chromatographic analysis of bile pigments as their native tetrapyrroles and as their dipyrrolic azosulfanilate derivatives

A reversed-phase high-performance liquid chromatographic (HPLC) analysis of bile pigments is described that provides baseline separation of the major bilirubin conjugates found in bile. The advantage of the technique is that the bile pigments can be analyzed directly as their native tetrapyrroles without prior solvent extractions or derivatization. The use of ammonium acetate in place of sodium salts permits preparative isolation and lyophilization of the pigments for mass spectroscopy. The derivatization of the pigments as their dipyrrolic azosulfanilates with subsequent HPLC analysis demonstrates baseline separation of the endo- and exovinyl azodipyrroles and allows identification of that half of the tetrapyrrole which contains the conjugate in the instances of monoglycosides.

High-performance liquid chromatographic separation of bilirubin conjugates: the effects of change in molarity and pH

A fast, sensitive high-performance liquid chromatographic method has been developed for the separation and quantitation of biliary bile pigments; this utilizes a C18 reversed-phase column with two solvents, a buffer and an organic solvent, which were changed in a linear gradient from a polar to a less polar combination. Nine glycosidic conjugates of bilirubin as well as unconjugated bilirubin and a suitable internal standard, unconjugated mesobilirubin IX alpha, were all separated to baseline by gradient elution; the species eluted in a polar to less polar fashion. Increasing the molarity of the solvent decreased the binding of non-glucuronide pigments to the column, with a decrease in their retention times, whereas for bilirubin monoglucuronide they increased. Decrease in pH, similarly, preferentially increased bilirubin monoglucuronide retention times.

High-performance liquid chromatographic separation of bilirubin conjugates

A fast sensitive method for the isolation and quantitation of biliary bile pigments by reverse-phase high-performance liquid chromatography has been developed. Nine conjugates of bilirubin as well as unconjugated bilirubin and an internal standard, unconjugated mesobilirubin IX alpha, were all separated to baseline by gradient elution. The following sequence of eluted compounds was chemically identified by separating their ethyl anthranilate derivatives by thin-layer chromatography and by their enzymatic formation with UDP-bilirubin transferase and cosubstrate: bilirubin diglucuronide, bilirubin monoglucuronide monoglucoside, bilirubin monoglucuronide monoxyloside, bilirubin monoglucuronide (C-8, C-12), bilirubin diglucoside, bilirubin monoglucoside monoxyloside, bilirubin dixyloside, bilirubin monoglucoside (C-8, C-12), and bilirubin monoxyloside. The use of the commercially available mesobilirubin IX alpha as an internal standard was found to facilitate quantitation of the bilirubin conjugates.

Chromatographic analysis and structure determination of biliverdins and bilirubins

Recent applications of thin-layer chromatographic (TLC) and high-performance liquid chromatographic (HPLC) procedures has revealed an unexpected wide variety of naturally occurring unconjugated and conjugated bilirubins. Biliverdins seems to occur only in unconjugated forms, mainly as the IX alpha isomer. Several synthetic biliverdins and bilirubins present interesting models for biochemical and metabolic studies. Owing to recent recognition of the astounding heterogeneity of natural bilirubins and to the various artifactual changes that bile pigments can undergo, considerable confusion has existed, and still exists, with regard to the nomenclature of the bile pigments and their derivatives. To set a background for further discussion, the present review starts with a brief discussion of nomenclature and of the various characteristic forms of lability of the bile pigments. TLC and HPLC procedures for preparation and analysis of unconjugated biliverdins and bilirubins and their methyl ester and sugar ester conjugates, as well as procedures for analysis of bilirubin-protein conjugates, are then discussed. Since, in view of the lability and pronounced heterogeneity of bile pigments, it is important to assess the composition and nature of chromatographically isolated pigments, the review is concluded by a brief evaluation of various structural tests.

Human beta-glucuronidase. Measurement of its activity in gallbladder bile devoid of intrinsic interference

Our purpose is to develop a standard method for preparing the bile for beta-glucuronidase determination by removal of bile acids and conjugated bilirubin which interfere with its activity. The bile acids and conjugated bilirubin in their purified solutions and in the diluted gallbladder biles could be extracted completely with cholestyramine in powder form or tetrahexylammonium chloride (THAC) in chloroform or ethyl acetate. The enzyme was, however, partially precipitated with cholestyramine and denatured by chloroform but not by ethyl acetate. A standard procedure, therefore, includes extraction of the diluted gallbladder bile with THAC in ethyl acetate, followed by determination of the maximal velocity (Vmax) of the enzyme by a kinetic method employing phenolphthalein glucuronide as the substrate. The average Vmax of beta-glucuronidase in the 20 normal gallbladder biles was 165 +/- 86 nmol/min/ml (mean +/- SD), a 23.5-fold increase over the activity before extraction. The measured activity represented the true activity of the enzyme in the bile for recovery of activity of the enzyme added to the bile was practically complete.

Microsomal specificity underlying the differing hepatic formation of bilirubin glucuronide and glucose conjugates by rat and dog

Bilirubin monoglucuronide monoglucoside diester is one of the principal bilirubin conjugates in dog bile (and a lesser conjugate, in human bile), and bilirubin diglucoside is an occasional trace conjugate in dog bile whereas, in contrast, neither is detectable in rat bile. In order to investigate, in comparative fashion, the factors underlying the formation of glucuronide and glucose-containing conjugates, hepatic microsomes were isolated by differential centrifugation from the livers of both normal mongrel dogs and Sprague-Dawley rats, and their formation of bilirubin conjugates examined, in the presence of varying levels of UDP-glucuronate and UDP-glucose. Bilirubin and its conjugates were extracted and separated by high-performance liquid chromatography; a new methodology was devised, which clearly separates bilirubin diglucoside from bilirubin monoglucuronide, as well as bilirubin diglucuronide, the mixed monoglucuronide monoglucoside conjugate and bilirubin monoglucoside. At bilirubin levels of 12.5 microM, in the presence of equal amounts of both UDP-glucuronate and UDP-glucose, dog microsomes formed substantial amounts of both bilirubin diglucuronide and the mixed monoglucuronide-monoglucoside conjugate, and minor amounts of bilirubin monoglucuronide and bilirubin diglucoside. Microsomes from rat liver, under similar conditions, formed only bilirubin diglucuronide and bilirubin monoglucuronide. When only UDP-glucose was present, dog microsomes formed predominantly diglucoside and rat, predominantly monoglucoside. The findings imply that it is not the availability of the UDP-glycoside but rather the preference of the microsomal enzymic system for the different glycosidic nucleotides which dictates the varieties of bilirubin conjugates ordinarily formed in these two species.

Transport of conjugated bilirubin and other organic anions in bile: relation to biliary lipid structures

Using gel-permeation chromatography, we studied associative relationships between conjugated bilirubin and various biliary lipid particle species, including lecithin/cholesterol vesicles, mixed-lipid micelles, and simple bile salt micelles. Five other organic anions were comparably studied: phenol red, Evans blue, sulfobromophthalein, rose bengal, and indocyanine green. For compounds of intermediate hydrophobicity, including conjugated bilirubin, the dominant association was with a bile salt/organic anion hybrid particle of dimensions larger than that of a simple pure bile salt micelle. Vesicular association was found to be dominant only for the most hydrophobic organic anions, indocyanine green and rose bengal; conversely, the most hydrophilic anion, phenol red, showed no vesicular association. Accordingly, a strong positive correlation (P less than 0.001) was found between percent vesicular association and degree of hydrophobicity of the organic anion. Alkaline conditions (eluant pH 9) decreased or prevented vesicular hydrophobic interaction with all anions. We conclude that two important particulate mechanisms for transport in bile of conjugated bilirubin and other water-soluble anions are bile salt/organic anion hybrid particles and vesicles. For most organic anions of intermediate hydrophobicity, including conjugated bilirubin, the bile salt/organic anion hybrid particle is the dominant transport vehicle.

The excretion pattern of biliverdin and bilirubin in bile of the small skate (Raja erinacea)

Bile pigment composition (biliverdin, bilirubin and their conjugates) was analyzed in stored gallbladder bile and newly synthesized hepatic bile from the small skate (Raja erinacea). During a five day period of captivity, gallbladder volume remained relatively constant while bilirubin and biliverdin content increased two to three fold. Biliverdin which accounted for 50% of the pigments did not increase as a percentage of tetrapyrroles during this period. The relative proportion of bilirubin and its conjugates also remained constant, averaging 65% for bilirubin monoglucuronide, 30% for bilirubin diglucuronide and 5% for unconjugated bilirubin as measured by HPLC methods. Intravenous administration of biliverdin resulted in significant increases in the biliary excretion of both biliverdin and all bilirubin tetrapyrroles. Insignificant quantities of 3H-biliverdin were detected in hepatic bile following the intravenous administration of 3H-bilirubin. These studies indicate that the small skate excreted both biliverdin and bilirubin conjugates in bile and that the biliverdin was not produced by in vitro oxidation of bilirubin or its metabolites.

Study of bilirubin metabolism by high-performance liquid chromatography: stability of bilirubin glucuronides

The stabilities of bilirubin (BR) glucuronide, monoglucuronide (BMG), and diglucuronide (BDG) were studied under various conditions by HPLC. In aqueous media, BMG showed a pronounced lability and was easily transformed into equimolar BDG and BR. It was proved by direct analysis of tetrapyrrole isomers that BDG and BR were formed from dipyrrole exchange of BMG molecules. All reducing agents examined (sodium ascorbate, cysteine, GSH, dithiothreitol, NADH, and NADPH) suppressed the transformation of BMG into BDG and BR. Bovine serum albumin and rat liver cytosol fractions also stabilized BMG strongly. BDG was fairly stable in aqueous media as compared with BMG. When BMG was incubated both with and without liver plasma membranes (N2 fraction) from Wistar rats, the formation rates of BDG and BR in both incubation mixtures were exactly the same. The composition of BDG and BR isomers was the same in both mixtures. Also, heat denaturation of the plasma membranes did not affect formation rates. Moreover, the reaction was completely inhibited by sodium ascorbate. These findings indicate that rat liver plasma membranes have no enzyme activity for BDG formation from BMG.

Reverse-phase h.p.l.c. separation, quantification and preparation of bilirubin and its conjugates from native bile. Quantitative analysis of the intact tetrapyrroles based on h.p.l.c. of their ethyl anthranilate azo derivatives

We describe a facile and sensitive reverse-phase h.p.l.c. method for analytical separation of biliary bile pigments and direct quantification of unconjugated bilirubin (UCB) and its monoglucuronide (BMG) and diglucuronide (BDG) conjugates in bile. The method can be 'scaled up' for preparative isolation of pure BDG and BMG from pigment-enriched biles. We employed an Altex ultrasphere ODS column in the preparative steps and a Waters mu-Bondapak C18 column in the separatory and analytical procedures. Bile pigments were eluted with ammonium acetate buffer, pH 4.5, and a 20 min linear gradient of 60-100% (v/v) methanol at a flow rate of 2.0 ml/min for the preparative separations and 1.0 ml/min for the analytical separations. Bile pigments were eluted in order of decreasing polarity (glucuronide greater than glucose greater than xylose conjugates greater than UCB) and were chemically identified by t.l.c. of their respective ethyl anthranilate azo derivatives. Quantification of UCB was carried out by using a standard curve relating a range of h.p.l.c. integrated peak areas to concentrations of pure crystalline UCB. A pure crystalline ethyl anthranilate azo derivative of UCB (AZO . UCB) was employed as a single h.p.l.c. reference standard for quantification of BMG and BDG. We demonstrate that: separation and quantification of biliary bile pigments are rapid (approximately 25 min); bile pigment concentrations ranging from 1-500 microM can be determined 'on line' by using 5 microliters of bile without sample pretreatment; bilirubin conjugates can be obtained preparatively in milligram quantities without degradation or contamination by other components of bile. H.p.l.c. analyses of a series of mammalian biles show that biliary UCB concentrations generally range from 1 to 17 microM. These values are considerably lower than those estimated previously by t.l.c. BMG is the predominant, if not exclusive, bilirubin conjugate in the biles of a number of rodents (guinea pig, hamster, mouse, prairie dog) that are experimental models of both pigment and cholesterol gallstone formation. Conjugated bilirubins in the biles of other animals (human, monkey, pony, cat, rat and dog) are chemically more diverse and include mono-, di- and mixed disconjugates of glucuronic acid, xylose and glucose in proportions that give distinct patterns for each species.

Radioassay of UDP-glucuronyltransferase-catalysed formation of bilirubin monoglucuronides and bilirubin diglucuronide in liver microsomes

A radioassay for specific determination of the rates of UDP-glucuronic acid-dependent conversion of bilirubin into the two isomeric (C-8, C-12) bilirubin monoglucuronides and bilirubin diglucuronide is described and illustrated by its application to rat liver microsomes. The method is based on measurement of the relative amounts of radiolabel in unesterified bilirubin and its mono- and di-esterified reaction products after incubation with [14C]bilirubin as substrate. This analysis is performed by the alkaline-methanolysis procedure, combined with one of two t.l.c. systems developed in order to enhance the sensitivity, accuracy and precision of the radioassay. Results for rates of total bilirubin glucuronide formation obtained with the new assay and the standard enzyme assay based on the ethyl anthranilate diazo-method were identical. However, the sensitivity of the latter technique is approx. 10-fold lower than that of the radioassay.

Methods for determination of conjugated bilirubin in rat faeces

Conjugated bilirubin was prepared from the faeces of germ-free (GF) rats by three different preparative methods. The bilirubin conjugate preparations were coupled with diazotized ethyl anthranilate and the formed ethyl anthranilate azopigments were quantified spectrophotometrically and separated by thin-layer chromatography (tlc). The most polar azopigment was purified by tlc and subjected to ammonolysis followed by tlc of the released saccaride. As a result of this procedure, only glucuronic acid was detected as the conjugating saccaride thus indicating that the most polar azopigment prepared from GF rat faeces was the delta ethyl anthranilate azopigment. Reference azopigments were prepared from GF rat small intestinal contents and subjected to separation by tlc. The azopigment pattern was very similar to the pattern obtained with the faecal azopigment preparations and a maximum of ten separated azopigment spots were detected. The findings indicated that, in addition to bilirubin glucuronides, other bilirubin conjugates with unknown structure are excreted with the faeces of GF rats. One of the preparative methods used for the preparation of conjugated bilirubin from GF rat faeces was tested on faeces from conventional (CONV) rats. From these preparations, no ethyl anthranilate azopigments were formed, thus indicating that faeces from CONV rats is devoid of conjugated bilirubin.

Analysis of bilirubins in biological fluids by extraction and thin-layer chromatography of the intact tetrapyrroles: application to bile of patients with Gilbert's syndrome, hemolysis, or cholelithiasis

A method was developed to extract quantitatively the bilirubins from bile, urine, serum, stool, and preparations from liver with a chloroform-ethanol mixture at pH 1.8 in the presence of ascorbic acid and NaCl. Extracted pigment was submitted to thin-layer chromatography, and the separated bilirubins were either immediately eluted and determined spectrophotometrically or individually converted to ethyl anthranilate azo derivatives for thin-layer chromatographic analysis of each isolated pigment band. Bilirubins in duodenal bile of eight healthy adults comprised 1.5 +/- 1.3% unconjugated bilirubin-IX alpha, 69 +/- 6% bilirubin diglucuronide, and 16 +/- 4% bilirubin monoglucuronides. Mixed diconjugates containing one glucuronosyl moiety and either one xylosyl or one glucosyl group amounted to 10 +/- 3%. Most samples (6 of 8) contained trace amounts (0.6 +/- 0.6%) of unconjugated bilirubin-IX beta, in agreement with nearly exclusive cleavage of heme at the alpha-meso position. The composition of the bilirubins in bile was normal in 6 patients with cholesterol gallstones, 4 with chronic hepatitis, and 3 with hemolysis. In duodenal bile of individuals with Gilbert's syndrome (n = 10), the concentration of bilirubin conjugates was comparable to that in healthy adults, but the proportion of bilirubin diglucuronides (52 +/- 8%) was decreased. The concentration of unconjugated bilirubin-IX alpha showed a fair positive correlation with that of bilirubin monoglucuronide and was increased in half of the patients with Gilbert's syndrome.

Bilirubin kinetics in intact rats and isolated perfused liver. Evidence for hepatic deconjugation of bilirubin glucuronides

Most previous compartmental models describing bilirubin transport and metabolism in the liver have been validated solely by analysis of the plasma disappearance of radiolabeled bilirubin in human subjects. We now have determined the transport kinetics of a bilirubin tracer pulse by analysis of plasma, liver, and bile radioactivity data from 30 intact rats. Plasma [3H]bilirubin disappearance was best described by the sum of three exponentials, and a six-compartment model, derived by simulation analysis, was necessary and adequate to describe all experimental data. Examination of the injected radiolabeled bilirubin by extraction with hexadecyltrimethylammonium bromide and thin-layer chromatography revealed that 6.6% (mean) of the original pigment had been degraded to labeled nonbilirubin derivatives during preparation of the tracer dose. This material exhibited a significantly longer half-life (mean 50.6 min) of the plasma terminal exponential than that of authentic radiobilirubin (20.6 min). In isolated perfused rat liver, the kinetics of [3H]bilirubin in perfusate and bile readily fitted the proposed model. Compatibility of the model with the data obtained, both in the isolated liver and in vivo, required that a fraction of bilirubin conjugated in the liver be deconjugated and returned to the plasma. Deconjugation of bilirubin glucuronides was evaluated directly by infusion of bilirubin monoglucuronides, containing 14C in the glucuronosyl group, into rats with an external bile fistula. Since metabolic degradation of hydrolyzed 14C-labeled glucuronic acid yields 14CO2, this was measured in expired air. Whereas 86% of the administered labeled pigment was recovered in bile, 7% of the label appeared in 14CO2. These findings directly validate a portion of the proposed kinetic model and suggest that hepatic deconjugation of a small fraction of bilirubin glucuronides is a physiological event. Deconjugation may also account, at least in part, for the presence of increased concentrations of unconjugated bilirubin in the plasma of patients with cholestasis.

beta-Glucuronidase-resistant bilirubin glucuronide isomers in cholestatic liver disease--determination of bilirubin metabolites in serum by means of high-pressure liquid chromatography

"Direct reacting bilirubin" in serum of patients with cholestatic liver disease and in serum of bile duct-ligated rats consists of a complex mixture of bilirubin metabolites. These metabolites were studied by means of high-pressure liquid chromatography. Bilirubin glucuronides in normal bile are beta-glycosidic 1-O-acyl conjugates which are completely hydrolyzed on incubation with beta-glucuronidase. Cholestatic serum contains glucuronide and non-glucuronide bilirubin metabolites. The glucuronides were only partially hydrolyzable with beta-glucuronidase. Compernolle et al. [11] showed that the 1-O-acyl bond of bilirubin glucuronides is labile and prone to migrate from the C1 position at the glucuronosyl residue to positions C2, C3 and C4. The isomerisation products are non-beta-glycosidic, beta-glucuronidase-resistant conjugates. The main beta-glucuronidase-resistant conjugates in cholestatic serum were characterized as: non-beta-glycosidic bilirubin monoglucuronide, non-beta-glycosidic diglucuronide and a diglucuronide isomer with beta-glycosidic and non-beta-glycosidic glucuronosyl groups. Moreover, a substantial amount of bilirubin monoglucoside monoglucuronide was detected in cholestatic human serum.

Synthesis of biliverdin and bilirubin 1-O-acyl-beta-D-glucopyranuronic acids

Biliverdin and bilirubin mono- and di-beta-glucuronides were prepared by nucleophilic substitution of the 1-O-mesyl derivative of alpha-ethoxyethyl-protected glucuronic acid (compound II) with the tetrabutylammonium salts of biliverdin and bilirubin. Removal of the acetal-protecting groups by mild acid treatment yielded biliverdin glucuronides, which were reduced to bilirubin glucuronides. Depending on reaction conditions the pure beta-anomers or mixtures highly enriched in the beta-anomers were obtained. The biliverdin and bilirubin glucuronides were identical with pigments derived from bile. They were characterized as the IX alpha isomers and the beta-anomers by alkaline hydrolysis, n.m.r. spectroscopy, hydrolysis with beta-glucuronidase and conversion into dipyrrolic azopigments. Model reactions of the 1-O-mesylate (II) with other nucleophiles also were performed, i.e. the acetate anion and various alcohols.

Blue light and bilirubin excretion

Blue light converts bilirubin in the skin of jaundiced rats to metastable geometric isomers that are transported in blood and excreted in bile. The same reaction probably occurs in jaundiced babies exposed to light, particularly during treatment with phototherapy. Excretion of unisomerized bilirubin is prevented by intramolecular hydrogen bonding, and the pigment has to be metabolized to more polar derivatives to be excreted efficiently.

Analysis of bilirubin and bilirubin mono- and di-conjugates. Determination of their relative amounts in biological samples

1. A novel method for determination of the relative amounts of unconjugated bilirubin and sugar mono- and di-conjugates of bilirubin in biological samples, including serum, is described and illustrated by its application to the analysis of bilinoids in rat bile. 2. The method is based on specific conversion of the carbohydrate conjugates of bilirubin into the corresponding mono- or di-methyl esters by base-catalysed transesterification in methanol. Under the selected reaction conditions, unconjugated biliru-in remains intact and no dipyrrole exchange in the bilinoids is detectable; transesterification of bilirubin mono- or di-glucuronide is virtually complete (approx. 99%), and sponification is negligible (less than 1%); recovery of the pigments is approx. 95%. 3. The reaction products bilirubin and its methyl esters are separated by t.l.c. and determined spectrophotometrically; the two isomeric bilirubin-IX alpha monomethyl esters are separated and therefore can be determined individually. 4. Reference bilirubin mono- and di-methyl esters have been synthesized and characterized, and the two isomers of bilirubin-IX alpha monomethyl ester and bilirubin dimethyl ester were obtained individually, in crystalline form. 5. With this new method, virtually all bilinoids (over 99%) in normal rat bile have been found to be conjugated, with diconjugates (71%) predominating. A significantly increased proportion of monoconjugates is present in bile collected from heterozygous Gunn rats or from normal rats that were refused with large amounts of bilirubin.

An accurate and sensitive analysis by high-pressure liquid chromatography of conjugated and unconjugated bilirubin IX-alpha in various biological fluids

An accurate and sensitive method was developed for the complete separation of the native tetrapyrroles, such as bilirubin and its mono- and di-conjugates of glucuronic acid, glucose and xylose, by ion-pair reversed-phase high-pressure liquid chromatography. The application of this method was demonstrated by the analysis of bile pigments in human bile and urine, and the method also makes it possible to estimate very low UDP-glucuronyltransferase activity, such as is found in the human foetal and neonatal liver.

Self-association of unconjugated bilirubin-IX alpha in aqueous solution at pH 10.0 and physical-chemical interactions with bile salt monomers and micelles

Spectrophotometric measurements of bilirubin-IX alpha in water and in aqueous/organic solvent mixtures at pH 10.0 as a function of bilirubin-IX alpha concentration (approx. 0.6--400 microM) are consistent with the formation of dimers (KD - 1.5 microM) in dilute (less than 10 microM) aqueous solution and further self-aggregation to multimers at higher concentrations. Added urea (to 10M) and increases in temperature (to 62 degrees C) obliterate the dimer-multimer transition at 10 microM, but added NaCl (to 0.30 M) promotes strong aggregation of dimers over a narrow concentration range, suggesting a 'micellization' phenomenon. Concentrations of dioxan or ethanol greater than 60% (v/v) in water were required to obtain the absorption spectrum of bilirubin-IX alpha monomers, suggesting that both hydrophobic and electrostatic (pi-orbital) interactions are involved in stabilizing the dimeric state in water. Micellar concentrations of sodium dodecyl sulphate induced spectrophotometric shifts in the dimer absorption spectrum of bilirubin-IX alpha consistent with progressive partitioning of bilirubin-IX alpha monomers into a relatively non-polar region of the micelles and allowed a deduction of the apparent critical micellar concentration that closely approximated the literature values. The pattern of bilirubin IX alpha association with bile salts is complex, since the absorption spectrum shifts hypsochromically below and bathochromically above the critical micellar concentration of the bile salts. Consistent with these observations, bilirubin IX alpha appears to bind to the polar face of bile salt monomers and to the polar perimeter of small bile salt micelles. At higher bile salt concentrations some-bilirubin-IX alpha monomers partition into the hydrophobic interior of the bile salt micelles. Our results suggest that under physiological conditions the natural conjugates of bilirubin-IX alpha may exhibit similar physical chemical properties in bile, in that dimers, highly aggregated multimers and bile salt-associated monomers may co-exist.

Bilirubin diglucuronide synthesis by a UDP-glucuronic acid-dependent enzyme system in rat liver microsomes

Incubation of rat liver homogenate or microsomal preparations with bilirubin or bilirubin monoglucuronide with (BMG) resulted in formation of bilirubin diglucuronide (BDG). Both synthesis of BMG and its conversion to BDG were critically dependent on the presence of UDP-glucuronic acid. Pretreatment of the animals with phenobarbital stimulated both reactions. When 33 microM bilirubin was incubated with microsomal preparations from phenobarbital-treated rats, 80-90% of the substrate was converted to bilirubin glucuronides; the reaction products consisted of almost equal amounts of BMG and BDG. When phenobarbital pretreatment was omitted or when the substrate concentration was increased to 164 microM bilirubin, proportionally more BMG and less BDG were formed. Homogenate and microsomes from homozygous Gunn rats neither synthesized BMG nor converted BMG to BDG. These findings in vitro suggest an explanation for the observations in vivo that, in conditions of excess bilirubin load or of genetically decreased bilirubin UDP glucuronosyltransferase (EC 2.4.1.17) activity, proportionally more BMG and less BDG are excreted in bile.

Comparative aspects of bile pigment formation and excretion

Breakdown of haem which is of key importance in most organisms, involves oxidative CO-evolving cleavage of the macrocyclic ring with formation of biliverdin-IX. In two major pathways established so far formation of biliverdin-IX alpha is followed by (a) biliary secretion or (b) reduction to bilirubin-IX alpha, formation of more hydrophilic derivatives (usually glycosidic conjugates) and biliary secretion. The scattered comparative information available indicates marked species variation with regard to the methin-bridge carbon atom removed from haem and the metabolic site of cleavage, the nature of bilirubin conjugates and the developmental sequence of maturation of enzyme activities and transport proteins involved in the chain of events leading from breakdown of haem to the excretion of the final end products.

Structure revision of disaccharidic conjugates of bilirubin-IX alpha in human bile and identification of phenylazo derivatives B4, B5, and B6 as 2-, 3- and 4-O-acylglucuronides

Aniline azopigments B4, B5 and B6, derived from conjugates of bilirubin-IX alpha in human bile, and previously characterized as disaccharidic esters [Kuenzle (1970) Biochem. J. 119, 387-394 and 411-435], were analysed by using t.l.c. and mass spectrometry. The compounds were identified as partially separated mixtures of 2-, 3- and 4-O-acylglucuronide positional isomers. The 1-O-acylglucuronide was not detected in the mixtures and was the only compound hydrolysed with beta-glucuronidase. Further scrutiny of structural assignments made by Kuenzle [(1970) Biochem. J. 119, 411-435] led to identification of the lactone and hexuronic acid derivatives that were obtained from azopigment B5 along with glucuronolactone and glucuronic acid. A branched-chain structure, i.e. 3-C-hydroxy-methyl-D-riburonic acid, was assigned previously, but the derivatives have now been identified as various incompletely silylated forms of glucuronolactone and glucuronic acid. Several trimethylsilyl derivatives glucuronolactone were isolated and characterized by n.m.r. and mass spectrometry.

Rearranged glucuronic acid conjugates of bilirubin-IX alpha in post-obstructive bile. Structure elucidation of azopigments beta and gamma as ethyl anthranilate N-glycosides derived from 2-, 3- and 4-o-acyl glucuronides

Azopigment analysis was performed on conjugates of bilirubin-IXalpha in bile of man and rats obtained after obstruction of the bile duct or in bile incubated under N2. The azopigments beta and gamma, formed by applying a pH 2.7 diazonium reagent containing an excess of ethyl anthranilate, correspond to rearranged ethyl athranilate N-glucuronides having the azodipyrrole acyl group on positions 2, 3 and 4 of the sugar. These assignments were verified, first by conversion of the structurally known 2-, 3- and 4-O-acyl glucuronide azopigments, unsubstituted at C-1, into ethyl anthranilate N-glucuronide reference compounds, and second, by mass spectrometry of trimethylsilyl ether methyl ester derivatives of unknown and reference compounds. The C-1 ethyl anthranilate group of the N-glucuronides triggers characteristics fragmentation reactions of the carbohydrate moiety revealing the position of the azodipyrrole O-acyl group.

Glucuronic acid conjugates of bilirubin-IXalpha in normal bile compared with post-obstructive bile. Transformation of the 1-O-acylglucuronide into 2-, 3-, and 4-O-acylglucuronides

Structures have been determined for bilirubin-IXalpha conjugates in freshly collected bile of normal rats, dogs and man and in post-obstructive bile of man and rats. The originally secreted conjugate has been characterized as azopigment (I), i.e. a 1-O-acyl-beta-d-glucopyranuronic acid glycoside. Conversion of the acetylated methyl ester of azopigment (I) into methyl 2,3,4-tri-O-acetyl-1-bromo-1-deoxy-beta-d-glucopyranuronate (V) indicates the pyranose ring structure for the carbohydrate and a C-1 attachment for the bilirubin-IXalpha acyl group. Alternative procedures for deconjugation of azopigment (I) and its derivatives are also described. In post-obstructive bile, the 1-O-acylglucuronide is converted into 2-, 3- and 4-O-acylglucuronides via sequential intramolecular migrations of the bilirubin acyl group. The following approach was utilized. (1) The tetrapyrrole conjugates were cleaved to dipyrrolic aniline and ethyl anthranilate azopigments, and the azopigments were separated as the acids or methyl esters. (2) The isomeric methyl esters were characterized by mass spectral analysis of the acetates and silyl ethers. (3) The free glycosidic function was demonstrated by 1-oxime and 1-methoxime derivative formation. (4) The position of the dipyrrolic O-acyl group was determined for the methyl esters by protecting the free hydroxyl groups of the glucuronic acid moieties as the acetals formed with ethyl vinyl ether and by further conversion of the carbohydrates into partially methylated alditol acetates. These were analysed by using g.l.c.-mass spectrometry. The relevance of the present results with regard to previous reports on disaccharidic conjugates is discussed. Details of procedures for the formation of chemical derivatives for g.l.c. and mass spectrometry have been deposited as Supplementary Publication SUP 50081 (15 pages) at the British Library Lending Division, Boston Spa, West Yorkshire LS23 7BQ, U.K., from whom copies can be obtained on the terms indicated in Biochem. J. (1978), 169, 5.