Origin of mammalian biliprotein and rearrangement of bilirubin glucuronides in vivo in the rat.

In hepatobiliary disease bilirubin becomes bound covalently to serum albumin, producing a nondissociable bile pigment-protein complex (biliprotein). To elucidate the mechanism of biliprotein formation we studied the bile pigment composition of blood from animals with experimental cholestasis and carried out comparative studies on the rate of biliprotein formation in vivo and in vitro during incubation of bilirubin glucuronides with albumin. Bile duct ligation in the rat and guinea pig led to rapid accumulation in the circulation of bilirubin, heterogeneous bilirubin esters of glucuronic acid, and a biliprotein that migrated along with albumin on high performance liquid chromatography. When the obstruction was removed, biliprotein remained longer in the circulation than did the other bile pigment species. Biliprotein and heterogeneous bilirubin esters of glucuronic acid were not formed in bile duct-ligated homozygous Gunn rats but they were formed when bilirubin glucuronides were incubated with Sprague-Dawley rat serum or human serum albumin at 37 degrees C in vitro. Bilirubin glucuronide rearrangement in vitro was accompanied by nonenzymic hydrolysis. We conclude that the formation of biliprotein in vivo is probably nonenzymic and suggest that mammalian biliprotein is formed by acyl migration of bilirubin from a bilirubin-glucuronic acid ester to a nucleophilic site on albumin.

Ion-pair high-performance liquid chromatographic procedure for the quantitative analysis of theophylline in serum samples.

An ion-pair chromatographic system is described for the separation of theophylline and related xanthines from serum samples. The mobile phase consisted of 0.02 M tetrabutylammonium ion and 0.015 M Tris buffer in water-acetonitrile-methanol (93:3.5:3.5, v/v/v) at a precisely controlled pH (7.50 +/- 0.02, adjusted with hydrochloric acid). The flow-rate was 1.2 ml/min through a 15 cm X 4.6 mm I.D., 5-micron reversed-phase column (Ultrasphere C18 ion pair). Xanthines were extracted from serum (100 microliter) with 1 ml of acidified chloroform-isopropanol (95:5, v/v). After reconstitution in 200 microliter of mobile phase, the extracted xanthines, including theophylline, caffeine, theobromine and 1.7-dimethylxanthine, were baseline-resolved in less than 15 min. The method correlates well with a common clinical immunoassay for theophylline (EMIT Syva, r2 = 0.999) and yields excellent recovery and precision (98-101% and better than 2% at therapeutic levels, respectively). In addition, the use of the ion-pair chromatography mode eliminates many of the interferences noted in the published literature for the common reversed-phase separations of theophylline.

Separation of bilirubin species in serum and bile by high-performance reversed-phase liquid chromatography.

A high-performance, reversed-phase liquid chromatographic (HPLC) procedure has been developed for the separation of at least three major bilirubin fractions in bile and four fractions in human serum. This procedure was unlike most others, in that serum was not totally deproteinized prior to injection onto the HPLC column; instead, serum was treated with an excess of sodium sulfate solution to precipitate primarily proteins larger than albumin. Injection of the filtered and diluted supernatant onto a reversed-phase column then resulted in the separation of the bilirubin species in a 24-min gradient elution run. Both the initial aqueous acidic mobile phase and the final isopropyl alcohol-based mobile phase contained 5% methoxyethanol (v/v) to facilitate elution of albumin still present in the treated sample. Bilirubin species eluting from the column were detected by absorbance at 450 nm. Results of a number of chromatographic separations of pathological sera indicated a wide variation in the relative proportions of the four bilirubin fractions observed. A correlation of the sum of the areas of the bilirubin peaks observed by HPLC was found with the total bilirubin value obtained by a standard reference procedure.

Degradation of ethylenediaminetetraacetic acid by mictobial populations from an aerated lagoon.

The ferric chelate of ethylenediaminetetraacetic acid (EDTA) was biologically degraded by a mixed population of microorganisms present in an aerated lagoon receiving this chemical in its feed. As determined radiorespirometrically, 28% of the acetate-2-C and 30% of the ethylene position of the ammonium ferric chelate of [14C]EDTA was recovered as 14CO2 after 5 days. In a separate experiment using gas liquid chromatography and the sodium ferric chelate, as much as 89% disappearance of EDTA (0.1% wt/vol) was observed during a similar time period. Optimum 14CO2 evolution was observed at a pH value between 7 and 8 and at room temperature. Degradation of NH4Fe-[2-14C]EDTA was stimulated by the addition of either unlabeled NaFe-EDTA, nitrilotriacetic acid or ethylenediamine, and inhibited by the addition of a variety of different sugars and amino acids. Consistent with the biological nature of this degradation, little or no 14CO2 evolution was observed after heat treatment of the microorganisms at 100 C for 10 min, or after the addition of antibiotics to the incubation mixtures. Gas-liquid chromatography and mass spectral analyses were performed to demonstrate EDTA disappearance and to identify possible intermediates of EDTA degradation.

Quantitative liquid-chromatographic estimation of bilirubin species in pathological serum.

Earlier we described a "high-performance" liquid-chromatographic procedure for separating the four distinct fractions of bilirubin (unconjugated, monoconjugated, diconjugated, and protein-bound) in pathological human serum (J. Chromatogr. 226: 391-402, 1981). We have modified the prechromatography precipitation of the serum globulins required in that method and have measured the bilirubin content of the precipitate spectrophotometrically. On average, the precipitate contained less than 10% of the total bilirubin in the serum samples. Adding the value obtained for the precipitate to that obtained by chromatography for the individual bilirubin fractions gave an estimate of the concentration of the total bilirubin in the sample. For 357 samples from 132 patients, this total value correlated well with that obtained by the Jendrassik-Gróf diazo procedure (slope = 1.00; r = 0.995, linear least-squares fit). The CV for the total and fractional bilirubin measurements was, on average, less than or equal to 5% for pathological sera. Serum sampled at different times from the same patient showed significant changes in the distribution of bilirubin among the four fractions.

Isolation and preliminary characterization of a fraction of bilirubin in serum that is firmly bound to protein.

We have isolated from pathological sera a bilirubin fraction (delta) that is very tightly, if not covalently, bound to protein, most likely albumin. This delta fraction absorbed at a lambda max of 433 nm in the visible spectrum, between the lambda max of unconjugated (alpha) and that of conjugated (Bc) bilirubin when measured in solutions containing albumin. However, unlike the other bilirubin species, this fraction could not be separated from the proteins in serum by exhaustive ultrafiltration in the presence of caffeine/benzoate solution. In the Jendrassik-Grof diazo procedure for bilirubin analysis, the delta fraction gave a large direct reaction (76-89% of the total reaction). Yet, when relatively hydrophobic azo dyes were formed by reaction of the delta fraction with the diazonium salt of dichloroaniline, only 50% of the dyes were extractable from aqueous solution. On chromatography the rest remained associated with protein. Of the extractable dye, more than 70% was accounted for by two liquid-chromatographic peaks with retentions identical with those of azo dyes formed from unconjugated bilirubin. This delta fraction was not appreciably separated from protein by treatment with strong acid or base, or by prolonged digestion with various enzymes. Finally, in a highly denaturing solvent (urea/mercaptoethanol), this fraction was not dialyzable through a membrane with a 12 000-dalton cutoff.

Degradation of the Ferric Chelate of EDTA by a Pure Culture of an Agrobacterium sp.

A pure culture of an Agrobacterium sp. (deposited as ATCC 55002) that mineralizes the ferric chelate of EDTA (ferric-EDTA) was isolated by selective enrichment from a treatment facility receiving industrial waste containing ferric-EDTA. The isolate grew on ferric-EDTA as the sole carbon source at concentrations exceeding 100 mM. As the degradation proceeded, carbon dioxide, ammonia, and an unidentified metabolite(s) were produced; the pH increased, and iron was precipitated from solution. The maximum rate of degradation observed with sodium ferric-EDTA as the substrate was 24 mM/day. At a substrate concentration of 35 mM, 90% of the substrate was degraded in 3 days and 70% of the associated chemical oxygen demand was removed from solution. Less than 15% of the carbon initially present was incorporated into the cell mass. Significant growth of this strain was not observed with uncomplexed EDTA as the sole carbon source at comparable concentrations; however, the ferric chelate of propylenediaminetetraacetic acid (ferric-PDTA) did support growth.

The clinical importance of a protein-bound fraction of serum bilirubin in patients with hyperbilirubinemia.

A directly reacting fraction of bilirubin that is probably covalently bound to albumin (albumin-bound bilirubin) has recently been described. To determine its clinical importance we used a new high-performance liquid-chromatography technique to measure it in the serum of 200 patients with hyperbilirubinemia from various causes. Albumin-bound bilirubin was an important fraction (8 to 90 per cent) of total bilirubin in patients with hepatocellular and cholestatic jaundice as well as in patients with the Dubin-Johnson syndrome. It was not detected in normal volunteers, neonates with physiologic jaundice, or patients with Gilbert's disease or hemolysis. Thus, albumin-bound bilirubin appears in serum when hepatic excretion of conjugated bilirubin is impaired. It becomes a larger component of serum bilirubin as jaundice subsides, delaying resolution of this disorder and causing bilirubin to persist in plasma after it has disappeared from the urine.

Estimation of unconjugated, conjugated, and "delta" bilirubin fractions in serum by use of two coated thin films.

We used two coated thin films to measure the concentrations of unconjugated, conjugated, and total bilirubin as well as bilirubin covalently bound to albumin ("delta" bilirubin) in more than 400 serum samples. We measured the unconjugated and conjugated species by determining their reflection densities at two wavelengths (400 and 460 nm) on a coating designed for the enhanced spectral measurement of bilirubin but which does not register the delta form. Total bilirubin was measured by use of a diazo-based thin film (Clin Chem 29: 37-41, 1983). We estimated the concentration of delta bilirubin by subtracting the sum of unconjugated and conjugated bilirubin from the concentration of total bilirubin. All measurements agree well with those by comparative methods, as shown by linear regression. Slopes ranged from 0.92 to 1.02, correlation coefficients from 0.935 and 0.998. Linear combinations of these values can also be used to compute other results; e.g., the sum of conjugated and delta bilirubin can be considered to be an estimate of "direct"-reacting bilirubin.