Studies on the Determination of Bile Pigments I. Standard of Purity for Bilirubin

Size: px

Start display at page:

Download "Studies on the Determination of Bile Pigments I. Standard of Purity for Bilirubin"

Miranda Flynn
5 years ago
Views:

1 Studies on the Determination of Bile Pigments I. Standard of Purity for Bilirubin Richard J. Henry, S. L. Jacobs, and Neil Chiamori HE ACCURACY of the standardization for determination of bilirubin is at present open to question. This is a result of the fact that commercially available bffirubin preparations vary considerably in their chromogenicity in the azobilirubin reaction (1) and because a reliable standard of purity has never been established. Furthermore, difficulty in preparing biirubin solutions has led many workers to the use of artificial standards. The investigation reported in this article is concerned with evaluation of purity of a number of commercial bifirubin preparations, preparation of pure biirubin, and study of the purity of the purified product. EXPERIMENTAL AND RESULTS DETERMINATION OF ABSORPTIVITIES OF COMMERCIAL PRODUCTS Commercial preparations of bilirubin were dissolved in chloroform, A. R. grade, by shaking in the dark at room temperature. Stock solutions of 3.0 mg./50 ml. were prepared with the use of a microbalance. Dilutions of 1:10 were made, and absorbancy at 453 m, the absorption maximum, was read in a Beckman DU spectrophotometer against a blank of chloroform. Absorptivities (molar extinction coefficients) based on a molecular weight of 584 are listed in Table 1. It has been stated (2) that the color of biirubin is no criterion of purity, but rather that it represents the physical state of the preparation. Nevertheless, the only two commercial preparations that were orange, those of Pfanstiehl and Matheson, Coleman & Bell, had ab- Prom the Bio-Science Research Foundation, Los Angeles. Received for publication July 8,

530 HENRY ft AL. Clinical Chemistry sorptivities in chloroform of 58,900 and 58,600, respectively. All other commercial preparations were brown and had absorptivities of 56,800 or less.

2 530 HENRY ft AL. Clinical Chemistry sorptivities in chloroform of 58,900 and 58,600, respectively. All other commercial preparations were brown and had absorptivities of 56,800 or less. The Pfanstiehl and Matheson, Coleman & Bell products were completely soluble in chloroform. All other commercial products contained some insoluble brown material. The absorptivities of azobilirubin formed from the commercial bilirubins were determined by the method of Malloy and Evelyn (3) and are presented in Table 1. The bilirubins were dissolved in 0.2% Table 1. DATA ON COMMERCIAL AND RECRYSTALLIZED BILIRUBINS Absorptivity of Ratio, Absorptivity at azobizirubin at absorp. bii. in CHC12 453,n,z in 545 mjs Prodwct nwi8ture CECIS (MaUoy-Evelyn) ab8orp. azobii. Eastman Kodakt ,600 53, purified 1X 60,700 63, Eastman Kodak (Lot 22) 56,800 59, Pfanstiehl (Lot 4612A) 58,900 58, purified 1X 60,700 64, purified 2X 60,300 64, Glogau (Lot 70603) 55,600 55, purified 1X 59,100 63, purified 2X 59,600 64, Hoffmann-La Roche (Lot ,500 59, B511067) Harlecol ,400 60, Mann (Lot B1468) ,800 59, Matheson, Coleman & Bell ,600 63, (Lot 36349) *Calculated on basis of an equivalent amount of bilirubin; not truly the absorptivity of the color product. Corrected for decrease of 3 per cent in volume when CH3OH is added. Not corrected for moisture. tno specific lot numbers for these products. Na2CO3 in the dark at room temperature and were used within 1 hr. of preparation. The reaction mixtures were composed of 1 ml. of diazo reagent and 4 ml. of water, to which was added a mixture of 4.9 ml. of methanol and 0.1 ml. of the bilirubin solution containing 20 g. of the product. Readings were made at 40 mm. at 545 m/l, the absorption peak, against a water blank. Color reached a stable maximum by 30 mm. The values given in Table 1 are calculated on the basis of an equivalent amount of bilirubin, and hence the values are not true absorptivities of the colored products. The values are corrected for the 3 per cent decrease in volume occurring when equal volumes of methanol and aqueous solution are mixed. Impure prod-

Vol. 6, No. 6, 1960 DETERMINATION OF BILE PIGMENTS 531 ucts were more easily soluble n Na2CO3 than in CHC13, although a small amount of insoluble material always remained.

3 Vol. 6, No. 6, 1960 DETERMINATION OF BILE PIGMENTS 531 ucts were more easily soluble n Na2CO3 than in CHC13, although a small amount of insoluble material always remained. Preparations of high purity were completely and rapidly soluble in 0.2% Na2CO3. Moisture contents of several of the commercial products are also given in Table 1. These were determined from losses in weight after drying over P205 in a vacuum desiccator. For practical purposes the moisture content may be ignored. Absorptivities for the commercial products are uncorrected for moisture. EFFECT OF CHLOROFORM ON AZOBILIRUBIN COLOR Since bilirubin is practically insoluble in methanol, it is customary in standardizing the Malloy-Evelyn technic to dissolve bilirubin in chloroform and then to employ a dilution in methanol for the standard. Bilirubin does not precipitate out of the dilutions ordinarily employed. If one employs a standard equivalent to 5 mg. of biirubin per 100 ml. of serum in the Malloy-Evelyn test (3), the final volume of the test (9.7 ml.) contains 0.05 ml. chloroform. Figure 1 shows the effect of chloroform on intensity of the azo color in standards of the Malloy-Evelyn technic as they were run (final concentrations of methanol, 50%). Points are averages of seven experiments and represent the percentage of color formed with zero concentration of chloroform. Each standard contained 20 g. of bilirubin added in 0.1 ml. of a 0.2% Na9CO3 solution. This amount of Na2CO3 had no measurable effect on the final ph of the test. Even 0.05 ml. of chloroform significantly inhibited the color intensity. For this reason, S ro ndd 90 - Malloy-Evelyn technic. a5 80 I I I I ML. CHCI3

4 532 HENRY ft AL. Clinical Chemistry aqueous carbonate solutions of bilirubin were used in this study of azobilirubin absorptivities. PURIFICATION OF COMMERCIAL BILIRUBIN PRODUCTS In patented procedures (4-6) an organic solvent extract of bilirubin is treated with a weak base to remove strongly acidic impurities; the bilirubin is extracted with Na2CO3, forming the sodium salt; ethanol is added, and the bilirubin is precipitated by acidification. The procedure for purification employed in this study was as follows: Two hundred mffligrams of bilirubin were dissolved in 400 ml. of chloroform by shaking for about 30 mm. Insoluble material was removed by filtration. The solution was washed twice with 40-mi. portions of 5% sodium bicarbonate. The chloroform phase was ifitered, and the biirubin was extracted twice with 40-mi. portions of 5% sodium carbonate containing a pinch of sodium hydrosulfite. The combined aqueous phase and emulsion that sometimes appeared at the interface of the solvent layers was washed with 100 ml. of chloroform, about 50 ml. of water were added to aid in separation of the phases, and the aqueous portion was ifitered. The bilirubin was then extracted with 100 ml. of chloroform containing 5 ml. of glacial acetic acid. The aqueous phase was extracted with an additional 40 ml. of chloroform, and the solvent extracts were combined and filtered. After drying over anhydrous Na2SO4, the solution was evaporated with the aid of a stream of nitrogen at less than 40#{176} to a volume of 5-10 ml. Much of the bifirubin crystallized during this process. About 1.5 volumes of ether were added, and, after standing in the refrigerator overnight, the bilirubin was collected by centrifugation and the product washed twice with 5-mi. portions of ether. The purified biirubin was dried overnight over anhydrous CaC12 in a vacuum desiccator. An average recovery of 50 per cent was obtained with this method. Values for the absorptivities of the preparations are given in Table 1. A second purification of commercial products appeared to yield no significant increase in absorptivities over those obtained in the first purification. All purified preparations were completely and rapidly soluble in CHC13 and 0.2% Na2CO3. The mean absorptivity of the purified products in chloroform was 60,100 (range 59,100-60,700). The mean absorptivity of the azobilirubin was 64,100 (range 63,500-64,600). The ratio, absorptivity of bilirubin in CHC13/absorptivity of azo-

Vol. 6, No. 6, 1960 DETERMINATION OF BILE PIGMENTS 533 biirubin, varied from 0.87 to 1.01 (mean of 0.945) for the unpurified commercial products and from 0.93 to 0.96 (mean of 0.

5 Vol. 6, No. 6, 1960 DETERMINATION OF BILE PIGMENTS 533 biirubin, varied from 0.87 to 1.01 (mean of 0.945) for the unpurified commercial products and from 0.93 to 0.96 (mean of 0.94) for the purified preparations. TEST OF PURITY Purified preparations exhibited no melting point; this conilrms a previous report (2). No differences were detectable in the shapes of ultraviolet absorption curves of solutions of pure and impure bilirubins in 0.2% sodium carbonate. Infrared spectra were obtained on commercial (llogau and Pfanstiehl twice purified biirubin preparations, using a Model 13, Perkin-Elmer spectrophotometer. Figure 2 is the spectrum obtained on 0.5 mg. of purified bilirubin incorporated in a KBr wafer. This spectrum differed in no significant way from that obtained on the unpurified Glogau bilirubin. Saturated solutions of biirubin in CHC13 and Cd4 yielded none but solvent peaks. Craven et al. (7) published a curve obtained with a CC14 solution of biirubin. The cause of this discrepancy is not clear. In washing chloroform solutions of impure bilirubins with 5% sodium bicarbonate, it was noted that brown-colored material was extracted. Some of the products were subjected to the following test: One milligram of bilirubin was shaken slowly in the dark for 25 mm. with 40 ml. of C11C13 and 5 ml. of 5% sodium bicarbonate. An aliquot of the chloroform phase was suitably diluted with more chloroform, and the visible absorption curves of this chloroform dilution and the sodium bicarbonate phase were obtained in a Perkin- Elmer recording spectrophotometer, Model 4000-A. The absorption CM C :::: z06c! (i - - : I:II1I ll :3 7 b 9 10 WAVELENGTH (MICRONS) IEE Fig. 2. Infrared spectrum of bilirubin in a KBr wafer. III!I

534 HENRY FT AL. Clinkal Chemistry Table 2. ABSORBANCY DATA AFTER PARTITIONING BETWBEN Cm.oRoFoaM AND SoDIuM BICARBONATE Product CHC1 phase NaHCO, phase A at 415 A at 458 ina mp.

6 534 HENRY FT AL. Clinkal Chemistry Table 2. ABSORBANCY DATA AFTER PARTITIONING BETWBEN Cm.oRoFoaM AND SoDIuM BICARBONATE Product CHC1 phase NaHCO, phase A at 415 A at 458 ina mp.at 1:8 dilutton Eastman Kodak purified 1X Glogau purified 1X purified 2X Pfanstiehl purified 1X purified 2X peak of the chloroform phases occurred at 453 m in all cases. The absorption peaks of the sodium bicarbonate phases always appeared between 415 and 425 m. Table 2 shows the results obtained with different samples of bilirubin. Absorbancy values of the sodium bicarbonate phase are corrected to a volume of 40 ml. (1:8 dilution) to permit direct comparison with calculated values for the undiluted chloroform phase. Significant amounts of colored material were extracted from impure bilirubin into sodium bicarbonate, whereas a very small amount was obtained from those of high purity. A sodium bicarbonate extract, showing an absorption peak at m, was acidified and extracted back into chloroform. The peak now appeared at 435 m. The sodium bicarbonate extracts rich in the brown material gave a positive diazo reaction. Since this material was not extractable from the chloroform into water and since the sodium bicarbonate solutions gave a positive diazo reaction only after the addition of methanol, it was concluded that the material was not bilirubin glucuronide. Two bilirubin preparations (Eastman Kodak, Lot 22, and Glogau, twice purified), were chromatographed on Whatman No. 1, paper* using (1) chloroform, (2) 1% ammonium hydroxide in n-butanol and (3) n-butanol-acetic acidwater (8:2:2) in an effort to demonstrate contaminants in the unpurified preparation. In addition, 1% ammonium hydroxide in iibutanol and solvent systems A and B of Cole, Lathe, and Billing (8) were used with siliconized Whatman No. 1 paper. Descending chromatograms were run in the dark in nitrogen-flushed tanks. With CHC13 alone, the bilirubin migrated, and there was more yellow trailing with the impure preparation than with the purified one. In the ammonia-butanol system, using both siliconized and unsiliconized *H. Reeve Angel & Co.

Vol. 6, No. 6. 1960 DETERMINATION OF BILE PIGMENTS 535 paper, streaking from the origin occurred with both bilirubin preparations.

7 Vol. 6, No DETERMINATION OF BILE PIGMENTS 535 paper, streaking from the origin occurred with both bilirubin preparations. The purified bilirubin did not migrate in the other solvent systems, but in the case of the impure bilirubin, a small amount of yellow material migrated as a streak from the origin. The strips were dried and dipped in neutral diazo reagent. In every case the yellow streaks gave a positive reaction. The position of the streaks relative to bilirubin suggest that the colored impurities were more polar. This is in agreement with the experiments in which bilirubin preparations were partitioned between C11C13 and sodium bicarbonate. The fact that appreciable increases in absorptivity of azobilirubin were achieved in the process of purification of some of the commercial products would indicate that a major fraction of the impurities removed either did not give the diazo reaction or at least yielded a markedly lower absorptivity in the diazo reaction. CONCLUSIONS Suggested standards for pure bilirubin are as follows: 60,100 for absorptivity at 453 m in chloroform; 64,100 for absorptivity (based on an equivalent amount of bilirubin, molecular weight of 584) of azobilirubin at 545 m, when the diazo reaction is carried out by the technic presented. Because the ratio, absorptivity of bilirubin in CHC13/absorptivity of azobilirubin, varied significantly between impure preparations (Table 1), it is concluded that standardization of the determination of bilirubin with an impure bilirubin preparation and correction of the standard from the absorptivity of that preparation in chloroform is not valid. The following alternatives are available for standardization of bilirubin tests: (1) Standardization from the absorptivity of azobilirubin, as suggested above; this approach is not generally advisable unless a reference standard is unavailable; (2) checking the azobilirubin absorptivity of the commercial product and, if not close to that of pure bilirubin, then purification, or correction of the absorbancy obtained on the standard (A8) as follows. 64,100 Corrected A. =... X A. obtained absorptivity obtained DISCUSSION Absorptivities of supposedly pure bilirubin in chloroform reported in the literature have varied significantly: 55,000-56,000 (2), 53,200-

8 536 HENRY ft AL. Clinical Chemistry 56,600 (9), 59,800 (1), and 60,000 (10). In most of the reported research involving bilirubin determinations, results have been obtained using commercial biirubin preparations of varying purity. It is hoped that this presentation will constitute a step toward establishment of a standard of purity for bilirubin that in turn should lead to more accurate bilirubin determinations. SUMMARY The purity of a number of commercial biirubin preparations has been examined. A technic of purification of bilirubin is presented and certain characteristics of the product examined. Tentative standards of purity for bilirubin are proposed. REFERENCES 1. O Hagan, J. E., Hamilton, T., LeBreton, E. C., and Shaw, A. E., GUn. Chem. 3, 609 (1957). 2. Lemberg, B., and Legge, J. W., Hematin Compounds and Bile Pigments. New York, Interscience Pub., Malloy, H. T., and Evelyn, K. A., J. Biol. Chem. 119, 481 (1937). 4. Sifferd, R. H., Porsche, J. D., and Sohns, P. J. (Armour and Co.), U. S. Patent 2,381,574 (1943). 5. Porsche, J. D., and Sifferd, B. H. (Armour and Co.), U. S. Patent 2,363,471 (1944). 6. Porsche, J. D., and Soims, F. J. (Armour and Co.), U. S. Paten.t 2,336,716 (1945). 7. Craven, C. W., Reissman, K. B., and Chinn, H. I., Anal. Chem. 24, 1214 (1952). 5. Cole, P. 0., Lathe, G. H., and Billings, B. H., Btochem. J. 57, 514 (1954). 9. Heilmeyer, L., Spectrophotometry in Medicine. London, Huger Ltd., Henry, B. J., Golub, 0. J., Berkman, S., and Segalove, M., Am. J. GUn. Pathal. 23, 841 (1953).

A STUDY OF THE METABOLISM OF THEOBROMINE, THEOPHYLLINE, AND CAFFEINE IN MAN* Previous studies (1, 2) have shown that after the ingestion of caffeine

A STUDY OF THE METABOLISM OF THEOBROMINE, THEOPHYLLINE, AND CAFFEINE IN MAN* BY HERBERT H. CORNISH AND A. A. CHRISTMAN (From the Department of Biological Chemistry, Medical School, University of Michigan,