A simple and sensitive assay of total serum bile acids

79

Clinica Chimica Acta, 70 (1976) 79-86 0 Elsevier Scientific Publishing Company,

Amsterdam

- Printed in The Netherlands

CCA 7755

A SIMPLE AND SENSITIVE ASSAY OF TOTAL SERUM BILE ACIDS

FUMIKO MASHIGE a, KAZUHIRO

IMAI b and TOSHIAKI

OSUGA c**

a Central Clinical Laboratory, Tokyo University Hospital, Hong0 7-3-1, Bunkyo-ku, Tokyo 113, b Faculty of Pharmaceutical Sciences, University of Tokyo, Hong0 7-3-1, Bunkyo-ku, Tokyo 113 and c Department of Internal Medicine, University of Tsukuba School of Medicine Sakura-mum, Zbaraki-ken 300-31 and Central Clinical Laboratory, Tokyo University Hospital, Hong0 7-3-1, Bunkyo-ku, Tokyo 113 (Japan) (Received

January 15,1976)

Summary A simple and sensitive method was developed for the quantification of serum total 3cu-hydroxy bile acids. 0.1 ml of serum was mixed with tris(hydroxymethyl) aminomethane hydrochloric acid buffer and heated at 67°C for 30 min. To the solution were added 3a!-hydroxysteroid:oxidoreductase (EC 1.1.1.50; 3aHSD), NAD, diaphorase (EC 1.6.4.3) and resazurin. The mixture was incubated at 20°C for 1 h. The resultant fluorescence of resorfin was measured at 580 nm with the excitation at 560 nm. The blank value was obtained after the same treatment of another 0.1 ml of the same serum without 3a-HSD. A linear relationship was obtained between the amount of bile acids and the fluorescence intensities in the range of 1 to 150 pmol/l. The recovery of bile acids added to the serum was 81.4 + 2.5 (SD.) % for cholate, chenodeoxycholate and deoxycholate. The bile acid content in the serum was 48.8 E.tmol/l with a standard deviation of kO.42 and a coefficient of variation of +0.87% in 10 replicate determinations. The mean bile acid content of normal fasting male sera was 8.0 pmol/l(3.612.6 pmol/l, n = 12) and of female sera 6.8 pmol/l(3.2-12.7 I.tmol/l, n = 13).

Introduction There are few satisfactory methods for the estimation of serum bile acids despite the growing interest in the relationship between the serum level and liver cell function. Iwata et al. [l] applied an enzymatic method for the determination of total * To whom correspondence should be addressed. Abbreviations: C. cholic acid: CDC. chenodeoxycholic hydroxysteroid:oxidoreductaee

(EC 1.1.1.60).

acid; DC, deoxycholic

acid; 3&H3D,

30r-

80

bile acids in blood, using NAD-linked 3ar-hydroxysteroid : oxidoreductase (EC 1.1.1.50, 3aHSD) from Pseudomonas testosteroni that had originally been developed for 3a-hydroxysteroids [ 21. Since then, several investigators [ 3,4] have modified the method and Murphy et al. [ 51 used the fluorescence of NADH to increase the sensitivity for the measurement of total bile acids in the serum. However, all the methods mentioned above need the prelimit extraction of bile acids either with a column [3] or organic solvents [ 51, which is time consuming and often gives low recoveries. In this paper we describe a new, simple and sensitive assay procedure for serum total bile acids without any prior extraction. The principle of the method is as follows: Serum enzymes are inactivated by heating, bile acids are converted to 3-0~0 bile acids with 3~HSD with the concomitant reduction of NAD to NADH, and then the hydrogen of the generated NADH is transferred by diaphorase (EC 1.6.4.3) to resazurin to yield the fluorophore, resorfin [6]. Finally, the fluorescence of resorfin is measured and this is proportional to the amount of bile acids. The method was applied to human sera. Materials and methods Reagents and samples

3~HSD * (supplied from Nyegaard Co., Oslo) was dissolved in distilled water. The specificity of the enzyme for 3a-hydroxy bile acids, C, CDC and DC has been shown elsewhere 171. Resazurin sodium (Tokyo Tanabe Seiyaku Co.) was dissolved in distilled water. NADdiaphorase ** (Sigma Co.) which contained 240 mg of NAD, 20 U of diaphorase and phosphate was dissolved in 40 ml of distilled water. The pH of the solution was 7.4 (0.065 M phospha~). Sodium cholate (Mikunikagaku Co.), sodium chenodeoxycholate and sodium deoxycholate (Tokyo Kasei Co.) were dissolved in distilled water and used as standards. Human sera were collected from normal fasting volunteers and patients with various diseases who were admitted to Tokyo University Hospital. Modified method of Schwarz et al. [3]

One ml of serum was diluted with 10 volumes of 0.1 M NaOH in physiologic saline and applied to an Amberlite XAD-2 column (0.5 cm X 5 cm). The column was washed with 5 ml of distilled water and eluted with 7 ml of ethanol at a flow rate of 0.2-0.3 ml per min. The eluate was evaporated to dryness at 70°C under reduced pressure. The residue was dissolved in 1 ml of methanol and two aliquots of 0.4 ml were removed. Samples were treated with the Sterognost-& Kit (Nyegaard Co., Oslo) which contained 3o-HSD, NAD and hydrazine hydrate. The intensity of fluorescence of the NADH produced was measured at 450 nm with excitation at 350 run. * 1 I.U.

is the amount of enzyme that will convert 1 pnol of deoxycholate per min at 26% and PH 9.6. ** 1 U equals a decrease in absorbance of 1.00 per min of 2.6-dichlorophenolindophenol at 25°C.

81

Results (A) Preliminary investigation 3a-HSD concentration. Two tubes were prepared for a test and its blank. One tube contained a mixture of 0.1 ml 150 E.IMDC-Na, 2.0 ml 0.05 M Tris/ HCl buffer (pH 9.5), 0.5 ml NADdiaphorase solution (NAD 3 mg and diaphorase 0.25 U), 0.2 ml 100 PM resazurin and 0.2 ml &-HSD solution (total 3.0 ml). The other tube contained the same reagents except for having distilled water instead of 3~HSD. The solutions were incubated at 25°C. The fluorescence was measured at 580 nm with excitation at 560 nm using a Shimadzu RF-501 recording fluorospectrometer (Shimadzu Seisakusho, Ltd.). The relative fluorescence intensities were obtained from the difference between the two tubes. The reaction curves are shown in Fig. 1. The reaction maximum was reached by 20 min and 0.0048 I.U. of 3a-HSD was sufficient to convert 15 nmol DC-Na in the assay system. According to a preliminary experiment, most of the sera from the patients with liver diseases were found to contain at most 150 /.LMof bile acids. Therefore, 0.0048 I.U. of 3a-HSD seemed to be enough for 0.1 ml of serum. NAD-diaphorase concentration. To the mixture of 0.1 ml 150 FM DC-Na, 1.5 ml 0.05 M Tris/HCl buffer (pH 9.5), 0.2 ml 3~HSD solution (0.0048 I.U.) and 0.2 ml 100 PM resazurin sodium were added to various volumes of NADdiaphorase solution (NAD 6 mg/ml and diaphorase 0.5 U/ml). The volume of the mixture was adjusted to 3.0 ml with 0.065 M phosphate buffer (pH 7.4). The reaction was performed at 25°C. The fluorescence was measured at 580 nm with excitation at 560 nm. A blank experiment was also performed without 3~-HSD. The reaction graphs are shown in Fig. 2. Judging from the curves the reaction seems to be influenced by the increased concentration of NAD. However, since pure diaphorase was not available it could not be confirmed. To shorten the reaction period the concentration of diaphorase selected was 0.25 U. Influence ofpH on the reaction. A mixture of 0.1 ml 200 PM DC-Na, 0.2 ml 3a-HSD (0.0048 I.U.), 0.5 ml NAD-diaphorase (NAD 3 mg and diaphorase 0.25 U), 0.15 ml 200 PM resazurin sodium and f.05 ml 0.05 M Tris/HCl buffer was in-

100

J

1

IO

20

30

Fig. 1. Effect of concentration 0.0144: 6. 0.0096. e, 0.0040. *

40

50

of 3eHSD

60

70

80

90 mm

on the reaction. 3cAISD

I.U.13 ml: a. 0.0240;

b. 0.0192;

c,

82

I

10

20

30

40

50

60

70

80

90mzn

Fig. 2. Effect of concentration of NADdiaphorase on the reaction. U); B, 4.5 mg (0.375 U): c, 3.0 mg (0.250 U); d, 1.5 mg (0.175 U).

NAD-diaPhoraee/3

ml: a. 6 mg (0.500

cubated at 22°C. The blank values were obtained after the experiment without 3a-HSD. The optimal pH of the solution was found to be about 9, using Tris/ HCl buffer of pH 9.5, as shown in Fig. 3. ~~f~~e~lce of temper~t~re on the reaction. A mixture of 0.1 ml 130 PM DCNa, 2.05 ml 0.05 M Tris buffer (pH 9.5), 0.2 ml 3al-HSD (0.0048 I.U.), 0.5 ml NAD-diaphorase solution (NAD 3 mg and diaphorase 0.25 U) and 0.15 ml 100 PM resazurin sodium was incubated at 20, 25 or 37°C. The blank values were also obtained as mentioned above. The lowest temperature was found to be best for the reaction, probably because of the instability of the resorfin produced at the higher temperatures. Thus, the temperature selected was 20°C. Inactivation of serum enzymes. The pH values of incubation mixtures were adjusted to 8.0, 8.5, 9.0, 9.5 and 10.0. At pH 8.5 the mixture became turbid, while it was transparent above pH 9.5. Therefore, the pH selected was 9.5. At pH 9.5, sera were inactivated by heating at 67°C for 30 min, but not at 60°C for 1 h. Therefore, inactivation was carried out at 67°C for 30 min. (B) Standard method

Two tubes were prepared for a serum and its blank. 0.1 ml of serum and 2.0 ml 0.05 Tris/HCl buffer (pH 9.5) were added to each tube. The tubes were

I

10

Fig. 3. Influence buffers.

2b

3-O

4b

of pH on the reaction.

5b

6bmin

a. pH 8.0; b, pH 8.5: o. pH S.O:d. pH 9.5: e. pH 10.0. Trls/HCl

83

150 pi-4

10 Fig. 4. Calibration

20

,uM

curve of DC-Na.

heated at 67°C for 30 min and cooled to room temperature. Then 0.9 ml of a mixture of 0.2 ml 3a-HSD (0.0048 I.U.), 0.5 ml NADdiaphorase (NAD 3 mg and diaphorase 0.25, U) and 0.2 ml 100 PM resazurin was added to one tube and the same mixture with distilled water instead of 3a-HSD was added to the other. The solution was incubated at 20°C for 1 h. The fluorescence intensity was measured at 580 nm with excitation at 560 nm. From the difference of fluorescence intensities between the two tubes the amount of bile acid was calculated using the calibration curve made with the standard solution (Fig. 4). A linear relationship between the amount of bile acids and the fluorescence intensities was obtained in the range of 1 to 150 PM. The excitation and emission spectra obtained from a serum were identical with those from the standard solution (Fig. 5). Recovery. 0.1 ml of 5 PM of each bile acid (C, CDC, Glyco-DC and Tauro-DC) was added to 0.1 ml of a serum. The recovery was 81.4 f 2.5 (S.D.) % for DC. The same range of recoveries were also obtained for the other bile acids. Comparison with the method of Schwarz et al. [3]. A good correlation (r = 0.99) was observed between the present method and that of Schwarz et al. using the same sera which showed no obvious jaundice (Fig. 6). The sensitivity of the present method was several times higher than that of Schwa& et al. [ 31. Precision. Ten replicate determinations were performed on serum samples from two patients. The mean value, standard deviation and coefficient of variation were 48.8 PM, k0.42 and kO.87%, respectively, for one serum. The corresponding values were 118 PM, +1.9 and *1.6% for the other.

84 ‘100

EXCITATION

Fig. 6. The excitation and emission spectra. -, (serum).

i0

lb0

standard solution (DC-Na); - - - - - -, sample solution

ltip.4

Method of Schworz et al. Fig. 6. Correlation between the method of Schwarz et al. [81 and the present method. y = 1.1x + 9.9. r = 0.99. n = 26.

85

(C) Normal values Total bile acids in the serum of normal fasting subjects ranged from 3.6 to 12.6 PM with a mean of 8.0 E.~M for 12 males and 3.2 to 12.7 PM with a mean of 6.8&M for 13 females. The values obtained by the present method were slightly higher than those observed by Murphy et al. [ 51. Discussion The first step, i.e. heat inactivation of serum enzymes, is essential to the present method. There are many enzymes affecting NAD-linked oxidoreduction that generate NADH to make the reaction proceed. Some of them seem to be heat stable because 1 h of heating at 60” C was not sufficient for inactivation of some sera. However, all were inactivated by 30 min at 67°C. This simple treatment of serum eliminated any laborious extraction using a column [3] or organic solvents [ 51. The interference of bilirubin, which is often encountered in the fluorescence measurement of NADH (excitation at 350 nm and emission at 460 nm), was eliminated in the present method, since the absorption of bilirubin exists around 340 nm, while the fluorescence of resorfin is at 580 nm with the excitation at 560 nm. Since the amount of lithocholic acid in serum has been suggested as being scanty [ 81, lithocholic acid was excluded from our study. The present method can determine only 3at-hydroxy bile acids. Sulfated bile acids that were elevated in hepatobili~ diseases [9] cannot be measured without hydrolysis. The presence of 3a-hydroxysteroids other than bile acids has been reported to be less than a few nmol/l in serum and moreover almost all of them were conjugated [lo]. Therefore, the present method is assumed to be specific for serum bile acids. The method proved to be simple, rapid and reliable compared with the other methods for the determination of serum bile acids which are complex, cumbersome and require relatively large amounts of serum. By contrast, this method needs only 0.2 ml of serum owing to the high sensitivity (0.1 nmol of bile acid in the incubation mixture is me~u~ble). With this assay system a large number of samples could be handled in clinical laboratories and also application to an automated system may be possible. Acknowledgements The technical assistance of Miss Etsuko Yanagisawa and Miss Setsuko Shioya is gratefully appreciated. We also wish to express our deep thanks to Professor M. Yamanaka and colleages in the general unit of the Central Clinical Laboratory, Tokyo University Hospital, for their support. References 1 Iwata, T. and Yamasaki, K. (1964) J. Biochem. (Tokyo) 66.424 2 Hurlock, B. and Talalay, P. (1967) J. Biol. Chem. 227, 37 3 Schwarz, H.P., Bergmann. k.V. and Paumgartier, G. (1974) Clin. Chim. Acta 50,197

86 4 5 6 7 8 9 lo

Panveliwalla. D., Lewis, B., Wooton. I.D.P. and Tabaqchali. S. (1970) J. Clin. Pathol. 23. 309 Murphy, G.H., Billing, B.H. and Baron. D.N. (1970) J. Clin. Pathol. 23. 594 Guilbault, G.G. and Kramer, D.N. (19651 Anal. Chem. 37.1219 Osuga. T. and Mashlge, F. (19761 Isaku no Ayumi 96.16 Pellizzsxi, E.D.. O’Nail, F.S., Farmer. R.W. and Fabre, Jr., L.F. (1973) Clin. Chem. 19.246 k&&no. I., Hashimoto, H., Shlnozaki, K., Yoshino. K. and Nakagawa, 3. (1975) Gastroenterology 68, 545 6amuels, L.T. and Eik-Nes, K.B. (1968) in Metabolism of steroid hormones (Greenberg, D.M., ed.1. Vol. 2, pp. 169, Academic Press, New York, London