A bioluminescent assay for 12-α-hydroxy bile acids using immobilized enzymes

ANALYTICAL

BIOCHEMISTRY

133,244-250

(I 983)

A Bioluminescent Assay for 12-cY-Hydroxy Acids Using Immobilized Enzymes

Bile

JUERGEN SCHOELMERICH,*,’ JOHN E. HINKLEY,* IAN A. MAcDoNALD,t ALAN F. HOFMANN,$ AND MARLENE DELucA*~~ *Departments of Chemistry and &Uedicine, University of California, San Diego, La Jolla, California 92093 and tDepartment of Medicine, Dalhousie University, Halifax, Canada Received March 7, 1983 A bioluminescent assayfor 12-cr-hydroxy bile acids was developed using enzymes coimmobilized onto Sepharose 4B. The immobilized enzymes used were a bacterial 12-a-hydroxysteroid dehydrogenase, bacterial luciferase, and NADPHPMN oxidoreductase or bacterial diaphorase. The assay was specific for 12+hydroxy bile acids and the lower limit of detection was 4 pmol/O.S ml assay volume with a linear range of 4 to 2000 pmol. Intraassay precision was from 7.8 to 8.2%. Values obtained with this assayshowed good agreement with those obtained by gas-liquid chromatography. The system using diaphorase was not stable at 4°C in the absence of added thiol compounds, but could be stabilized by the addition of glutathione (0.5 mM). The assay is a convenient, a rapid, and an extremely sensitive method for the measurement of 12-o-hydroxy bile acid concentrations in the serum of patients or experimental animals. KEY WORDS: bile acids; bioluminescent assay; immobilized enzymes; hydroxysteroid dehydrogenase.

The determination of fasting and postprandial serum bile acids (BA)3 is under active study to evaluate its clinical value in assessing liver disease or ileal malfunction (l-4). To date, widespread application of bile acid measurement methods has been limited by the sensitivity of methods such as gas-liquid chromatography (GLC) or enzymatic determinations or by the difficulty and lack of specificity of competitive binding techniques such as radioimmunoassay (RIA) and enzyme immunoassay (5- 13). The sensitivity of enzymatic methods has been increased modestly by using NADH fluorescence rather than ab’ Present address: Department of Internal Medicine, University of Freiburg, West Germany. * To whom correspondence should be addressed: Department of Chemistry, M-00 1, University of California, San Diego, La Jolla, Calif. 92093. 3 Abbreviations used: OH BA, hydroxy bile acid; BA, bile acid(s); GLC, gas-liquid chromatography; HSD, hydroxysteroid dehydrogenase; BSA, bovine serum albumin; DTT, dithiothreitol. 0003-2697183 $3.00 Copyright 0 1983 by Academx Press. Inc. All rights of reproduction in any form reserved.

sorbance (14,15) or by the addition of fluorescent redox indicators ( 16,17), as discussed recently in more detail (1,18). Bioluminescence assays have been described for testosterone and androsterone using either 3-a-hydroxysteroid dehydrogenase (HSD) or 3-&HSD coimmobilized with NADHFMN oxidoreductase and bacterial luciferase ( 19). We have recently reported a successful adaptation of this approach to the measurement of primary bile acids using coimmobilized 7(u-HSD (18). We describe here the develop ment, characterization, and validation of a bioluminescent assay for 12-cr-hydroxy bile acids (12-a-OH BA) using 12+HSD coimmobilized with bacterial luciferase and either NADPHFMN oxidoreductase or diaphorase. MATERIALS

AND

METHODS

NADP, NADPH, and bacterial diaphorase were purchased from Boehringer-Mannheim, Mannheim, West Germany; FMN, decanal, 244

BIOLUMINESCENT

ASSAY

OF

and dithiothreitol (DTT) were from Sigma, St. Louis, Missouri; Sepharose 4B was from Pharmacia, Piscataway, New Jersey; and cyanogen bromide was from Eastman-Kodak, Rochester, New York. NADPHFMN oxidoreductase and bacterial luciferase were prepared and assayed as previously described (20,2 1). Bile acids were obtained from different sources and had a purity of 98% on thin-layer chromatography ( 18,22). 12-CX-HSD was prepared as recently reported (23). The immobilization of the enzymes on cyanogen bromide-activated Sepharose 4B was done as described previously (18,19). The first system consisted of a mixture of 0.5 ml luciferase ( 10 mg/ml, 3 X lo* relative light units/ ml), 0.9 ml of NADPHFMN oxidoreductase (1.1 mg/ml, 3.3 U/ml), 1.O ml of 12-(u-HSD (1.3 mg/ml, 3 U/ml), and 10 mg of bovine serum albumin (BSA) in 1.6 ml of sodium pyrophosphate buffer, 0.1 M, pH 8.0, EDTA 1 IIIM. The second mixture contained 0.5 ml luciferase, 0.5 ml diaphorase ( 19.2 mg/ml, 18 U/ml), and 1.O ml 12-(r-HSD. The volume was brought to 4 ml by the addition of 2 ml of pyrophosphate buffer. After overnight dialysis against 4 liters of the same buffer, 3 ml of the mixture was reacted with 1 g of the activated Sepharose. After reaction ( 16 h, 4°C) and washing as described before ( 18), the immobilized enzymes were suspended at a ratio of 1 g Sepharose/ 10 ml storage buffer. For the oxidoreductase system, phosphate buffer, 0.1 M, pH 7.0, containing BSA (10.2 g/dl), DTT (0.1 mM), and sodium azide (0.02 g/dl) was used. With diaphorase the storage buffer contained glycerol (lo%, v/v) and sodium azide (0.02 g/dl). Later preparations of the immobilized enzymes containing diaphorase had 0.5 mrvr glutathione added to the buffer. Efficiency of the immobilization procedure was measured as described ( 18,19). 12-cu-HSD activity was determined by adding 50 ~1 of enzyme mixture or Sepharose-bound enzymes to 1 ml of phosphate buffer (0.1 M, pH 7.0) containing NADP (2 mM) and cholyltaurine (2 mM) and measuring the increase in absorbance at 340 nm. Diaphorase activity was as-

I2-a-HYDROXY

BILE

ACIDS

245

sayed by following the decrease in absorbance at 340 nm when 50 ~1 of enzyme mixture or Sepharose-bound enzymes were added to 1 ml of phosphate buffer containing NADH (0.2 m&Q and FMN (0.13 mM). The oxidation of NADH in the absence of added FMN was subtracted from that in the presence of FMN. Protein concentrations were calculated from the absorbance at 280 nm. The principle of the bioluminescent assay is identical to that recently described for androsterone or bile acids ( 18,19,24). The assay was carried out by combining 420 ~1 of phosphate buffer (0.1 M, pH 7.0) containing 0.004% decanal with 10 ~1 of NADP (20 mM), 10 ~1 of FMN (2 mrvr), and 10 or 20 ~1 of immobilized enzymes (1 g/ 10 ml). After the tube was mixed, a background reading was obtained, and then 1O-60 ~1 of standard solution or a diluted serum sample ( 1:4- 1:64, v/v with phosphate buffer) was added and the tube was mixed for 5 s on a Vortex mixer and the peak light emission read. Alternatively, the maximum peak height after repeated shaking and reading was used. Light emission of twice the background signal was defined as the detection limit. All light measurements were done in an Aminco Chem Glow photometer. Diluted sera were heated for 15 min at 68 “C prior to measurements, which were carried out at room temperature, as described previously for 7-(Y-OH BA (18). Values of serum samples were directly read from standard curves using cholyltaurine (0.4-200 PM). Standard curves were also constructed for deoxycholyltaurine, deoxycholylglycine, and deoxycholate as well as for cholylglycine and cholate. Specificity of the assay was assessedby using samples of other BA or steroids lacking a 12cr-hydroxy group. Intraassay precision was determined by 10 replicate measurements of two different serum samples with low and high BA concentrations during 1 day. Recovery was measured by adding aliquots of cholyltaurine (lo- 100 pM) to serum samples with a low (0.70 PM), a medium (2.83

246

SCHOELMERICH

surements with those obtained by GLC. These serum samples from patients with various liver diseases were generously supplied by Dr. G. van Berge Henegouwen, Municipal Hospital, Arnhem, The Netherlands. Individual BA had been measured in the samples by GLC as reported by his laboratory (5).

/ 100

-

/

l

z

E

g s

ET AL.

lo-

RESULTS r =0.9965

The amount of active enzyme obtained on the Sepharose is expressed as a percentage of the amount in the initial coupling mixture. The yields were 12% for the luciferase, 50% for the oxidoreductase, and 53% for the 12(Y-HSD. With the diaphorase preparation they were 14, 32, and 41%, respectively. Kinetics of the assay as judged by light output were identical for standard and serum samples. A linear response of light intensity with BA concentration was obtained from 1 to 200 PM for all 12+OH BA studied with both immobilized systems. No differences between conjugated and unconjugated BA and between different 12-a-OH BA were found. Figure 1 gives the standard curve obtained with a fresh diaphorase preparation which was more sensitive (0.4-200 PM) because of lower background (0.6 light units) than that containing NADPH:FMN oxidoreductase (l-200 PM, background 2 light units) for identical samples. The linearity and sensitivity of the assay were

y = 1.060x +ooLT74 l l/ 0.4

, 2

20

200

BA CONCENlRATlON(~M)

FIG. 1. Typical standard curve using the diaphorase preparation and cholyltauxine (0.4-200 I.~M), first peak method. PM), and a relatively

high (9.37 PM) BA concentration. Samples were assayed before and after addition of the standard to calculate recovet-y percentage. Stability of the preparations was assessed by determination of the amount of light obtained with 200 pM cholyltaurine during a period of 8 weeks. The preparation containing diaphorase was not stable, and the effect of added mercaptoethanol or glutathione (OS1.OmM) was studied for the remaining period. The assay was validated on serum samples by comparing results of bioluminescent mea60 r

c

b

a

50 g 3 5 9 g 's g

\ 40302010 -

2 BA

4

6

1 s

n. B

\

I 10 A

I 12

I 14

I 16

I 18

1 20 MIN. A

FIG. 2. Typical time course of three different assaymethods using the diaphorase preparation. (a) Maximum peak height obtained after repeated mixing; (b) first peak after initial mixing (c) peak after injection of NADP into the assay tube. B = background; A = assay of a standard (20 PM cholyltaurine).

BIOLUMINESCENT

ASSAY OF 12-a-HYDROXY

TABLE 1 INTRAASSAYPRE~ISIONFORTWODIFFERENT BILE ACID CONCENTRATIONS

LowBA High BA

Mean (FM)

SD

n

cv @)

4.32 38.84

0.34 3.28

10 10

1.8 8.2

Note. Intraassay precision using the diaphorase system for two different BA concentrations, mean + standard deviation (SD), coefficient of variation (CV), first peak method.

similar when either the first peak height or the maximum peak height method was used. Figure 2 shows a typical time course of light emission for a standard using either the maximum peak height obtained after repeated mixing (a) or the first peak height (b). The latter system is dependent on shaking intensity and time when diaphorase is used. If the reaction is started with the injection of NADP, the response shown in Fig. 2c is obtained. Since the injection method gives similar results but is less time consuming, it is preferable. The assay was found to be specific for 12a-hydroxy compounds, as there was no light produced by androsterone, cortisol, a 12-keto BA, chenodeoxycholate and its conjugates,

247

BILE ACIDS

muricholate or hyodeoxycholate, at concentrations as high as 10 mM. Lactate, pyruvate, acetoacetate, Bhydroxybutyrate, and malate (lo-20 InM) did not interfere with the assay. The intraassay precision at two different BA concentrations was satisfactory (Table 1). Recovery of added standards was good and similar for all three concentrations used (Table 2). The preparation containing NADPH:FMN oxidoreductase was stable over a period of 8 weeks when stored at 4”C, but that prepared with diaphorase lost considerable sensitivity during the same period. The addition of /3mercaptoethanol (1 mM) to the preparation, although resulting in a somewhat higher background, stabilized the system. Use of glutathione (0.5 mM) rather than mercaptoethanol lowered the background and also maintained activity. Figures 3a-c show data comparing the results obtained using the bioluminescent assay with those obtained by GLC on human serum samples. The agreement was equally good for both preparations and both assay methods used. DISCUSSION

The results presented indicate that 12-eOH BA can be measured with excellent spec-

TABLE 2 RECOVERYOFADDEDCHOLYLTAUR~NESTANDARDSTOVARIOUSSERUMSAMPLES Initial BA concentration (PM)

Cholytaurine added (NM)

BA measured (PM)

Difference (PM)

Recovery (%)

9.37 9.37 9.37

0 25 50

9.37 34.41 58.60

0 25.04 49.23

100.0 loo.2 98.5

2.83 2.83 2.83

0 5 10

2.83 7.58 12.47

0 4.75 9.64

100.0 95.0 96.4

0.70 0.70 0.70

0 5 10

0.70 5.64 10.81

0 4.94 10. I 1

100.0 98.8 101.1

Note. Recovery of serum bile acids when different amounts of cholyltaurine standards were added. Diaphomse system, first peak method.

248

SCHOELMERICH

ET AL.

GA CONCENTRdTlON(~MI (GLC)

1

I

10 100 GA CONCENTRATlON(pM) (GLC)

1000

1 1 IO 100 GA CONCENTFlATlON(~M) (GtCJ

1 1000

FIG. 3. Validation of the bioluminescent assaycomparing it with GLC on serum samples. (a) Oxidoreductase preparation, first peak method; (b) diaphorase preparation, maximum peak height method; (c) diaphorase preparation, first peak method.

BIOLUMINESCENT

ASSAY OF I 2-(u-HYDROXY

ificity and extreme sensitivity by an enzymecoupled bioluminescent assay utilizing an immobilized 12-cu-HSD isolated from Clostridia (23). In contrast to the 7-a-OH BA bioluminescent assay previously reported (1 S), a commercially available diaphorase was used as well as the NADPH:FMN oxidoreductase isolated from Beneckeu harveyi (21). Use of the diaphorase improves sensitivity of the assay since background is less. However, it has to be mentioned that diaphorase-containing systems are very sensitive to differences in mixing. This disadvantage can be overcome by performing the assay as an injection assay in which the NADP is injected into the premixed tube to start the reaction. Comparable sensitive assays have been performed using flow cells (25), thereby proving the reusability of the immobilized enzymes. Thus automated versions of the assay are possible. The specificity, intraassay precision, and recovery of added standards were comparable to other systems (7,26-29). The sensitivity was as good as that of competitive binding assays and superior to all other methods published so far (28-32).

BILE ACIDS

249

the value of knowing the BA profile in serum in assessing liver disease is questionable (1,35,39,40).

However, the 12-a assay may also be of use in animal experiments, especially with species in which 12-a-OH BA are predominant such as the rabbit and the dog. Plasma BA have been shown to be an indicator of liver damage in experimental models (41). Because of its rapidity, this assay could even be used as a short-term follow-up in experimental systems such as isolated liver perfusion or isolated hepatocytes. The value of serum BA for screening or diagnostic purposes continues to be under active investigation (42-47). Neither their value in assessing liver disease nor in quantification of ileal dysfunction has been finally determined. The development of this 12-a-OH bioluminescent assay is an additional step toward development of a battery of inexpensive, rapid, convenient, sensitive, and specific assays for BA which will facilitate future studies in humans and animals to define the clinical and experimental utility of BA determinations (1). ACKNOWLEDGMENTS

Utility

of the Assay

The 12-a-HSD assay offers the advantage of determining both conjugated and unconjugated bile acids and thus permits total 12a-OH BA determinations on serum without any chemical procedure. The 12-a-assay should be completely specific since the only 12-a-steroids are BA. Further, in contrast to 3-a and 7-a positions, the 12-a position is rarely sulfated in cholestasis (33-35). As has been proposed with enzymatic endpoint determinations (36,37), a combination of this assay with that recently reported for 7-a-OH BA ( 18) and one using immobilized 3-(Y-HSD as described for androsterone ( 19) should allow the determination of the major circulating serum BA of most vertebrates. To achieve this, problems arising from the differing affinity of 3-a-HSD and 7-a-HSD for certain BA (36,38) must be solved. And, despite earlier claims,

Supported in part by NIH Grant AM 2 I506 and the National Science Foundation. J.S. was supported by a research grant from the Thyssen Foundation, West Germany.

REFERENCES

5. 6. 7. 8. 9.

Hofmann, A. F. (1982) Heputology 2, 5 12-5 17. Hofmann, A. F. (1977) Clin. Gastroenterol. 6, 3-24. Gilmore, 1. T., and Hofmann, A. F. (1980) Gastroenteroiogy 18, 177-179. Heubi, J. E., Bali&et-i, W. F., Partin, J. C., Schubert, W. K., and Suchy, F. J. (1980) .I. Lab. Clin. Med. 95,23 l-240. van Berge Henegouwen, G. P., Reuben, A., and Brandt, K-H. (1974) Cfin. Chim. Actu 54, 249261. Ali, S. S., and Javitt, N. B. ( 1970) Cunad. J. Biochem. 48, 1054-1057. Sandbcrg, D. H., Sjovall, J., Sjovall, K., and Turner, D. A. (1965) J. Lipid Rex 6, 182-192. Makino, I., and Sjovall, J. (1972) Anal. Lett. 5, 34 1. Iwata, T., and Yamasaki, K. (1964) J. Biochem. (Tokyo) 56, 424-43 1.

250

SCHOELMERICH

10. Fausa, O., and Skalhegg, B. A. ( 1977) &and. .I, Gustroenterol. 12, 441-447. 11. Simmonds, W. J., Korman, M. G., Go, V. L. W., and Hofmann, A. F. (1973) Gastroenterology 65, 705-711. 12. Roda, A., Roda, E., Aldini, R., Capelli, M., Festi, D., Sama, C., Mazzella, G., Morselli, A. M., and Barbara, L. (1980) Clin. Chem. 26, 1677-1682. 13. Spenney, J. G., Johnson, B. J., Hirschowitz, B. I., Mihas, A., and Gibson, R. (1977) Gastroenterology 72, 305-3 11. 14. Murphy, G. M., Billing, B. A., and Baron, D. N. (1970) J. Clin. Pathol. 23, 594-598. 15. Schwartz, H. P., von Betgmann, K., and Paumgartner, G. (1974) C/in. Chim. Acta 50, 197-206. 16. Mashige, F., Imai, K., and Osuga, T. (1974) C’fin. Chim. Acta 70,79-86. 17. Mashiie, F., Tanaka, N., Maki, A., Kamei, S., and Yamanaka, M. (1981) C’fin. Chem. 27, 1352-1356. 18. Roda, A., Kricka, L. J., DeLuca, M., and Hofmann, A. F. (1982) J. LipidRes. 23, 1354-1361. 19. Ford, J., and DeLuq M. (198 1) Anal. B&hem. 110, 43-48. 20. Jablonski, E., and DeLuca, M. (1977) Biochemistry 16,2932-2936. 21. Baldwin, T. O., Nicoli, M. Z., Becvar, J. E., and Hastings, J. W. (1975) J. Biol. Chem. 250,2763-2768. 22. Roda, A., Hofmann, A. F., and Mysels, K. J. (1983) J. Biol. Chem. 258,6362-6370. 23. Macdonald, I. A., Jellett, J. F., and Mahony, D. E. (1979) J. Lipid Rex 20, 234-239. 24. DeLuca, M. (1982) in Luminescent Assays: Perspectives in Endocrinology and Clinical Chemistry (Serio, M., and Pazzagli, M., eds.), pp. 115-123, Raven Press, New York. 25. Kricka, L. J., Wienhausen, G. K., Hinkley, J. E., and DeLuca, M. (1983) Anal. B&hem. 129,392-397. 26. Setchell, K. D. R., and Matsui, A. (1983) Clin. Chim. Acta 127, l-17. 27. Ikeda, M., Nanba, N., Hayakawa, S., and Ohmori, S. (1980) J. Clin. Chem. C/in. B&hem. 18, 407411. 28. Starkey, B. J., and Marks, V. (1982) Clin. Chim. Acta 119, 165-177. 29. Kamada, S., Maeda, M., Tsuji, A., Umezawa, Y., and

ET AL.

30. 31. 32. 33. 34. 35.

36. 37. 38.

39. 40. 41. 42. 43. 44. 45. 46. 47.

Kurahashi, T. (1982) J. Chromatogr. 239, 773783. Beppu, T., Seyama, Y., Kasama, T., and Yamakawa, T. (1981) J. Biochem. 89, 1963-1973. Gnishi, S., Itoh, S., and Ishida, Y. (1982) B&hem. J. 204, 135-139. Linnet, K. (1982) Stand. J. Clin. Lab. Invest. 42, 455-460. van Berge Henegouwen, G. P., Brandt, K-H., Eyssen, H., and Parmentier, G. (1976) Gut 17, 861-869. Raedsch, R., I.auterbur& B. H., and Hofmann, A. F. (198 1) Dig. Dis. Sci. 26, 394-401. Stiehl, A., Becker, M., Czygan, P., Frohling, W., Kommerell, B., Rotthauwe, ,H. W., and Senn, M. (1980) Eur. J. Clin. Invest. 10, 307-316. Macdonald, I. A., Williams, C. N., and Mahony, D. E. (1974) Anal. Biochem. 57, 127-136. Macdonald, I. A., Williams, C. N., and Musial, B. C. (1980) J. Lipid Res. 21, 38 l-385. Macdonald, I. A., Williams, C. N., and Mahony, D. E. (1974) in Advances in Bile Acid Research (Matem, S., Hackenschmidt, J., Back, P., and Gerok, W., eds.), pp. 99-104, Schattauer Verlag, New York. Linnet, K., and Kelbaek, H. (1982) Stand. J. Gastroenterol. 17, 919-924. Editorial (1982) Lancet 2, 1136-l 138. Anwer, M. S., Engelking, L. R., Gronwall, R., and Klentz, R. D. (1976) Rex Vet. Sci. 20, 127-130. Aldini, R., Roda, A., Festi, D., Mazzella, G., Morsel& A. M., Sama, C., Roda, E., Scopinaro, N., and Barbara, L. (1982) Gut 23, 829-834. Matsui, A., Psacharopoulos, H. T., Mowat, A. P., Portmann, B., and Murphy, G. M. (1982) J. Clin. Pathol. 35, 101 l-1017. Heuman, R., Sjcdahl, R., Tobiamon, P., and Tagesson, C. (1982) Stand. J. Gastroenterol. 17, 137-140. Aldini, R., Roda, A., Festi, D., Sama, C., Mazzella, G., Bazzoli, F., Morselli, A. M., Roda, E., and Barbara, L. (1982) Dig. Dis. Sci. 27, 495-502. Samuelson, K., Aly, A., Johansson, C., and Norman, A. (1982) &and. J. Gastroenterol. 17, 121-128. Ewerth, S. (1982) Stand. J. Gastroenterol. 17, 781784.