heterologous enzyme immunoassay for serum androstenediol

Clinica Chimica Acta 296 (2000) 193–201www.elsevier.com/ locate /clinchim

Heterologous enzyme immunoassay for serumandrostenediol

a , a a*Noriko Tagawa , Junko Tamanaka , Aya Fujinami ,b b c cToshiko Kiguchi , Takeaki Naito , Toru Takano , Nobuyuki Amino ,

aYoshiharu KobayashiaClinical Chemistry Laboratory, Kobe Pharmaceutical University, 4-19-1, Motoyamakita-machi,

Higashinada-ku, Kobe, 658-8558, JapanbMedicinal Chemistry Laboratory, Kobe Pharmaceutical University, Kobe, Japan

cDepartment of Laboratory Medicine D2, Osaka University Medical School, 2-2, Yamadaoka, Suita City,565-0871, Japan

Received 12 October 1999; received in revised form 29 February 2000; accepted 8 March 2000

Abstract

A heterologous enzyme immunoassay for serum androstenediol (Adiol: 3b, 17b-dihydroxy-androst-5-ene) was established. The combination of anti-Adiol antiserum raised in rabbit againstAdiol 7-O-(carboxymethyl)oxime (Adiol 7-CMO) conjugated bovine serum albumin (Adiol7-CMO-BSA) and Adiol 7-iminomethylcarboxylic acid conjugated alkaline phosphatase was usedfor the assay. The sensitivity of the heterologous assay system was superior to that of ahomologous assay system in which an antibody raised in rabbit against Adiol 7-CMO-BSA andenzyme labeled antigen, Adiol 7-CMO conjugated alkaline phosphatase, were used. The minimal

21 21amount of Adiol detected was 0.4 ng ml and the measurable range was from 0.4 to 150 ng ml .21Intra-assay coefficients of variation (C.V.) were 8.6% (1.5260.13 ng ml , mean6S.D., n 5 10)

21 21and 6.7% (13.460.9 ng ml , n 5 10). Inter-assay C.V. were 12.9% (1.6360.21 ng ml , n 5 8)21and 11.5% (12.261.4 ng ml , n 5 8). A linear relation was observed between the serum sample

dilution and the Adiol concentration. For recovery study, authentic Adiol was added to serum21sample (original concentration: 1.43 ng ml ). The calculated final Adiol concentration was 2.99

21ng ml . The recovery was 98.6% (n 5 5). The Adiol concentrations in healthy subjects measured21 21by the proposed assay (male: 1.160.3 ng ml (mean6S.D.), range: 0.7–1.7 ng ml , age: 22–50,

21 21n 5 10; female: 0.660.4 ng ml , range: 0.2–1.6 ng ml , age: 23–48, n 5 20) were consistentwith reported values. 2000 Elsevier Science B.V. All rights reserved.

*Corresponding author. Tel: 1 81-78-441-7558; fax: 1 81-78-441-7559.E-mail address: [email protected] (N. Tagawa)

0009-8981/00/$ – see front matter 2000 Elsevier Science B.V. All rights reserved.PI I : S0009-8981( 00 )00261-8

194 N. Tagawa et al. / Clinica Chimica Acta 296 (2000) 193 –201

Keywords: Androstenediol; Enzyme immunoassay; Heterologous EIA; Homologous EIA

1. Introduction

Androstenediol (Adiol: 3b, 17b-dihydroxy-androst-5-ene) was first isolatedfrom adrenal vein blood by Hirschmann [1]. For many years, research on Adioland its sulfate focused on its role as an intermediate in sex steroid biosynthesis.Recently, Padgett et al. proposed that Adiol functions to augment the responseof the immune system or to counteract or modulate the immunosuppressiveeffects of corticosteroids such as cortisol [2]. In women, 100%, and in men,approximately 60%, of Adiol is derived from dehydroepiandrosterone (DHEA: 3b-hydroxy-androst-5-ene-17-one), which is the most abundant steroid in serum[3]. However, the physiological significance of Adiol in serum has yet to befully elucidated.

Several methods for measuring serum Adiol have been reported such ascompetitive protein binding assay (CPBA) [4,5], radioimmunoassay [6–8], gaschromatography (GC) [9] and gas chromatography–mass spectrometry (GC–MS) [10]. Enzyme immunoassay (EIA) is more sensitive than CPBA, GC andGC–MS. In this paper, we described an EIA for serum Adiol. We also discussedimprovements to the assay sensitivity achieved with a heterologous EIA system.To our knowledge, this is the first report on the measurement of serum Adiol byEIA. The proposed method is sufficiently sensitive, reproducible and cost-effectiveness for use in routine measurement of serum Adiol.

2. Materials and methods

2.1. Materials

The chemicals used in this study were obtained from the following sources:Adiol and 3b, 17b-diacetoxy-androst-5-ene, Steraloids Inc. (Newport, RI, USA);glycine, carboxymethoxylamine hemihydrochloride and a kit for measuring the

activity of alkaline phosphatase ‘Alkaline Phospha K-Test ’, Wako PureChemical Industry (Osaka, Japan); bovine serum albumin (BSA) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide HCl (EDAC), Nacalai Tesque Inc.(Kyoto, Japan); 5% HCl–methanol (w/v%), GL Science Inc. (Tokyo, Japan);

Sephadex G-25 fine , Pharmacia Biotech (Uppsala, Sweden); ASPEC pak(disposable octadecyl silane cartridge, 20 mm 3 0.9 mm I.D.) column, M&SInstruments Trading Inc. (Osaka, Japan); calf intestinal alkaline phosphatase

N. Tagawa et al. / Clinica Chimica Acta 296 (2000) 193 –201 195

(EC 3.1.3.1.), Boehringer Mannheim (Mannheim, Germany); anti-Adiol rabbitantiserum (raised in rabbit against Adiol 7-O-(carboxymethyl)oxime-BSA),Biogenesis (Poole, UK); second antibody (anti-rabbit g-globulin goat an-tiserum), Eiken Immunochemical Institute (Tokyo, Japan). Steroid-free serumwas prepared according to a method described by Heyns et al. [11]. Water waspurified with a Milli-Q system, Nippon Millipore Ltd (Tokyo, Japan) and otherchemicals were of analytical grade.

Synthesis of 3b, 17b-dihydroxy-5-androstene-7-one (Adiol 7-one) from 3b,17b-diacetoxy-androst-5-ene was carried out as described by Defaye et al. [12].3b, 17b-Dihydroxy-androst-5-ene 7-O-(carboxymethyl)oxime (Adiol 7-CMO)was prepared from Adiol 7-one by a method described by Fajkos et al. [13].

Serum samples were obtained from normal healthy subjects (ten males, age:22–50; 20 females, age: 23–48) and informed consent was obtained from allsubjects. No cases of patients with endocrine disease were included in this study.Serum samples were kept frozen at 2 408C until assayed.

2.2. Synthesis of Adiol 7-iminomethylcarboxylic acid (Adiol 7-ICA)

21A solution of glycine (25 mg) in 3 ml of 1 mol l NaOH was added to 3 mlof ethanol containing Adiol 7-one (50 mg) and the mixture was refluxed for 1 h.The solvent was evaporated under vacuum, ethyl acetate was added to theresidue and the product was extracted with water. Then the aqueous solution waswashed with ethyl acetate and was acidified with c-HCl. The solution wasextracted with ethyl acetate. The organic solution was washed with water anddried over anhydrous Na SO . Finally, the solvent was again evaporated under2 4

vacuum and the resulting residue (2 mg) was used as ligand for conjugation ofalkaline phosphatase without further purification.

2.3. Conjugation of alkaline phosphatase to Adiol 7-CMO and Adiol-7-ICA

Adiol 7-CMO or Adiol 7-ICA (2 mg) was dissolved in 100 ml of dimethyl-formamide and then EDAC (2 mg) was added. The reaction mixture was stirredat room temperature for 1 h. Then the mixture was added to a solution of

21alkaline phosphatase (0.5 mg) in 0.5 ml of 0.07 mol l phosphate buffer (pH8.0). The mixture was gently stirred at room temperature for 3 h. The resultingsolution was applied to a Sephadex G-25 fine column (400 mm 3 10 mm I.D.),

21which was eluted with 0.07 mol l phosphate buffer (pH 7.4). The fractionscontaining the alkaline phosphatase conjugated Adiol derivatives were collectedand stored at 48C. Under these conditions, the conjugates were stable for at least6 months.


2.4. Preparation of serum standards

21Adiol stock solution, 100 ng ml , in methanol was serially diluted withmethanol. Three hundred microliters of methanol standard (1.22 pg-10 ng/100ml) and 200 ml of steroid free serum were added to tubes. These standards werecalled ‘serum standards’, and were prepared just before assay.

2.5. Sample preparation for EIA

Serum standards or serum samples (200 ml) were extracted with 1 ml ofmethanol three times (when the serum concentration of Adiol was under thedetection limit, the volume of serum sample (0.5–1 ml) was increased). Thesolvent was combined and evaporated under vacuum. The residue was dissolvedwith 0.2 ml of 50% methanol. Then, the solvent was applied to a solid-phase

extraction column, ASPEC pak column (20 mm 3 0.9 mm I.D.), precon-ditioned with 4 ml of methanol and 8 ml of water. The column was washed with8 ml of water and the sample was eluted with 6 ml of 72% ethanol from thecolumn. The eluate that contained unconjugated steroids was evaporated undervacuum. The residue was dissolved in 300 ml of HPLC mobile phase solution(water: acetonitrile 5 7: 3 (v /v)) and 250 ml of the sample was subjected toHPLC. The HPLC conditions are described below. The eluate between 12.6 and15.4 min was collected and referred to as the Adiol fraction. The solvent wasevaporated under vacuum and the residue was dissolved in 250 ml of methanol.Aliquots of 100 ml of this solution were transferred into two tubes forduplication. After evaporation of the solution, EIA was performed.

2.6. HPLC conditions

HPLC was performed with two pumps (Waters, model 6000A and 510), a UVdetector (Japan Spectroscopic Co. Ltd, model Uvidec-100-IV) operating at 210nm, an auto sampler (Waters, model 710B) and a column (Wako Pure ChemicalIndustry, Wakosil-II 5C18-HG 150 mm 3 4.6 mm I.D.). Separations were

21performed at 408C and at a flow-rate of 1 ml min with a linear gradient ofacetonitrile (2%/min) from 30 to 70% in water. The mobile phase solvent waschanged back to the initial composition for the next run.

2.7. Procedure for EIA

A sample purified by HPLC as above was used for EIA. To the tube, 500 mlof anti-Adiol antiserum diluted with 0.07 M phosphate buffer (pH 7.4)containing 0.25% BSA, and 100 ml of alkaline phosphatase conjugated Adiol


21diluted with 0.07 mol l phosphate buffer (pH 7.4) containing 0.5% normalrabbit serum and 0.25% BSA, were added. The tube was stood at roomtemperature for 1 h. Then, 100 ml of the second antibody diluted with 0.07

21mol l phosphate buffer (pH 7.4) containing 0.25% BSA was added and themixtures were allowed to stand at 48C overnight. Finally, 2 ml of water wasadded and the reaction mixtures were centrifuged at 1000 3 g for 60 min. Theenzyme activity of the resulting precipitate which was washed twice with water

as above was measured with an Alkaline Phospha K-Test . A standard curvewas fitted by quadratic logit-log equation [14]. The program for the curve fittingwas kindly provided by Dr. K. Ichihara.

3. Results and discussion

3.1. Comparison of the sensitivity and detection limit of homologous andheterologous EIA

It is reported that more sensitive assays can be obtained by employing a ’site’or ’bridge’ heterologous EIA system rather than homologous one [15,16]. Weinvestigated the assay sensitivity of both systems. For the homologous system,the combination of antibody and enzyme-labeled steroid was anti-Adiol an-tiserum raised in rabbit against Adiol 7-CMO-BSA (Biogenesis: Poole, UK) andAdiol 7-CMO conjugated alkaline phosphatase. For the heterologous system, thecombination was the same antibody and Adiol 7-ICA conjugated alkalinephosphatase (Fig. 1). As shown in Fig. 2, the sensitivity of the heterologous EIAsystem was superior to that of the homologous one. The detection limit wasdetermined from the Adiol concentration of 2 1 standard deviation at the 0concentration of Adiol (n 5 5). The detection limit for the homologous and

21 21heterologous assay was 4.5 ng ml and 0.4 ng ml , respectively. We chose theheterologous EIA system for measurement of serum Adiol concentrations.

3.2. Evaluation of the present EIA

A typical calibration curve of serum Adiol is shown in Fig. 3. The minimal21amount of Adiol detected was 0.4 ng ml and the measurable range was from

210.4 to 150 ng ml . Two serum samples were diluted serially with Adiol freeserum (steroid free serum) and their Adiol concentrations were determined bythe present method (heterologous system). A linear relation was observedbetween the dilution and the Adiol concentration for each sample (data notshown). The accuracy was estimated from the recovery of Adiol from serum.Authentic Adiol was added to the serum sample (original concentration: 1.43

21 21ng ml ). The calculated final Adiol concentration was 2.99 ng ml . The


Fig. 1. Ligands for the homologous and heterologous ADiol EIA systems. BSA: bovine serumalbumin, Alp: alkaline phosphatase. [I] androstenediol 7-O-(carboxymethyl)oxime conjugatedBSA, [II] androstenediol 7-O-(carboxymethyl)oxime conjugated Alp, [III] androstenediol 7-iminomethylcarboxylic acid conjugated Alp.

Fig. 2. Calibration curves for ADiol obtained by homologous (closed triangle) and heterologous(closed circle) EIA. Each curve was obtained using serially diluted ADiol solution in methanol.


Fig. 3. A typical calibration curve for serum ADiol EIA. Vertical bar shows the mean6S.D.(n 5 5). M: male, F: female.

average recovery was 98.6% (n 5 5). The intra-assay coefficients of variation21(C.V.) were 8.6% (1.5260.13 ng ml , mean6S.D., n 5 10) and 6.7%

21(13.460.9 ng ml , n 5 10). The inter-assay C.V. were 12.9% (1.6360.2121 21ng ml , n 5 8) and 11.5% (12.261.4 ng ml , n 5 8).

According to the manufacturer’s data sheet, the cross-reactivity of the anti-Adiol antiserum used was 1.5% (dehydroepiandrosterone), 0.5% (testosterone),0.25% (DHEA) and 0.14% (androstenedione). As the present EIA was carriedout after the separation of these cross-reactive compounds by HPLC, they didnot affect the serum concentrations of Adiol.

3.3. Adiol concentrations in sera from healthy adults

Adiol concentrations in sera from healthy adults (age: male 22–50, n 5 10;female 22–48, n 5 20) were measured by the present method. There was nosignificant difference in age between the sexes. Serum concentrations of Adiol in

21 21males (1.160.3 ng ml (mean6S.D.), range: 0.7–1.7 ng ml ) were sig-21nificantly higher (P , 0.001) than those in females (0.660.4 ng ml , range:


210.2–1.6 ng ml ). These values are consistent with those obtained by radioim-munoassay [6,8,17], competitive protein binding assay [5] and gas chromatog-raphy–mass spectrometry [10].

Acknowledgements

This work was supported by Grants-in-Aid for Scientific Research to N.T.(no. 11771516) from the Ministry of Education, Science, Sports and Culture,Japan.

References

[1] Hirschmann H, Courcy C, Levy RP, Miller KL. Adrenal precursors of urinary 17-ketoste-roids. J Biol Chem 1960;235:PC48.

[2] Padgett DA, Loria RM, Sheridan JF. Endocrine regulation of the immune response toinfluenza virus infection with a metabolite of DHEA-androstenediol. J Neuroimmunol1997;78:203–11.

[3] Bird C, Murphy J, Boroomend K, Finnis W, Dressel D, Clark A. Dehydroepiandrosterone:Kinetics of metabolism in normal men and women. J Clin Endocrinol Metab 1978;47:818–22.

[4] Rosenfield RL, Otto P. Androstenediol concentrations in human peripheral plasma. J ClinEndocrinol Metab 1972;35:818–22.

[5] Demisch K, Magnet W, Neubauer M, Schoffling K. Studies about unconjugated andros-tenediol in human peripheral plasma. J Clin Endocrinol Metab 1973;37:129–34.

[6] Loriaux DL, Lipsett MB. Radioligand assay for D5-3b-hydroxysteroids. II. 5-Androstene-3b,17b-diol and 3b,17b-dihydroxy-5-pregnen-20-one. Steroids 1972;19:681–7.

[7] Rosemary C, Bonney MJ, Scanlon DL, Jones PA, Beranek MJ, Reed VHTJ. The interrela-tionship between plasma 5-ene adrenal androgens in normal women. J Steroid Biochem1984;20:1353–5.

[8] Belanger A, Candas B, Dupont A, Cusan L, Diamond P, Gomez JL, Labrie F. Changes inserum concentrations of conjugated steroids in 40- to 80-year-old men. J Clin EndocrinolMetab 1994;79:1086–90.

[9] Sirinathsinghji DJS, Mills IH. Concentration patterns of plasma dehydroepiandrosterone, D5-androstenediol and their sulphates, testosterone and cortisol in normal healthy women and inwomen with anorexia nervosa. Acta Endocrinol 1985;108:255–60.

[10] Dikkeschei LD, Willemse PHB, Wolthers BG, De Ruyter-Buitenhuis AW, Nagel GT.Delta-5-androstenediol and its sulphate in serum and urine of normal adults and patients withendocrine diseases. Clin Endocrinol 1993;39:475–82.

[11] Heyns W, Baelen H, Moor PD. Study of steroid-protein binding by means of competitiveadsorption: application to cortisol binding in plasma. Clin Chim Acta 1967;18:361–70.

[12] Defaye G, Luche MJ, Chambaz EM. Microbiological 7- and 15-hydroxylations of C-19steroids. J Steroid Biochem 1978;9:331–6.

[13] Fajkos J, Pouzar V,Veres K. Synthesis of (19E)-17b-hydroxy-3-oxoandrost-4-en-19-al 19-(O-carboxymethyl)oxime. Collect Czech Chem Commun 1990;55:2086–94.


[14] Ichihara K, Yamamoto T, Kumahara Y, Miyai K. An improved processing of radioimmunoas-say data by means of a desk-top calculator (1) Comparison of regression procedures appliedto selected kinds of radioimmunoassay. Clin Chim Acta 1997;79:331–40.

[15] Van Weemen BK, Schuurs AHWM. The influence of heterologous combinations of antiserumand enzyme-labeled estrogen on the characteristics of estrogen enzyme-immunoassays.Immunochemistry 1975;12:667–70.

[16] Hosoda H, Yoshida H, Sakai Y, Miyairi S, Nambara T. Sensitivity and specificity in enzymeimmunoassay of testosterone. Chem Pharm Bull 1980;28:3035–40.

[17] Kirschner MA, Sinhamahapatra S, Zucker IR, Loriaux L, Nieschlag E. The production,origin and role of dehydroepiandrosterone and D5-androstenediol as androgen prehormonesin hirsute women. J Clin Endocrinol Metab 1973;37:183–9.

heterologous enzyme immunoassay for serum androstenediol

Documents