Enzyme immunoassay for antibodies to thyroxine in human serum using synthesized antibody as a calibrator

Journal oflmmunologicalMethods, 115 (1988) 263-268

263

Elsevier JIM04992

Enzyme immunoassay for antibodies to thyroxine in human serum using synthesized antibody as a calibrator R. Y a m a m o t o t, S. K i m u r a 1 a n d Y. Ishizuki 2 I Department of Research and Development, Amano Pharmaceutical Co., Kunotubo, Nishiharu, Nishikasugai, A ichi 481, Japan and 2 Ishizuki Thyroid Clinic, Sakae 2, Naka-Ku, Nagoya 460, Japan

(Received 3 May 1988, revised received 21 July 1988, accepted 24 July 1988)

We prepared human I g G labeled with rabbit antibodies to thyroxine, and used it for the assay of antibodies to thyroxine in human serum as a calibrator. The assay system consisted of thyroxine labeled with /3-D-galactosidase and a microcolumn containing goat antibodies to human I g G immobilized on Sepharose 4B. Each serum sample or standard serum containing human I g G labeled with anti-thyroxine antibodies was incubated with the enzyme-labeled thyroxine, and the reaction mixture was passed through the microcolunm. The column was washed to remove the unbound label, and enzyme activity of the bound label was assayed. The minimum detectable amount of the anti-thyroxine antibodies was 0.3 n g / a s s a y tube. The analytical recoveries of human I g G labeled with the antibodies added to human serum were from 103% to 108%. We measured anti-thyroxine antibodies in human sera from 187 patients with autoimmune thyroid diseases. The antibodies were detected in 32 serum samples, and the values in positive samples varied from 0.31 to 2.31 ~tg/ml. On the other hand, in 58 sera from patients with non-thyroid diseases, the antibodies were detected in only two samples. Key words: Thyroxine; Enzyme immunoassay; Microcolumn; (Antibody)

Introduction The presence of autoantibodies to hormones, hormone receptors or various tissue components has been reported (Belyavin and Trotter, 1959; Berson and Yalow, 1959; Smith and Hall, 1981). In thyroid diseases, autoantibodies to thyroid hormones, thyroglobulin, microsome or T S H receptor have been known (Robbins et al., 1956; Belyavin and Trotter, 1959; Smith and Hall, 1981).

Correspondence to: R. Yamamoto, Department of Research and Development, Amano Pharmaceutical Co., Kunotubo, Nishiharu, Nishikasugai, Aichi 481, Japan. Abbreviations: T4 thyroxine; IgG, immunoglobulin G; TSH, thyroid-stimulating hormone; MCHA, anti-microsome antibody.

Previously we developed an enzyme immunoassay for thyroid hormones ( Y a m a m o t o et al., 1982), and this assay method was applicable to measurement of antibodies in serum. More recently we developed a two-site enzyme immunoassay with use of rabbit F(ab')2-1abeled goat antibodies (Yamamoto et al., 1986). This technique for synthesis of the hybrid immunoglobulin was useful for preparation of animal antibody-human I g G conjugate, which could be used in assays of antibodies in human serum as a calibrator. Here we describe an enzyme immunoassay for antibodies to thyroxine in h u m a n serum with use of human I g G labeled with rabbit antibodies to thyroxine, as a calibrator, and a microcolumn containing anti-human I g G antibodies immobilized on Sepharose 4B.

0022-1759/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

264

Materials and methods

Antigen, antibodies and human immunoglobulin G (IgG) t-thyroxine (crystalline) was purchased from Sigma Chemicals Co., St. Louis, MO. Rabbit antiserum to thyroxine and goat antiserum to human IgG (whole IgG) were obtained from Medical and Biological Laboratories, Nagoya, Japan. The affinity constant of antiserum to thyroxine was 1.2 × 109 l / m o l by our enzyme immunoassay (Yamamoto et al., 1982). Pooled human serum was obtained from healthy blood donors. The IgG fractions were isolated from antiserum or human serum by precipitation with ammonium sulfate, dialysis, and chromatography on DEAE-cellulose column (Kato et al., 1979). Antigen-enzyme conjugate L-thyroxine (T4) was coupled to fl-D-galactosidase (EC 3.2.1.23, from Escherichia coli, Boehringer Mannheim, Mannheim, F.R.G.) by use of 4-(maleimidomethyl)cyclohexane-l-carboxylic acid succinimide ester (Zieben Chemical Co., Tokyo, Japan) (Yamamoto et al., 1981a). The amounts of the conjugate were expressed in terms of units of enzyme activity, and 1 U of activity was defined as that which hydrolized 1 btmol of o-nitrophenyl-fl-D-galactoside/min under the conditions described previously (Yamamoto et al., 1982). The conjugate was found to contain 4 - 5 mol of Tn/mol of the enzyme (Yamamoto et al., 1982a). Antibody-human IgG conjugate Human IgG was coupled to Fab' fragments of anti-T4 antibodies, prepared by reducing F(ab')2 fragments with 2-mercaptoethylamine (Kato et al., 1976), according to the method described previously (Yamamoto et al., 1986). 7 mg of human IgG in 2 ml of 0.1 mol/1 sodium phosphate buffer (pH 7) was mixed with 150 /~1 of 2 g/1 4-(maleimidomethyl)cyclohexane-l-carboxylic acid succinimide ester solution and incubated at 30 ° C for 60 rain. The reaction mixture was then applied to a column of Sephadex G-25 (Pharmacia Fine Chemicals, Uppsala, Sweden) equilibrated with 0.1 mol/1 sodium phosphate buffer, pH 7, to remove unreacted ester. The ester-treated IgG was

incubated at 3 0 ° C for 90 min with Fab' fragments of anti-T4 antibodies (20 mg). The coupling reaction was stopped by addition of 0.1 mol/1 2-mercaptoethylamine (10 ~tl). The reaction mixture was applied to a column of Sephacryl S-300 (Pharmacia Fine Chemicals) equilibrated with 0.1 mol/1 sodium phosphate buffer, pH 7, to separate the antibody-human IgG conjugate from unreacted Fab' or IgG. The molecular weight of the conjugate was determined to about 500000 by chromatography on Sephacryl S-300 column described above.

Anti-human lgG antibodies immobilized on Sepharose 4B We mixed 100 mg of the goat antibodies to human IgG (IgG fractions), in 20 ml of 0.1 m o l / I sodium phosphate buffer (pH 8), with 10 ml of CNBr-activated Sepharose 4B (Pharmacia Fine Chemicals), and stirred at 4 ° C overnight (Yamamoto et al., 1982). The antibodies immobilized on Sepharose 4B were then washed with 0.1 mol/1 Tris-HC1 buffer (pH 8), and stored at 4 ° C . When we used the antibody-immobilized Sepharose 4B as B / F separation column, we washed it with buffer G, described below, and loaded 0.1 ml of it into a microcolumn (4 × 8 mm, with a funnel-shaped buffer reservoir on the top) (Yamamoto et al., 1983). Standard serum The molar concentration of the antibody-human IgG conjugate was calculated from the extinction coefficient at 280 nm (E280 nm= 15.0) and its molecular weight. The value was converted to the weight concentration of human IgG (molecular weight 150000). The content of true anti-T4 antibody-labeled human IgG was determined by passing the conjugate through a column of Ta-immobilized Sepharose 4B. The percentage of that in the conjugate was 1.6%. Then the antibody-human IgG conjugate was diluted in a pooled human serum. Buffer for immunoassay (buffer G) Buffer G was sodium phosphate buffer (10 mmol/1, p H 7) containing 0.3 mol of NaC1, 1 mmol of MgC12, 1 g of bovine serum albumin (Cohn Fraction V, Armour Pharmaceutical Co.,

265 Chicago, IL), 5 g of digested gelatin, and 1 g of NaN 3 per liter. The digested gelatin was prepared by digesting gelatin (Difco Laboratories, Detroit, MI) with protease T 1 (Amano Pharmaceutical Co., Nagoya, Japan) ( K a t o et al., 1980).

1,0 ~I_ .......

I

. . . . . . .

'

........

'

o

Subjects 187 patients with autoimmune thyroid diseases and 58 patients with non-autoimmune thyroid diseases were included in this study. Patients were classified into the following groups according to clinical evaluation, serum TSH, and anti-microsome antibodies (MCHA): MCHA-positive group consisting of hyperthyroidism (n = 124), hypothyroidism (n = 29), and chronic thyroiditis (n = 34); MCHA-negative control group consisting of patients with simple goiter or adenomatous goiter (n = 58).

Immunoassay procedure We diluted standard serum or sample serum 100-fold with buffer G and mixed 100 /~1 of the diluted serum with 500/~1 of buffer G containing the antigen-enzyme conjugate (80 m U / m l ) . After incubating the mixture at 3 7 ° C for 120 rain, 500 #1 of it was applied to the B / F separation column described above. We washed each column twice with 1 ml of buffer G and placed each one on the top of a new test tube (15 x 100 mm). Then we applied 500 /~1 of 8.3 m m o l / 1 o-nitrophenyl-fl-Dgalactoside (dissolved in buffer G) to the column, and allowed the enzyme reaction to proceed at room temperature (25 o C) for 90 min. The reaction was stopped by washing it with 2 ml of 80 m m o l / 1 sodium carbonate. We measured the absorbance of the eluate in the test tube, which contained o-nitrophenol, at 420 nm with a Shimazu Spectrophotometer UV-240.

Results

Fig. 1 illustrates a typical standard curve for our enzyme immunoassay. The difference of the dose responses of fl-D-galactosidase activity was regarded as significant between 0 and 0.3 n g / a s say tube ( P = 0.005). Therefore we regarded the minimum detectable amount of the assay as 0.3 n g / a s s a y tube. The dose responses of the enzyme

$_

o

0.i

0 L', . . . . . . . 0

1

,r i0

Anti-thyroxine antibody (ng/tube)

Fig. 1. Typical standard curve for anti-thyroxine antibody. Anti-thyroxine antibody-human IgG conjugate was used as a calibrator.

activity were obtained at between 0.3 and 10 ng of the antibodies per assay tube. We tested the precision of the assay b y assaying three sera (concentrations of antibodies; 0.50, 1.05 and 1.75 /~g/ml) ten times in one assay (within-run) or in duplicate in seven consecutive assays (betweenrun). The coefficients of variation in each assay were less than 6.9% (6.5, 4.9 and 5.7% in withinrun, and 6.9, 5.8 and 6.5% in between-run respectively). The effect of sample dilution in the assay is shown in Fig. 2. We assayed three different sera diluted with the pooled h u m a n serum. A linear relation between the antibody concentration and the dilution factor was observed. In experiments on analytical recovery of the antibody-human I g G conjugate or that of h u m a n antibodies, we used five specimens of h u m a n sera containing 0.32-1.05 /~g of endogenous antibodies/ml. We assayed each sample in duplicate with and without added antibody-human I g G conjugate (1.2-4.8 /zg/ml) or the h u m a n antibodies ( I g G fractions) purified from serum ( 0 . 9 / l g / m l ) , the antibody concentration of which was determined with the present assay, and estimated the analytical recovery from the standard curve. The average recoveries ranged from 98.1% to 108% (Table I). In order to test the effect of hemoglobin or bilirubin on the assay, we assayed five specimens

266

TABLE II EFFECT OF THYROXINE OR HUMAN IgG ON THE ANTIBODY ASSAY

~2

Sample ro

Absorbance at 420 nm a

1 2 3 4d

=~1 ox 4~ I

0 8

4

i i

2 Dilution

(folds)

Fig. 2. Effect of sample dilution on the assay. Three serum samples were diluted as indicated with a pooled human serum, and the diluted samples were subjected in duplicate to the imrnunoassay.

of human serum containing 0.31-1.75 #g of endogenous antibodies/ml in duplicate with and w i t h o u t a d d e d h u m a n h e m o g l o b i n (5 m g / m l ) o r bilirubin (200 /~g/ml). The hemoglobin and bilirubin did not interfere with the assay. To make sure that the serum component bound t o t h e B/F s e p a r a t i o n c o l u m n w a s t h e a n t i - T 4 antibody, we added T 4 or human IgG to three antibody-positive sera and a serum from a healthy individual, and tested the effect of them on the a s s a y . A s s h o w n i n T a b l e II, t h e b i n d i n g o f t h e antigen-enzyme conjugate was inhibited with T 4 or human IgG. Then, we measured the molecular weight of the serum component by a molecular sieve chromatography o n a T S K gel G 3 0 0 0 S W c o l u m n ( T o y o S o d a Co., T o k y o , J a p a n ) . T h e

No addition

+ Thyroxine b

+ Human lgG c

0.171 0.170 0.236 0,052

0.060 0.076 0.065 0.053

0.057 0.050 0.058 0.049

a The antibodies were assayed according to the method described in the text and the absorbance of the eluate from the column was measured. b 2.5 ~tg of thyroxine (T4) was added to the assay tube. c 5.7 mg of human IgG was added to the assay tube. d Serum from a healthy individual.

m o l e c u l a r w e i g h t o f it w a s a b o u t 160 0 0 0 ( F i g . 3). Therefore, we presumed that we measured IgG w h i c h c o u l d b i n d t o T 4, i.e., a n t i - T 4 a n t i b o d i e s . W e m e a s u r e d a n t i - T 4 a n t i b o d i e s i n 245 p a t i e n t s w i t h t h y r o i d d i s e a s e s . A s s h o w n i n T a b l e III, t h e a n t i b o d y c o n c e n t r a t i o n s i n 58 p a t i e n t s w i t h n o n a u t o i m m u n e t h y r o i d d i s e a s e s ( c o n t r o l s ) w e r e less t h a n 0 . 3 / ~ g / m l e x c e p t f o r t w o s a m p l e s (3.4%). O n t h e o t h e r h a n d , 32 s a m p l e s f r o m t h e p a t i e n t s w i t h

*Molecular weight l

m

-ql, Ferritin (450K)*

e4

,

,

Catalase Aldolase Serum albumin (240K) (16OK) (67K)

Antibody

"--..,

c x

TABLE I ANALYTICAL RECOVERY OF ANTI-THYROXINE ANTIBODY ADDED TO HUMAN SERUM Anti-thyroxine antibody (p g/ml) Added

Recovered

1.2 a 2.4 ~ 4.8 ~ 0.9 b

1.29 5:0.0435 c 2.46 +0.128 4.93 +0.470 0.883 5:0.0441

280nm

o

i

Z

Recovery (%) 108 5:3.63 c 103 +5.33 103 -+9.79 98.1 __.4.90

Antibody-human IgG conjugate. b Human antibodies purified from serum. c M e a n + SD (n = 5). a

2

10

20 Retention time (min)

Fig. 3. Molecular sieve chromatography of a human serum containing anti-thyroxine antibodies. 200 ~1 of the serum (or each marker protein) was applied at a flow rate of 0.8 m l / m i n to a TSK gel G3000SW column (7.5 × 60 mm, Toyo Soda Co.) equilibrated with 0.25 mol/1 potassium phosphate buffer, pH 7.

267 T A B L E III A N T I - T H Y R O X I N E A N T I B O D Y LEVELS IN S E R U M F R O M P A T I E N T S W I T H T H Y R O I D D I S O R D E R Patients

N u m b e r of patients Anti-thyroxine antibody (t~g/ml)

Autoimmune thyroid disease (n =187) Chronic thyroiditis (n = 34) Hyperthyroidism (n = 124) Hypothyroidism (n = 29) Non-autoimmune thyroid disease (n = 58)

Undetectable

Detectable

< 0.3

0.3 < < 1.0

1.0 < < 2.0

2.0 <

25 (13.4) b

5 (2.7) b

2 (1.1) b

155 (82.8)

a

27 (79.5)

6 (17.6)

1 (2.9)

0 (0.0)

104 (83.9)

15 (12.1)

3 (2.4)

2 (1.6)

24 (82.8)

4 (13.8)

1 (3.4)

0 (0.0)

56 (96.6)

2 (3.4) b

0 (0.0)

0 (0.0)

" N u m b e r (and %). b The difference in the n u m b e r of antibody-positive patients between two groups can be regarded as significant with Fisher's test ( P = 0.004).

autoimmune thyroid diseases (17.1%) were contained detectable amounts of the antibodies, and the antibody concentrations in those sera varied widely from 0.31 to 2 . 3 1 / t g / m l . The difference in the number of antibody-positive samples between two groups could be regarded as significant with Fisher's test ( P = 0.004).

Discussion The present enzyme immunoassay is practical, simple, and sensitive for determining the concentration of antibodies in serum. This method is generally applicable to the quantitative determination of antibodies. We have confirmed that antibodies to triiodothyronine, thyroglobulin or insulin can be measured with the same method (data was not shown). The B/F separation column containing 0.1 ml of anti-human I g G antibodies immobilized on Sepharose 4B was found to bind 10-30/~g of I g G ( Y a m a m o t o et al., 1981b). Therefore, we could use a reasonable volume (1 /~l/assay tube) of serum samples for the assay, sufficient to achieve a sensitive assay of antibodies in serum. The antibodies immobilized on Sepharose 4B, the antigen-enzyme

conjugate and the antibody-human I g G conjugate are stable at 4 ° C for at least 1 year. In Table II, the absorbance of sample 2 was not completely suppressed by addition of T 4, while those of other sample were suppressed completely. This result may related to the variety of affinity constants of the autoantibodies. We did not estimate the affinity constant of each sample in the present study. The effect of the affinity constant of autoantibodies on the assay and that of the calibrator are forthcoming subjects to be investigated. At the same time, we must investigate the effect of endogenous T 4 in sample serum, while we added 3 x 10 -12 mol of T 4 as antigen-enzyme conjugate (about 100000-fold of free T 4 or 15-fold of total "1"4) to the reaction mixture, and the difference in the number of antibody-positive sera between a Ta-low group (below 150 /~g/1) and a T 4 - h i g h group (over 150 /~g/1) in patients with Graves' disease or between euthyroid patients and T4-high patients with Graves' disease could not be regarded as significant with Yates-modified X 2 test. In our investigation, the anti-T 4 antibodies were detected in about 17% in the patients with autoimmune thyroid diseases, and the percentages of the antibody-positive sera were comparable to each

268

other among the autoimmune thyroid diseases: chronic thyroiditis, hyperthyroidism, and hypothyroidism. It is an interesting subject that the relationship among anti-T4 antibodies and other autoantibodies. The relationship between the anti-thyroid hormone antibodies and anti-thyroglobulin antibodies was discussed by several investigators (Byfield et al., 1985; Sakata et al., 1986). However, in our preliminary experiment, the concentration of anti-T4 antibodies did not correlate with that of anti-thyroglobulin antibodies, which was determined with the present method (data was not shown). There is now considerable evidence that the hyperthyroidism of Graves' disease is due to antibodies to TSH receptor (Smith, 1981). And Hashimoto's thyroiditis is an autoimmune disease in which antibodies to several components of thyroid are present (Roitt et al., 1956). However the mechanism of the production of antibodies to thyroid hormones in the body and its effect on metabolism are still not completely clarified. The present method is useful for investigations concerning autoantibodies and autoimmune diseases.

References Belyavin, G. and Trotter, W.R. (1959) Investigation of thyroid antigen reacting with Hashimoto sera. Evidence for an antigen other than thyroglobulin. Lancet ii, 648. Berson, S.A. and Yalow, R.S. (1959) Quantitative aspects of the reaction between insulin and insulin-binding antibody. J. Clin. Invest. 38, 1996. Byfield, P.G.H., Clingan, D. and Himsworth, R.L. (1985) Structural basis for the reaction of 3,5,3'-triiodothyroninespecific antibodies with thyroxine-containing thyroglobulin. Biochem. J. 228, 155. Kato, K., Fukui, H., Harnaguchi, Y. and Ishikawa, E. (1976) Enzyme-linked immunoassay: Conjugation of the Fab' fragment of rabbit IgG with fl-D-galactosidase from E. coli and its use for immunoassay. J. Immunol. 116, 1554.

Kato, K., Umeda, Y., Suzuki, F. and Kosaka, A. (1979) Interference in a solid-phase enzyme immunoassay system by serum factors. J. Appl. Biochem. 1,479. Kato, K., Umeda, Y., Suzuki, F. and Kosaka, A. (1980) Improved reaction buffers for solid-phase enzyme immunoassay without interference by serum factors. Clin. China. Acta 120, 261. Robbins, J., Rail, J.E. and Rawson, R.W. (1956) An unusual instance of thyroxine-binding by human serum gamma globulin. J. Clin. Endocrinol. Metab. 16, 573. Roitt, I.M., Doniach, D., Campbell, P.N. and Hudson, R.V. (1956) Auto-antibodies in Hashimoto's disease (lymphadenoid goiter). Lancet ii, 820. Sakata, S., Nakamura, S., Komaki, T., Matsuda, M., Kojima, N., Tarutani, O. and Miura, K. (1986) Autoimmune recognition of thyroglobulin and thyroid hormones in rabbits. J. Protein Chem. 5, 407. Smith, B.R. (1981) Thyrotropin Receptor Antibodies. Receptors and Recognition, Series B, Vol. 13. Chapman and Hall, London. Smith, B.R. and Hall, R. (1981) In: J.J. Langone and H. Van Vunakis (Eds.), Methods in Enzymology, Vol. 74, Part C, Measurement of Thyrotropin Receptor Antibodies. Academic Press, New York, p. 405. Yamamoto, R., Hattori, S., Inukai, T., Matuura, A., Yamashita, K., Kosaka, A. and Kato, K. (1981a) Enzyme immunoassay for thyroxine and triiodothyronine in human serum, with use of a covalent chromatographic separation method. Clin. Chem. 27, 1721. Yamamoto, R., Umeda, Y., Kosaka, A. and Kato, K. (1981b) Enzyme immunoassay for antibodies in serum using a covalent chromatographic method for separation of the bound label. J. Biochem. 89, 223. Yamamoto, R., Hattori, S., Matuura, A., Naka, Y., Kosaka, A. and Kato, K. (1982) Enzyme immunoassay for thyroxine and triiodothyronine with an improved column-separation method. J. Appl. Biochem. 4, 168. Yamamoto, R., Kimura, S., Hattori, S., Ishiguro, Y. and Kato, K. (1983) Column enzyme immunoassay for secretory immunoglobulin A in serum. Clin. Chem. 29, 151. Yamamoto, R., Kimura, S., Matuura, A., Fukuda, Y., Hayakawa, T. and Kato, K. (1986) Two-site enzyme immunoassay for fetoprotein involving column chromatography. J. Immunol. Methods 87, 197.