A new assay for thyroglobulin concentration in serum using monoclonal antibodies against synthetic peptides

Clinica Chimica Acta 298 (2000) 69–84 www.elsevier.com / locate / clinchim

A new assay for thyroglobulin concentration in serum using monoclonal antibodies against synthetic peptides a, b a b Ryoji Kato *, Masayuki Maruyama , Tetsuo Sekino , Yoshio Kasuga a

Shinshu University, School of Allied Medical Science, 3 -1 -1 Asahi, Matsumoto 390 -8621, Japan b Department of Surgery, Shinshu University School of Medicine, Matsumoto 390 -8621, Japan

Received 5 October 1999; received in revised form 15 February 2000; accepted 29 February 2000

Abstract The concentration of thyroglobulin (Tg) measured by radioimmunoassay (RIA) or enzymelinked immunosorbent assay (ELISA) is greatly affected by the presence of anti-Tg autoantibodies in sera. We developed a new assay for detecting Tg in the presence of high concentrations of anti-Tg autoantibodies. A 48-kDa fragment was purified from Tg after treatment with V8 protease. This fragment did not appear to bind to two types of monoclonal antibodies (57Ab and 28D3) against a peptide in the C-terminus (amino acids 2735–2748) of Tg and intact Tg, respectively, by ELISA and Western blot analysis. In contrast, anti-Tg autoantibody or anti-Tg polyclonal antibody reacted well with this fragment. Our new ELISA used 57Ab as a solid phase antibody and 28D3 as a antibody conjugated to horseradish peroxidase. Buffer containing purified 48-kDa fragment was used to neutralize autoantibodies against Tg. With this assay, the recovery of Tg was 84.0–89.6% in normal healthy donors (n 5 5) in the presence of immunoglobulin G (IgG) purified from sera positive for anti-Tg autoantibody, and 76.2–104.4% in patient sera Grave’s disease (n 5 15). Furthermore, the Tg concentrations in sera from patients with Grave’s disease (n 5 20) ranged from 25 to 526 ng / ml, even though the Tg concentration, as measured by a commercial RIA did not exceed 55 ng / ml. There was good agreement between Tg concentrations measured by new Tg-ELISA and commercial Tg-RIA in sera that were negative for anti-Tg autoantibody. Overall, Abbreviations: DAB, 3,39-diaminobenzidine; ELISA, enzyme-linked immunosorbent assay; HRP, horseradish peroxidase; PVDF, polyvinylidene difluoride; P-MAb(s), anti-peptide monoclonal antibody (ies); P-PAb(s), anti-peptide polyclonal antibody (ies); Tg, thyroglobulin; TgF, 48-kDa fragment of Tg; Tg-AAb(s), antithyroglobulin human autoantibody (ies); Tg-MAb(s), anti-thyroglobulin monoclonal antibody (ies); Tg-PAb(s), anti-thyroglobulin polyclonal antibody (ies); TMB, 3,39,5,59-tetramethylbenzidine; V8, Staphylococcus aureus V8 protease *Corresponding author. Tel.: 1 81-263-372-390; fax: 1 81-263-372-370. E-mail address: [email protected] (R. Kato) 0009-8981 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0009-8981( 00 )00258-8

70

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

our new ELISA containing a Tg fragment to neutralize the presence of autoantibodies, showed good sensitivity and precision, and may be useful for routine use. Further investigations with the new assay should allow wider assessment of the prevalence and pattern of thyroid autoimmunity or thyroid neoplasms.  2000 Elsevier Science B.V. All rights reserved. Keywords: Thyroglobulin peptide; Thyroglobulin ELISA; Monoclonal antibody; Anti-thyroglobulin autoantibody

1. Introduction Thyroglobulin (Tg) is one of the major thyroid proteins with an apparent molecular mass of 660 kDa. High Tg concentrations indicate thyroid cancer [1–5]. Radioimmunoassay (RIA) [1–11] or enzyme immunoassays (EIA) [12] have been used for measuring Tg. However, these assays are affected by the presence of anti-Tg autoantibodies in the sera. Assays which attempt to remove the interference of anti-Tg autoantibodies using anti-Tg monoclonal antibody can detect Tg only in the presence of low concentrations of anti-Tg autoantibodies [12–15]. Consequently, we developed a new assay for measuring Tg in the presence of high concentrations of anti-Tg autoantibodies. This assay uses a Tg fragment and monoclonal antibodies produced by immunization with synthetic.

2. Materials and methods

2.1. Subjects Sera containing anti-Tg autoantibodies were obtained from 90 patients with Grave’s disease, 20 with Hashimoto’s disease, 30 with thyroid cancer, and 15 with benign thyroid tumors. Sera without autoantibodies were obtained from 10 patients with Grave’s disease, 10 patients with thyroid cancer, and 20 normally healthy volunteers. All subjects were informed of the study and consented to participate. Tg or anti-Tg autoantibody concentrations in the sera were measured by either a passive particle agglutination assay for thyroglobulin (Serodia-ATG, Fujirebio, Japan), ELISA [12] or RIA (Eiken Chemical, Japan), according to the manufacturer’s recommended procedures.

2.2. Synthesis of thyroglobulin peptides The amino acid sequences of synthetic peptides were chosen to represent the most antigenic or functionally important regions of Tg by the Super Mirror

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

71

Fig. 1. Production of synthetic peptides on thyroglobulin. Eight different peptides corresponding to amino acid residues 1–15, 134–152, 1000–1018, 1474–1492, 1824–1842, 2561–2579, and 2735–2748 were synthesized by the Multiple Antigen Peptide System (MAP method). No of amino acids on Tg molecule.

method [16]. Eight different peptides corresponding to amino acid residues 1–15, 134–152, 1000–1018, 1474–1492, 1824–1842, 2561–2579, and 2735– 2748 were synthesized (Fig. 1) by the Multiple Antigen Peptide System (MAP method) [17].

2.3. Preparation of Tg from surgically obtained thyroid tissue A total of 100 g of fresh thyroid tissue was obtained with informed consent from surgically removed thyroid glands of patients with Grave’s disease. The samples were kept frozen at 2 808C or in liquid nitrogen until use. The Tg was purified from the tissue according to the method of Shulman and Armenia [18]. Briefly, the thyroid tissue was homogenized in PBS (50 mmol / l phosphatebuffered saline, pH 7.2), containing 1 mmol / l phenylmethylsulfonyl fluoride (PMSF, Sigma, USA), and then centrifuged at 100 000 3 g for 60 min at 48C. The supernatant was mixed with an equal volume of saturated ammonium sulfate and incubated for 30 min at 48C, then centrifuged at 10 000 3 g for 30 min at 48C. The resulting pellet containing the Tg was dissolved in PBS and dialyzed against PBS overnight at 48C. Macromolecular fractions were further purified by the Fast Performance Liquid Chromatography (FPLC) method using SuperdexE 75 column (Pharmacia LKB, Sweden). Finally, the Tg was purified by absorbing a small amount

72

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

of contaminating IgG using a Protein A Sepharose 4B column (Pharmacia LKB, Sweden). The purified Tg was quantitated by an enzyme-linked immunosorbent assay (ELISA) [12] and then used as the antigen for the immunization of mice to produce monoclonal antibodies.

2.4. Digestion of Tg with V8 protease Staphylococcus aureus V8 protease (Sigma) was dissolved at 400 mg / l in 0.1 mol / l Tris–HCl, pH 7.9. One hundred ng of the Tg in 100 ml of PBS were mixed with 10 ml of the protease solution and incubated at 378C for 1 h. The reaction was stopped by the addition of 3,5-isochumaline (3,5-DIC, Sigma). A 48-kDa fragment (TgF) was purified from the total Tg fragments (V8-Tg) by preparative SDS–PAGE using a 10% slab gel. Some TgF were also purified by FPLC using a SuperdexE 75 column (Pharmacia LKB).

2.5. Production of anti-Tg murine polyclonal and monoclonal antibodies Balb / C mice were immunized with either purified intact Tg or synthetic peptides each representing a different antigenic region of Tg. They were given booster injections at 3-week intervals until the concentrations of anti-thyroglobulin polyclonal antibody(Tg-PAb) or anti-synthetic peptides polyclonal antibody(P-PAb) in sera were sufficiently high. The production and cloning of hybridomas were performed according to the previously reported method [19,20]. Selected hybridoma clones were inoculated into the peritoneal cavity of Balb / C mice and monoclonal antibodies were purified from their ascites by affinity chromatography using a Protein A Sepharose 4B column (Prosep A, Pharmacia LKB).

2.6. Measurement of the antibody titers by ELISA Microtiter plates (Nunc, Denmark) were coated with either 5 mg / well of purified Tg or 200 ng / well of one of the synthetic peptides in PBS at 48C overnight. The wells were blocked with 1% fetal calf serum (FCS) in PBS (FCS buffer) at 378C for 1 h. One hundred ml of FCS buffer and 50 ml of the anti-peptide monoclonal antibodies (P-MAb), anti-Tg monoclonal antibodies (Tg-MAb), anti-Tg human autoantibodies (Tg-AAb), or serially diluted Tgstandards were added together to the wells and incubated at 258C for 30 min. The plates were washed 3 times with 0.05% Tween 20 in PBS (washing buffer) and then incubated with 1:10 4 dilution of horseradish peroxidase (HRP)conjugated anti-mouse IgG (Dako A / S, Denmark) or anti-human IgG (Dako) at 258C for 30 min. After three washes with washing buffer, 100 ml of 3,39,5,59tetramethylbenzidine (TMB, Kirkegaad & Perry Laboratory, USA) were added

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

73

to each well and incubated at 258C for 30 min. Finally, 50 ml of 1 mol / l phosphoric acid (H 3 PO 4 ) were added to stop the reaction and the optical density at 490 nm was measured on an ELISA-Auto-Reader (Sanko Junyaku, Japan) [21]. The optical densities of the blank wells were always less than 0.05 and were subtracted from those of the samples.

2.7. Epitope analyses of monoclonal antibodies specific for Tg by competitive ELISA Microtiter plate wells were coated with a small amount of Tg (5 ng / well) and reacted with the excess amount of unconjugated anti-Tg monoclonal antibodies or patients’ sera containing anti-Tg autoantibodies at 258C for 1 h. One hundred ml of HRP-conjugated P-MAb or Tg-MAb were added to the wells containing unconjugated antibodies and incubated at 258C for 1 h. After washing, the color was developed and recorded as described above. Inhibition ratios (%) regarding anti-Tg monoclonal antibodies were calculated as follows: OD 490 with unconjugated antibodies / OD 490 without unconjugated antibodies 3 100 (%). For patients whose sera were positive for anti-Tg autoantibodies, the results were classified into four groups according to their inhibition ratios (%) as follows: groups A ( , 10), B (10–30), C (30–50) and D (50 . ).

2.8. Measurement of Tg by ELISA Microtiter plates were coated with 5 mg / well of purified 57Ab(P-MAb) in PBS at 48C overnight. The wells were blocked with FCS buffer at 48C overnight. One hundred ml of FCS buffer and 50 ml of each sample were added together into the wells and incubated at 48C for overnight. The plates were washed 3 times with washing buffer and then incubated with HRP-conjugated 28D3(Tg-MAb) at 258C for 60 min. After three washes with washing buffer, 100 ml of TMB were added to each well and incubated at 258C for 30 min. Finally, 50 ml of 1 mol / l H 3 PO 4 were added to stop the reaction and the optical density at 490 nm was measured on an ELISA-Auto-Reader.

2.9. Measurement of Tg by commercial RIA Fifty ml of samples, 200 ml of 125 I-labelled Tg-MAb and one of the anti-TgAb-coated beads were added together into the tubes, mixed with vortex mixer, and then incubated at 258C overnight. The tubes were washed 3 times with 1 ml of washing buffer and the isotope activities were measured by scintillation counter (Aloka, Japan).

74

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

2.10. Absorption of the antibodies with the 48 kDa Tg fragment Fifty ml each of various concentrations of P-MAb, Tg-MAb, or Tg-AAb in FCS buffer were added to the Tg-coated wells (5 ng / well) of a microtiter plate and incubated with 100 ml of buffer containing 1 mg / ml of the TgF for 1 h at 258C. After several washings with a washing buffer, the plates were incubated with 1:10 4 dilution of HRP-conjugated P-MAb, Tg-MAb, or Tg-AAb for 1 h at 258C. After washing, TMB was added to each well and the optical density (OD490) was measured as described above.

2.11. Western blotting of Tg and V8 -Tg The Tg fragments generated by V8 protease digestions were electrophoresed on a 12% slab gel under non-reducing conditions [22]. The separated proteins were electrotransferred to PVDF membrane (Bio-Rad, USA) at 48C [23]. The membrane was blocked at 258C for 30 min in 1:5 diluted BlockAce (Dainihon Seiyaku, Japan) in PBS. The membrane was washed three times with washing buffer. The membrane was cut into strips and incubated with a 5 3 10 2 2 10 3 dilution of anti-Tg murine monoclonal antibodies or 10 2 dilution of the patients’ sera containing Tg-AAb at 378C overnight. After washing 3 times, the membrane was incubated at 378C for 1 h with a 2 3 10 3 dilution of either HRP-conjugated rabbit anti-mouse IgG (Kirkegaad & Perry Laboratory) or HRP-conjugated protein A (Kirkegaad & Perry Laboratory). After the final four washes, diaminobendizene (DAB; Nakarai Kagaku, Japan) and H 2 O 2 were added as the substrate for HRP.

2.12. Analysis of the N-terminal amino acid sequence of the 48 -kDa TgF The 48-kDa TgF was electrophoresed on a 12% slab gel under non-reducing conditions. The separated 48-kDa TgF were electrotransferred to PVDF membrane, which were stained with Coomassie brilliant blue. The band, possibly corresponding to the 48-kDa TgF on membrane was cut out and analysed by amino acid sequencer (Hitachi, Japan)

3. Results

3.1. The characteristics of anti-peptide antibodies ( P-PAb and P-MAb) properties of peptide antibodies Eight types of P-PAb were obtained by immunization using all synthetic peptides. Anti-2179 antibody reacted strongly with Tg and the synthetic peptide

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

75

Table 1 The characteristics of anti-peptide polyclonal antibodies against peptides (P-Pab) based on an analysis of reactivity to Tg and peptides by ELISA P-Pab

Antigen

Anit-2177 Anti-2178 Anti-2179 Anti-2180 Anti-2181 Anti-2182 Anti-2183 Anti-2184 a

Tg

Synthetic peptides

0.307 , 0.010 1.926 a , 0.0010 0.013 0.144 1.606 , 0.010

0.247 0.383 1.561 a 0.030 , 0.010 1.514 0.634 0.658

Anti-2179 antibody reacted strongly with both Tg and synthetic peptides (2179).

(Table 1). Six types of P-MAb were obtained by immunization using 2179 synthetic peptide. Among P-MAbs, 57Ab reacted strongly with the Tg and the synthetic peptide. The subclass of 57Ab was IgG2b.

3.2. Analysis of epitope on Tg by inhibition assay on ELISA The reactivity of 57Ab to Tg was not inhibited by Tg-MAb including 28D3 (1.5–7.5%), while 57Ab strongly inhibited the reaction between 57Ab (91.3%) and Tg. Furthermore 57Ab exhibited no inhibition to each between Tg-MAbs (0.7–8.3%) and Tg (Table 2). Table 3 shows the epitope analysis on Tg by the inhibition assay using sera positive for anti-Tg autoantibody on ELISA, Table 2 Epitope analysis on Tg by the inhibition assay on ELISA using monoclonal antibodies (M-Ab) Unconjugated M-Ab

57Ab 29E6 108F5 76C6 78E9 86D4D3 88F4 74E7 28D3 a b

Conjugated (M-Ab)a 57Ab

29E6

108F5

76C6

78E9

86D4D3

88F4

74E7

28D3

91.3 b 4.2 1.5 5.0 4.9 3.7 7.5 5.0 3.5

6.3 94.2 b 2.0 4.9 8.9 22.3 28.5 44.0 6.0

0.7 0.2 97.5 b 1.0 0 1.0 2.5 5.0 3.8

4.3 2.2 3.5 91.1 b 8.2 5.3 1.5 4.3 NT

1.9 1.7 3.8 0 94.9 b 3.6 22.5 4.1 NT

4.2 1.2 0.7 1.7 2.5 98.2 b 17.3 7.4 NT

8.3 10.9 8.2 6.5 7.1 33.7 97.5 b 14.5 NT

0.9 12.8 0.4 7.4 7.9 11.4 13.8 98.1 b NT

1.7 7.9 5.0 NT NT NT NT NT 95.9 b

Inhibition ratio (%) 5 (OD 490 with unconjugated M-Ab / without unconjugated M-Ab) 3 100. Strong inhibition. NT, not tested.

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

76

Table 3 Epitope analysis on Tg by the inhibition assay on ELISA using anti-Tg autoantibodies (Tg-Aab) MAb

P-Mab Tg-Mab

a

Name

57Ab 28D3 29E6 86D4D3 108F5 76C6 74E7 88F4 78E9

Inhibition ratio (%)a

No. of cases with more than 10% inhibition

, 10

11–30

31–50

. 50

117 116 114 110 105 95 91 90 74

3 4 6 10 15 25 24 21 38

0 0 0 0 0 0 5 6 7

0 0 0 0 0 0 0 3 1

Inhibition ratio (%) 5 (OD 490 autoantibodies) 3 100.

3 (2.5%) 4 (3.3%) 6 (5.0%) 10 (8.3%) 15 (12.5%) 25 (20.8%) 29 (24.1%) 30 (25.0%) 46 (38.3%)

with anti-Tg autoantibodies / OD490 without anti-Tg

demonstrating marked inhibition. The reactivity of Tg with 57Ab(P-MAb), 28D3 or 29E6(Tg-MAb) was not significantly inhibited.

3.3. Analysis of epitope on Tg by Western blot The reactivities of the intact Tg or Tg treated with V8 protease (V8-Tg) with each monoclonal antibody were analyzed by Western blot. Of them, 57Ab of P-MAb and 28D3 of Tg-MAb did not react to the 48-kDa TgF. The 660-kDa protein, that was identified to be Tg by protein staining, reacted well with each antibody (Fig. 2). Fig. 3 showed that a 48-kDa band representing a Tg fragment reacted well with Grave’s disease sera and Hashimoto’s sera. The Tg-AAb titer in sera was 10 3 2 2 – 6 for Grave’s disease (n 5 5) and 10 3 2 7 – 18 for Hashimoto’s disease (n 5 5), respectively. In the case of the 36- or 30-kDa fragments, these bands were identified as having a high Tg-AAb titer for Hashimoto’s sera.

3.4. Analysis of N-terminal amino acid sequence of the 48 -kDa TgF The amino acid sequence of N-terminus of the 48-kDa TgF was consistent with the sequence in the middle position (amino acids 1112–1121) of Tg molecule (Thy–Ala–Arg–Leu–Glu–Ala–Ser–Gly–Ala–Gly–).

3.5. The reactivity of peptide antibodies with 48 -kDa TgF on ELISA TgF (inhibition ratio of 87.7% for Tg-AAb and 78.2% for 76C6) significantly inhibited the reactivity of Tg with Tg-AAb or 76C6 (Tg-MAb). However, the

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

77

Fig. 2. Western blot analysis for the reactivity of monoclonal antibodies with Tg or V8-Tg. Intact Tg and Tg treated with V8 protease (V8-Tg) were blotted onto nitrocellulose membrane using protein staining (lane V8), reactivity with 57Ab of P-MAb (lane 1), Tg-Pab (lane 2), 28D3 of Tg-Mab (lane 3), 108F5 of Tg-MAb (lane 4), 70E4 of TgMAb (Lane 5), and 76C7 of Tg-MAb (lane 6). A 48-kDa band and small fragment bands reacted well with each Tg-MAb except for 57Ab of P-MAb (Lane 1, 57Ab with P-MAb). Lane Mk, molecule standard marker.

48-kDa TgF did not appear to inhibit the reactivity between the Tg and 57Ab (Table 4).

3.6. Tg measurement by ELISA employing a new determination system The coefficient of variation in this assay is shown in Table 5. The lower limit of detection for this ELISA was approximately 10 ng / ml for the Tg standard solutions after reaction for 1 h at 258C. There was no interference of triiodothyronine (T3, up to 100 ng / ml), thyroxine (T4, up to 10 mg / ml) or thyroid-stimulating hormone (TSH, up to 1 mU / ml) to the new Tg-ELISA. There was a good correlation (r 5 0.98) between the new Tg-ELISA and the commercial Tg-RIA in sera without Tg-AAb (Fig. 4). On the contrary, a poor correlation (r 5 0.49) was observed when sera with Tg-AAb were analyzed (Fig.

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

78

Fig. 3. Western blot analysis for the reactivity of anti-Tg autoantibodies with Tg or V8-Tg. A 48-kDa band representing a Tg fragment reacted well with Grave’s disease sera and Hashimoto’s sera. In the case of the 36- or 30-kDa fragments, these bands were identified as high titer of TgAb for Hashimoto’s sera. Lane 1, Grave’s disease (anti-TgAb titer 10 3 2 6 ); lane 2, Hashimoto’s disease (anti-TgAb titer 10 3 2 9 ); lane 3, 70E4 with Tg-MAb (anti-TgAb titer 10 3 2 18 ). V8, Tg treated with V8 protease; Tg, intact Tg; Mk, molecule standard marker.

Table 4 The reactivity of peptide antibodies with TgF on ELISA Antibodies

57Ab Tg-AAb b Tg-MAb(76C6)b Control (buffer) a b

OD values

Inhibition ratio (%)a

2 TgF

1 TgF

0.375 0.398 0.357 0.030

0.388 0.049 0.078 0.038

Inhibition ratio (%) 5 100 2 (added TgF / non-added TgF) 3 100. Tg-Aab, anti-Tg autoantibody; Tg-Mab, anti-Tg monoclonal antibody.

3.5 87.7 78.2 –

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

79

Table 5 Intraassay and interassay reproducibility of the new Tg-ELISA

L M H

Intra-assay variation (n 5 20)

Inter-assay variation (n 5 20)

Mean6S.D.

CV%

Mean6S.D.

CV%

15.161.8 69.364.8 215.469.5

11.9 6.9 4.4

13.862.4 67.965.8 219.1612.1

17.9 8.5 5.5

4B), indicating that the new Tg ELISA system could detect Tg concentrations which were not detected by the commercial Tg RIA. In the presence of immunoglobulin G (IgG) purified form of sera positive for Tg-AAb, the Tg concentrations showed 84.0–89.6% of variation in the case of normal healthy donors (n 5 5) and 76.2–104.4% in the case of Grave’s disease (n 5 15), suggesting that anti-Tg antibody had little effect on this Tg determination system (data not shown). Consequently, the Tg measured by this method in sera from Grave’s disease (n 5 20) ranged from 25 to 526 ng / ml despite results of no more than 55 ng / ml by the commercial Tg-RIA. The results of the Tg measurement using the new Tg-ELISA in sera from patients with thyroid cancer accompanied by distant metastases or patients with thyroid carcinoma without distant metastasis both before and after operations are shown in Fig. 5. For patients without metastases, the decreasing concentration of Tg was clearly evident after operations, whereas

Fig. 4. Correlation between new Tg-ELISA and commercial RIA. (A) Good correlation (r 5 0.98) was observed between the new Tg-ELISA and the commercial Tg-RIA in samples from thyroid cancer and Grave’s disease in sera without Tg-AAb. (B) Less correlation (r 5 0.49) between them was observed when sera with Tg-AAb were analyzed.

80

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

Fig. 5. Tg concentrations both before and after operation for thyroid cancer. In patients without metastases, the decreasing concentration of Tg was clearly evident after operations, whereas a marked change was observed before and after surgery in only three patients with metastases.

a marked change was observed before and after surgery in only three patients with metastases.

4. Discussion Assays for Tg in sera have varied since Hjort [24] first used passive particle agglutination in 1961. It has been reported that the measurement of Tg was most useful in the case of thyroid cancer when the sera were analyzed from patients with distant metastases after the surgical operation [3–7,25]. To date, many attempts have been made without success to overcome the difficulty of detecting Tg in the presence of autoantibody against Tg [12–15,26– 34]. Therefore, the aim of the present study was to develop a new method for detecting Tg in the presence of anti-thyroglobulin autoantibody. First, eight different regions (15–20 amino acids) including hydrophobic parts

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

81

of Tg were selected for synthesizing peptides by Super Mirror Analysis [16]. These peptides were then used to immunize mice to obtain polyclonal or monoclonal antibodies. Six anti-peptide monoclonal antibodies were acquired from one of eight peptides (2179). Of these, 57Ab was used in the present study for the following reasons: (1) 57Ab reacted well with intact Tg; (2) 57Ab did not influence the reaction of Tg and Tg-AAb by ELISA; (3) 57Ab did not appear to bind to three of the fragments of the Tg (48, 36 and 30 kDa) that bound well to the Tg-AAb by Western blot analysis; and (4) 57Ab recognizes the C-terminus on the primary amino acid sequence of the Tg molecule. The reason for the less reactivity between 57Ab and Tg after treatment with V8 protease is unclear at present. We can only speculate that the C-terminus of Tg, which may be an epitope for 57Ab, disappeared from the Tg as a smaller fragment and lost its antigenicity to 57Ab. The 28D3 as Tg-MAb were also selected for the labelled antibody in Tg ELISA for the same reason as 57Ab. Furthermore, the epitopes recognized by Tg-AAb might be situated in the middle position of Tg, where 57Ab or 28D3 could not bind. On the other hand, it may be evident from the epitope analysis in Table 2 that the epitopes recognized by 57Ab appeared to be different from those by 28D3. To clarify these observations by Western blot, neutralization studies were carried out using the 48-kDa fragment (TgF) purified from V8-Tg by FPLC. The addition of the TgF of Tg did not neutralize the 57Ab but anti-Tg autoantibodies did. These results suggest that the Tg-AAb had some ability to recognize not only the conformational structures in Tg but, to a small extent, some restricted regions on Tg larger than the peptide length. This is in agreement with the findings reported by Nye et al. [35] who suggested that the reactivity of Tg-AAb with Tg after treatment with DTT studied by the immunoblotting method was weaker than the reactivity of intact Tg. Our findings also support the concept that thyroglobulin after treatment with trypsin [33], plasmin [33], or V8 protease [36] reduced its antigenicity to Tg-AAb. On the basis of these preliminary experiments, we have designed a new assay for measuring Tg employing an ELISA with 57Ab as solid phase and 28D3 as a conjugate in the presence of purified TgF of Tg (1 mg / ml) in buffers in the first phase of this assay. The precision of this assay was good, with inter- and intra-assay coefficients of variation of no more than 10%. Using the sera from patients with thyroid cancer after total thyroidectomy, we could detect significantly increased concentrations (P , 0.05) of Tg in the sera from patients with distant metastases in comparison with patients without distant metastases for our assay based on a Tg fragment. In addition, there was good agreement between the Tg results in our assay and the commercial assays (r 5 0.977, n 5 25) in sera without anti-Tg autoantibody. Furthermore, to evaluate the effect of the Tg-AAb on our assay, the Tg concentrations were measured on the serum, negative for Tg-AAb, which was mixed with IgG-purified sera showing a high

82

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

titer of Tg-AAb from a patient with Grave’s disease in different dilutions. The Tg concentrations in the sera were not inhibited by the addition of the Tg-AAb, resulting in confirming the usefulness of our ELISA method. Similar results were obtained using the sera from patients with autoimmune thyroid disease containing Tg-AAb. Consequently, an assay is now available that will allow easy, reliable, routine assessment of Tg concentrations in various kinds of thyroid disorders such as auto-immune thyroid disease (AITD), subacute thyroiditis or thyroid cancer. This type of assessment is particularly important in AITD with high concentrations of Tg-AAb in the sera, as Tg in these sera could not have been detected by previous assays due to the interference of Tg-AAb [30–32]. Further investigations of the prevalence of Tg in the presence of autoantibody against Tg are necessary.

Acknowledgements The authors wish to thank Professor N. Yoshioka (Akita University, Japan) for his critical review and helpful suggestions regarding our manuscript. We are also grateful to Professor R. Dunstan (Curtin University, Perth, WA) for his generous assistance with the English in the manuscript.

References [1] Torrigiani G, Doniach D, Roitt IM. Serum thyroglobulin concentrations in healthy subjects and in-patients with thyroid disease. J Clin Endcrinol Metab 1969;29:305–14. [2] Van Herle AJ, Uller RP, Matthews NL, Brown J. Radioimmunoassay for measurement of thyroglobulin in human serum. J Clin Invest 1973;52:1320–7. [3] Van Herle A, Uller RP. Elevated serum thyroglobulin; a marker of metastases in differentiated thyroid carcinomas. J Clin Invest 1975;56:272–7. [4] Pancini F, Pinchera A, Giani C, Grasso L, Doveri F, Bascheri L. Serum thyroglobulin in thyroid carcinoma and other thyroid disorders. J Endocrinol Invest 1980;3:283–92. [5] Panza N, Lombardi G, DeRosa M, Pacilio G, Lapenta L, Salvatore M. High serum thyroglobulin concentrations. Cancer 1987;60:2233–6. [6] Schaadt B, Feldt-Rasmussen U, Rasmusson B et al. Assessment of the influence of thyroglobulin (Tg) autoantibodies and other interfering on the use of serum Tg as tumor marker in differentiated thyroid carcinoma. Thyroid 1995;5:165–70. [7] Ladenson PW. Optimal laboratory testing for diagnosis and monitoring of thyroid nodules, goiter, and thyroid cancer. Clin Chem 1996;42:183–7. [8] Ochi Y, Hachiya T, Yoshimura M, Miyazaki T. Radioimmunoassay for estimation of thyroglobulin in human serum. Endocrinol Japon 1975;22:351–6. [9] Schneider AB, Pervos R. Radioimmunoassay of human thyroglobulin: effect of antithyroglobulin autoantibodies. J Clin Endocrinol Metab 1978;47:126–37.

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

83

[10] Bodlaender P, Arjonilla JR, Sweat R, Twomey SL. A practical radioimmunoassay of thyroglobulin. Clin Chem 1978;24:267–71. [11] Bayer MF, Kriss JP. Immunoradiometric assay for serum thyroglobulin: semiquantitative measurement of thyroglobulin in antithyroglobulin positive sera. J Clin Endocrinol Metab 1979;49:557–64. [12] Kato R, Noguchi S, Noguchi A. A human serum thyroglobulin determination with monoclonal antibody one-step assay: minimum interference of autoantibody. Endocrinol Japon 1987;34:171–8. [13] Piechaczyk M, Bldet L, Pau B, Bastide JM. Novel immunoradiometric assay of thyroglobulin in serum with use of monoclonal antibodies selected for lack of cross-reactivity with autoantibody. Clin Chem 1989;35:422–4. [14] Schulz W, Benker G, Bethyser H, Stempka L, Hyner M. An autoimmuno-dominant thyroglobulin epitope characterized by a monoclonal antibody. J Endocrinol Invest 1992;15:25–30. [15] Marquet PY, Daver A, Sapin R et al. Highly sensitive immunoradiometric assay for serum thyroglobulin with minimal interference from autoantibodies. Clin Chem 1996;42:258–62. [16] Robson B, Suzuki E. Conformational properties of amino acid residues in globular proteins. J Mol Biol 1996;107:327–56. [17] Tam JP, Lu YA. Vaccine engineering; enhancement of immunogenicity of synthetic peptide vaccines related to hepatitis in chemically defined models consisting of T and B cell epitopes. Proc Natl Acad Sci USA 1989;86:9084–8. [18] Shulman S, Armenia J. Studies on thyroid proteins. The components of hog thyroid tissue, and the preparation of purified thyroglobulin by column chromatography. J Biol Chem 1963;238:2723–31. [19] Reading CL. Theory and methods for immunization in culture and monoclonal antibody production. J Immunol Methods 1982;53:261–91. ¨ [20] Kohler G, Milistein E. Continuous culture of fused cells secreting antibodies of predefined specificity. Nature 1975;256:495–7. [21] Kato R, Noguchi S, Noguchi A. A sensitive EIA of antithyroglobulin autoantibodies. In: Fifth International Congress of Immunology, 1983, p. 77. [22] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. [23] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350–4. [24] Hjort T. Determination of serum thyroglobulin by a hemagglutination inhibition test. Lancet 1961;i:1264. [25] Shlossberg AH, Jacobson JC, Ibbertson HK. Serum thyroglobulin the diagnosis and management of thyroid carcinoma. Clin Endocrinol 1979;10:17–27. [26] Mariotti S, Cupini C, Giani C et al. Evaluation of a solid-phase immunoradiometric assay (IRMA) for serum thyroglobulin: effect of antithyroglobulin autoantibody. Clin Chim Acta 1982;123:347–55. [27] Piechaczyk M, Chardes T, Oamille M, Pau B. Production and characterization of monoclonal antibodies against human thyroglobulin. Hybridoma 1985;4:361–7. [28] Piechaczyk M, Bouanani M, Salhi SL, Bldet L. Antigenic domains on the human thyroglobulin molecule recognized by autoantibodies in patients’ sera and by natural autoantibodies isolated from the sera of healthy subjects. Clin Immunol Immunopathol 1987;45:114–21. [29] Ligabue A, Poggioli MC, Zacchini A. Interference of specific autoantibodies in the assessment of serum thyroglobulin. J Nucl Biol Med 1993;37:273–9.

84

R. Kato et al. / Clinica Chimica Acta 298 (2000) 69 – 84

[30] Mariotti S, Barbesino G, Caturegli M. Assay of thyroglobulin in serum with thyroglobulin autoantibodies: An unobtainable goal? J Clin Endocrinol Metab 1995;80:468–72. [31] Spencer CA, Wang CC. Thyroglobulin measurement. techniques, clinical benefits, and pitfalls. Endocrinol Metab Clin North Am 1995;24:841–63. [32] Spencer CA, Takeuchi M, Kazarosyan M. Current status and performance goals for serum thyroglobulin assay. Clin Chem 1996;42:164–73. [33] Ruf J, Carayon P, Sarles-Philip F, Kourilsky F, Lissitzky S. Specificity of monoclonal antibodies against human thyroglobulin comparison with autoimmune antibodies. EMBO J 1983;2:1821–6. [34] Furmaniak J, Smith BR. The structure of thyroid autoantigens. Autoimmunity 1990;7:63–80. [35] Nye L, Pontes De, Carvalho LC, Ritt IM. Restrictions in the response to autologous thyroglobulin in the human. Clin Exp Immunol 1980;41:252–63. [36] Kohno Y. Antigenic determinants of thyroglobulin recognized in autoimmunue response. Clin Endocrinol (Japan) 1984;35:37–44, (In Japanese).