| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
RSR Ltd., Avenue Park, Pentwyn, Cardiff CF23 8HE, UK;
2
Cosmic Corporation, Tomisaka Bldg., 2-7-3 Koishikawa, Bunkyo-ku, Tokyo 112, Japan;
a author for correspondence: fax 44-1222-732704
Hyperthyroidism in Graves disease is attributable to autoantibodies to the thyroid-stimulating hormone receptor (TSHR), and measurement of these TSHR autoantibodies (TRAbs) can be useful in disease diagnosis and management (1)(2)(3)(4). Usually TRAbs are detected by bioassays based on cultured cells or by receptor assays based on 125I-labeled TSH (1)(2)(3)(4). The most widely used receptor assay (5) uses detergent-solubilized porcine TSHR with TRAbs to inhibit the TSHR-125I-TSH interaction, and polyethylene glycol (PEG) to separate receptor-bound and free labeled TSH by precipitation. Recently, we developed a monoclonal antibody (MAb) to the porcine TSHR COOH terminus that can be used to attach the TSHR to a solid phase, and we describe the use of this antibody to develop an ELISA for TRAbs. In the assay, TRAbs compete for binding to the TSHR with biotinylated bovine TSH [bovine TSH being readily available and having a much higher biological activity than human TSH (6)]. TSH-biotin binding is then monitored using a streptavidin-peroxidase conjugate.
The C-terminal end of the porcine TSHR (last 160 amino acids) (3) was expressed in Escherichia coli as a fusion protein with glutathione S-transferase and used to produce MAbs as described previously (7). One of the MAbs had a relatively high affinity (5 x 109 L/mol) for detergent-solubilized porcine TSHR, and the purified IgG was used to coat ELISA plate wells by incubation overnight at 4 °C. In these experiments, 100 µL of 0.01 g/L antibody in 100 mmol/L NaHCO3, pH 9.2, was used. After washing and coating with 10 g/L bovine serum albumin (Sigma-Aldrich), the plates were washed again with assay buffer (10 mmol/L Tris-HCl, pH 7.4, 50 mmol/L NaCl, 1 g/L bovine serum albumin, 1 mL/L Triton X-100). Detergent-solubilized porcine TSHR (100 µL) diluted in assay buffer was then added, and the plates were incubated for 30 min with shaking at room temperature. The TSHR preparation was then removed, and the wells were washed with assay buffer. The receptor-coated wells were then used immediately or dried, sealed in foil pouches, and stored at 2–8 °C.
In the assay, 100 µL of test sera and 10 µL of mouse serum (to neutralize any anti-mouse IgG present in the test sera) were added to duplicate receptor-coated wells, followed by incubation at room temperature for 2 h. Sera were then removed (without washing at this stage), and 100 µL (5 ng) of biotinylated bovine TSH (TSH-bi; RSR Ltd.) in assay buffer added. After further incubation for 15 min at room temperature, excess TSH-bi was removed (without washing at this stage), 100 µL of streptavidin-peroxidase conjugate (Sigma-Aldrich) was added, and incubation was continued for 20 min. The wells were then washed once with assay buffer and once with distilled water. The peroxidase substrate tetramethylbenzidine (100 µL) was added to the wells, followed by, after 20 min, the addition of 2 mol/L H2SO4 and measurement of absorbance at 450 nm.
In the presence of sera from healthy blood donors, an absorbance of ~2 was observed. This was reduced in a dose-dependent manner to ~0.3 by TRAbs or by bovine TSH. For example, thyroid-stimulating antibody first international standard 90/672 (National Institute for Biologic Standards and Controls, Hertfordshire, UK) at 3, 10, and 30 units/L gave 15%, 37%, and 70% inhibition of TSH binding, respectively [inhibition of TSH binding expressed as: 100 x (1 - ratio of absorbance at 450 nm of test sample to absorbance at 450 nm of a pool of healthy blood donor sera)].
Sera from individual healthy blood donors (n = 65), double-stranded DNA antibody-positive patients with systemic lupus erythematosus (n = 10), rheumatoid factor-positive patients with rheumatoid arthritis (n = 10), and patients with Hashimoto thyroiditis (n = 15) showed inhibition of TSH binding of <10%.
We compared the ELISA and the conventional TRAb receptor assay based on
125I-labeled TSH and PEG precipitation
(5) in 56 sera from patients suspected of having Graves
disease (Fig. 1
). Of the 56 sera studied, 38 were clearly positive by
125I assay, and the same 38 were clearly positive
by ELISA (inhibition of TSH binding >15% in both assays). Two sera
were borderline positive in both assays (12–13% inhibition), and one
serum was borderline positive by 125I assay (11%
inhibition) but negative by ELISA (3% inhibition). At this stage it
was not clear which result (borderline positive or negative) would
better reflect the outcome of treatment. The remaining 15 sera were
negative in both the radioactive assay (inhibition of binding, -4% to
8%) and the ELISA (inhibition of binding, -6% to 9%). We used the
same 56 sera to compare the ELISA and a TRAb bioassay based on
stimulation of cAMP production in isolated porcine thyroid cells
(8) (reagents from Yamasa Ltd.); r =
0.73. Thirty-four of the 56 sera were positive by bioassay
compared with 38 of 56 by ELISA. One sample was positive by bioassay
(285% stimulation; positive = greater than 180%) but only
borderline positive by ELISA (12% inhibition) and by radioactive assay
(12% inhibition). The 56 patients in the study were receiving or had
received treatment for hyperthyroidism; consequently, in the case of
the five sera with discrepant bioassay and ELISA results, it was not
clear which result better reflected the underlying disease activity
[see Refs. (1)(2) for a review of the relative
clinical effectiveness of TRAb measurement by inhibition of TSH-binding
assay and by bioassay].
|
In terms of TRAb ELISA precision, typical variations in absorbances at 450 nm (in the same assay run) were 1.67 ± 0.05, 1.1 ± 0.04, 0.76 ± 0.025, and 0.51 ± 0.028 (means ± SD; n = 28).
Intraassay imprecisions (CVs) were 9.0%, 4.5%, and 2.3% at mean inhibitions of TSH binding of 30%, 42%, and 71%, respectively (n = 28), and interassay CVs were 6.8%, 3.3%, and 2.3% at mean inhibitions of TSH binding of 24% (n = 10), 45% (n = 6), and 69% (n = 10), respectively. These values of intra- and interassay imprecision are similar to those observed for the radioactive assay (5).
Occasional sera from patients with Graves disease contain TSH
antibodies (9)(10) that form complexes with
labeled TSH in the PEG-based receptor assay for TRAb. These
TSH-antibody complexes are precipitated with PEG in addition to
TSHR-TSH complexes, giving an increase in precipitated
125I compared with healthy blood donor sera.
Consequently, TRAb measurements with the PEG method are difficult to
make in samples containing TSH antibodies. To investigate the effects
of TSH antibodies in the ELISA, measurements were made in sera from 48
treated Graves disease patients selected for the presence of antibodies
reactive with labeled bovine TSH. The effects of anti-TSH antibodies
clearly evident in the PEG method (mean inhibition of TSH binding,
-33.3%; range, -13% to -87%) were not observed in the
ELISA (mean inhibition of TSH binding, 18.7%; range, -0.4% to 47%).
This was presumably attributable to the removal of the test serum
samples containing TSH antibodies after incubation in the
receptor-coated wells. Consequently, TRAb concentrations were
measurable in samples containing TSH antibodies with the ELISA, and in
the series studied, 38 of 48 sera showed 
10% inhibition (Table 1
).
|
Our results indicate that porcine TSHR immobilized on ELISA plates, using a MAb to the receptor’s COOH terminus, and TSH-biotin can be used to create a TRAb ELISA. The ELISA has similar sensitivity and precision to the current radioactive test but has some major advantages, including easy automation and absence of interference from antibodies to TSH.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
