| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Medicine (K.G., S.K., J.S., J.R.W.), Allegheny University Hospitals, Allegheny General, Pittsburgh, Pennsylvania 15212-4772; Institute of Endocrinology (A.D.B., A.Be., A.A.S.), 2nd University of Naples, Naples, Italy; Endocrine Research Laboratory (S.W.), Hospital de Sant Pau, Autonomous University of Barcelona, Barcelona, Spain; Department of Veterans Affairs Medical Center and Department of Biochemistry and Biophysics (B.C., B.A.C.A.), University of California at San Francisco, California 94121; Cattedra di Endocrinologia (M.S.), Universita di Parma, Parma, Italy; and Department of Clinical and Experimental Medicine "F Magrassi" (A.Bi.), 2nd University of Naples, Naples, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eye muscle membrane proteins of 63–67 kDa are possible markers of ophthalmopathy in patients with thyroid autoimmunity (8, 9, 10). Of these, one, originally assigned a molecular mass of 64 kDa on the basis of electrophoretic mobility, is most closely associated with active eye muscle inflammation (3, 9). Recently, we identified this 64-kDa protein as the flavoprotein (Fp) subunit of mitochondrial succinate dehydrogenase (11), which has a molecular mass of 67 kDa, based on amino acid sequence (12). Preliminary data from Western blotting also confirmed serum antibodies against the purified subunit in 67% of patients with TAO, 12% of those with Graves’ hyperthyroidism, and 8% of normal subjects (11). We have now developed an enzyme-linked immunosorbent assay (ELISA) that incorporates the highly homologous beef heart succinate dehydrogenase as antigen and have tested for reactive serum antibodies in patients with TAO and Graves’ hyperthyroidism without evident ophthalmopathy, including 12 patients who developed ophthalmopathy after treatment of their hyperthyroidism. A very close relationship between eye muscle disease and serum anti-Fp subunit antibodies is demonstrated. On the basis of our findings, it is proposed that TAO comprises two main subgroups: namely, ocular myopathy (which is associated with anti-Fp antibodies) and congestive ophthalmopathy (where eye muscle damage does not occur and the antibodies are not detected).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sera from patients with TAO and Graves’ hyperthyroidism have been tested for antibodies reactive against the Fp subunit of mitochondrial succinate dehydrogenase. The screening was in 2 parts. In the first part were patients with TAO classified according to the activity and severity of their eye disease and whether or not they had clinical evidence for eye muscle involvement. In the second part, we tested serial serum samples from 16 patients with Graves’ hyperthyroidism who had ophthalmopathy at the onset (4 patients) or developed the eye disorder after treatment of the hyperthyroidism with methimazole (12 patients), these being a subgroup of over 100 patients without eye disease studied earlier by De Bellis et al (13); and from 4 other patients who did not develop ophthalmopathy. In this study, blood was drawn initially, then at approximately 3 monthly intervals for up to 24 months. Informed written consent was obtained from all patients and normal subjects studied.
Clinical subjects
These included: 1) 96 patients with ophthalmopathy associated
with Graves’ hyperthyroidism (74 females and 22 males, 27–79 yr old;
mean age, 53 yr), none of whom were being treated with corticosteroids
at the time of study; 2) 18 patients with Graves’ hyperthyroidism,
without ophthalmopathy or orbital ultrasound evidence for eye muscle
enlargement (13 females and 5 males, 26–65 yr old; mean age, 43 yr), 4
of whom were studied prospectively for up to 24 months; 3) 16 patients
with Graves’ hyperthyroidism, 12 of whom developed ophthalmopathy
after treatment of the hyperthyroidism and 4 of whom had eye disease at
the onset (14 females and 2 males, 24–48 yr old; mean age, 35 yr),
studied prospectively; 4) 15 patients with Hashimoto’s thyroiditis (12
females and 3 males, 29–69 yr old; mean age, 44 yr); 5) 8 patients
with nonimmunologic thyroid disorders: namely, 2 with thyroid cancer
and 6 with multinodular goiter (8 females and 1 male, 30–77 yr
old; mean age, 45 yr); 6) 25 patients with type I diabetes mellitus (6
females and 19 males, 14–68 yr old; mean age, 28 yr), without evidence
of ophthalmopathy; and 7) 32 healthy subjects (24 females and 8 males,
22–52 yr old; mean age, 39 yr), with no personal or family history of
thyroid disease, ophthalmopathy, or other autoimmune disease, as
controls. Patients with ophthalmopathy were classified according to an
activity index (AI, 0–7) proposed by a committee of the International
Thyroid Associations (14). Severity of the congestive changes was
classified as: active (AI > 4), with or without eye muscle disease;
moderately active (AI 1–3), with or without eye muscle disease; or
chronic (AI 0), disease stable and duration more than 1 yr, with or
without eye muscle disease. Full ophthalmologic examination, including
measurement of eye muscle function and performance of orbital
ultrasound, was carried out on all patients with ophthalmopathy. For
the purposes of this study, ocular myopathy was defined as: diplopia
and reduced eye movement in one or more of the main gazes associated
with marked eye muscle volume increase on orbital ultrasound; and
congestive ophthalmopathy as: nil or minimal eye muscle enlargement
with, usually, a fibrotic appearance, as described by Ossoinig (15),
features of periorbital inflammation (e.g. chemosis, lid
swelling, conjunctival injection), and no diplopia or reduced eye
movements. The diagnoses of Graves’ hyperthyroidism, Hashimoto’s
thyroiditis, and type I diabetes mellitus were made according to
standard clinical criteria and were confirmed by appropriate laboratory
testing. The main clinical findings, including details of treatments
given for hyperthyroidism and ophthalmopathy, and the anti-Fp
antibody status, both initially and on follow up, of the 20 patients
with Graves’ hyperthyroidism studied prospectively (see above), are
shown in Table 1
.
|
Succinate dehydrogenase was solubilized by perchlorate treatment of succinate:coenzyme Q oxidoreductase (complex II of the respiratory chain) (16), which had been isolated from beef heart mitochondria by the method of Baginsky and Hatefi (17). The enzyme, which contains an Fp and an iron-sulfur subunit (18, 19), was more than 90% pure, based on gel analysis and content of covalently-bound flavin adenine dinucleotide (20). Pure Fp subunit was separated on SDS-polyacrylamide gels and excised according to the method of Merli et al. (21) and used as antigen in ELISA.
ELISA
The method has been described in previous publications from this laboratory (22, 23). Tests were performed, in triplicate, in 96-well plates. The optimal concentration of purified Fp subunit was found, in preliminary checkerboard assays, to be 1 µg/mL; and optimal serum dilution, 1:50. The second antibody was an alkaline phosphatase-labeled goat antihuman IgG, diluted 1:1500. As control, we used PBS instead of antigen, serum, or secondary antibody. Results were expressed as optical density (OD) at 410 nmol/L, and a positive test was taken as an OD 410 nmol/L > 0.200, the upper limit of normal for 10 control sera. The intraassay coefficient of variation for 6 positive and 6 negative sera was 5%. The interassay coefficient of variation for 6 positive control sera was 10%.
Statistical analysis
Differences in prevalences of serum autoantibodies against Fp,
between patient groups and normals, were assessed statistically using
-square tests and Yeats’ correction for small numbers. Differences
between mean (± SD) values for the groups were assessed
using Student’s t tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Aliquots of standard positive
and negative sera were included in each ELISA as internal controls.
Anti-Fp antibodies were detected in 73% (24 of 33) of patients with
active ophthalmopathy of recent onset and clinical evidence for
involvement of the eye muscle disease, as compared with 25% (3 of 12)
of those without eye disease. The corresponding values with and without
eye disease were 50% (7 of 14) and 37% (6 of 16), respectively, for
patients with moderately active ophthalmopathy of recent onset; and 0%
(0 of 10) and 9% (1 of 11), respectively, for patients with chronic
eye disease. Tests were also positive in 14% (2 of 14) of patients
with Graves’ hyperthyroidism without evident ophthalmopathy, in 20%
(3 of 15) of patients with Hashimoto’s thyroiditis, in 12.5% (1 of 8)
of patients with nonimmunologic thyroid disorders, in 12% (3 of 25) of
patients with type I diabetes mellitus, and in 12% (4 of 32) of age-
and sex-matched normal subjects. The prevalences were significantly
greater than for normals for patients with: 1) active TAO and eye
muscle disease (-square test, P < 0.001); 2)
moderately active TAO and eye muscle disease (P <
0.05); and 3) moderately active TAO without eye muscle disease
(P < 0.05), respectively, but not for the other groups
(P = NS). Mean (± SD)) OD values for these
3 groups (0.280 ± 0.10, 0.210 ± 0.06, and 0.198 ±
0.065, respectively) were significantly increased, compared with
normals (0.148 ± 0.09; t test, P <
0.001, P < 0.01, P < 0.05,
respectively) but not for the other groups (Table 1
). Twenty-three
patients with euthyroid Graves’ disease, defined as ophthalmopathy in
the absence of overt thyroid dysfunction, were tested for anti-Fp
antibodies. Tests were positive in all 7 patients with active
ophthalmopathy and eye muscle involvement, with an overall prevalence
of 31% (P = NS, compared with normals) but negative in
the 16 who were tested more than 12 months after the onset of their eye
symptoms (results not shown).
|
) developed diplopia, eye muscle
dysfunction, and marked eye muscle volume increase, as observed by
orbital ultrasound, with or without congestive changes, from 1.5–6
months after beginning treatment; and 6 patients (nos. 7–12, Table 1)
developed severe congestive ophthalmopathy but no eye muscle
dysfunction and no (or only minimal) eye muscle volume increase, on
orbital ultrasound, after 4.5–10 months. We also studied 4 patients
who did not develop ophthalmopathy during a 12-month period after
radioiodine treatment. The results of antibody tests, which were
conducted simultaneously at the end of the study, are summarized in
Fig. 1, in which individual patients are
indicated. Three replicates of each sample were tested, and all sera
were tested in a single assay. Anti-Fp subunit antibodies were detected
in 4 of the 6 patients who developed eye muscle dysfunction after
diagnosis of the hyperthyroidism and during treatment with antithyroid
drugs, and borderline levels were observed in 2 patients (Fig. 1A). The
antibodies were first detected at the time of treatment of the
hyperthyroidism in 2 of these patients (nos. 4 and 5) and after 3 and 5
months, respectively, in 2 (nos. 2 and 3). In the other 2 patients
(nos. 1 and 6), levels increased and became borderline-positive after 2
and 5 months, respectively. Overall, the antibodies predicted the
development of eye muscle disease in 5 of the 6 patients in whom this
occurred (see Table 1). In contrast, anti-Fp subunit antibodies were
not detected in any of the 6 patients who developed connective tissue
inflammation but no eye muscle disease (nos. 7–12) (Fig. 1B). Four
patients (nos. 13–16) had ophthalmopathy at the onset, of whom 3 (nos.
13, 15, and 16) had eye muscle disease. Anti-Fp antibodies were
detected in all 3 of these latter patients, initially in 2 (nos. 15 and
16) and after 5 months in 1 patient (no. 13) (Fig. 1C). Anti-Fp
antibodies were not detected in any of 4 patients with Graves’
hyperthyroidism who did not develop ophthalmopathy during a 12-month
period after radioiodine treatment (Fig. 1D).
|
. Serum antibody
levels fell, in parallel with improvement in the severity of the eye
muscle disease, in all three patients who had eye muscle disease
initially (nos. 13, 15, and 16) and in the two (nos. 4 and 5) who
developed it during treatment of the hyperthyroidism. Levels increased
slightly in the one patient (no. 16) whose ophthalmopathy worsened
during steroid therapy. Antibody levels were always negative in one
patient (no. 9). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). What is the significance of anti-Fp antibodies? First, our findings support the hypothesis, proposed by Solovyena (36), that TAO comprises two main subgroups of patients. As defined by our data, there would be patients with eye muscle involvement and serum antibodies reactive against the Fp subunit of succinate dehydrogenase, which we call ocular myopathy, with or without associated congestive changes (e.g. chemosis, lid swelling, epiphora, and conjunctival injection), as opposed to those patients with only congestive disease in whom anti-Fp antibodies are not detected, which we call congestive ophthalmopathy. Although it might be argued that the minimal increase in eye muscle volume, on orbital ultrasound (observed in four of the six patients with congestive ophthalmopathy) reflected eye muscle disease, we propose that this is part of a generalized orbital connective tissue and fat swelling and, in the absence of diplopia and eye muscle dysfunction, not indicative of eye muscle damage. Moreover, in the patients with eye muscle dysfunction, clinically, the increase in eye muscle volume, on orbital ultrasound, was always marked and involved all muscles. Ossoinig (15) found a close correlation between high reflectivity of the extraocular muscles (on orbital ultrasonography) and myositis, whereas more irregular or low reflectivity was correlated with fibrosis, the end result of congestive ophthalmopathy. In our study, patients with the nonfibrotic pattern of reflectivity were all positive for anti-Fp antibodies, whereas tests were always negative in those showing low muscle reflectivity.
We further propose that release of succinate dehydrogenase from necrosed eye muscle cells and formation of antibodies against its Fp subunit are specific markers of extraocular muscle autoimmunity. The cause of the connective tissue inflammation, which may be the only abnormality in perhaps a third of patients with ophthalmopathy, is unknown but may reflect effects of proinflammatory cytokines on the orbital fibroblasts, as suggested by several workers (2, 25, 37, 38, 39). Though the TSH-receptor has been advocated by some as a target antigen in the orbital connective tissue (40, 41, 42), this is controversial (26).
Second, although recent evidence shows that some autoantibodies against surface (membrane) antigens and cognate receptors are internalized (43), it is unclear whether a globular antibody molecule directed against an intracellular protein would be able to penetrate the intact cell, and subcellular membranes, to bind to its target antigen. We have demonstrated mitochondrial abnormalities, on electron microscopic examination of eye muscle fibers, in patients with TAO (30). Though these findings in ophthalmopathy are consistent with a direct effect of antibodies targeting succinate dehydrogenase, the observed abnormalities are more likely to be secondary to muscle fiber necrosis, sensitization to succinate dehydrogenase released from necrosed muscle fibers being secondary. In support of this, we have recently demonstrated anti-Fp antibodies in serum from 50% of patients with ocular myasthenia gravis (45), suggesting that the antibodies are nonspecific markers of an autoimmune reaction against the extraocular muscle fiber. The possibility that anti-Fp antibodies are also markers of systemic skeletal muscle inflammation has not been addressed.
Third, anti-Fp subunit antibodies seem to accurately predict the
development of eye muscle disease in patients with Graves’
hyperthyroidism. The antibodies were detected before the onset of eye
muscle disease in five of the six patients who developed this during
treatment of their hyperthyroidism (Table 1). This confirms our earlier
findings, using Western blotting, which showed a close relationship
between ophthalmopathy and serum antibodies against a porcine eye
muscle membrane antigen of 64-kDa protein (5, 8). Patients with
Graves’ hyperthyroidism could be screened for serum anti-Fp antibodies
to identify those who are predisposed to develop ophthalmopathy, with
the possibility of preventing the eye disorder. This test could be used
to address the controversy regarding the possible effects of
radioiodine treatment (46). In patients with Graves’ hyperthyroidism,
but no eye signs, the detection of serum anti-Fp antibodies could be an
indication for prophylactic coverage of radioiodine therapy,
e.g. with corticosteroids. This question will be addressed
in a prospective multicenter study of patients with Graves’
hyperthyroidism treated with radioiodine or antithyroid drugs. The test
could also be used to monitor treatment of patients with active TAO. In
this study, we showed a close correlation between changes in the
clinical status of the eye muscle component of ophthalmopathy and serum
levels of anti-Fp antibodies after treatment with steroids; in the one
patient whose ophthalmopathy worsened while being treated with
steroids, levels of antibodies increased.
Although anti-Fp antibodies are closely associated with eye muscle damage in patients with Graves’ hyperthyroidism, the antibodies were also found in 12% of patients with insulin-dependent diabetes mellitus, 20% of patients with Hashimoto’s thyroiditis, and 12% of normal subjects, potentially limiting the usefulness of the test. We have recently shown that such false-positive reactivity is attributable to recognition of the flavine adenine dinucleotide cofactor used by succinate dehydrogenase, and some other mitochondrial enzymes; whereas in patients with TAO and eye muscle disease, reactivity is against the Fp molecule but usually not flavine adenine dinucleotide (de Bellis et al., submitted). Once the TAO-specific epitope on the Fp molecule has been identified, a test incorporating synthetic peptide can be developed for clinical use.
In conclusion, we have shown a close relationship between serum antibodies against the Fp subunit of mitochondrial succinate dehydrogenase and eye muscle disease (but not congestive changes) in patients with Graves’ hyperthyroidism. We propose that the test be used to predict which patients with Graves’ hyperthyroidism are predisposed to development of ophthalmopathy after treatment with radioiodine or antithyroid drugs, and to monitor treatment of patients with active eye muscle disease. Before recommending the test for the diagnosis of euthyroid Graves’ disease, we must test sera from patients with the most important alternative diagnoses (including orbital myositis, pseudotumor, and orbital malignancies).\.
|
Footnotes |
|---|
1 This work was supported by research grants from Toray-Fuji Bionics Inc., Tokyo, Japan; The Pennsylvania Lions Sight
Conservation and Eye Research Foundation Inc.; and National Institutes
of Health Grant HL-16251 (B.A.C.A.). 
2 Drs. Gunji and De Bellis contributed equally to the work reported
in this manuscript.
Received May 28, 1998.
Revised August 27, 1998.
Revised December 4, 1998.
Accepted January 19, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Bartalena, A. Pinchera, and C. Marcocci Management of Graves' Ophthalmopathy: Reality and Perspectives Endocr. Rev., April 1, 2000; 21(2): 168 - 199. [Abstract] [Full Text]
|
|
|
|
 |
|
 G. Amato, M. Rotondi, I. Salzano, A. De Bellis, G. De Felice, C. Costagliola, A. Bellastella, and C. Carella Extraocular Muscle Antibodies Positivity as the Only Serum Marker of Euthyroid Graves' Ophthalmopathy following Subacute Thyroiditis: Case Report J. Clin. Endocrinol. Metab., March 1, 2000; 85(3): 950 - 952. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
