| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Institute for Immunology, Pathology and Molecular Biology (A.K., C.H., R.A.), Hamburg D-22339, Germany; and Department of Clinical Endocrinology (F.S.), Hannover Medical School, Hannover D-30173, Germany
Address all correspondence and requests for reprints to:: Arno Kromminga, Ph.D., Institute for Immunology, Pathology and Molecular Biology, Lademannbogen 61–63, D-22339 Hamburg, Germany. E-mail: kromminga{at}labor-keeser-arndt.de
 |
Abstract |
|---|
 Top Abstract IntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
Thus, we have recloned the full-length 1D-complementary DNA and assessed its expression levels in 90 healthy and diseased human thyroids. Only Graves’ patients suffering from TAO (n = 29) displayed a significant, 2.1-fold increase of 1D expression levels (P = 0.029), compared with normal controls (n = 9), as assessed using the Mann-Whitney U-test for paired, nonnormally distributed samples. In contrast, a decrease of 1D expression (to 40% of control normal values) was confined to thyroid autonomy (n = 19, P = 0.032). In all other diseased human thyroids, including Graves’ thyroids from patients not suffering from clinically overt TAO (n = 9), 1D expression levels were not different from the healthy controls. 1D gene expression was demonstrated in both healthy (n = 10) and diseased (n = 10) eye muscle tissues.
Furthermore, a recombinant protein derived from baculovirus-infected Sf9 insect cells was purified under both nondenaturing and denaturing conditions. While under nondenaturing conditions, the molecular mass of recombinant 1D was determined to be 85 kDa; denaturing isolation yielded the expected 64-kDa protein. Autoantibodies against denatured 1D protein were not detectable in sera of diseased or healthy subjects. Immunoreactivity against the 85-kDa, nondenatured protein, evaluated in a panel of 222 different human sera, showed that 82% of Graves’ patients suffering from TAO had autoantibodies against recombinant 1D, whereas only 5% of the healthy controls were positive for antibodies against 1D.
Taken together, our results demonstrate a high disease sensitivity and specificity of recombinant, nondenatured 1D, to distinguish Graves’ disease with or without TAO from other forms of thyroid and/or eye disease. Prospective studies will have to show whether autoantibodies against 1D can also be used as a prognosticator of TAO.
|
Introduction |
|---|
|
TopAbstractIntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
The nature of the autoantigenic link between the thyroid and the orbit has not been established yet. One possibility is a shared thyroid-orbit antigen leading to an autoimmune response against thyroid and orbit in Graves’ patients. One potential candidate is the TSH-R. TSH-R expression was demonstrated in Graves’ thyroids (2) and retroocular cultured fibroblasts from patients with TAO (3).
Another potentially shared thyroid-orbit antigen is the 1D protein cloned and characterized by Dong and co-workers (4). This membrane-associated 64-kDa protein, consisting of 572 amino acids, was shown to be expressed in human thyroid and human eye muscle tissue (4, 5, 6). However, 1D expression was only investigated in a relatively small number of specimens, and the clinical diagnosis was not indicated (4, 5, 6). Analysis of serum antibodies against this 64-kDa protein was carried out by several groups, indicating that only one third to two thirds of sera from Graves’ patients with TAO reacted with this protein (5, 7, 8, 9). In addition, reactivity in healthy controls was either not determined (8) or positive, in up to 53% (5), thus questioning the significance of these findings. Better results were obtained when another 64-kDa protein was used (10). Thirty to 67% of patients suffering from stable and active TAO, respectively, were positive for antibodies against this 64-kDa protein, which was identified to be the flavoprotein subunit of the succinate dehydrogenase complex (10).
To increase the disease specificity and sensitivity, we recloned the 1D antigen, assessed its expression levels in a total of 90 healthy and diseased human thyroid glands and 20 healthy and diseased human eye muscle tissues, generated the corresponding recombinant protein under nondenaturing and denaturing conditions, and evaluated its immunoreactivity in a panel of 222 different human sera.
|
Methods and Subjects |
|---|
|
TopAbstractIntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
Amplification and cloning of 1D-complementary DNA (1D-cDNA). Molecular biological methods, not described in detail, were performed as described by Sambrook et al. (11). Primers for all PCRs were synthesized by Perkin-Elmer/Applied Biosystems (Foster City, CA), and DNA sequence data of 1D (4) were retrieved from EMBL DNA data bank (Cambridge, UK) (accession number X54162). For amplification of the full-length 1D-cDNA, a nested PCR was applied on a human thyroid cDNA library (Clontech, San Diego, CA) (30 cycles of 95 C, 30 sec; 58 C, 30 sec; 72 C, 3 min 30 sec), introducing restriction sites for BamHI and SalI. The restricted PCR product was ligated into linearized pUC18, resulting in recombinant pAK39,19. The presence of insert was verified by DNA sequence analysis, following the dideoxy method (12) using the DyeDideoxy M13 Primer kit (Perkin Elmer/Applied Biosystems) and a 377 DNA sequence analyzer (Perkin-Elmer/Applied Biosystems).
Heterologous expression of 1D and its immunoreactivity. Insert DNA of recombinant pAK39,19 was subcloned into the linearized transfer vector pBlueBacHis2a (pBBH2a, Invitrogen, San Diego, CA), resulting in pBBH39,19,5. Sf9 insect cells (Invitrogen) were cotransfected, using the recombinant pBBH39,19,5 and linearized baculovirus AcMNPV DNA (Bac-N-BlueTM, Invitrogen) (13), and recombinant baculovirus clones were identified by blue/white screening and purified by plaque assay. For expression, recombinant baculovirus was added to multplicity of infection (MOI) of 0.5, and 3 days postinfection recombinant hexahistidine-tagged protein was purified from Sf9 insect cells, under both nondenaturing and denaturing conditions, using immobilized Ni-NTA agarose (Qiagen, Hilden, Germany). Protein concentrations were measured by Bradford assay (Biorad, Hercules, CA). Autoantibodies against purified 1D protein were detected by Western blot analysis. Therefore, 160 ng of 1D protein were immobilized on a nitrocellulose membrane stripe and were incubated with dilutions of human sera (1:100) overnight. Bands were visualized after incubation with antihuman IgG peroxidase conjugate (1:2000) by adding 4-chloro-1-naphthol/H2O2. All fine chemicals for SDS-PAGE and immunoblotting not mentioned were purchased from Serva (Heidelberg, Germany).
Messenger RNA (mRNA) analysis. For gene expression
experiments in thyroid tissues, total RNA was isolated from thyroid
fragments that had been frozen in liquid nitrogen immediately after
surgical removal. RNA extraction and Northern blot analysis were
performed as described previously (14). One hundred nanograms of a
961-bp PstI fragment of the pAK39,19 cDNA clone were labeled
with 
-[32P]-deoxycycidine triphosphate (Amersham,
Braunschweig, Germany). RNA integrity and the amount of RNA loaded per
lane were assayed by measuring 18S and 28S recombinant RNA (rRNA) in
ethidium bromide stains. Autoradiographic signals were quantified by
laser-scanning densitometry (Pharmacia-LKB, Freiburg, Germany) and were
compared with signals of oligonucleotide 28S rRNA.
For gene expression experiments in eye muscle, deep-frozen tissue was grinded to homogeneity. Cell lysis and isolation of total RNA were performed according to the manufacturer’s instructions (PureScript, GentraSystems, Minneapolis, MN). RT reaction was applied on total RNA using random hexamers for first-strand cDNA synthesis. Thirty cycles of seminested PCR were applied on first-strand cDNA (95 C, 30 sec; 62 C, 30 sec; 72 C, 1 min 30 sec).
Patients
Clinical parameters. Written consent was obtained from all participating patients after approval of the local ethics committee.
The following serum parameters were assessed in all participants: total T3, total T4, and T4-binding globulin (RIAs by Ciba Corning, Germany). The ratio of total T4 and T4-binding globulin was used to calculate the free T4 (FT4) index. Basal TSH levels were measured with an immunoluminescence assay (ILMA, Ciba Corning, Fernwald, Germany). Serum levels of TPO antibodies (TPOAb) and Tg antibodies (TgAb) were measured with an enzyme-linked immunosorbent assay by Elias, Freiburg, Germany. TSH binding-inhibiting Ig (TBII) was determined with a radioreceptor ligand assay (TRAK, B.R.A.H.M.S., Berlin, Germany).
Using Northern blot analysis, mRNA expression was assessed in samples from thyroid tissues from a total of 90 patients: 26 with Graves’ disease and TAO, 9 with Graves’ without TAO, 5 with Hashimoto’s, 15 with nontoxic goiters, 19 with thyroid autonomy, 7 with thyroid carcinomas, and 9 healthy controls. The diagnosis of Graves’ was made by clinical signs and symptoms of hyperthyroidism, elevated thyroid hormone levels, suppressed TSH levels, and diffusely increased technetium-99 m pertechnetate uptake in thyroid scintigraphy. All Graves’ patients were either suffering from TAO and/or were positive for TBII. TAO was diagnosed by clinical symptoms and grouped according to the modified NO SPECS classification (15). In Graves’ patients with eye symptoms compatible with but not specific for TAO, nuclear magnetic resonance of the orbital regions was performed, thus establishing or eliminating the diagnosis of TAO. Graves’ patients with or without TAO did not significantly differ from each other in all other clinical parameters (data not shown). The diagnosis of Hashimoto’s was based on the existence of an increased thyroid volume and altered thyroid texture, as assessed by ultrasound, decreased thyroid hormone levels, increased TSH levels, and evidence of TPOAb and/or TgAb. Patients with nontoxic goiter were euthyroid; had an increased thyroid volume; were TPOAb-, TgAb-, and TBII-negative; and had a decreased daily urinary iodine excretion. The diagnosis of thyroid autonomy was based upon clinical symptoms of hyperthyroidism, elevated thyroid hormone levels, suppressed TSH levels, increased 99 m pertechnetate uptake, absence of TAO, and TBII negativity. Clinical parameters in patients with thyroid autonomy did not significantly differ from those in Graves’ patients with or without TAO (data not shown). Patients with thyroid carcinomas had large cold nodules with the definitive diagnosis established postoperatively through histology. Patients with a healthy thyroid generally had to undergo surgery because of primary hyperparathyroidism.
Lymphocyte infiltration was histologically assessed and semiquantitatively graded without knowledge of the clinical and experimental data.
Applying RT-PCR, mRNA expression was assessed in undissected eye muscle tissue from 10 Graves’ patients with TAO undergoing eye surgery and in eye muscle tissue from 10 patients with neither thyroid disease nor TAO who underwent eye surgery for correction of diplopia. In terms of sex and age distribution, these two groups did not significantly differ from each other (data not shown).
Using Western blot analysis, a total of 222 human sera were analyzed: 33 sera from Graves’ patients with TAO, 43 sera from Graves’ patients without clinically overt TAO, 4 sera from Hashimoto patients, 11 sera from patients with atrophic thyroiditis, 17 sera from patients with euthyroid ATD (euthyroidism, normal TSH and significant TPOAb and/or TgAb levels in the serum), 16 sera from patients with nontoxic goiter, and 8 sera from patients with thyroid autonomy. As negative controls, we used sera from patients without thyroid disease, both clinically and in terms of serum parameters: 20 sera from patients suffering from systemic lupus erythematodes (SLE) with high autoantibody titers of antinuclear antibodies (ANA) and high antidouble-stranded DNA (anti-dsDNA), 10 sera from patients with high ANA and low anti-dsDNA, and 20 sera from patients suffering from primary biliary cirrhosis (PBC) with high antimitochondrial antibodies. An additional 40 sera, obtained from healthy blood donors who were negative for all of the autoantibodies mentioned above, served as negative controls.
For statistical analysis, the Mann-Whitney U-test for paired, nonnormally distributed samples was used to compare parameters from the patients in different groups. A P value of less than 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
Exploiting the 1D-cDNA sequence previously published by Dong
et al. (4), a nested PCR applied on a human thyroid cDNA
library resulted in a distinct product of 1766 (Fig. 1a
). After cloning into linearized pUC18,
DNA sequence analysis revealed the identity of our clone and the
1D-cDNA sequence previously published by Dong et al.
(4).
|
. Figure 3
summarizes the 1D expression levels in all 90 human thyroid tissues
investigated. Graves’ thyroids display a wide variation of 1D RNA
levels, ranging from those that are slightly reduced to those that are
increased by a factor of 4- to 5-fold. Mean 1D expression levels in
those Graves’ patients that were suffering from TAO (n = 26) is
2.1-fold and thus significantly higher (P = 0.029) than
in the 9 healthy controls. The mean 1D expression level of the 9
healthy thyroids was arbitrarily defined as 1. In contrast, the
1.5-fold increase in 1D expression levels in Graves’ patients without
TAO (n = 9) was not significant (P = 0.453) in
relation to the 9 healthy controls. Further, 1D expression in
Hashimoto’s thyroids, nontoxic goiter, thyroid autonomies, and thyroid
carcinoma is not different from 1D expression in healthy thyroids. The
increase of 1D expression in thyroids from Graves’ patients suffering
from TAO, and the decrease in thyroid autonomies, are therefore
significant and specific.
|
|
). 1D gene expression was analyzed
in human eye muscle tissue from Graves’ patients suffering from TAO
(n = 10) and from patients with neither thyroid disease nor TAO
who underwent eye surgery for correction of diplopia (n = 10). RNA
integrity was confirmed by coamplifying a 268-bp fragment of ß-globin
(data not shown). Isolation of recombinant 1D from Sf9 insect cells
The 1D-cDNA was subcloned into the linearized transfer vector
pBlueBacHis-2a, resulting in the recombinant transfer vector
pBBH39,19,5. Sf9 insect cells were cotransfected by pBBH39,19,5 and
linearized AcMNPV baculovirus DNA. After plaque assay, infected Sf9
insect cells were harvested 3 days post infection, and recombinant 1D
was purified under both nondenaturing and denaturing conditions, which
were controlled with the help of SDS-PAGE and Western blot analysis
(Fig. 4a). While under nondenaturing
conditions, recombinant 1D was detected, with a molecular mass of 85
kDa (lane 1 in Fig. 4b); denaturing isolation yielded the expected
64-kDa band (lane 2 in Fig. 4b). Aggregation as a possible explanation
for the relatively large size of the 85-kDa protein can be excluded
(data not shown).
|
Immunoreactivity of recombinant 1D protein was examined by Western
blot analyses. Autoantibodies against denatured 1D protein were
detectable in sera of diseased and in healthy subjects only with low
sensitivity, even though the 4-fold amount of antigen was used (data
not shown). In a panel of 222 human sera (Fig. 5), autoantibodies against recombinant
nondenatured 1D were detected in 27 of 33 Graves’ sera (82%) from
patients with TAO, whereas only 24 of 43 Graves’ sera (56%) from
patients without TAO were anti-1D-positive. None of the 4 sera from
Hashimoto patients (0%), 3 of 11 sera from patients with atrophic
thyroiditis (27%), 4 of 17 sera from patients with euthyroid ATD
(24%), 5 of 16 sera from patients with nontoxic goiter (31%), and 2
of 8 sera from patients with thyroid autonomy (25%) were positive for
anti-1D antibodies. As controls, we tested 20 sera from patients
suffering from SLE with high ANA and high anti-dsDNA antibody titer
(10% positive), 10 sera with high ANA and low anti-dsDNA (none
positive), 20 sera from patients suffering from PBC with high
antimitochondrial antibodies (10% positive), and 40 healthy blood
donors (5% positive).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
Next to this discrepancy in electrophoretic migration, there is a discrepancy in the immunoreactivity of recombinant 1D, depending on the isolation procedure. Autoantibodies against denatured 1D protein were neither detectable in sera of diseased nor healthy subjects, even though the 4-fold amount of antigen was used (data not shown). On the other hand, autoantibodies against nondenatured recombinant 1D protein were found in 27 of 33 (82%) human sera from Graves’ patients suffering from TAO, which is in the same order of magnitude as TBII positivity in Graves’ patients in general (17). In contrast, healthy controls and patients suffering from other autoimmune, nonthyroid diseases, such as SLE or PBC, were positive in only 5 and 10%, respectively. This compares favorably with other studies where up to 53% of the normal controls were immunoreactive (5). Twenty-three of 43 sera (56%) from Graves’ patients without clinical signs of TAO reacted with 1D. Because neither we nor the authors of theses studies mentioned conducted a clinical follow-up of their patients, it remains unclear whether positivity for the recombinant 1D represents subclinical TAO and/or precedes the development of clinically overt ophthalmopathy. We did not correlate 1D thyroid expression levels and anti-1D autoantibody titers with eye changes of TAO, because the NO SPECS classification (class 0 to class VI) only incompletely describes eye disease activity (15).
In addition, some patients suffering from thyroid diseases other than Graves’ were anti-1D positive, ranging from 24% (euthyroid ATD) to 31% (nontoxic goiter). This is in line with previous observations where the existence of autoimmune thyroid phenomena was not necessarily correlated with ATD. Up to 10–15% of normal women may have thyroid autoantibodies, but only 10% (18) to 30% (19) of these women develop clinically overt thyroid disease in the course of their lives. Furthermore, in nontoxic goiter, a thyroid disease normally not associated with autoimmunity, intrathyroidal autoimmune phenomena [such as expression of the major histocompatibility complex class II (20), invariant chain (21), and expression of various cytokines (2, 22, 23, 24, 25)] have been described.
1D expression could be demonstrated in a total of 90 human thyroid
glands and 20 human eye muscles. Interestingly, despite the high degree
of overlap, Graves’ thyroids from patients suffering from TAO were the
only ones displaying increased 1D expression levels (2.1-fold), whereas
other diseased thyroids (such as thyroids from Graves’ patients
without TAO) showed normal 1D expression levels (Fig. 3). Therefore,
the pattern of 1D gene expression is comparable with the pattern of
TSHR gene expression, where an increase of TSHR mRNA levels (2.2-fold)
was limited to Graves’ thyroids, as well (2). Lymphocyte infiltration
was the same in all thyroids investigated and, therefore, can not
account for these differences in 1D expression. Therefore, these
differences in 1D expression are thought to be significant and
specific.
Because 1D transcripts were also detectable in human eye muscles (Fig. 1b), our data support the idea that a shared thyroid-orbit antigen,
i.e. 1D, may constitute an autoantigenic link in thyroid and
eye disease in Graves’ patients. According to this model, Graves’
patients with an increased 1D expression in their thyroids may develop
autoantibodies against this antigen that subsequently cross-react with
the human eye muscle (26). At this point, however, it remains to be
shown whether 1D, on top of being an indicator, may also play a role in
the pathogenesis of TAO. Also, it is unclear whether an increase in 1D
expression in thyroids and eye muscle tissues from Graves’ patients
suffering from TAO correlates with anti-1D serum levels. Despite the
high prevalence of autoantibodies against recombinant 1D, as shown in
this study, the role of T cell-mediated autoimmune response is still a
matter of investigation.
In summary, our data show that autoantibodies against recombinant 1D are a useful marker of clinically overt TAO and may help to distinguish Graves’ disease, with or without TAO, from other forms of thyroid and/or eye disease. Prospective studies will have to show whether autoantibodies against recombinant 1D can also be used as a prognosticator, i.e. to determine whether Graves’ patients with TAO will respond favorably to treatment.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 9, 1998.
Accepted April 29, 1998.
|
References |
|---|
|
TopAbstractIntroductionMethods and SubjectsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Fujarewicz, M. Jarzab, M. Eszlinger, K. Krohn, R. Paschke, M. Oczko-Wojciechowska, M. Wiench, A. Kukulska, B. Jarzab, and A. Swierniak A multi-gene approach to differentiate papillary thyroid carcinoma from benign lesions: gene selection using support vector machines with bootstrapping Endocr. Relat. Cancer, September 1, 2007; 14(3): 809 - 826. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
