| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division and Research Unit of Endocrinology (A.D.C., R.D.P., C.M., V.D.F.), Istituto di Ricovero e Cura a Carattere Scientifico Casa Sollievo della Sofferenza General Hospital, 71013 San Giovanni Rotondo (Foggia); the Section of Cell Regulation Metabolic Diseases Branch, National Institute of Diabetes, Digestive and Kidney Diseases, National Institutes of Health (K.T., L.D.K.), Bethesda, Maryland 20892; and the Department of Cell Biology and Oncology, Istituto di Ricerche Farmacologiche Mario Negri, Consorzio Mario Negri Sud (D.C.), 66030 S. Maria Imbaro (Chieti), Italy
Address all correspondence and requests for reprints to: Dr. Alfredo Di Cerbo, Division and Research Unit of Endocrinology, IRCCS Casa Sollievo della Sofferenza, 71013 San Giovanni Rotondo (Foggia), Italy. E-mail: adicerb{at}tin.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Epitope analysis of Graves’ IgG has been performed extensively in the last few years using site-directed mutagenesis, chimeras of the human TSHR (hTSHR) and LH-CG receptor (LH-CGR), and receptor peptides (6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). Several of these studies have shown that the N-terminal portion of the extracellular domain of TSHR contains functional epitopes for autoantibodies stimulating AC and increasing cAMP intracellular levels (6, 7, 14, 15, 16, 17) and have suggested that epitope subtyping may have clinical relevance (14, 15, 16, 17). Thus, a heterogeneous epitope distribution is found in GD patients that are more likely to become euthyroid after antithyroid drug treatment, whereas GD patients with homogeneous epitope distribution are less responsive to antithyroid drugs (15, 16). An IgG exhibiting a heterogeneous epitope is defined as one whose antibody population is not solely directed against the N-terminus of the extracellular domain (15, 16).
The molecular mechanism of PLA2 activation by Graves’ IgG
is not clear. For example, it is not unequivocally clear that
PLA2-stimulating IgG bind to the TSHR and if so whether
they activate the same or different epitopes of TSHR as IgG that
increase cAMP levels. To address this question we used Chinese hamster
ovary (CHO) cells stably transfected with the wild-type hTSHR or a
hTSHR chimera with residues 8–165 (Mc1+2) substituted by equivalent
residues of the LH-CGR (Fig. 1A
). Both
cAMP production and AA release induced by IgG from 32 patients with GD
in these 2 cellular systems was measured. We compared these results
with other clinical parameters of the patients.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hormones and tissue culture media were obtained from Sigma Chemcial Co. (St. Louis, MO) or Life Technologies, Inc. (Grand Island, NY). Protein A-Sepharose CL4B was purchased from Pharmacia LKB Biotechnology (Uppsala, Sweden); Minicon B15 concentrators were obtained from Amicon Division (W.R. Grace & Co., Danvers, MA). [3H]thymidine, [3H]AA, and cAMP RIA kits were purchased from New England Nuclear Corp. (Boston, MA). Purified TSH was obtained from the National Hormone and Pituitary Program [University of Maryland (Baltimore, MD) and NIH (Bethesda, MD)].
Cells
FRTL5 cells (Interthyr Research Foundation, Baltimore, MD), a continuous line of diploid functioning rat thyroid cells, were cultured as previously described in 6H medium containing TSH (4, 5). For cAMP assay and [3H]thymidine incorporation, FRTL5 cells were plated in 96-well plates, fed fresh medium 48 h later, and then incubated with medium deprived of TSH (5H medium) (4, 5).
hTSHR cloning, construction of chimeric receptors, stable transfection of CHO cells with both the hTSHR and the hTSHR-LH/CGR chimera, as well as selection and expansion of the most responsive clones to bTSH were described previously (15). For cAMP assays, cells were plated in 96-well plates (5 x 103 cells/well), grown to 100% confluence, then incubated with growth medium deprived of TSH (5H medium). For AA release assays, cells were plated in 12-well plates (4–5 x 104 cells/well) and grown to 50–60% confluence in 6H medium.
Patients
Sera were obtained from 32 patients with active GD randomly
selected from a group of 104 enrolled in our previous study (5) and
from 10 normal subjects. The characteristics of patients were described
previously (5) and are summarized in Table 1.
|
IgG purification, AA release, and cAMP assays, and [3H]thymidine incorporation measurements
IgG were affinity purified using protein A-Sepharose CL-4B columns (4).
AA release was measured as previously described (4). Briefly, assays were performed in 0.5 mL Hank’s Balanced Salt Solution (pH 7.4) containing fatty acid-free BSA (2 mg/mL; HBSS-BSA) in the absence of TSH. TSH and Graves’ IgG were added to the cells for 15 min at 37 C. The incubation medium was then aspirated and evaluated for released radioactivity. Normal and Graves’ IgG were tested in duplicate at 0.001, 0.01, 0.1, and 1 mg/mL concentrations. Graves’ IgG-induced AA release was normalized vs. basal AA release (HBSS-BSA only). On the average, about 2000 cpm [3H]AA were released in control wells.
Extracellular cAMP levels were evaluated by a commercial RIA. Assays were performed in NaCl-free, sucrose-enriched isotonic HBSS (5.4 mmol/L KCl, 1.3 mmol/L CaCl2, 0.8 mmol/L MgSO4, 0.3 mmol/L Na2HPO4, 0.4 mmol/L KH2PO4, and 0.1% glucose) containing 20 mmol/L HEPES (pH 7.4), 0.1% BSA, 0.5 mmol/L 3-isobutyl-1-methylxanthine (15). TSH and purified IgG were added to the cells for 60 min at 37 C, at which point the incubation medium was aspirated and stored frozen below -20 C until cAMP assay. Normal and Graves’ IgG were tested in triplicate at 0.001, 0.01, 0.1, and 1 mg/mL concentrations. The Graves’ IgG-induced cAMP was normalized vs. basal cAMP production (HBSS-BSA-3-isobutyl-1-methylxanthine only).
[3H]Thymidine incorporation was performed in FRTL5 cells as previously described (5).
The cut-off value discriminating between [3H]AA release, cAMP production, and [3H]thymidine incorporation induced by pathological and normal IgG in CHO-hTSHR and FRTL5 cells and by pathological IgG in CHO-hTSHR and CHO-Mc1+2-transfected cells was calculated using receiver operating characteristic curve analysis (19).
Measurement of thyroid hormones and autoantibodies
Serum levels of total thyroid hormones and TSH were measured using commercial kits [Amersham Pharmacia Biotech (Aylesbury, UK) and Travenol (Cambridge, UK)]. The ranges of serum concentrations of T3 and T4 in normal subjects were 80–180 ng/dL (1.23–2.8 nmol/L) and 4.5–12 µg/dL (58–155 nmol/L), respectively; that of TSH serum concentrations was 0.2–4 µU/mL. TSH binding-inhibiting Ig (TBII) was evaluated by a commercial RRA (Henning, Berlin, Germany). We considered the test positive when the TBII value was greater than 10 U/L, which is 2 SD above the mean value we previously reported for 54 healthy subjects (5).
Thyroid ultrasonography
Thyroid volume was measured by thyroid ultrasonography in all GD patients and normal subjects. All ultrasound examinations were performed by the same physician (A.D.C.), using a Toshiba SAL-38AS dynamic image scanner (Tokyo, Japan) with a linear 7.5-MHz transducer. The volume of each lobe was calculated according to the method of Brunn et al. (20).
Statistical analysis
Continuous data are presented as the mean ± SD
unless otherwise stated. Comparisons between two groups were performed
using Student’s t test or Mann-Whitney rank sum test if the
data were not normally distributed. The differences between more than
two groups were assessed by one-way ANOVA applied to ranked data
(equivalent to Kruskal-Wallis test). Two-way ANOVA was used to test the
statistical significance of the differences between TSH-stimulated
arachidonic acid release and cAMP production in CHO cells transfected
with wild-type hTSHR (WT cells) and hTSHR-rat LH/CG receptor chimera
(Mc1+2 cells). Pairwise comparison between groups at each dose point
was assessed with the Fisher’s least significant difference
post-hoc test. Categorical variables were tested with the
2 method, and the Yates’ correction was applied when
appropriate. P < 0.05 was considered statistically
significant. All statistical analyses were carried out using the SYSTAT
statistical program for Macintosh, version 5.2 (SYSTAT, Inc., Evanston,
IL) (21).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although PLA2 activity and AA release in FRTL5 cells
is suggested to be TSHR dependent (4, 5), this has not been formally
proven in TSHR transfected cells. To address this concern, we developed
optimal assay conditions and evaluated PLA2-stimulating
activity using recombinant hTSHR-transfected cells exposed to
increasing concentrations of TSH. As shown in the Fig. 1B
, CHO cells
transfected with wild-type hTSHR, but not control cells with no TSHR,
are able to release AA from membrane phospholipids when challenged with
TSH. The same concentrations of TSH could also induce AA release in
Mc1+2-transfected CHO cells (Fig. 1B).
When the ability of TSH to stimulate cAMP production in the same CHO
cells containing wild-type TSHR or in CHO cells stably transfected with
the Mc1+2 chimera (Fig. 1C) was measured, the maximal cAMP response to
TSH was reduced in Mc1+2 chimera, but there was a slight increase in
response sensitivity at lower TSH concentrations (Fig. 1C). These data
are similar to those reported previously (15). Our data thus show that
PLA2 activity and AA release are TSHR dependent and that
substitution of the N-terminal region of the TSHR extracellular domain
results in transfectants that retain PLA2 as well as
AC-stimulating activity when assayed with TSH.
Graves’ IgG stimulation of PLA2 vs. cAMP activity in CHO cells transfected with wild-type hTSHR and the Mc1+2 chimera
Activity in cAMP assays and clinical correlations. Studies
using chimeras of the hTSHR and LH-CG receptor (LH-CGR) have shown that
most of the patients from a Korean cohort with GD have AC-stimulating
IgG whose activity depends on the N-terminal portion of the TSHR
extracellular domain (residues 8–165) (15, 16). To ensure that this
was true in our patients and for subsequent comparative purposes, we
initially measured the ability of different concentrations of IgG from
GD patients to activate AC signal transduction in the CHO-hTSHR/chimera
assay system. We found that 26 of 32 IgG tested (81%) were positive in
CHO cells transfected with wild-type hTSHR (mean cAMP concentration,
304% over the basal) at a 1 mg/mL concentration (Tables 1 and 2). Patients whose IgG increased cAMP
levels (group A1 in Table 2) had greater TBII activities,
[3H]thymidine incorporation, and serum T3 and
T3/T4 molar ratio; larger goiters; and a higher
prevalence of ophthalmopathy than patients whose IgG did not increase
cAMP levels (group B1 in Table 2). As previously reported (6, 7, 15, 16) the AC-stimulating activity was largely lost when IgG from these
patients was used in the Mc1+2 cell system (Fig. 2A). As a group, a significant (55%)
reduction in cAMP production was found (P < 0.0001;
range, 0–84% reduction); individually (Table 1), 20 of 26 patients
(77%) had AC-stimulating antibodies that require epitopes on the
N-terminal portion of the TSHR extracellular domain (residues 8–165).
When we compared the clinical parameters of patients whose IgG
exhibited residual activity in the chimera [group A1 (+/+) in Table 3] with those of patients who had activity in CHO-hTSHR cells only,
i.e. patients whose IgG lost their stimulatory activity in
the chimera and whose activity depended solely on epitopes in the
N-terminal portion of the extracellular domain [group A1 (±) in Table 3], we found that clinical characteristics of the 2 groups of patients
were slightly different. Patients in the group with a heterogeneous
population of antibodies [Table 3, group A1 (+/+)] had higher serum
T3 values and had slightly, albeit not statistically
different, greater T3/T4 molar ratios, larger
goiters, and a higher prevalence of ophthalmopathy. We could thus
identify a subgroup of patients with slightly more severe disease based
only on the heterogeneity of the cAMP-stimulating antibodies, as was
the case in the Korean studies (15, 16). As will be evident below, when
heterogeneity of both cAMP and AA release antibodies was considered,
the severity of the disease was more apparent.
|
|
|
and 2). Detailed examination of individual IgG showed that 11 of 14 IgG
with a decreased response in AA release at 0.1 mg/mL were able to
better stimulate AC at that concentration (Fig. 2, B vs. A).
This finding was our initial hint that individual serum samples are
heterogeneous in terms of signal transduction even when both
activities were present in the same patient.
When the clinical characteristics of patients whose IgG were able to
stimulate AA release in CHO-hTSHR cells (group A2 in Table 2) were
compared with those of patients whose IgG did not (group B2 in Table 2), the patients from group A2 displayed larger goiters and a higher
prevalence of ophthalmopathy. Thus, the AA release assay also, albeit
to a lesser degree than cAMP production, identified patients with more
severe GD (Table 2).
As a group, a significant reduction in AA release was found
(P = 0.022; range, 0–83% reduction) when AA release
assay was performed in the CHO-Mc1+2-transfected cells (Table 2). Based
on the comparison of the activities of individual IgG (Table 1), we
found that 11 of 23 patients (48%) had PLA2-stimulating
antibodies that require epitopes on the N-terminal portion of the TSHR
extracellular domain (residues 8–165).
We divided patients into two groups based on IgG activity in both
CHO-hTSHR cells and CHO-Mc1+2-transfected cells, i.e.
patients whose IgG exhibited residual activity in chimera [group A2
(+/+) in Table 3] vs. patients whose IgG lost their
stimulatory activity in chimera and whose activity depended on epitopes
in the N-terminal portion of the extracellular domain [group A2 (±)
in Table 3]. We found that patients in group A2 (+/+), who had IgG
whose epitopes were heterogeneous, tended to have greater TBII
activities and higher [3H]thymidine incorporation, and
serum T3 levels, but these were not statistically
significant mean increases when considered as a group (Table 3).
Comparison of AA release and cAMP production in CHO cells transfected with wild-type TSHR and Mc1+2 chimeras
We compared the activities of the 32 IgG in CHO cells
transfected with wild-type TSHR at a 1 mg/mL concentration both for AA
release and cAMP production. We found that 20 GD patients (63%) had
antibodies able to stimulate both PLA2 and AC, whereas 6
and 3 patients’ IgG were active only on AC or PLA2,
respectively (Table 1 and Fig. 3). Three
patients’ IgG (9.4%) were devoid of stimulating activity in both
PLA2 and AC assays (Table 1 and Fig. 3). These data
indicate that some patients with GD (9 of 32) have autoantibodies that
selectively activate only AC and PLA2 pathways. In
addition, the absence of a linear correlation between the two
activities (Fig. 3) suggests that independent antibodies from the same
patients stimulate AC or PLA2.
|
).
|
, 9 of the 20 IgG (45%) lost their
stimulatory activities in both AA release and cAMP assays; 6 and 1 IgG,
respectively, maintained their ability to increase AA release or cAMP
levels; and 4 IgG maintained their activities in both assays. Once
again, these data confirm the fact that Graves’ IgG are heterogeneous
in their ability to activate different signal transduction pathways.
Moreover, they demonstrate for the first time in a large cohort that
autoantibodies that activate different signal transduction pathways can
recognize different epitopes. Finally the data show, also for the first
time, that patients who have IgG with residual stimulating activity for
AC and PLA2 in the chimera, i.e. epitope
heterogeneity for both transduction pathways, have high TBII activity,
high serum levels of T3, a higher
T3/T4 molar ratio, and a tendency for higher
[3H]thymidine incorporation (Table 5).
|
|
It has been previously shown (15) that IgG-stimulating activities
measured in the CHO-25-hTSHR system tend to correlate with
T3 and T4 serum concentrations, whereas
stimulating TSHRAb activity assayed in the FRTL5 system tends to
correlate with goiter size. This observation led to the suggestion that
stimulating TSHRAb activities measured in the 2 different assay systems
may not represent identical populations of stimulating antibodies (15).
We also compared cAMP production in the 2 systems. Using the mean value
for stimulating TSHRAb activity assayed in FRTL5 cells and in the
CHO-25-hTSHR system as an arbitrary reference point for the 32 IgG
simultaneously evaluated in both assays, we found that 8 patients had
high values in both assays, 16 patients had low values in both assays,
3 patients had high values in CHO-25-hTSHR assay and low values in
FRTL5 cells, and 5 patients had low values in the CHO-25-hTSHR assay
and high values in FRTL5 cells (Table 1). Thus, 24 of the 32 patients
(75%) had concordantly high or low values in both assays, whereas 25%
had a different behavior, i.e. high in 1 assay and low in
the other. When patients were divided into the above 4 groups we found
that patients with high values in CHO-25-hTSHR exhibited greater TBII
activities and a tendency toward higher [3H]thymidine
incorporation and serum T3 levels, although this was not a
significant change when mean values between groups were compared (Table 6, bold values). Patients with
high values in FRTL5 cells tended to have larger goiters (Table 6,
bold and italicized values).
|
). Thus,
23 of the 32 patients (72%) had concordantly high or low values in
both assays, whereas 28% had a different behavior, i.e.
high in one assay and low in the other. When patients were divided in
these 4 groups, we found that patients with high values in CHO-25-hTSHR
assays tended to have a higher prevalence of ophthalmopathy (Table 7,
bold values), but no significant difference between TBII
activities, [3H]thymidine incorporation, serum
T3, T3/T4 molar ratio, or goiter
size (Table 7). IgG capable of
stimulating AA release in both assay systems exhibited, however, the
highest [3H]thymidine incorporation (Table 7), suggesting
that simultaneous measurement of AA release in both cell systems could
select a subpopulation of IgG able to stimulate thyroid cell
growth.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this report we present data consistent with previous studies (15, 16, 28) showing that the substitution of residues 8–165 of TSH receptor with homologous residues of LH/CG receptor results in a loss of the ability of Graves’ IgG to stimulate cAMP production, whereas cAMP signal transduction induced by TSH was preserved. Thus, Graves’ IgG and TSH require different TSHR domains with respect to AC stimulation.
It has previously been demonstrated that the PLA2 pathway is involved in the mechanism of action of Graves’ IgG (4, 5, 29), and that PLA2 stimulation and AA metabolites, particularly cyclooxygenase products, play a role in the regulation of thyroid cell growth in GD (5, 30). The molecular mechanism involved in stimulation of the PLA2/AA/PGE2 pathway has not been elucidated. Whereas much work performed in the last few years has been directed at showing which residues on the TSHR are responsible for Graves’ IgG activation of AC, no comparable data have been produced for PLA2. Thus, measurements of AA release induced by Graves’ IgG in a CHO cell model transfected with wild-type and mutant human TSH receptors, were likely to help clarify this point.
In this article we show that, like AC and PLC, the hTSHR is coupled to PLA2 in CHO-25-hTSHR cells and that the substitution of residues 8–165 abolishes the PLA2-stimulating effect of many IgG. We also present the novel result that substitution of residues 8–165 does not, in contrast, modify the stimulating effect of TSH on AA release. Thus, the data reported herein clearly show that, as in the case of AC activity (15), the PLA2-stimulating activity of Graves’ IgG, but not that of TSH, is largely eliminated with substitution of the N-terminal portion of the extracellular domain of the TSHR. Like previous work showing that some cAMP-stimulating TSHRAb have epitopes other than the N-terminal portion of the TSHR extracellular domain (15), we found herein that 11 of 23 (48%) IgG require epitopes on the N-terminus of the TSHR, whereas the other 12 IgG have epitope heterogeneity when measured in the AA release assay. A possible explanation for this finding could be that some PLA2-stimulating IgG, like TSH, interact with regions of the TSHR other than N-terminal portion of the extracellular domain. This possibility has been suggested in previous studies showing that the extracellular portion of LH/CGR has distinct sites that are important for the agonist activity and high affinity binding (31), and that stimulating TSHRAb bind peptides directed to the C-terminal portion of the TSHR (13, 32). Extensive studies performed in the last few years have shown that the high affinity TSH-binding site as well as the epitopes of the TSHR-blocking antibody associated with hypothyroidism are located on the C-terminal portion of the extracellular domain (6, 7, 28, 33).
Several aspects of the present work confirm and extend our previous data obtained in different cell systems. Thus, they show that Graves’ IgG signal transduction activity is heterogeneous and that the heterogeneity of IgG activity corresponds to a clinical heterogeneity of Graves’ patients (5). Moreover, our data clearly show that IgG activating both cAMP and arachidonic acid signal systems in CHO-hTSHR cells correlate with goiter size, as previously described in FRTL5 cells (4, 5). These data extend the results published by Kim et al. (15), who showed that patients in the epitope heterogeneity group have higher TBII activity, although they are more responsive to antithyroid drug therapy.
As previously described for the FRTL5 cell system (4), when Graves’ IgG were analyzed for their TBII activity and for their ability to stimulate AC or PLA2, no significant correlation was found between cAMP production or AA release. This suggests that even in CHO-hTSHR cell system, cAMP production, AA accumulation, and TBII assays do not measure the same populations of IgG. Early studies comparing FRTL5 cells and human thyroid primary cultures (34) and FRTL5 cells and CHO-hTSHR cells (15) showed that about 30% of Graves’ IgG represent different functional populations.
In conclusion, we show that PLA2 and AC stimulating IgG, as measured in both FRTL5 and CHO-hTSHR cells, correlate with different manifestations of GD (see Results). In addition, the present report shows that in CHO-hTSHR cells the TSHR is coupled to arachidonate as well as cAMP signals and that the N-terminal region of the extracellular domain of the TSHR is crucial for both AC- and PLA2-stimulating activity of Graves’ IgG. Moreover, it provides evidence demonstrating that AC- and PLA2-stimulating antibodies represent a heterogeneous population when assayed in CHO cells containing the hTSHR or the Mc1+2 hTSHR-LH/CGR chimeric receptor.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 These authors contributed equally to this work.
Received March 3, 1999.
Revised May 20, 1999.
Accepted May 25, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Ohto, N. Uozumi, T. Hirabayashi, and T. Shimizu Identification of Novel Cytosolic Phospholipase A2s, Murine cPLA2{delta}, {epsilon}, and {zeta}, Which Form a Gene Cluster with cPLA2{beta} J. Biol. Chem., July 1, 2005; 280(26): 24576 - 24583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Xing, H. Usadel, Y. Cohen, Y. Tokumaru, Z. Guo, W. B. Westra, B. C. Tong, G. Tallini, R. Udelsman, J. A. Califano, et al. Methylation of the Thyroid-stimulating Hormone Receptor Gene in Epithelial Thyroid Tumors: A Marker of Malignancy and a Cause of Gene Silencing Cancer Res., May 1, 2003; 63(9): 2316 - 2321. [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 |
