| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Immunology, Erasmus MC, 3000 DR Rotterdam, The Netherlands
Address all correspondence and requests for reprints to: Prof. Dr. Hemmo A. Drexhage, Department of Immunology, Erasmus MC, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. E-mail: h.drexhage{at}erasmusmc.nl.
The classical and well-known signs and symptoms of Graves’ disease are hyperthyroidism, diffuse goiter, and ophthalmopathy ("thyroid eye disease"), forming the so-called "Merseburger triad" of signs and symptoms. A less well-known sign and symptom is pretibial myxedema, which occurs in only a minority of patients.
Of the three classical signs and symptoms, thyroid eye disease is the most difficult to treat, and the eye condition forms a considerable burden for many of the patients because it is often painful and is practically always a cosmetically distressing disorder. In some instances, the ophthalmopathy leads to diplopia and sight loss, the latter due to damage of the optic nerve. Thyroid eye disease in its clinically overt form with gritty eyes, photophobia, chemosis, and clear exophthalmos occurs in 25–50% of cases of Graves’ disease; however, subclinical changes are detectable in almost all cases of Graves’ disease. Also, eye signs do not always occur at the same time as hyperthyroidism and goiter (1).
Medical treatment of thyroid eye disease (steroids, orbital radiotherapy, and surgery) has progressed little in the past 25 yr, mainly because little is known about its precise pathogenesis. The histopathology picture of the disorder is formed by an edema and predominantly mononuclear infiltration of the retro-orbital soft fatty tissues as well as of the retro-orbital muscles. The resulting swelling of these tissues leads to the proptosis of the eyes because the narrow bony orbit has not enough space to accommodate the edematous swollen, infiltrated, and inflamed tissues (2).
Because the eye disease is so closely connected to Graves’ disease and because this thyroid disorder is generally considered to be due to the activation of thyrocytes by thyroid-stimulating antibodies reacting to the TSH receptor (TSH-R), it will come as no surprise that one in general has focused in the studies on the pathogenesis of thyroid eye disease on self-reactivity of the immune system to retro-orbital tissue antigens.
One of the first immune theories explored in the 1960s and early 1970s was the theory of Kriss (28) based on an anatomical connection between the lymphatic system of the eyes and that of the thyroid. This was thought to make a retrograde flow possible from the thyroid to the orbits of one of the major thyroid antigens present in lymph thyroglobulin. An antibody reaction to thyroglobulin, which is frequent in Graves’ disease, would then explain the inflammation of the retro-orbital tissues. Present literature showing only a poor binding of thyroglobulin to orbital structures makes this mechanism less likely to play a prominent role (3).
In the 1970s, antibodies to striated muscle cells were detected in the serum of patients with thyroid eye disease, and of some of these antibodies the antigen has been found (4). However, antibodies to striated muscle cells can be found in many conditions in which muscle cells are destroyed, and one presently considers that at least the majority of these muscle-directed antibodies (if not all) are not the cause, but the consequence of the inflammatory reaction of the retro-orbital muscles (4).
In the 1980s, a third theory was developed based on the full recognition at that time that autoantibodies to the TSH-R were an important causal mechanism of Graves’ hyperthyroidism and goiter. These antibodies are capable of stimulating the hormone release and growth of thyroid cells. This receptor antibody theory on thyroid eye disease widened the concept of thyroid receptor antibodies to the contents of the eye: Rotella et al. (5) were able to show that Igs prepared from the sera of Graves’ disease patients were capable of stimulating the collagen biosynthesis of cultured human skin fibroblasts, and the authors assumed that these were at least in part ophthalmopathy-stimulating antibodies. Indeed, more or less at the same time one started to accept the fibroblast as one of the important targets of the immune attack in the Graves’ eye because Graves’ orbital fibroblasts expressed major histocompatibility complex class II and other important immune molecules (6). However, it was also generally believed that the TSH-R was not expressed on cells other than thyroid cells. Hence, the theory of TSH-R-related antibodies stimulating orbital fibroblast as a prime cause of thyroid eye disease was refuted. Moreover, exophthalmus is not part of neonatal Graves’ disease, the transient hyperthyroid disorder due to the transplacental passage of TSH-R antibodies from a Graves’ mother to her newborn. Nevertheless, a role for autoreactive immune cells to orbital components and in particular to antigens on orbital fibroblasts was still open.
In the 1990s, orbital fibroblasts turned out to be special fibroblasts. They belong in majority to the so-called subset of Thy-1-negative fibroblasts, which are capable of differentiating into mature adipocytes when treated with peroxisome proliferator-activated receptor 
agonists as well as with cAMP-enhancing agents (7). Thy-1-positive fibroblasts do not but are able to differentiate into myo-fibroblasts when treated with TGF-ß. There are strong indications that Thy-1-negative orbital fibroblasts differentiating into adipocytes start to express the full, functional TSH-R in appreciable amounts (8). It must be borne in mind, however, that the presence of nonfunctional TSH-R peptides is sufficient to drive a T-cell-mediated autoreactive immune reaction.
In this issue of Endocrinology, Han and Smith (9) report that interferon (IFN)- (produced by Th1 cells) and IL-4 (produced by Th2 cells) both up-regulate the in vitro production of hyaluronan from IL-1ß-stimulated orbital fibroblasts by enhancing the enzyme hyaluronan synthase-2. IL-1ß is an important proinflammatory molecule produced by monocytes, macrophages, and dendritic cells, also instrumental in the stimulation of T cells. Hyaluronan is an important hydrophilic glycosaminoglycan molecule playing a major role in the edematous reaction of the Graves’ retrobulbar tissues.
In previously reported studies, the group of Smith has shown that orbital fibroblasts show an exaggerated response to IL-1ß (9, 10) (most likely due to a reduced ability to produce IL-R antagonist, one of the natural IL-1 antagonists) (11) and produce not only excessive amounts of hyaluronan, but also of modulators of matrix metalloproteinase activity (12) and of the inflammation regulating molecule prostaglandin synthase 2 resulting in an enhanced production of prostaglandin E2 (11). In the present report, Han and Smith (9) show that IFN- and IL-4 both attenuate the IL-1ß-induced higher production of prostaglandin synthase 2. Because T cells predominate in later stages of the orbitopathy, the authors conclude from their data that the T-cell-dominating infiltrate in later stages of the orbitopathy still has an edema eliciting profile, yet has reached another stage of immune regulation.
In addition to this work on cytokine-induced activation of orbital fibroblasts, the group of Smith has revived in recent years the IGF-I receptor (IGF-IR) as a possible target of the Graves’ autoimmune reaction (13). In 1994, Weightman et al. (14) had described the IGF-IR as a possible target of the "ophthalmopathy inducing antibodies." The group of Smith found the serum of Graves’ patients to contain antibodies to displace IGF-I from its binding sites on orbital fibroblasts (which naturally over express the IGF-IR), to activate the receptor and to exacerbate the production of the chemokine RANTES, of the cytokine IL-16, but also of the glycosaminoglycan hyaluronan.
Compiling these data from Smith and those collected by others in recent and past years, a picture is emerging in which a concerted action between antigen-presenting cells, T cells, B cells, and the special retro-bulbar preadipocyte is essential in the pathogenesis of thyroid eye disease. The immune cells use as communication language in these interactions at least two autoantigens: the TSH-R and the IGF-IR (and peptides derived of these autoantigens) (Fig. 1
).
|
When autoreactive Th1 and Th2 cells and autoantibodies specific for the TSH-R and IGF-IR have been produced, they are able to encounter their respective antigens in the orbit. The IGF-IR antibodies will be instrumental to induce an enhanced production of hyaluronan, yet this is not sufficient to precipitate the eye disease as neonatal Graves’ disease learns us. The IGF-IR antibodies are, however, in the view of Smith et al. also able to induce the production of RANTES and proinflammatory IL-16 from orbital fibroblasts. These factors will facilitate a local infiltration and activation of dendritic cells, macrophages, and T cells. The dendritic cells and macrophages are able to produce cytokines like IL-1ß and TGF-ß (16) capable to exaggerate the receptor antibody-driven hyaluronan production. Among the infiltrated T cells, there will be TSH-R and IGF-IR peptide-specific T cells. For such accumulated autoreactive T cells, there is no need to recognize the full IGF-IR or TSH-R; on the contrary, they solely react to antigenic peptides derived of the receptors and displayed on antigen-presenting cells in the orbit. When they do so, they will start to produce cytokines like IL-4 (the Th2 cells) and IFN- (the Th1 cells). IFN- will (further) activate the locally accumulated macrophages, and both IFN- and IL-4 will aid in a further activation of the hyaluronan production.
Obviously, in the pathogenesis of pretibial myxedema, mechanisms very similar to those in the orbit might play a role targeting the pretibial fibroblasts (17). But does the recognition of this multistep and concerted action of antibodies and cellular immune components in the pathogenesis of thyroid eye disease also have consequences for our ideas on the pathogenesis of the other two signs and symptoms of the Merseburger triad, i.e. hyperthyroidism and goiter development?
At present, it is generally accepted that a polyclonal antibody response to the TSH-R involving antibodies that agonize and/or antagonize the function of the TSH-R determine the extent of hyperthyroidism and goiter formation in Graves’ disease. These antibodies affect the adenyl cyclase but also other second messenger routes (e.g. the phospho-inositol turnover cycle) in the thyrocyte (18, 19). But is that the whole story? Why is it that neonatal hyperthyroidism is not an exact titration of the hyperthyroidism of the mother if the TSH-R antibodies play the decisive role? Why is it that TSH-R immunization in mice does clearly elicit TSH-R antibodies but only results in a rather mild hyperthyroid and goitrous disease with very questionable eye signs (20)? Would antigen-presenting cells and T-cell-immune reactivity to the IGF-IR and the TSH-R also play a role in Graves’ hyperthyroidism and goiter formation? There are reports suggesting that there is a direct effect of infiltrated and activated leukocytes in Graves’ goiters on the control of adjacent thyroid cell proliferation via the interaction of integrin signaling (21). Moreover, in vitro thyroid cell growth and hormone secretion are influenced by coculture with antigen-presenting cells, and this process is mediated by cytokines such as IL-1ß and IL-6 (22). Also, IL-1ß and T-cell-derived cytokines are able to influence in vitro thyroid cell function and growth (23). And what is the effect, if any, of the IGF-IR antibodies on the thyroid cells in Graves’ disease? It is almost unanimously accepted that TSH and IGF-I synergistically activate thyrocyte proliferation, and p66Shc has been described as a key molecule in this synergistic activation in one of the recent issues of this journal (24). Could antibodies to the TSH-R and IGF-IR also act synergistically in this way? There is indeed some evidence that IGF-IR antibodies might induce the proliferation in thyroid cells in vitro (25). Or are IGF-IR antibodies capable of targeting preadipocyte-like fibroblasts in the thyroid? Although Thy-1- negative fibroblasts could not be found in the thyroid (26), there is a report that IFN- increases thyroidal fat in mice (27).
So, we are left with the question of whether TSH-R antibodies are indeed the decisive driving force behind the signs and symptoms of Graves’ disease or if there is perhaps more that meets the eye in Graves’ disease and its Merseburger triad of symptoms.
 |
Footnotes |
|---|
Received October 13, 2005.
Accepted for publication October 19, 2005.
|
References |
|---|
 Top References |
|---|
in cultured retroocular fibroblasts derived from patients with Graves’ ophthalmopathy and pretibial dermopathy. J Clin Endocrinol Metab 73:307–313[Abstract] (PPAR), and thyrotropin receptor by PPAR agonist in human orbital preadipocyte fibroblasts. J Clin Endocrinol Metab 87:2352–2358. J Immunol 174:3072–3079. Int J Exp Pathol 86:97–106[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  A. Iagaru and I. R. McDougall Treatment of Thyrotoxicosis J. Nucl. Med., March 1, 2007; 48(3): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Salvi, G. Vannucchi, I. Campi, N. Curro, D. Dazzi, S. Simonetta, P. Bonara, S. Rossi, C. Sina, C. Guastella, et al. Treatment of Graves' disease and associated ophthalmopathy with the anti-CD20 monoclonal antibody rituximab: an open study Eur. J. Endocrinol., January 1, 2007; 156(1): 33 - 40. [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 |
