Adipogenesis in Thyroid Eye Disease

Abstract

purpose. Adipogenesis contributes to the pathogenesis of thyroid eye disease (TED). Thyrotropin receptor (TSHR) transcripts are present in orbital fat. This study was conducted to determine whether they are expressed as functional protein, and if so, whether this is restricted to TED orbits or to a particular stage in adipocyte differentiation.

methods. Samples of fat were obtained from 18 TED-affected orbits and 4 normal orbits, and 9 were obtained from nonorbital locations. Frozen sections were examined by immunocytochemistry using monoclonal antibodies specific for the human TSHR. Samples were disaggregated and the preadipocytes separated from the mature by differential centrifugation and cultured in serum-free or DM and examined for morphologic changes, oil red O and TSHR staining, and TSH-induced cyclic adenosine monophosphate (cAMP) production.

results. Marked immunoreactivity was observed in frozen sections from all three TED samples and faint staining in both normal orbital fat samples. In vitro, 1% to 5% of preadipocytes displayed TSHR immunoreactivity in five of six TED and two of three normal orbital samples and in three of five nonorbital samples. Differentiation, was induced in all 14 orbital samples. Three of four nonorbital samples contained occasional differentiated cells. Fifty percent to 70% of differentiating cells demonstrated receptor immunoreactivity. Two of three TED and four of four nonorbital preadipocytes in DM and/or mature adipocytes displayed a TSH-mediated increase in cAMP.

conclusions. The results indicate that orbital fat TSHR transcripts are expressed as protein, which can be functional. This is not aberrant in TED orbits, although expression may be upregulated. The majority of preadipocytes undergoing differentiation express the receptor, indicating a key role for this population in one mechanism for increasing orbital volume.

Thyroid eye disease (TED) is an autoimmune condition most frequently associated with Graves’ disease (GD). ¹ In GD, the thyrotropin receptor (TSHR) is the target of stimulating autoantibodies (TSAB), which mimic the action of TSH by increasing intracellular cyclic adenosine monophosphate (cAMP) levels. This leads to increased thyrocyte function and growth, independent of the hypothalamic-pituitary-thyroid axis. ²  

The pathogenesis of TED is less clear. Most of the signs and symptoms can be explained by the increase in volume of the orbital contents. The extraocular muscles (EOMs) become grossly enlarged, mostly because of edema. There are few microscopically visible changes in the myocytes themselves, although in advanced disease they become heavily fibrosed. ¹ Apart from the edema, two further mechanisms increase orbital volume: the production of glycosaminoglycans (by orbital fibroblasts), which absorb water and swell ³ and hyperplasia of the adipose tissue. ⁴ Together, these cause proptosis and compression of the optic nerve, which can result in diplopia and in extreme cases, loss of sight. 

What of the biologic mediators? Adipogenesis is induced by hormones and factors outside the immune system, although interleukin (IL)-11, interferon-γ, and tumor necrosis factor (TNF)-α have an inhibitory effect. ⁵ Certainly, edema and glycosaminoglycan production can be stimulated by cytokines elaborated by the infiltrating immune cells ⁶ but considerable gaps remain in our knowledge, including the absence of a target autoantigen, which would explain why activated T lymphocytes home to the orbit. 

Evidence is accumulating in favor of the antigen in GD, the TSHR, also having a role in the pathogenesis of TED. From the clinical standpoint, TED is usually found in GD patients who have the highest titers of TSAB, although patients with euthyroid eye disease have also been described. ⁷ It has been reported that eye disease is exacerbated after radioiodine ablation of the thyroid gland, but this is not a universal finding. ⁸ Recently, we have described an animal model for TED, induced in BALBc mouse recipients of TSHR-primed T cells, which is further support for the receptor’s playing a major role. ⁹  

Before the molecular cloning of the TSHR, TSH binding sites, TSH-mediated adenylate cyclase activity, and TSH-induced lipolysis had been reported in orbital and other fat depots in rodents. ^{10 11 12 13} Results for human tissues have been more controversial, with some investigators failing to show TSH binding to extrathyroidal tissues, ¹⁰ whereas others have demonstrated low ¹⁴ or high-affinity binding ¹⁵ to human adipocyte membranes. The lipolytic effects of TSH are more evident with brown fat, and this may explain why TSH-induced lipolysis in humans is detectable in the neonate but is virtually extinguished by the age of 10 years. ¹⁶  

The application of molecular methods has led to the confirmation of many of the earlier functional studies. In rodents, adipose tissues from several locations have been shown to express TSHR transcripts. ¹⁷ In a series of experiments, a functional TSHR was cloned from rat fat cDNA ¹⁸ ; receptor expression was shown to be associated with differentiating preadipocytes, first by using a cell line ¹⁹ and subsequently with freshly isolated rat preadipocytes. ^{20 21} The same group has investigated the effect of TSH on rat preadipocyte cultures and shown that TSHR transcriptional control in these cells is different from that observed for the thyroidal TSHR in vitro. ²²  

In humans, recent Northern analysis data have revealed clear TSHR transcripts in infant abdominal fat, but the levels are substantially reduced in the equivalent adult tissue. ²³ In human disease, several methods, including reverse transcription–polymerase chain reaction (RT-PCR), ²⁴ liquid hybridization analysis, ²⁵ and Northern blot analysis, ²⁶ have indicated that TSHR transcripts may be present in the orbit. Conclusions are sometimes conflicting, because tissues were analyzed after a period in culture versus tissue ex vivo and the primers used were able to detect TSHR variant transcripts or because contaminating genomic DNA was amplified. ²⁷ In this context, in earlier studies (reviewed in Reference 27) we found no evidence for full-length TSHR transcripts in orbital tissues, but these did not include orbital fat. We have demonstrated TSHR transcripts in a single specimen of orbital fat from a patient with TED by Northern blot analysis ²⁶ when signals in normal orbital and abdominal fat were at the limit of detection. 

In recent studies, immunocytochemistry (ICC) using a monoclonal antibody to the membrane-spanning region of the TSHR has demonstrated immunoreactivity for the receptor in orbital fibroblasts in culture and a small number of TED tissue specimens ex vivo. ²⁸ A recent study from the same group suggests that functional receptor expression is induced by differentiation. ²⁹  

The purpose of this study has been to investigate the role of the TSHR in adipogenesis, by determining whether human orbital fat TSHR transcripts are expressed as functional protein. If yes, is this aberrant in TED and/or associated with a particular stage in the preadipocyte to mature adipocyte differentiation pathway? 

Materials and Methods

Sample Collection and Patients Studied

All tissues were obtained with informed consent. Orbital adipose tissues (OAT, predominantly white fat) were from 18 patients with TED undergoing surgical decompression for their eye disease, from 2 with normal fat who were treated for other noninflammatory conditions, and from 2 postmortem donors who had no eye disease. Nonorbital fat samples (non-OAT)—six omental, one perirenal, and two cervical—were obtained during routine surgery. 

Samples for cryostat sectioning were snap frozen in liquid nitrogen and stored at −80°C until use. In all instances tissues for culture were kept moist on physiological saline–soaked gauze and maintained on ice until the disaggregation process began (<24 hours). 

Eleven of eighteen patients with TED had active disease, and 17 were female. Six of 18 had undergone thyroidectomy, 6 had received radioiodine, and 6 had been treated with antithyroid drugs for thyroid disease. A further 2 of the 18 were euthyroid. All treatment for thyroid disease, with the exception of thyroxine replacement, had been completed at least 2 years before sample collection and analysis. Six of 18 had received no previous treatment for eye disease, 7 had been treated with steroids, and 5 had received orbital radiotherapy. 

The study was performed with the ethical approval of the Bro Taff Health Authority and adhered to the Declaration of Helsinki regarding research involving human subjects. 

Immunocytochemistry

Five-micrometer cryostat sections were plunged into acetone and air dried. They were then incubated for 15 minutes at room temperature in phosphate-buffered saline (PBS)-Tween followed by 45 seconds with 0.228% periodic acid, to inhibit endogenous peroxidase, and were washed in PBS-0.02% gelatin-0.05% Tween (PGT). Sections were incubated overnight, at 4°C with primary antibody, washed three times in PGT, and incubated using the ABC indirect immunoperoxidase staining system (Vector, Burlingame, CA), according to the manufacturer’s instructions. All sections were counterstained with Meyer’s hematoxylin, dehydrated, and mounted in DPX mountant medium (Fisher Scientific UK Ltd, Loughborough, UK). 

In all experiments, sections of GD thyroid and colon were included as positive and negative control samples, respectively. Specimens were also subjected to the ICC procedure but using an irrelevant isotype–matched control monoclonal antibody or omitting the first antibody. 

Antibodies Used in the Study

We used three monoclonal antibodies, BA8 and 3G4 (both used at 1:100) ³⁰ and A10 (used at 1:500). BA8 reacts with conformational epitopes present only on the human TSHR; 3G4 recognizes a linear epitope, residues 352-362, found in the rat, canine, and murine receptors; and A10 binds amino acids 22-35. ³¹  

Disaggregation and Collagenase Digestion

The adipose tissues were finely chopped and collagenase digestion solution (3 mg/ml collagenase type II, 1.5% bovine serum albumin[ BSA] in Hanks’ balanced salt solution) was added at a ratio of 1:4. Disaggregation was at 37°C, until the tissue appeared completely digested when it was filtered through gauze. Two to 3 ml of phthalic acid dinoyl ester was added followed by centrifugation at 1500 rpm at 4°C for 5 minutes to separate mature adipocytes from preadipocytes. ³²  

Mature adipocytes form a layer on the top of the oil, whereas the preadipocytes form the pellet, along with red blood cells which were lysed by ammonium chloride (0.154 M in 10 mM potassium carbonate and 0.1 mM EDTA). After a further centrifugation at 1500 rpm at 4°C, for 5 minutes, the cell pellet was resuspended in serum-containing medium (SCM) and transferred to tissue culture dishes. Some of the cells were plated on coverslips and subjected to ICC within 5 to 10 days (see below). 

Culture Conditions and Differentiation Protocol

The preadipocytes were grown in SCM (Dulbecco’s modified Eagle’s medium [DMEM]-Ham’s F-12 1:1 medium containing 10% fetal calf serum[ FCS]), plated in 60-mm or 100-mm dishes depending on the sample size, and incubated at 37°C in the presence of 5% CO₂ with medium changes routinely every 2 days. 

Cultures in dishes were trypsinized and split 1:4 after 5 to 10 days when approximately 75% confluent. In all cases only a single passage was performed. The cells were plated on coverslips and maintained in culture in SCM to increase cell number; in some cases confluence was achieved. 

At this point, some of the coverslips were maintained in SCM, and the remainder were washed three times with sterile PBS to remove any FCS, and serum-free medium (SFM) alone or containing differentiation reagents was added. The differentiation medium (DM) contained biotin (33 mM), pantothenic acid (17 μm), apotransferrin (10 μg/ml), T3 (0.2 nM), cortisol (100 nM), and insulin (500 nM). Iso-butyl-methyl-xanthine (IBMX; 0.25 mM) was also added to the DM on day 1 of differentiation. The medium was exchanged on day 3 and after than, every third day, with DM without IBMX. Differentiation took place over 10 to 14 days. After this period, the cells were examined for morphologic changes and oil red O staining (described later). They also underwent ICC for TSHR expression, as described earlier but with the inclusion of an avidin-biotin blocking step. 

Control cell lines were also subjected to the same culturing and differentiation protocols. JP09, a Chinese hamster ovary (CHO) cell line expressing approximately 10⁴ receptors per cell, provided a positive control for TSHR immunoreactivity, and a skin fibroblast cell line, HCA2 (a gift from James R. Smith) was used as a negative control. 

Oil Red O Staining

Saturated oil red O was diluted to 0.3% and filtered immediately before use. Coverslips for oil red O staining were fixed in 60% isopropanol and stained with oil red O in a closed container for 10 to 15 minutes. The background was cleared by rinsing in 60% isopropanol, and then the samples were washed in water. The cells were counterstained, washed in water, and mounted in gel (Glycergel; Dako, High Wycombe, UK). 

TSH-Induced cAMP Production at Various Stages of Adipogenesis

In some experiments, 5 × 10⁵ freshly isolated preadipocytes and adipocytes were seeded separately into 12-well plates in SCM. Forty-eight hours later, they were incubated for 2 hours in SCM (preadipocytes as adherent cells but mature adipocytes in suspension) containing 10⁻⁴ M IBMX and 0, 1, or 10 mU/ml bovine TSH. cAMP was extracted using 0.1 M HCl and measured using an in-house radioimmunoassay. ³³ TSH-induced cAMP production was also measured in preadipocytes cultured for 14 days in SFM or DM, as described earlier. Both cell populations were switched to SCM for the 2-hour incubation with IBMX and TSH. In addition, TSH-induced cAMP production was measured in patients with GD and multinodular goitre (MNG) by disaggregation and culturing of thyroid cells in SCM for 48 hours, followed by 2-hour incubation with IBMX and TSH. 

The results of (at least duplicate) experiments were calculated as femtomoles cAMP per well and expressed as a percentage of the 0 TSH value (100%). 

Results

TSHR Immunoreactivity in Sections of Orbital Adipose Tissue

When BA8, 3G4, or A10 was applied to sections of thyroid, intense staining was obtained at the basolateral surface of the follicular cells, in keeping with the known location of the TSHR, which was absent in sections in which the first antibody was omitted or substituted with an isotype control (Fig. 1A 1B ). No specific staining was obtained when the TSHR monoclonal antibodies were applied to a range of nonthyroidal tissues (data not shown). 

Adipose tissue sections from two normal orbits examined displayed faint but visible immunoreactivity with BA8 (Fig. 1C ; note also the homogenous cell size and that some of the cells did not stain). In contrast, marked staining in the thin rim of cytoplasm in the classic adipose “honeycomb” was obtained in the three TED adipose tissues that were sectioned (Fig. 1D) , particularly in the smaller fat cells. Furthermore, areas of strongest staining were often where several cells juxtapose and many nuclei were present, suggestive of a different cell population between the mature adipocytes. The degree of staining did not seem to be affected by the proximity of the infiltrating lymphocytes (Fig. 1F) . No staining was obtained in the absence of the first antibody (Fig. 1E) . 

A Small Proportion of Preadipocytes Express the TSHR

Fifteen TED, two normal orbital, and nine nonorbital fat samples were cultured. After 5 to 10 days in culture the preadipocytes were elongated cells with the appearance of fibroblasts (Fig. 2A ). More rounded cells were also present in four TED fat samples and, unlike other preadipocyte cultures, these samples had cells containing small lipid droplets demonstrated by oil red O staining (Fig. 3C ), before culture in the DM. 

When cultures could be examined after 5 days, 1% to 5% of the cells in seven orbital and three nonorbital samples showed positive labeling, which took the form of “peppering” with the BA8, A10, and/or 3G4 monoclonal antibodies specific for the human TSHR (Figs. 3A 3B) and was not obtained in the absence of primary antibody (data not shown). The highest percentages of TSHR-positive cells were observed in the samples with oil red O staining. In specimens in which the period in culture was extended to 10 days, because of the small sample size, positive-staining cells were not detected. In the JP09-positive control, more than 90% showed positive staining (Fig. 3D) , which was absent on the negative control HCA2 fibroblast cell line (data not shown). 

The Majority of Preadipocytes Differentiating to Mature Adipocytes Express the TSHR

When confluent preadipocytes from TED or normal orbits or nonorbital adipose tissues were maintained for a further 10 to 14 days in 10% SCM or SFM alone, there were no morphologic changes or oil red O–positive cells, and TSHR expression, assessed by BA8–3G4-A10 staining, became undetectable. 

When cultured in SFM containing differentiation factors, 14 of 14 of the orbital and 3 of 4 of the nonorbital preadipocyte cultures showed signs of differentiation. Morphologic changes involved the cells’ becoming more rounded, and small vacuoles appeared (Fig. 2B) . With time the vacuoles enlarged but were shown to be lipid filled from early on by the positive oil red O staining, (Fig. 3F) . Furthermore, the majority of cells demonstrating these morphologic changes also expressed the TSHR, although differentiating cells negative for BA8–3G4-A10 staining were present (Fig. 3E) . 

The degree of differentiation induced varied, with the highest in TED orbital and the lowest in nonorbital fat. In addition, all differentiating cultures contained a proportion of cells with the vacuolated morphology of a developing adipocyte but completely devoid of oil red O staining lipid droplets (Fig. 3E) . For comparison, a sample of mature adipocytes obtained from the top layer after centrifugation is shown in Figure 2C . There was no correlation between the degree of induced differentiation and the differing previous treatments of either thyroid or eye disease. No morphologic changes that indicated adipocyte differentiation or oil red O staining were induced in the JP09 or HCA2 cell lines (data not shown). Results of all preadipocyte cultures are summarized in Table 1 . 

Mature Adipocytes and Differentiating Preadipocytes Express a Functional TSHR

TSH-induced cAMP production was measured in four cell populations representing different stages in adipogenesis in three samples of TED orbital fat and four nonorbital fats, although it was not possible to test all four populations using a single sample. Results were compared with GD and MNG thyroid cells. 

Preadipocytes, either freshly isolated or cultured in SFM, were unresponsive to TSH at 1 and 10 mU/ml. The majority of preadipocyte samples in DM and freshly isolated mature adipocytes demonstrated increased cAMP levels with 10 mU/ml TSH. This contrasts with the thyroid cells, which were responsive at 1 mU/ml TSH and more. Results of a typical experiment are shown in Table 2 . In these preliminary studies, no difference in TSH responsiveness was observed between samples of TED OAT and non-OAT, as summarized in Figure 4 . 

Discussion

Our results provide clear evidence that the TSHR is expressed, at several differentiation stages in the orbital adipose compartment and demonstrate that this expression is not restricted to TED orbits. Furthermore, we have shown that, at least in the later stages of differentiation, the TSHR is functional, both in orbital and nonorbital fat. 

A small percentage of preadipocytes in all fat depots examined expressed the TSHR. The increased percentage in the TED fats was associated with signs of active adipogenesis, evidenced by oil red O staining, indicating that receptor expression is increased by differentiation. This parallels recent findings in rodent adipose tissue. ²⁰  

In vitro adipocyte differentiation can be stimulated in preadipocytes from a number of locations, including the orbit. ⁴ However, this is the first demonstration that the majority of cells entering the differentiation program express the TSHR, even in a protocol containing no TSHR agonists. Furthermore, the results obtained in vitro are supported by the ex vivo finding that most intense TSHR immunoreactivity is present on the smallest, and presumably the most recently formed, adipocytes in TED orbital tissue sections. 

The work lends further support for the receptor’s being an autoimmune target in TED, although orbital TSHR expression is not aberrant. Thus, some other explanation must be found for the occurrence of eye disease only in some patients with GD, even though all Graves’ disease is the result of an autoimmune response to the receptor. 

If aberrant expression cannot provide an explanation, a second possibility is that the receptor expression is upregulated. This is certainly supported by our Northern analysis data ²⁶ and the recent report of an RNase protection assay that showed increased TSHR transcripts in TED. ²⁹ Clearly, receptor expression is increased during the differentiation process, but whether this is preceded by upregulation in the number or intensity of preadipocytes that express receptor is not yet known. 

A further possibility is that receptor function in TED pre- or mature adipocytes differs from that of other fat depots. If a functional receptor is expressed on orbital preadipocytes, circulating TSAB could stimulate their proliferation (increasing the percentage of receptor-positive cells) and/or differentiation. Conversely, functional receptor on mature fat cells would result in lipolysis, as in the neonate. ¹⁶ If this process exceeded adipogenesis. there would be no increase in orbital fat, as is the case in some TED patients. Our preliminary data suggest that preadipocytes do not respond to TSH by increasing cAMP levels, indicating either that the receptor is not functional or is not coupled to adenylate cyclase. Alternatively, the proportion of TSHR-expressing preadipocytes may be too small to produce a measurable increase in cAMP. In other studies, the equivalent of our freshly isolated preadipocytes in the rat responded slightly to TSH ²⁰ which was also shown to increase proliferation but to decrease differentiation of these cells. ²¹  

In contrast, TSH was found to stimulate TSHR expression in TED orbital preadipocyte fibroblasts. ³⁴ Our results from mature adipocytes and differentiating preadipocytes (i.e., that 10 mU/ml TSH were required to stimulate cAMP production, whereas thyroid follicular cells responded to 1 mU/ml in the same protocol) are similar to those obtained comparing rat tissues. ²⁰ This implies that the adipose compartment is unresponsive to physiological TSH concentrations. 

The final possibility is that the receptor merely serves as a beacon to enable receptor-specific immune cells to home to the orbit and that TED is the result of a particular type of autoimmune response to the TSHR. Our animal model tends to support this theory, ⁹ because orbital disease is induced only in mice with severe T-helper cell type 2 (Th2) thyroiditis characterized by abundant B cells and T cells expressing IL-4 and not in a strain showing destructive Th1 thyroiditis. However, if this were the case, all fat depots, not just those in the orbit, would be potential targets of the Th2 receptor autoreactivity. Accumulation of fat in the orbit would have a more pronounced effect because of the constricted space. Alternatively, the balance between adipogenesis and lipolysis could vary in different fat depots so that no net accumulation occurs in some depots, whereas adipose tissue builds up in the orbit. Future studies investigating the function of the TSHR in the minority of receptor-expressing preadipocytes may shed more light on this issue. 

Ours is the first study to concentrate on the orbital adipose compartment, whose importance was first sign-posted by the work of Feliciello et al. ²⁴ Studies from another group on early-passage orbital fibroblasts and preadipocytes in a differentiation protocol have yielded conflicting results, ^{28 29} perhaps because of the application of different TSHR monoclonal antibodies with which staining was also obtained on adrenal and thymus. ³⁵ We have applied three monoclonal antibodies to the TSHRs recognizing diverse epitopes and have obtained essentially the same result with all three. 

In conclusion, we provide clear evidence for functional TSHR expression in the orbital adipose compartment which is not aberrant in TED or restricted to the orbit but is upregulated in TED. Furthermore, because the diameter of a preadipocyte is 30 to 40 μm, whereas that of a fat cell is up to 120 μm, we have demonstrated the importance of the TSHR-expressing preadipocyte population in at least one of the mechanisms responsible for increasing the volume of the orbital contents in TED. 

 Supported by a University of Wales College of Medicine, Division of Medicine Studentship and Brahms Diagnostica, Berlin, Germany.

 Submitted for publication December 27, 1999; revised June 6, 2000; accepted June 16, 2000.

 Commercial relationships policy: F (ML); N (all others).

 

Figure 1.

 

Immunocytochemical staining of 5-μm cryostat sections. (A) Graves’ thyroid in the absence of primary antibody; (B) Graves’ thyroid with BA8: staining at basolateral edge of the thyrocyte (arrow); (C) normal orbital adipose tissue with BA8: faint staining in many adipocytes of homogenous size; (D) TED orbital adipose tissue with BA8: staining in most adipocytes, which were of heterogenous size, where cells juxtapose (arrow); (E) TED orbital adipose tissue in the absence of primary antibody; and (F) TED orbital adipose tissue with BA8: staining was not dependent on the lymphocytic infiltrate. Magnification, ×400.

Figure 2.

 

Phase-contrast micrographs showing various stages of adipocyte differentiation. (A) TED orbital preadipocytes freshly isolated before differentiation: elongated cells that resemble fibroblasts; (B) TED orbital preadipocytes after 14 days in DM: rounded morphology, with droplets appearing in many of the cells (arrow); and (C) TED orbital mature adipocytes freshly isolated: rounded cells with a large lipid vacuole. Magnification, ×400.

Figure 3.

 

ICC staining at various stages during adipocyte differentiation. (A) TED orbital preadipocytes, before differentiation stained with A10: elongated fibroblast-like cells (arrow); (B) TED orbital preadipocytes before differentiation stained with BA8: isolated fibroblast-like cells and more rounded cells with droplets (arrow); (C) TED orbital preadipocytes, before differentiation stained with oil red O: in many of the cells, small lipid droplets indicative of the differentiation process’s having begun in vivo; (D) JP09 cell line expressing TSHR with BA8: more than 90% of cells labeled; (E) TED orbital preadipocytes after 14 days in DM and stained with 3G4: rounded cells containing many vacuoles and showing intense labeling but no TSHR immunoreactivity visible in some differentiated cells (arrow); (F) TED preadipocytes after 14 days in DM stained with oil red O: lipid droplets in cells with the morphology of adipocytes. Magnification, (A) ×200; (B through F) ×400.

Table 1.

 

Summary of Preadipocyte Culture and Differentiation

Tissue	Preadipocytes			After Differentiation
		TSHR+	ORO+	±	+	++	ORO+	TSHR+
TED OAT	5/6	4/7	4/12	5/12	3/12	8/8	8/8
N OAT	2/3	0/1		2/2		2/2	2/2
non OAT	3/5	1/3	2/4	1/4		2/2	2/2

 Summary of preadipocyte culture and differentiation assessed in orbital adipose tissue from 15 patients with thyroid eye disease (TED OAT) and 2 with normal orbits (N OAT) compared with 5 samples of nonorbital (non-OAT) fat. Pradipocytes refers to the pellet obtained after density gradient centrifugation and examined after 5 days in culture before the differentiation protocol. After differentiation describes the preadipocytes after a period of up to 14 days in the differentiation medium. TSHR+ gives the number of samples in which TSHR immunoreactivity was observed using BA8, 3G4, and/or A10 monoclonal antibodies. ORO+ gives the number of samples in which oil red O staining lipid droplets were observed. Degree of differentiation was also assessed morphologically. ±, one to two small differentiated cells per field; +, up to three large differentiated cells per field;++ , more than three large differentiated cells per field.

Table 2.

 

TSH-Induced cAMP Production

TSH (mU/ml)	Thyroid SCM	TED OAT			non-OAT
				Pre SFM	Pre DM	Pre SCM	Mat SCM
0	360 (390, 330)	90 (93, 87)	105 (99, 111)	289 (282, 297)	60 (78, 45)
1	915 (990, 840)	111 (108, 114)	207 (210, 201)	197 (212, 179)	70 (60, 78)
10	2820 (2310, 3300)	75 (63, 90)	324 (375, 273)	210 (204, 214)	270 (186, 360)

 Data are expressed as femtomoles per well and are from a representative experiment. cAMP production was measured in all tissues over 2 hours in SCM at 37°C in the presence of 10⁻⁴ M iso-butyl-methyl-xanthine and varying doses of thyrotropin. Thyroid, primary culture of multinodular goiter follicular cells; TED OAT, thyroid eye disease orbital adipose tissue; non-OAT, nonorbital adipose tissue; Pre SFM, preadipocytes cultured 14 days in SFM; Pre DM, preadipocytes cultured 14 days in differentiation medium; Pre SCM, freshly isolated preadipocytes in SCM; Mat SCM, freshly isolated mature adipocytes in SCM. Data are the mean of duplicate experiments, calculated from the individual values in parentheses.

Figure 4.

 

TSH-induced cAMP production expressed as a percentage, where the 0 TSH value is 100%. Thyroid: filled diamonds, Graves Disease; open diamonds, multinodular goiter. TED OAT: thyroid eye disease orbital adipose tissue: triangles, preadipocytes in DM; squares, freshly isolated mature adipocytes. Variation in shading indicates samples from different individuals. Non-OAT: nonorbital adipose tissue; triangles and squares as for TED OAT. Preadip: preadipocytes either freshly isolated or cultured in SFM: open circles, TED OAT; closed circles, nonOAT.

The authors thank Sabine Costagliolam (Brussels, Belgium) and Paul Banga (London, United Kingdom) for providing monoclonal antibodies; Malcolm Wheeler (Cardiff, United Kingdom) for thyroid tissue; The Blond McIndoe Center (Brighton, United Kingdom), for samples of normal orbital adipose tissue; and Gilbert Vassart for cell lines.