Induction of Thyroid-Stimulating Hormone Receptor Autoimmunity in Hamsters
Abstract*7{)1e?, 百拇医药
Female Chinese hamsters (n = 10) were immunized with Chinese hamster ovary (CHO) cells that expressed the human TSH receptor (TSHR) to generate a model of Graves’ disease. TSHR-autoantibodies (TSHR-Ab) were determined by CHO-TSHR. Two hamsters with stimulating TSHR-Ab showed thyrocyte hypertrophy associated with a focal lymphocytic infiltration. CHO-TSHR were then stimulated with interferon to enhance major histocompatibility complex class II expression. However, after immunization no stimulating TSHR-Ab were detected, but blocking TSHR-Ab were found in three of five animals. The thyroid glands from these hamsters showed marked thinning of thyroid epithelial cells, indicative of early thyroid atrophy consistent with a TSHR blocking antibody, but no lymphocytic infiltration. Lastly, female Armenian hamsters were immunized with an adenovirus construct incorporating wild-type TSHR. High titers of TSHR-Ab were induced effectively, but the thyroid hypertrophy observed was not associated with a lymphocyte infiltration.
In summary, we demonstrated that the hamster could serve as a model of TSHR autoimmunity and that an adenoviral vector produced higher levels of TSHR-Ab than more conventional immunization with cells. The data also indicated that the intrathyroidal cellular immunity in this model was not related to TSHR-Ab formation and was an independent reflection of the T-cell immune response to TSHR antigen..ky, 百拇医药
Introduction.ky, 百拇医药
THE THYROID GLAND is a common target organ in human autoimmunity. Prominent antigens shown to be involved include the TSH receptor (TSHR), thyroid peroxidase, and thyroglobulin, unique to the thyroid gland. TSHR autoantibodies (TSHR-Ab) may enhance thyroid function whether they act as agonists and stimulate the receptor (Graves’ disease) or inhibit the TSHR (atrophic thyroiditis) by acting as antagonists. Hence, the clinical presentation reflects the ratio of the two types of TSHR-Ab (1). Immune intervention strategies for Graves’ disease require a detailed knowledge of how TSHR-specific T cells and TSHR-Ab interact with the receptor and modify the clinical course of the disease. Such studies should lead to a better insight into the pathogenesis, better prediction of the clinical course, and eventually, better treatment. However, TSHR-Ab in human Graves’ disease have proven difficult to study because their serum concentrations are low (2) and, therefore, much effort has been expended on developing an appropriate animal model.
After several initial unsuccessful attempts by several groups to develop such a model by direct immunization with recombinant TSHR ectodomain protein, Shimojo et al. (3) used immunization of mice with syngeneic fibroblasts expressing both human TSHR (hTSHR) and major histocompatibility complex (MHC) class II antigen, which was confirmed by us (4) and others (5). In this mouse model, hyperthyroid animals (~ 25% of total) developed goiters, thyroid epithelial hypertrophy, and serum-stimulating TSHR-Ab activity. However, these hyperthyroid mice lacked the intrathyroidal lymphocytic infiltration seen in Graves’ disease and had no extrathyroidal responses to the TSHR, such as lymphocytic infiltration in extrathyroidal muscles (3, 4, 5). Another recent approach has involved genetic immunization of mice with an hTSHR-expressing vector (6). Hyperthyroidism associated with stimulating TSHR-Ab was also seen in approximately 20% of wild mice, and this model showed both intrathyroidal and retro-ocular lymphocytic infiltration (6). Unfortunately, several laboratories have been unable to confirm this model (7, 8). Recently, Nagayama et al. (8) generated a Graves’ disease mouse model using an adenovirus vector incorporating the hTSHR, in which approximately 50% of immunized BALB/c mice showed hyperthyroidism associated with hypertrophy of the thyroid gland. Once again, however, there was no intrathyroidal lymphocytic infiltration.
We have undertaken a novel strategy for generating a more appropriate model for Graves’ disease. We suspected that the main reason for the lack of lymphocyte infiltration in the murine syngeneic model was the low expression of the immunogenic hTSHR antigen. Hence, we first modified this approach by using a Chinese hamster ovary (CHO) cell line that expressed high levels of functional hTSHR on the plasma membrane (CHO-TSHR; Refs. 9 and 10) to immunize Chinese hamsters. We found that we were able to induce Graves’ disease in the Chinese hamster that was associated with intrathyroidal lymphocytic infiltration but in low frequency. We, therefore, attempted to further improve the model by enhancing MHC class II expression on CHO-TSHR cells and also by TSHR antigen exposure using adenovirus immunization.?mto, 百拇医药
Materials and Methods?mto, 百拇医药
Cells and virus?mto, 百拇医药
JPO9 cells [Chinese hamster ovarian cells stably expressing human TSH receptor (CHO-hTSHR); Ref. 9 ] and JPO2 cells [control CHO cells lacking hTSHR (CHO)] were kindly provided by Dr. G. Vassart (Louvain Medical School, Brussels, Belgium). These cells were maintained in Ham’s F-12 medium (Mediatech, Herndon, VA) supplemented with 10% fetal bovine serum (Mediatech), 100 U/ml penicillin and streptomycin (Life Technologies, Inc., Grand Island, NY), and 400 µg/ml geneticin (Life Technologies, Inc.) at 37 C in 5% CO2. Adenovirus incorporating full-length hTSHR (AdTSHR), kindly provided by Dr. Y. Nagayama (Nagasaki University School of Medicine, Nagasaki, Japan), was propagated, purified, and measured as described previously (8).
Immunization of group 1np!+^fx, 百拇医药
Six- to 8-wk-old female Chinese hamsters (n = 12) were purchased from Cytogen Research and Development, Inc. (West Roxbury, ME). All experiments were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Before the immunization, CHO cells were detached from the culture dish using 1 mM EDTA plus 1 mM EGTA (Sigma, St. Louis, MO) and incubated with Ham’s F-12 medium containing 1 mg/ml mitomycin C (Boehringer Manheim, Indianapolis, IN) at 37 C for 2 h. These treated CHO cells were mixed with pertussis toxin (Sigma; 0.18 mg/animal) and alums (Pierce Chemical Co., Rockford, IL; 30 µl/animal) and injected ip (1 x 107 cells per animal) every 2 wk for 12 wk.np!+^fx, 百拇医药
Immunization of group 2np!+^fx, 百拇医药
Female Chinese hamsters (6–8 wk old; n = 11) were used in the second set of immunizations. Before the immunization, CHO and CHO-TSHR cells were stimulated with 1000 U/ml human IFN (kindly provided by PBL Biomedical Laboratories, San Diego, CA) for 48 h. The protocol was similar to the first set of immunizations. In brief, IFN-treated cells were detached and treated with mitomycin C. After mixing with pertussis toxin (0.18 mg/animal) and alum (30 µl/animal), IFN-treated cells (1 x 107 cells per animal) were used to immunize animals two to three times, as stated, each 2 wk apart and all given ip. All animals were killed at 7 wk after immunization.
Immunization of group 3ym[88, http://www.100md.com
Six- to 8-wk-old female Armenian hamsters (n = 10) were purchased from Cytogen Research and Development, Inc. Animals were immunized with 5 x 1011 virus particles of AdTSHR per animal im every 3 wk, and for three times in all. Animals were killed after 18 wk of immunization.ym[88, http://www.100md.com
Thyroid hormone assayym[88, http://www.100md.com
Blood taken from each animal was used to measure the T4 levels by a neonatal paper assay (Diagnostic Products Corp., Los Angeles, CA). Normal T4 levels of the Chinese hamster (<4.6 µg/dl) and the Armenian hamsters (<3.5 µg/dl) were estimated by measuring T4 in preimmune blood (n = 15 in the Chinese hamster and n = 10 in the Armenian hamster) in multiple assays.ym[88, http://www.100md.com
Histologyym[88, http://www.100md.com
The thyroids of killed animals were fixed in 10% formalin in PBS and stained with hematoxylin and eosin. A thyroiditis grade was assigned to each sample as described previously (11). The severity of intrathyroidal infiltration was graded as follows: 0.5, small focal areas of inflammatory cells; 1.0, focal collections of mononuclear cells; 2.0, diffuse infiltration of mononuclear cells approximately 40% or less of thyroid tissue examined; and 3.0, diffuse infiltration more than 40% of thyroid tissue.
Detection of TSHR-Ab by flow cytometry+'v-)37, http://www.100md.com
In preliminary studies, it was observed that Chinese hamsters immunized with CHO-TSHR generated strong immune responses against CHO cell membrane antigens. To minimize such non-hTSHR-Ab, 1 µl of hamster sera from test animals was diluted 1:100 with 0.1% BSA/0.01% sodium azide/PBS and absorbed on 1 x 106 CHO cells at 4 C overnight on a rotator. As shown in Fig. 1A, the nonspecific antibodies in an immunized hamster serum were significantly reduced after three rounds of absorption. Thus, every serum was absorbed three times before flow cytometry analysis. Absorbed supernatants were aliquoted and incubated with either CHO cells or CHO-TSHR for 1 h at 4 C on a rotator. After washing, cells were incubated with fluorescein isothiocyanate (FITC)-labeled antihamster IgG antibody (Boehringer Manheim; 1:100) for 60 min at 4 C. Fluorescence intensity was measured using a FACScan flow cytometer (BD PharMingen, San Diego, CA). Because background staining on CHO cells was highly variable in each sample due to nonspecific antibodies against CHO cells, more than 15 arbitrary units of mean fluorescent intensity on CHO-TSHR over CHO was judged as positive. To measure TSHR-Ab directly in sera from animals not immunized with CHO cells, 1 µl of sera was diluted and incubated with either CHO-TSHR or CHO without immunoabsorption. FITC-labeled antihamster IgG antibody (Boehringer Manheim) or FITC-labeled anti-Armenian hamster IgG2+3 (PharMingen; 1:100) and biotin-conjugated anti-Armenian hamster IgG1 (PharMingen; 1:100) were used as secondary antibodies. Avidin-conjugated phycoerythrin (Sigma; 1:100) was additionally used to obtain double color staining. Because immunized animal sera did not generate nonspecific staining on CHO cells, average plus 3 SD of mean fluorescent intensity on CHO cells by each serum was obtained and considered nonspecific (<4.5 in IgG1 and <5.0 in IgG2+3). Mean fluorescent intensity more than this range on CHO-TSHR cells was considered positive.
fig.ommitteed+$l], http://www.100md.com
Figure 1. Reduction of nonspecific binding of serum antibodies from CHO-TSHR immunized Chinese hamster by preabsorption on CHO cells as assessed by flow cytometry. A, Sera obtained from CHO-TSHR-immunized Chinese hamsters showed strong antibody responses against CHO membrane antigen (solid line). After three rounds of absorption on CHO cells, these non-TSHR-Ab were significantly reduced (dashed line). Background staining by preimmune serum is also shown (filled area). B, Staining of CHO-TSHR cells (thick line) compared with staining of CHO cells (thin line) by serum from one hamster immunized with CHO-TSHR after three rounds of absorption on CHO cells.+$l], http://www.100md.com
Detection of TSHR-Ab by competition assay+$l], http://www.100md.com
CHO-TSHR cells (5 x 104) were seeded into 96-well plates 1 d before assay. The next day, the cells were washed with modified Hanks’ solution (12) and incubated with increasing doses of bovine TSH (for generating a standard curve; Sigma) or test hamster serum (1:10 dilution) with 125I-labeled TSH (Kronus, Boise, ID) at 37 C for 1.5 h. The radioactivity remaining on CHO-TSHR cells after three washings was measured after cell lysis in 1 N NaOH. Competition (percentage) was determined as the ratio to the competition of preimmune serum. More than 15% competition was judged as positive.
Detection of TSHR-Ab by cAMP assay4lw{d}&, http://www.100md.com
To detect stimulating TSHR-Ab, CHO-TSHR cells were seeded in 96-well plates at 4 x 104 cells per well the day before the assay. The next day, 50 µl of hamster serum diluted 1:10 with Ham’s F-12 medium were added to the cells. Cells were incubated for 2 h at 37 C. Medium was discarded, and the intracellular cAMP was measured by ELISA according to the manufacturer’s protocol (Amersham Pharmacia Biotech, Newark, NJ). Stimulation was expressed as the percentage increase of cAMP production by test serum divided by cAMP production by preimmune serum. More than 200% stimulation was judged as positive. A similar protocol was used to detect blocking TSHR-Ab. CHO-TSHR cells were incubated with hamster serum (1:10 dilution) in addition to 10 µU/ml of bovine TSH. Blocking was expressed as the percentage reduction of cAMP produced by test sera divided by control sera. More than 20% blocking was considered positive.
RT-PCR for hamster MHC class II9|8;}, 百拇医药
Because hamster MHC class II sequences were unavailable, the primers used to amplify Chinese hamster MHC class II were based on published data from other species, choosing a region highly conserved among a wide variety of mammalian species (13). The oligonucleotides we designed to detect conserved region of MHC class II ß chain were as follows: forward primer, 5'-CGCTTCGACAGCGACGTGGG-3'; reverse primer, 5'-TCCTTGCCATTCCGGAACCATC-3'. To confirm that the primers could amplify the expected region, cDNA was made from rat and Chinese hamster spleen cells with or without treatment of 5 µg/ml of concanavalin A (ConA; Sigma) for 48 h. These cDNAs were used as templates to test the primers for PCR amplification. An actin fragment was also amplified as a housekeeping gene. PCR primers for actin were as follows: 5'-GACGAGGCCCAGAGCAAGAGAG-3' and 3'-GTTGTGGGGTCGGTACATGCA-5'.9|8;}, 百拇医药
Detection of MHC class II expression on CHO and CHO-TSHR
RNA was extracted from CHO and CHO-TSHR with or without treatment of 1000 U/ml of human, rat, or mouse IFN{gamma} (kindly provided by PBL Biomedical Laboratories). cDNA was PCR-amplified by MHC class II primers, using actin primers as a control.3p, http://www.100md.com
Statistical analysis3p, http://www.100md.com
Student’s t test and Fisher’s exact probability test were used to examine statistical significance.3p, http://www.100md.com
Results3p, http://www.100md.com
Immunization with CHO-TSHR (group 1)3p, http://www.100md.com
Animals were immunized every 2 wk with either CHO cells lacking hTSHR (n = 2) or CHO-TSHR cells (n = 10; group 1), and serum T4 levels were measured as an indicator of stimulating TSHR-Ab. As shown in Table 1, after 9 wk of immunization, one of the hamsters immunized with CHO-TSHR showed marked hyperthyroidism. This animal seemed thyrotoxic as evidenced by nervousness, hyperactivity, and increased demand for water and food and was at risk of arrhythmic death, which has been occasionally observed in mice (4). T4 levels from two additional animals were also elevated. The sera obtained from these studies were assessed for the presence of TSHR-Ab by flow cytometry against CHO-TSHR cells. Sera were first absorbed onto control CHO cells to reduce background CHO cell membrane binding (Fig. 1A) and then bound to CHO-TSHR cells. Sera from hamsters immunized with control CHO cells had no TSHR-Ab on flow cytometry. However, as summarized in Table 1, four sera from CHO-TSHR-immunized hamsters contained TSHR-Ab by flow cytometry analysis (Fig. 1B). Four sera were also positive in the cAMP generation assay, and four were detectable by TSH-binding competition (Table 1). cAMP-stimulating activity from the most hyperthyroid hamster was also shown up to a 1:1000 dilution (Fig. 2).
fig.ommitteed^w, 百拇医药
Table 1. Autoimmune parameters in Chinese hamsters immunized with CHO-TSHR (group 1)^w, 百拇医药
fig.ommitteed^w, 百拇医药
Figure 2. Serial dilution study of the hamster serum. The most hyperthyroid Chinese hamster serum was serially diluted and examined for its ability to stimulate cAMP production in CHO-TSHR cells.^w, 百拇医药
Macroscopically, the thyroid gland from two of the four hamsters with stimulating TSHR-Ab showed goiter formation with vascular engorgement (Graves’ hamster; Fig. 3). Histologically, these thyroids had hypertrophy of follicular epithelial cells with papillary protrusion and colloid droplet accumulation. In addition, these thyroids were found to have a focal lymphocyte infiltration (Fig. 4) with no follicular destruction.^w, 百拇医药
fig.ommitteed^w, 百拇医药
Figure 3. Hyperthyroidism (goiter) in Graves’ hamsters. The thyroid from an animal subjected to the CHO-TSHR immunization protocol (right) was significantly enlarged and rich in vascularity, compared with the thyroid from a CHO-immunized hamster (left).
fig.ommitteedi, 百拇医药
Figure 4. Thyroid hypertrophy in Graves’ hamsters. Normal follicular structure in a control Chinese hamster thyroid (A) contrasts markedly with severe hypertrophy of thyrocytes (B) and a focal lymphocyte infiltration (C and D) seen in Graves’ hamsters. The hypertrophy seen in B was associated with papillary protrusion and colloid droplets accumulation. Magnification: A, x100; B and C, x200; D, x400.i, 百拇医药
Immunization with IFN-stimulated CHO-TSHR (group 2)i, 百拇医药
In an attempt to enhance the immune response against TSHR antigen, IFN was used to augment MHC class II expression on CHO-TSHR cells before immunization (group 2). Because hamster IFN and sequences for MHC class II were not available, we tested human, rat, and mouse IFN on CHO cells using MHC class II mRNA as an indicator of the effect of IFN. PCR primers were first tested on cDNA taken from the Chinese hamster spleen cells and rat spleen cells as a positive control. As shown in Fig. 5, MHC class II fragments were successfully amplified from rat and hamster spleen cells, and the PCR products appeared to be increased by ConA treatment. These PCR-amplified fragments were sequenced. The rat product was identical to the published sequence (13), and the PCR fragment of the Chinese hamster MHC class II showed considerable homology to rat (sequence identity, 85.9%; amino acid identity, 75.4%), which confirmed its identity. Having established the efficacy of the PCR, CHO and CHO-TSHR were stimulated with human, rat, or mouse IFN to examine whether IFN could up-regulate MHC class II expression on these cells. As shown in Fig. 5, mRNA expression of the MHC class II ß chain was barely detectable in control CHO and CHO-TSHR cells. However, after stimulation with human, rat, or mouse IFN, MHC class II mRNA expression was markedly increased (Fig. 5). These PCR products were sequenced and showed approximately 79% identity in nucleic acid sequence and approximately 65% in amino acid to that of the Chinese hamster, respectively. In addition, it should be noted that IFN treatment did not affect TSHR protein expression on CHO-TSHR, as assessed by flow cytometry using several monoclonal antibodies against hTSHR - and ß-subunits (data not shown).
fig.ommitteed50y!e, 百拇医药
Figure 5. RT-PCR detection of MHC class II. Spleen cells (left) from either a rat (R) or a Chinese hamster (H) were treated with or without ConA (5 µg/ml) for 48 h, and cDNA was PCR-amplified by primers for the MHC class II ß-chain (based on rat sequence) and for actin as a housekeeping gene. In the case of CHO cells, RT-PCR was performed using cDNA generated from CHO (middle) and CHO-TSHR cells (right) with or without IFN treatment. Human (Hu), rat (R), or mouse (M) IFN (1000 U/ml) was used to treat these CHO cell lines for 48 h before RNA extraction. cDNA was not added in blank (Bl).50y!e, 百拇医药
In group 2, animals (n = 9) were immunized with human IFN-stimulated CHO-TSHR cells, and human IFN-stimulated CHO cells were immunized as controls (n = 2). The hamsters in this second set showed severe erosive dermatitis on the abdomen after 4 wk of immunization, and four animals died at this time. T4 levels and TSHR-Ab were studied when the animals were killed (Table 2). T4 levels were all normal in this group, although one of the sera showed TSHR-Ab by flow cytometry, and three sera contained TSHR-Ab detected by competition assay. We also noticed that sera from the CHO-TSHR-immunized animals significantly suppressed basal cAMP levels in CHO-TSHR cells (P < 0.01), and three sera had significant TSHR blocking activity as evidenced by inhibition of TSH-induced generation of cAMP (Table 2).
fig.ommitteed%hrsqw[, http://www.100md.com
Table 2. Autoimmune parameters in Chinese hamsters immunized with IFN-stimulated CHO-TSHR (group 2)%hrsqw[, http://www.100md.com
Thyroid specimens from group 2 showed no evidence of thyroid stimulation and no lymphocyte infiltration. However, thyroid epithelial cells from CHO-TSHR-immunized hamsters were remarkably thin when compared with thyroid epithelial cells from CHO immunized hamsters (Fig. 6). Reduced colloid content was also apparent. These changes were most evident in the hamster with the strongest blocking TSHR-Ab. Thus, we concluded that such thyroids were atrophic due to blocking TSHR-Ab. Thyroid atrophy was seen in three of five animals in group 2 and was not seen in group 1. This difference was statistically significant (P < 0.05).%hrsqw[, http://www.100md.com
fig.ommitteed%hrsqw[, http://www.100md.com
Figure 6. Atrophy of thyroid epithelial cells in animals immunized with IFN-stimulated CHO-TSHR cells. The thyroid glands from the Chinese hamsters immunized with IFN-stimulated CHO-TSHR cells showed thinning of the thyroid epithelial cells with reduced colloid content (B and D), compared with thyroid glands from animals immunized with IFN-stimulated CHO cells (A and C), which appeared normal. Magnification: A and B, x100; C and D, x200.
Eye specimens from animals with strong TSHR-Ab in groups 1 and 2 were examined, but no abnormal retro-ocular findings, such as lymphocytic infiltration, were seen except an occasional lacrimal gland infiltrate that was also seen in controls (data not shown).l, http://www.100md.com
Immunization with AdTSHR (group 3)l, http://www.100md.com
On the basis of the results from groups 1 and 2, we concluded that the hamster could be used as an animal model of TSHR autoimmunity. We then attempted a better immunization protocol aimed at obtaining higher levels of TSHR-Ab. Recently, Nagayama et al. (8) generated a new mouse model of Graves’ disease using an AdTSHR. In their model, up to 50% of BALB/c mice immunized with AdTSHR developed Graves’ hyperthyroidism, which was much higher than approximately 25% disease induction seen in other models (3). However, this adenovirusinduced mouse model still lacked intrathyroidal lymphocytic infiltration. In group 3, we combined the advantage of the hamster with this same adenovirus construct. In addition to this change, and with the aim of eventually obtaining monoclonal antibodies, we employed the Armenian hamster that is known to provide more stable hybridomas and to fuse with a higher efficiency (14).
Animals (n = 10) were immunized with 5 x 1011 particles per animal of AdTSHR. No abnormal behavior consistent with Graves’ hyperthyroidism was seen during the immunization protocol. T4 levels measured after 12 wk of immunization are summarized in Table 3. As seen, T4 levels from four animals were elevated after 18 wk of follow-up. Sera were also examined for TSHR-Ab using flow cytometry analysis with CHO-TSHR cells. In contrast to CHO-TSHR immunization in groups 1 and 2, no immunoabsorption was needed to measure TSHR-Ab by flow cytometry (data not shown). Sera were strongly positive for TSHR-Ab when using a secondary antibody specific to Armenian hamster IgG1 and IgG2+3 (Fig. 7). As seen in Table 3, the nine sera positive for TSHR-Ab were all IgG2+3. The mean fluorescent intensity of positive samples on CHO-TSHR was 46.2 ± 10.5 (n = 9) in IgG2+3 vs. 8.3 ± 3.0 (n = 3) in IgG1 (P < 0.001). IgG1 TSHR-Ab were only detected in three animals in low titer. Thus, TSHR-Ab generated in this model were predominantly IgG2+3. TSHR-Ab detected by competition assay were also seen in all sera. The percentage competition was more than 90% in nine animals, and one showed moderate competition (Table 3). Stimulating TSHR-Ab were also detected in two animals by using the cAMP response in CHO-TSHR. Four of the thyroid specimens from group 3 also showed thyroid hypertrophy associated with papillary protrusion (Fig. 8). However, there was no intrathyroidal lymphocyte infiltration as seen earlier in group 1. There were also no retro-ocular findings seen.
fig.ommitteed1y$':, 百拇医药
Table 3. Autoimmune parameters in the Armenian hamsters immunized with AdTSHR (group 3)1y$':, 百拇医药
fig.ommitteed1y$':, 百拇医药
Figure 7. IgG isotype-specific TSHR-Ab determined by flow cytometry in sera from the Armenian hamster immunized with AdTSHR (group 3). IgG2+3 TSHR-Ab (A), but not IgG1 (C), were detected in serum from animal no. 3 (Table 3); however, in serum from animal no. 8, both IgG2+3 (B) and IgG1 (D) TSHR-Ab were detected. The thin and thick lines indicate staining on CHO and CHO-TSHR cells, respectively. IgG1 was measured in FL2 and IgG2+3 in FL1.1y$':, 百拇医药
fig.ommitteed1y$':, 百拇医药
Figure 8. Thyroid hypertrophy in the Armenian hamsters. In contrast to the normal thyroid follicular structure (A and C), the thyroid epithelial cells from hyperthyroid hamsters showed marked hypertrophy associated with papillary protrusion (B and D). No lymphocytic infiltration was seen. Magnification: A and B, x100; C and D, x200.1y$':, 百拇医药
Comparison of TSHR-Ab induced in the three immunization protocolh|, 百拇医药
To compare the levels of TSHR-Ab determined by flow cytometry, all data were obtained using the same secondary antibody against hamster IgG (Table 4). Compared with group 1, we expected stronger TSHR antigen exposure associated with enhanced MHC class II expression on CHO-TSHR cells in group 2. However, there was no statistical difference in the levels of TSHR-Ab or efficiency of TSHR-Ab induction between groups 1 and 2. It should be noted that this comparison was likely to be influenced by the fact that group 2 immunization was not as extensive as group 1 because of the adverse effects of the immunization regime. Compared with CHO-TSHR immunization (groups 1, 2, and 1+2), TSHR-Ab determined by competition assay after AdTSHR (group 3) was significantly stronger (P < 0.001), and TSHR-Ab induction was also significantly more effective (P < 0.001). The level of TSHR-Ab determined by flow cytometry was comparable between group 1+2 and group 3, but such TSHR-Ab were more effectively induced in group 3 compared with group 1 (P < 0.005), group 2 (P < 0.02), and group 1+2 (P < 0.001). Thus, in the hamster model, adenovirus immunization induced TSHR-Ab, as determined by competition assay and flow cytometry, more successfully than CHO-TSHR immunization.
fig.ommitteed7*cw1w, 百拇医药
Table 4. TSHR-Ab induced in three immunization protocols7*cw1w, 百拇医药
Discussion7*cw1w, 百拇医药
We first attempted to generate a model of Graves’ disease in the hamster by immunizing with CHO cells stably overexpressing hTSHR antigen. Even with Chinese hamsters, we found a marked immune response to control CHO cell (Fig. 1A). However, after CHO-TSHR immunization, two hamsters showed marked thyroid hypertrophy and a focal lymphocyte infiltrate (Figs. 3 and 4) and also had stimulating TSHR-Ab in their sera. Stimulation of CHO-TSHR with human IFN before the immunization (group 2) appeared to enhance MHC class II mRNA in CHO-TSHR cells when assessed by semiquantitative RT-PCR (Fig. 5), but this did not lead to an enhancement of the development of Graves’ disease in this model. Only blocking TSHR-Ab were detected in hamsters treated in this way, and the thyroid glands from these animals showed atrophy and thinning of the thyroid epithelial cells (Fig. 6). To further improve the hamster model, adenovirus immunization was employed to obtain more effective exposure of TSHR antigen (group 3). We found that TSHR-Ab were detected by both flow cytometry and competition assay in all animals immunized (Table 3). However, the thyroid hypertrophy seen in these animals was not associated with an intrathyroidal lymphocytic infiltration (Fig. 8).
Initial attempts to generate a Graves’ disease animal model used immunization with recombinant hTSHR-extracellular domain protein from prokaryotic and eukaryotic cell expression systems (15). Although strong immune responses against the hTSHR were obtained, these TSHR-Ab antibodies only inhibited binding of TSH to the TSHR, even when associated with a thyroidal lymphocytic infiltration (16). One study reported generating stimulating antibodies associated with hyperthyroidism using hTSHR ectodomain protein, but the histological findings were of a destructive thyroiditis that seemed at odds with the observations (17). We and others previously concluded that an improved protocol would incorporate immunization with full-length hTSHR protein that had undergone appropriate post-translational processing and folding (15). This assumption was the basis of both the Shimojo mouse model (3) and DNA immunization (6). In the former model, TSHR protein was expressed on mammalian fibroblasts and in the latter, the TSHR protein was expressed on the host muscle. However, there has been difficulty in reproducing the DNA immunization model (7, 8), although TSHR-reactive T cells have been detected (7). One of the reasons for these difficulties may have been low TSHR antigen production and low exposure to the immune system compared with direct protein immunization (18). Indeed, low expression of the TSHR on muscle after DNA immunization has recently been shown (8).
Based on these observations, we undertook new strategies for generating an animal model for Graves’ disease. We first used a CHO cell line stably expressing high levels of functional hTSHR on the plasma membrane (CHO-TSHR). This cell line, named JPO9 (9), is highly sensitive to TSH stimulation (9, 10, 19), and the present study showed that the cells expressed MHC class II mRNA constitutively (Fig. 5). Although CHO cells were derived from the Chinese hamster, the strong background staining on flow cytometry using sera from immunized hamsters indicated that these outbred animals recognized many CHO cell membrane proteins as foreign antigens (Fig. 1A). This indicated that cell immunization induces strong nonspecific immune reaction in both a syngeneic model (20) and an allogeneic model (this study). Thus, the major advantage of the approach was the high density of functional TSHR protein available to the immune system. We succeeded in inducing intrathyroidal lymphocytic infiltration along with stimulating TSHR-Ab production in this model, but the rate of disease development was still only approximately 20%. As in previous studies (21), our findings demonstrated that the immune system recognized a variety of epitopes on the TSHR as evidenced by the results of a variety of assays for TSHR-Ab. In addition, the failure to induce stimulating TSHR-Ab in a large percentage of animals may simply have reflected the variable genetic susceptibility of hamsters to Graves’ disease (6, 8).
In an attempt to address some of these limitations, a second immunization protocol was performed using IFN to up-regulate MHC class II expression on antigen-presenting cells (Fig. 5). Although TSHR-Ab and thyroid histology in group 2 were examined at an earlier time point than in group 1, thyroid atrophy along with blocking TSHR-Ab were observed in three of five animals. This finding suggested that IFN stimulation skewed the host immune reaction toward the generation of blocking rather than stimulating TSHR-Ab. IFN is well known to enhance expression of MHC classes I and II. It also regulates expression of co-stimulatory molecules, such as B7-1 and B7-2, as well as affecting antigen processing and presentation (22, 23). Because IFN did not affect TSHR protein expression levels on CHO-TSHR cells (data not shown), we did not identify any other mechanism(s) in which the host immune reaction was skewed by IFN-stimulated CHO-TSHR. However, altered expression of co-stimulation molecules on CHO-TSHR cells and/or the different TSHR epitopes presented on IFN-stimulated CHO-TSHR cells from those on untreated cells might be involved, in addition to the quantity of MHC molecules. Human leukocyte antigen class II expression on the Graves’ thyroid gland has been widely investigated (24), suggesting its involvement in thyroid autoantigen presentation. Successful immunization of fibroblasts expressing both TSHR antigen and MHC class II, but not fibroblasts expressing only TSHR antigen, in the Shimojo model (3) further indicated an important role for MHC class II in the initiation of Graves’ disease. Graves’ hamsters in this study were generated by using CHO-TSHR cells that appeared to constitutively express MHC class II. Indeed, MHC class II overexpression by IFN stimulation prompted the generation of blocking TSHR-Ab rather than stimulating TSHR-Ab. Taken together, these results confirmed the important role(s) of MHC class II in Graves’ disease induction, and also indicated that IFN may modulate the immune response to TSHR antigen.
We concluded that the hamster could serve as an animal model of TSHR autoimmunity using CHO-TSHR cells and proceeded to further improve the model by immunizing with an adenovirus construct incorporating TSHR DNA. Here, we expected a more effective exposure of the immune system to TSHR antigen that could have induced Graves’ disease in higher frequency and still be associated with an intrathyroidal lymphocytic infiltration. Indeed, in these animals the TSHR-Ab levels were significantly greater, confirming that the AdTSHR immunization triggered a stronger humeral immune response to TSHR antigen. This difference may also have been due to a greater genetic susceptibility in the Armenian hamster in addition to the immunization protocol. However, both the Chinese and the Armenian hamsters remain outbred, suggesting this was an unlikely explanation. The resulting thyroid hypertrophy was not associated with an intrathyroidal lymphocytic infiltration. Their TSHR-Ab Ig subtype, determined by flow cytometery, suggested that these animals did have an immune response that could be classified as preferentially T helper, but this clearly did not lead to lymphocytic infiltration. Furthermore, because these hamsters had no intrathyroidal lymphocytic infiltration, it was also clear that neither stronger exposure of TSHR antigen nor high levels of TSHR-Ab were an indication of such infiltration.
In summary, we demonstrated that the hamster could serve as an animal model of TSHR autoimmunity. Our study suggested a modulating role for cytokines and MHC class II expression in the induction of Graves’ disease. We also demonstrated that an adenovirus vector was a highly effective immunogen for the induction of TSHR autoimmunity, achieving significantly greater levels of TSHR-Ab. In addition, our data suggested that an intrathyroidal lymphocytic infiltration was not directly related to levels of TSHR-Ab or exposure to more TSHR antigen. Nevertheless, the hamster offers promising insights as a model for thyroid autoimmune disease.})/$8, 百拇医药
Acknowledgments})/$8, 百拇医药
We thank Dr. Y. Nagayama, Nagasaki University School of Medicine, Nagasaki, Japan, for providing AdTSHR.})/$8, 百拇医药
Received June 4, 2002.})/$8, 百拇医药
Accepted for publication November 4, 2002.})/$8, 百拇医药
References})/$8, 百拇医药
Vassart G, Dumont JE 1992 The thyrotropin receptor and the regulation of thyrocyte function and growth. Endocr Rev 13:596–611
Jaume JC, Kakinuma A, Chazenbalk GD, Rapoport B, McLachlan SM 1997 Thyrotropin receptor autoantibodies in serum are present at much lower levels than thyroid peroxidase autoantibodies: analysis by flow cytometry. J Clin Endocrinol Metab 82:500–507?%, http://www.100md.com
Shimojo N, Kohno Y, Yamaguchi K, Kikuoka S, Hoshioka A, Niimi H, Hirai A, Tamura Y, Saito Y, Kohn LD, Tahara K 1996 Induction of Graves-like disease in mice by immunization with fibroblasts transfected with the thyrotropin receptor and a class II molecule. Proc Natl Acad Sci USA 93:11074–11079?%, http://www.100md.com
Kita M, Ahmad L, Marians RC, Vlase H, Unger P, Graves PN, Davies TF 1999 Regulation and transfer of a murine model of thyrotropin receptor antibody mediated Graves’ disease. Endocrinology 140:1392–1398?%, http://www.100md.com
Jaume JC, Rapoport B, McLachlan SM 1999 Lack of female bias in a mouse model of autoimmune hyperthyroidism (Graves’ disease). Autoimmunity 29:269–272?%, http://www.100md.com
Costagliola S, Many MC, Denef JF, Pohlenz J, Refetoff S, Vassart G 2000 Genetic immunization of outbred mice with thyrotropin receptor cDNA provides a model of Graves’ disease. J Clin Invest 105:803–811
Pichurin P, Yan XM, Farilla L, Guo J, Chazenbalk GD, Rapoport B, McLachlan SM 2001 Naked TSH receptor DNA vaccination: a TH1 T cell response in which interferon- production, rather than antibody, dominates the immune response in mice. Endocrinology 142:3530–3536[a\\z, 百拇医药
Nagayama Y, Kita-Furuyama M, Ando T, Nakao K, Mizuguchi H, Hayakawa T, Eguchi K, Niwa M 2002 A novel murine model of Graves’ hyperthyroidism with intramuscular injection of adenovirus expressing the thyrotropin receptor. J Immunol 168:2789–2794[a\\z, 百拇医药
Perret J, Ludgate M, Libert F, Gerard C, Dumont JE, Vassart G, Parmentier M1990 Stable expression of the human TSH receptor in CHO cells and characterization of differentially expressing clones. Biochem Biophys Res Commun 171:1044–1050[a\\z, 百拇医药
Michelangeli VP, Munro DS, Poon CW, Frauman AG, Colman PG 1994 Measurement of thyroid stimulating immunoglobulins in a new cell line transfected with a functional human TSH receptor (JPO9 cells), compared with an assay using FRTL-5 cells. Clin Endocrinol (Oxf) 40:645–652[a\\z, 百拇医药
Imaizumi M, Pritsker A, Unger P, Davies TF 2002 Intrathyroidal fetal microchimerism in pregnancy and postpartum. Endocrinology 143:247–253
Davies TF, Yang C, Platzer M 1987 Cloning the Fisher rat thyroid cell line (FRTL-5): variability in clonal growth and 3, 5'-cyclic adenosine monophosphate response to thyrotropin. Endocrinology 121:78–83v3, 百拇医药
Robertson KA, McMaster WR 1985 Complete structure of a rat RT1 E ß chain: extensive conservation of MHC class II ß chains. J Immunol 135:4095–4099v3, 百拇医药
Lengweiler S, Smyth SS, Jirouskova M, Scudder LE, Park H, Moran T, Coller BS 1999 Preparation of monoclonal antibodies to murine platelet glycoprotein IIb/IIIa (IIbß3) and other proteins from hamster-mouse interspecies hybridomas. Biochem Biophys Res Commun 262:167–173v3, 百拇医药
Davies TF, Vlase H, Kita M 1999 The search for an animal model for Graves’ disease. In: Shoenfeld Y, ed. The decade of autoimmunity. New York: Elsevier Science; 43–50v3, 百拇医药
Costagliola S, Many MC, Stalmans-Falys M, Tonacchera M, Vassart G, Ludgate M 1994 Recombinant thyrotropin receptor and the induction of autoimmune thyroid disease in BALB/c mice: a new animal model. Endocrinology 135:2150–2159
Kaithamana S, Fan J, Osuga Y, Liang SG, Prabhakar BS 1999 Induction of experimental autoimmune Graves’ disease in BALB/c mice. J Immunol 163:5157–5164:4:|, 百拇医药
Ito K, Takeuchi Y, Kato S 2000 Strain-dependent antibody response induced by DNA immunization. Immunol Lett 74:245–250:4:|, 百拇医药
Patibandla SA, Dallas JS, Seetharamaiah GS, Tahara K, Kohn LD, Prabhakar BS 1997 Flow cytometric analyses of antibody binding to Chinese hamster ovary cells expressing human thyrotropin receptor. J Clin Endocrinol Metab 82:1885–1893:4:|, 百拇医药
Yan XM, Guo J, Pichurin P, Tanaka K, Jaume JC, Rapoport B, McLachlan SM 2000 Cytokines, IgG subclasses and costimulation in a mouse model of thyroid autoimmunity induced by injection of fibroblasts co-expressing MHC class II and thyroid autoantigens. Clin Exp Immunol 122:170–179:4:|, 百拇医药
Vlase H, Nakashima M, Graves PN, Tomer Y, Morris JC, Davies TF 1995 Defining the major antibody epitopes on the human thyrotropin receptor in immunized mice: evidence for intramolecular epitope spreading. Endocrinology 136:4415–4423:4:|, 百拇医药
Boehm U, Klamp T, Groot M, Howard JC 1997 Cellular responses to interferon-. Annu Rev Immunol 15:749–795:4:|, 百拇医药
York IA, Rock KL 1996 Antigen processing and presentation by the class I major histocompatibility complex. Annu Rev Immunol 14:369–396:4:|, 百拇医药
Davies TF, Piccinini LA, Roman SH, Hirose W, Neufeld DS 1988 Role of MHC class II antigen expression in thyroid autoimmunity. Ann NY Acad Sci 546:151–163(Takao Ando Misa Imaizumi Peter Graves Pamela Unger and Terry F. Davies)
Female Chinese hamsters (n = 10) were immunized with Chinese hamster ovary (CHO) cells that expressed the human TSH receptor (TSHR) to generate a model of Graves’ disease. TSHR-autoantibodies (TSHR-Ab) were determined by CHO-TSHR. Two hamsters with stimulating TSHR-Ab showed thyrocyte hypertrophy associated with a focal lymphocytic infiltration. CHO-TSHR were then stimulated with interferon to enhance major histocompatibility complex class II expression. However, after immunization no stimulating TSHR-Ab were detected, but blocking TSHR-Ab were found in three of five animals. The thyroid glands from these hamsters showed marked thinning of thyroid epithelial cells, indicative of early thyroid atrophy consistent with a TSHR blocking antibody, but no lymphocytic infiltration. Lastly, female Armenian hamsters were immunized with an adenovirus construct incorporating wild-type TSHR. High titers of TSHR-Ab were induced effectively, but the thyroid hypertrophy observed was not associated with a lymphocyte infiltration.
In summary, we demonstrated that the hamster could serve as a model of TSHR autoimmunity and that an adenoviral vector produced higher levels of TSHR-Ab than more conventional immunization with cells. The data also indicated that the intrathyroidal cellular immunity in this model was not related to TSHR-Ab formation and was an independent reflection of the T-cell immune response to TSHR antigen..ky, 百拇医药
Introduction.ky, 百拇医药
THE THYROID GLAND is a common target organ in human autoimmunity. Prominent antigens shown to be involved include the TSH receptor (TSHR), thyroid peroxidase, and thyroglobulin, unique to the thyroid gland. TSHR autoantibodies (TSHR-Ab) may enhance thyroid function whether they act as agonists and stimulate the receptor (Graves’ disease) or inhibit the TSHR (atrophic thyroiditis) by acting as antagonists. Hence, the clinical presentation reflects the ratio of the two types of TSHR-Ab (1). Immune intervention strategies for Graves’ disease require a detailed knowledge of how TSHR-specific T cells and TSHR-Ab interact with the receptor and modify the clinical course of the disease. Such studies should lead to a better insight into the pathogenesis, better prediction of the clinical course, and eventually, better treatment. However, TSHR-Ab in human Graves’ disease have proven difficult to study because their serum concentrations are low (2) and, therefore, much effort has been expended on developing an appropriate animal model.
After several initial unsuccessful attempts by several groups to develop such a model by direct immunization with recombinant TSHR ectodomain protein, Shimojo et al. (3) used immunization of mice with syngeneic fibroblasts expressing both human TSHR (hTSHR) and major histocompatibility complex (MHC) class II antigen, which was confirmed by us (4) and others (5). In this mouse model, hyperthyroid animals (~ 25% of total) developed goiters, thyroid epithelial hypertrophy, and serum-stimulating TSHR-Ab activity. However, these hyperthyroid mice lacked the intrathyroidal lymphocytic infiltration seen in Graves’ disease and had no extrathyroidal responses to the TSHR, such as lymphocytic infiltration in extrathyroidal muscles (3, 4, 5). Another recent approach has involved genetic immunization of mice with an hTSHR-expressing vector (6). Hyperthyroidism associated with stimulating TSHR-Ab was also seen in approximately 20% of wild mice, and this model showed both intrathyroidal and retro-ocular lymphocytic infiltration (6). Unfortunately, several laboratories have been unable to confirm this model (7, 8). Recently, Nagayama et al. (8) generated a Graves’ disease mouse model using an adenovirus vector incorporating the hTSHR, in which approximately 50% of immunized BALB/c mice showed hyperthyroidism associated with hypertrophy of the thyroid gland. Once again, however, there was no intrathyroidal lymphocytic infiltration.
We have undertaken a novel strategy for generating a more appropriate model for Graves’ disease. We suspected that the main reason for the lack of lymphocyte infiltration in the murine syngeneic model was the low expression of the immunogenic hTSHR antigen. Hence, we first modified this approach by using a Chinese hamster ovary (CHO) cell line that expressed high levels of functional hTSHR on the plasma membrane (CHO-TSHR; Refs. 9 and 10) to immunize Chinese hamsters. We found that we were able to induce Graves’ disease in the Chinese hamster that was associated with intrathyroidal lymphocytic infiltration but in low frequency. We, therefore, attempted to further improve the model by enhancing MHC class II expression on CHO-TSHR cells and also by TSHR antigen exposure using adenovirus immunization.?mto, 百拇医药
Materials and Methods?mto, 百拇医药
Cells and virus?mto, 百拇医药
JPO9 cells [Chinese hamster ovarian cells stably expressing human TSH receptor (CHO-hTSHR); Ref. 9 ] and JPO2 cells [control CHO cells lacking hTSHR (CHO)] were kindly provided by Dr. G. Vassart (Louvain Medical School, Brussels, Belgium). These cells were maintained in Ham’s F-12 medium (Mediatech, Herndon, VA) supplemented with 10% fetal bovine serum (Mediatech), 100 U/ml penicillin and streptomycin (Life Technologies, Inc., Grand Island, NY), and 400 µg/ml geneticin (Life Technologies, Inc.) at 37 C in 5% CO2. Adenovirus incorporating full-length hTSHR (AdTSHR), kindly provided by Dr. Y. Nagayama (Nagasaki University School of Medicine, Nagasaki, Japan), was propagated, purified, and measured as described previously (8).
Immunization of group 1np!+^fx, 百拇医药
Six- to 8-wk-old female Chinese hamsters (n = 12) were purchased from Cytogen Research and Development, Inc. (West Roxbury, ME). All experiments were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Before the immunization, CHO cells were detached from the culture dish using 1 mM EDTA plus 1 mM EGTA (Sigma, St. Louis, MO) and incubated with Ham’s F-12 medium containing 1 mg/ml mitomycin C (Boehringer Manheim, Indianapolis, IN) at 37 C for 2 h. These treated CHO cells were mixed with pertussis toxin (Sigma; 0.18 mg/animal) and alums (Pierce Chemical Co., Rockford, IL; 30 µl/animal) and injected ip (1 x 107 cells per animal) every 2 wk for 12 wk.np!+^fx, 百拇医药
Immunization of group 2np!+^fx, 百拇医药
Female Chinese hamsters (6–8 wk old; n = 11) were used in the second set of immunizations. Before the immunization, CHO and CHO-TSHR cells were stimulated with 1000 U/ml human IFN (kindly provided by PBL Biomedical Laboratories, San Diego, CA) for 48 h. The protocol was similar to the first set of immunizations. In brief, IFN-treated cells were detached and treated with mitomycin C. After mixing with pertussis toxin (0.18 mg/animal) and alum (30 µl/animal), IFN-treated cells (1 x 107 cells per animal) were used to immunize animals two to three times, as stated, each 2 wk apart and all given ip. All animals were killed at 7 wk after immunization.
Immunization of group 3ym[88, http://www.100md.com
Six- to 8-wk-old female Armenian hamsters (n = 10) were purchased from Cytogen Research and Development, Inc. Animals were immunized with 5 x 1011 virus particles of AdTSHR per animal im every 3 wk, and for three times in all. Animals were killed after 18 wk of immunization.ym[88, http://www.100md.com
Thyroid hormone assayym[88, http://www.100md.com
Blood taken from each animal was used to measure the T4 levels by a neonatal paper assay (Diagnostic Products Corp., Los Angeles, CA). Normal T4 levels of the Chinese hamster (<4.6 µg/dl) and the Armenian hamsters (<3.5 µg/dl) were estimated by measuring T4 in preimmune blood (n = 15 in the Chinese hamster and n = 10 in the Armenian hamster) in multiple assays.ym[88, http://www.100md.com
Histologyym[88, http://www.100md.com
The thyroids of killed animals were fixed in 10% formalin in PBS and stained with hematoxylin and eosin. A thyroiditis grade was assigned to each sample as described previously (11). The severity of intrathyroidal infiltration was graded as follows: 0.5, small focal areas of inflammatory cells; 1.0, focal collections of mononuclear cells; 2.0, diffuse infiltration of mononuclear cells approximately 40% or less of thyroid tissue examined; and 3.0, diffuse infiltration more than 40% of thyroid tissue.
Detection of TSHR-Ab by flow cytometry+'v-)37, http://www.100md.com
In preliminary studies, it was observed that Chinese hamsters immunized with CHO-TSHR generated strong immune responses against CHO cell membrane antigens. To minimize such non-hTSHR-Ab, 1 µl of hamster sera from test animals was diluted 1:100 with 0.1% BSA/0.01% sodium azide/PBS and absorbed on 1 x 106 CHO cells at 4 C overnight on a rotator. As shown in Fig. 1A, the nonspecific antibodies in an immunized hamster serum were significantly reduced after three rounds of absorption. Thus, every serum was absorbed three times before flow cytometry analysis. Absorbed supernatants were aliquoted and incubated with either CHO cells or CHO-TSHR for 1 h at 4 C on a rotator. After washing, cells were incubated with fluorescein isothiocyanate (FITC)-labeled antihamster IgG antibody (Boehringer Manheim; 1:100) for 60 min at 4 C. Fluorescence intensity was measured using a FACScan flow cytometer (BD PharMingen, San Diego, CA). Because background staining on CHO cells was highly variable in each sample due to nonspecific antibodies against CHO cells, more than 15 arbitrary units of mean fluorescent intensity on CHO-TSHR over CHO was judged as positive. To measure TSHR-Ab directly in sera from animals not immunized with CHO cells, 1 µl of sera was diluted and incubated with either CHO-TSHR or CHO without immunoabsorption. FITC-labeled antihamster IgG antibody (Boehringer Manheim) or FITC-labeled anti-Armenian hamster IgG2+3 (PharMingen; 1:100) and biotin-conjugated anti-Armenian hamster IgG1 (PharMingen; 1:100) were used as secondary antibodies. Avidin-conjugated phycoerythrin (Sigma; 1:100) was additionally used to obtain double color staining. Because immunized animal sera did not generate nonspecific staining on CHO cells, average plus 3 SD of mean fluorescent intensity on CHO cells by each serum was obtained and considered nonspecific (<4.5 in IgG1 and <5.0 in IgG2+3). Mean fluorescent intensity more than this range on CHO-TSHR cells was considered positive.
fig.ommitteed+$l], http://www.100md.com
Figure 1. Reduction of nonspecific binding of serum antibodies from CHO-TSHR immunized Chinese hamster by preabsorption on CHO cells as assessed by flow cytometry. A, Sera obtained from CHO-TSHR-immunized Chinese hamsters showed strong antibody responses against CHO membrane antigen (solid line). After three rounds of absorption on CHO cells, these non-TSHR-Ab were significantly reduced (dashed line). Background staining by preimmune serum is also shown (filled area). B, Staining of CHO-TSHR cells (thick line) compared with staining of CHO cells (thin line) by serum from one hamster immunized with CHO-TSHR after three rounds of absorption on CHO cells.+$l], http://www.100md.com
Detection of TSHR-Ab by competition assay+$l], http://www.100md.com
CHO-TSHR cells (5 x 104) were seeded into 96-well plates 1 d before assay. The next day, the cells were washed with modified Hanks’ solution (12) and incubated with increasing doses of bovine TSH (for generating a standard curve; Sigma) or test hamster serum (1:10 dilution) with 125I-labeled TSH (Kronus, Boise, ID) at 37 C for 1.5 h. The radioactivity remaining on CHO-TSHR cells after three washings was measured after cell lysis in 1 N NaOH. Competition (percentage) was determined as the ratio to the competition of preimmune serum. More than 15% competition was judged as positive.
Detection of TSHR-Ab by cAMP assay4lw{d}&, http://www.100md.com
To detect stimulating TSHR-Ab, CHO-TSHR cells were seeded in 96-well plates at 4 x 104 cells per well the day before the assay. The next day, 50 µl of hamster serum diluted 1:10 with Ham’s F-12 medium were added to the cells. Cells were incubated for 2 h at 37 C. Medium was discarded, and the intracellular cAMP was measured by ELISA according to the manufacturer’s protocol (Amersham Pharmacia Biotech, Newark, NJ). Stimulation was expressed as the percentage increase of cAMP production by test serum divided by cAMP production by preimmune serum. More than 200% stimulation was judged as positive. A similar protocol was used to detect blocking TSHR-Ab. CHO-TSHR cells were incubated with hamster serum (1:10 dilution) in addition to 10 µU/ml of bovine TSH. Blocking was expressed as the percentage reduction of cAMP produced by test sera divided by control sera. More than 20% blocking was considered positive.
RT-PCR for hamster MHC class II9|8;}, 百拇医药
Because hamster MHC class II sequences were unavailable, the primers used to amplify Chinese hamster MHC class II were based on published data from other species, choosing a region highly conserved among a wide variety of mammalian species (13). The oligonucleotides we designed to detect conserved region of MHC class II ß chain were as follows: forward primer, 5'-CGCTTCGACAGCGACGTGGG-3'; reverse primer, 5'-TCCTTGCCATTCCGGAACCATC-3'. To confirm that the primers could amplify the expected region, cDNA was made from rat and Chinese hamster spleen cells with or without treatment of 5 µg/ml of concanavalin A (ConA; Sigma) for 48 h. These cDNAs were used as templates to test the primers for PCR amplification. An actin fragment was also amplified as a housekeeping gene. PCR primers for actin were as follows: 5'-GACGAGGCCCAGAGCAAGAGAG-3' and 3'-GTTGTGGGGTCGGTACATGCA-5'.9|8;}, 百拇医药
Detection of MHC class II expression on CHO and CHO-TSHR
RNA was extracted from CHO and CHO-TSHR with or without treatment of 1000 U/ml of human, rat, or mouse IFN{gamma} (kindly provided by PBL Biomedical Laboratories). cDNA was PCR-amplified by MHC class II primers, using actin primers as a control.3p, http://www.100md.com
Statistical analysis3p, http://www.100md.com
Student’s t test and Fisher’s exact probability test were used to examine statistical significance.3p, http://www.100md.com
Results3p, http://www.100md.com
Immunization with CHO-TSHR (group 1)3p, http://www.100md.com
Animals were immunized every 2 wk with either CHO cells lacking hTSHR (n = 2) or CHO-TSHR cells (n = 10; group 1), and serum T4 levels were measured as an indicator of stimulating TSHR-Ab. As shown in Table 1, after 9 wk of immunization, one of the hamsters immunized with CHO-TSHR showed marked hyperthyroidism. This animal seemed thyrotoxic as evidenced by nervousness, hyperactivity, and increased demand for water and food and was at risk of arrhythmic death, which has been occasionally observed in mice (4). T4 levels from two additional animals were also elevated. The sera obtained from these studies were assessed for the presence of TSHR-Ab by flow cytometry against CHO-TSHR cells. Sera were first absorbed onto control CHO cells to reduce background CHO cell membrane binding (Fig. 1A) and then bound to CHO-TSHR cells. Sera from hamsters immunized with control CHO cells had no TSHR-Ab on flow cytometry. However, as summarized in Table 1, four sera from CHO-TSHR-immunized hamsters contained TSHR-Ab by flow cytometry analysis (Fig. 1B). Four sera were also positive in the cAMP generation assay, and four were detectable by TSH-binding competition (Table 1). cAMP-stimulating activity from the most hyperthyroid hamster was also shown up to a 1:1000 dilution (Fig. 2).
fig.ommitteed^w, 百拇医药
Table 1. Autoimmune parameters in Chinese hamsters immunized with CHO-TSHR (group 1)^w, 百拇医药
fig.ommitteed^w, 百拇医药
Figure 2. Serial dilution study of the hamster serum. The most hyperthyroid Chinese hamster serum was serially diluted and examined for its ability to stimulate cAMP production in CHO-TSHR cells.^w, 百拇医药
Macroscopically, the thyroid gland from two of the four hamsters with stimulating TSHR-Ab showed goiter formation with vascular engorgement (Graves’ hamster; Fig. 3). Histologically, these thyroids had hypertrophy of follicular epithelial cells with papillary protrusion and colloid droplet accumulation. In addition, these thyroids were found to have a focal lymphocyte infiltration (Fig. 4) with no follicular destruction.^w, 百拇医药
fig.ommitteed^w, 百拇医药
Figure 3. Hyperthyroidism (goiter) in Graves’ hamsters. The thyroid from an animal subjected to the CHO-TSHR immunization protocol (right) was significantly enlarged and rich in vascularity, compared with the thyroid from a CHO-immunized hamster (left).
fig.ommitteedi, 百拇医药
Figure 4. Thyroid hypertrophy in Graves’ hamsters. Normal follicular structure in a control Chinese hamster thyroid (A) contrasts markedly with severe hypertrophy of thyrocytes (B) and a focal lymphocyte infiltration (C and D) seen in Graves’ hamsters. The hypertrophy seen in B was associated with papillary protrusion and colloid droplets accumulation. Magnification: A, x100; B and C, x200; D, x400.i, 百拇医药
Immunization with IFN-stimulated CHO-TSHR (group 2)i, 百拇医药
In an attempt to enhance the immune response against TSHR antigen, IFN was used to augment MHC class II expression on CHO-TSHR cells before immunization (group 2). Because hamster IFN and sequences for MHC class II were not available, we tested human, rat, and mouse IFN on CHO cells using MHC class II mRNA as an indicator of the effect of IFN. PCR primers were first tested on cDNA taken from the Chinese hamster spleen cells and rat spleen cells as a positive control. As shown in Fig. 5, MHC class II fragments were successfully amplified from rat and hamster spleen cells, and the PCR products appeared to be increased by ConA treatment. These PCR-amplified fragments were sequenced. The rat product was identical to the published sequence (13), and the PCR fragment of the Chinese hamster MHC class II showed considerable homology to rat (sequence identity, 85.9%; amino acid identity, 75.4%), which confirmed its identity. Having established the efficacy of the PCR, CHO and CHO-TSHR were stimulated with human, rat, or mouse IFN to examine whether IFN could up-regulate MHC class II expression on these cells. As shown in Fig. 5, mRNA expression of the MHC class II ß chain was barely detectable in control CHO and CHO-TSHR cells. However, after stimulation with human, rat, or mouse IFN, MHC class II mRNA expression was markedly increased (Fig. 5). These PCR products were sequenced and showed approximately 79% identity in nucleic acid sequence and approximately 65% in amino acid to that of the Chinese hamster, respectively. In addition, it should be noted that IFN treatment did not affect TSHR protein expression on CHO-TSHR, as assessed by flow cytometry using several monoclonal antibodies against hTSHR - and ß-subunits (data not shown).
fig.ommitteed50y!e, 百拇医药
Figure 5. RT-PCR detection of MHC class II. Spleen cells (left) from either a rat (R) or a Chinese hamster (H) were treated with or without ConA (5 µg/ml) for 48 h, and cDNA was PCR-amplified by primers for the MHC class II ß-chain (based on rat sequence) and for actin as a housekeeping gene. In the case of CHO cells, RT-PCR was performed using cDNA generated from CHO (middle) and CHO-TSHR cells (right) with or without IFN treatment. Human (Hu), rat (R), or mouse (M) IFN (1000 U/ml) was used to treat these CHO cell lines for 48 h before RNA extraction. cDNA was not added in blank (Bl).50y!e, 百拇医药
In group 2, animals (n = 9) were immunized with human IFN-stimulated CHO-TSHR cells, and human IFN-stimulated CHO cells were immunized as controls (n = 2). The hamsters in this second set showed severe erosive dermatitis on the abdomen after 4 wk of immunization, and four animals died at this time. T4 levels and TSHR-Ab were studied when the animals were killed (Table 2). T4 levels were all normal in this group, although one of the sera showed TSHR-Ab by flow cytometry, and three sera contained TSHR-Ab detected by competition assay. We also noticed that sera from the CHO-TSHR-immunized animals significantly suppressed basal cAMP levels in CHO-TSHR cells (P < 0.01), and three sera had significant TSHR blocking activity as evidenced by inhibition of TSH-induced generation of cAMP (Table 2).
fig.ommitteed%hrsqw[, http://www.100md.com
Table 2. Autoimmune parameters in Chinese hamsters immunized with IFN-stimulated CHO-TSHR (group 2)%hrsqw[, http://www.100md.com
Thyroid specimens from group 2 showed no evidence of thyroid stimulation and no lymphocyte infiltration. However, thyroid epithelial cells from CHO-TSHR-immunized hamsters were remarkably thin when compared with thyroid epithelial cells from CHO immunized hamsters (Fig. 6). Reduced colloid content was also apparent. These changes were most evident in the hamster with the strongest blocking TSHR-Ab. Thus, we concluded that such thyroids were atrophic due to blocking TSHR-Ab. Thyroid atrophy was seen in three of five animals in group 2 and was not seen in group 1. This difference was statistically significant (P < 0.05).%hrsqw[, http://www.100md.com
fig.ommitteed%hrsqw[, http://www.100md.com
Figure 6. Atrophy of thyroid epithelial cells in animals immunized with IFN-stimulated CHO-TSHR cells. The thyroid glands from the Chinese hamsters immunized with IFN-stimulated CHO-TSHR cells showed thinning of the thyroid epithelial cells with reduced colloid content (B and D), compared with thyroid glands from animals immunized with IFN-stimulated CHO cells (A and C), which appeared normal. Magnification: A and B, x100; C and D, x200.
Eye specimens from animals with strong TSHR-Ab in groups 1 and 2 were examined, but no abnormal retro-ocular findings, such as lymphocytic infiltration, were seen except an occasional lacrimal gland infiltrate that was also seen in controls (data not shown).l, http://www.100md.com
Immunization with AdTSHR (group 3)l, http://www.100md.com
On the basis of the results from groups 1 and 2, we concluded that the hamster could be used as an animal model of TSHR autoimmunity. We then attempted a better immunization protocol aimed at obtaining higher levels of TSHR-Ab. Recently, Nagayama et al. (8) generated a new mouse model of Graves’ disease using an AdTSHR. In their model, up to 50% of BALB/c mice immunized with AdTSHR developed Graves’ hyperthyroidism, which was much higher than approximately 25% disease induction seen in other models (3). However, this adenovirusinduced mouse model still lacked intrathyroidal lymphocytic infiltration. In group 3, we combined the advantage of the hamster with this same adenovirus construct. In addition to this change, and with the aim of eventually obtaining monoclonal antibodies, we employed the Armenian hamster that is known to provide more stable hybridomas and to fuse with a higher efficiency (14).
Animals (n = 10) were immunized with 5 x 1011 particles per animal of AdTSHR. No abnormal behavior consistent with Graves’ hyperthyroidism was seen during the immunization protocol. T4 levels measured after 12 wk of immunization are summarized in Table 3. As seen, T4 levels from four animals were elevated after 18 wk of follow-up. Sera were also examined for TSHR-Ab using flow cytometry analysis with CHO-TSHR cells. In contrast to CHO-TSHR immunization in groups 1 and 2, no immunoabsorption was needed to measure TSHR-Ab by flow cytometry (data not shown). Sera were strongly positive for TSHR-Ab when using a secondary antibody specific to Armenian hamster IgG1 and IgG2+3 (Fig. 7). As seen in Table 3, the nine sera positive for TSHR-Ab were all IgG2+3. The mean fluorescent intensity of positive samples on CHO-TSHR was 46.2 ± 10.5 (n = 9) in IgG2+3 vs. 8.3 ± 3.0 (n = 3) in IgG1 (P < 0.001). IgG1 TSHR-Ab were only detected in three animals in low titer. Thus, TSHR-Ab generated in this model were predominantly IgG2+3. TSHR-Ab detected by competition assay were also seen in all sera. The percentage competition was more than 90% in nine animals, and one showed moderate competition (Table 3). Stimulating TSHR-Ab were also detected in two animals by using the cAMP response in CHO-TSHR. Four of the thyroid specimens from group 3 also showed thyroid hypertrophy associated with papillary protrusion (Fig. 8). However, there was no intrathyroidal lymphocyte infiltration as seen earlier in group 1. There were also no retro-ocular findings seen.
fig.ommitteed1y$':, 百拇医药
Table 3. Autoimmune parameters in the Armenian hamsters immunized with AdTSHR (group 3)1y$':, 百拇医药
fig.ommitteed1y$':, 百拇医药
Figure 7. IgG isotype-specific TSHR-Ab determined by flow cytometry in sera from the Armenian hamster immunized with AdTSHR (group 3). IgG2+3 TSHR-Ab (A), but not IgG1 (C), were detected in serum from animal no. 3 (Table 3); however, in serum from animal no. 8, both IgG2+3 (B) and IgG1 (D) TSHR-Ab were detected. The thin and thick lines indicate staining on CHO and CHO-TSHR cells, respectively. IgG1 was measured in FL2 and IgG2+3 in FL1.1y$':, 百拇医药
fig.ommitteed1y$':, 百拇医药
Figure 8. Thyroid hypertrophy in the Armenian hamsters. In contrast to the normal thyroid follicular structure (A and C), the thyroid epithelial cells from hyperthyroid hamsters showed marked hypertrophy associated with papillary protrusion (B and D). No lymphocytic infiltration was seen. Magnification: A and B, x100; C and D, x200.1y$':, 百拇医药
Comparison of TSHR-Ab induced in the three immunization protocolh|, 百拇医药
To compare the levels of TSHR-Ab determined by flow cytometry, all data were obtained using the same secondary antibody against hamster IgG (Table 4). Compared with group 1, we expected stronger TSHR antigen exposure associated with enhanced MHC class II expression on CHO-TSHR cells in group 2. However, there was no statistical difference in the levels of TSHR-Ab or efficiency of TSHR-Ab induction between groups 1 and 2. It should be noted that this comparison was likely to be influenced by the fact that group 2 immunization was not as extensive as group 1 because of the adverse effects of the immunization regime. Compared with CHO-TSHR immunization (groups 1, 2, and 1+2), TSHR-Ab determined by competition assay after AdTSHR (group 3) was significantly stronger (P < 0.001), and TSHR-Ab induction was also significantly more effective (P < 0.001). The level of TSHR-Ab determined by flow cytometry was comparable between group 1+2 and group 3, but such TSHR-Ab were more effectively induced in group 3 compared with group 1 (P < 0.005), group 2 (P < 0.02), and group 1+2 (P < 0.001). Thus, in the hamster model, adenovirus immunization induced TSHR-Ab, as determined by competition assay and flow cytometry, more successfully than CHO-TSHR immunization.
fig.ommitteed7*cw1w, 百拇医药
Table 4. TSHR-Ab induced in three immunization protocols7*cw1w, 百拇医药
Discussion7*cw1w, 百拇医药
We first attempted to generate a model of Graves’ disease in the hamster by immunizing with CHO cells stably overexpressing hTSHR antigen. Even with Chinese hamsters, we found a marked immune response to control CHO cell (Fig. 1A). However, after CHO-TSHR immunization, two hamsters showed marked thyroid hypertrophy and a focal lymphocyte infiltrate (Figs. 3 and 4) and also had stimulating TSHR-Ab in their sera. Stimulation of CHO-TSHR with human IFN before the immunization (group 2) appeared to enhance MHC class II mRNA in CHO-TSHR cells when assessed by semiquantitative RT-PCR (Fig. 5), but this did not lead to an enhancement of the development of Graves’ disease in this model. Only blocking TSHR-Ab were detected in hamsters treated in this way, and the thyroid glands from these animals showed atrophy and thinning of the thyroid epithelial cells (Fig. 6). To further improve the hamster model, adenovirus immunization was employed to obtain more effective exposure of TSHR antigen (group 3). We found that TSHR-Ab were detected by both flow cytometry and competition assay in all animals immunized (Table 3). However, the thyroid hypertrophy seen in these animals was not associated with an intrathyroidal lymphocytic infiltration (Fig. 8).
Initial attempts to generate a Graves’ disease animal model used immunization with recombinant hTSHR-extracellular domain protein from prokaryotic and eukaryotic cell expression systems (15). Although strong immune responses against the hTSHR were obtained, these TSHR-Ab antibodies only inhibited binding of TSH to the TSHR, even when associated with a thyroidal lymphocytic infiltration (16). One study reported generating stimulating antibodies associated with hyperthyroidism using hTSHR ectodomain protein, but the histological findings were of a destructive thyroiditis that seemed at odds with the observations (17). We and others previously concluded that an improved protocol would incorporate immunization with full-length hTSHR protein that had undergone appropriate post-translational processing and folding (15). This assumption was the basis of both the Shimojo mouse model (3) and DNA immunization (6). In the former model, TSHR protein was expressed on mammalian fibroblasts and in the latter, the TSHR protein was expressed on the host muscle. However, there has been difficulty in reproducing the DNA immunization model (7, 8), although TSHR-reactive T cells have been detected (7). One of the reasons for these difficulties may have been low TSHR antigen production and low exposure to the immune system compared with direct protein immunization (18). Indeed, low expression of the TSHR on muscle after DNA immunization has recently been shown (8).
Based on these observations, we undertook new strategies for generating an animal model for Graves’ disease. We first used a CHO cell line stably expressing high levels of functional hTSHR on the plasma membrane (CHO-TSHR). This cell line, named JPO9 (9), is highly sensitive to TSH stimulation (9, 10, 19), and the present study showed that the cells expressed MHC class II mRNA constitutively (Fig. 5). Although CHO cells were derived from the Chinese hamster, the strong background staining on flow cytometry using sera from immunized hamsters indicated that these outbred animals recognized many CHO cell membrane proteins as foreign antigens (Fig. 1A). This indicated that cell immunization induces strong nonspecific immune reaction in both a syngeneic model (20) and an allogeneic model (this study). Thus, the major advantage of the approach was the high density of functional TSHR protein available to the immune system. We succeeded in inducing intrathyroidal lymphocytic infiltration along with stimulating TSHR-Ab production in this model, but the rate of disease development was still only approximately 20%. As in previous studies (21), our findings demonstrated that the immune system recognized a variety of epitopes on the TSHR as evidenced by the results of a variety of assays for TSHR-Ab. In addition, the failure to induce stimulating TSHR-Ab in a large percentage of animals may simply have reflected the variable genetic susceptibility of hamsters to Graves’ disease (6, 8).
In an attempt to address some of these limitations, a second immunization protocol was performed using IFN to up-regulate MHC class II expression on antigen-presenting cells (Fig. 5). Although TSHR-Ab and thyroid histology in group 2 were examined at an earlier time point than in group 1, thyroid atrophy along with blocking TSHR-Ab were observed in three of five animals. This finding suggested that IFN stimulation skewed the host immune reaction toward the generation of blocking rather than stimulating TSHR-Ab. IFN is well known to enhance expression of MHC classes I and II. It also regulates expression of co-stimulatory molecules, such as B7-1 and B7-2, as well as affecting antigen processing and presentation (22, 23). Because IFN did not affect TSHR protein expression levels on CHO-TSHR cells (data not shown), we did not identify any other mechanism(s) in which the host immune reaction was skewed by IFN-stimulated CHO-TSHR. However, altered expression of co-stimulation molecules on CHO-TSHR cells and/or the different TSHR epitopes presented on IFN-stimulated CHO-TSHR cells from those on untreated cells might be involved, in addition to the quantity of MHC molecules. Human leukocyte antigen class II expression on the Graves’ thyroid gland has been widely investigated (24), suggesting its involvement in thyroid autoantigen presentation. Successful immunization of fibroblasts expressing both TSHR antigen and MHC class II, but not fibroblasts expressing only TSHR antigen, in the Shimojo model (3) further indicated an important role for MHC class II in the initiation of Graves’ disease. Graves’ hamsters in this study were generated by using CHO-TSHR cells that appeared to constitutively express MHC class II. Indeed, MHC class II overexpression by IFN stimulation prompted the generation of blocking TSHR-Ab rather than stimulating TSHR-Ab. Taken together, these results confirmed the important role(s) of MHC class II in Graves’ disease induction, and also indicated that IFN may modulate the immune response to TSHR antigen.
We concluded that the hamster could serve as an animal model of TSHR autoimmunity using CHO-TSHR cells and proceeded to further improve the model by immunizing with an adenovirus construct incorporating TSHR DNA. Here, we expected a more effective exposure of the immune system to TSHR antigen that could have induced Graves’ disease in higher frequency and still be associated with an intrathyroidal lymphocytic infiltration. Indeed, in these animals the TSHR-Ab levels were significantly greater, confirming that the AdTSHR immunization triggered a stronger humeral immune response to TSHR antigen. This difference may also have been due to a greater genetic susceptibility in the Armenian hamster in addition to the immunization protocol. However, both the Chinese and the Armenian hamsters remain outbred, suggesting this was an unlikely explanation. The resulting thyroid hypertrophy was not associated with an intrathyroidal lymphocytic infiltration. Their TSHR-Ab Ig subtype, determined by flow cytometery, suggested that these animals did have an immune response that could be classified as preferentially T helper, but this clearly did not lead to lymphocytic infiltration. Furthermore, because these hamsters had no intrathyroidal lymphocytic infiltration, it was also clear that neither stronger exposure of TSHR antigen nor high levels of TSHR-Ab were an indication of such infiltration.
In summary, we demonstrated that the hamster could serve as an animal model of TSHR autoimmunity. Our study suggested a modulating role for cytokines and MHC class II expression in the induction of Graves’ disease. We also demonstrated that an adenovirus vector was a highly effective immunogen for the induction of TSHR autoimmunity, achieving significantly greater levels of TSHR-Ab. In addition, our data suggested that an intrathyroidal lymphocytic infiltration was not directly related to levels of TSHR-Ab or exposure to more TSHR antigen. Nevertheless, the hamster offers promising insights as a model for thyroid autoimmune disease.})/$8, 百拇医药
Acknowledgments})/$8, 百拇医药
We thank Dr. Y. Nagayama, Nagasaki University School of Medicine, Nagasaki, Japan, for providing AdTSHR.})/$8, 百拇医药
Received June 4, 2002.})/$8, 百拇医药
Accepted for publication November 4, 2002.})/$8, 百拇医药
References})/$8, 百拇医药
Vassart G, Dumont JE 1992 The thyrotropin receptor and the regulation of thyrocyte function and growth. Endocr Rev 13:596–611
Jaume JC, Kakinuma A, Chazenbalk GD, Rapoport B, McLachlan SM 1997 Thyrotropin receptor autoantibodies in serum are present at much lower levels than thyroid peroxidase autoantibodies: analysis by flow cytometry. J Clin Endocrinol Metab 82:500–507?%, http://www.100md.com
Shimojo N, Kohno Y, Yamaguchi K, Kikuoka S, Hoshioka A, Niimi H, Hirai A, Tamura Y, Saito Y, Kohn LD, Tahara K 1996 Induction of Graves-like disease in mice by immunization with fibroblasts transfected with the thyrotropin receptor and a class II molecule. Proc Natl Acad Sci USA 93:11074–11079?%, http://www.100md.com
Kita M, Ahmad L, Marians RC, Vlase H, Unger P, Graves PN, Davies TF 1999 Regulation and transfer of a murine model of thyrotropin receptor antibody mediated Graves’ disease. Endocrinology 140:1392–1398?%, http://www.100md.com
Jaume JC, Rapoport B, McLachlan SM 1999 Lack of female bias in a mouse model of autoimmune hyperthyroidism (Graves’ disease). Autoimmunity 29:269–272?%, http://www.100md.com
Costagliola S, Many MC, Denef JF, Pohlenz J, Refetoff S, Vassart G 2000 Genetic immunization of outbred mice with thyrotropin receptor cDNA provides a model of Graves’ disease. J Clin Invest 105:803–811
Pichurin P, Yan XM, Farilla L, Guo J, Chazenbalk GD, Rapoport B, McLachlan SM 2001 Naked TSH receptor DNA vaccination: a TH1 T cell response in which interferon- production, rather than antibody, dominates the immune response in mice. Endocrinology 142:3530–3536[a\\z, 百拇医药
Nagayama Y, Kita-Furuyama M, Ando T, Nakao K, Mizuguchi H, Hayakawa T, Eguchi K, Niwa M 2002 A novel murine model of Graves’ hyperthyroidism with intramuscular injection of adenovirus expressing the thyrotropin receptor. J Immunol 168:2789–2794[a\\z, 百拇医药
Perret J, Ludgate M, Libert F, Gerard C, Dumont JE, Vassart G, Parmentier M1990 Stable expression of the human TSH receptor in CHO cells and characterization of differentially expressing clones. Biochem Biophys Res Commun 171:1044–1050[a\\z, 百拇医药
Michelangeli VP, Munro DS, Poon CW, Frauman AG, Colman PG 1994 Measurement of thyroid stimulating immunoglobulins in a new cell line transfected with a functional human TSH receptor (JPO9 cells), compared with an assay using FRTL-5 cells. Clin Endocrinol (Oxf) 40:645–652[a\\z, 百拇医药
Imaizumi M, Pritsker A, Unger P, Davies TF 2002 Intrathyroidal fetal microchimerism in pregnancy and postpartum. Endocrinology 143:247–253
Davies TF, Yang C, Platzer M 1987 Cloning the Fisher rat thyroid cell line (FRTL-5): variability in clonal growth and 3, 5'-cyclic adenosine monophosphate response to thyrotropin. Endocrinology 121:78–83v3, 百拇医药
Robertson KA, McMaster WR 1985 Complete structure of a rat RT1 E ß chain: extensive conservation of MHC class II ß chains. J Immunol 135:4095–4099v3, 百拇医药
Lengweiler S, Smyth SS, Jirouskova M, Scudder LE, Park H, Moran T, Coller BS 1999 Preparation of monoclonal antibodies to murine platelet glycoprotein IIb/IIIa (IIbß3) and other proteins from hamster-mouse interspecies hybridomas. Biochem Biophys Res Commun 262:167–173v3, 百拇医药
Davies TF, Vlase H, Kita M 1999 The search for an animal model for Graves’ disease. In: Shoenfeld Y, ed. The decade of autoimmunity. New York: Elsevier Science; 43–50v3, 百拇医药
Costagliola S, Many MC, Stalmans-Falys M, Tonacchera M, Vassart G, Ludgate M 1994 Recombinant thyrotropin receptor and the induction of autoimmune thyroid disease in BALB/c mice: a new animal model. Endocrinology 135:2150–2159
Kaithamana S, Fan J, Osuga Y, Liang SG, Prabhakar BS 1999 Induction of experimental autoimmune Graves’ disease in BALB/c mice. J Immunol 163:5157–5164:4:|, 百拇医药
Ito K, Takeuchi Y, Kato S 2000 Strain-dependent antibody response induced by DNA immunization. Immunol Lett 74:245–250:4:|, 百拇医药
Patibandla SA, Dallas JS, Seetharamaiah GS, Tahara K, Kohn LD, Prabhakar BS 1997 Flow cytometric analyses of antibody binding to Chinese hamster ovary cells expressing human thyrotropin receptor. J Clin Endocrinol Metab 82:1885–1893:4:|, 百拇医药
Yan XM, Guo J, Pichurin P, Tanaka K, Jaume JC, Rapoport B, McLachlan SM 2000 Cytokines, IgG subclasses and costimulation in a mouse model of thyroid autoimmunity induced by injection of fibroblasts co-expressing MHC class II and thyroid autoantigens. Clin Exp Immunol 122:170–179:4:|, 百拇医药
Vlase H, Nakashima M, Graves PN, Tomer Y, Morris JC, Davies TF 1995 Defining the major antibody epitopes on the human thyrotropin receptor in immunized mice: evidence for intramolecular epitope spreading. Endocrinology 136:4415–4423:4:|, 百拇医药
Boehm U, Klamp T, Groot M, Howard JC 1997 Cellular responses to interferon-. Annu Rev Immunol 15:749–795:4:|, 百拇医药
York IA, Rock KL 1996 Antigen processing and presentation by the class I major histocompatibility complex. Annu Rev Immunol 14:369–396:4:|, 百拇医药
Davies TF, Piccinini LA, Roman SH, Hirose W, Neufeld DS 1988 Role of MHC class II antigen expression in thyroid autoimmunity. Ann NY Acad Sci 546:151–163(Takao Ando Misa Imaizumi Peter Graves Pamela Unger and Terry F. Davies)