Evidence that the C terminus of the A subunit suppresses thyrotropin receptor constitutive activity

The TSH receptor (TSHR), unlike the LH receptor (LHR), has considerable ligand-independent adenylyl cyclase activity, a feature of pathophysiological importance. The TSHR ectodomain partially suppresses constitutive activity, an effect reversed by trypsin treatment of intact cells. Localizing the functional site of trypsin action would provide insight into how the TSHR ectodomain exerts its constraint. For this purpose, we examined the effect of trypsin on intact cells expressing a series of modified TSHR. Trypsin did not increase cAMP production by a chimeric TSH-LH receptor involving substitution of TSHR residues 261-418 (the ectodomain C terminus). In contrast, with the wild-type TSHR, trypsin enhanced constitutive activity despite mutation of the following potential tryptic cleavage sites [arginine (R) and lysine (K) residues]: 1) K565, K651, K660 in the extracellular loops of the serpentine region; 2) B subunit juxtamembrane residues K371, K401, K415; 3) A subunit residues R310, R312, K313. We previously excluded K337 and K339 from being implicated in TSHR tryptic activation. By exclusion, only one R/K cluster remains as a possible target for the functional effect of trypsin, namely K287, K290, K291, and R293. Mutation of this cluster is incompatible with TSHR cell surface expression. However, tryptic clipping at this locus would reproduce a previously demonstrated structural effect of trypsin on the TSHR, removal of about a 2-kDa polypeptide fragment extending downstream from the locus to the C terminus of the A subunit. Taken together, these data suggest that the C terminus of the A subunit functions as a suppressor of TSHR constitutive activity.

The thyrotropin receptor autoantigen in Graves disease is the culprit as well as the victim

Graves disease, a common organ-specific autoimmune disease affecting humans, differs from all other autoimmune diseases in being associated with target organ hyperfunction rather than organ damage. Clinical thyrotoxicosis is directly caused by autoantibodies that activate the thyrotropin receptor (TSHR). The etiology of Graves disease is multifactorial, with nongenetic factors playing an important role. Of the latter, there is the intriguing possibility that the molecular structure of the target antigen contributes to the development of thyroid-stimulatory autoantibodies (TSAb's). Among the glycoprotein hormone receptors, only the TSHR undergoes intramolecular cleavage into disulfide-linked subunits with consequent shedding of some of the extracellular, autoantibody-binding A subunits. Functional autoantibodies do not arise to the noncleaving glycoprotein hormone receptors. Recently, TSAb's were found to preferentially recognize shed, rather than attached, A subunits. Here we use a new adenovirus-mediated animal model of Graves disease to show that goiter and hyperthyroidism occur to a much greater extent when the adenovirus expresses the free A subunit as opposed to a genetically modified TSHR that cleaves minimally into subunits. These data show that shed A subunits induce or amplify the immune response leading to hyperthyroidism and provide new insight into the etiology of Graves disease.

Insight into antibody responses induced by plasmid or adenoviral vectors encoding thyroid peroxidase, a major thyroid autoantigen

Plasmid and adenoviral vectors have been used to generate antibodies in mice that resemble human autoantibodies to the thyrotrophin receptor. No such studies, however, have been performed for thyroid peroxidase (TPO), the major autoantigen in human thyroiditis. We constructed plasmid and adenovirus vectors for in vivo expression of TPO. BALB/c mice were immunized directly by intramuscular injection of TPO-plasmid or TPO-adenovirus, as well as by subcutaneous injection of dendritic cells (DC) infected previously with TPO-adenovirus. Intramuscular TPO-adenovirus induced the highest, and TPO-plasmid the lowest, TPO antibody titres. Mice injected with TPO-transfected DC developed intermediate levels. Antibodies generated by all three approaches had similar affinities (Kd approximately 10(-9)M) and recognized TPO expressed on the cell-surface. Their epitopes were analysed in competition assays using monoclonal human autoantibodies that define the TPO immunodominant region (IDR) recognized by patients with thyroid autoimmune disease. Surprisingly, high titre antibodies generated using adenovirus interacted with diverse TPO epitopes largely outside the IDR, whereas low titre antibodies induced by DNA-plasmid recognized restricted epitopes in the IDR. This inverse relationship between antibody titre and restriction to the IDR is likely to be due to epitope spreading following strong antigenic stimulation provided by the adenovirus vector. However, TPO antibody epitope spreading does not occur in Hashimoto's thyroiditis, despite high autoantibody levels. Consequently, these data support the concept that in human thyroid autoimmunity, factors besides titre must play a role in shaping an autoantibody epitopic profile.

The cysteine-rich amino terminus of the thyrotropin receptor is the immunodominant linear antibody epitope in mice immunized using naked deoxyribonucleic acid or adenovirus vectors

Experimental Graves' disease is more effectively produced by immunization approaches involving in vivo TSH receptor (TSHR) expression than by conventional immunization with TSHR protein and adjuvant. Unlike conformational epitopes that are extremely difficult to define, linear epitopes can be readily assessed using synthetic peptides. TSHR linear epitopes are well characterized in conventionally immunized animals, but there is no information for animals vaccinated with TSHR DNA in plasmid or adenovirus vectors. We used synthetic peptides to characterize linear epitopes in mice immunized by in vivo expression of TSHR DNA. TSHR adenovirus-injected mice had higher antibody levels than TSHR DNA-vaccinated mice. However, the dominant peptide recognized in both groups was the TSHR cysteine-rich N terminus (residues 22-41). Sera from TSHR adenovirus-immunized (but not TSHR DNA-vaccinated) mice interacted to a lesser extent with peptides encompassing residues 352-401, which include the region deleted following TSHR cleavage as well as the ectodomain juxta-membrane region. Although antibodies characterized using synthetic peptides are probably TSH blockers or nonfunctional, stimulating antibodies may recognize linear components in a conformational epitope. The cysteine-rich TSHR N terminus is functionally important in the action of stimulating TSHR autoantibodies in humans. The immunodominance of the same region in immunized mice suggests that this region may also be immunodominant in humans.

Targeted restoration of cleavage in a noncleaving thyrotropin receptor demonstrates that cleavage is insufficient to enhance ligand-independent activity

Two unusual features of the TSH receptor (TSHR) ectodomain are its intramolecular cleavage at the cell surface into disulfide-linked subunits and its constraint of ligand-independent (constitutive) activity inherent to the serpentine region. Whether ectodomain cleavage alters the level of TSHR constitutive activity is an important unanswered question. To address this issue, we used a TSHR engineered so as not to undergo spontaneous cleavage into subunits (deletion of amino acid residues 317-366 and GQE(367-369)NET substitution). Into this noncleaving TSHR (termed TSHR-D1-NET), we introduced thrombin recognition motifs (termed Thr 6 and Thr 18) at the site of spontaneous cleavage. Treatment of intact Chinese hamster ovary cells expressing TSHR-D1-NET-Thr 6 and -Thr 18 with thrombin induced cleavage into A and B subunits, as determined by (125)I-TSH covalent cross-linking. Nevertheless, constitutive activity of the thrombin-cleaved TSHR was unaltered. The level of TSHR constitutive activity was, therefore, fully dissociated from intramolecular cleavage into subunits. Trypsin treatment of the same cells expressing the noncleaving TSHR also generated disulfide-linked A and B subunits but, in contrast to thrombin, enhanced TSHR constitutive activity. Therefore, the activating effect of trypsin appears to involve clipping at an additional, as-yet unidentified, site. In summary, our data demonstrate that TSHR cleavage is, by itself, insufficient to reduce TSHR ectodomain constraint on ligand-independent constitutive activity. These data are consistent with other evidence that A subunit shedding consequent to TSHR cleavage is a critical factor in enhancing TSHR constitutive activity.

Prevention of autoantibody-mediated Graves'-like hyperthyroidism in mice with IL-4, a Th2 cytokine

Graves' hyperthyroidism has long been considered to be a Th2-type autoimmune disease because it is directly mediated by autoantibodies against the thyrotropin receptor (TSHR). However, several lines of evidence have recently challenged this concept. The present study evaluated the Th1/Th2 paradigm in Graves' disease using a recently established murine model involving injection of adenovirus expressing the TSHR (AdCMVTSHR). Coinjection with adenovirus expressing IL-4 (AdRGDCMVIL-4) decreased the ratio of Th1/Th2-type anti-TSHR Ab subclasses (IgG2a/IgG1) and suppressed the production of IFN-gamma by splenocytes in response to TSHR Ag. Importantly, immune deviation toward Th2 was accompanied by significant inhibition of thyroid-stimulating Ab production and reduction in hyperthyroidism. However, in a therapeutic setting, injection of AdRGDCMVIL-4 alone or in combination with AdCMVTSHR into hyperthyroid mice had no beneficial effect. In contrast, coinjection of adenoviruses expressing IL-12 and the TSHR promoted the differentiation of Th1-type anti-TSHR immune responses as demonstrated by augmented Ag-specific IFN-gamma secretion from splenocytes without changing disease incidence. Coinjection of adenoviral vectors expressing IL-4 or IL-12 had no effect on the titers of anti-TSHR Abs determined by ELISA or thyroid-stimulating hormone-binding inhibiting Ig assays, suggesting that Ab quality, not quantity, is responsible for disease induction. Our observations demonstrate the critical role of Th1 immune responses in a murine model of Graves' hyperthyroidism. These data may raise a cautionary note for therapeutic strategies aimed at reversing Th2-mediated autoimmune responses in Graves' disease in humans.

A major role for non-major histocompatibility complex genes but not for microorganisms in a novel murine model of Graves' hyperthyroidism

The etiology of Graves' disease is multifactorial. We investigated the role of genetic and environmental factors on the susceptibility to Graves' hyperthyroidism using a new murine model. Intramuscular injection of recombinant adenovirus expressing the thyrotropin receptor (AdCMVTSHR) induces Graves'-like hyperthyroidism (thyrotropin receptor [TSHR] antibodies, elevated thyroxine, and diffuse goiter) in more than 50% of female BALB/c mice. The relative contributions of major histocompatibility complex (MHC) and non-MHC genes on the susceptibility to hyperthyroidism were studied by immunizing BALB/c (H-2d), BALB.K (H-2k), and DBA/2J (H-2d) mice with AdCMVTSHR. Hyperthyroidism developed in approximately 50% of BALB/c and BALB.K mice but only 5% of DBA/2J mice, indicating a major role for non-MHC genes in disease development. The effect of environmental microorganisms was evaluated by comparing disease incidence in BALB/c mice maintained in pathogen-free conditions versus those in nonsterile, conventional housing, as well as by coadministering microorganism components (Escherichia coli lipopolysaccharide or yeast zymosan A) as adjuvants with AdCMVTSHR. Neither type of exposure to environmental pathogens influenced disease induction. In conclusion, non-MHC genes, but not infectious organisms, play a major role in the etiology of this novel murine model of Graves' disease.

Dendritic cells infected with adenovirus expressing the thyrotrophin receptor induce Graves' hyperthyroidism in BALB/c mice

Dendritic cells (DCs) are the most potent antigen-presenting cells and a prerequisite for the initiation of primary immune response. This study was performed to investigate the contribution of DCs to the initiation of Graves' hyperthyroidism, an organ-specific autoimmune disease in which the thyrotrophin receptor (TSHR) is the major autoantigen. DCs were prepared from bone marrow precursor cells of BALB/c mice by culturing with granulocyte macrophage-colony stimulating factor and interleukin-4. Subcutaneous injections of DCs infected with recombinant adenovirus expressing the TSHR (but not beta-galactosidase) in syngeneic female mice induced Graves'-like hyperthyroidism (8 and 35% of mice after two and three injections, respectively) characterized by stimulating TSHR antibodies, elevated serum thyroxine levels and diffuse hyperplasitc goiter. TSHR antibodies determined by ELISA were of both IgG1 (Th2-type) and IgG2a (Th1-type) subclasses, and splenocytes from immunized mice secreted interferon-gamma (a Th1 cytokine), not interleukin-4 (a Th2 cytokine), in response to TSHR antigen. Surprisingly, IFN-gamma secretion, and induction of antibodies and disease were almost completely suppressed by co-administration of alum/pertussis toxin, a Th2-dominant adjuvant, whereas polyriboinosinic polyribocytidylic acid, a Th1-inducer, enhanced splenocyte secretion of IFN-gamma without changing disease incidence. These observations demonstrate that DCs efficiently present the TSHR to naive T cells to induce TSHR antibodies and Graves'-like hyperthyroidism in mice. In addition, our results challenge the previous concept of Th2 dominance in Graves' hyperthyroidism and provide support for the role of Th1 immune response in disease pathogenesis.

Thyroglobulin-thyroperoxidase autoantibodies are polyreactive, not bispecific: analysis using human monoclonal autoantibodies

Autoantibodies (Ab) to thyroglobulin (Tg) and to thyroid peroxidase (TPO) are reported to share common epitopes, and an assay for bispecific TgPOAb has been developed that may distinguish between different clinical presentations of thyroid autoimmunity. We sought to clone TgPOAb from an Ig gene combinatorial library constructed from B cells infiltrating the thyroid of a patient with TgPOAb. As described for isolating serum TgPOAb, we panned the phage display library by alternating from Tg- to TPO-coated ELISA wells. After panning, the library was enriched for TgPO-binding phage. Of 526 clones tested for expressed Ab, most were negative; 3 clones were specific for Tg, and 5 clones specifically recognized TPO. Antibody from a single clone, encoded by a non-Tg, non-TPO Ig heavy chain gene, bound both Tg and TPO (TgPO activity). However, this antibody also bound equally well to nonthyroid antigens. In conclusion, enrichment for Tg- and TPO-binding phage was largely attributable to phage specific for either Tg or TPO. This finding, albeit from a single patient, questions previous observations of serum TgPOAb prepared by affinity chromatography. Combined with the isolation of a polyreactive monoclonal antibody, our data provide powerful evidence against shared, cross-reactive epitopes on 2 major thyroid autoantigens.

Evidence that the complement control protein-epidermal growth factor-like domain of thyroid peroxidase lies on the fringe of the immunodominant region recognized by autoantibodies

There is no consensus regarding the location of the immunodominant region (IDR) on thyroid peroxidase (TPO) recognized by the majority of autoantibodies. Strong evidence indicates that it lies upstream of amino acid 741. However, an epitope has been localized to downstream residues 742-848 encompassing a disulfide-rich complement control protein (CCP)-like and epidermal growth factor (EGF)-like domain. To determine whether these domains comprise part of the IDR, we used a recombinant CCP/EGF-like polypeptide to screen a thyroid B-cell-derived immunoglobulin gene phage display library. Two unusual TPO autoantibodies were isolated. Neither was among the 83 clones previously obtained by panning the same library on native or denatured TPO, or TPO with the IDR masked. Fab from these clones bound native TPO, one with high affinity (Kd 6 x 10(-10) M), and both recognized TPO expressed on the surface of mammalian cells. Phage-expressing multiple copies of the antibody (multivalent), but not monovalent Fab from these clones, bound to the CCP/EGF polypeptide. Most important, inhibition of TPO binding by autoantibodies to the IDR indicated that the epitopes of the two new autoantibodies overlap with this region. The value of these two rare clones lies in the insight they provide into the location of the TPO IDR. From their binding characteristics, we deduce that the CCP/EGF-like domain lies on the fringe of the TPO immunodominant region.

Localization of the thyroid peroxidase autoantibody immunodominant region to a junctional region containing portions of the domains homologous to complement control protein and myeloperoxidase

Thyroid peroxidase (TPO) autoantibody epitopes are largely restricted to an immunodominant region (IDR) on the extracellular region of the native molecule. Localization of the IDR has been a longstanding and difficult goal. The TPO extracellular region comprises a large myeloperoxidase-like domain, linked to the plasma membrane by two smaller domains with homology to complement control protein (CCP) and epidermal growth factor (EGF), respectively. Recent studies have focused on the CCP- and EGF-like domains as the putative location of the TPO autoantibody IDR. To address this issue, we attempted to express on the surface of transfected cells native TPO in which the CCP- and EGF-like domains were deleted, either together or individually. We used a quartet of human monoclonal autoantibodies that define the TPO IDR, as well as polyclonal TPO autoantibodies in patients' sera, to detect these mutated TPO molecules by flow cytometry. The combined CCP/EGF-like domain deletion did not produce a signal with TPO autoantibodies but did not traffic to the cell surface. In contrast, both monoclonal and polyclonal autoantibodies recognized TPO with the juxtamembrane EGF-like domain deleted equally as well as the wild-type TPO on the cell surface. TPO with the CCP-like domain deleted expressed normally on the cell surface, as determined using the polyclonal mouse antiserum. Nevertheless, this modified TPO molecule was recognized very poorly by both the human monoclonal autoantibodies and the polyclonal autoantibodies in patients' sera. In conclusion, we have clearly excluded the juxtamembrane EGF-like domain as being part of the IDR. In contrast, a component of the CCP-like domain does contribute to the IDR. These data, together with findings from other studies, localize the TPO autoantibody IDR to the junction of the CCP-like domain and the much larger myeloperoxidase-like domain on TPO.

Peptide scanning for thyrotropin receptor T-cell epitopes in mice vaccinated with naked DNA

Vaccinating mice with DNA encoding the thyrotropin receptor (TSHR), the major autoantigen in Graves' disease, induces memory T cells that secrete interferon-gamma (IFN-gamma) in response to TSHR antigen. We used a panel of 29 synthetic TSHR peptides encompassing the ectodomain and three extracellular loops to identify T-cell epitopes after TSHR-DNA vaccination of BALB/c, NOD.H-2h4, and AKR/N mice. These strains were chosen because of their previous use in animal models of thyroid autoimmunity. In initial studies, challenge of splenocytes with TSHR protein induced IFN-gamma and tumor necrosis factor-alpha (TNF-alpha) production in all three strains of mice. BALB/c mice recognized three peptides, all in the TSHR A subunit. These peptides differed from the four peptides recognized by nonobese diabetic (NOD mice NOD H-2h4). Three of the latter were also in the A subunit. The fourth was within the intervening C peptide region excised on TSHR cleavage into A and B subunits. Because of high and erratic responses in AKR/N mice, their TSHR T-cell epitopes could not be determined. In summary, we report that TSHR DNA vaccination of BALB/c and NOD.H-2h4 mice, with different major histocompatibility complex (MHC) class II genes (I-Ad and I-Ak, respectively), recognize restricted, nonoverlapping TSHR T-cell epitopes, nearly all in the TSHR A subunit.

Human thyroperoxidase folds in one complex B-cell immunodominant region

Human thyroperoxidase (TPO) ectodomain is successively made of myeloperoxidase-, complement control protein repeat-, and epidermal growth factor-like gene modules. However, the TPO immunodominant region targeted by autoantibodies from patients with an autoimmune thyroid disease has not been mapped on the molecule. Here, we used two purified recombinant TPO peptides produced in eukaryotic cells, which correspond to the major first and the further two gene modules of TPO. We compared by ELISA their respective immunoreactivity with that of the recombinant soluble TPO containing all the three gene modules. We used well-characterized murine and human TPO monoclonal antibodies and human autoantibodies affinity-purified from a large pool of patients' sera. We found that the TPO immunodominant region was susceptible to denaturation and required the integrity of the molecule to be correctly expressed. We concluded that TPO B-cell autoepitopes are made by amino acids from the three gene modules, which fold in one highly conformational immunodominant region.

Thyroid-stimulating autoantibodies in Graves disease preferentially recognize the free A subunit, not the thyrotropin holoreceptor

Graves disease is directly caused by thyroid-stimulating autoantibodies (TSAb's) that activate the thyrotropin receptor (TSHR). We observed upon flow cytometry using intact cells that a mouse mAb (3BD10) recognized the TSHR ectodomain with a glycosidylphosphatidylinositol (ECD-GPI) anchor approximately tenfold better than the same ectodomain on the wild-type TSHR, despite the far higher level of expression of the latter. The 3BD10 epitope contains the N-terminal cysteine cluster critical for TSAb action. Consequently, we hypothesized and confirmed that TSAb (but not thyrotropin-blocking autoantibodies [TBAb's]) also poorly recognize the wild-type TSHR relative to the ECD-GPI. Despite poor recognition by TSAb of the holoreceptor, soluble TSHR A subunits (known to be shed from surface TSHR) fully neutralized autoantibody-binding activity. These data indicate that the epitope(s) for TSAb's, but not for TBAb's, are partially sterically hindered on the holoreceptor by the plasma membrane, the serpentine region of the TSHR, or by TSHR dimerization. However, the TSAb epitope on the soluble A subunit is freely accessible. This observation, as well as other evidence, supports the concept that A subunit shedding either initiates or amplifies the autoimmune response to the TSHR, thereby causing Graves disease in genetically susceptible individuals.

Immune deviation away from Th1 in interferon-gamma knockout mice does not enhance TSH receptor antibody production after naked DNA vaccination

TSH receptor (TSHR) DNA vaccination induces high TSHR antibody levels in BALB/c mice housed in a conventional facility. However, under pathogen-free conditions, we observed a Th1 cellular response to TSHR antigen characterized by interferon-gamma (IFN gamma) production. In the present study we investigated the effect on TSHR DNA vaccination of diverting the cytokine milieu away from Th1 using 1) IFN gamma knockout BALB/c mice, and 2) wild-type mice covaccinated with DNA for the TSHR and for IFN gamma/receptor-Fc protein that prevents IFN gamma from binding to its receptor. Neither approach enhanced TSHR antibody levels, although splenocyte IFN gamma production in response to TSHR antigen was absent (IFN gamma knockouts) or reduced (IFN gamma receptor-Fc). Moreover, production of IL-2, another Th1 cytokine, but not Th2 cytokines, indicated that neither strategy overcame the Th1 bias of im DNA vaccination. Importantly, splenocyte production of IFN gamma and IL-2 provides a sensitive detection system for TSHR-specific T cells. Unexpectedly, higher TSHR antibody levels developed in rare mice. High titer animals had TSHR-specific responses of both Th2 and Th1 types, whereas low titer animals had Th1-restricted TSHR responses. The heterogeneity of responses induced by TSHR DNA vaccination in mice may provide insight into the titers and IgG subclasses of spontaneous autoantibodies in humans.

Naked TSH receptor DNA vaccination: A TH1 T cell response in which interferon-gamma production, rather than antibody, dominates the immune response in mice

Two approaches have been developed to induce TSH receptor antibodies in mice with properties resembling those in Graves' disease, the Shimojo model of injecting live fibroblasts coexpressing the TSH receptor and major histocompatibility complex antigen Class II, and TSH receptor-DNA vaccination. Thyroid-stimulating antibodies appear to occur less commonly after DNA vaccination, but there has been no direct comparison of these models. We performed a three-way comparison of 1) AKR/N and 2) BALB/c mice vaccinated with TSH receptor-DNA and 3) AKR/N mice injected with fibroblasts expressing the TSH receptor and the major histocompatibility complex antigen class II of AKR/N mice. TSH receptor-DNA vaccinated mice had low or undetectable levels of TSH receptor antibodies determined by ELISA or flow cytometry. Nonspecific binding precluded comparisons with sera from Shimojo mice by these assays. TSH binding inhibition and thyroid-stimulating antibody were undetectable in TSH receptor-DNA vaccinated mice. In Shimojo mice, TSH binding inhibition was positive in approximately 60%, and thyroid-stimulating antibodies were positive in hyperthyroid animals. Unlike the negative antibody data, splenocytes from TSH receptor-vaccinated (but not Shimojo) mice proliferated and produced the Th1 cytokine interferon-gamma in response to TSH receptor antigen. In conclusion, DNA vaccination is less effective at inducing TSH receptor antibodies than the Shimojo approach, but it permits the future characterization of TSH receptor-specific T cells generated without adjuvant.

A full biological response to autoantibodies in Graves' disease requires a disulfide-bonded loop in the thyrotropin receptor N terminus homologous to a laminin epidermal growth factor-like domain

We observed amino acid homology between the cysteine-rich N terminus of the thyrotropin receptor (TSHR) ectodomain and epidermal growth factor-like repeats in the laminin gamma1 chain. Thyroid-stimulating autoantibodies (TSAb), the cause of Graves' disease, interact with this region of the TSHR in a manner critically dependent on antigen conformation. We studied the role of the cluster of four cysteine (Cys) residues in this region of the TSHR on the functional response to TSAb in Graves' patients' sera. As a benchmark we also studied TSH binding and action. Removal in various permutations of the four cysteines at TSHR positions 24, 29, 31, and 41 (signal peptide residues are 1-21) revealed Cys(41) to be the key residue for receptor expression. Forced pairing of Cys(41) with any one of the three upstream Cys residues was necessary for trafficking to the cell surface of a TSHR with high affinity TSH binding similar to the wild-type receptor. However, for a full biological response to TSAb, forced pairing of Cys(41) with Cys(29) or with Cys(31), but not with Cys(24), retained functional activity comparable with the wild-type TSHR. These data suggest that an N-terminal disulfide-bonded loop between Cys(41) and Cys(29) or its close neighbor Cys(31) comprises, in part, the highly conformational epitope for TSAb at the critical N terminus of the TSHR. Amino acid homology, as well as cysteine pairing similar to the laminin gamma1 chain epidermal growth factor-like repeat 11, suggests conformational similarity between the two molecules and raises the possibility of molecular mimicry in the pathogenesis of Graves' disease.

Human monoclonal autoantibodies to B-cell epitopes outside the thyroid peroxidase autoantibody immunodominant region

Human autoantibodies to thyroid peroxidase (TPO) interact with a restricted or immunodominant region (IDR) on intact TPO. However, a smaller proportion of polyclonal serum TPO autoantibodies bind outside this region. To isolate monoclonal nonimmunodominant region (non-IDR) TPO autoantibodies, we screened a thyroid-derived immunoglobulin gene phage display library while "epitope masking" the TPO IDR with four human TPO monoclonal autoantibodies that define the IDR. Among 31 non-IDR autoantibodies obtained (expressed as Fab), 8 representatives were analyzed further based on their restriction digestion profiles. All are encoded by almost identical H chains (VH3 family), with extremely long D regions, paired with three different types of light chains. In contrast, IDR TPO Fab from the same patient utilize seven different heavy chains (VH1 and VH5 families) paired nonpromiscuously with different light chains. Use of VH5 genes has not been reported previously for TPO autoantibodies. Both non-IDR and IDR Fab bind specifically to TPO and not to other proteins. The non-IDR Fab affinities for TPO are moderately high (Kd 1-2 x 10(-9) M), somewhat lower than those for most IDR Fab (Kd 1-4 x 10(-10) M). The epitopes of the three types of non-IDR Fab overlap with each other, indicating a major role for their heavy chain in TPO binding. Most importantly, the epitopes of non-IDR Fab are recognized by patients' serum autoantibodies. In summary, we provide the first insight into the immunoglobulin genes, affinities and epitopes of human monoclonal autoantibodies that bind outside the TPO-immunodominant region.

Insight into thyrotropin receptor cleavage by engineering the single polypeptide chain luteinizing hormone receptor into a cleaving, two subunit receptor

To gain insight into the thyrotropin hormone (TSH) receptor (TSHR) cleavage, we sought to convert the noncleaving luteinizing hormone (LH) receptor (LHR) into a cleaved, two-subunit molecule. For this purpose, we generated a series of LHR mutants and chimeric LH-TSH receptors. Cleavage of mature, ligand binding receptors on the cell surface was determined by covalent 125I-labeled hCG crosslinking to intact, stably transfected mammalian cells. We first targeted a cluster of three N-linked glycans in the LHR (N295, N303, N317) in a region corresponding to the primary TSHR cleavage site, which has only one N-linked glycan. Elimination by mutagenesis of the most strategic N-linked glycan (LHR-N317Q) generated only a trace amount of LHR cleavage. Removal of the other N-linked glycans had no additive effect. A much greater degree of cleavage ( approximately 50%) was evident in a chimeric LH-TSHR in which the juxtamembrane segment of the LHR (domain E; amino acids 317-367) was replaced with the corresponding domain of the TSHR (residues 363-418). Similarly cleaving LHR were created using a much smaller component within this region, namely LHR-NET317-319 replaced with TSHR-GQE367-369, or by substitution of the same three amino-acid residues with AAA (LHR-NET317-319AAA). In summary, our data alter current concepts regarding TSHR cleavage by suggesting limited (not absent) amino-acid specificity in a region important for TSHR cleavage (GQE367-369). The data also support the concept of a separate and distinct downstream cleavage site 2 in the TSHR.

Autoimmune response to the thyroid in humans: thyroid peroxidase--the common autoantigenic denominator

Autoimmunity to thyroid peroxidase (TPO), manifest as high affinity IgG class autoantibodies, is the common denominator of human thyroid autoimmunity, encompassing patients with overt hyper- or hypothyroidism as well as euthyroid individuals with subclinical disease. The identification and cloning of TPO (the "thyroid microsomal antigen") provided the critical tool for analyzing B and T cell reactivity to this major thyroid autoantigen. In particular, the availability of immunoreactive TPO permitted the isolation of essentially the entire repertoire of human monoclonal antibodies, a feat unparalled in an organ-specific autoimmune disease. These recombinant autoantibodies (expressed as Fab) provide insight into the genes encoding their H and L chains as well as the conformational epitopes on TPO with which serum autoantibodies interact. Analyses of TPO autoantibody epitopic "fingerprints" indicate a lack of epitope spreading as well as a genetic basis for their inheritance. Limited data are available for the responses and cytokine profiles of T cells to endogenously processed TPO. Moreover, the role of thyroid cells in initiating the autoimmune response to TPO, and of B cells in expanding and/or modulating the response of sensitized T cells, has yet to be established. Finally, because autoantibody (and likely T cell) responses to TPO parallel those to TSH receptor and thyroglobulin, manipulation of T and B cell responses to TPO may provide the basis for the development of immunospecific therapy for autoimmune thyroid disease in general.

A prion-like shift between two conformational forms of a recombinant thyrotropin receptor A-subunit module: purification and stabilization using chemical chaperones of the form reactive with Graves' autoantibodies

A secreted recombinant TSH receptor (TSHR) ectodomain variant (TSHR-289) neutralizes TSHR autoantibodies in Graves' disease, but is heterogeneous in containing both immunologically active and inactive molecules and is also unstable. We have now purified each form of TSHR-289 using sequential affinity chromatography with a mouse mAb (3BD10) specific for the inactive form, and a mAb to C-terminal His residues that recognizes both forms. The immunological difference between active and inactive TSHR-289 was unrelated to primary amino acid sequence or carbohydrate content and was, therefore, attributable to its folded state. The epitopes for Graves' autoantibodies and 3BD10 overlap, and both are destroyed by denaturation. Therefore, reciprocal binding by autoantibodies and 3BD10 to conformational determinants involving the same TSHR segment suggests a prion-like shift between two folded states of the molecule. Despite purification, immunologically active TSHR-289 remained labile, as determined by loss of autoantibody, and gain of 3BD10, recognition. However, using chemical chaperones we have, for the first time, been able to stabilize purified TSHR antigen in immunologically intact form. In summary, purification of immunologically active and stable antigen in milligram quantities provides a powerful tool for future diagnostic and therapeutic studies in Graves' disease.

Reassessment of the location of the thyrotropin receptor 50 amino acid "insertion" provides evidence in favor of a second downstream cleavage site

Cleavage of thyrotropin receptors (TSHR) on the cell surface into disulfide-linked A and B subunits involves deletion of an intervening region that corresponds approximately to a 50 amino acid "insertion" in the TSHR relative to the noncleaving luteinizing hormone/choriogonadotropin receptor (LH/CGR). The location of this insertion is imprecise because of the relatively low homology between the two receptors in this region. We tested the hypothesis that the TSHR 50 amino acid insertion was further downstream than we previously concluded, a possibility that would relocate the crucial LH/CGR glycan at N291 relative to the position of the TSHR insertion, and that would mitigate against the 50 amino acid insertion playing a role in TSHR intramolecular cleavage. Thus, we transferred the LH/CGR glycan at amino acid 291 from downstream (N367) to upstream of the 50 amino acid insertion (N317) in the TSHR, leaving this insertion intact. TSHR cleavage persisted. Moreover, deletion of amino acid residues 320-366 in addition to the upstream N291 substitution (ALN317-319NET) also did not prevent cleavage. On the other hand, deletion of three contiguous downstream residues (GQE367-369) in the TSHR 50 amino acid insertion abolished receptor cleavage into subunits. In summary, the present data are consistent with our previous location of the TSHR 50 amino acid insertion and, therefore, do not undermine evidence for the involvement of this insertion in TSHR cleavage. In addition, the data regarding TSHR residues GQE367-369 (far downstream of cleavage site 1) support the controversial possibility of a secondary cleavage site downstream of the insertion.

Search for the autoantibody immunodominant region on thyroid peroxidase: epitopic footprinting with a human monoclonal autoantibody locates a facet on the native antigen containing a highly conformational epitope

Autoantibodies to thyroid peroxidase (TPO) are the hallmark of the humoral autoimmune response in human autoimmune thyroiditis (Hashimoto's thyroiditis). The majority of TPO autoantibodies in individual patients' sera interact with a restricted immunodominant region on TPO. Although this region can be mapped, previous studies have failed to localize its position on the TPO molecule. We, therefore, used a footprinting approach that can localize a highly conformational, discontinuous epitope on a very large molecule. Extensive biotinylation ( approximately 15 biotins/molecule protein) of lysine residues on the surface of purified, native TPO resulted in loss of multiple tryptic cleavage sites, as determined by analysis of tryptic polypeptide fragments on reverse-phase HPLC. TPO was then complexed with a monoclonal human autoantibody Fab (TR1.9) before biotinylation. After dissociation from TR1.9, TPO was recovered by gel filtration. A trypsin site, previously observed to be lost after TPO biotinylation, was restored when biotinylation was performed on the TPO-TR1.9 complex. The epitope-protected lysine (K) was present in a 30-aa TPO fragment that, by N-terminal sequencing, was found to be K713. Altered recognition by TR1.9 of a TPO-myeloperoxidase chimeric molecule involving this region supported the epitope protection data. In conclusion, we provide the first identification of an amino acid residue (K713) comprising part of an epitope within the TPO immunodominant region. This focal residue localizes the facet on the large, highly complex TPO molecule that contains the immunodominant region and provides the basis for rational guided mutagenesis studies to more fully characterize this region.

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

AKR/N mice injected with fibroblasts expressing MHC class II (RT4.15HP cells) and the TSH receptor (TSHR) develop antibodies similar to those in Graves' disease. We were unable to analyse the subclass of these antibodies because of unexpectedly high non-specific binding by ELISA or flow cytometry. The non-specific binding reflected generalized immune activation which occurred even when the fibroblasts did not express the TSHR. However, the IgG subclasses were determined for thyroid peroxidase (TPO) antibodies induced using TPO-expressing RT4.14HP cells and found to be IgG2a > IgG1. This Thl pattern is consistent with spontaneous secretion of interferon-gamma (but not IL-4 or IL-10) by splenocytes from injected mice. The Th1 bias was related to fibroblast injection because conventional immunization of the same mouse strain with purified TPO and adjuvant induced a Th2 response (IgG1 >> IgG2a). Further, untransfected fibroblasts themselves induced powerful, non-specific proliferative responses when used as antigen-presenting cells (APC) in vitro. Flow cytometry revealed that the RT4.15HP fibroblasts (and TSHR- and TPO-transfected derivatives) expressed B7-1. Unexpected constitutive expression of this key molecule may bypass the requirement for up-regulation of other costimulatory molecules involved in T cell stimulation. Our data support the concept that RT4.15HP fibroblasts present the TSHR (or TPO), at least for initiating the immune response. However, the accompanying generalized immune stimulation creates difficulties for analysis of TSHR-specific T and B lymphocytes. On the other hand, extension of the model to TPO, an easier antigen to study, will facilitate analysis of murine T cell responses likely to resemble those in human thyroid autoimmunity.

Presence of interleukin 4 or interleukin 10, but not both cytokines, in pancreatic tissue of two patients with recently diagnosed diabetes mellitus type I

Studies in the NOD mouse model suggest that development of diabetes mellitus type I can be prevented and established disease cured by deviation towards a Th2-type response. To obtain insight into whether this approach may be applicable to human disease, we investigated the Th1/Th2 cytokine balance in pancreatic tissue from two patients with diabetes of recent onset (Case 1, accidental death; Case 2, ketoacidosis). Using the polymerase chain reaction to amplify reverse-transcribed cDNA, signals for actin and CD36 confirmed mRNA integrity and the presence of T cells in pancreatic tissue from both patients and from a control. IFN-gamma cDNA was also amplified from all three tissues. However, IL-4 (but not IL-10) cDNA, was amplified from the pancreas of Case 1. Conversely, IL-10 (but not IL-4) cDNA was amplified from the the pancreas of Case 2. The control pancreas yielded specific signals for both IL-4 and IL-10. Our data extend the limited database on Th1 and Th2 cytokine expression in human pancreatic tissue from recently diagnosed diabetics. Moreover, together with previous observations, our findings raise the possibility that the lack of both IL-4 and IL-10 may be associated with the development of IDDM in humans.

Evidence that cleavage of the thyrotropin receptor involves a "molecular ruler" mechanism: deletion of amino acid residues 305-320 causes a spatial shift in cleavage site 1 independent of amino acid motif

Some TSH receptors (TSHR) on the cell surface cleave into A and B subunits. Cleavage at upstream Site 1 is followed by the proteolytic excision of an intervening C peptide region terminating at a downstream Site 2. Although present evidence suggests that Site 1 lies between amino acid residues 303 and 317, the mechanism and exact amino acid(s) involved in cleavage are unknown. Previous amino acid substitutions at Site 1 failed to abrogate cleavage. We, therefore, performed deletion mutations within this region. Cleavage of cell surface TSHR, detected by 125I-TSH cross-linking to intact cells, was not prevented by deletion of four individual segments within the Site 1 cleavage region (delta305-308, delta309-312, delta313-316, delta317-320). However, deletion of the entire region (delta305-320) reduced the extent of cleavage and shifted the cleavage site upstream of the glycan at amino acid residue N302. Elimination of this glycan (N302Q substitution) reversed the effect of deleting amino acid residues 305-320 on TSHR cleavage, suggesting that reduced cleavage at the new, upstream cleavage site was caused by steric hindrance by the glycan at N302. In summary, deletion, as opposed to mutagenesis, of the TSHR cleavage Site 1 region produces a spatial shift in TSHR cleavage Site 1 from downstream to upstream of the glycan at N302. These observations provide strong evidence that TSHR cleavage at this site does not occur at a particular amino acid motif and suggests that cleavage involves a "molecular ruler" mechanism involving cleavage at a fixed distance from a protease attachment site.

Elephantiasic pretibial myxedema: insight into and a hypothesis regarding the pathogenesis of the extrathyroidal manifestations of Graves' disease

The basis for the extrathyroidal manifestations of Graves' ophthalmopathy (GO) and dermopathy are not well understood. We describe immunohistochemical studies on the skin of a patient with an extreme, elephantiasic form of Graves' dermopathy that developed after periods of prolonged standing with dependent edema. Excision of part of the lesion with subsequent skin grafting from a normal donor site resulted in recurrence of the disease at the original site as well as in development of disease at the donor site. A murine monoclonal antibody reacted with the thyrotropin receptor (TSHR) or a cross-reacting protein in fibroblast-like cells in the patient's upper dermis and, surprisingly, with dermal cells from unaffected individuals. The patient's dermis containing lymphoid follicles comprising B cells and CD3+, CD4+ T cells, with few CD8+ T cells. CD21+ cells (most likely follicular dendritic cells) were also present in the dermis. Based on past and present observations, we raise an unifying hypothesis to explain the diverse extrathyroidal manifestations of Graves' disease and their apparent lack of association with TSHR autoantibodies. As opposed to the present concept that these phenomena relate to site-specific properties on preadipocytes or fibroblasts, we suggest that clinically evidence GO and dermopathy are primarily caused by local factors (particularly in the orbit) superimposed on a systemic, low-grade connective tissue inflammation.

Evidence for antigen presentation to sensitized T cells by thyroid peroxidase (TPO)-specific B cells in mice injected with fibroblasts co-expressing TPO and MHC class II

Injection of AKR/N mice with fibroblasts co-expressing MHC class II and TPO in the absence of adjuvant induces IgG-class TPO antibodies that resemble spontaneously arising human thyroid autoantibodies. We have used this model to examine the effect of iodide on TPO antibody induction as well as to analyse the interaction between T and B cells. Despite its importance as a major environmental factor in thyroid autoimmunity, variable iodide intake had no detectable effects on TPO antibody levels, lymphocytic infiltration of the thyroid or thyroid hormone levels. In terms of T cell responsiveness, splenocytes from TPO fibroblast-injected mice, but not from control mice, proliferated in response to TPO. Intriguingly, B cell-depleted splenocytes (mainly T cells without reduction of macrophages) proliferated in response to TPO only when co-cultured with irradiated autologous splenocytes from TPO fibroblast-injected mice but not from control mice. These data suggest that TPO-specific B cells are involved in antigen presentation to sensitized T cells and are supported by the ability of spleen cells from TPO cell-injected (but not control) mice to secrete TPO antibodies spontaneously in culture. In conclusion, we provide the first evidence for the presence of thyroid autoantigen-specific B cells and their ability to present their autoantigen to sensitized T cells in mice induced to develop TPO antibodies resembling autoantibodies in humans.

Subunit structure of thyrotropin receptors expressed on the cell surface

We studied cell surface thyrotropin receptor (TSHR) by biotinylating proteins on the surface of metabolically labeled, intact cells. In addition to TSHR cleaved into A and B subunits, mature single-chain receptors with complex carbohydrate were also present on the cell surface. A low A/B subunit ratio indicated partial shedding of extracellular A subunits from transmembrane B subunits. TSHR cleavage at upstream site 1 (within amino acid residues 305-316) would generate a B subunit of 51-52 kDa. However, only smaller B subunits (40-46 kDa) were detected, corresponding to N termini from residues approximately 370 (site 2) extending downstream to the region of B subunit insertion into the plasma membrane. The intervening C peptide region between sites 1 and 2 could not be purified from TSHR epitope-tagged (c-myc) within this region. However, the small proportion of B subunits recovered with a c-myc antibody were larger (45-52 kDa) than the majority of B subunits recovered with a C-terminal antibody. In conclusion, our study provides the first characterization of cell surface TSHR including their A and B subunits. Single-chain, mature TSHR do exist on the cell surface. The C peptide lost during intramolecular cleavage disintegrates rapidly following cleavage at upstream site 1 of the single-chain TSHR into A and B subunits. N-terminal disintegration of the B subunit pauses at site 2, but then progresses downstream to the vicinity of the plasma membrane, revealing a novel mechanism for A subunit shedding.

A direct binding assay for thyrotropin receptor autoantibodies

There is, at present, no assay in clinical use for the direct assay of autoantibody binding to the thyrotropin receptor (TSHR). We now describe a direct thyrotropin receptor autoantibody binding assay (DTAb) using a secreted form of the TSHR ectodomain (TSHR-289) without the need for antigen purification. The assay compensates for the low TSHR autoantibody concentration in serum by capturing a relatively large amount of patient immunoglobulin G (IgG) on high-capacity beads, a reversal of standard methods that typically first immobilize antigen. TSHR-289 captured by Graves' IgG was detected in a colorimetric reaction using a biotinylated murine monoclonal antibody to the poly-histidine tail engineered into the antigen. By this approach, sera from 11 normal individuals provided a mean optical density (OD) value of 0.20 +/- 0.08 SD (range 0.06-0.33). Of 38 sera from unselected patients with a history of Graves' disease (untreated and treated), 29 (76%) generated OD values > 0.37 (2 SD above the mean for the normal sera), the highest being OD 1.38. Surprisingly, 3 of 13 (23%) sera from TPO autoantibody-positive patients with Hashimoto's thyroiditis also provided values > 2 SD above the normal sera. The extent of direct autoantibody binding to the TSHR correlated closely with the thyrotropin binding inhibition (TBI) values (r = 0.881; p < 0.001). One serum was clearly positive in only the direct binding assay and another in only the TBI assay. The data obtained with the direct binding assay correlated less well with the thyroid-stimulating antibody (TSAb) assay (r = 0.582; p < 0.001). In summary, we describe a new direct DTAb assay that correlates more closely with the TBI than with the TSI assays. Future studies in a large series of clinically defined patients will be needed to evaluate the clinical utility of the DTAb assay.

On the functional importance of thyrotropin receptor intramolecular cleavage

We examined the relationship between TSH receptor (TSHR) cleavage into two subunits and ligand-independent, constitutive activity characteristic of this receptor. Because of homology to the thrombin receptor-tethered ligand, we focused initially on a region in the vicinity of the second, downstream cleavage site of the TSHR ectodomain. We introduced into the wild-type TSHR three mutations. One mutation, TSHR(GQE(367-369)NET) prevents cleavage at site 2. The other two mutations, ELK(369-371)T-Y (TSHR-E1a2) and NPQE(372-375)SAIF (TSHR-E1b), introduce major changes into the potential tethered ligand. Basal, steady state intracellular cAMP levels in cloned, stably transfected Chinese hamster ovary cells were expressed as a function of the number of receptors (cAMP/receptor). None of these three mutations decreased ligand-independent constitutive activity, thereby excluding the tethered ligand hypothesis as well as a requirement for cleavage at site 2 in this process. Turning to the more upstream site 1 in the TSHR ectodomain, we examined a receptor (TSHR-delta50AA) with deletion of a unique 50-amino acid insertion (residues 317-366) that appears to be involved in cleavage at this site. Constitutive cAMP production was similar to that of the wild-type TSHR. Finally, we studied a TSHR mutant that cleaves at neither site 1 (deletion of residues 317-366) nor site 2 (GQE(367-369)NET substitution) and, therefore, does not cleave into A and B subunits. Again, the basal, constitutive level of cAMP production was similar to that of the wild-type TSHR. In summary, contrary to the prevailing hypothesis based on several lines of evidence, TSHR cleavage into subunits is not associated with constitutive, ligand-independent activity.

Comparison of recombinant human thyrotropin receptors versus porcine thyrotropin receptors in the thyrotropin binding inhibition assay for thyrotropin receptor autoantibodies

Thyrotropin receptor autoantibodies (TRAb) are most commonly measured in a thyrotropin-binding inhibition (TBI) assay using solubilized porcine thyrotropin receptors (pTSHR). Recently, we reported modifications in recombinant human thyrotropin receptor (hTSHR) production and extraction that made substitution of this antigen for the pTSHR practical. We now report the first comparison of the behavior in a TBI assay of the recombinant, solubilized hTSHR with the pTSHR in a large series of clinically characterized patients with autoimmune thyroid disease. We studied 227 patients with Graves' disease (32 untreated patients, 156 patients receiving antithyroid medications, 24 patients in remission, 9 patients with recurrence of disease, and 6 thyroidectomized patients), as well as 32 patients with Hashimoto's thyroiditis and 28 normal individuals. In patients with untreated Graves' disease, 29 of 32 (90.6%) were TBI positive with either antigen, although two sera gave discrepant data in the two assay. Of the patients receiving antithyroid drugs, 94 of 156 (60.3%) were positive with the pTSHR and 106 of 156 (67.9%) were positive with the hTSHR TBI assay (p < 0.05%). In all other respects, however, there was no difference between the two TBI assays. Neither assay performed well in providing clinical guidance in the remission or relapse of disease. Of the 24 Graves' patients in remission, 75.0% and 79.2% were TBI negative with the hTSHR and pTSHR assays, respectively. The TBI assay at the time of relapse was even less informative; 6 of 9 (66.7%) being TBI negative in the pTSHR assay and 3/9 (33.3%) being negative in the hTSHR assay. In TBI assays with both species of TSHR, 3 of 32 hypothyroid patients with Hashimoto's thyroiditis were TBI positive. In summary, production of the recombinant hTSHR is now a practical reality and this antigen can clearly substitute at least as well for the pTSHR in the imperfect, although most commonly used, TBI assay. It is, therefore, likely that the hTSHR will supplant the pTSHR in this important assay. However, the use of the hTSHR rather than pTSHR does not appear to provide a major advantage, at least in terms of TBI assay sensitivity, specificity and predictive value.

Balance of Th1/Th2 cytokines in thyroid autoantibody synthesis in vitro

Thyroid-infiltrating lymphocytes spontaneously synthesize IgG class thyroid autoantibodies while blood lymphocytes require activation to produce the same autoantibodies. Surprisingly, the stimulus commonly used to induce autoantibodies by blood lymphocytes, Pokeweed mitogen (PWM), inhibits autoantibody synthesis by thyroid lymphocytes. To address this dichotomy, we investigated the Th1: Th2 cytokine balance in relation to thyroid peroxidase (TPO) autoantibody production in cultures of thyroid, lymph node and blood lymphocytes. The characteristic PWM effect on TPO autoantibody production by thyroid and blood lymphocytes (10 day cultures) was confirmed. Cytokine measurements in these cultures revealed that PWM increased IFN-gamma production by thyroid, lymph node and blood lymphocytes. However, PWM enhanced IL-4 levels in lymphocytes from blood and lymph node but not in thyroid lymphocytes. Moreover, the IL-4: IFN-gamma ratios in short- and long-term cultures were higher for PBMC and lymph node lymphocytes than for thyroid lymphocytes. In summary, PWM shifts the cytokine balance towards Th2 for blood lymphocytes and towards Th1 for thyroid lymphocytes. The shift towards Th1 in the target organ is associated with reduced autoantibody synthesis. Our observations suggest that "immune deviation" towards Th2 as a form of therapy in Graves' disease could project the patient from the frying pan into the fire.

Rarity of autoantibodies to a major autoantigen, thyroid peroxidase, that interact with denatured antigen or with epitopes outside the immunodominant region

The nature of the autoantibody repertoire to the dominant autoantigen in human autoimmune thyroid disease is controversial. There is evidence that autoantibodies to thyroid peroxidase (TPO) interact with overlapping conformational epitopes in an immunodominant region and binding to denatured (DN) protein is decreased. Contrary data demonstrate TPO autoantibody reactivity with DN-TPO or polypeptide fragments. However, none of the TPO-specific, human monoclonal autoantibodies isolated to date preferentially recognize denatured autoantigen. We therefore searched an immunoglobulin gene phage display library for human autoantibodies that bind TPO denatured by reduction and alkylation (DN-TPO). Thyroid-infiltrating B cells from a typical TPO autoantibody-positive patient were the source of mRNA for library construction. Surprisingly, the library enriched after panning on DN-TPO, as well as a panel of individual clones, preferentially bound native (N)-TPO. Of 13 clones selected using DN-TPO or N-TPO, 12 clones recognized the TPO immunodominant region. Moreover, regardless of selection with N-TPO or DN-TPO, their heavy and light chains were encoded by similar VDJ and Vkappa combinations. One clone (DN4), isolated using DN-TPO, did not interact with the TPO immunodominant region and its H chain derives from a different VH gene. Although DN4 binds specifically to TPO, its affinity is low, unlike the high affinities of other human TPO autoantibodies. In conclusion, human monoclonal autoantibodies that preferentially recognize denatured TPO could not be isolated from an immunoglobulin gene library despite selection with denatured protein. Our findings demonstrate the bias of the human B cell repertoire towards recognition of an immunodominant region on the conformationally intact form of a major thyroid autoantigen.

Cellular thyroid peroxidase (TPO), unlike purified TPO and adjuvant, induces antibodies in mice that resemble autoantibodies in human autoimmune thyroid disease

Autoantibodies to several protein antigens in human autoimmunity interact with a restricted range of epitopes, whereas diverse epitopes are recognized by antibodies induced in animals using antigen and adjuvant. To examine the basis for this difference, we compared the qualitative nature of antibodies developing in AKR/N mice injected with purified thyroid peroxidase (TPO) and adjuvant or with TPO expressed on major histocompatibility complex (MHC) class II+ fibroblasts. Mice injected with purified TPO had higher TPO antibody levels than TPO+/class II+ fibroblast-treated mice. Despite lower titers, recipients of TPO+/class II+ cells developed very high affinity antibodies (Kd = approximately 10(-10) M), comparable with those of human TPO autoantibodies and about 10-fold higher than those in purified TPO plus adjuvant-immunized mice. Moreover, more than 90% of TPO antibodies in TPO+/class II+ fibroblast-injected mice, compared with only approximately 50% in TPO plus adjuvant-immunized mice, were to the immunodominant region recognized by patients' autoantibodies. Consistent with this epitopic restriction, TPO+/class II+ fibroblast-injected mice had TPO antibody epitopic fingerprints similar to those of human autoantibodies. In conclusion, mice injected with TPO+/class II+ fibroblasts, but not those injected with purified TPO and adjuvant, develop antibodies closely resembling autoantibodies in human disease. These observations indicate that some animal models based on conventional immunization may not be representative of human diseases with a major humoral component.

The shed thyrotropin receptor is primarily a carboxyl terminal truncated form of the A subunit, not the entire A subunit

The TSH receptor (TSHR) sheds its A subunit, particularly when cells are cultured in serum-poor medium. This shed A subunit is reported to be smaller than the cell-associated receptor because of the loss of glycan without change in its polypeptide core. Contrary to previous deductions, we now find that the 'small' shed A subunit has lost a glycan moiety because of the proteolytic clipping of a small C-terminal fragment containing an Asn-linked glycan. Moreover, this lost peptide fragment contains cysteine residues likely involved in A subunit linkage to the membrane-associated B subunit. Progressive lowering of the serum concentration in culture medium accentuates the process. Therefore, 'small' A subunit shedding does not appear related to a physiological mechanism involving disulfide bond reduction. On the other hand, we detected, for the first time, shedding of a lesser amount of normal-sized, in addition to small, A subunits, especially by cells cultured in standard serum concentrations.

Evidence for genetic transmission of thyroid peroxidase autoantibody epitopic "fingerprints"

Autoimmune thyroid disease is characterized by the tendency to cluster in families and by IgG class autoantibodies to antigens such as thyroid peroxidase (TPO). The epitopes recognized by polyclonal serum autoantibodies can be quantitatively fingerprinted using four recombinant human TPO autoantibodies (expressed as Fab) that define A and B domain epitopes in an immunodominant region. To determine whether these fingerprints are genetically transmitted, we analyzed fingerprints of 63 members of 7 multiplex Old Order Amish families and 17 individuals from 4 Hashimoto thyroiditis families. Inhibition of serum autoantibody binding to [125I]TPO by the recombinant Fab was used to assess recognition of the TPO immunodominant region (4 Fab combined) and recognition of domain A or B (individual Fab). Complex segregation analysis was performed using a unified model (POINTER). For the 4 Fab combined inhibition phenotype, the no transmission model was rejected (chi2(4) = 20.67; P < 0.0032), and the most parsimonious model includes a major gene effect. More importantly, evidence for genetic transmission was obtained for the phenotype defined by the ratio of inhibition by subdomain Fab B1:B2. Thus, for this ratio (reflecting recognition of the B domain), the no transmission model was rejected chi2(4) = 63.59; P < 0.000008). Moreover, the polygenic hypothesis could be rejected, but not the major locus hypothesis, suggesting that major genes might be involved in familial transmission of this trait. In conclusion, our findings suggest that autoantibody recognition of the TPO immunodominant region and the TPO B domain is genetically transmitted. These data may open the way to the identification by candidate analysis or positional cloning of at least one gene responsible for the development of Hashimoto's thyroiditis.

A mouse monoclonal antibody to a thyrotropin receptor ectodomain variant provides insight into the exquisite antigenic conformational requirement, epitopes and in vivo concentration of human autoantibodies

We used the secreted TSH receptor (TSHR) ectodomain variant TSHR-289 (truncated at amino acid residue 289 with a 6-histidine tail) to investigate properties of TSHR autoantibodies in Graves' disease. Sequential concanavalin A and Ni-chelate chromatography extracted milligram quantities of TSHR-289 (approximately 20-40% purity) from the culture medium. Nanogram quantities of this material neutralized the TSH binding inhibitory activity in all 15 Graves' sera studied. We generated a mouse monoclonal antibody (mAb), 3BD10, to partially purified TSHR-289. Screening of a TSHR complementary DNA fragment expression library localized the 3BD10 epitope to 27 amino acids at the N-terminus of the TSHR, a cysteine-rich segment predicted to be highly conformational. 3BD10 preferentially recognized native, as opposed to reduced and denatured, TSHR-289, but did not interact with the TSH holoreceptor on the cell surface. Moreover, mAb 3BD10 could extract from culture medium TSHR-289 nonreactive with autoantibodies, but not the lesser amount (approximately 25%) of TSHR-289 molecules capable of neutralizing autoantibodies. Although the active form of TSHR-289 in culture medium was stable at ambient temperature, stability was reduced at 37 C, explaining the mixture of active and inactive molecules in medium harvested from cell cultures. In conclusion, studies involving a TSHR ectodomain variant indicate the exquisite conformational requirements of TSHR autoantibodies. Even under "native" conditions, only a minority of molecules in highly potent TSHR-289 preparations neutralize patients' autoantibodies. Therefore, Graves' disease is likely to be caused by even lower concentrations of autoantibodies than previously thought. Finally, reciprocally exclusive binding to TSHR-289 by human autoantibodies and a mouse mAb with a defined epitope suggests that the extreme N-terminus of the TSHR is important for autoantibody recognition.

Thyrotropin receptor cleavage at site 1 involves two discontinuous segments at each end of the unique 50-amino acid insertion

Among the glycoprotein hormone receptors, only the thyrotropin receptor (TSHR) cleaves (at two sites) into disulfide-linked A and B subunits. A 50-amino acid insertion unique to the TSHR ectodomain (residues 317-366) plays no role in ligand binding or signal transduction, but its deletion abrogates cleavage at Site 1, closely upstream of the insertion. We sought to define the region within the 50-amino acid tract involved in TSHR cleavage at Site 1. Mutation of small segments within this region previously failed to prevent cleavage at Site 1. We, therefore, divided the 50-amino acid insertion into quartiles and deleted each one individually (TSHR residues 317-327, 328-338, 339-350, and 351-362). As determined by covalent cross-linking of 125I-TSH to intact cells expressing the mutant receptors, none of these deletions prevented TSHR cleavage at Site 1. Neither did larger deletions of quartiles 1 + 2, 2 + 3, and 3 + 4. However, qualitative differences in the extent of receptor cleavage suggested that quartiles 1 and 4 were playing a greater role in cleavage at Site 1 than were the middle two quartiles. In support of this hypothesis, deletion of these two discontinuous segments almost completely eliminated TSHR cleavage at Site 1. In conclusion, intramolecular cleavage at Site 1 requires the presence of the N-terminal and C-terminal quartiles of the 50-amino acid insertion unique to the TSHR. Taken together with previous observations, our data suggest that this tract may provide a discontinuous binding site for a protease that clips the TSHR at Site 1.

Cytokine profiles of in vivo activated thyroid-infiltrating T cells cloned in the presence or absence of interleukin 4

We compared Th1 and Th2 cytokines secreted by randomly selected, intrathyroidal CD4+ T cell clones isolated from a patient with Graves' disease using IL-4 + IL-2 versus IL-2 alone. Prior to T cell isolation, PCR of cDNA from the intact thyroid tissue generated IL-4 and IL-10, but not IFN-gamma, products. As controls, IL-4, IFN-gamma and IL-10 cDNA was amplified from stimulated, but not unstimulated, PBMC. All 21 of the nine IL-2 clones and twelve IL-2 + IL-4 clones isolated from the thyroid tissue were CD4+. With the exception of one clone in the IL-2 group, all clones produced IL-10 on stimulation with anti-CD3 and phorbol-12-myristate 13-acetate (PMA) with similar mean values for both groups. The majority of clones in both groups also produced IFN-gamma and IL-4 after stimulation. However, the IL-4:IFN-gamma ratios were significantly higher in clones isolated using IL-2 + IL-4 than in those isolated with IL-2 alone. Furthermore, the distribution of Th1, Th0 and Th2 type clones, defined by their IL-4:IFN-gamma ratios, was also significantly different between those isolated using IL-2 alone and those isolated using IL-2 + IL-4. Of note, although Th0 clones predominated in both groups, Th1 clones were only obtained using IL-2 alone and Th2 clones were only obtained with IL-2 + IL-4. In conclusion, the presence of IL-4 together with IL-2 induces a shift away from a Th1-towards a Th2-response in T cells cloned from in vivo-activated thyroid-infiltrating lymphocytes. This difference in cytokine profile emphasizes that IL-4 is required for cloning T cells representative of a response involving both Th1 and Th2 cells, as occurs in autoimmune thyroid disease.

The greater glycan content of recombinant human thyroid peroxidase of mammalian than of insect cell origin facilitates purification to homogeneity of enzymatically protein remaining soluble at high concentration

Structural studies on thyroid peroxidase (TPO), a major thyroid autoantigen, require milligram amounts of pure protein. We found that the human TPO ectodomain (amino acid residues 1-848) generated in insect cells did not remain in solution at high concentrations after affinity purification. In contrast, the TPO ectodomain secreted by mammalian (Chinese hamster ovary) cells, although generated to a lesser extent (1 vs. 8 mg/liter), remained in solution at high concentration (10 mg/ml) after purification to homogeneity. This purified material was well recognized by TPO autoantibodies, but lacked enzymatic activity. We attempted to restore activity by culturing the Chinese hamster ovary cells in the presence of added heme. TPO enzymatic activity was clearly detected in conditioned medium from cells cultured in hematin and hemin, but not in protoporphyrin IX (all at 1 mg/liter). Heme prosthetic group incorporation into affinity-purified TPO was highest for hematin and hemin, but unchanged for protoporphyrin IX (OD 410/280 nm ratios of 0.25, 0.23, and 0.14, respectively). Enzymatic activity was now evident with hemin (mean +/- SE, 27.2 +/- 2.6; n = 3; guaiacol units/mg protein), hematin (24.1 +/- 1.6), and, to a lesser extent, protoporphyrin IX (3.6 +/- 0.2). Culturing cells in 20 mg/liter hematin, the maximum concentration tolerated, increased enzymatic activity even further (45.6 +/- 0.6 guaiacol units/mg protein). All purified TPO preparations were homogeneous on PAGE and of similar size (105 kDa). Enzymatic deglycosylation showed a complex carbohydrate contribution of 13 kDa (unlike the 2.3 kDa in insect cell TPO). In conclusion, this is the first report on the purification to homogeneity of recombinant human TPO of mammalian cell origin. Unlike TPO generated in insect cells, mammalian TPO remains soluble at high concentration, possibly because of its greater carbohydrate content. This enzymatically active, recombinant human TPO may be useful for future structural studies.

Relationship between autoantibody epitopic recognition and immunoglobulin gene usage

An immunodominant region recognized by serum autoantibodies has been defined on the autoantigen thyroid peroxidase (TPO) using recombinant human TPO-specific Fab or a panel of mouse MoAbs. We have now analysed the epitopic relationships between the four recombinant Fab that identify the A and B domains of the TPO immunodominant region and (i) the mouse TPO MoAb as well as (ii) nine new TPO-specific Fab isolated independently. Competition between mouse MoAbs and recombinant Fab for binding to 125I-TPO revealed three patterns. First, for MoAbs 15, 59, 64 and 18, TPO binding was virtually abolished (approximately 90%) by Fab which define the A domain of TPO, with less inhibition by B domain Fab. Second, for MoAbs 2, 9 and 47, the Fab competed much less for TPO binding, and, when detectable, inhibition was predominantly with B domain Fab (65-20%). Third, for MoAbs 53, 30, 1, 24 and 40, none of the Fab competed effectively for 125I-TPO binding. Thus, the epitopes for MoAbs 18, 59, 64 and 15 correspond to those of the A domain defined by the human Fab, and the epitopes for MoAbs 2, 9 and 47 correspond to those of the B domain. In the second part of the study, competition studies demonstrated that the epitopes of nine new Fab corresponded to those of the four Fab that define the immunodominant region. For four new Fab, TPO binding was inhibited to a greater extent by B- than by A-domain Fab (65-95% versus <50%). In contrast, for five new Fab the A-domain Fab were more effective inhibitors (approximately 90%) than the B-domain Fab. In addition, consistent with previous observations, all five new Fab with 02/012 kappa L chains, but none of the new Fab with non-O2/O121 chains, interacted with A-domain epitopes. In conclusion, we have established the epitopic relationships between recombinant human Fab and mouse MoAbs that define the TPO immunodominant region on TPO. Further, analysis of recombinant TPO Fab isolated from patients on three continents strengthens the paradigm of a relationship between autoantibody epitopic recognition and immunoglobulin gene usage.