Thyroxine-Binding Antibodies Inhibit T Cell Recognition of a Pathogenic Thyroglobulin Epitope1
http://www.100md.com
免疫学杂志 2005年第5期
Abstract
Thyroid hormone-binding (THB) Abs are frequently detected in autoimmune thyroid disorders but it is unknown whether they can exert immunoregulatory effects. We report that a THB mAb recognizing the 5' iodine atom of the outer phenolic ring of thyroxine (T4) can block T cell recognition of the pathogenic thyroglobulin (Tg) peptide (2549–2560) that contains T4 at aa position 2553 (T4(2553)). Following peptide binding to the MHC groove, the THB mAb inhibited activation of the Ak-restricted, T4(2553)-specific, mouse T cell hybridoma clone 3.47, which does not recognize other T4-containing epitopes or noniodinated peptide analogues. Addition of the same THB mAb to T4(2553)-pulsed splenocytes largely inhibited specific activation of T4(2553)-primed lymph node cells and significantly reduced their capacity to adoptively transfer thyroiditis to naive CBA/J mice. These data demonstrate that some THB Abs can block recognition of iodine-containing Tg epitopes by autoaggressive T cells and support the view that such Abs may influence the development or maintenance of thyroid disease.
Introduction
Thyroid hormone-binding (THB)4 Abs are frequently detected in patients with autoimmune thyroid disorders or in animal models of experimental autoimmune thyroiditis (EAT) (reviewed in Refs.1 and 2). THB Abs recognize thyroxine (T4) or triiodothyronine (T3) residues on distinct hormonogenic sites of thyroglobulin (Tg) and are being viewed as a subset of Tg-reactive autoantibodies. Development of THB Abs does not usually result in systemic hypothyroidism because of the ability of thyroid-pituitary feedback mechanisms to compensate for hormone neutralization (1, 2). However, THB Abs are clinically quite important because they interfere with the measurement of T3 and/or T4 and may occasionally hinder appropriate diagnosis and treatment (3, 4, 5). THB Abs raise important questions at the basic level as well, because the parameters regulating the immunogenicity of the T3 and T4 moieties, which are considered as haptens covalently linked to the Tg molecule, remain mostly unknown. Also, anti-Tg Abs may exert modulatory effects in EAT via their influence on the generation of pathogenic Tg peptides (6).
In Tg, the hormonogenic site at amino acid position 2553, at which a Tyr residue has been replaced by T4 via intramolecular coupling of two iodotyrosines, is one target of THB Abs (7). This site contributes to the formation of a B cell determinant but it is also encompassed within the Ak-binding, 12mer T cell epitope (2549–2560) that was originally discovered to cause EAT by Roitt’s group (8, 9). This 12mer peptide at aa position 2553 (T4(2553)) is rather unique among the thirteen so far mapped pathogenic Tg sequences (10), in that it is the only known Tg peptide that contains T4 and elicits strong proliferative as well as cytotoxic T cell responses (8, 9, 11, 12, 13). T4 is an integral part of this epitope because replacement of T4 by a Tyr residue abolishes T cell recognition (8). Furthermore, the four iodine atoms within the two-phenyl ring structure of T4 appear to sufficiently modify the peptide-MHC complex so as to elicit a distinct subset of thyroid-infiltrating T cells that recognize only the iodinated determinant (10, 12, 14, 15). These observations strongly supported the view that posttranslational modifications of Tg form a determinant that elicits both THB Abs and autoaggressive T cells.
In this study, we set out to examine whether THB Abs, represented by two mAbs, could influence T cell recognition of the T4(2553) peptide. This aim was made feasible by the generation of the 3.47 T cell hybridoma clone that was serendipitously found to recognize the T4(2553) epitope (6). In particular, we wanted to investigate whether THB Abs could detect the T4(2553) peptide even after its binding to the Ak molecule and block its MHC-restricted presentation to T cells. Such evidence would provide support for an immunoregulatory role of THB Abs in the disease process.
Materials and Methods
Ag presentation
The preparation of F-moc-L-thyronine, F-moc-L-thyroxine, and the synthesis of thyronine (T0)- or T4-containing peptides were as previously described (12). The sequences of these 12mer peptides, conserved between the murine and human species, were: Tg(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12) NH2-NIFE T4 QVDAQPL-NH2, T4(5); Tg(2549–2560) NH2-STDD T0/T4 ASFSRAL-NH2, T0/T4(2553); and Tg(2559–2570) NH2-ALENATRD T4 FII-NH2, T4(2567). The composition and relative purity of each peptide were verified by time-of-flight mass analysis on a Bio-Ion 20 Analyzer (Applied Biosystems) and, in some cases, by amino acid analysis of peptide hydrolysates (6 N HCl, 22 h, 121°C) on a Beckman 6300 amino acid analyzer (Beckman Coulter). Purity was determined to be >90%. The mouse Tg peptides (2494–2510) Ac-GLINRAKAVKQFEESQG-NH2 and (2694–2711) Ac-CSFWSKYIQTLKDADGAK-NH2, as well as the mouse lysozyme peptide (46–62), were synthesized at the Alberta Peptide Institute (Edmonton, Alberta, Canada) and purified as previously described (16, 17). Mouse Tg was purified from homogenates of thyroid glands of outbred ICR mice (Harlan Bioproducts for Science) by gel filtration over a Sepharose CL-4B column as described (17). Tg was dissolved in PBS, filter-sterilized, and stored at –20°C until use. Free T4 (Sigma-Aldrich) was dissolved in 99:1 of absolute methanol and 30% ammonium hydroxide at a stock concentration of 5 mg/ml.
Cell lines and T cell activation assay
The T cell hybridoma clone 3.47 was generated following a modified method of Perkins et al. (18). Briefly, lymph node cells (LNC) from CBA/J mice (The Jackson Laboratory), immunized with mouse Tg emulsified in CFA (with Mycobacterium butyricum; Difco), were further stimulated in vitro with the same Ag and fused with BW5147 –– cells (a kind gift of P. Marrack, National Jewish Medical and Research Center, Denver, CO) using polyethylene glycol 1500 (Boehringer Mannheim). Screening and cloning were performed as previously described (16). Expression of TCR and CD4 molecules on the 3.47 cells was verified by direct FACS analysis (data not shown). The TA3 cell line, produced by fusion of B cells from CAF1 mice with the M12.4.1 BALB/c B lymphoma (16, 19), was kindly provided by L. Glimcher (Harvard University, Cambridge, MA). The IL-2-dependent CTLL-2 line was purchased from American Type Culture Collection. All cells were cultured in DMEM (Invitrogen Life Technologies) supplemented with 10% FBS (Harlan Bioproducts
for Science), 10 mM HEPES buffer, 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (all from Invitrogen Life Technologies), and 5 x 10–5 M 2-ME (Sigma-Aldrich). CTLL-2 cells were grown in culture media supplemented with 10% supernatant from Con A-activated rat spleen cells.
In the activation assay, 105 3.47 T cells and 105 TA3 cells or macrophages were cultured in flat-bottom wells of 96-well plates, with or without various Ags, in a total volume of 200 μl. Following a 24 h incubation, 100 μl of supernatant was removed and stored at –20°C for assessment of IL-2 release, as measured by the proliferation of CTLL-2 line using [3H]thymidine (DuPont). Peritoneal macrophages were harvested from mice 4 days after i.p. injection of 2.5 ml of 3% thioglycolate per mouse.
Purification of mAbs and ELISA
The 55H8 and 91A1 mAbs (both IgG1) were derived from splenic B cells of BALB/c mice immunized with human Tg, following fusion with the mouse myeloma NSO. The mAbs were purified from ascites fluid by affinity chromatography on protein G-Sepharose 4 Fast Flow columns (Pharmacia). The purified material was concentrated to 1–2 mg/ml in PBS, filter-sterilized and stored in 50% glycerol at –20°C until use. Isotyping was performed with the ISO-2 kit (Sigma-Aldrich). The hybridomas specific for Ak (10-3.6.2, IgG2a) or Ek (14-4-4S, IgG2a) were purchased from American Type Culture Collection and the respective mAbs were purified from culture supernatants by affinity chromatography, as described earlier. ELISA was performed by coating microtiter polyvinyl chloride plates (Dynatech Laboratories) with 1 μg/well of Tg or Tg peptides and by using an alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma-Aldrich) as the second Ab. Absorbance at 405 nm was measured using a microplate reader (Vmax; Molecular Devices).
LNC proliferation assay and assessment of EAT
Female, 6- to 8-wk-old CBA/J mice were s.c. immunized with 100 μg of T4(2553) in 100 μl of CFA emulsion and 9–12 days later, the draining LNC were restimulated in a recall assay in vitro. All experimental procedures were reviewed and approved by the Animal Care Committee at Memorial University of Newfoundland. In blocking assays, mitomycin C-treated splenic cells (107 cells/ml) were used as APC following incubation with 10 μg/ml T4(2553) peptide for 6 h at 37°C. After washing away the free peptide, the T4(2553)-pulsed splenic cells (107 cells/ml) were cultured with or without 60 μg/ml blocking mAbs for 1 h. The APC were then washed and added (4–6 x 105 cells/well) to an equal number of T4(2553)-primed LNC for 96 h in a total volume of 200 μl of DMEM supplemented with 1% normal mouse serum. During the last 18 h of culture, 1 μCi of [3H]thymidine was added to each well. Cell harvesting and the counting of incorporated radioactivity were done as previously described (20). Adoptive transfer of EAT by peptide-primed LNC was assessed 13 days posttransfer. Effector cell activation conditions were slightly modified from conditions earlier described: mitomycin C-treated splenic APC (2 x 107/ml) were pulsed for 4 h with peptide (30 μg/ml), washed, incubated with 100 μg/ml blocking mAb for 2 h, washed, and subsequently cultured with peptide-primed LNC at 2:1 effector to APC ratio. Four days later, 107 live cells, separated by Ficoll centrifugation, were i.p. injected into each recipient. Fixation, embedding, and sectioning of the thyroids were performed as previously described (21). The mononuclear cell infiltration index was scored as follows: 0, no infiltration; 1, interstitial accumulation of cells between two or three follicles; 2, one or two foci of cells at least the size of one follicle; 3, extensive infiltration 10–40% of total area; 4, extensive infiltration 40–80% of total area; and 5, extensive infiltration >80% of total area. Each mouse was assigned the average infiltration index observed from both thyroid lobes (at least 10 sections per lobe were read and the maximum infiltration index per lobe was scored). The results were analyzed by the nonparametric Wilcoxon rank-sum test.
Results
The Ak-restricted T cell hybridoma clone 3.47 recognizes a Tg peptide containing a unique hormonogenic site
The 3.47 T cell hybridoma clone was initially selected for its reactivity to mouse Tg-pulsed TA3 cells but upon continuous culture it lost responsiveness to intact mouse Tg (data not shown). However, it was serendipitously found that 3.47 was strongly activated by the conserved Tg peptide T4(2553), whereas it was unreactive to two other Tg peptides (2494–2711) (Fig. 1A). The 3.47 clone is Ak-restricted because activation was blocked by an Ak-specific but not Ek-specific mAb (Fig. 1B). To test whether the presence of T4 within any peptide backbone is sufficient to activate the 3.47 clone, we examined the activation properties of two additional T4-containing 12mer peptides of Tg: the T4(5) (aa 1-12) and T4(2567) (aa 2559-2570). As shown in Fig. 1C, TA3 APC pulsed with those peptides or free T4 could not stimulate 3.47 to secrete IL-2. Lack of activation was not due to lack of peptide binding to Ak because the T4(2567) peptide inhibited 3.47 activation by T4(2553) peptide in a competitive inhibition assay to t
he same extent as the mouse lysozyme peptide (46–62), which is known (19) to bind strongly to Ak (Fig. 1D). The T4(5) peptide and free T4 were not inhibitors suggesting very weak or no binding to Ak. The pathogenic 9mer Tg peptide (2496–2504) that binds to Ek (10) also showed no inhibitory effect (data not shown). These results confirmed that the amino acid residues surrounding T4 were essential for Ak-binding and/or optimal TCR contact that would lead to the activation of the 3.47 clone.
FIGURE 1. The 3.47 T cell hybridoma clone specifically recognizes the Tg peptide T4(2553) and is Ak-restricted. TA3 cells were used as APC. Points indicate mean of triplicate wells and denote radioactive thymidine uptake (cpm) by the IL-2-dependent CTLL cells. A, IL-2 secretion following activation of the 3.47 clone by the Tg peptides is shown. B, 3.47 T cell activation is shown in the presence of increasing concentrations of mAbs specific for Ak or Ek Ags and 150 nM of T4(2553) peptide. Control cultures, without mAbs, yielded a mean of 13,120 cpm. C, Activation of 3.47 T cells by T4(2553) but not other T4-containing peptides of Tg. D, Competitive inhibition of 3.47 T cell activation in the presence of 37.5 nM of T4(2553) and increasing amounts of inhibitor peptides. Control culture cpm mean without inhibitory peptide is 14,830; without Ag is 530.
The 3.47 T cell clone recognizes an iodinated determinant
To assess whether iodine atoms contributed to the formation of the epitope recognized by the 3.47 clone, we compared the activation profiles obtained following recognition of the T4(2553) peptide vs the T0(2553) peptide analog that lacks the four iodine atoms within the T4 structure. It was found that 3.47 cells were activated by TA3 pulsed with T4(2553), but were unresponsive to TA3 pulsed with the noniodinated analog (Fig. 2A). The T0(2553) peptide could almost completely block the activation of 3.47 by T4(2553) (Fig. 2B) suggesting strong binding of the T0-containing analog to the Ak molecule. These results clearly demonstrate that the 3.47 clone recognizes a determinant modified by iodine atoms within the two-phenolic ring structure of T4, and also suggest that the iodine atoms do not critically influence the binding of the T4(2553) peptide to the Ak molecule.
FIGURE 2. The 3.47 clone recognizes a Tg determinant modified by iodine atoms. TA3 cells were used as APC. A, Lack of 3.47 T cell activation by the thyronine-containing analog peptide T0(2553). B, Competitive inhibition of 3.47 activation in the presence of 25 nM of T4(2553) and increasing amounts of the competitor analog peptide T0(2553). Data denote cpm mean of triplicate wells and are representative of three separate experiments. Control culture cpm mean values are the same as in Fig. 1D.
Iodine atoms on the phenolic ring of T4 contribute to the formation of determinants recognized by the T4-reactive mAbs 55H8 and 91A1
We have previously reported the generation of a Tg-specific mAb 55H8 that binds to the T4(2553) peptide (12). We proceeded to examine whether 55H8 is T4-reactive by monitoring its binding against other T4-containing peptides. It was found that 55H8 reacted similarly against T4(5) and T4(2567) whereas it did not bind to the T0(2553) analog (Fig. 3A). These data suggested that the 55H8 mAb is T4-specific and that the iodine atoms on the phenolic ring of T4 are essential for such recognition. This was further confirmed by a competitive inhibition ELISA in which free T4, or T4-containing peptides but not T0(2553), completely inhibited the binding of 55H8 to mouse Tg (Fig. 3C).
FIGURE 3. A and B, Reactivity of 55H8 and 91A1 mAbs to the indicated Tg peptides encoding hormonogenic regions. ELISA plates were coated at 1 μg of peptide per well. C and D, Competitive inhibition of mAb binding to mouse Tg using 1.0 μg/ml 55H8 (C) or 5.0 μg/ml 91A1 (D) in the presence of increasing amounts of free T4 or Tg peptide competitors. Points indicate mean of triplicate wells; SD values were <10% of the mean. OD values in the absence of inhibitory peptides were 1.48 (C) and 0.947 (D).
When we compared the reactivity profile of 55H8 with that of a second T4-specific mAb 91A1, we obtained an overall similar pattern of response with some differences. As shown in Fig. 3B, mAb 91A1 reacted with all the three T4-containing peptides but not with T0(2553) peptide on a solid phase despite a weaker binding to peptides T4(5) and T4(2553). Similarly, in competitive inhibition, all of the T4-containing peptides and free T4 molecule could inhibit binding of 91A1 to mouse Tg (Fig. 3D), although such inhibition was detectable at higher concentrations of peptides and T4 molecule than those for 55H8. This was partly due to the fact that a 5-fold higher concentration (5 μg/ml) of 91A1 was used in the inhibition assay to achieve a maximum OD value of 0.947 in the absence of inhibitor. Thus, the two T4-specific mAbs, 55H8 and 91A1, exhibited an overall similar reactivity profile to T4-containing peptides, but the 91A1 seemed to recognize T4 with a different affinity and/or such recognition was more influenced by neighboring amino acids than that of 55H8.
The T4-reactive mAbs 55H8 and 91A1 differ in their fine specificity
The T4 molecule contains four iodine atoms, two at positions 3 and 5 of the inner phenolic ring and another two at positions 3' and 5' of the outer ring. The hormone T3 lacks the iodine atom at the 5' position. To examine the fine specificities of 55H8 and 91A1, we conjugated rabbit serum albumin (RSA) with T4 or T3 and tested the binding of these mAbs to the conjugates by ELISA. The mAb 55H8 bound strongly to T4-RSA but not to T3-RSA (Fig. 4A), whereas 91A1 bound equally well to both conjugates (Fig. 4, A and B). Both mAbs bound equally well to mouse Tg. These data suggested that the iodine atom at the 5' position of the outer ring is an integral component of the determinant recognized only by 55H8, and not by the 91A1 mAb.
FIGURE 4. The mAbs 55H8 (A) and 91A1 (B) differ in their fine specificity. ELISA reactivities were assessed using Tg, or conjugates of T3 or T4 with RSA (1 μg/well) in the presence of increasing concentrations of mAbs shown.
CR recognition of the T4(2553)-Ak complex is blocked by the 55H8 mAb
It is well known that peptide-specific Abs do not commonly recognize their ligand within peptide-MHC complexes due to sequestration of the peptide within the MHC groove. Due to the bulkiness of the T4 moiety within T4(2553), and the requirement for T4 in the TCR triggering of the 3.47 clone, we envisaged that the T4-reactive 55H8 mAb might bind to the peptide-MHC complex and block peptide recognition by this T cell clone. To test this, TA3 cells (Fig. 5A) or peritoneal thioglycollate-stimulated macrophages (Fig. 5B) were pulsed with 2 μg/ml T4(2553) for 6 h, washed, and subsequently cultured with 3.47 T cells in the presence of increasing amounts of 55H8 and 91A1 mAbs or a control Tg-specific mAb of the same IgG subclass (3B3). Activation of 3.47 was strongly inhibited by 55H8, whereas the 91A1 and the control mAb 3B3, used at equimolar amounts, had no effect. A similar blocking effect of 55H8 was seen following fixation of T4(2553)-pulsed TA3 APC with glutaraldehyde (Fig. 5C). These data suggested that some T4-specific Abs, exemplified by 55H8, can block recognition of the pathogenic T4(2553) epitope by T cells.
FIGURE 5. The T4-specific mAb 55H8 blocks recognition of the T4(2553) peptide by the 3.47 clone. The APC line TA3 (A) or thioglycollate-stimulated peritoneal macrophage from CBA mouse (B) were pulsed with T4(2553) (2 μg/ml) for 6 h, washed and cultured with 3.47 cells in the presence of increasing concentration of the mAbs as shown. C, TA3 cells were pulsed with T4(2553) as in A, and were then fixed with 0.05% glutaraldehyde (F) for 30 s, or left untreated (N). Inhibition of the 3.47 cell activation was monitored in the presence of increasing concentrations of the mAbs shown. Points indicate the cpm mean of triplicate wells. Mean values of control wells in the absence of mAbs were (cpm): 14,097 (A); 31,555 (B), and 23,450 (C). Background values in the absence of T4(2553) peptide ranged from 80 to 400 cpm. Similar results were obtained in three separate experiments.
55H8 inhibits activation of T4(2553)-primed LNC and adoptive transfer of EAT
The potential of the 55H8 mAb to inhibit T cell activation was subsequently tested in a proliferation assay using T4(2553)-primed LNC from CBA/J mice. Mitomycin C-treated syngeneic splenocytes were pulsed with T4(2553), washed, and cultured with T4(2553)-primed LNC in the presence of 55H8 or 91A1. As shown in Fig. 6A, T4(2553)-pulsed, but not unpulsed, splenocytes induced a strong proliferative LNC response that was significantly inhibited only by 55H8 and not by the 91A1 or 3B3 control mAbs. Subsequent adoptive transfer of these cell populations into syngeneic CBA mice showed that effector cells cultured with the 55H8-treated, T4(2553)-pulsed splenocytes induced EAT with an average thyroidal lymphocyte infiltration index of 1.42 ± 0.58 (Fig. 6B). In contrast, the T4(2553)-primed LNC, cultured with the nontreated or the 91A1-treated APC, elicited a significantly higher lymphocytic infiltration of the thyroid with an average infiltration index of 2.42 ± 0.58 (p < 0.05). The large inhibition of the T4(2553)-specific T cell activation by 55H8 and the reduction of the pathogenic potential of such cells in vivo strongly suggested an immunoregulatory role for T4-reactive Abs following the formation of hormonogenic peptide: MHC complexes resulting from the processing of Tg or Tg fragments.
FIGURE 6. Activation of T4(2553)-specific pathogenic T cells is blocked by the 55H8 mAb. A, T4(2553)-primed LNC were cultured with mitomycin C-treated splenocytes that were pulsed with 10 μg/ml T4(2553) peptide for 6 h, washed and subsequently blocked with 60 μg/ml mAb 55H8, 91A1, or 3B3 () or no mAb () for 1 h. LNC proliferation was measured by adding [3H]thymidine during the last 18 h of a 4-day culture. Splenocytes not pulsed with peptide were used as controls (). B, Adoptive transfer of EAT by 107 T4(2553)-primed LNC following their in vitro restimulation with mitomycin C-treated, T4(2553)-pulsed syngeneic (CBA/J) splenocytes in the presence of 100 μg/ml 55H8 and 91A1 mAbs. EAT severity was scored on day 13 after transfer and the data were analyzed by the nonparametric Wilcoxon rank-sum test.
Discussion
Among the six hormonogenic sites described in human Tg, corresponding to tyrosyls 5, 685, 1290, 2553, 2567, and 2746 (22), the one at position 2553 has attracted special attention because peptides encompassing this site are recognized by human autoantibodies (7) and stimulate the adoptive transfer of murine EAT by T cells primed in vivo with the same peptide or mouse Tg (7, 8, 9, 10, 11, 12, 13, 14). In this report, we provide evidence that the mAb 55H8, which requires the 5' iodine atom of the outer phenolic ring of T4 for binding to Tg, markedly inhibits the activation of T cells responding to the pathogenic T4(2553) peptide. This suggests that the B cell and T cell responses targeting this site can be antagonistic at the effector stage.
The 55H8 and 91A1 mAbs, binding to T4 and T3, respectively, represent only certain subsets of all THB Abs that include Abs reactive to both T3 and T4 or Abs binding to T3, T4, and reverse T3 (1). Both mAbs were generated using human Tg as an immunogen, but it is unknown whether T4 at position 2553 contributed solely to their induction, because there are at least two hapten-like T4 residues exposed on the surface of the Tg "carrier" (23, 24). Flanking residues can contribute to the antigenicity of hormonogenic sites (7) and the inhibition data suggested differences between 55H8 and 91A1 binding to Tg that could be accounted for by steric effects of T4, surrounding residues and/or by binding affinity differences between the two mAbs. Conversely, the binding of 55H8 and 91A1 to Tg was completely inhibited by free T4 in our study, suggesting that the B cell determinants recognized by these mAbs are localized within the thyroxyl residue. In contrast, the 3.47 clone was not activated by free T4, or T4-containing
peptides other than T4(2553), in agreement with previous studies with hormonogenic peptides, emphasizing the role of amino acid residues surrounding T4 for MHC binding or TCR contact (12, 15).
THB Abs are detected both in human autoimmune thyroid disorders and in models of spontaneous thyroiditis (1, 2, 25, 26, 27). However, they usually exert no appreciable effects on metabolic status because of: 1) their relatively low binding affinity as compared with physiologic blood proteins such as thyroxine-binding globulin; and 2) the capacity of the thyroid gland to compensate for hormone neutralization via the thyroid-pituitary feedback mechanism (1, 2). In addition, hormonogenic sites on Tg may constitute minor antigenic determinants serologically because THB Abs are found only in a small portion of patients with anti-Tg Abs (2). The relative autonomy of THB Ab production has also been reported in mice, in which the Ir genes controlling the response to human Tg vs thyroid hormone seem to be different (2, 28). Although the molecular basis of these observations remains obscure it is becoming increasingly clear that THB Abs, once formed, can exert immunoregulatory influences. For example, we have previously shown that 55H8 when bound to Tg suppresses the generation of T4(2553) in APC processing the immune complex (6). We have hypothesized that this suppressive effect could be due to peptide sequestration and prevention from loading on Ak molecules, as predicted by the T cell to B cell reciprocity hypothesis (29, 30). In this report, 55H8 has been shown to block recognition of T4(2553) after the peptide is bound in the MHC groove. This steric hindrance effect can be explained by a protrusion of the thyroxyl side chain above the groove to allow binding of 55H8 and the strict dependence of 3.47 activation on the presence of iodine atoms within the peptide-MHC complex because this clone does not react to the T0(2553) analog. Abs accessing peptides bound to MHC class I (31) or class II (32, 33, 34) molecules have been previously reported.
Our results are in agreement with previous studies showing that LNC from CBA/J mice primed with the T4(2553) peptide could not be cross-stimulated in vitro with the T0(2553) analog and this regimen failed to generate effector cells that could adoptively transfer EAT (12). This suggested that the bulky iodine atoms of T4 (atomic radius of iodine is 133 pm) modify the peptide-MHC complex sufficiently so as to elicit a distinct subset of T cells, probably represented by the 3.47 clone described in this report, that recognize only the iodinated determinant. Also, our results are in partial agreement with those of Dawe et al. (15) who showed that the activation of a hybridoma clone responding to T4(2553) but not to T0(2553) could be blocked by an anti-T4 mAb. However, in that study, the iodine atoms within the T4 moiety were deemed necessary for peptide binding to MHC because the T0(2553) analog could not block activation of the T4(2553)-specific clone. Our data (Fig. 2) do not support this interpretation because T0(2553) can almost completely block the activation of the 3.47 clone suggesting that the iodine atoms are critical in forming a TCR-contacting residue. The mild immunopathogenicity of the T0(2553) peptide in CBA/J hosts (12) also strongly suggests that the iodine atoms of the thyroxyl side chain are not necessary for peptide-MHC binding. Removal of iodine from the tryptic human Tg fragment (2513–1713) has been reported to convert epitope(s), encompassed within this fragment, from immunogenic to tolerogenic (35), emphasizing the immunomodulatory potential of iodine incorporated in Tg.
Our current and previous (6) findings highlight the interplay between humoral and cellular immunity in autoimmune thyroiditis. They demonstrate that THB Abs, represented by 55H8, can prevent the activation of T cells responding to the pathogenic T4(2553) epitope, either via peptide sequestration (6) or blockade of peptide-MHC complexes. Abs specific for complexes of As and an encephalitogenic determinant of myelin basic protein have analogously been reported to block EAE (32). The T4(2553) can activate T cells following priming with intact Tg in vivo but it has been classified as subdominant because T4(2553)-primed T cells seldom respond to intact Tg processed in vitro (8, 9, 10, 12). Nonetheless, tyrosyl 2553 is one of the first sites to be iodinated in Tg (22). Given that enhanced iodine incorporation increases the immunogenicity of Tg(36, 37, 38), it is plausible that the T4(2553) peptide becomes an early target of autoaggressive T cells in autoimmune thyroiditis, following the subsequent coupling of a "donor" iodotyrosyl to form the hormonogenic site at position 2553. TBH Abs blocking the activation or effector function of such cells may help in the prevention or amelioration of the disease.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Dr. P. Marrack for the BW5147 – – cells and Dr. L. Glimcher for the TA3 cell line.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported by Grants from the Canadian Institutes for Health Research (to G.C.) and the National Institutes of Health Grant DK 45960 (to Y.M.K.). Y.D.D. is a recipient of a Memorial University PhD Fellowship.
2 Current address: Division of Immunoregulation, Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121.
3 Address correspondence and reprint requests to Dr. George Carayanniotis, Faculty of Medicine, Memorial University of Newfoundland, 300 Prince Philip Drive, St. John’s, Newfoundland A1B 3V6, Canada. E-mail address: gcarayan{at}mun.ca
4 Abbreviations used in this paper: THB, thyroid hormone binding; Tg, thyroglobulin; EAT, experimental autoimmune thyroiditis; T0, thyronine; T3, triiodothyronine; T4, thyroxine; RSA, rabbit serum albumin; LNC, lymph node cell.
Received for publication May 6, 2004. Accepted for publication December 23, 2004.
References
Benvenga, S., F. Trimarchi, J. Robbins. 1987. Circulating thyroid hormone autoantibodies. J. Endocrinol. Invest. 10:605
Sakata, S.. 1994. Autoimmunity against thyroid hormones. Crit. Rev. Immunol. 14:157.
Despres, N., A. M. Grant. 1998. Antibody interference in thyroid assays: a potential for clinical misinformation. Clin. Chem. 44:440.
Sapin, R., F. Gasser, A. Boehn, M. Rondeau. 1995. Spuriously high concentration of serum free thyroxine due to anti-triiodothyronine antibodies. Clin. Chem. 41:117.
Shimon, I., C. Pariente, J. Shlomo-David, Z. Grossman, J. Sack. 2003. Transient elevation of triiodothyronine caused by triiodothyronine autoantibody associated with acute Epstein-Barr-virus infection. Thyroid 13:211.
Dai, Y., K. A. Carayanniotis, P. Eliades, P. Lymberi, P. Shepherd, Y. Kong, G. Carayanniotis. 1999. Enhancing or suppressive effects of antibodies on processing of a pathogenic T cell epitope in thyroglobulin. J. Immunol. 162:6987.
Sakata, S., T. Ogawa, Y. Kimata, H. Takuno, H. Maekawa, M. Matsuda, O. Tarutani, K. Okuda. 1991. Synthesis and examination of antigenicity of four hormonogenic sites and two non-hormonogenic sites of human thyroglobulin. Mol. Cell. Endocrinol. 79:93
Champion, B. R., K. R. Page, N. Parish, D. C. Rayner, K. Dawe, G. Biswas-Hughes, A. Cooke, M. Geysen, I. M. Roitt. 1991. Identification of a thyroxine-containing self-epitope of thyroglobulin which triggers thyroid autoreactive T cells. J. Exp. Med. 174:363.[
Hutchings, P. R., A. Cooke, K. Dawe, B. R. Champion, M. Geysen, R. Valerio, I. M. Roitt. 1992. A thyroxine-containing peptide can induce murine experimental autoimmune thyroiditis. J. Exp. Med. 175:869.
Carayanniotis, G.. 2003. The cryptic self in thyroid autoimmunity: the paradigm of thyroglobulin. Autoimmunity 36:423.
Carayanniotis, G., Y. C. Kong. 2000. Pathogenic thyroglobulin peptides as model antigens: insights on the induction and maintenance of autoimmune thyroiditis. Int. Rev. Immunol. 19:557.
Dawe, K. I., P. R. Hutchings, M. Geysen, B. R. Champion, A. Cooke, I. M. Roitt. 1996. Unique role of thyroxine in T cell recognition of a pathogenic peptide in experimental autoimmune thyroiditis. Eur. J. Immunol. 26:768.
Carayanniotis, G., E. Chronopoulou, V. P. Rao. 1994. Distinct genetic pattern of mouse susceptibility to thyroiditis induced by a novel thyroglobulin peptide. Immunogenetics 39:21.
Chronopoulou, E., G. Carayanniotis. 1992. Identification of a thyroiditogenic sequence within the thyroglobulin molecule. J. Immunol. 149:1039.
Perkins, D. L., G. Berriz, T. Kamradt, J. A. Smith, M. L. Gefter. 1991. Immunodominance: intramolecular competition between T cell epitopes. J. Immunol. 146:2137.
Glimcher, L. H., T. Hamano, R. Asofsky, D. H. Sachs, M. Pierres, L. E. Samelson, S. O. Sharrow, W. E. Paul. 1983. IA mutant functional antigen-presenting cell lines. J. Immunol. 130:2287.
Verginis, P., M. M. Stanford, G. Carayanniotis. 2002. Delineation of five thyroglobulin T cell epitopes with pathogenic potential in experimental autoimmune thyroiditis. J. Immunol. 169:5332.
Rao, V. P., A. E. Kajon, K. R. Spindler, G. Carayanniotis. 1999. Involvement of epitope mimicry in potentiation but not initiation of autoimmune disease. J. Immunol. 162:5888.[Abstract/Free Full Text]
Lamas, L., P. C. Anderson, J. W. Fox, J. T. Dunn. 1989. Consensus sequences for early iodination and hormonogenesis in human thyroglobulin. J. Biol. Chem. 264:13541.[Abstract/Free Full
Byfield, P. G., D. Clingan, R. L. Himsworth. 1984. Exposure of thyroxine residues in human thyroglobulin: two-site binding studies. Biochem. J. 219:405.
Sakata, S., K. Okuda, Y. Aihara, Y. Tanaka, S. Nakamura, T. Komaki, O. Tarutani, K. Miura. 1986. Genetic control of the production of anti-thyroid hormone antibodies in mice immunized with human thyroglobulin. Immunol. Invest. 15:481
Berzofsky, J. A.. 1983. T-B reciprocity: an Ia-restricted epitope-specific circuit regulating T cell-B cell interaction and antibody specificity. Surv. Immunol. Res. 2:223
Ozaki, S., J. A. Berzofsky. 1987. Antibody conjugates mimic specific B cell presentation of antigen: relationship between T and B cell specificity. J. Immunol. 138:4133.
Apostolopoulos, V., G. Chelvanayagam, P. X. Xing, I. F. McKenzie. 1998. Anti-MUC1 antibodies react directly with MUC1 peptides presented by class I H2 and HLA molecules. J. Immunol. 161:767.
Aharoni, R., D. Teitelbaum, R. Arnon, J. Puri. 1991. Immunomodulation of experimental allergic encephalomyelitis by antibodies to the antigen-Ia complex. Nature 351:147.
Murphy, D. B., S. Rath, E. Pizzo, A. Y. Rudensky, A. George, J. K. Larson, C. A. Janeway, Jr. 1992. Monoclonal antibody detection of a major self peptide: MHC class II complex. J. Immunol. 148:3483.[Abstract]
Nygard, N. R., C. Bono, L. R. Brown, J. Gorka, K. S. Giacoletto, W. T. Schaiff, M. B. Graham, D. W. McCourt, M. Kabeer, V. L. Braciale, et al 1991. Antibody recognition of an immunogenic influenza hemagglutinin-human leukocyte antigen class II complex. J. Exp. Med. 174:243
Gardine, C. A., F. Gentile, C. Pellegrini, F. Giallauria, G. Torelli, T. Kouki, L. DeGroot. 2003. Multiple fragments of human TG are capable of inducing oral tolerance to whole human TG. J. Endocrinol. Invest. 26:294
Champion, B. R., D. C. Rayner, P. G. Byfield, K. R. Page, C. T. Chan, I. M. Roitt. 1987. Critical role of iodination for T cell recognition of thyroglobulin in experimental murine thyroid autoimmunity. J. Immunol. 139:3665
Dai, Y. D., V. P. Rao, G. Carayanniotis. 2002. Enhanced iodination of thyroglobulin facilitates processing and presentation of a cryptic pathogenic peptide. J. Immunol. 168:5907
Sundick, R. S., D. M. Herdegen, T. R. Brown, N. Bagchi. 1987. The incorporation of dietary iodine into thyroglobulin increases its immunogenicity. Endocrinology 120:2078.(Yang D. Dai Petros Eliade)
Thyroid hormone-binding (THB) Abs are frequently detected in autoimmune thyroid disorders but it is unknown whether they can exert immunoregulatory effects. We report that a THB mAb recognizing the 5' iodine atom of the outer phenolic ring of thyroxine (T4) can block T cell recognition of the pathogenic thyroglobulin (Tg) peptide (2549–2560) that contains T4 at aa position 2553 (T4(2553)). Following peptide binding to the MHC groove, the THB mAb inhibited activation of the Ak-restricted, T4(2553)-specific, mouse T cell hybridoma clone 3.47, which does not recognize other T4-containing epitopes or noniodinated peptide analogues. Addition of the same THB mAb to T4(2553)-pulsed splenocytes largely inhibited specific activation of T4(2553)-primed lymph node cells and significantly reduced their capacity to adoptively transfer thyroiditis to naive CBA/J mice. These data demonstrate that some THB Abs can block recognition of iodine-containing Tg epitopes by autoaggressive T cells and support the view that such Abs may influence the development or maintenance of thyroid disease.
Introduction
Thyroid hormone-binding (THB)4 Abs are frequently detected in patients with autoimmune thyroid disorders or in animal models of experimental autoimmune thyroiditis (EAT) (reviewed in Refs.1 and 2). THB Abs recognize thyroxine (T4) or triiodothyronine (T3) residues on distinct hormonogenic sites of thyroglobulin (Tg) and are being viewed as a subset of Tg-reactive autoantibodies. Development of THB Abs does not usually result in systemic hypothyroidism because of the ability of thyroid-pituitary feedback mechanisms to compensate for hormone neutralization (1, 2). However, THB Abs are clinically quite important because they interfere with the measurement of T3 and/or T4 and may occasionally hinder appropriate diagnosis and treatment (3, 4, 5). THB Abs raise important questions at the basic level as well, because the parameters regulating the immunogenicity of the T3 and T4 moieties, which are considered as haptens covalently linked to the Tg molecule, remain mostly unknown. Also, anti-Tg Abs may exert modulatory effects in EAT via their influence on the generation of pathogenic Tg peptides (6).
In Tg, the hormonogenic site at amino acid position 2553, at which a Tyr residue has been replaced by T4 via intramolecular coupling of two iodotyrosines, is one target of THB Abs (7). This site contributes to the formation of a B cell determinant but it is also encompassed within the Ak-binding, 12mer T cell epitope (2549–2560) that was originally discovered to cause EAT by Roitt’s group (8, 9). This 12mer peptide at aa position 2553 (T4(2553)) is rather unique among the thirteen so far mapped pathogenic Tg sequences (10), in that it is the only known Tg peptide that contains T4 and elicits strong proliferative as well as cytotoxic T cell responses (8, 9, 11, 12, 13). T4 is an integral part of this epitope because replacement of T4 by a Tyr residue abolishes T cell recognition (8). Furthermore, the four iodine atoms within the two-phenyl ring structure of T4 appear to sufficiently modify the peptide-MHC complex so as to elicit a distinct subset of thyroid-infiltrating T cells that recognize only the iodinated determinant (10, 12, 14, 15). These observations strongly supported the view that posttranslational modifications of Tg form a determinant that elicits both THB Abs and autoaggressive T cells.
In this study, we set out to examine whether THB Abs, represented by two mAbs, could influence T cell recognition of the T4(2553) peptide. This aim was made feasible by the generation of the 3.47 T cell hybridoma clone that was serendipitously found to recognize the T4(2553) epitope (6). In particular, we wanted to investigate whether THB Abs could detect the T4(2553) peptide even after its binding to the Ak molecule and block its MHC-restricted presentation to T cells. Such evidence would provide support for an immunoregulatory role of THB Abs in the disease process.
Materials and Methods
Ag presentation
The preparation of F-moc-L-thyronine, F-moc-L-thyroxine, and the synthesis of thyronine (T0)- or T4-containing peptides were as previously described (12). The sequences of these 12mer peptides, conserved between the murine and human species, were: Tg(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12) NH2-NIFE T4 QVDAQPL-NH2, T4(5); Tg(2549–2560) NH2-STDD T0/T4 ASFSRAL-NH2, T0/T4(2553); and Tg(2559–2570) NH2-ALENATRD T4 FII-NH2, T4(2567). The composition and relative purity of each peptide were verified by time-of-flight mass analysis on a Bio-Ion 20 Analyzer (Applied Biosystems) and, in some cases, by amino acid analysis of peptide hydrolysates (6 N HCl, 22 h, 121°C) on a Beckman 6300 amino acid analyzer (Beckman Coulter). Purity was determined to be >90%. The mouse Tg peptides (2494–2510) Ac-GLINRAKAVKQFEESQG-NH2 and (2694–2711) Ac-CSFWSKYIQTLKDADGAK-NH2, as well as the mouse lysozyme peptide (46–62), were synthesized at the Alberta Peptide Institute (Edmonton, Alberta, Canada) and purified as previously described (16, 17). Mouse Tg was purified from homogenates of thyroid glands of outbred ICR mice (Harlan Bioproducts for Science) by gel filtration over a Sepharose CL-4B column as described (17). Tg was dissolved in PBS, filter-sterilized, and stored at –20°C until use. Free T4 (Sigma-Aldrich) was dissolved in 99:1 of absolute methanol and 30% ammonium hydroxide at a stock concentration of 5 mg/ml.
Cell lines and T cell activation assay
The T cell hybridoma clone 3.47 was generated following a modified method of Perkins et al. (18). Briefly, lymph node cells (LNC) from CBA/J mice (The Jackson Laboratory), immunized with mouse Tg emulsified in CFA (with Mycobacterium butyricum; Difco), were further stimulated in vitro with the same Ag and fused with BW5147 –– cells (a kind gift of P. Marrack, National Jewish Medical and Research Center, Denver, CO) using polyethylene glycol 1500 (Boehringer Mannheim). Screening and cloning were performed as previously described (16). Expression of TCR and CD4 molecules on the 3.47 cells was verified by direct FACS analysis (data not shown). The TA3 cell line, produced by fusion of B cells from CAF1 mice with the M12.4.1 BALB/c B lymphoma (16, 19), was kindly provided by L. Glimcher (Harvard University, Cambridge, MA). The IL-2-dependent CTLL-2 line was purchased from American Type Culture Collection. All cells were cultured in DMEM (Invitrogen Life Technologies) supplemented with 10% FBS (Harlan Bioproducts
for Science), 10 mM HEPES buffer, 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (all from Invitrogen Life Technologies), and 5 x 10–5 M 2-ME (Sigma-Aldrich). CTLL-2 cells were grown in culture media supplemented with 10% supernatant from Con A-activated rat spleen cells.
In the activation assay, 105 3.47 T cells and 105 TA3 cells or macrophages were cultured in flat-bottom wells of 96-well plates, with or without various Ags, in a total volume of 200 μl. Following a 24 h incubation, 100 μl of supernatant was removed and stored at –20°C for assessment of IL-2 release, as measured by the proliferation of CTLL-2 line using [3H]thymidine (DuPont). Peritoneal macrophages were harvested from mice 4 days after i.p. injection of 2.5 ml of 3% thioglycolate per mouse.
Purification of mAbs and ELISA
The 55H8 and 91A1 mAbs (both IgG1) were derived from splenic B cells of BALB/c mice immunized with human Tg, following fusion with the mouse myeloma NSO. The mAbs were purified from ascites fluid by affinity chromatography on protein G-Sepharose 4 Fast Flow columns (Pharmacia). The purified material was concentrated to 1–2 mg/ml in PBS, filter-sterilized and stored in 50% glycerol at –20°C until use. Isotyping was performed with the ISO-2 kit (Sigma-Aldrich). The hybridomas specific for Ak (10-3.6.2, IgG2a) or Ek (14-4-4S, IgG2a) were purchased from American Type Culture Collection and the respective mAbs were purified from culture supernatants by affinity chromatography, as described earlier. ELISA was performed by coating microtiter polyvinyl chloride plates (Dynatech Laboratories) with 1 μg/well of Tg or Tg peptides and by using an alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma-Aldrich) as the second Ab. Absorbance at 405 nm was measured using a microplate reader (Vmax; Molecular Devices).
LNC proliferation assay and assessment of EAT
Female, 6- to 8-wk-old CBA/J mice were s.c. immunized with 100 μg of T4(2553) in 100 μl of CFA emulsion and 9–12 days later, the draining LNC were restimulated in a recall assay in vitro. All experimental procedures were reviewed and approved by the Animal Care Committee at Memorial University of Newfoundland. In blocking assays, mitomycin C-treated splenic cells (107 cells/ml) were used as APC following incubation with 10 μg/ml T4(2553) peptide for 6 h at 37°C. After washing away the free peptide, the T4(2553)-pulsed splenic cells (107 cells/ml) were cultured with or without 60 μg/ml blocking mAbs for 1 h. The APC were then washed and added (4–6 x 105 cells/well) to an equal number of T4(2553)-primed LNC for 96 h in a total volume of 200 μl of DMEM supplemented with 1% normal mouse serum. During the last 18 h of culture, 1 μCi of [3H]thymidine was added to each well. Cell harvesting and the counting of incorporated radioactivity were done as previously described (20). Adoptive transfer of EAT by peptide-primed LNC was assessed 13 days posttransfer. Effector cell activation conditions were slightly modified from conditions earlier described: mitomycin C-treated splenic APC (2 x 107/ml) were pulsed for 4 h with peptide (30 μg/ml), washed, incubated with 100 μg/ml blocking mAb for 2 h, washed, and subsequently cultured with peptide-primed LNC at 2:1 effector to APC ratio. Four days later, 107 live cells, separated by Ficoll centrifugation, were i.p. injected into each recipient. Fixation, embedding, and sectioning of the thyroids were performed as previously described (21). The mononuclear cell infiltration index was scored as follows: 0, no infiltration; 1, interstitial accumulation of cells between two or three follicles; 2, one or two foci of cells at least the size of one follicle; 3, extensive infiltration 10–40% of total area; 4, extensive infiltration 40–80% of total area; and 5, extensive infiltration >80% of total area. Each mouse was assigned the average infiltration index observed from both thyroid lobes (at least 10 sections per lobe were read and the maximum infiltration index per lobe was scored). The results were analyzed by the nonparametric Wilcoxon rank-sum test.
Results
The Ak-restricted T cell hybridoma clone 3.47 recognizes a Tg peptide containing a unique hormonogenic site
The 3.47 T cell hybridoma clone was initially selected for its reactivity to mouse Tg-pulsed TA3 cells but upon continuous culture it lost responsiveness to intact mouse Tg (data not shown). However, it was serendipitously found that 3.47 was strongly activated by the conserved Tg peptide T4(2553), whereas it was unreactive to two other Tg peptides (2494–2711) (Fig. 1A). The 3.47 clone is Ak-restricted because activation was blocked by an Ak-specific but not Ek-specific mAb (Fig. 1B). To test whether the presence of T4 within any peptide backbone is sufficient to activate the 3.47 clone, we examined the activation properties of two additional T4-containing 12mer peptides of Tg: the T4(5) (aa 1-12) and T4(2567) (aa 2559-2570). As shown in Fig. 1C, TA3 APC pulsed with those peptides or free T4 could not stimulate 3.47 to secrete IL-2. Lack of activation was not due to lack of peptide binding to Ak because the T4(2567) peptide inhibited 3.47 activation by T4(2553) peptide in a competitive inhibition assay to t
he same extent as the mouse lysozyme peptide (46–62), which is known (19) to bind strongly to Ak (Fig. 1D). The T4(5) peptide and free T4 were not inhibitors suggesting very weak or no binding to Ak. The pathogenic 9mer Tg peptide (2496–2504) that binds to Ek (10) also showed no inhibitory effect (data not shown). These results confirmed that the amino acid residues surrounding T4 were essential for Ak-binding and/or optimal TCR contact that would lead to the activation of the 3.47 clone.
FIGURE 1. The 3.47 T cell hybridoma clone specifically recognizes the Tg peptide T4(2553) and is Ak-restricted. TA3 cells were used as APC. Points indicate mean of triplicate wells and denote radioactive thymidine uptake (cpm) by the IL-2-dependent CTLL cells. A, IL-2 secretion following activation of the 3.47 clone by the Tg peptides is shown. B, 3.47 T cell activation is shown in the presence of increasing concentrations of mAbs specific for Ak or Ek Ags and 150 nM of T4(2553) peptide. Control cultures, without mAbs, yielded a mean of 13,120 cpm. C, Activation of 3.47 T cells by T4(2553) but not other T4-containing peptides of Tg. D, Competitive inhibition of 3.47 T cell activation in the presence of 37.5 nM of T4(2553) and increasing amounts of inhibitor peptides. Control culture cpm mean without inhibitory peptide is 14,830; without Ag is 530.
The 3.47 T cell clone recognizes an iodinated determinant
To assess whether iodine atoms contributed to the formation of the epitope recognized by the 3.47 clone, we compared the activation profiles obtained following recognition of the T4(2553) peptide vs the T0(2553) peptide analog that lacks the four iodine atoms within the T4 structure. It was found that 3.47 cells were activated by TA3 pulsed with T4(2553), but were unresponsive to TA3 pulsed with the noniodinated analog (Fig. 2A). The T0(2553) peptide could almost completely block the activation of 3.47 by T4(2553) (Fig. 2B) suggesting strong binding of the T0-containing analog to the Ak molecule. These results clearly demonstrate that the 3.47 clone recognizes a determinant modified by iodine atoms within the two-phenolic ring structure of T4, and also suggest that the iodine atoms do not critically influence the binding of the T4(2553) peptide to the Ak molecule.
FIGURE 2. The 3.47 clone recognizes a Tg determinant modified by iodine atoms. TA3 cells were used as APC. A, Lack of 3.47 T cell activation by the thyronine-containing analog peptide T0(2553). B, Competitive inhibition of 3.47 activation in the presence of 25 nM of T4(2553) and increasing amounts of the competitor analog peptide T0(2553). Data denote cpm mean of triplicate wells and are representative of three separate experiments. Control culture cpm mean values are the same as in Fig. 1D.
Iodine atoms on the phenolic ring of T4 contribute to the formation of determinants recognized by the T4-reactive mAbs 55H8 and 91A1
We have previously reported the generation of a Tg-specific mAb 55H8 that binds to the T4(2553) peptide (12). We proceeded to examine whether 55H8 is T4-reactive by monitoring its binding against other T4-containing peptides. It was found that 55H8 reacted similarly against T4(5) and T4(2567) whereas it did not bind to the T0(2553) analog (Fig. 3A). These data suggested that the 55H8 mAb is T4-specific and that the iodine atoms on the phenolic ring of T4 are essential for such recognition. This was further confirmed by a competitive inhibition ELISA in which free T4, or T4-containing peptides but not T0(2553), completely inhibited the binding of 55H8 to mouse Tg (Fig. 3C).
FIGURE 3. A and B, Reactivity of 55H8 and 91A1 mAbs to the indicated Tg peptides encoding hormonogenic regions. ELISA plates were coated at 1 μg of peptide per well. C and D, Competitive inhibition of mAb binding to mouse Tg using 1.0 μg/ml 55H8 (C) or 5.0 μg/ml 91A1 (D) in the presence of increasing amounts of free T4 or Tg peptide competitors. Points indicate mean of triplicate wells; SD values were <10% of the mean. OD values in the absence of inhibitory peptides were 1.48 (C) and 0.947 (D).
When we compared the reactivity profile of 55H8 with that of a second T4-specific mAb 91A1, we obtained an overall similar pattern of response with some differences. As shown in Fig. 3B, mAb 91A1 reacted with all the three T4-containing peptides but not with T0(2553) peptide on a solid phase despite a weaker binding to peptides T4(5) and T4(2553). Similarly, in competitive inhibition, all of the T4-containing peptides and free T4 molecule could inhibit binding of 91A1 to mouse Tg (Fig. 3D), although such inhibition was detectable at higher concentrations of peptides and T4 molecule than those for 55H8. This was partly due to the fact that a 5-fold higher concentration (5 μg/ml) of 91A1 was used in the inhibition assay to achieve a maximum OD value of 0.947 in the absence of inhibitor. Thus, the two T4-specific mAbs, 55H8 and 91A1, exhibited an overall similar reactivity profile to T4-containing peptides, but the 91A1 seemed to recognize T4 with a different affinity and/or such recognition was more influenced by neighboring amino acids than that of 55H8.
The T4-reactive mAbs 55H8 and 91A1 differ in their fine specificity
The T4 molecule contains four iodine atoms, two at positions 3 and 5 of the inner phenolic ring and another two at positions 3' and 5' of the outer ring. The hormone T3 lacks the iodine atom at the 5' position. To examine the fine specificities of 55H8 and 91A1, we conjugated rabbit serum albumin (RSA) with T4 or T3 and tested the binding of these mAbs to the conjugates by ELISA. The mAb 55H8 bound strongly to T4-RSA but not to T3-RSA (Fig. 4A), whereas 91A1 bound equally well to both conjugates (Fig. 4, A and B). Both mAbs bound equally well to mouse Tg. These data suggested that the iodine atom at the 5' position of the outer ring is an integral component of the determinant recognized only by 55H8, and not by the 91A1 mAb.
FIGURE 4. The mAbs 55H8 (A) and 91A1 (B) differ in their fine specificity. ELISA reactivities were assessed using Tg, or conjugates of T3 or T4 with RSA (1 μg/well) in the presence of increasing concentrations of mAbs shown.
CR recognition of the T4(2553)-Ak complex is blocked by the 55H8 mAb
It is well known that peptide-specific Abs do not commonly recognize their ligand within peptide-MHC complexes due to sequestration of the peptide within the MHC groove. Due to the bulkiness of the T4 moiety within T4(2553), and the requirement for T4 in the TCR triggering of the 3.47 clone, we envisaged that the T4-reactive 55H8 mAb might bind to the peptide-MHC complex and block peptide recognition by this T cell clone. To test this, TA3 cells (Fig. 5A) or peritoneal thioglycollate-stimulated macrophages (Fig. 5B) were pulsed with 2 μg/ml T4(2553) for 6 h, washed, and subsequently cultured with 3.47 T cells in the presence of increasing amounts of 55H8 and 91A1 mAbs or a control Tg-specific mAb of the same IgG subclass (3B3). Activation of 3.47 was strongly inhibited by 55H8, whereas the 91A1 and the control mAb 3B3, used at equimolar amounts, had no effect. A similar blocking effect of 55H8 was seen following fixation of T4(2553)-pulsed TA3 APC with glutaraldehyde (Fig. 5C). These data suggested that some T4-specific Abs, exemplified by 55H8, can block recognition of the pathogenic T4(2553) epitope by T cells.
FIGURE 5. The T4-specific mAb 55H8 blocks recognition of the T4(2553) peptide by the 3.47 clone. The APC line TA3 (A) or thioglycollate-stimulated peritoneal macrophage from CBA mouse (B) were pulsed with T4(2553) (2 μg/ml) for 6 h, washed and cultured with 3.47 cells in the presence of increasing concentration of the mAbs as shown. C, TA3 cells were pulsed with T4(2553) as in A, and were then fixed with 0.05% glutaraldehyde (F) for 30 s, or left untreated (N). Inhibition of the 3.47 cell activation was monitored in the presence of increasing concentrations of the mAbs shown. Points indicate the cpm mean of triplicate wells. Mean values of control wells in the absence of mAbs were (cpm): 14,097 (A); 31,555 (B), and 23,450 (C). Background values in the absence of T4(2553) peptide ranged from 80 to 400 cpm. Similar results were obtained in three separate experiments.
55H8 inhibits activation of T4(2553)-primed LNC and adoptive transfer of EAT
The potential of the 55H8 mAb to inhibit T cell activation was subsequently tested in a proliferation assay using T4(2553)-primed LNC from CBA/J mice. Mitomycin C-treated syngeneic splenocytes were pulsed with T4(2553), washed, and cultured with T4(2553)-primed LNC in the presence of 55H8 or 91A1. As shown in Fig. 6A, T4(2553)-pulsed, but not unpulsed, splenocytes induced a strong proliferative LNC response that was significantly inhibited only by 55H8 and not by the 91A1 or 3B3 control mAbs. Subsequent adoptive transfer of these cell populations into syngeneic CBA mice showed that effector cells cultured with the 55H8-treated, T4(2553)-pulsed splenocytes induced EAT with an average thyroidal lymphocyte infiltration index of 1.42 ± 0.58 (Fig. 6B). In contrast, the T4(2553)-primed LNC, cultured with the nontreated or the 91A1-treated APC, elicited a significantly higher lymphocytic infiltration of the thyroid with an average infiltration index of 2.42 ± 0.58 (p < 0.05). The large inhibition of the T4(2553)-specific T cell activation by 55H8 and the reduction of the pathogenic potential of such cells in vivo strongly suggested an immunoregulatory role for T4-reactive Abs following the formation of hormonogenic peptide: MHC complexes resulting from the processing of Tg or Tg fragments.
FIGURE 6. Activation of T4(2553)-specific pathogenic T cells is blocked by the 55H8 mAb. A, T4(2553)-primed LNC were cultured with mitomycin C-treated splenocytes that were pulsed with 10 μg/ml T4(2553) peptide for 6 h, washed and subsequently blocked with 60 μg/ml mAb 55H8, 91A1, or 3B3 () or no mAb () for 1 h. LNC proliferation was measured by adding [3H]thymidine during the last 18 h of a 4-day culture. Splenocytes not pulsed with peptide were used as controls (). B, Adoptive transfer of EAT by 107 T4(2553)-primed LNC following their in vitro restimulation with mitomycin C-treated, T4(2553)-pulsed syngeneic (CBA/J) splenocytes in the presence of 100 μg/ml 55H8 and 91A1 mAbs. EAT severity was scored on day 13 after transfer and the data were analyzed by the nonparametric Wilcoxon rank-sum test.
Discussion
Among the six hormonogenic sites described in human Tg, corresponding to tyrosyls 5, 685, 1290, 2553, 2567, and 2746 (22), the one at position 2553 has attracted special attention because peptides encompassing this site are recognized by human autoantibodies (7) and stimulate the adoptive transfer of murine EAT by T cells primed in vivo with the same peptide or mouse Tg (7, 8, 9, 10, 11, 12, 13, 14). In this report, we provide evidence that the mAb 55H8, which requires the 5' iodine atom of the outer phenolic ring of T4 for binding to Tg, markedly inhibits the activation of T cells responding to the pathogenic T4(2553) peptide. This suggests that the B cell and T cell responses targeting this site can be antagonistic at the effector stage.
The 55H8 and 91A1 mAbs, binding to T4 and T3, respectively, represent only certain subsets of all THB Abs that include Abs reactive to both T3 and T4 or Abs binding to T3, T4, and reverse T3 (1). Both mAbs were generated using human Tg as an immunogen, but it is unknown whether T4 at position 2553 contributed solely to their induction, because there are at least two hapten-like T4 residues exposed on the surface of the Tg "carrier" (23, 24). Flanking residues can contribute to the antigenicity of hormonogenic sites (7) and the inhibition data suggested differences between 55H8 and 91A1 binding to Tg that could be accounted for by steric effects of T4, surrounding residues and/or by binding affinity differences between the two mAbs. Conversely, the binding of 55H8 and 91A1 to Tg was completely inhibited by free T4 in our study, suggesting that the B cell determinants recognized by these mAbs are localized within the thyroxyl residue. In contrast, the 3.47 clone was not activated by free T4, or T4-containing
peptides other than T4(2553), in agreement with previous studies with hormonogenic peptides, emphasizing the role of amino acid residues surrounding T4 for MHC binding or TCR contact (12, 15).
THB Abs are detected both in human autoimmune thyroid disorders and in models of spontaneous thyroiditis (1, 2, 25, 26, 27). However, they usually exert no appreciable effects on metabolic status because of: 1) their relatively low binding affinity as compared with physiologic blood proteins such as thyroxine-binding globulin; and 2) the capacity of the thyroid gland to compensate for hormone neutralization via the thyroid-pituitary feedback mechanism (1, 2). In addition, hormonogenic sites on Tg may constitute minor antigenic determinants serologically because THB Abs are found only in a small portion of patients with anti-Tg Abs (2). The relative autonomy of THB Ab production has also been reported in mice, in which the Ir genes controlling the response to human Tg vs thyroid hormone seem to be different (2, 28). Although the molecular basis of these observations remains obscure it is becoming increasingly clear that THB Abs, once formed, can exert immunoregulatory influences. For example, we have previously shown that 55H8 when bound to Tg suppresses the generation of T4(2553) in APC processing the immune complex (6). We have hypothesized that this suppressive effect could be due to peptide sequestration and prevention from loading on Ak molecules, as predicted by the T cell to B cell reciprocity hypothesis (29, 30). In this report, 55H8 has been shown to block recognition of T4(2553) after the peptide is bound in the MHC groove. This steric hindrance effect can be explained by a protrusion of the thyroxyl side chain above the groove to allow binding of 55H8 and the strict dependence of 3.47 activation on the presence of iodine atoms within the peptide-MHC complex because this clone does not react to the T0(2553) analog. Abs accessing peptides bound to MHC class I (31) or class II (32, 33, 34) molecules have been previously reported.
Our results are in agreement with previous studies showing that LNC from CBA/J mice primed with the T4(2553) peptide could not be cross-stimulated in vitro with the T0(2553) analog and this regimen failed to generate effector cells that could adoptively transfer EAT (12). This suggested that the bulky iodine atoms of T4 (atomic radius of iodine is 133 pm) modify the peptide-MHC complex sufficiently so as to elicit a distinct subset of T cells, probably represented by the 3.47 clone described in this report, that recognize only the iodinated determinant. Also, our results are in partial agreement with those of Dawe et al. (15) who showed that the activation of a hybridoma clone responding to T4(2553) but not to T0(2553) could be blocked by an anti-T4 mAb. However, in that study, the iodine atoms within the T4 moiety were deemed necessary for peptide binding to MHC because the T0(2553) analog could not block activation of the T4(2553)-specific clone. Our data (Fig. 2) do not support this interpretation because T0(2553) can almost completely block the activation of the 3.47 clone suggesting that the iodine atoms are critical in forming a TCR-contacting residue. The mild immunopathogenicity of the T0(2553) peptide in CBA/J hosts (12) also strongly suggests that the iodine atoms of the thyroxyl side chain are not necessary for peptide-MHC binding. Removal of iodine from the tryptic human Tg fragment (2513–1713) has been reported to convert epitope(s), encompassed within this fragment, from immunogenic to tolerogenic (35), emphasizing the immunomodulatory potential of iodine incorporated in Tg.
Our current and previous (6) findings highlight the interplay between humoral and cellular immunity in autoimmune thyroiditis. They demonstrate that THB Abs, represented by 55H8, can prevent the activation of T cells responding to the pathogenic T4(2553) epitope, either via peptide sequestration (6) or blockade of peptide-MHC complexes. Abs specific for complexes of As and an encephalitogenic determinant of myelin basic protein have analogously been reported to block EAE (32). The T4(2553) can activate T cells following priming with intact Tg in vivo but it has been classified as subdominant because T4(2553)-primed T cells seldom respond to intact Tg processed in vitro (8, 9, 10, 12). Nonetheless, tyrosyl 2553 is one of the first sites to be iodinated in Tg (22). Given that enhanced iodine incorporation increases the immunogenicity of Tg(36, 37, 38), it is plausible that the T4(2553) peptide becomes an early target of autoaggressive T cells in autoimmune thyroiditis, following the subsequent coupling of a "donor" iodotyrosyl to form the hormonogenic site at position 2553. TBH Abs blocking the activation or effector function of such cells may help in the prevention or amelioration of the disease.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Dr. P. Marrack for the BW5147 – – cells and Dr. L. Glimcher for the TA3 cell line.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This study was supported by Grants from the Canadian Institutes for Health Research (to G.C.) and the National Institutes of Health Grant DK 45960 (to Y.M.K.). Y.D.D. is a recipient of a Memorial University PhD Fellowship.
2 Current address: Division of Immunoregulation, Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121.
3 Address correspondence and reprint requests to Dr. George Carayanniotis, Faculty of Medicine, Memorial University of Newfoundland, 300 Prince Philip Drive, St. John’s, Newfoundland A1B 3V6, Canada. E-mail address: gcarayan{at}mun.ca
4 Abbreviations used in this paper: THB, thyroid hormone binding; Tg, thyroglobulin; EAT, experimental autoimmune thyroiditis; T0, thyronine; T3, triiodothyronine; T4, thyroxine; RSA, rabbit serum albumin; LNC, lymph node cell.
Received for publication May 6, 2004. Accepted for publication December 23, 2004.
References
Benvenga, S., F. Trimarchi, J. Robbins. 1987. Circulating thyroid hormone autoantibodies. J. Endocrinol. Invest. 10:605
Sakata, S.. 1994. Autoimmunity against thyroid hormones. Crit. Rev. Immunol. 14:157.
Despres, N., A. M. Grant. 1998. Antibody interference in thyroid assays: a potential for clinical misinformation. Clin. Chem. 44:440.
Sapin, R., F. Gasser, A. Boehn, M. Rondeau. 1995. Spuriously high concentration of serum free thyroxine due to anti-triiodothyronine antibodies. Clin. Chem. 41:117.
Shimon, I., C. Pariente, J. Shlomo-David, Z. Grossman, J. Sack. 2003. Transient elevation of triiodothyronine caused by triiodothyronine autoantibody associated with acute Epstein-Barr-virus infection. Thyroid 13:211.
Dai, Y., K. A. Carayanniotis, P. Eliades, P. Lymberi, P. Shepherd, Y. Kong, G. Carayanniotis. 1999. Enhancing or suppressive effects of antibodies on processing of a pathogenic T cell epitope in thyroglobulin. J. Immunol. 162:6987.
Sakata, S., T. Ogawa, Y. Kimata, H. Takuno, H. Maekawa, M. Matsuda, O. Tarutani, K. Okuda. 1991. Synthesis and examination of antigenicity of four hormonogenic sites and two non-hormonogenic sites of human thyroglobulin. Mol. Cell. Endocrinol. 79:93
Champion, B. R., K. R. Page, N. Parish, D. C. Rayner, K. Dawe, G. Biswas-Hughes, A. Cooke, M. Geysen, I. M. Roitt. 1991. Identification of a thyroxine-containing self-epitope of thyroglobulin which triggers thyroid autoreactive T cells. J. Exp. Med. 174:363.[
Hutchings, P. R., A. Cooke, K. Dawe, B. R. Champion, M. Geysen, R. Valerio, I. M. Roitt. 1992. A thyroxine-containing peptide can induce murine experimental autoimmune thyroiditis. J. Exp. Med. 175:869.
Carayanniotis, G.. 2003. The cryptic self in thyroid autoimmunity: the paradigm of thyroglobulin. Autoimmunity 36:423.
Carayanniotis, G., Y. C. Kong. 2000. Pathogenic thyroglobulin peptides as model antigens: insights on the induction and maintenance of autoimmune thyroiditis. Int. Rev. Immunol. 19:557.
Dawe, K. I., P. R. Hutchings, M. Geysen, B. R. Champion, A. Cooke, I. M. Roitt. 1996. Unique role of thyroxine in T cell recognition of a pathogenic peptide in experimental autoimmune thyroiditis. Eur. J. Immunol. 26:768.
Carayanniotis, G., E. Chronopoulou, V. P. Rao. 1994. Distinct genetic pattern of mouse susceptibility to thyroiditis induced by a novel thyroglobulin peptide. Immunogenetics 39:21.
Chronopoulou, E., G. Carayanniotis. 1992. Identification of a thyroiditogenic sequence within the thyroglobulin molecule. J. Immunol. 149:1039.
Perkins, D. L., G. Berriz, T. Kamradt, J. A. Smith, M. L. Gefter. 1991. Immunodominance: intramolecular competition between T cell epitopes. J. Immunol. 146:2137.
Glimcher, L. H., T. Hamano, R. Asofsky, D. H. Sachs, M. Pierres, L. E. Samelson, S. O. Sharrow, W. E. Paul. 1983. IA mutant functional antigen-presenting cell lines. J. Immunol. 130:2287.
Verginis, P., M. M. Stanford, G. Carayanniotis. 2002. Delineation of five thyroglobulin T cell epitopes with pathogenic potential in experimental autoimmune thyroiditis. J. Immunol. 169:5332.
Rao, V. P., A. E. Kajon, K. R. Spindler, G. Carayanniotis. 1999. Involvement of epitope mimicry in potentiation but not initiation of autoimmune disease. J. Immunol. 162:5888.[Abstract/Free Full Text]
Lamas, L., P. C. Anderson, J. W. Fox, J. T. Dunn. 1989. Consensus sequences for early iodination and hormonogenesis in human thyroglobulin. J. Biol. Chem. 264:13541.[Abstract/Free Full
Byfield, P. G., D. Clingan, R. L. Himsworth. 1984. Exposure of thyroxine residues in human thyroglobulin: two-site binding studies. Biochem. J. 219:405.
Sakata, S., K. Okuda, Y. Aihara, Y. Tanaka, S. Nakamura, T. Komaki, O. Tarutani, K. Miura. 1986. Genetic control of the production of anti-thyroid hormone antibodies in mice immunized with human thyroglobulin. Immunol. Invest. 15:481
Berzofsky, J. A.. 1983. T-B reciprocity: an Ia-restricted epitope-specific circuit regulating T cell-B cell interaction and antibody specificity. Surv. Immunol. Res. 2:223
Ozaki, S., J. A. Berzofsky. 1987. Antibody conjugates mimic specific B cell presentation of antigen: relationship between T and B cell specificity. J. Immunol. 138:4133.
Apostolopoulos, V., G. Chelvanayagam, P. X. Xing, I. F. McKenzie. 1998. Anti-MUC1 antibodies react directly with MUC1 peptides presented by class I H2 and HLA molecules. J. Immunol. 161:767.
Aharoni, R., D. Teitelbaum, R. Arnon, J. Puri. 1991. Immunomodulation of experimental allergic encephalomyelitis by antibodies to the antigen-Ia complex. Nature 351:147.
Murphy, D. B., S. Rath, E. Pizzo, A. Y. Rudensky, A. George, J. K. Larson, C. A. Janeway, Jr. 1992. Monoclonal antibody detection of a major self peptide: MHC class II complex. J. Immunol. 148:3483.[Abstract]
Nygard, N. R., C. Bono, L. R. Brown, J. Gorka, K. S. Giacoletto, W. T. Schaiff, M. B. Graham, D. W. McCourt, M. Kabeer, V. L. Braciale, et al 1991. Antibody recognition of an immunogenic influenza hemagglutinin-human leukocyte antigen class II complex. J. Exp. Med. 174:243
Gardine, C. A., F. Gentile, C. Pellegrini, F. Giallauria, G. Torelli, T. Kouki, L. DeGroot. 2003. Multiple fragments of human TG are capable of inducing oral tolerance to whole human TG. J. Endocrinol. Invest. 26:294
Champion, B. R., D. C. Rayner, P. G. Byfield, K. R. Page, C. T. Chan, I. M. Roitt. 1987. Critical role of iodination for T cell recognition of thyroglobulin in experimental murine thyroid autoimmunity. J. Immunol. 139:3665
Dai, Y. D., V. P. Rao, G. Carayanniotis. 2002. Enhanced iodination of thyroglobulin facilitates processing and presentation of a cryptic pathogenic peptide. J. Immunol. 168:5907
Sundick, R. S., D. M. Herdegen, T. R. Brown, N. Bagchi. 1987. The incorporation of dietary iodine into thyroglobulin increases its immunogenicity. Endocrinology 120:2078.(Yang D. Dai Petros Eliade)