Search PubMed⌕ Search

Biomedical subjects

T F Davies

Publications and source records attributed to T F Davies.

At least 73 records · Page 4Linked to original sources

Induction of thyroid autoantibodies in naive mice by idiotypic manipulation.

Previously, we have shown that it is possible to induce, in naive mice, systemic autoimmune diseases (e.g., antiphospholipid syndrome and systemic lupus erythematosus) by idiotypic manipulation. In the present study we expanded our experience to examine whether idiotypic manipulation could be utilized to induce organ-specific autoantibodies and a disease mediated by cellular mechanisms, namely, experimental autoimmune thyroiditis. Fifteen BALB/c mice were immunized with a monoclonal mouse anti-human thyroglobulin (hTg) antibody; controls were immunized with an irrelevant mouse IgG. The mice immunized with anti-hTg antibody developed, 6 weeks after immunization, autoantibodies to human thyroglobulin, but not to dsDNA, cardiolipin, or myeloperoxidase. The presence of specific autoantibodies was associated with low production of thyroid hormones, and during a follow-up of 20 weeks the mice did not develop characteristic histological signs of thyroiditis. We conclude that idiotypic manipulation can induce anti-thyroglobulin autoantibodies.

Animals↗

The role of T cells expressing TcR V beta 13 in autoimmune thyroiditis induced by transfer of mouse thyroglobulin-activated lymphocytes: identification of two common CDR3 motifs.

The transfer of lymphocytes from mouse thyroglobulin (mTg)-immunized CBA/J (H-2k) mice following in vitro activation with mTg initiates experimental autoimmune thyroiditis (EAT) in syngeneic recipients. We have analyzed the T cell receptor (TcR) V gene families used by the intrathyroidal lymphocytic infiltrate of such mice. Using a radiolabeled RT-PCR technique with oligonucleotides detecting 17 mouse TcR V beta gene families to examine the heterogeneity of the amplified V-D-J (CDR3) fragments, we demonstrated that only the TcR V beta 13 amplifications consistently showed two similar homogeneous CDR3 sizes consistent with two clonally expanded T cell populations. Sequencing of the homogeneous RT-PCR products from these V beta 13++ populations confirmed the presence of clonal expanded T cells and identified two recurrent CDR3 motifs LTGKDTQ and LGEQ present in six of the seven samples. Both these motifs had been found as contributors to the T cell population in our previous studies of CBA/J mouse thyroiditis induced by active immunization with heterologous human (h) Tg. These data suggest that the autoepitope recognized was shared between hTg and mTg. It appears, therefore, that in transfer thyroiditis the intrathyroidal T cell clonal proliferation follows the homing of V beta 13 antigen-specific T cells which have been expanded by a brief (3 day) in vitro activation to mTg and utilize two distinct CDR3 motifs. CDR3 size heterogeneity in many of the other expressed V gene families also suggested the accumulation and recruitment of selected bystander T cells responding to additional but limited Tg or other self epitopes, perhaps on the basis of CDR3 shape rather than sequence. Such T cells may also have integral roles in the development of autoimmune thyroiditis.

Amino Acid Sequence↗

Prospective study of T cell receptor utilization following the induction of murine thyroiditis.

Murine experimental autoimmune thyroiditis (EAT) is a well established model of autoimmune disease initiated by immunization with thyroglobulin. We have previously analyzed the T cell receptor (TcR) V gene families used by the intrathyroidal lymphocytic infiltrate in CBA/J mice with well established thyroiditis EAT and have implicated T cells expressing the mTcR V beta 13 gene family. We have now proceeded to examine the time course of mTcR V gene family use following immunization with mTg. We used a radiolabelled RT-PCR technique with oligonucleotides detecting 17 mouse TcR V beta gene families to examine the heterogeneity of the amplified V-D-J (CDR3) fragments. As previously, the TcR V beta 13 amplifications showed the expression of two similar homogeneous CDR3 sizes consistent with two clonally expanded T cell populations. However, such T cell clonal expansion was observed to peak at day 25 and by 90 days had markedly diminished despite the continuing presence of extensive histologic infiltration. An additional immunization with mTg at 63 days failed to maintain the mTcR V beta 13 clonal presence. Further confirmation of these observations was obtained by direct analysis of intrathyroidal T cells rescued from mice with EAT. Such intrathyroidal T cells, 25 days after mTg, demonstrated a marked increase in mTcR V13 expressing T cells to 9.4% compared to 2% of T cells in peripheral blood. It appeared, therefore, that in EAT the accumulation of V13 expressing T cells was a transient phenomenon which peaked at 25 days after immunization. The persistence of an intrathyroidal infiltration indicated that such T cells must have been accompanied by the accumulation and recruitment of additional selected bystander T cells. Such non-specific T cells may also have an integral role in the progression of autoimmune thyroiditis.

Animals↗

Multimeric complex formation by the thyrotropin receptor in solubilized thyroid membranes.

The TSH receptor (TSHR) has a large glycosylated ectodomain comprising the amino-terminal half of the molecule (394 of 743 residues) implicated in TSH binding, as well as autoantibody recognition in Graves' disease. In this study we employed antibodies specific for the amino-terminus (Ab1), midportion (Ab2), and carboxyl-terminus (Ab3) of the TSHR-ectodomain, previously mapped using recombinant receptor proteins, to detect the natural receptor present in detergent-solubilized porcine thyroid cell membranes via immunoblotting. Several forms of the receptor were detected. In reduced samples Ab1 detected full-length holoreceptors present in both nonglycosylated and glycoslylated forms of apparent molecular masses 80 and 90 kDa, respectively, as well as apparent dimeric nonglycosylated and dimeric glycosylated holoreceptor forms resistant to reduction. Also detected by Ab1 were a glycosylated amino-terminal 47- to 52-kDa fragment of the holoreceptor (gly alpha-subunit), reduced to 42 kDa (alpha-subunit) by Endo F deglycosylation. Ab2 detected all of the same forms. Ab3 detected primarily a carboxy-terminal, nonglycosylated fragment of 35 kDa (beta-subunit). In unreduced samples, the recognition pattern was unchanged with Ab1. Ab2 detected monomeric and dimeric beta-subunits, as well as higher order complexes. The different TSHR forms present in unreduced preparations were resolved by ammonium sulfate precipitation, confirming their autonomy. The data demonstrate the presence of multiple forms of the natural TSHR. Their roles in TSH action and TSHR autoimmunity require further exploration.

Ammonium Sulfate↗

Intrathyroidal T cell accumulation in Graves' disease: delineation of mechanisms based on in situ T cell receptor analysis.

We have explored the pattern of T cell clonal and nonclonal expansion within the thyroid glands of 7 patients with surgically treated hyperthyroid Graves' disease. Radiolabeled RT-PCRs were performed with 18 V alpha and 21 V beta oligonucleotides as forward primers and 32 P-labeled constant (C) region oligonucleotides as reverse primers, giving 273 experiments in all. Peripheral blood mononuclear cells showed CDR3 banding patterns involving at least 6-12 distinct bands per V gene family. However, 2 distinct banding patterns were seen with intrathyroidal T cell samples. The first was a limited number of bands (< 6), representing within V gene family restriction, seen in 38/153 (25%) PCR positive samples. The second was the presence of markedly enhanced bands representing either clonal expansion or accumulation of T cells using the same V gene with the same CDR3 length and observed in 15/153 (10%) of samples. Further examination of the enhanced bands by sequencing of gel-selected PCR-amplified V gene products showed the presence of both T cell clonal expansion, as evidenced by similar CDR3 sequences, and in a few samples, heterogenous T cell populations, as evidenced by differing CDR3 sequences, yet all sharing the same V gene family. These data support the hypothesis that the human intrathyroidal T cell population in Graves' disease was selected by two distinct mechanisms. The first was associated with the TcR V gene invariant product and represented by nonclonal accumulation of T cells sharing the same V gene. The second mechanism responsible for T cell accumulation was specific antigen recognition by the human T cell receptor CDR3 region and was represented by clonally expanding T cells.

Amino Acid Sequence↗

Monoclonal antibodies to the human TSH receptor: epitope mapping and binding to the native receptor on the basolateral plasma membrane of thyroid follicular cells.

We have characterized four murine monoclonal antibodies (mAbs) to the extracellular domain of the human TSH receptor (TSH-R.E), the target autoantigen of Graves' disease. Recombinant TSH-R.E used as immunogen, was produced in E. coli as a fusion protein with glutathione-S-transferase or in a baculovirus-insect cell system, as a non-fusion glycoprotein. To increase the epitope specificity of the mAbs, two different strains of mice (H-2(b) and H-2(d)) were immunized. The epitopes recognized by the mAbs were characterized by immunoblotting with various recombinant constructs of TSH-R.E and by binding to overlapping synthetic peptides of the receptor. The four IgG mAbs characterized recognized epitopes localized to different regions on the TSH-R.E; amino acids 22-35 (A1O and A11, both IgG2b from H-2(b) animals), amino acids 402-415 (A7, IgG2b from H-2(b) animals) and amino acids 147-228 (A9, IgG1 from H-2(d) animals). Immunolocalization studies showed that mAb A9 recognized TSH-R.E on unfixed cryostat sections, where binding was localized to the basolateral plasma membrane of thyroid follicular cells, suggesting that this antibody reacts with the native receptor on thyroid cells. The binding of the mAbs A7, A10 and A11 was also restricted to the basal surface of thyroid cells, but only after acetone fixation of the sections, implying that the epitopes recognized on the amino and carboxyl terminus of the extracellular region of the receptor are not accessible on the native molecule. None of the mAbs stimulated cyclic AMP responses in COS-7 cells transiently transfected with full-length functioning TSH-R.E, whilst weak inhibition of binding of radiolabelled TSH to porcine membranes in a radioreceptor assay was apparent with mAb A10 and A11, but only at high concentrations of IgG. The ability of mAb A9 to bind to the native receptor without stimulating activity or inhibition of TSH binding suggests that antibody can bind to the central region of the TSH-R.E without perturbing receptor function. The availability of mAbs that recognize epitopes on different regions of the extracellular domain of TSH-R will lead to a better understanding of the autoantigenic regions on TSH-R implicated in disease activity.

Amino Acid Sequence↗

Thyroid-specific T cells in the Wistar rat: 3. Induction of anergy by a syngeneic thyroid cell line.

Using a syngeneic Wistar rat model we have shown that the Wistar rat thyroid (WRT) cell line causes significant and specific proliferation of lymph node T cells from normal Wistar rats, and of splenic T cells from a thyroiditis prone line of BB/W rats, when cultured in the presence of irradiated feeder cells. These T cell responses were associated with a marked increase in the number of CD8+ T cells. However, using normal Wistar rat T cells which had been previously exposed to WRT cells, rested and then re-exposed to WRT cells as antigen, we consistently found that the T cell population had been rendered unreactive, or anergic, to further thyroid cell stimulation. However, if recombinant rat IL-2 was added to the cultures, then T cell responsivity was seen on re-exposure to WRT cells. The lymphopenic BB/W rat also had T cells which showed a primary T cell response to the WRT cell line accompanied by a marked increase in CD8+ T cells. In contrast to the Wistar rat T cells, the BB/W T cells retained a proliferative responsiveness to WRT cells on re-exposure although such responsiveness could also be markedly enhanced with IL-2. These data suggested that antigen-mediated inhibitory signals were induced in normal Wistar rat T cells by the syngeneic WRT cell line, independent of the presence of co-stimulatory molecules. Furthermore, the thyroiditis prone BB/W rat T cells appeared to be less responsive to such anergy induction, perhaps contributing to their susceptibility to autoimmune thyroid disease.

Animals↗

Infections and autoimmune endocrine disease.

The literature examined in this review points to the possible involvement of infectious agents in the pathogenesis of autoimmune endocrine diseases, primarily autoimmune thyroid disease and diabetes mellitus. Various mechanisms have been proposed to explain induction of autoimmunity by infection but it seems that three possibilities may be important in individuals susceptible to developing autoimmune disease: molecular mimicry (perhaps to retroviruses); polyclonal T cell activation (by an endogenous superantigen or an infecting organism); and MHC class II antigen induction. It seems reasonable that all three mechanisms operate together or separately in different individuals. Data continue to accumulate in favour of infectious agents being important initiators of autoimmune disease.

Animals↗

Defining the major antibody epitopes on the human thyrotropin receptor in immunized mice: evidence for intramolecular epitope spreading.

To evaluate the B cell response to the extracellular domain of the human TSH receptor (hTSHR-ecd), we used recombinant hTSHR-ecd to immunize BALB/c mice (group A) and CBA/J mice (groups B and C). Mice from groups A and B were boosted once, and mice from group C received three antigen boosts. All individual mice developed highly specific hTSHR-ecd antibodies (hTSHR-ecd-Ab), confirmed by Western blot analyses. The B cell epitopes recognized by these murine hTSHR-ecd-Ab were mapped by enzyme-linked immunoassays using 26 synthetic overlapping peptides spanning the entire mature hTSHR-ecd [amino acids (aa) 22-415], i.e. without the signal sequence. Although all BALB/c and CBA/J mice antisera recognized peptide 1 (aa 22-41), the hyperimmunized CBA/J mice (group C) demonstrated recognition of additional peptides (numbers 21-26) clustered toward the carboxyl-terminus of the hTSHR-ecd (aa 322-415). Furthermore, group C serum blocked the binding of [125I]bTSH to native porcine TSHR, whereas sera from groups A and B were inactive. We were also able to map the B cell epitopes of antisera from rabbits immunized repeatedly with hTSHR-ecd and found the same recognition pattern of peptide 1 and additional peptides clustered near the carboxyl-terminus of the hTSHR-ecd (aa 322-341 and 367-415). These rabbit antisera also inhibited the binding of [125I]bTSH to native porcine TSHR. These data provide a comprehensive B cell epitope-mapping study of induced hTSHR-ecd-Ab and demonstrate intramolecular spreading of the epitopes recognized. Although the N-terminal region was highly antigenic, repeated immunization induced hTSHR-ecd-Ab targeted to a region critical for TSH binding.

Animals↗

Folding of the recombinant human thyrotropin (TSH) receptor extracellular domain: identification of folded monomeric and tetrameric complexes that bind TSH receptor autoantibodies.

We have analyzed protein folding and disulfide bond formation in the extracellular domain of the human TSH receptor (hTSHR-ecd) expressed in Escherichia coli. This domain, which begins at the amino-terminus and ends at residue 415, is a major autoantigen in human autoimmune thyroid disease. Refolding of reduced and denatured hTSHR-ecd occurred in polyacrylamide gels treated with 0.25 M KCl for visualization of protein bands. Under conditions of partial renaturation, at least three forms of the hTSHR-ecd were resolved by reelectrophoresis using sodium dodecyl sulfate-polyacrylamide gel electrophoresis: 1) unfolded monomers, 2) folded monomers, and 3) tetramers. Disulfide bond formation was implicated in both folding and tetramerization, as reduction of these forms produced only unfolded monomers. A natural variant of the hTSHR (v1.3), sharing 231 N-terminal amino acids with hTSHR-ecd, formed folded monomers and dimers on renaturation, but not tetramers, implicating one or more of the five cysteine residues residing between positions 231-415 in the association of dimers into tetramers. Binding of three different sources of hTSHR antibodies to these various forms of the hTSHR-ecd was assessed by immunoblotting using: 1) murine monoclonal antibodies (MAbs) generated against hTSHR-ecd, 2) rabbit polyclonal antisera generated against overlapping synthetic peptides spanning residues 37-71 of the hTSHR-ecd, and 3) human immunoglobulin G from patients with Graves' disease and detectable hTSHR-Ab. One of the MAbs, shown to recognize residues 21-35, and the rabbit polyclonal antibodies bound to all three forms of the hTSHR-ecd. Some of the hTSHR autoantibodies bound predominantly to the monomeric forms of the hTSHR, but autoantibodies were also identified that recognized tetrameric hTSHR-ecd. These data demonstrate that hTSHR-Abs recognize differing nonlinear and linear epitopes in the hTSHR-ecd and provide a methodology that should be useful for further defining the structural requirements for folding of this functionally and immunologically important domain of the hTSHR.

Animals↗

Human autoantibodies to the thyrotropin receptor: recognition of linear, folded, and glycosylated recombinant extracellular domain.

To examine the heterogeneity of autoantibodies to the human TSH receptor (hTSHR), we evaluated 20 sera from patients with Graves' disease for their recognition of prokaryotic (unglycosylated) and eukaryotic (insect cell glycosylated) recombinant hTSHR extracellular domain (ecd) in an unfolded (linear) and a folded (nonlinear) state. With the prokaryotic antigen, 12 (60%) bound folded hTSHR ecd monomer, 8 (40%) bound to the unfolded monomer, and 3 (15%) bound to a tetrameric species. Such binding to different hTSHR antigens was not mutually exclusive. In addition, 7 (35%) sera showed an apparently higher reactivity for the folded than the unfolded monomer. When reacted against the glycosylated insect cell hTSHR ecd, 9 (45%) sera recognized both the unfolded and folded monomer, and 5 (25%) recognized the tetrameric form. In all of our testing, 17 of the 20 sera (85%) bound to 1 or more of the recombinant hTSHR ecd antigens, and the recognition pattern appeared to be heterogeneous in at least 4 (20%) of the serum samples, with hTSHR antibodies recognizing linear, folded, and glycosylated hTSHR ecd monomers. We conclude, therefore, that patients with Graves' disease have autoantibodies that recognize multiple epitopes on the hTSHR ecd and that it is possible to classify them according to their recognition of linear, folded, and glycosylated products.

Autoantibodies↗

Endogenous antigen presentation by autoantigen-transfected Epstein-Barr virus-lymphoblastoid cells: T cell receptor N-region hydrophobicity relates to thyroid antigen recognition.

Seven human T cell lines from a patient with Graves' disease were raised against endogenously generated human thyroid peroxidase (hTPO) with stimulation indices ranging from 2.1 to 7.6. Clonal expansion within these T cell lines was demonstrated by sequencing multiple bacterial colonies containing RT-PCR-generated fragments derived from the expressed hTcRs. Some lines had more than one human T cell receptor (hTcR) alpha and beta chain mRNAs as judged by RT-PCR. Stopcodons present in several hTcR sequences indicated that only one V alpha and one V beta gene were translated. Both the V alpha/beta gene families and the J alpha/beta gene segments differed amongst the lines and no characteristic recognition sequences were discernable in the CDR3 regions. Using Kyte-Doolittle analysis we found hydrophobic peaks in most N alpha-regions (but not N beta regions) suggesting that hydrophobic interactions may be important in the recognition of hTPO. However, increasing affinity values, as measured by SI, were strongly correlated with decreasing hydrophobicity in the N alpha region (1st order regression, r = -0.93138, p < 0.01). Thus, lower affinity, self-reactive, T cells may be more hydrophobic ('sticky') in their N alpha regions while higher affinity cells may be characterized by TcRs with lower hydrophobicity. These findings demonstrate a substantial role for hydrophobic interactions in hTPO-reactive T cell receptors and further support a role for the TcR alpha chain in the recognition of thyroid autoantigen.

Antigen Presentation↗

Thyroglobulin-induced murine thyroiditis assessed by intrathyroidal T cell receptor sequencing.

Thyroid glands from CBA/J (H-2k) mice, with human thyroglobulin (Tg)-induced thyroiditis, were removed and total cellular RNA was prepared and reverse transcribed as target cDNA for PCRs to detect 17 murine T cell receptor (mTCR) V beta families. Southern analysis of the resulting PCR fragments showed a marked bias in the intrathyroidal expression of mTCR V beta gene families with only 10 detectable V beta gene families of the 17 tested. Four Tg-specific CBA/J T cell lines also showed a biased use of mTCR V beta gene families with only 7 V beta gene families detected of the 17 tested. Sequence analysis showed evidence of oligoclonal expansion of the intrathyroidal T cell populations, and some of the CDR3 sequences found in the thyroid infiltrates were identical, or similar, to examples of the patterns of CDR3 sequences seen in the Tg-specific lines. Overall, however, there was marked N-D-N sequence variation in both the lines and thyroid infiltrates. These data indicated that 1) mTcR V gene analysis of intact thyroid tissue reflected the thyroid Ag-specific T cells infiltrating the thyroid gland; 2) confirmed the phenomenon of oligoclonal expansion in early thyroiditis; 3) showed that certain V beta gene families may be overemployed in the T cell recognition of Tg in the context of H-2k; and 4) revealed, by the varied patterns of the CDR3 sequences, that the V gene itself may be more important in thyroid Ag recognition than N-D-N region motifs.

Amino Acid Sequence↗

Preservation of functioning human thyroid "organoids" in the severe combined immunodeficient mouse. III. Thyrotropin independence of thyroid follicle formation.

The severe combined immunodeficient (scid) mouse allows the in vivo reconstitution of thyroid follicles from thyroid monolayer cells when transplanted sc within an extracellular basement membrane matrix. After 2-3 weeks, these human thyroid organoids show an active microfollicular histology and secrete human thyroglobulin into the murine serum in response to the administration of recombinant human TSH. Furthermore, such organoids survive for more than 3 months in this functional state. To assess whether thyroid follicular reconstitution was TSH dependent, we examined organoid follicular reconstruction in T3-induced hyperthyroid scid mice, in which endogenous murine TSH was presumed to be totally suppressed. By providing water with 12 micrograms/ml T3, we increased the murine serum T3 levels from a mean of 1.9 nmol/liter in controls to greater than 12.0 nmol/liter. After 3 weeks, thyroid cells derived from normal human thyroid monolayers were suspended in an extracellular basement membrane matrix, and the suspension was transplanted sc into scid mice (with or without T3-induced hyperthyroidism) and allowed to reconstitute. Histological examination 4 weeks later showed a similar degree of thyroid follicle formation in mice treated with or without T3, indicating that the hyperthyroid state had caused no interference with thyroid follicle reconstitution. This was further confirmed by transmission electron microscopy, which demonstrated normal human thyroid follicle cell polarity in the organoids of both euthyroid and hyperthyroid mice. In addition, using an extracellular basement membrane preparation with reduced growth factor (epidermal growth factor, insulin-like growth factor-I, and platelet-derived growth factor) levels also allowed normal thyroid follicle formation in T3-fed mice. These data demonstrate that in vivo thyroid follicle formation is TSH independent and that extrathyroidal epidermal growth factor, insulin-like growth factor-I, and platelet-derived growth factor may be relatively unimportant. The factors and molecular mechanisms leading to follicular reorganization of adult human thyroid cells remain to be determined, but are likely to depend largely on intrathyroidal growth factor secretion and cell-cell interaction.

Animals↗