Alvaro A. Giraldo

T-cell regulation in autoimmune thyroiditis.

The mechanisms of induction and maintenance of self-tolerance are still problematical. Several basically different strategies have been proposed to explain the bodys apparent refusal to respond immunologically to its own constituents. 1. Self-reactive lymphocytes may be eliminated by clonal deletion during fetal life, by later clonal abortion or anergy, by receptor blockade or by other antigen-dependent methods of inducing and maintaining unresponsiveness. 2. Active immunoregulatory procedures, similar to those that normally limit immunological responses, may prevent autoimmunity by suppression. The leading candidates as active regulatory mechanisms are the generation of a distinct subpopulation of suppressor T lymphocytes, the production of specific anti-idiotypic antibodies or a combination thereof; that is, induction of antiiodiotypic cytotoxic T lymphocytes (Fig. 1). 3. Both clonal deletion and active suppression may be involved, providing double insurance against self-destruction by immunological means. The decisive question for distinguishing the several mechanisms of selfrecognition is whether lymphocytes that recognize unaltered self-antigens persist in normal animals. Clonal deletion theories in all of their various forms predict that they do not, whereas the concept of active regulation implies the persistence of functional, self-reactive lymhocytes in normal as well as disease states. To study this question we have been analyzing two different models of autoimmune thyroiditis. The first is experimental autoimmune thyroiditis

Regulation of experimental autoimmune thyroiditis: Mapping of susceptibility to the I-A subregion of the mouse H-2

Experimental autoimmune thyroiditis (EAT) is induced in certain mouse strains by injection of mouse thyroglobulin (MTg) with a suitable adjuvant (Rose et al. 1971). The severity of thyroiditis can be ascertained by the extent of mononuclear cell infiltration into the thyroid. Mice of the H-2 k haplotype displayed the highest incidence of severe disease, compared with 10w incidence of mild disease in mice of the b haplotype (Vladutiu and Rose 1971, Rose et al. 1981). The titers of MTg antibodies were less discriminating than histopathologic findings. Lymphocyte proliferative responses as well as the production of antibody and thyroid lesions are under H-2 control (Christadoss et al. 1978). The principal gene controlling susceptibility to EAT, Ir-Tg, was mapped to the K and/or I-A region (Tomazic et al. 1974). At that time, suitable intra-H-2 recombinant strains were not available to permit more precise localization of the gene. Recently we have found that genes at the D end modify the severity of thyroid lesions (Kong et al. 1979). Other studies of a K-region mutant, B6.C-H-2 bin1 (H1), have suggested that a gene in this region regulates susceptibility to thyroiditis (Maron and Cohen 1979). To distinguish the relative contribution of the K and I -A regions to EAT, several new intra-H-2 recombinant strains and H-2K-region mutants have been used to pinpoint the Ir-Tg gene. In this brief communication, we report the definitive mapping of the major Ir-Tg gene to the I-A subregion. Our intra-H-2 recombinant study further suggests that the K region, like the D region, may modify the incidence of thyroiditis. In contrast to the finding of Maron and Cohen (1979), our use of 6 K b region mutants, including bml, did not show an increased incidence of EAT.

Concurrent Induction of Antitumor Immunity and Autoimmune Thyroiditis in CD4+CD25+ Regulatory T Cell–Depleted Mice

When CD4+ CD25+ regulatory T cells are depleted or inactivated for the purpose of enhancing antitumor immunity, the risk of autoimmune disease may be significantly elevated because these regulatory T cells control both antitumor immunity and autoimmunity. To evaluate the relative benefit and risk of modulating CD4+ CD25+ regulatory T cells, we established a new test system to measure simultaneously the immune reactivity to a tumor-associated antigen, neu, and an unrelated self-antigen, thyroglobulin. BALB/c mice were inoculated with TUBO cells expressing an activated rat neu and treated with anti-CD25 monoclonal antibody to deplete CD25+ cells. The tumors grew, then regressed, and neu-specific antibodies and IFN-gamma-secreting T cells were induced. The same mice were also exposed to mouse thyroglobulin by chronic i.v. injections. These mice produced thyroglobulin-specific antibody and IFN-gamma-secreting T cells with inflammatory infiltration in the thyroids of some mice. The immune responses to neu or thyroglobulin were greater in mice undergoing TUBO tumor rejection and thyroglobulin injection than in those experiencing either alone. To the best of our knowledge, this is the first experimental system to assess the concurrent induction and possible synergy of immune reactivity to defined tumor and self-antigens following reduction of regulatory T cells. These results illustrate the importance of monitoring immune reactivity to self-antigens during cancer immunotherapy that involves immunomodulating agents, and the pressing need for novel strategies to induce antitumor immunity while minimizing autoimmunity.

Resistance to experimental autoimmune thyroiditis: L3T4+ cells as mediators of both thyroglobulin-activated and TSH-induced suppression.

Mechanisms suppressive to induction of murine experimental autoimmune thyroiditis (EAT) can be activated by pretreatment with tolerogenic doses of mouse thyroglobulin (MTg) or prior TSH infusion to raise circulatory MTg levels. MTg-activated suppressor T cells (Ts), shown earlier to be Thy-1+ and probably I-J+, were further characterized by in vivo administration of paired rat monoclonal antibodies to distinct epitopes on the L3T4 or Lyt-2 molecule, either on the day of, or subsequent to, initiation of the tolerogenic regimes. The cells required at the time of MTg pretreatment were L3T4+, Lyt-2- and low anti-L3T4 doses had no effect on their activation. The cells that mediated the strong MTg-induced resistance following pretreatment were also L3T4+; their suppressor function could only be abrogated by depletion of L3T4+, but not Lyt-2+, cells. Injection of cyclophosphamide (20-100 mg/kg) either prior to EAT induction or after Ts activation did not affect the severity of disease. Similarly, the suppressor state evoked by TSH infusion could only be abrogated by anti-L3T4 treatment. These findings indicate that both MTg-activated and TSH-induced suppression are mediated by L3T4+ cells. We hypothesize that MTg-specific Ts are present in normal, EAT-susceptible mice in low numbers to contribute to the maintenance of self-tolerance and that they are stimulated by increased levels of circulatory MTg to expand/differentiate and mediate the marked resistance to EAT induction.

Activation of cytotoxic T cells and effector cells in experimental autoimmune thyroiditis by shared determinants of mouse and human thyroglobulins.

Previous studies have shown that T cells from genetically susceptible mice developing experimental autoimmune thyroiditis (EAT) proliferate in response to restimulation with mouse thyroglobulin (MTg) in vitro and differentiate into cells cytotoxic for syngeneic thyroid monolayers. To examine further the effector cells involved in pathogenesis and the determinants on MTg responsible for their activation, spleen cells (SC) and lymph node cells (LNC) from mice immunized with MTg or human (H) Tg, and adjuvant (complete Freunds adjuvant (CFA) or lipopolysaccharide (LPS] were cultured in vitro with MTg or HTg. Control cultures were incubated with concanavalin A (Con A) or purified protein derivative (PPD). The in vitro-activated cells which proliferated in response to MTg, HTg, or Con A adoptively transferred thyroiditis to normal recipients, whereas cells transferred directly without in vitro culture were very ineffective. The capacity to transfer EAT was abrogated by irradiation (1500 R), and SC from CFA-immunized control mice which responded in vitro to PPD stimulation did not transfer thyroiditis. The serum titers of MTg autoantibodies were uniformly low and were not correlated with severity of disease. The localization of EAT-effector (precursor) cells depended upon the site of immunization; they were found in the spleens after inguinal (subcutaneous) or systemic (intravenous) immunizations, but were present in the popliteal lymph nodes after hind footpad injections. Both homologous MTg and heterologous HTg functioned as in vivo sensitizing antigen and in vitro activating antigen for each other; such cultured cells transferred thyroiditis in vivo and became cytotoxic for thyroid monolayers in vitro. These findings show that shared determinants are autoantigenic and thyroiditogenic, and support the hypothesis that EAT-effector cells responsible for initiating thyroid damage include cytotoxic cells.

Role of mouse and human class II transgenes in susceptibility to and protection against mouse autoimmune nhyroiditis

Abstract Mouse experimental autoimmune thyroiditis (EAT), a model for Hashimoto’s thyroiditis, is induced by immunizing with mouse thyroglobulin (MTg). To study the extent of H2A involvement in EAT, we introduced AaAb genes from susceptible k mice into resistant or intermediately susceptible strains which do not express H2E molecules. Thyroiditis was severe in resistant B10.M (H2f) mice carrying the double transgene AakAbk. Likewise, thyroid infiltration was significantly extended in intermediate B10.Q (H2q) mice with the same transgene. To examine the effect of H2E molecules in the presence of H2A-mediated susceptibility, we introduced an Eaktransgene into E– B10.S mice to express the Eβs molecule and observed significant reduction in EAT severity in B10.S(E+) mice. On the other hand, the presence of an Ebd transgene in B10.RQB3 (H2Aq) mice resulting in the expression of H2Eβd molecules did not alter EAT susceptibility, suggesting a role for Eb gene polymorphism in protection against EAT. We have shown recently that the HLA-DRB1*0301 (DR3) transgene conferred EAT susceptibility to B10.M as well as class II-negative B10.Ab0 mice. However, we report here that the HLA-DQB1*0601 (DQ6b) transgene in B10.M or HLA-DQA1*0301/DQB1*0302 (DQ8) transgene in class II-negative Ab0 mice did not. These studies show the differential effects of class II molecules on EAT induction. Susceptibility can be determined when class II molecules from a single locus, H2A or HLA-DQ, are examined in transgenic mice, but the overall effect may depend upon the presence of both class II molecules H2A and H2E in mice and HLA-DQ and HLA-DR in humans.

Resistance of experimental autoimmune thyroiditis induced by physiologic manipulation of thyroglobulin level

The role of circulatory mouse thyroglobulin (MTg) level in activating mechanisms suppressive to induction of experimental autoimmune thyroiditis (EAT) was studied by two regimens to strengthen normal maintenance of self-tolerance in genetically susceptible mice. One was to administer graded doses of exogenous MTg either 7 days apart or daily for 10 days and then challenge the animals with MTg + LPS. The other was to infuse TSH via an osmotic pump for 7 days. The steady TSH infusion for 7 days resulted in an increase in MTg level peaking on Day 3. Such kinetics of MTg concentration in response to TSH coincided with enhanced resistance to EAT induction. After an initial rapid clearance rate of t1/2 of 3 hr, tolerogenic doses of exogenous MTg sustained similar levels for 2-3 days. In contrast, subtolerogenic doses declined to baseline levels in 2 days or less. Clearance can be best explained by a two-compartment model for distribution with an initial alpha phase (t1/2 about 3 hr), followed by a beta phase (t1/2 about 10 hr). We conclude that, for the prevention of EAT induction in the presence of potent adjuvants (CFA or LPS), a threshold, but above baseline, level of either exogenous or endogenous MTg, represented by the beta phase, is required for a critical period (greater than 2-3 days) to activate suppressor mechanisms over and above homeostatic regulation. Whether MTg concentration raised by TSH (TRH) administration activates suppressor T cells as observed after the injection of a tolerogenic dose of MTg remains to be determined.

T-cell subsets in the thyroids of mice developing autoimmune thyroiditis☆

To examine the role of T-cell subsets in the development of thyroid lesions, female CBA/J mice were immunized with 60 micrograms mouse thyroglobulin (MTg) in 0.1 ml complete Freunds adjuvant in both hind footpads. The thyroids were removed 12-21 days later, pooled, and dispersed. The cell suspension was examined by membrane immunofluorescence for the distribution of Thy-1+, Lyt-1+, Lyt-2+, and sIg+ lymphocytes. For comparison, peripheral blood leukocytes (PBL) from the same animals were similarly examined. Throughout this 10-day interval, B cells in the thyroid were consistently below 5%, whereas B cells represented 19-24% of PBL. Thy-1+ cells in PBL ranged from 45 to 59%, whereas Thy-1+ cells in the thyroid were 37-50%. However, only thyroidal T cells showed a consistent decline with time and were replaced gradually by cells without T or B cell markers. In particular, there was a clear shift in the Lyt-1+:Lyt-2+ ratio from about 7 down to 2 in the thyroid as the early predominance of Lyt-1+ cells was followed by a relative increase in Lyt-2+ cells. Our results show that there is an accumulation of Lyt-1+ and Lyt-2+ cells in the infiltrated thyroid. These cells may include MTg-reactive, helper, and cytotoxic T cells which localize (or differentiate) in the thyroid and initiate the lesions.

HLA-DR and HLA-DQ polymorphism in human thyroglobulin-induced autoimmune thyroiditis: DR3 and DQ8 transgenic mice are susceptible

In contrast to H2-based susceptibility to experimental autoimmune thyroiditis (EAT) induced with thyroglobulin (Tg), human leukocyte antigen (HLA) association with Hashimotos thyroiditis, the human counterpart, is less clear, and determining association is further complicated by DR/DQ linkage disequilibrium. Previously, we addressed the controversial implication of HLA-DR genes by introducing HLA-DRA/DRB1*0301 (DR3) transgene into endogenous class II negative H2Ab(0) mice. EAT induction with either human (h) or mouse (m) Tg demonstrated the permissiveness of DR3 molecules for shared Tg epitopes. Here, we examined the participation of HLA-DQ genes by introducing DQA1*0301/DQB1*0302 (DQ8) transgene into class II negative Ab(0) or class I and II negative beta(2)m((-/-)) Ab(0) mice. About 50% and 80% of HLA-DQ8(+) Ab(0) and beta(2)m(-) Ab(0) mice, respectively, developed moderate EAT after hTg immunization, but only minimal response to mTg. The hTg presentation to hTg-primed cells was blocked by anti-DQ mAb in vitro. By contrast, HLA-DRB1*1502 (DR2) and *0401 (DR4) transgenes contributed little to hTg induction. Similarly, DQA1*0103/DQB1*0601 or DQA1*0103/DQB1*0602 (DQ6) transgenic Ab(0) mice were unresponsive to hTg induction and carried no detectable influence in DQ8/DQ6 double transgenic mice. Thus, both HLA-DR and -DQ polymorphism exists for hTg in autoimmune thyroiditis. The use of defined single or double transgenic mice obviates the complications seen in polygenic human studies.

Superiority of thyroid peroxidase DNA over protein immunization in replicating human thyroid autoimmunity in HLA-DRB1*0301 (DR3) transgenic mice

Murine experimental autoimmune thyroiditis (EAT), characterized by thyroid destruction after immunization with thyroglobulin (Tg), has long been a useful model of organ‐specific autoimmune disease. More recently, porcine thyroid peroxidase (pTPO) has also been shown to induce thyroiditis, but these results have not been confirmed. When (C57BL/6 × CBA)F1 mice, recently shown to be susceptible to mouse TPO‐induced EAT, were immunized with plasmid DNA to human TPO (hTPO) and cytokines IL‐12 or GM‐CSF, significant antibody (Ab) titres were generated, but minimal thyroiditis was detected in one mouse only from the TPO + GM‐CSF immunized group. However, after TPO DNA immunization of HLA‐DR3 transgenic class II‐deficient NOD mice, thyroiditis was present in 23% of mice injected with TPO + IL‐12 or GM‐CSF. We also used another marker for assessing the closeness of the model to human thyroid autoimmunity by examining the epitope profile of the anti‐TPO Abs to immunodominant determinants on TPO. Remarkably, the majority of the anti‐TPO Abs was directed to immunodominant regions A and B, demonstrating the close replication of the model to human autoimmunity. TPO protein immunizations of HLA‐DR3 transgenic mice with recombinant hTPO did not result in thyroiditis, nor did immunization of other mice expressing HLA class II transgenes HLA‐DR4 or HLA‐DQ8, with differential susceptibility to Tg‐induced EAT. Moreover, our efforts to duplicate exactly the experimental procedures used with pTPO also failed to induce thyroiditis. The success of hTPO plasmid DNA immunization of DR3+ mice, similar to our reports on Tg‐induced thyroiditis and thyrotropin receptor DNA‐induced Graves’ hyperthyroidism, underscores the importance of DR3 genes for all three major thyroid antigens, and provides another humanized model to study autoimmune thyroid disease.